0% found this document useful (0 votes)
53 views312 pages

The Quiet Revolution - The Electrification of Rural Ireland, - Michael J - Shiel - Dublin, Ireland, 1984 - O'Brien Press, Limited, The - 9780862780562 - Anna's Archive

The document titled 'The Quiet Revolution' by Michael J. Shiel discusses the electrification of rural Ireland from 1946 to 1976, detailing its significant social, economic, and cultural impacts. It highlights how the introduction of electricity transformed daily life, improved agricultural efficiency, and facilitated rural development. The book serves as a historical account of the challenges and successes faced during this transformative period in Irish history.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
53 views312 pages

The Quiet Revolution - The Electrification of Rural Ireland, - Michael J - Shiel - Dublin, Ireland, 1984 - O'Brien Press, Limited, The - 9780862780562 - Anna's Archive

The document titled 'The Quiet Revolution' by Michael J. Shiel discusses the electrification of rural Ireland from 1946 to 1976, detailing its significant social, economic, and cultural impacts. It highlights how the introduction of electricity transformed daily life, improved agricultural efficiency, and facilitated rural development. The book serves as a historical account of the challenges and successes faced during this transformative period in Irish history.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 312

iT

THE QUIET REVOLUTION


Co. Monaghan in the 1940s — the EsB workers are objects ot great curiosity.
THE QUIET REVOLUTION

, , The Electrification
T102 of
SO) Rural Ireland
154 1946-1976

MICHAELJ. SHIEL

CD
D
THE O’BRIEN PRESS
DUBLIN
First published 1984 by The O’Brien Press Ltd.,
20 Victoria Road, Rathgar, Dublin 6, Ireland.

© The Electricity Supply Board


All rights reserved. No part of this book may be reproduced or
utilised in any form or by any means, electronic or mechanical,
including photocopying, recording or by any information storage
and retrieval system without permission in writing from the
publisher.

British Library Cataloguing in Publication Data


Shiel, Michael
The quiet revolution.
1. Electric utilities — Social aspects —
Ireland— History 2. Dwellings — Ireland
— Power supply — History
I. Title
621.31°09415 HD9685172

ISBN 0-86278-056-X The O’Brien Press.

Acknowledgements
We would like to thank the following for permission to reproduce
photographs: The National Library p. 17; G. A. Duncan 189; The
Catholic Communications Centre 255; Ulster Folk and Transport
Museum 263; C. W. Vanston 217; Irish Photographic Enterprises
frontispiece; Maurice Curtain 217; Ronnie Persse for cartoon 133.
We also thank the following for permission to quote copyright
material: Brandon Book Publishers; The Irish Times; Kenny Fine
Bindings; The Irish Press; Burns and Oates. Every effort has been
made to contact copyright holders; if any oversight has occurred we
ask the holders of such copyright to contact the publisher.

Cover design: Michael O’Brien.


Photograph by Frank Fennell.
Book design: Michael O’Brien.
Edited by Ide ni Laoghaire.
Copy-edited by Siobhan Parkinson.
Typesetting: Cahill Printers Ltd., Dublin.
Printed and bound by O’Brien Promotions Ltd., Dublin.
CONTENTS

List of Illustrations 6 16 The Voluntary Organisations 180


Acknowledgements 7 17 WateronTap 194
Foreword — DrTom Walsh 8 18 Electricityonthe Farm 213
Introduction 11 19 RuralIndustries 229
Prologue 13 20 Financing — The Post-
1 Electricity forthe Nation 14 Development Phase 237
2 TheShannon Scheme 16 21 A Tale of Two Parishes 253
3 Rural Electrification — Epilogue 261
Overtures 23 Appendices
4 Payingthe Piper 31 1 The White Paper 264
5 Building the Organisation 48 2 Definition of ‘Rural’ 269
6 REONews 67 3 The Fluctuating Minimum
7 Material Matters 74 Return 271
8 Shipping the Poles 87 4 Fixedand Unit Charges 274
9 ‘Sing the Peasantry and then. . .” 5 Things Chiefly Technical 276
The People — 1946 105 6 Three-Phase versus
10 MovingIn 118 Single-Phase 290
11 Problems 131 7 Corrosion 293
12 SwitchingOn 142 8 Castletownbere — An Adventure in
13 Threading the Way 150 Rural Electrification 294
14 Spreading the Message 156 Notes 297
15 Exhibitions and Competitions 168 Index 302

TABLES
1 ess trading results, 1946 to 1954 42
2 Size distribution of farms in the various counties 110
3 Number of agricultural holdings exceeding one acre 111
4 Areas completed and consumers connected 1958-1965 153
5 Domestic appliance ownership among rural electrification consumers 166
6 Annual electricity consumption by rural electrification consumers 1964-1979 167
7 Rural water supplies in Ireland at beginning of 1980 212
8 Number and percentage of farms of over five acres in each size group in
Great Britain and Ireland in 1956 214
9 Minimum return required 272
ILLUSTRATIONS

Co. Monaghan in the 1940s __frontispiece Canon John Hayes 130


The first rural consumer 12 Backsliders 133
Getting water from the river 17 The ‘switch-on’; the 100,000th
T. McLaughlin; T. McGilligan; Callan’s consumer 145
induction coil 19 The direct approach; spreading the
Se4n Lemass 27 message 159
Song of the wires 33 The mobile showrooms 161
The work-horse; starting out 49 Early-morning demonstration 165
The ‘wiring-gang’; landing the electricity The Munster Agricultural Show 1952 171
cable 55 ESB/ICA farm kitchen; grooming
Working aloft 59 demonstration 175
W.F. Roe; P.J. Dowling 66 Dr Muriel Gahan 189
REO News 69 Carrying water home; the gianttap 197
Linesman 75 Waterontap 209
First batch of transformers 85 Pump installation demonstration 211
MV Make; erecting the first pole 91 Milling and grinding; pigrearing 217
The skinning crew; coaster delivering Cleaning; milking 223
poles; transportation 95 The farm workshop 225
The farm wife 107 Agriculture-based industnes; farm
Rosses Point 1947; new electric equipment 231
pump 115 The Black Valley 241
Entertaining the crew; installing a Glencolmcille 255
radio 121 Milking 263
Advertising the ruralscheme 124
ACKNOWLEDGEMENTS

IT WOULD HAVE BEEN IMPOSSIBLE to attempt this book without the co-
operation of a multitude of former Rural Electrification colleagues. Help
was given not only freely but with enthusiasm and delight that a signifi-
cant chapter in the life of the Irish rural community was to be recorded.
Space would not permit the mentioning by name of all contributors
but neither can some names be omitted. Particularly valuable was the
contribution of the late William F. Roe who despite failing sight and
consequent lack of access to written notes recalled with amazing clarity
the early days of the Scheme. Patrick J. Dowling who succeeded Bill
Roe as Engineer in Charge was also extremely helpful. Patrick Commins
of the Rural Sociology Department of An Foras Taltintais gave me
immense assistance both by his advice and by providing research mate-
rial. Another tower of strength was Patrick Bolger, County Development
Officer for Co. Donegal, and author of The Irish Co-operative Movement,
Its History and Development. Patrick scrutinised my early efforts page
by page and his uninhibited but most constructive criticism helped
immensely in sharpening up the text.
I must also acknowledge the help of the officials of the various
voluntary rural organisations and that of Dr M. T. Casey of Maynooth
College for his information on Nicholas Callan. Thanks are also due to
former REO colleague Tom Timlin who prepared the groundwork for
Appendix 6 (Things Chiefly Technical) with such competence and to
Mrs Annie Joye for allowing me access to her unique collection of press
cuttings of the early days of ‘Rural’.
A very special word of thanks is due to Mrs Kathleen Fitzgerald (née
Robinson) who converted my semzi-illegible drafts, redrafts and re-
redrafts into a most professionally typed final manuscript.
Finally, the story would have been much the poorer but for the rich
lode of contemporary information, comment and anecdote available for
the mining in the fourteen volumes of the REO NEWS covering the first
fourteen years of the Scheme, a publication which rates a chapter to
itself in the book.
I would like to dedicate this book to my wife Delma, to thank her for
her support and for the countless cups of tea and coffee that kept me
going at difficult times.

Michael J. Shiel,
December 1983
Foreword

THIS IS A BOOK OF UNIQUE VALUE, aptly named. It records in considerable


detail not only the bringing of light and power to our rural areas, but
also the economic, social and cultural revolution brought about by this
event. The story told in this book has also, however, a significance
beyond the direct effects on the farm and the home and the creation of
a more meaningful, rewarding and abundant life for those on the land.
It points to many other advantages and lessons arising from this great
project.
For generations, the way of life in rural Ireland had changed but little.
The same season-by season operations were carried out by human and
animal power, often under conditions of great hardship. Farming for
many was of a subsistence nature. Life in the rural home had followed
the same pattern over many generations. With candle and lamp condi-
tioning living there, many of the housewife’s tasks were onerous and
time-consuming, sometimes amounting to downright drudgery. Then
came rural electrification bringing power into the homes and on to the
farms, lessening the burden on the housewife, shortening the time of
many chores, providing light and heat at the turn of a simple switch. On
the farm it provided the means for much greater efficiency in many
operations and a base for the application of modern technology. Again
at the turn of a switch the time for such activities as milking, grinding,
milling and cleaning could be more than halved, apart from creating
greater basic efficiency. It would have been impossible, for instance, to
have brought about the great advance in dairying in the last two decades
without electrical power.
A means was provided for rural development with power for industries
and services which were then badly needed to stem emigration. The
forge could be converted into a workshop to serve the needs of farm
mechanisation. People from cities and towns could enjoy urban living
standards in the countryside. Electricity facilitated social activities in
halls and other places where people gathered. In effect, rural society
was opened up, parochialism and insularity reduced, and rural people
were provided with amenities and opportunities previously the prerog-
ative of urban dwellers.
While all of these advances were not realised at once, the potential
was created. Indeed, looked at in retrospect, the changes were pene- ©

8
trating and, as can now be seen, were cumulative. I know that there are
some who lamented the passing of the old times and their attributes, but
for anyone born and reared in rural Ireland, with experience of day-to-
day living there and the hard lives of many, the change has been
remarkable and rewarding. As one reads this book, the way in which
rural people were provided with new opportunities and the way windows
were opened on a much broader spectrum of life emerges clearly.
There were, I believe, many less tangible but nevertheless important
benefits from rural electrification which are of relevance and value to
national development. High on such a list are the confidence and self-
reliance created by the successful prosecution primarily through our own
national effort — engineering, planning and management skills — of this
mammoth exercise.
As one reads about the details of organisation, of the surmounting of
the problems of pole provision, transport and various technical con-
straints on the innovative input, the picture clearly emerges of an
operation carried out by highly-motivated staff, with pride in their work
and determination to do a good job. These attributes are badly needed
today. Reading this book, I began to realise how privileged I was to
know some of the main actors in this scene.
These days we hear much talk and criticism — much of it so ill-founded
— of the state-supported or sponsored enterprises. Here, however, we
see a project carried through under conditions presenting many con-
straints and barriers by such a body. It is an example of outstanding
enterprise by the public service. There is an important lesson here for
the policy-makers and planners. What has to be done should be clearly
defined, the required resources identified and decided, and then the
organisation concerned should be given a clear go-ahead and allowed to
get on with the job. This does not always happen. Fortunately in this
case, constraints which could have interfered seriously with this great
project were overcome because the organisation and people responsible
were determined and highly motivated.
The successful carrying out of this scheme required the involvement
of many people other than the staff concerned. Many rural families had
to be contacted and convinced: there were reservations against installa-
tion on the part of many who would ultimately benefit. The cost factor
was a serious consideration for many farming people with limited
resources. It was necessary to surmount such barriers and the achieve-
ment of progress in this respect, through a system of local committees
deeply involving the local community, was a master stroke. It is a fine
example of what can be achieved by working positively with people
through a participative community approach. Here again there is a lesson

9
to be learned in dealing with many current problems in terms of enriching
and effectively utilising the democratic process.
Another related aspect of this co-operative approach is the extent to
which the Board effectively mobilised, with beneficial results, resources
in other organisations. Again, this is an example well worth following
where the discipline and capacity to involve others are such essential
ingredients for success.
There is no doubt that this project has achieved much and if more, in
terms of farming and rural development, has not resulted it is not because
of any shortcoming in its organisation and prosecution. The position in
this respect is quite clear. From the agricultural point of view, this
scheme provided the base for the application of modern technology on
our farms and in the processing plants concerned with agriculture. With
as yet only about one-third of our farmers applying modern technology
in a productive way, there is obviously much more scope for develop-
ment. This position can and must be changed by removing the present
barriers to progress. In this respect, education, advice, training and
demonstrations, so effectively used in this scheme, are pivotal for
success.
I feel that it is especially to be regretted that with the first important
step taken in rural development through this scheme, the scheme was
not further advanced on the basis of a well-planned, creative programme
of integrated rural action. This could have ensured the most effective
deployment of our resources of land and people. Perhaps some day we
will have the vision and courage to proceed along that road. The scheme
is a concrete example of what a well-planned input into rural living can
do.
Finally, this project was a great challenge to those involved. This story
shows that the challenge was met and answered. The Board has rightly
ensured that the history and achievements of this project have been
recorded for posterity. It is one of the major advances in our country in
this century and perhaps in social and societal terms the most significant
since the famine, apart from the abolition of landlordism. Michael Shiel
has made a penetrating analysis of the scene through his record of fact,
sidelight and anecdote, personal and otherwise. He has provided a most
interesting and stimulating book for which he and the Board merit deep
appreciation and thanks.

Dr Tom Walsh
Former Director of ACOT and of An Foras Taliintais

10
Introduction

THE STORY OF RURAL ELECTRIFICATION IN IRELAND is the story of


political, financial and technical decisions; the procurement of construc-
tion materials in a situation of world scarcity; the recruitment, organi-
sation, training and motivation of staff. More importantly, however, it
is the story of a rural people left by history and circumstance in a
depressed and backward state accentuated by a recent debilitating world
war, their initial reaction to innovation, their doubts, the persuasion
process and finally the culmination in the acceptance and utilisation of
the new energy to achieve a higher living standard and to avail to the
maximum of the new market opportunities opening up for their produce.
In telling the story the vital contribution of the various instructional
research and development bodies, the voluntary rural organisations and
the improving market situation to the transformation of rural Ireland
will be highlighted — a transformation that was not of course entirely
due to the advent of rural electrification. Rural electrification was,
however, one of the first agents of this change and one of the most
essential of the infrastructural developments that made it possible. In
those early post-war years it could fairly be said that it was the knocking
on the farmer’s door by the rural electrification canvasser that first
sounded the knell of those old, entrenched, conservative and cautious
attitudes, which had kept so many small farmers in the subsistence
bracket, their wives condemned to a life of drudgery and so many of
their children fated to take passage on the emigrant ship.
Fully to document the human background against which the scheme
was carried out would be beyond the scope of this book or, indeed, the
capacity of the writer. However, the social, economic and cultural factors
which operated at the start and which evolved during the course of the
scheme play such an important part that they surface constantly in the
telling of the story. Indeed, the promotion of rural electrification could
be regarded as a classic case-study in the communication and eventual
acceptance of innovation in a largely ultra-conservative society. It is to
be hoped that the necessarily brief outlines sketched in the following

11
chapters will prompt some social scientist or historian to probe more
deeply into this fruitful field of research.
Free use is made throughout of quotations from contemporary official
and newspaper reports, magazine articles, speeches and letters. We are
fortunate in having had scribes who at the time not only recorded the
facts but who captured the flavour, as a modern writer could not, of rural
Ireland immediately after the war and the reactions of its less sophisti-
cated society to the first rumblings of this quiet but momentous revolution
which was so completely to change its way of life.

I ;iV | x 4 s és
The first rural consumer — putting away the old oil lamp i McCullagh’s public
house, Oldtown, 15 January 1947.

12
Prologue
‘Somebody — I cannot remember who — switched on the lights. . .’

THE PLACE WAS OLDTOWN in north County Dublin. The date was 15
January 1947, a cold, windy winter’s evening with patches of snow. The
body of the small village hall was packed with local people, while up on
the stage sat a group of leading citizens, the parish priest and senior ESB
officials. The occasion was the switching on for the first time of electricity
under the new Rural Electrification Scheme. There was, however, a
snag of which the audience was unaware. Severe winter storms had
caused a last-minute fault in the supply line and even now, as the
speeches commenced, a line crew was working frantically in the pitch
darkness over a mile away to put things right.
At the back of the stage was mounted a large switch, which, when
operated, should illuminate the hall and village with the new light. The
hands of the clock now showed eight, the scheduled time for the ‘switch
on’. As the Engineer-in-Charge, W. F. Roe, commenced his speech he
kept one eye on a small table at the side. There sat a gramophone
turntable, connected to the still inanimate supply line. Anxiety sharp-
ened as talk time was running out. Suddenly and unobtrusively the
turntable commenced to rotate; the pick-up dropped onto the record,
and legend has it that a very relieved Bill Roe concluded his speech to
the strains of ‘Cockles and Mussels, Alive, Alive — O’!
A blessing was invoked. The switch was thrown. The hall burst into
light and Oldtown passed into the history books as the first village in
Ireland to be electrified under the Rural Electrification Scheme.
At the end of the year, R. M. Smylie (‘Quidnunc’), discussing the
memorable events of 1947 in The Irish Times, closed his column as
follows:

But how many of these things will be remembered in, say, 2047? I dare
swear that if any event is recorded in the history books (taught through
the medium of Russian) it will be none of those I have mentioned; rather
it will be one which has passed almost unnoticed, amid the turmoil of the
year.
Somebody — I cannot remember who - switched on the lights in some
village — I cannot remember where — and rural electrification took her
bow. And if that does not mean more to the country than all the rest of the
year’s events put together, I shall be very surprised indeed.

15
CHAPTER ONE

Electricity for the Nation

IT COULD BE FAIRLY CLAIMED that one of the fathers of rural electrification


in Ireland in the twentieth century was a man born in an Irish rural
community in the closing days of the eighteenth century.
Nicholas Joseph Callan was born at Darver near Dundalk, Co. Louth,
on 20 December 1799. He went on to study for the priesthood at St
Patrick’s College, Maynooth. At college his intense interest in the pursuit
of scientific knowledge brought him far beyond the traditional curricu-
lum. His genius and devotion to scientific discovery eventually led to his
appointment as Professor of Natural Philosophy at St Patrick’s College
and to a series of basic discoveries and inventions, particularly in the
field of electromagnetics. He achieved considerable international rec-
ognition, but far less than was warranted by the fundamental importance
of his work.’
Callan’s invention of the induction coil in 1836 and his pioneering
work contributed greatly to the development and evolution of the power
transformer without which the widespread distribution of electricity
would have been well-nigh impossible, especially in sparsely populated
rural areas. His output of ideas and inventions was prolific. In 1837 he
made a fundamental discovery, the principle of self-excitation in
dynamo-electric machines. His communication on this is dated May-
nooth, 20 February 1838 — twenty-eight years before Werner van Sie-
mens reported his observations on the same principle to the Berlin
Academy of Science. The Encyclopaedia Britannica of 1860 estimated
the lifting capacity of Callan’s giant electromagnet of 1834 — which,
happily, may still be seen in St Patrick’s College museum — at over one
tonne.
It would perhaps be too much to expect that Callan, labouring away
in the tranquility of his Maynooth laboratory, set as it was in an Ireland
where millions of depressed poverty-stricken peasant farmers were strug-
gling for mere existence, could ever have envisaged that a century later

14
Electricity for the Nation

the fruits of his labours would be employed to help provide a better life
for their descendants. The journey from the induction coils of Maynooth
in the 1830s to the distribution transformers of rural Ireland in the 1940s
was indeed to be a long one, but the experiments of this dedicated
professor, himself of rural stock, may fairly be regarded as the first vital
steps on that journey.
Following the general pattern of the times, the practical development
and utilisation of electricity in Ireland was first confined to the cities and
larger towns. Callan was credited with causing some amusement among
his colleagues in the 1830s by predicting the use of electricity for lighting.
He would have been vindicated when in 1880 the electric filament lamp
was invented almost simultaneously by Edison and Swan. In that year
the Dublin Electric Light Co. was set up and in the same year an
experimental arc-lamp was erected outside the offices of the Freeman’s
Journal in Prince’s St. in Dublin. The following year saw seventeen
public lights — arc-lamps — in the vicinity of Kildare Street, Dawson
Street and St Stephen’s Green in Dublin. The first provincial town in
Ireland to have public electric lighting was Carlow, supplied from a
generator in a flour mill some four miles away. The year was 1889.
Charles Stewart Parnell, who was addressing a large meeting there on
the night of the ‘switch-on’, used the new light as a symbol of a new and
free Ireland in his speech.
Slowly at first, and then more rapidly, electricity spread to all the
principal towns in Ireland, supplied in some cases by the local authority
and in others by privately owned supply companies. By 1925 there were
161 separate electricity undertakings in the country. These operated on
many different standards and voltages and almost all were DC systems.
This type of separate development, beneficial as it was to the cities and
towns concerned, could not form any basis for the general extension of
electricity to the rural community. This necessarily had to await the
establishment of a national alternating current electricity grid with ade-
quate generating capacity. It was in order to provide this nation-wide
electricity supply system that the concept of the Shannon Scheme was
developed in the early years of the newly created Irish Free State.

15
CHAPTER 2

The Shannon Scheme

THE SHANNON SCHEME was the brainchild of a young Irish engineer,


Thomas A. McLaughlin, who left Ireland in December 1922 to work
with the German firm of Siemens Schuckert. In his work on the supply
of electricity to various parts of Germany, he became convinced that
electricity was the key to the economic uplift needed by Ireland. At this
time the Irish Free State, in the aftermath of the Great War, the War of
Independence and the Civil War, was in a very depressed condition.
McLaughlin carried out a large amount of preliminary work on his own
before approaching Siemens or the Irish government, but when he did
it was with a very carefully thought-out scheme of utilising the waters of
the River Shannon to generate cheap electricity for the whole country.
The core of his scheme was the harnessing of the fall of the river, not at
a number of separate points as had been mooted hitherto, but utilising
almost the whole fall at one point, Ardnacrusha.
In concentrating on the Shannon, McLaughlin was flying in the face
of a strong Dublin lobby, which was recommending that the Liffey
should be the first river to be developed in this fashion, as the big
demand for electricity would come from the Greater Dublin region. The
extension of supply to outside the metropolitan area did not appear to
be a high priority.
On the other hand, the McLaughlin/Siemens report clearly envisaged
the inclusion of Irish farms in its proposed scheme.

Ireland is, no doubt, as were all the other countries, faced with the problem
of lessening the dullness and hardship of the farmer’s life and, in addition,
with the problem of labour on the farm. The remedy abroad is electricity
and no doubt in Ireland it will serve the same end. It brings to the farmer
first of all the most pleasant and rational form of lighting, serves to brighten
up his home and, by lighting his yard, barns and cow-houses, makes it
much easier and more agreeable for him to work there on a dark night or
an early winter’s morning. Secondly, it provides for the most efficient

16
, See OL PREM
© ee

The daily drudgery of getting water from the river in early twentieth
century Ireland.

17
The Quiet Revolution

application of labour, as the electrical drive is the only possible economic


mechanical drive for farm machinery, especially on a small farm."

After many discussions the project, which had now been developed in
detail in conjunction with Siemens, was accepted by the government of
the new Irish Free State. There was a lot of criticism from politicians,
business interests and newspapers, Irish and British. The Irish Inde-
pendent and The Irish Times when not actually hostile expressed doubts
about the size and timing of the venture while the Morning Post of
London said, ‘The present needs of Southern Ireland cannot be more
than about 50 million units per annum, whereas the scheme provides for
150 million units. The Irish people . . . . with such an excess of power
.. .. may all be electrocuted in their beds.’ (Apropos of this and many
other similar prognostications made at the time, it is worth recording
that this figure of 150 million units was reached in 1935 and that by 1970
annual consumption of electricity from rural consumers alone exceeded
1,000 million units).
Patrick McGilligan, Minister for Industry and Commerce, was enthu-.
siastic from the start and undertook the extremely difficult task of
piloting the Shannon Scheme through the political storm. On 13 August
1925 the contract for the scheme, which was to cost £5.2 million, was
signed between the government and Siemens Schuckert. The time for
completion was three and a half years.” The official opening was on 22
July 1929, and the first current commenced to flow in October of that
year.
The Electricity Supply Board was established on 11 August 1927, with
the objective of operating, managing and maintaining the Shannon
Scheme and distributing and selling its output on a national scale. It also
got the task, which the government regarded as being of key importance,
of promoting and encouraging the purchase and use of electricity and of
controlling, co-ordinating and improving its supply, distribution and
sale.
In setting up the first state-sponsored business organisation to manage
the nation’s electricity supply, the government of the day showed con-
siderable foresight. Banking and business interests and even the Farmers’
Party were vocal in advocating private enterprise. To many the concept
of a state-run undertaking was anathema. One newspaper described it
as ‘the first fruits of Bolshevism in this country’.? While the government
did for a while examine the possibilities of involving the private sector,
there were major difficulties.
In the first place, large sums would have to be raised to finance the
huge capital expenditure involved in the widespread extension of the

18
Top Left: Dr Thomas McLaughlin. Top Right: Patrick McGilligan. Above:
Callan’s induction coil, which can be seen in Maynooth College museum.

19
The Quiet Revolution

national electricity network. In addition, considerable sums would be


required as compensation in the takeover of existing undertakings. The
return on such investment would of its very nature be extremely long
term. In the perilous economic conditions following the recent birth of
the State, such long-term capital was unlikely to be available from Irish
investors. The possibility of attracting foreign capital was investigated
and dismissed as it was obvious that foreign investors would only be
attracted on their own hard-nosed commercial terms which would not
harmonise with the social and economic objectives of the scheme.
Furthermore, the idea of delivering control of such a vital national
resource to a foreign corporation was in those early days of independence
completely unacceptable. As investigations proceeded, it became
obvious that nothing short of a publicly owned organisation could suc-
cessfully overcome all the obstacles in the way of achieving the ultimate
object of the Shannon Scheme, that of providing a cheap, reliable,
integrated electricity service nation-wide.
An obvious option was the setting up of a separate government
department. It was considered, however, that the detailed accountability
of a government department would not afford the flexibility and freedom
of decision required. What the government was looking for was an
organisation which would be answerable to the Oireachtas and which
would implement national policy as directed, but which would have
maximum freedom to deploy its resources, exercise its business judge-
ment and make its own commercial decisions. The result of the search
was the creation of a state-sponsored or ‘semi-state’ Corporation which
was to be a blue-print for many other such bodies in years to come.
The Electricity (Supply) Act 1927, setting up the EsB gave the Board
powers to sell electricity either in bulk to other distributors or directly to
the consumers, but shortly after its constitution the Board made decisions
which were to have a profound effect on the pace of development.

It would not sell in bulk to other distributors of electricity. It would retail


directly to the customer on a ‘non-profit-making’ basis (i.e. after making
suitable provision for interest, sinking fund, depreciation and other such
charges).
It would acquire all existing electricity undertakings.
It would develop its own technical expertise in the design, construction,
and operation of electricity systems.

These decisions formed the basis for the creation of a single integrated
and country-wide electricity supply system and for the development of

20
The Shannon Scheme

native ‘know-how’, which were of immense value when the time came
in getting rural electrification off the ground.
The decision not to sell in bulk even to the large, experienced
public-supply authorities such as Dublin Corporation and Rathmines
and Pembroke Townships was opposed violently. The Ess, however,
backed by the government, regarded the proper development of a
national electricity supply as something that should not be subject to
municipal boundaries or local politics. Time has shown that this decision,
contentious as it was at the time, was undoubtedly the right one. It
permitted a single-minded national approach to electricity development,
untrammelled by the necessity to deal with and harmonise the interests
of a multitude of local supply authorities, as is still the case in a number
of western European countries.
It was also decided to decentralise all customer-oriented activities,
including power-line and substation construction. This was done by
setting up sub-organisations on a geographical basis — ‘Districts’ — and
giving these the requisite resources and the maximum autonomy in
providing a full electricity service. The boundaries of these Districts and
the smaller ‘Areas’ into which they were subdivided were determined by
the needs of the system and of consumers. They did not necessarily have
to follow county or urban boundaries.

PROMOTIONAL STRATEGIES

In setting up the EsB, the government was fully aware that the physical
construction of a supply system would not in itself be sufficient. The
approach of the average householder to electricity was, it could be said,
extremely circumspect. The price per unit charged by existing suppliers
was in the region of one shilling (Sp; equivalent to about £1 in 1983
terms). The EsB was able to offer electricity at about 1p per unit
throughout the 1930s. However, price alone was not enough to ensure
the best use of the new ‘Shannon’ current. The consumer needed to be
educated, advised and persuaded to make the Shannon Scheme a paying
proposition. This was regarded by the new Board as the most important
of its tasks. A number of strategies were developed.

The Ess defined its pricing policies as being promotional for growth ‘to
encourage consumers to forsake their conservative habits’.
A sales organisation with a contract and wiring department was estab-
lished with the object of reducing the risk of defective wiring and
ensuring safe operation.

21
The Quiet Revolution

Outlets for the sale of tested electrical appliances were opened in all the
main cities and towns.
The Board did not consider its task fulfilled with the arrival of the supply
at the consumer’s meter. It also set up a service repair organisation for
virtually all types of electrical apparatus to ensure that appliances were
repaired and returned to service in as short a time as possible and at an
acceptable cost.

These promotional efforts were effective. By the outbreak of World


War II 170,000 consumers had been connected and were using 320
million units per annum, almost double the output of the Shannon
Scheme. The bulk of the remainder was generated in the Dublin coal-
fired station at the Pigeon House. This achievement was confined almost
entirely to the cities and towns but again, as with many other activities,
the expertise developed was of great value when the time came to make
the same promotional effort among the rural community. Even during
the war years some expansion was achieved, so that by 1946 the number
of consumers had reached 240,000, using 380 million units per annum.
Again, however, the great majority were ‘urban’ consumers. The 400,000
rural dwellings had been virtually untouched.

iy
CHAPTER THREE

Rural Electrification —
Overtures

THE ULTIMATE OBJECT OF THE SHANNON SCHEME — supply on a nation-wide


basis, rural as well as urban — had not been forgotten. As early as 1925,
Professor Boyle of University College, Cork, wrote an article in the
journal Studies on ‘The Possibilities of Electricity in Agriculture’.
Against the background of the proposed scheme, he developed in some
detail the economic advantage of using electricity in various farmyard
tasks such as grain-crushing, root-pulping, chaff-cutting, milking, milk-
separating and churning. He finished his article by pointing out that ‘If
the farmer can be shown that the utilisation of electricity saves him time
and money and makes life more tolerable, the rural demand is likely to
surpass the experts’ expectations’.
In April 1927 Sean McEntee, himself an engineer and later to become
Minister for Finance in the first Fianna Fail government in 1932, read a
paper on rural electrification before the Irish Centre of the Institution of
Electrical Engineers. The minutes of the centre record that, owing to
the importance of the paper and of the subject dealt with, it was decided
to postpone discussion to a special meeting to be held a fortnight later.
A perusal of the minutes of the Irish Centre of the Institution around
this period discloses an increasingly high degree of interest by electrical
engineers in rural electrification. In March 1928 a paper was read by R.
Borlase Matthews on ‘Electric Ploughing’. Discussion broadened into
the general question of electricity in agriculture. In December 1931 two
British members, E. W. Dickinson and H. W. Grimmet, came to Dublin
specially to present a paper on “The Design of a Distribution System in
a Rural Area’ and in February 1938 J. S. Pickles read a paper on ‘Rural
Electrification’. Thomas McLaughlin was elected Chairman of the Irish
Centre of the Institution in 1940 and, as was to be expected, his opening
address was mainly on the subject of rural electrification. In March 1942
J. C. Costello presented a very detailed treatise to the centre on the
economics, suggested technical design and utilisation of a rural electri-

23
The Quiet Revolution

fication scheme. In February 1944 a further paper entitled “Some Aspects


of Rural Electrification’ was presented by J. F. OD. McFaul and J. M.
F. Higgins.
McLaughlin never forgot that the completion of the Shannon Scheme
and the extension of ‘Shannon’ electricity to the cities and towns of
Ireland was only part of the task. Many many times during the thirties
he raised the matter of rural electrification at board meetings (he had
been appointed an executive director and member of the board by the
government). In April 1936, at his instigation, a schedule of rates of
charge for supply to villages of not more than 250 population and isolated
consumers in rural areas was produced. This provided for a ‘fixed charge’
based on the floor area of the dwelling-house and, at a lower rate, of the
farm out-offices. In addition there was a kilowatt-hour (kWh) or ‘unit’
charge of 1.25d. per unit for general domestic purposes (1d. per unit for
cooking and 3d. per unit for water heating). The fixed charge for farm
premises, including out-offices, worked out at about 75% more than the
corresponding charge in urban locations (which was based on Poor Law
Valuation of the premises). For the average small farm, it came to 30s.
(£1.50) per two-month period. This in itself was not unreasonable; the
snag lay in the high cost of making the initial connection.
The Board required that total expected revenue, taken over two years,
had to be equal to or greater than the capital cost of connection. This
requirement was generally referred to as the ‘2 to 1 ratio’. Any balance
in the capital cost not so covered had to be contributed by the house-
holder. While this was not too onerous where houses were clustered
together, requiring no capital contribution in most towns and villages, it
had the effect of excluding practically all isolated rural premises. There
the capital costs involved would be so high that there would be no
possibility of generating the required level of revenue.
In 1937 McLaughlin proposed that the ratio should be extended to 4
to 1 from the existing 2 to 1 level, pointing out that a ratio of 5 to 1 was
quite common in other countries. The Board, however, postponed
consideration of the proposal until the question of the large capital
involved in a comprehensive rural electrification scheme was definitely
settled. It should be noted that even the extension of the ratio as
proposed would achieve a very modest advance in rural electricity
development, as supply to most rural houses would still require very high
capital contributions from the householders. Examples given by
McLaughlin for premises half a kilometre to two kilometres from an
eet ten thousand volt (10kV) line quoted contributions of from £78
to £280.

24
Rural Electrification—Overtures

This assessment of contributions was on the basis of a strict accounting


approach to the costs involved and again there is a record of McLaughlin
pleading for a mitigation of this and the adoption of marginal costing.
However, it was obvious to the Board that at best any mitigation it could
apply would only benefit those householders within comparatively easy
reach of existing electricity networks. Only a very small proportion of
rural dwellings lay within even the two kilometre range. The capital
contribution for the remainder — the vast majority — would be so high as
to be unthinkable.
It should be stressed that the widespread penetration of electricity into
rural areas depended on the prior development of the EsB transmission
and distribution system. A backbone of transmission and distribution
lines had first to be built to carry electricity to the cities and towns. This
was the task on which the Board concentrated in the thirties and early
forties. It involved very heavy capital expenditure (about £18 million by
1946), which was advanced by the government on a totally repayable
basis at around 5% interest (the ordinary bank rate at the time). In order
to meet its interest and repayments and at the same time hold down the
price of electricity, the EsB had to fix strict financial conditions to the
connection of new consumers.
These arrangements, which had been worked out for urban develop-
ment would be totally unsuitable for rural conditions. A drastically new
approach to organisation, planning and financing was needed. McLaugh-
lin was fully aware of this and his constant raising of rural electrification
at Board meetings was undoubtedly intended to stir the Board and
ultimately the government to action.
The government itself was under increasing political pressure to do
something for the farmers. Sean Lemass, Minister for Industry and
Commerce, was an ardent advocate of rural electrification as a means of
improving the lot of the rural dweller. In May 1939 he requested the EsB
to prepare plans to supply rural areas with electricity and to make
proposals regarding finance etc. The EsB immediately undertook a
detailed investigation. This was directed by Dr McLaughlin who was
assisted by Patrick J. Dowling and Alphonsus J. McManus. As well as
being qualified engineers, these two men were farmers’ sons with a good
appreciation of the problems of the rural dweller. They came from
opposite ends of the country: Paddy Dowling grew up on a large farm in
County Carlow, while Alfie McManus’s family farmed in the Lagan
valley area of east Donegal. By September 1939 good progress had been
made but the outbreak of World War II brought other priorities to the
fore.

25
The Quiet Revolution

In the early years of the war the EsB, with a very restricted staff, was
preoccupied with keeping electricity supply available to the community
under very difficult conditions. It could not see its way to devote much
time or resources to planning rural electrification for what then appeared
to be a distant and uncertain future. Nevertheless, the government,
beset though it was with grim wartime problems, still regarded post-war
rural electrification as one of its priorities. Lemass subsequently recalled
(Dail Eireann, 24 January 1945):

In 1939 when I ceased to be Minister for Industry and Commerce [he had
been appointed Minister for Supplies] and the present Minister for Local
Government was taking over from me I drew his attention to certain
matters which were outstanding. One of the matters outstanding was this
report from the EsB on rural electrification. During his period as Minister
for Industry and Commerce, Mr McEntee pursued the Board — almost to
the point of friction — to produce this report.

Lemass, while remaining Minister for Supplies, took back the Industry
and Commerce portfolio in August 1941. In the autumn of 1942 he wrote
to ask the EsB formally if its plans for rural electrification were ready.
The reaction in the EsB was one of astonishment. 1942 had been a very
difficult year and the mere maintaining of electricity supply had taxed
the resources — and resourcefulness — of the Board to their limits. An
acute drought in the early part of the year had severely restricted the
output of Ardnacrusha and had thrown most of the load on to the coal-
burning Pigeon House station. This depended on irregular supplies of
very poor British coal — pithead rejections, in fact, as neutral Eire was
very far down on the coal priority list - which was extremely difficult to
burn. A severe electricity shortage threatened. New connections of
domestic electricity consumers were discontinued, those for industrial
purposes closely vetted and restrictions introduced for all existing con-
sumers. The German armies were at the gates of Stalingrad and the
prospects for rural electrification could not have appeared more remote.
Nevertheless, it was obvious that Lemass meant business. McLaughlin
and his team set to the work with such effect that by 22 December 1942
a comprehensive report had been completed and delivered to the Depart-
ment of Industry and Commerce. The report, which was subsequently
published in a slightly edited and revised form as a White Paper, is
outlined in some detail in Appendix 1. At this stage it will suffice to give
the main features, which were as follows: an analysis of the problem of
rural electrification in Ireland; a review of its progress in other countries;
an outline of the proposed method of development on an ‘Area’ basis; .

26
a great debt.

74}
The Quiet Revolution

a detailed treatment of the costs and the possibility of recouping these


from revenue (about 12% of the capital cost per annum would be
required); the conclusion that a capital subsidy would be necessary and
that in order to keep costs and subsidy down to a reasonable level,
supply under the scheme could only be offered to 86% of rural premises.
Of these it was expected that only about 80% would elect to take supply
so that the scheme as put to the government provided for supply to
280,000 premises, i.e. 69% of the estimated 403,000 dwellings in rural
Ireland. The estimated cost of supplying these was £14 million at pre-
war prices.
The government, undoubtedly pressed on by Lemass, wasted little
time in considering the report. On 26 August 1943 the Esp received a
letter from the Department of Industry and Commerce approving the
proposals subject to a number of stipulations and giving certain govern-
ment undertakings, which may be summarised as follows.

Current ‘rural’ tariff rates to apply (i.e. the 1936 schedule), subject to
any war increases generally applicable.
Electricity to be supplied to all rural premises where the capital cost of
connection did not exceed sixteen times the prospective revenue from
the fixed annual part of the tariff.
The State to compensate the Ess for any shortfall in the required 12%
return on the capital expenditure as yielded by the revenue from the
fixed annual part of the tariff (based on 5% interest rates).
The Esp to plan at once to commence construction in the maximum
number of centres simultaneously (not less than one in each county) as
soon as supplies of material became available.
Subject to above, priority to be given to the most remunerative areas.
Legislation to be prepared forthwith to enable the scheme to proceed.
The Board to plan to complete the scheme within ten years of the date
on which adequate supplies became available.
The maximum amount of equipment to be of Irish manufacture.

On 28 August a statement was issued by the Government Information


Bureau to the effect that work on the scheme would commence as soon
as materials became available after the war. The report was published in
the form of a very detailed and well-illustrated White Paper of 114 pages
in August of the following year, 1944.

28
Rural Electrification—Overtures

On 29 November 1944 the Minister for Industry and Commerce, Sean


Lemass, introduced the Electricity (Supply) (Amendment) Bill 1944, a
massive document covering hydro-electric schemes, acquisition and man-
agement of fisheries (by the EsB), particular powers and duties of the
Board and advances to the Board out of the central fund. The section
(number 41) covering the Rural Electrification Scheme, which was to be
one of the largest projects undertaken by the Ess, was surprisingly curt,
comprising less than half a page — four short paragraphs — in a document
of seventy-seven pages. However, it was sufficient to launch the scheme.
The section provided for an initial advance of £5 million out of the
central fund, of which a half would be repaid to the fund out of monies
provided by the Oireachtas and half by the Ess. The Bill, when passed,
became the 1945 Act of similar title.
In practice, it was agreed that the EsB repayments would be phased
over a period of fifty years for which purpose a sinking fund was set up.
The Act provided for the repayment of the other half of the advances
(the subsidy) to the central fund out of voted monies, the purpose being
to ensure that the initial borrowings to cover the subsidy moiety of the
advances would be repaid out of revenue.”
Even though the section dealing with rural electrification occupied but
a small portion of the total Bill, the debate on it was an extended one
with many long contributions by spokesmen for all parties. These
included Patrick McGilligan, by now in opposition, who had in earlier
years piloted through the Acts setting up the Shannon Scheme and the
ESB. He recalled some of the pressures exerted on him to drop the
Shannon Scheme: ‘... we were prayed by businessmen, by chambers of
commerce and by every newspaper in the country to stop this thing and
to cut the losses, to pay the German firm something to get out of the
whole business’. There was some criticism of certain aspects such as the
fixed charge, but in general there was almost complete approval and a
strong awareness of the significance of rural electrification in improving
the lot of rural dwellers. The contribution of Deputy James Larkin
Junior in quite a short speech towards the end of the debate was typical
in demonstrating this awareness particularly in the area of social change:

In our country, electrification is more than merely producing light and


power; it is bringing light into darkness. The great value of this proposed
scheme of rural electrification is not that we are going to have farming by
electricity instead of by hand, but that we are going to put into the homes
of our people in rural areas a light which will light up their minds as well
as their homes. If we do that, we will have brought a new atmosphere and

By)
The Quiet Revolution

a new outlook to many of these people. It often appals me to think that we


have just under 2,000,000 men and women, the majority of whom are in
many physical ways, debarred from mental development, the application
of their mental faculties, just because they have not got the elementary
physical help of a decent light with which to read. If we can get them light
and nothing else, then I think we have brought about a great change.

In the Senate debate of 7 March 1945 the same awareness of the


benefits of the proposed scheme was manifest on all sides of the chamber.
In his closing speech in the Senate the Minister emphasised the role
which electrification could play in easing the lot of the rural housewife:

I hope to see the day that when a girl gets a proposal from a farmer she
will enquire not so much about the number of cows but rather concerning
the electrical appliances she will require before she gives her consent
including not merely electric light but a water heater, an electric clothes
boiler, a vacuum cleaner and even a refrigerator.

30
CHAPTER FOUR

Paying the Piper

THE WHITE PAPER (see Appendix 1) had made it clear that the primary
problems to be solved in floating the Rural Electrification Scheme were
financial ones. The very low income levels of most rural dwellers meant
that rates of charge which would be adequate to service the very high
capital investment would be beyond the means of the majority of
householders to pay. Some form of subsidy was necessary. In the heavily
industrialised countries such as Britain and Germany the proportion of
rural dwellers was small and this subsidy could be provided by urban
electricity consumers in the form of a small and tolerable increase in
their electricity charges. In Ireland, however, in 1946, there were only
about 250,000 urban electricity consumers. To expect these to subsidise
the electrification of over 400,000 isolated rural dwellings, or even the
280,000 mentioned in the White Paper, was, in the EsB’s book, com-
pletely out of the question. In the initial years of the scheme this principle
appeared to be accepted by the government of the day, but later
governments took a different approach.
Throughout the thirty-odd years of the scheme the question of the
amount of subsidy required and who would provide it remained a
constant issue between the EsB and the government. In simple terms it
could be said that the EsB was prepared and even anxious to extend
supply to every premises, however remote, provided it received sufficient
government subsidy to prevent the scheme becoming too big a burden
on EsB funds. The government, on the other hand, had to meet many
other calls upon the public purse, which restricted the amount available
for rural subsidy. Nevertheless, as time went on there was intense
political pressure on the government to extend supply deeper and deeper
into the uneconomic areas. Naturally much of this pressure was trans-
ferred to the EsB and we have the story of various government
approaches, ESB reactions, subsequent agreements, disagreements, com-
promises and, on occasions, government directives. The end result was

31
The Quiet Revolution

that a scheme which started out in 1946 to supply 69% of the rural
premises in the State over a ten-year period aided only by a government
subsidy finished thirty years later with 98% of all premises connected,
but with the help of a large cross-subsidy from the urban electricity
consumers. ae
Those thirty years saw many ebbs and flows in the economic tide, in
capital availability, in interest rates and in the fortunes of the farming
community. Nevertheless, once started, the work of rural electrification
went on without a break. At some times more monetary assistance was
available from the central government than at others. At some times the
ESB was chided by the government for going too slowly; at others, warned
of the unavailability of capital to meet all its proposals. At times political
pressures to go beyond what was provided for in the statutes, or what
was possible from available resources, were intense; at others it looked
as if the completion of the scheme was a receding target as more new
houses centinued to be built in rural areas while demands for improve-
ment and strengthening of supply to existing consumers cut deeply into
the resources available. Rising capital costs were always a problem. The
original estimate based on pre-war prices for connecting 280,000 con-
sumers was just over £14 million (an average of £50 per consumer). By
1946, when the first pole was erected, costs had escalated by 50%, and
by the end of the initial phase in 1962 the average capital cost of
connecting a new consumer had risen to £150. Each rise in capital costs,
which was not necessarily accompanied by a corresponding rise in fixed
charges, had ramifications in the economics of the scheme. The ratio of
fixed charge revenue to capital cost, on which the ‘coverage’ (i.e. the
percentage of householders to whom supply could be offered at standard
rates of charge) was based, was frequently upset. (See Appendix 3 for
a fuller treatment of this.)

THE INITIAL SCHEME

In 1946 when a start appeared to be possible capital costs had increased


by 50% on pre-war estimates, while the increase allowed by the Minister
in fixed charges was only 20%. This decision upset from the very outset
the basis of the economic calculations. It was now assessed that a capital
subsidy of 89% would be required to give the originally envisaged
‘coverage’ (i.e. supply to 280,000 consumers without capital contribu-
tion). The Esp at this stage continued to insist that rural electrification as
an enterprise would have to stand on its own feet with the help of
whatever subsidy was available. There was to be no question of a shortfall
in the rural account being subsidised internally from surpluses in the

52
na
if

oe ms Nh -
re i
H

Their weapons are climbers and pliers


Their battles are fought up where high tension shoots
And danger’s the song of the wires.
Anon., ‘REO News’, December 1950

63
The Quiet Revolution

non-rural accounts. Rising capital costs would thus require a higher


subsidy, so long as the rural fixed charges were not allowed to rise in the
same ratio. At the beginning, this also appeared to be the approach of
government, but in later years a change in the economic climate and in
government outlook compelled the Ess to alter its stance. -
In 1946, however, this was still for the future. In these initial stages
the government, while acknowledg ing the EsB’s case, was reluctant to
increase the subsidy of 50% incorporate d in the 1945 Act without actual
experience of events. In even agreeing to a fixed subsidy at all it was
going against the advice of the Department of Finance. That department,
mindful no doubt of earlier problems in the controlling of EsB expendi-
ture, had counselled against any fixed rate of subsidy. In a memorandum
dated 2 March 1944 it had drawn attention to the uncertainty of the
degree of subsidisation needed. This issue arose because the return
required by the terms of the White Paper (12% per annum) and the
annual fixed charge revenue provided for were merely estimates. The
memorandum suggested that as an alternative ta a fixed rate of subsidy,
the Minister for Finance would ‘whenever he thought proper’ decide on
the amount of subsidy necessary to enable the Board to break even. It
expressed the belief that in the early years with the ‘best’ areas being
developed first, no subsidy at all might be required.
As could be expected, the Ess did not relish the implications of this
suggestion. If accepted, every aspect of the scheme which affected
financial results — rates of charge, construction costs, working costs,
overheads, etc. — would be subject to detailed investigation by the
Department of Finance. A system of dual management would, in fact,
be set up. The Ess advised the government that if this were to be so the
Board could not be expected to carry responsibility for the success of the
scheme. On the other hand, a fixed subsidy would throw the whole
responsibility for sound management squarely on to the EsB. The govern-
ment took the point. It accepted the principle of fixed subsidy and thus
removed the threat of detailed financial control by the Department of
Finance. In his speech in the Senate debate on 7 March 1945 Sean
Lemass stressed his trust in the EsB’s own control: ‘We can, I think, with
confidence entrust them with the task of preparing these plans and
carrying them through without any of the safeguards which were consid-
ered necessary in the early stage of the Board’s existence.’ g
, Nevertheless, the government was still reluctant to agree to an increase
in the rate of subsidy above the 50% incorporated in the 1945 Act. In
July 1946 it instructed the EsB to go ahead with the scheme on the basis
of a 20% increase in (pre-war) fixed charge rates and said that ‘should
the actual expenditure on the scheme disclose justification for a subsidy

34
Paying the Piper

in excess of the 50% provided for in the Electricity (Supply) (Amend-


ment) Act 1945, the additional subsidy will be paid out of the Transition
Development Fund for a period of two years and will subsequently be
provided by means of amending legislation’. In the same letter, the ESB
was asked to proceed with the erection of the network with all possible
speed.
The Department of Finance had indeed been justified in highlighting
the tenuous nature of the financial projections in the White Paper. The
economics of the scheme were very finely balanced and, because of the
lack of time and resources to carry out more widespread field work (the
Report was prepared in the height of the war), were based mainly on
desk research. There was no certainty about how things would work out
on the ground. The fact that selection of the order of area development
was on the basis of ‘the best annual fixed charge as a percentage of
capital’ might tend to bring the areas with the best return to the fore in
the early years of development, giving a high initial return and perhaps
a misleading impression of the eventual out-turn. As the poorer and
more remote areas were reached, the return on the capital might drop
drastically. On the other hand, increasing familiarity with the progress
and benefits of the scheme might tend to give a higher number of
‘acceptances’ in the areas developed later, and thereby a higher return
so that the best areas might not necessarily be those developed first. No
one could be sure at this stage and the decision of the government to
Keep its powder dry was probably correct.
Some control on expenditure was established by providing in the Act
for only £5 million i.e. one quarter of the total estimated cost, thereby
ensuring that the progress of the scheme came under Dail scrutiny at
any early date. From the Esp’s point of view, this control was of course
vastly preferable to the close, continuous scrutiny of its expenditure by
the Department of Finance which had earlier loomed up as a possibility.

EARLY SNAGS

By the start of the scheme agreement had been reached between the
government and the Ess to offer supply free of capital contribution to all
householders in a selected area where’ the capital cost involved did not
exceed 173 times the annual fixed charge revenue. This represented a
return on capital of only 5.7% — the extreme case. Other, more favour-
ably situated customers in the area would give a better return. In fact
the White Paper had estimated an average return country-wide of 9.7%
based on pre-war costs, but the 50% post-war increase in capital costs as

55
The Quiet Revolution

against the allowed 20% increase in fixed charges had brought this figure
down to 7.76%. a
But what of the premises returning less than the minimum of 5.77?
According to the terms of the scheme, houses where the return was less
than this even marginally were excluded completely from the benefits of
the scheme. They would by definition fall outside the 86% of all premises
for which the scheme was intended to cater. It quickly became obvious
that such a sharp cut-off point was unacceptable. Many would-be elec-
tricity customers were being informed that they did not qualify for
inclusion as their ‘return’ fell below the requisite 5.7%, sometimes by
only a small amount. To the unfortunate householders thus excluded,
particularly in the marginal cases, this appeared as a stupid bureaucratic
quibble and it was not long before irresistible pressure developed for a
change in the system.

THE SPECIAL SERVICE CHARGE

In December 1947, just a year after commencement, the demand for


change was met by the introduction of the special service charge (ssc).
This provided for an annual supplementary payment by householders
whose premises fell outside the required minimum figure. The special
charge was such that when added to the ‘normal’ fixed charge (according
to the schedule) it gave an overall return on the capital cost of 7.76%
(the average national return provided for). The supplementary payment
would thus be small for houses just outside the limit, growing progres-
sively larger for the more remote premises.
At the time this was accepted as a satisfactory approach to the problem.
Later these special service charges were to become the subject of much
criticism and controversy. Many rural Dail deputies and county council-
lors understandably could not see the justice of a practice which, in their
eyes, penalised those of their constituents who lived in worthy townlands
but by chance found themselves at a distance from the power-line, which
meandered erratically through the countryside, ‘designed no doubt by
an intoxicated engineer’, as one aggrieved householder put it!

SUBSIDY

Throughout the whole of the scheme the question of adequate subsidy


was constantly an issue between the EsB and the government. The
accounting approach of the EsB with regard to capital expenditure on the
rural scheme was determined by the fact that any capital costs not met
by non-repayable grants would have to be met from borrowings. For the’

36
Paying the Piper

first ten years these monies were provided by the government’s central
fund and thereafter by loans floated on the national and international
money markets. Provision would have to be made out of revenue for the
payment of interest on such loans and for their eventual redemption. It
was agreed with the government that a fifty-year period would be allowed
for redemption of central fund loans and an appropriate sinking fund
was set up. In addition, a depreciation fund (based on an average life of
223 years) was created to enable replacement of assets to take place. As
in the case of the sinking fund this also required appropriate annual
payments from revenue. (This double provision was to become a con-
troversial item in later years.) Finally, there were a number of ‘other’
costs (operation, maintenance and repair of the network, general admin-
istration etc.) which also had to be met from revenue. It was estimated
on the basis of experience with urban networks that an annual figure of
4% of capital expenditure would be required to meet these.
In the early years interest rates varied from 23% to over 5%. On the
basis of an interest rate of 33% which obtained from 1947 to 1952 the
annual charges would amount to 11.15% of the capital cost of the
network, thus:

% of capital
cost
Interest 3.25%
Sinking fund (50 years) 0.82%
Depreciation (223 years) 3.08%
Other costs (operation, maintenance, repair,
administration etc.) 4.00%
Total 11.15%

It is worth stressing that once a network is erected these charges must be


met irrespective of the actual quantity of electricity used. They thus form
the basis of the fixed charge portion of the electricity tariff.
Under these circumstances the effect of a capital subsidy was to relieve
the interest and sinking fund charges. Thus a 50% subsidy would reduce
these two charges by 50% or in the above example by 2.03% of capital
costs. It is also worth noting that a 100% capital subsidy (i.e. a free grant
of the complete cost) would still leave annual charges of 7.08% of capital
to be met by the fixed charges. An area returning less than this on its
capital cost would therefore involve the EsB in a continuous annual loss
even if the total cost was met by a free grant. This issue was to rear its
head constantly as the scheme progressed and it was found that rapidly

37
The Quiet Revolution

rising costs reduced the returns from many areas to levels far below this
minimum. .
In the early months of 1949, however, when the scheme was just over
two years old (about thirty areas had been completed and eleven thou-
sand customers connected) it was hoped that the average return from
areas developed in future years would be around 8.5%. This would
require a capital subsidy of 65% to break even.’ By this stage materials
were more freely available and the EsB was in a position to step up the
rate of construction. Before committing itself to this higher capital
expenditure the Board wished to settle the question of adequate subsidy.
It requested that the 50% rate be increased to the 65% figure. A
request was also made for the implementation of the undertaking, given
in 1946, that the government would top up the subsidy shortfall on the
first two years’ working. However, a new government was now in power
and, following the devaluation of sterling in 1949 and the outbreak of
the Korean war in 1950, was experiencing great difficulty in providing
adequately for capital investment. It was therefore reluctant to imple-
ment the undertaking given by the previous government. The Board was
invited by the Taoiseach, J. A. Costello, to put its case to the cabinet
committee on capital expenditure of which he was chairman. On 20
October 1950 a deputation from the EsB met the committee. The govern-
ment’s difficulties were outlined and, as a result, the ESB agreed to
continue on the basis of a government undertaking to review the subsidy
at the end of 1952, i.e. at the end of a five-year in lieu of the original
two-year period.

SEPARATE RURAL ACCOUNTS

Around the end of 1949 the government requested that the figures
relating to the rural scheme be extracted from the Board’s main accounts
and balance sheet and shown separately in the annual accounts. This
would give it some basis for monitoring the out-turn experienced and
assessing the Board’s case for a higher subsidy. The request was not
received with any great enthusiasm by the Board. Naturally it was
keeping separate costings so as to control and monitor expenditure on
the scheme, but, as much of the EsB system was utilised to supply both
rural and non-rural customers, many of the figures used were necessarily
based on estimates and apportionments. To quote the Board’s Chief
Accountant, ‘financial accounts are required to be statements of fact. It
has.been proven adequately that exact facts in regard to the economy of
any location of supply, or any type of supply or any rate ef charge are
quite unascertainable.’ The EsB offered to produce not accounts as such

38
Paying the Piper

but an ‘informative statement’. This was not acceptable, however. A


further request for separate accounts drew the joint response from Chief
Engineer, Chief Accountant, and Engineer-in-Charge Rural Electrifi-
cation (6 January 1951) that ‘the Board should be reluctant to commit
itself to the production or publication of accounts purporting to show
true financial results and any figures supplied to State Departments
should be given as statistical statements and with all the proper
reservations’.
In June 1951 the government changed and Sean Lemass was once
more Minister for Industry and Commerce. He in turn continued to
press for accounts, but conceded that specific explanations and appor-
tionments could be shown where required. He emphasised that without
this information he would find it difficult to assess the EsB case for extra
subsidy. The Board was checkmated, and in October the basis and
presentation of a rural revenue account was agreed and an account for
1950/51 sent to the Minister. This distributed the total debits to the
production account between non-rural and rural revenue accounts. The
rural revenue account was charged on the basis of the estimated bulk
supply of units and of the estimated rural proportion of the total system
peak demand, the first being used for the distribution of fuel and
operating costs and the second for capital and administrative charges.
Other expenses such as maintenance and repair were based on direct
working allocations. On this basis, a surplus of £22,700 was shown.
There was, however, a particular point made by the Board with regard
to the depreciation charges which, if agreed, would have reduced this
figure to almost zero. In the account depreciation was charged, according
to standard practice, on the opening capital investment. In view of the
heavy and rapidly expanding rate of rural construction there was a strong
case for basing depreciation instead on the average capital expenditure
for the year. This would result in an extra depreciation charge of the
order of £20,000 which would virtually absorb the surplus.

MINIMUM RETURN REDUCED TO 5%

By the end of 1951 costs had risen as a result of devaluation and price
increases. These increases had the result of pushing a large number of
premises hitherto returning the required 5.7% or better below this figure
and made them liable for special service charges. In order to relieve the
position the EsB decided on two important changes. Firstly, it dropped
the minimum 5.7% requirement to 5%, thereby restoring the proportion
of premises to which supply could be offered without service charge.”

39
The Quiet Revolution

Secondly, it eased the situation for the remaining premises so that the
ssc was now only that required to bring the return up to the new
minimum of 5% instead of the 7.76% hitherto demanded. In the absence
of an increase in the subsidy or an upward revision of the fixed charges
both these concessions had of course the effect of depressing the overall
economic return of the scheme.

SCOPE OF SUBSIDY EXTENDED

In July 1952 the government, at the request of the EsB, gave two very
important rulings on the interpretation of the Act of 1945. It ruled that
capital spent in bringing supply to premises in a rural area after comple-
tion of the initial development would qualify for subsidy. This ruling was
to become very important in the sixties and seventies when a very large
number of premises were connected under various ‘post development’
schemes.
The government also ruled that the cost of reinforcing the rural
distribution system (‘system improvements’) subsequent to initial devel-
opment qualified for subsidy. This ruling was of the greatest importance
in the development of the scheme and in the optimum use of available
capital at any time. It meant that in the early stages the EsB could save
on capital investment, or, more correctly, use the available capital to
connect a larger number of consumers. It did this by taking advantage
of the fact that in the initial period after connection the average new
electricity consumer’s demand tended to be low and generally took some
years to reach the level of established consumers. The requirements of
some grew very rapidly while those of other consumers showed very
little growth even after many years. It was not possible to predict
accurately how a particular consumer’s demand would develop, depend-
ing as it did on many factors. What the engineer could do, however, was
to install the minimum capacity to meet the consumer’s likely short-term
requirements, but design the network so that it could be reinforced at
minimum cost as and when the demand grew. The Board could thus
ensure that the minimum amount of capital investment was lying idle at
any time in the knowledge that when it subsequently reinforced the
system, it would not have forfeited the government subsidy involved in
the extra work. On the other hand, if it had been ruled that subsidy
applied only to initial development, prudence and the interests of the
consumer would have compelled the Board to provide at the outset for
the long-term potential demand at a much higher initial — and overall —
cost.

40
Paying the Piper
CHANGE IN BASIS FOR SELECTION
The White Paper had stressed that rational development required that
areas closest to the existing 10kV system should be given priority in
development. This was in fact substantially achieved by selecting those
areas giving the highest financial return. As with many other aspects of
the scheme, however, social considerations and resulting political pres-
sures challenged the rational approach. Areas with a high sign-up
frequently found themselves pushed down in the queue by areas with
lower acceptance rates because of the more favourable location of the
latter. Sean Lemass was particularly concerned that Gaeltacht areas,
mostly in remote locations, should not find themselves at the end of a
long waiting list and he continually pressed the EsB to revise its selection
procedure. In July 1953 a new selection system was agreed whereby the
percentage acceptance figure in an area up for selection carried equal
weight to the return on capital, thereby giving those remote areas with
a high sign-up an opportunity to move up in the queue.
Costs continued to increase. Interest rates, which in the initial stages
had been 23%-33%, had in 1953 reached 54%. With ever-rising material,
labour and money costs and a declining average return per area the gap
between annual fixed charge revenue and the annual costs which the
subsidy was intended to meet was widening. There was, however, no
slackening in the pace of development. In fact, at the insistence of the
government the pace was stepped up and by November the rate of
connections was 50% up on that of the previous year. For the year
ending 31 March 1954, sixty areas had been developed and 23,477
consumers connected bringing the total to 289 areas completed and
101,000 consumers connected which was over one-third of the target.
.By February 1954 capital costs were 50% higher than at the com-
mencement of the scheme in 1946 but fixed charges had been pegged at
the 1946 level. Most of the areas being selected at this time were
returning only about 5% and in a detailed report to the Board on the
finances of the scheme, the Engineer-in-Charge noted that ‘we have not
yet got into what might be called the “lean territory” ’. There was a
probability, he went on, that as the EsB moved into these areas the rural
scheme would built up a large revenue loss. If this happened the Board
would have to consider the questions of altering the rates of charge or
carrying some or all of the deficit on the rest of the EsB’s business. It was
not long before this prediction was fulfilled.
WITHDRAWAL OF SUBSIDY

In June 1954 an inter-party government took over once more in a


situation of great national financial difficulty. A fresh look was taken at

41
The Quiet Revolution

the overall working of the EsB and at the necessity for the rural subsidy.
The government’s attention focussed on the overall surpluses earned by
the EsB in the years following the introduction of rural electrification. As
shown in Table 1 these were substantial.

TABLE 1

ESB Trading Results 1946 to 1954 as per Annual Reports

Year Surplus Deficit

1946/47 £248,940 —
47/48 £ 74,996 —
48/49 £212 :158 --
49/50 £239,344 —
50/51 £514,107 —
S152 Ee 962 —
52/53 = £488 213
53/54 £233,209 —

The 1946/47 and 1947/48 surpluses eliminated an accumulated defi-


ciency in the net revenue account. The Board then decided that any
future surpluses should go towards building up a contingencies reserve.
By 31 March 1951 this reserve stood at £800,000 but was drawn upon
after the deficit in 1952/53 of £488,213. This deficit was attributed to an
abnormal drought, which reduced the hydro generation at Ardnacrusha
by 30% and increased fuel costs by almost £3 million. The rural electri-
fication account which first appeared for 1950/51 showed a surplus of
£22,700 and for the next three years showed deficits of £46,000, £10,000
and £30,000 respectively, minute amounts in the context of the total EsB
trading.
At the end of October the Es received a letter from the government
noting the substantial net surpluses achieved in every post-war year
except in the abnormally dry year of 1952/53. It appeared (to the
government) that the Board’s liability to repay 50% of the capital cost
of rural electrification including the payment of interest had had but little
effect on the profitability of its operations as a whole. In fact in the
previous year the Board could apparently have undertaken the full
liability for rural electrification and still have shown a clear net surplus
of over £100,000. The Minister, the letter went on, would be glad to
learn if the Board would agree that it should be possible without
detriment to the rate of future development of the scheme, to terminate

42
Paying the Piper

the provision of subsidy by the State, say from 1 January 1955. This was
not all. It was also proposed that the Board would ‘assume liability to
the Central Fund in respect of the moiety of advances for rural electri-
fication repayable from voted monies’, which totalled £2.1 million at 31
March 1954. In addition the Board would become responsible for the
repayment to the central fund of the full amount of advances for rural
electrification in the calendar year 1954 together with interest on each
such advance from the date of the advance to 31 December 1954. These
last two proposals meant not alone would the subsidy be withdrawn for
the future but the Board would have to repay about £4 million already
advanced plus accrued interest. (In the event the actual amount came to
£4.8 million.)
The Board’s reaction as conveyed to the Minister was one of ‘surprise
and anxiety’. (This was, to say the least, expressing it mildly!) The Board
emphasised that the question of subsidy had been examined exhaustively,
and agreed with the government of the day, before the rural development
scheme commenced. The scheme had been undertaken as a matter of
urgent national policy on a positive assurance that a capital subsidy of
not less than 50% would be given and the Board had committed itself
very heavily as a result. The question of subsidising the scheme from
non-rural activities had never been raised. On the contrary, it had been
taken as fundamental by the Board that the rural activities would not be
permitted to affect adversely the electricity costs of the Board’s urban
consumers who numbered 250,000 at the commencement of the scheme.
A carefully documented case demonstrated that many of the argu-
ments used by the Minister were not, in fact, based on the true situation
or did not coincide with the Board’s projections. It dealt in detail with
the necessity for building up adequate reserves and pointed out that the
current rapid growth of capital charges on new and costly installations
could lead to future revenue deficits and that materials and labour costs
were still rising, as indeed were fuel prices. The Ess also pointed out the
risk that higher rates of charge to rural consumers to compensate for the
loss of subsidy might well produce a situation of diminishing returns in
that fewer and fewer consumers would get connected, which could
possibly lead to a collapse of the scheme. Finally, it stressed that so long
as rural electrification continued, capital still had to be provided and
provided in the year the development took place. Savings to the com-
munity would not result from the withdrawal of subsidy; the effect would
be merely to change the incidence of the cost (i.e. from the taxpayer to
the urban electricity consumer). With the prospect of heavy fresh charges
to be borne, it would have to re-consider its whole policy on rural
electrification.

43
The Quiet Revolution

The Minister in turn expressed his feeling that the Board’s views were
unduly pessimistic. He considered, having regard particularly to the
results of the Board’s operations in the year ended 31 March 1954 that
the undertaking had reached the stage when it was in a position to bear
all proper interest and depreciation charges and provide adequate
reserves for contingencies without any further subvention from voted
monies.
Finally, on 8 March 1955, the debate was terminated abruptly when
the Minister formally advised the Board that the government had decided
that the subsidy should be discontinued and the EsB would be required
to meet the full charges except on capital already repaid to the central
fund from voted monies.
Thus, the capital subsidy, the case for which was so carefully developed
and documented in the White Paper and which was intended to meet the
extra ‘social’ costs of rural electrification, was withdrawn. From this time
on, even though in a couple of years the subsidy was to be restored, it
never again met the original objective. The ‘loss’ on rural electrification
increased steadily, as costs increased and as the minimum return on
capital was held far below an economic level, from £30,000 in 1952/53 to
£3.2 million in 1970/71, the last year in which separate rural accounts
appeared in the EsB annual report. By then the cumulative ‘loss’ recorded
amounted to over £19 million. In effect, that figure represented the
amount of cross-subsidy from non-rural consumers.?

THE PEAK IS PASSED

By 1956 the Ess could foresee a rapidly increasing financial deficit on


its rural activities. In August of that year there was a severe tightening
of the State capital budget and the EsB was requested to reduce its capital
demands on the central fund. As might be expected the Board consid-
ered, among other actions, reducing the pace of the rural scheme, which
for 1955/56 had accounted for over £4 million or 12% of the State’s
capital expenditure. It intimated as much to the government and indi-
cated that it was examining the possible cancellation of orders for
materials and a reduction in the number of rural construction crews from
forty to ten.
Very promptly the Board was advised that the Minister did not agree
with the proposed reduction; nor did he agree that any action should be
taken to cancel orders for materials. Any reduction in the expenditure
of the Board that might be necessary because of the current financial
difficulties should be achieved otherwise than at the expense of rural
electrification. Nevertheless, reduction was effected as capital expendi- .

44
Paying the Piper

ture dropped from £4.16 million in 1955/56 to £3.7 million in the following
year and to £2.5 million in 1957/58 with only about half the number of
new connections.
The reason for the phasing down of the construction rate was not
altogether financial. By the end of 1956 over two-thirds of the 280,000
connections envisaged in the scheme had been achieved and it was
considered necessary to ease down in a controlled manner so as to avoid
a sudden stop. This would bring serious redundancy problems, not alone
in the Board’s own labour force, but in all the native industries that had
grown up to supply materials for the scheme. (In 1957 69% of all material
used, amounting to £1,442,000, was provided by Irish suppliers.)

THE SPECIAL SERVICE CHARGE IS QUESTIONED

Amost as sacrosanct in the eyes of the Ess as the capital subsidy was
the principle of the special service charge. This allowed householders
who otherwise would find themselves excluded from the scheme, to be
included on paying a special additional annual charge. In August 1956
the government enquired of the Board as to the possibility of abolishing
the special service charge. A iong reply from the Board outlined the
history of the charge and its purpose. The number actually paying these
charges was comparatively small — about 6,000 — and their abolition in
the case of this number was not of any consequence. What was of
consequence was that if the Board was compelled to supply all applicants
without a special service charge, the most isolated premises could
demand a supply. In that eventuality it might be faced with connecting
about 40,000 very uneconomic consumers at an extra capital cost of £7
million (involving an increase in the annual loss on rural electrification
of about £500,000). It put these points very forcefully to the Minister
and the issue was dropped for the time being.

THE ‘POST DEVELOPMENT’ CONSUMERS

In the autumn of 1956, in view of the projected loss in the rural


revenue account for 1956/57 (close on £3 million), the growing activity
in connecting rural ‘post development” consumers now came under
particular scrutiny. These were householders who had been left out in
the initial development but who now were requesting connection. Even
though the main construction crews had moved on, it had been the
practice to accede to their requests wherever possible but at a somewhat
higher minimum return (now 6.6% as against 4.6% for first-time devel-
opment). The Board now decided that in view of the scarcity of capital

45
The Quiet Revolution

and the withdrawal of the subsidy, it must devote most of its available
resources to the programme of initial development of areas to which it
was committed. A householder in a developed area who had already
been offered and had refused supply and who now wished to be con-
nected must contribute a larger proportion of the cost. A minimum
return of 9% was now fixed for all future ‘post development’ consumers.
The Board also abolished the facility of special service charges for these
consumers. If the normal fixed charge on a premises did not realise the
9% return, a capital contribution was required from the householder to
make up the difference.
The decision was to provoke a strong reaction from the applicants,
their public representatives and the farming organisations. Sean Lemass
was not long back in his old office, with the change in government in
March 1957, before a deputation from the National Farmers’ Association
raised the issue. He in turn exhorted the Ess to mitigate the conditions
and restore the special service charge for the consumers in question. The
ESB took a hard line: it pointed out that with the subsidy abolished it was
in no position to yield such financial concessions, as to do so would
merely accelerate the plunge into the severe loss-making situation which
was forecast.
In a comprehensive survey of the financial position of the rural scheme,
the Board set out its predictions: by the 31 March 1958 £23 million would
have been spent on rural electrification with a £2 million subsidy from
the government (i.e. 9% instead of the 50% subsidy which was provided
for in the original enabling Act). The effect on the rural revenue account
would be as follows.

Retrospective withdrawal of £4.8m subsidy Loss of £245,000 p.a.


Withholding of £4.7m subsidy earned from
date of withdrawal to 31 March 1958 Loss of £300,000 p.a.

At 31 March 1958 annual deficit on rural elec-


trification would be £650,000 p.a.
At this date 184 areas would remain to be done
costing £7m and the estimated deficit on this
work would be £550,000 p.a.
The EsB would thus have spent £30m (subsidy
£2m) and would have incurred as a result an
annual deficit of £1,200,000 p.a.

46
Paying the Piper

In addition the cost over the years of connecting all remaining premises
under ‘post development’ would amount to a further £6 million.
Once again as in the case of its selection policy the ‘rational’ approach
of the EsB had to yield to political considerations. Sean Lemass would
not accept no for an answer. By the following February the capital
contribution from post development consumers was abolished; the
required minimum return was reduced from 9% to 6.6% and the special
service charge restored where the return fell below this.

THE SUBSIDY RESTORED

The case made by the Ess, even though it was a loser in this skirmish,
had obviously made some impression on the government. In December
1958 the Electricity (Supply) (Amendment) Act 1958 restored the sub-
sidy of 50% but only for capital expenditure incurred from 1 April 1958.
The shortfall of £9.5 million up to this date was never recovered.

47
CHAPTER FIVE

Building the Organisation

SEAN LEMASS in his introduction of the 1945 Electricity Bill in the Dail on
24 January had stressed that a job of the magnitude of the proposed
Rural Electrification Scheme had never before been undertaken. It
would use over one million poles and would involve the construction of
75,000 miles of new line (as against the total of about 2,000 miles which
then existed), the erection of 100,000 extra distribution transformers (as
against the current 1,200) and the connection of 280,000 new consumers.
Some doubts were voiced whether this huge task could possibly be
completed before the end of the century: in the Dail debate Deputy
Daniel Morrissey said that he had heard a period of seventy to eighty
years mentioned. The Minister, however, assured him that the aim was
to connect 280,000 consumers within ten years of materials becoming
generally available.
This presented a big challenge to the EsB engineering organisation.
Let us take one item, line construction: at its peak the rural programme
would require the construction of over 6,000 miles of line per year
involving the erection of 100,000 poles in forty different localities. The
fastest rate reached hitherto, at the height of the Shannon Scheme work,
was some 650 miles of line per annum. Procurement of the necessary
materials would likewise involve quantities far exceeding previous
experience, this with difficult post-war shortages of all classes of
materials.
The problems had been studied by the Chief Engineers of the Design,
Operations and Consumers Departments in the EsB, who identified two
important steps that would have to be taken to get the scheme off the
ground. The first was the setting up of a separate organisation concen-
trating on rural construction and development only, under the direction
of one leader of proven ability who would have senior (even Director)
status and who should be given a very high degree of autonomy and
authority covering design and construction, procurement of materials

48
Se ~
SSE

Above: The work-horse of the rural scheme, the Fordson 10 van.


Below: Starting out 1951 — John Murphy, with bicycle, and Seamus
Hayes.
The Quiet Revolution

and recruitment and training of staff. The second was a level of decen-
tralisation of the work itself far exceeding previous practice, which would
mean passing down to the field engineers as much responsibility and
authority as possible.
In October 1944 the Board commissioned P. J. Dowling, assistant to
Thomas McLaughlin, to prepare a report containing proposals on the
organisation of the scheme. He was requested to have it available in one
month.
One of the fruits of the early Board decision in 1929 to set up the
District organisation was now manifest. There existed an effective decen-
tralised framework onto which could be grafted the rural construction
and development structure required. There were twelve Districts in all,
but two of these — Dublin City and Cork City would not be involved in
rural electrification. The other ten Districts between them covered all
rural Ireland. These ‘rural’ Districts were Athlone, Cork No. 2, Dublin
No. 2, Dundalk, Galway, Limerick, Portlaoise, Sligo, Tralee and Water-
ford. The District administrative and operational headquarters was in
the city or town after which it took its name and the geographical
boundaries of the Districts were matched to the electrical ‘feeding’ and
operational layout of the network so that supply to every section and
every consumer within the boundaries was under the control of the staff
of that District headed up by the District Engineer. For this reason, ESB
and county boundaries did not always coincide.
When the main Shannon Scheme contractors departed in the early
thirties the EsB had to rely on its own resources to continue the extension
of the supply network to the towns and villages of Ireland. This it did by
developing skilled crews who worked out of the various District head-
quarters. When the big expansion was called for by the Rural Electrifi-
cation Scheme, these headquarter crews often provided the nucleus of
skilled and experienced supervisors, electricians and linesmen around
whom the new rural construction crews were formed.
A further asset of the District framework was that it provided a
readymade decentralised structure for the administrative and financial
control of the work. Thus, a very important component of the desired
framework already existed which had proved itself both efficient and
effective. The task now was to build on this and work out in some detail
how the total package of responsibilities and various levels of authority
should be divided between the Districts and Head Office.
The District Engineers were able to draw on their experience in
building up the urban network to assess the probable impact of the rural
work and the level of authority and responsibility that could be taken on
at District level. P. J. Dowling involved them in intense consultation and

50
Building the Organisation

by the end of the stipulated month had produced a report embodying a


comprehensive set of outline proposals on the organisation of the scheme
in the field. This was accepted in principle by the Board. The report
emphasised that, because of the magnitude of the work and its dispersed
nature, and because it involved such detailed dealings with so many
people, as much administrative work as possible should be carried out
locally, and as little as possible in Head Office.
A suggested break up of functions was as follows.

Local Functions:
Selection of areas on basis of rules devised by Head Office
Canvass and development
Layout of networks
Issue of wayleave notices and hearing of objections
Planning of construction
Requisition, storage and transport of materials
Erection of networks and connection of consumers
Local publicity and promotional schemes
Control of staff
All associated clerical and accountancy work including credit approval.

Head Office Functions:


Overall development of organisation and general control of staff
Network design standards and construction methods
Procurement of materials, tools and transport
Determination of rates of charge
Formulation of rules for selection of areas
Development of applications of electricity to agriculture
National publicity and assistance in local publicity
Monitoring of progress and financial results.

The report recommended the setting up of a separate rural organisa-


tion in each District. It considered that by doing so it would be possible
‘to develop a single-minded enthusiasm for the work which otherwise
would be impossible to achieve’. It was felt that in the absence of this
enthusiasm and concentration of effort on the one big job, rural electri-
fication, nothing like the desired rate of progress or financial results
could be obtained. In order to ensure co-operation and harmony between
the existing District ‘urban’ personnel and the new ‘rural’ group, and to
utilise fully the experience available among the former, it was proposed
that the District Engineer should be placed in charge of the rural

ol
The Quiet Revolution

electrification work in each District while still retaining control of the


urban work.
The size and composition of the geographical unit which would form
the rural area was then discussed with the conclusion that the parish,
with an average area of 25-30 square miles and containing about five
hundred premises, would be a suitable unit. Proposals were made on
various details of organisation in the selected areas and at District Office
level, most of which were incorporated into the final structure.
Dealing with Head Office, the report referred to the wealth of
experience and expertise which was already available in the various
departments. It suggested methods of utilising these resources without
impinging on the autonomy and responsibility of the Rural Office. It
discussed the degree of autonomy which should be given to the Rural
Office and its relationships with the Board, Head Office Departments,
and the Districts.

Finally, the report summarised its recommendations as follows.

That the parish should be the geographical unit.


That the work would be carried out by a temporary office set up in each
‘selected’ parish.
That these area offices should be subject to a separate Rural Electrifi-
cation Office in each District Headquarters.
That the District Engineer should be placed in charge of this office while
remaining in charge of the existing District Office dealing with urban
supply and the transmission system.
That a separate Rural Electrification Office should be set up in Head
Office.
That this office should correlate and be responsible for all rural electri-
fication activities in Head Office and the Districts.
That the Board should state what authority it wished to delegate to this
office.
That the Board should decide if a Director should be in charge of this
office and if so, that it should nominate that Director. Alternatively,
should the Board not adopt this course, it should nominate someone else
to teke charge and define the relationship of the Rural Office to the
Board.

52
Building the Organisation

The Board agreed to the recommendations in principle but decided


against putting a Director in overall charge of the scheme. It invited
applications from senior engineering staff for the position. William
Francis Roe, District Engineer, Cork City, was selected and appointed
on 19 January 1945.
W. F. Roe, a native of Kilkenny city, graduated in 1925 at the age of
twenty-one with first class honours in mechanical and electrical engin-
eering. After short periods with the Dublin Corporation Electricity
Department and the Shannon Board of Control (set up in 1925 by the
government to monitor the progress of the contractors to the Shannon
Scheme), he joined the newly formed Electricity Supply Board in 1928.
Here he specialised in the conversion of existing town electricity net-
works to EsB standards and he found himself working closely with a
contemporary — P. J. Dowling — who was responsible for building the
new distribution networks in towns that had not hitherto had electricity
supply. It was the commencement of a long association between the two
men, who thus found themselves architects of the EsB distribution system
from its conception.
With the setting up of the EsB’s District Organisation in 1929, W. F.
Roe embarked on a fifteen-year period of service as District Engineer,
in Portlaoise, Waterford and finally in Cork City. Wherever his job
brought him he involved himself in community affairs, and in 1938 he
met Fr John (later Canon) Hayes, founder of Muintir na Tire. This was
the commencement of a strong friendship and of a deep involvement in
a fast-growing movement whose aim was community development in
tural Ireland. He developed a profound knowledge of the social and
economic problems of the countryside and a strong commitment to their
amelioration.
Roe was a man of strong personality and a decisive style of manage-
ment. He had an extraordinary — and to some, disconcerting — ability to
express complicated issues in simple ‘countrified’ language. He had a
natural flair for leadership and a demonstrated commitment to the
improvement of the rural social and economic scene. These qualities
were now allied to extensive experience of the business of electricity
distribution. His first task was to recruit the nucleus of senior staff who
would assist him in the huge undertaking that lay ahead. His most
pressing need was for a high-calibre deputy and assistant in the overall
organisation, planning and supervision and he was very pleased when
Patrick J. Dowling was appointed.
P. J. Dowling, the farmer’s son from County Carlow, was also an
honours graduate in electrical and mechanical engineering. On joining
the esB he worked on the building of urban electricity distribution

53
The Quiet Revolution

networks for the new Shannon Scheme. This was followed by periods in
Athlone, Dublin City and Head Office, mostly working on consumers’
problems. His command of engineering economics and his analytical
ability involved him in various investigations with Dr McLaughlin and as
already recounted he played a major part in the production of the
original report on rural electrification subsequently to be issued as the
White Paper. In 1944, at the request of the Board, he had prepared a
comprehensive set of organisational proposals for the Rural Electrifi-
cation Scheme, most of which were subsequently adopted. In doing this
he had consulted all the District Engineers and had obtained many useful
suggestions, including a very thoughtful and valuable contribution from
the then District Engineer, Cork City, W. F. Roe now to be his imme-
diate chief. The Roe—Dowling partnership formed in 1945 was to survive
through a quarter of a century until both men retired almost simultane-
ously in 1969.
The third member of the original rural electrification triumvirate was
an accountant. Cornelius A. (Neil) O’Donoghue was seconded by the
ESB’s Chief Accountant to look after the specific accounting problems of
the scheme and to ensure their harmonisation with the Board’s general
accounting system. Memories of grave accounting problems in the fledg-
ling years of the EsB were still green. The whole emphasis had been on
getting the construction work done, at the expense of proper accounting
control. The new Rural Scheme would also involve an intense construc-
tion programme with correspondingly high capital expenditure. It was
determined by all concerned that there would be no repetition of the
accounting chaos of the early thirties and that strict accounting proce-
dures would be introduced to operate from the very start of the scheme.
Neil O’Donoghue had transferred to the Board from the private
electricity sector, moving from the local undertaking in his native Bandon
to be manager of the Dungarvan Electric Light and Power Company
prior to its absorption by the Ess. He had qualifications not only in
accountancy but in electrical engineering, having achieved graduate
membership of the Institution of Electrical Engineers and also member-
ship of the Chartered Institute of Secretaries. His talents and training
were thus not confined solely to accountancy matters. He knew the
technical as well as the business origins of every figure in his books. He
_ played an important part in getting the scheme off the ground in the
critical early years and his untimely death from a heart attack less than
two and a half years later, was a grievous loss to the rural electrification
organisation.
Annie Joye was the fourth member of the Rural Electrification Office
staff. A competent woman of strong personality, she acted as secretary

54
Above: The ‘wiring-gang’. Below: Landing the electricity cable from
the mainland to supply Valentia Island.
The Quiet Revolution

to the Engineer-in-Charge. As the organisation grew she built up an


efficient corps of REO typists, which she supervised until her departure
thirteen years later to take up a promotional position in the personnel
branch of the ess. During her period at the REO every new recruit was
inducted willy nilly into becoming a subscriber to Mrs Joye’s Mater
Hospital (Belfast) Football Pools!
On 14 March 1945, the three founder members of the new rural
electrification staff held their first meeting to develop a detailed plan for
carrying out the Scheme. There was no formal agenda. The trio had
available to them two documents, the EsB report on Rural Electrification
(the White Paper) and the report and recommendations on organisation
prepared by P. J. Dowling. W. F. Roe recalled:

We knew that the time programme was ten years, that over one million
poles and other items in equally astronomical numbers must be procured
and erected. The greater part of the world was still at war. We did not
know when the war would end or what the post-war world could be like.
Perhaps it was just as well. The Board had appointed us as the nucleus of
an organisation to carry out the rural scheme. The organisation’s functions
were to construct the networks, connect the consumers and see that the
job was carried out as economically as possible.’

No time was wasted and on 27 April Roe presented to the Board a


detailed set of proposals for the organisation of the Rural Office. The
document of fifty-two pages reflected the importance of establishing at
the very outset a clear understanding and agreement with the Board, on
objectives, responsibilities, levels of authority and relationships. It made
clear and detailed proposals about what authority should be retained by
the Board and what should be given to the Engineer in Charge. It also
dealt with areas where he in turn could pass certain levels of his authority
further down the line. The main issues dealt with were:
General arrangements for the Organisation of Rural Electrification
Responsibilities of the Rural Electrification Office
Activities and staff of the Rural Electrification Office
Rural activities and staff in District Offices
Activities and staff of Area Offices
Staff Control
Manufacture of materials
Relationship of the Rural Electrification Office to the Board and to the
various Departments of the Board
Recruitment and training of staff
Stages of expansion.

56
Building the Organisation

The final pages of the document listed steps which should be taken
immediately so that the Scheme could swing into action as soon as
materials became available. These included such items as the clear
definition of rural, the fixing of rates of charge, appointment of key staff,
mapping and technical design, detailed investigation of materials posi-
tion, location of stores, depots and determination of pay structures and
levels.
The proposals were considered in detail by the Board over the course
of seven meetings. Clarification was sought and given on several items.
In a few cases the Board retained more direct control than was proposed
but in general the proposals were agreed to with a small number of
alterations. The agreed document thus became the organisational blue-
print and was in no small way responsible for the speed and efficiency
with which the Rural Scheme swung into action and achieved its
objective.

RURAL ELECTRIFICATION OFFICE ORGANISATION

The first step was the setting up of the Rural Electrification Office
(REO), the specific ESB management organisation charged with the
carrying out of the scheme.” In charge was W. F. Roe who exercised
broad control over the whole activity. He had been given by the Board
a very high level of authority and discretion and delegated much of this
downwards to the District and Area Engineers. His deputy and assistant
was P. J. Dowling. Three Divisions concentrated on the main activities.
Materials Division was charged with securing all the materials for the
Scheme in the quantities required and ensuring that these were delivered
to the field crews when and where required. This involved the operation
of a large transport fleet. It was also responsible for ensuring that the
quality standards specified for the various items of material were
achieved and at the best possible prices. Finally, the Division was charged
with ensuring that the maximum amount of materials used was of Irish
manufacture.
Recruited from outside the EsB organisation to head this Division was
Patrick McDonald. He had been an engineer with Siemens Schuckert
for two years on the Shannon Scheme and in 1929 had transferred to the
ESB. In 1938 he had left to take up the position of manager of Siemens
Schuckert Manufacturing Co. Ireland. Returning to the EsB his
experience in the area of materials procurement for his previous company
over the difficult war period was to be invaluable, particularly in the
early post-war years when supplies of materials were still extremely
scarce.

7
The Quiet Revolution

Technical Division was responsible for developing the most suitable


electrical and mechanical design of the rural system, for the application
of the most effective technology and construction methods and for the
continuous assessment of standards of construction and productivity. In
the early years of the scheme, very close liaison was required between
Materials and Technical Divisions to cope with the recurrent shortages
of many items by adapting and substituting.
Harold Montgomery was appointed as Head of Technical Division.
He had worked in the Dublin Corporation Electricity Department and
was one of the first engineers recruited by the Ess in 1927. After a period
in Design Department he was transferred to District work. One of his
particular responsibilities was the technical training of the many young
engineers recruited for rural electrification. ‘Monty’, as he was affec-
tionately called, pioneered many of the design and construction methods
used on the scheme which contributed in a very large measure to holding
down the capital costs during a period of severe price escalation.
Development Division was involved in almost every aspect of relations
with consumers and with the public in general, with the setting up of the
selection process, the economic return of areas, terms for supply and
rates of charge. It also was responsible for investigation, development
and promotion of the application of electricity in the home and on the
farm.
Selected to head this Division in 1947 was Robert C. Cuffe. Starting
as a graduate apprentice with Messrs Metropolitan Vickers, Manchester,
he joined Design Department of the EsB in 1936, working in Lines and
Main System Divisions. His work on the effect of lightning discharges
on power lines won him his PhD. Having successfully established Devel-
opment Division in REO, he moved to System Operation Department to
head up its newly-formed Planning and Development Division. He was
succeeded in REO by John Francis Bourke, known to all and sundry as
‘J.F.’, who in the early years of the scheme had been mainly engaged on
the technical aspects, particularly the manual Design of Rural Networks
which became the rural engineers’ bible.
A prime task of the Development Division was awakening the interest
of the rural community in electricity and motivating the householders
not only to put it to use for lighting, but to raise living standards and to
improve production on the farm. One of the earliest appointments made,
therefore, was that of Publicity Officer for the Scheme. Patrick J. Ennis
of Consumers’ Department was appointed to the position in September
1946, charged with the development of the publicity and promotional
aspects on a nationwide basis. Travelling to all corners of the country,
he rapidly developed an effective set of relationships with the numerous

58
oe ete

Completely at ease while working aloft, thanks to well designed and reliable
climbing equipment.

59
The Quiet Revolution

voluntary and statutory bodies concerned with rural development. He


was assisted by a very professional corps of demonstrators and lecturers
who operated throughout the country during the course of the scheme,
attending agricultural shows, area demonstrations and local functions.
Rural Accounts were integrated with the general Accounts organisa-
tion of the EsB, under the supervision of Divisional Accountant ‘Neil’
O’Donoghue who, while seconded whole-time to the Rural Electrifica-
tion Office, still reported to the Chief Accountant. The particular and
often unique circumstances of rural electrification frequently required
special accounting consideration and O’ Donoghue ensured that harmon-
isation was maintained with the established system while still allowing
the new organisation to develop in a flexible and dynamic manner. He
also became responsible for the chartering of shipping to transport poles
from the Baltic to the Irish pole depots and to deliver materials by sea
to rural areas around the west and north-west coasts.
The REO, though it was a separate and practically autonomous entity,
was not completely isolated from other Ess Departments. If experience
or expertise pertinent to the problems of the rural scheme were available
in other Departments, they were freely sought and freely given. In this
way, the total resources of the EsB organisation were marshalled for the
benefit of the scheme.

THE DISTRICT ORGANISATION

It was at District level that the REO came face to face with the consumer.
Staff under the control of the District Engineer addressed meetings,
canvassed and organised, and submitted canvass results from areas for
the selection process. When an area was selected it was construction staff
accountable to the District Engineer who entered the lands to erect the
networks and connect up the houses to the system.
At District Headquarters a Rural Organisation Engineer (ROE) was
appointed as an assistant to the District Engineer solely for rural elec-
trification. He supervised the construction crews, usually three to five,
each of which was headed by a Rural Area Engineer (Rag). In addition,
the Rok had a small staff engaged in the preliminary canvassing, organ-
ising of areas, and the preparation and submission of canvassing results
for the selection. He ensured that the area crews functioned efficiently
and that standards of productivity and workmanship were maintained in
rural electrification work throughout the District.

60
Building the Organisation

THE AREA ORGANISATION

The area crew was the basic unit in the carrying out of the scheme.
Areas, usually based on the parish unit, were of 25-30 square miles. As
they came up in the selection process the area crew moved in, completed
the work in four to six months and moved on to the next in line. At
peak, forty such crews were operating simultaneously throughout the
ten rural Districts.
The rural area staff was made up of a more or less permanent nucleus
of various trades and skills which moved from area to area. Heading the
crew was the Rural Area Engineer who was assisted by a Rural Area
Clerk, a Rural Area Organiser and a Rural Area Supervisor with a
number of linesmen and other skilled workers. In addition, from forty
to sixty general workmen, in some cases up to one hundred, were
recruited locally for the duration of the work in each area.

THE AREA OFFICE

The Area Office was the headquarters for all activity in the area itself.
Initially the tendency was to select an office location in the geographical
centre of the area near a crossroads and erect a prefabricated wooden
building. Later on, however, experience showed that the advantages of
a central location were outweighed by the better communications avail-
able if the office was sited near a post office or in a village where a
telephone was available. The availability of a telephone was of great
importance, as the recurring shortages of materials items required the
RAE to be in constant touch with the supply situation so as to plan
substitutions or re-schedule the construction programme quickly. In the
immediate post-war situation subscriber telephones were scarce and
telephone calls from rural post offices frequently took hours to complete.
Much time was wasted if the telephone was any appreciable distance
from the Area Office as this meant that no other work could be attended
to while the connection was being awaited. In an effort to cope with the
problem a number of radio transmitter/receivers was purchased from
war-surplus stores. By modern standards these were very primitive in
range and clarity, but they did give a limited communication with the
District Office, which in turn could pass on or receive messages from
Head Office. However, it was usually possible, with the help of the local
committee, to secure offices and stores in a village where someone had
a phone. The radio transmitters were phased out and a loud shout down
the village street from the custodian of the local telephone brought the
RAE — or more often the Area Clerk — running when the call eventually
came through.

61
The Quiet Revolution

THE RURAL AREA ENGINEER

Recruiting to the engineering staff of the EsB had been virtually


discontinued during the war years and, when the rural scheme was
launched, very few mature or experienced engineers were available
among EsB staff for transfer to rural work. The few who were available
were of necessity allocated to the planning and organising work of the
Rural Electrification Office. None were available for work in the field.
In the post-war period not many experienced engineers were available
from outside. It became obvious to W. F. Roe that if he was to meet the
required timetable he would have to recruit and train young engineers,
most of them fresh from graduation, to head the field crews.
This policy, born of necessity, turned out to be one of the strengths of
the Scheme. What was required in an effective Rural Area Engineer was
not so much maturity and experience as resourcefulness, common sense,
energy and enthusiasm. These qualities were available in plenty in the
young graduates of the fifties and sixties, many of whom went straight
from college into rural electrification work. The reaction of an elderly
parish priest in the west to this situation was typical of many. On the
arrival of the construction crew, a member of the local committee
introduced the engineer who was to be in charge of the whole operation.
Obviously expecting a more mature man, the old P.P. blinked, sniffed
and exclaimed ‘Bless my soul! Has the boy been confirmed yet?”
In the early years ‘backsliders’ were a serious problem. This term was
applied to householders who had signed application forms but who when
the crew arrived in the area had changed their minds. ‘Backsliding’
generally resulted in a loss of fixed charge revenue without a corres-
ponding reduction in capital cost, so it was up to the RAE to try to recover
the position. This involved the exercise of his persuasive powers — a
subject unlikely to have been covered in his technological education.
Nevertheless, young men* who would have perhaps been too shy to
partake in a class debate in college quickly found themselves standing
on public platforms arguing the case for electrification and countering
objections and criticism from the floor.
While the responsibilities of the job were divergent and, for a young
person fresh from college, formidable, there were a number of important
factors in his favour. There was a strong element of crusade in the job
— the improvement in the lot of the rural dweller — which appealed to
youth and which was a very highly motivating factor. A high level of
authority was given. The objectives and constrictions, financial and
others, within which he had to operate were clearly set out. Apart from
occasional visits from his superiors in the District or Head Office, it was

62
Building the Organisation

left up to him to achieve the required results, but it was made clear from
the outset that a high standard of performance was required. A poor
performer could expect short shrift from W. F. Roe. The great majority
of the young engineers thrived on this combination of high motivation
and the high degree of responsibility and authority given to them. They
tackled the job with enthusiasm and, with their equally highly motivated
crews, achieved a rate of progress in rural electrification unparalleled in
any country in western Europe, at a cost which justified the trust placed
in them. That they and their crews also earned the respect and confidence
of the community among which they worked is demonstrated by the fact
that although over one million poles were erected, mostly on private
land, the number of disputes and wayleave objections which were not
settled on the spot but which rose to boardroom level could be counted
on the fingers of two hands.

THE RURAL AREA ORGANISER

Above all, rural electrification had to do with people, the people of


rural Ireland for whose benefit it had been conceived, to whose homes
and farms it was being brought, over whose land 75,000 miles of line
would stretch and on whose land over one million poles would be
erected. In one way or another the scheme would have an impact on
every rural housekeeper, householder and landowner in the country. It
was therefore of the greatest importance that in each area under con-
struction there should be an Ess officer whose responsibility lay in the
areas of relationships between the Board and the people it was serving.
As well as persuading potential consumers of the benefits of electricity
this officer would have to measure houses and assess the fixed charges,
get application forms signed, serve wayleaves and deal with objections
in the first instance, organise demonstrations of electrical equipment,
advise on selection and installation of electrical appliances and generally
act as liaison-officer between the Board and the consumer.
The recruitment of this new category of officer fortunately coincided
with the large scale demobilisation of the country’s war-time army. As
a result very many well-educated and adaptable young men of both
commissioned and non-commissioned officer rank applied and were
taken on as well as young people of equal quality from other walks of
life. A very short period of training sufficed to equip the new Area
Organisers for their job as the advance guard of the scheme, helping to
form local committees for the preliminary canvass and carrying out the
subsequent official canvass on which the order of selection of the area
was based.

63
The Quiet Revolution

When construction started the task of the Area Organiser was to check
the final acceptance position and try to persuade householders having
second thoughts not to backslide at this stage. The officer also had to
serve the wayleave notices on the landowners over whose land the lines
would run. These tasks were usually combined with that of ensuring that
houses were wired internally when supply was connected so that it could
be utilised immediately. With this constant contact with the householders
the ao became the best known Ess figure in the area.
Simultaneously with these duties the Area Organiser was constantly
promoting the use of electricity to improve the standard of living and to
improve productivity on the farm, mostly using simple inexpensive
methods in the early stages. At a later stage, the aos visited consumers
with sales vans fully equipped to demonstrate larger items of equipment
such as pumps, grain grinders, cookers and refrigerators. The experience
gained by the Area Organisers in dealing with people in different
situations stood them in good stead in later years when the scheme was
completed. Many of them became permanent salespeople and others
developed into specialist advisers in various applications of electricity.

THE AREA CLERK

Unlike most members of the area crew who were recruited specially
for the scheme, the Area Clerk was a member of the permanent account-
ing staff of the District, seconded to rural electrification for a period.
This ensured that the accounting and stores control procedures in the
area conformed to the standard procedures in force throughout the EsB
organisation. The area pay-roll and stores control were the special
responsibilities of the Area Clerk. As the REA and Ao were out ‘on the
line’ for most of the day, the Area Clerk was the only officer normally
available in the Area Office. This being so he also found himself involved
in a multitude of enquiries from callers in the course of the day. It was
usually he who had to ring Head Office on the local telephone to query
the latest material shortage and discuss possible alternatives or go
sprinting down the village street in answer to a shout from the custodian
of the phone when they called back. When aggrieved consumers or
landowners called to the office, it was usually the Area Clerk who had
to bear the first impact of their wrath. Being involved so intimately with
the custody of materials, their shortages and substitutions, the Area
Clerk developed a detailed knowledge of all hardware used in line
construction, the various pole sizes available, where they could be used
etc. and so was an invaluable source of reference.

64
Building the Organisation

Far from being a mere upholder of approved accounting procedures,


the Area Clerk was involved in almost every aspect of area work and
developed a very good relationship with the rest of the crew, agonising
with the Ao over backsliders, sympathising with the RAE when materials
ran short, and rejoicing with them all when the first consumer was
switched on. Many of these people who subsequently went on to occupy
senior accounting and administrative positions in the EsB organisation
still look back on their spell of rural duty as having given them a special
understanding of country life and living.

THE AREA SUPERVISOR

In direct control of all crew members actually engaged on construction


work was the Area Supervisor. In the early phases he was sometimes an
electrician or more often a senior linesman seconded from the District
construction and maintenance staff. Whatever his background the Area
Supervisor was selected for proven leadership qualities. It was he who
had to weld a usually inexperienced group of people into an efficient
working unit, employing a blend of carrot and stick.
Usually in the early stages of the Scheme the Area Supervisor had the
help of one or two experienced linesmen and a few semi-skilled men
who in turn trained some of the more promising of the locally-recruited
general workmen into the more skilled aspects of line work. These in
turn took charge of smaller units of unskilled men who dug holes, erected
poles, strung lines, erected transformers and serviced houses. The locally
recruited trainees were invited to move with the permanent crew to the
next area and so developed into skilled linesmen. Over the period of the
scheme there was a constant renewal upwards of the various levels of
skills as young recruits moved from unskilled to semi-skilled, to skilled,
to Linesman, to Chargehand Linesman and in many cases to Area
Supervisor or to senior Line Supervisor positions on the District staff.
One of the attributes of a good Area Supervisor was the ability to detect
qualities in young recruits which would fit them for the promotional
ladder.
The four officers described, Area Engineer, Organiser, Clerk and
Supervisor, together with a small number of skilled and semi-skilled
workers — about ten to fifteen in all — formed the nucleus of the rural
area crew and travelled from area to area. In each area, local men were
recruited for the less skilled work in excavation, pole erection etc., to
form a total crew usually of forty to eighty or sometimes even up to 100.
A crew of fifty could complete an area of four hundred consumers in
about five to six months. At the peak of construction, which was reached

65
The Quiet Revolution

in 1956, the total number of crews engaged in the work had reached
forty. In that year over 10,000 kilometres (6,000 miles) of line were
constructed involving the erection of 112,000 poles. Ninety-nine areas
were completed and over 34,000 new rural consumers connected.

W. F. Roe, left, first Engineer-in-Charge of rural electrification and P. J.


Dowling, his successor.

66
CHAPTER SIX

REO News

IN TACKLING what was then one of the largest development projects


ever envisaged in the country, W. F. Roe was keenly aware of the
importance of good communications with staff. If a high standard of
performance was to be achieved, the staff needed not alone to be well
briefed and motivated at the start, but to be constantly refreshed with
information on the progress of the scheme, advised of developments in
all aspects of the work, sustained when difficulties arose and motivated
to give of their best at all times.
As the peak of the scheme would find forty separate working units
each of from fifty to one hundred people of various disciplines and skills
scattered throughout the 26,000 square miles of the State, some in very
remote localities, day-to-day personal contact was out of the question.
In an intensely busy headquarters there was little time or opportunity
for developing a communications organ. W. F. Roe, recognising the
importance of such communication from an early stage, decided that a
start, however imperfect, would have to be made. One day in December
1947 in a typical gesture he called in a typist and, in his own words,
‘dictated the first issue of REO News from cover to cover’, three foolscap
pages which were then issued in stencilled form. The first paragraph of
this first issue reads:

In order to keep the rural staff informed of the progress of the Rural
Electrification Scheme, it is intended to issue REO News monthly. The
date of issue will be towards the end of each month when progress has
been reported to REO from each area and District. General information
regarding the various aspects of the scheme such as suggested construction
methods, general position of the supply of materials, and applications of
electricity to agriculture will be given and perhaps a social column so that
you will have some information about your friends in various parts of the
country. Contributions from members of the Area or District staffs will be
welcomed.

67
The Quiet Revolution

The issue then went on to give an overall review of the scheme. Areas
had now been selected in twenty-three counties and it was hoped to have
selections in the remaining three - Cavan, Monaghan and Donegal —
before the end of the month. With regard to materials, pole supplies
were reasonably secure with the purchase of 114,000 poles in Finland, of
which 95,000 had been delivered before the ice closed the Baltic ports.
These would supplement the much smaller quantities of Irish poles that
were available. The situation regarding other key items of material was
then gone into in some detail.
From small beginnings REO News blossomed into a sizeable monthly
of about twenty pages whose arrival was awaited with eagerness by the
field and district staff. After the first issue, W. F. Roe succeeded in
delegating responsibility for its production while retaining his prerogative
as Editor in Chief. Effectively the editorial responsibility was placed on
the shoulders of John Francis Bourke, assistant Head of Technical
Division. In 1950 he succeeded R. C. Cuffe as Head of Development
Division. There was a polite fiction that the identify of the editor was an
official secret, but even the rawest recruit was not left long in doubt
about the authorship of the pungent editorials and the pithy snippets
that laced the pages of the News, making it sometimes controversial,
sometimes pontificial, but never dull. J. F. — he was never known except
by the initials — remained editor and producer of the magazine for almost
its full fourteen years life.
The magazine retained its foolscap format for the first twelve issues
and then changed to the handier quarto size. Costs of production were
kept to the minimum and for twelve years it was produced in stencilled
form using the normal typing and duplicating facilities of REo. An
exception was the December issue, which was regarded as the Christmas
number. This was a fully printed product and was more light-hearted in
tone. 1953 saw an innovation in a printed cover page of good quality
glossy paper which enabled photographs of rural electrical appliances to
be reproduced and gave the publication a somewhat more professional
look. Finally, for the last two years of its life, REO News appeared in an
attractive fully printed form which made it possible to include a wider
range of material.
In the course of its life — from December 1947 to November 1961 —
REO News was to play a most important management role in informing,
educating and motivating the widely dispersed staff, in countering their
sense of isolation and in building.up a team spirit. It provided a vehicle
for the often spirited exchange of views, for criticism of performance of
management and field worker alike. It acted both as a suggestions box
and as a safety valve as its columns were open to all rural staff to make

68
“The old order changeth...
This is the last issue of “ R.E.O. NEWS”.

In December, 1947, the first number of the NEWS made its modest appearance

‘in order to keep the Rural staff informed of the progress of the Rural Electrifi-
cation Scheme”. For fourteen years it has fulfilled that promise.

With the end of the area development programme next year over 280,000 rural
consumers will be connected to our networks at a cost of over £30,000,000. This
was no small achievement. It was not always easy, but the co-operation and
enthusiasm of the staff carrying out the Scheme in the areas or planning and
helping it in the District and Head Offices overcame all obstacles. Now the job
is almost done and “ R.E.O. NEWS” has outlived its usefulness.

In the second issue of “ R.E.O. NEWS” (January, 1948) the following note
appeared :—

“ Quidnune ” discussing the memorable events of 1947 closes his column in the
“Trish Times” as follows :—

“But how many of these things will be remembered in, say 2047 ? I dare
swear that if any event is recorded in the history books (taught through the
medium of Russian) it will be none of those I have mentioned; rather, it will
be one which has passed almost unnoticed, amid the turmoil of the year.

Somebody—I cannot remember who—switched on the lights in some


village—I cannot remember where—and rural electrification took her bow.
And if that does not mean more to the country than all the rest of the year’s
events put together I shall be very surprised indeed.”

No comment is required. Though the Scheme is not quite finished “ R.E.O.


NEWS bows out with pride.

\.|. Bewtng.

69
The Quiet Revolution

suggestions and criticisms or simply to sound off and get rid of a head of
steam. In this way it provided a very effective feedback to management.
Through its leading articles in particular the magazine focussed the
attention of the staff on the broad aspects of the work. It praised or
criticised performance and exhorted all to greater efforts. It highlighted
areas where it felt improvement was required — construction costs and
standards, work output, vehicle mileage, accidents, appliance sales,
backsliders etc. In the area of work output pe: crew it was frankly
stakhanovite in its approach, setting up methc , of comparing the
performance of the different crews and so hoping to inculcate a sense of
rivalry. The first step was to establish a way of measuring work and of
expressing amount of work done, however different in character, in
terms of comparable ‘work units’. The monthly reports from the areas
showed the number of work units done and thus, month by month, the
performance of the crews both in output and cost could be compared.
A good average-sized crew working well could achieve over one
thousand work units in a four-week month and so the idea of the M
areas (M = 1,000) was developed. Each month the M areas were listed
together with the names of the engineer and supervisor concerned
together with suitable words of praise. While non-M areas were not
mentioned, they were of course conspicuously absent and motivated, it
was hoped, to earn the accolade at a future date. If a crew either larger
or smaller than average achieved M status the praise was muted or
heightened accordingly.
An item that proved popular was the Top Ten where the cumulative
performance of each crew (in terms of work units) from the beginning
of each year was logged. Each issue of REO News gave the placings of
the ten crews which currently had the highest cumulative outputs. The
crews and their placings changed from month to month for various
reasons not all under the control of the RAE. The prevalence of rock, for
example, could increase costs and reduce output. Similarly, completing
one area and transferring to the next could affect work output for the
month in which the change occurred. The placings at the end of the
twelve months were, however, generally accepted as a fair measure of
each crew’s efficiency.
While this may appear rather naive in these days of more sophisticated
management techniques and, indeed, more sophisticated worker-
response, it was effective in keeping the importance of high productivity
constantly to the fore. The Top Ten was invariably one of the first
features to which engineers and supervisors turned on receipt of their
copy of REO News.
Each of the three Headquarters Divisions — Materials, Technical and

70
REO News

Development — was given space in the journal to keep staff up to date in


its particular area.

MATERIALS DIVISION

In the first few years of the scheme in particular, materials for con-
struction were difficult to obtain and deliveries were sporadic in the
extreme. This made the achievement of high productivity in the field
difficult and sometimes impossible. If, however, the RAE was kept in
constant touch with the materials position — which materials were likely
to go short and when, or whether deliveries of some materials in short
supply could be expected and when — he could do much to counter the
effect of the shortages by the suitable re-deployment of his manpower
on the ground. The Materials Division Notes, which appeared in each
issue of REO News and which gave the most up-to-date position on each
item of stock were therefore of great value in allowing the most efficient
planning of the crews’ time and thereby keeping up the momentum of
the activity.

TECHNICAL DIVISION

The notes provided a continuous updating of the basic design manual,


Design of Rural Networks of which every RAE had a copy. New materials
coming on the market, new technology, practical construction problems
met in the field and their solutions, improvisation and substitution for
scarce materials were some of the subjects treated. A random sample
from just three of the over 160 sets of Technical Division Notes gives the
following:

Balancing of single phase loads, tree cutting, undesirable servicing,


faulty S.C.A. conductor, service poles.

Binding-in S.C.A. conductor, rural system improvements, bolted and


compression connections.
Records of P.O. crossings, new eaves fitting, H.T. construction on L.T.
lines, damage to roofs, earthing in rocky country.

DEVELOPMENT DIVISION

Materials and Technical Divisions had to do with the physical construc-


tion of the network. Development Division focussed on the consumers

71
The Quiet Revolution

themselves. It monitored the economics of whole areas and individual


extensions but the greatest part of its efforts went to ensuring that once
electricity was made available it was utilised to the greatest possible
extent. Demonstrations, exhibitions, participation in shows, educational
activities, liaison, with rural voluntary organisations, development of
applications of electricity to the rural home, to providing water on tap,
to agriculture and horticulture, promotion and sale of appliances — all
these activities and more came under the umbrella of Development
Division. Consequently its notes covered a wide field including such
items as, How to Measure a House on Your Own, Reports on Shows,
Sales campaigns, Conversion of Milk Separators to Electric Drive, Test
Reports on Appliances, Diagnosis of Pump Troubles and What the
Salesman Should Know about Welders. If these notes could be said to
have one recurrent theme it was the necessity for selling as much suitable
electrical equipment as possible to the newly connected consumers so as
to ensure that from the start they would reap the maximum benefits from
electrification.

FEEDBACK

A very important role of REO News was to provide a forum in which


the men in the field could exchange their experiences of dealing with
problems, criticise and make suggestions for changes in policy and
generally provide feedback to management about how the scheme was
progressing. This was of particular value in monitoring public reaction
to the scheme as the people in the field were in daily touch with a broad
cross-section of the community.

PROGRESS

The progress of the scheme was faithfully recorded in the pages of


REO News. As work commenced in each area, the number of poles
erected, length of line strung and consumers connected were recorded
month by month until the note ‘completed’ was entered and the crew
moved on to their next location. The grand total to date was also given
so that all readers were kept up to date on the overall picture.

SOCIAL

Finally and most importantly for staff members, the comings and
goings of their colleagues were chronicled. Newcomers to the staff were
welcomed by name, transfers from one area to another or to work areas

72
REO News

outside ‘rural’ were meticulously recorded. Engagements and marriages


were given full coverage, often in intimate and exquisite detail. When in
the course of time the marriages bore fruit, the newcomers were wel-
comed into the world. Thus REO News contributed to building an esprit
de corps which persisted long after the disbandment of the REO
organisation.
In November 1961 the last issue of REO News appeared. It had been
founded ‘to keep the Rural staff informed of the progress of the Rural
Electrification Scheme’. Now the pioneering work of initiating the rural
community into the benefits of electricity had been successfully com-
pleted and rural dwellers were approaching the sophistication of their
urban cousins in things electrical. The task of REO News was done and
it marched out with banners flying to make way for a new magazine,
Prospect, which would encompass a wider marketing spectrum under the
editorship of Michael V. O’Connor.

73
CHAPTER 7

Material Matters

THE PERIOD IMMEDIATELY AFTER THE WAR was not the most propitious time
for obtaining the wide range and large quantities of materials required
for a nation-wide rural electrification scheme. There had been no oppor-
tunity to accumulate stocks during the war and in the initial years,
keeping the construction crews supplied was a major problem. While a
small number of items — poles, conductor and transformers — accounted
for the bulk of materials (about 80% of the cost), there was a very large
range of other items, the absence of any one of which would prevent the
completion of the networks. Insulators with their pins and suspension
clamps, crossarms, stayrods and staywire, fuses and fuse isolators, con-
nectors and even nuts and bolts were just as vital. The position was
aggravated by the recurring failure of most suppliers (they had their own
difficulties) to deliver even reasonably close to the promised dates. Some
of the shortages would have had a far more serious effect if the Rural
Area Engineers had not shown their ability io adapt quickly to the
ever-changing materials situation by rearranging the work of their crews.
After the first couple of years the supply situation improved, but so
also did the demand as the number of crews was stepped up. It can be
said that at any point in the course of the scheme there were always
some components in short supply and much time and ingenuity were
devoted to developing substitutes or in rationing out available supplies
according to greatest needs. Very often unofficial bartering sessions were
held between neighbouring Rural Areas as items not immediately needed
were traded for those without which work would be held up.

POLES

The item that accounted for the greatest cost was poles. It was
estimated that over one million poles would be required and that when

74
Linesman, aloft, prepares for the stringing.

n5
The Quiet Revolution

construction reached its peak, over 100,000 poles per annum would be
erected. (In fact, in the year 1955/56 112,616 were erected.) A thorough
search for suitable wooden poles was undertaken in Ireland both in State
forests and in privately owned plantations. Some suitable supplies of
larch and Scots pine were located, generally in small parcels in woods
scattered throughout the country from Killeshandra in County Cavan,
to Tallow in County Waterford. By the end of 1945 a total of 6,000 poles
of Irish origin had been ordered. This, however, was only a small fraction
of requirements and it became obvious that if the scheme was to be
constructed using wooden poles, the bulk of these would have to come
from abroad.
A possible alternative to wood as a material for poles was reinforced
concrete as used extensively in some European countries. The govern-
ment was most anxious that this alternative should be examined as it
feared that sufficient wooden poles would not become available, even
from abroad, to commence the scheme. In the case of reinforced concrete
poles, on the other hand, apart from the reinforcing steel, all materials
required could be supplied from native sources. A detailed costing
exercise was carried out, which indicated that on the basis of post-war
prices, the use of concrete poles would add about £7 million to the cost,
i.e. £28 million as against £21 million.’ A small number of concrete poles
was manufactured by Moracrete Ltd., Crumlin, Co. Dublin, to gain
practical experience of manufacturing and erection costs. These were
erected at Fairyhouse in north County Dublin and in Ballytore rural area
in County Kildare, where they continue to give good service. However,
their weight made them difficult and expensive to handle and the price
differential remained and still remains in favour of wooden poles. Now-
adays, however, a high proportion of the wooden poles required is
supplied from Irish forests.

Source of Supply 0 The search for wooden poles was extended to


Scandanavia. Timber was available in Sweden and Norway, but in the
post-war period most of it was earmarked for home use and the govern-
ments concerned were keeping a tight rein on exports. Sweden in
particular was curtailing exports to Great Britain because of failure to
obtain reciprocal imports, and so in that country sterling was not a very
popular currency at the time.
The Swedish allocation in 1946 for all pole exports to Ireland, including
P 7 T requirements, was for between 5,000 and 10,690 poles at a very
high price. Norway had fairly large quantities on offer, but again the

76
Material Matters

price was unacceptably high. Finland offered the most favourable pros-
pect of obtaining sufficient poles at a reasonable price. Offers were
received through agents and by May 1946 16,700 Finnish poles had been
ordered to supplement the 6,000 obtained at home. This meant that
supplies for about sixteen areas were now secured.
In view of the huge quantities involved it was decided to make direct
contact with the principal Finnish pole suppliers and ascertain their
ability to meet large orders. In August 1946 Neil O’ Donoghue travelled
to Finland. He found that there was ample timber suitable for poles
standing in the Finnish forests. The Soviet Union had first call on all
Finnish products as war reparations, but at this time it appeared that it
had sufficient supplies of poles and was concentrating on manufactured
goods, ships, machinery and prefabricated housing. For this reason,
Finland’s ability to supply poles to Ireland was expected to be good.
Moreover, Finland was most anxious to sell for sterling, which of course
was the Irish trading currency of the time.
O’Donoghue made contact with the Finnish Pole Exporters Associa-
tion, a body set up to deal with the export of poles to countries with
which Finland had concluded agreements during the war years and which
was intended to function during the reparations period. Four members
of the association, who between them handled 70% to 75% of all pole
exports, arranged to deal with REO as a group. They agreed that one of
their number, Mr Onni J. Salovaara, would, as well as dealing on his
own behalf, act as agent for the other three with regard to all shipping
matters and correspondence.
After some days of intensive negotiation, agreement was reached for
the supply of 90,000 poles at a satisfactory price (averaging about £2 per
pole). W. F. Roe received a telegram from O’Donoghue advising him of
the final asking price and requesting permission to close the deal. He
decided to buy them but had a problem in conveying his agreement to
O’Donoghue without compromising the latter’s chances of improving
further on the price in the final sealing of the bargain. The difficulty was
that the only telegraphic address for O'Donoghue was the Salovaara
Office in Helsinki, so that any indications of satisfaction with the price
would immediately become known to the Finns. Not for the first or last
time the native language was called upon to help out. The telegram from
Dublin read: ‘Praghas maith go leor, Roe’ (‘Price satisfactory, Roe’).
O’Donoghue solemniy informed the group that they would have to
improve on the price to get the contract, which after some consultation
they did. O'Donoghue immediately telegraphed his boss: ‘Have done
deal. Get Board approval in ainm Dé’ (‘. . . in God’s name’). The first
major contract for the scheme had been successfully negotiated. Subse-

aa
The Quiet Revolution

quently the Finnish suppliers found it possible to offer a further 24,000


poles as an extension of this contract making the total quantity purchased
for delivery in the 1947 shipping season to 114,000.
While other Finnish suppliers subsequently came on the sceme, the
firm of Onni J. Salovaara and its associated companies remained major
suppliers of poles for rural electrification and a relationship of the highest
mutual respect was established which survived throughout the thirty
years of the Scheme. Many years later, Mr Salovaara got around to
asking about that first telegram from Dublin — the Finns had tried all the
standard international telegraphic codes without success — and was highly
amused to find that he had been stymied by the Irish language.
At all times preference was given to the purchase of Irish poles but
these were never available in anything approaching the quantities
needed; so up to one million poles were imported, almost entirely from
Finland. They were purchased at very keen prices, firstly because of the
large quantities required, and secondly because the REO conducted
business directly with principals and not through middlemen. A further
attraction to suppliers was the small sizes and assortment of poles that
were acceptable: by international standards the pole sizes required by
REO for rural work were generally on the small side; furthermore, REO
allowed a good latitude in regard to the numbers of the various sizes
acceptable to it. The result was that the pole producers did not have to
do so much ‘picking’ in their felling operations and found it easier to
handle the smaller sizes involved. This was reflected in the prices quoted.
While there was a strong element of competition among suppliers for
REO business, the Finnish government made sure, through a system of
export licences, that poles were not sold below what it considered a fair
price. On some occasions, in an effort to get business, small suppliers
quoted very low prices to find that they were unable to get export
licences. On other occasions, it was found necessary to split orders
between a number of suppliers to ensure that licences covering the
required total were granted.
Inspection of poles when felled was a major task. REO insisted that
every individual pole should be inspected and passed before shipping.
Initially this was done by REO’s own forester, Dermot Mangan, who
spent many long winter weeks in the snowbound Finnish forests travelling
by train, bus, taxi and sleigh, inspecting poles which were sometimes
buried in snow and at other times ‘had as much as an inch of melted
snow on them in the form of solid ice which made the measuring and
calipering no joke’. Eventually, as the quantities to be inspected grew
rapidly, a native Finnish forester, Viljo Rantala, was given training in
Ireland on precise REO requirements and returned as REO pole inspector

78
Material Matters

to Finland. This arrangement proved very satisfactory: there were even


some complaints from his fellow countrymen that he was too rigid. Paddy
McDonald, who was responsible for all material purchases, averred that
‘as a result, we practically never got a bad pole delivered out of a total
of one million’.
The first cargo of Finnish poles left the port of Hamina for Dublin at
the beginning of July 1946 on the decks of the $.S. Ashbel Hubbard.
This was followed immediately by 2,000 poles as deck cargo on the Irish
Larch and in the same month the S.S. Wicklow Head was reported as on
her way to Hamina to lift 13,000 poles for delivery to Cork. From then
on and for the next thirty years there was virtually never a time during
the ‘open water’ (ice-free period from May to December) in which
pole-carrying ships for rural electrification were not either loading in
Finnish ports, making the long voyage from the Baltic to Irish waters or
unloading at the three REO port depots of Dublin, Cork or Limerick.

Creosoting and Storage U The purchase of sufficient poles in bulk was


of course only the first link in the chain that terminated in serried ranks
of rural electrification poles marching across the Irish countryside. Bulk
storage and treatment depots had to be set up in Ireland and these were
established in Dublin, Cork and Limerick. The locations were selected
for their port reception facilities and for their strategic location with
regard to the distribution of the poles to the rural areas. In Dublin a
large area was rented from the Port and Docks Board at East Wall Road
convenient to the creosoting plant of Messrs T & C Martin. In Cork the
reception and storage area was on the river adjacent to the Ford Plant
and again convenient to an existing creosoting plant operated by Messrs
Eustace & Co. Creosoting as a method of treating raw poles for pres-
ervation was used practically exclusively. From time to time other
preservation methods were tried, such as various types of salts, but none
was adopted for general use. By 1947 it was necessary to open a third
depot at Limerick port and a new creosoting plant was put into operation
by Messrs Eustace through the Limerick firm of Spaights. Through these
three depots and treatment plants passed all the poles for the scheme
and apart from some problems in the immediate post-war years when
supplies of creosoting oil dried up periodically, there was never any
serious delay in the supplies of creosoted poles to the areas.
Delivery of creosoted poles to most areas was affected by a fleet of
tractors and specially constructed pole. trailers, but in the case of areas
in County Donegal and parts of the west coast small coasters were
chartered to transport the complete package of materials required.

79
The Quiet Revolution

CONDUCTOR — ‘THE WIRES’

The original design for the rural networds was mainly for copper
conductor and by October 1945 British firms were once again open to
orders for this. It quickly became evident, however, that the rapidly
rising price of copper would seriously escalate construction costs and
alternatives were urgently examined. These included aluminium, alu-
minium alloy, steel-cored aluminium and cadmium copper. Early in 1947
the very important decision was made to use steel-cored aluminium (sca)
almost universally, with all-aluminium conductor (i.e. without the steel
core) for some of the low voltage networks where, because of short
spans, the mechanical tensions were low. In the case of small, lightly
loaded, high-voltage spur lines, it was intended to use galvanised steel
stranded conductor, but the acute post-war steel shortage made this very
difficult to obtain. Light gauge sca was substituted with satisfactory
results.
The first orders for the aluminium conductor were placed with the
Aluminium Union of Canada and deliveries commenced in 1948.
Because of the acute dollar shortage there was some difficulty in getting
clearance to place the orders with the Canadian company, and in his
regular monthly report to the Board in January 1948 the Engineer-in-
Charge noted that ‘the first consignment of scA from Canada is at sea.
Permission to place the second order with Canadian suppliers amounting
to $160,000 [Canadian] was only obtained from the Department of
Finance at the latest moment to hold our position on the supplier’s order
book.’

The Aberdare Electric Co. Ltd. O Alternative sources of supply were


located in Britain, Belgium and Germany, but at this stage an Irish
supplier had entered the scene, the Aberdare Electric Company, later
to develop into the Irish industrial giant of Unidare Ltd. When, in 1944,
the government’s intentions regarding rural electrification were
announced, the possibilities of providing some of the equipment by an
Irish company were investigated by a number of Irish businessmen who
founded a consortium for the purpose. The Aberdare Electric Company
was launched in November 1947 with an issued capital of £210,000.”
The initial technical know-how regarding the manufacture of overhead
electrical conductor and transformers was provided by bringing on to
the new Board two members of the Board of Aberdare Cables Ltd. of
South Wales — hence the name of the Company. By the mid-1950s, the
activities of the Company had expanded into many fields outside the
electrical industry. For this reason and to avoid confusion with the Welsh

80
Material Matters

company and to emphasise its independence, the title of the Irish


company was changed to Unidare Ltd.’ It has long diversified from its
initial concentration on electrical equipment, but it freely acknowledges
that its springboard was the considerable initial market for its products
provided by the Rural Electrification Scheme, particularly transformers
and overhead conductors. Later came the manufacture of covered con-
ductor and underground cables, storage heaters, water pumps and piping
and a large range of accessories for line construction all developed in
very close liaison with the EsB.
The original machinery for the manufacture of conductors was
designed for copper so that when the price rises caused a switch to
steel-cored aluminium conductor there were problems in converting the
plant. A very big problem was the welding together of lengths of high
tensile steel core for the sca conductor. This was overcome and in
August 1949 the company received an order for half the year’s require-
ments of sca for the scheme. In addition, the EsB gave the company a
further three months to negotiate for further supplies of aluminium bars
in order to be in a position to quote for the balance.
The devaluation of sterling in 1949 had a serious effect on the price of
raw materials for the conductor manufacture. Copper increased in price
from £107 per ton to £140. With regard to aluminium the monthly Rural
Electrification Report for September 1949 notes: “The British Alumin-
ium Metal Price has increased from £93 per ton to £112. Fortunately,
the Aberdare Company had purchased the aluminium for our last order
before devaluation so that we are covered on conductor for about nine
months ahead.’

TRANSFORMERS

After poles and conductor, the third important item of material was
transformers. Owing to the distances involved and the dispersed nature
of the dwellings the most economic method of distribution was at 10,000
volts (1CkV) mostly single-phase. To reduce this voltage to the domestic
working voltage of 220 volts required transformers. Unlike the urban
situation where one large transformer —- 400kKVA or 640kVA
[kVA—Kilovolt-Amperes) for example — would serve several hundred
premises, the rural transformers were generally expected to serve groups
of from one to ten houses with four or five houses being the most usual
number. This necessitated a very large number (up to 100,000) trans-
formers of small size. The most common sizes in the initial stages were
24kVA, 3kVA, 5kVA, 15kVA and, for groups larger than normal,
33kVA.

81
The Quiet Revolution

American units were initially the cheapest available and considerable


orders were placed with Westinghouse in the U.S. As with the conductor,
however, the dollar shortage and the devaluation of sterling in 1949
made the American equipment expensive and difficult to obtain. Enquir-
ies in Great Britain resulted in quotations of almost double the American
price and there were some indications that a British ‘ring’ was in oper-
ation. The search was extended to Belgium, Sweden, Switzerland and
Czechoslovakia without very much success.
By the end of 1946 orders for 1,500 transformers had been placed with
American and British firms, and it was expected that about one half of
these would be delivered by the end of the following year. The difficulties
in obtaining the American transformers and the slow delivery and very
high price of alternative British and Continental supplies had raised the
question of manufacture in Ireland. One British manufacturer (Brush)
had already made overtures. By mid-1947 the prices of American trans-
formers had increased to almost the British level and the delivery position
was still difficult.
In May 1948 an order for five hundred 5KVA transformers was placed
with South Wales Switchgear, an associated company of Aberdare
Cables Ltd. On 7 November 1947 the Aberdare Electric Co. Ltd. had
been formed in Ireland to establish and develop a large-scale electrical
equipment manufacturing industry with two members of the Board
coming from the board of Aberdare Cables Ltd. By 1949 the company
was in a position to quote for transformers and in November of that year
the Ess placed their first order with it for three thousand SkVA trans-
formers, about twelve months requirements. Manufacture and deliveries
commenced immediately and on 22 December the first Irish-made trans-
former was erected in the Man-O-War rural area in north County
Dublin. It is well to record that in the development and testing of these
transformers there was a very close liaison between the Company and
the EsB thus ensuring high quality and close matching to EsB require-
ments. The amount of Irish manufacture (as distinct from assembly of
already manufactured parts) was very high; tanks were rolled and welded
from the sheet, shot-blasted and spray-painted; cores were assembled
from stampings; coils were wound on the premises. In fact, the trans-
former lid casting and the bushings were about the only manufactured
parts imported. The first batch of transformers was delivered in January
1950.

ACEC (Ireland) Ltd. O In pre-war years the Belgian Company, Atelier


de Constructions Electriques de Charleroi (acec), had done business

82
Material Matters

with the Ess. When Paddy McDonald visited Belgium and Holland in
October 1946 in search of materials to enable the scheme to commence,
he found that AcEc was just recovering from the disruption of the war
with most of its immediate output designated for urgent home
requirements.
Nevertheless, Jean Kottgen, chief of the export department, expressed
an acute interest in the long-term market for transformers and electric
motors which the rural electrification scheme was opening up in Ireland.
The upshot of the meeting was a more thorough investigation of the Irish
market by the company which noted particularly the Government stip-
ulation that as far as possible, Irish materials should be used on the
scheme. This resulted in the establishment in 1951 of acgc (Ireland)
Ltd., a wholly owned subsidiary of Acsc, S.A., of Belgium, with a factory
at Tycor in Waterford City, to manufacture transformers both for the
rural electrification scheme and for general distribution requirements.
Like the Aberdare Company, established four years earlier, Acec had a
company policy to manufacture to the maximum extent rather than just
to assemble. In fact the only finished bought in components were
bushings and tap changing mechanisms.
This policy ensured not alone in high level of local employment but
laid the foundations for subsequent expansion into a much larger and
wider range of products. Initially, concentration was on 3kVA single-
phase transformers for the rural scheme and the first of these was erected
in St John’s rural area in County Kilkenny in July 1952. The range of
transformers manufactured was extended gradually over the years to
38k V/10kV and then to the more recent 110k V/38kV giants of 31.5MVA
(31,500kVA) capacity. Again as with Unidare, the very close liaison and
co-operation established in the early days of the rural scheme has been
maintained to the benefit of both organisations and of the Irish electrical
industry in general.
For a period, the Waterford factory manufactured electric motors of
up to 30hp in three-phase and up to 3hp in single phase. These latter
were designed specially for rural conditions and many thousands were
used to provide motive power for pumps, grain grinders and milking
machines in rural areas. However, with the lifting of trade barriers,
competition from large-volume foreign manufacturers resulted in the
Irish motors being priced off the market and the main product apart
from transformers now coming from the Waterford company is electrical
flourescent fittings. These are produced in a modern plant in the factory
complex and account for a substantial part of turnover. In 1982 the
company employed three hundred people.

83
The Quiet Revolution

OTHER MATERIALS

Apart from the three major items — poles, conductor and transformers
— there was a host of other items, the absence of any one of which could
hold up work. These included ironwork of various kinds, struts, ties for
headgear, insulator pins, suspension clamps, shackles, shackle straps,
earth rods, stay rods, and a myriad of nuts and bolts of various sizes,
connectors of various kinds, dead-end thimbles, insulators for H T (high
tension) and L T (low tension) fuses, air break switches.
Immediately after the war, manufacturers, especially those in Europe,
had to face a major task of rebuilding or re-adapting their plants for
peace-time work and of locating supplies of raw materials, very often
from totally new sources. Life so many other industries the electrical
industry had to undergo a difficult period of re-adjustment which took
several years. Even ten years after the war there were recurring shortages
of various items of material, which sometimes caused serious hold-ups
in construction work. A compensating feature was the large stock of
war-surplus material released on to the market. While this had been
produced for the war effort, it was possible, by intelligent adaptation, to
use much of it for peace-time purposes. A good example on the rural
scheme was the use of British Admiralty surplus minesweeping hawser
as a Staywire. Owing to its hardness the multi-strand hawser could not
be spliced in the conventional manner. An alternative technique was
evolved using wire rope clips which permitted the use of this immensely
strong but difficult-to-handle material as a substitute for stay wire. About
eight hundred miles of hawser were used in this manner before supplies
of conventional stay wire were resumed. W. F. Roe recalled later that
just as delivery had been completed, an enquiry came from the British
Admiralty to ask if half the consignment could be returned. Obviously
its adaptability for uses other than minesweeping had been discovered.
Needless to say, he was not disposed to meet the request.
The procurement of sufficient materials to allow a start to be made
necessitated numerous visits to manufacturers in Great Britain and on
the Continent as well as to the many war-surplus depots which were
springing up as the huge wartime installations were decommissioned.
There were some initial difficulties for a country which had been neutral
in the conflict. In Britain itself some degree of reserve and even hostility
was encountered. Ireland’s neutrality had been criticised and misrepre-
sented by Winston Churchill in his widely disseminated victory speech
to the nation. De Valera’s reply had got little publicity in Britain and
Paddy McDonald, the man responsible for procuring the materials,
recalled that in his early meetings with suppliers much time was taken

84
‘Made in Waterford’ — The first batch of 250 3kVA rural electrifi-
cation transformers manufactured by AcEc awaiting collection June
>)

1952.

85
The Quiet Revolution

up with remedying this omission. Subsequently, excellent relations were


developed with many key British suppliers who, being first and foremost
keen businessmen, looking to the future rather than the past, recognised
the large potential export market provided by a scheme of such magni-
tude and made special efforts to meet REO’s requirements.
On the Continent the search was no less intense. On one trip alone,
in late 1946, McDonald visited no less than thirty-six Dutch and Belgian
firms seeking supplies of copper and steel conductor, galvanised steel
headgear and accessories, insulators, insulated cable, transformers and
switchgear. He had limited success, one firm hesitating to quote when
asked if it would accept sterling in payment. His assessment was that the
supply position would not ease out until 1948. However, through per-
sistence and adaptation, sufficient materials were procured to allow the
scheme to get under way as planned.
Gradually, as more Irish firms started manufacturing the proportion
of materials obtained from Irish suppliers increased. Indeed in some
cases the spread of rural electrification itself enabled Irish-made products
to be offered. A shortage of steel for crossarms — used by the tens of
thousands — necessitated the substitution of native oak crossarms, but
these were obtained only with difficulty until the development of the
Dundrum rural area in County Tipperary, early in 1951. The Department
of Lands installed a sawmill powered from the new rural electricity
network in the State forest which had good stands of suitable oak. Ample
supplies of oak crossarms of good quality were soon on their way.
It was the conscious policy of the REO to purchase as much as possible
of its material from Irish suppliers and to encourage in every way possible
the setting up of new Irish industries to meet its needs. By 1957 the
Board was able to report that for the year 1956/57, out of £2,090,000
spent on rural electrification materials, £1,442,000, or 69% had been
purchased from Irish suppliers. By 1961 this had risen to 82% and by
1965 to 84%.

86
CHAPTER EIGHT

Shipping the Poles

THE BALTIC TRADE

THE SHIPPING OF ONE MILLION POLES was in itself a huge operation.


At the outset, the Rural Electrification Office staff knew very little about
the niceties of ship chartering. The necessity to have huge quantities of
poles shipped promptly and cheaply provided the incentive to learn
quickly. Shipping from the Baltic was a particular problem, as from early
December to about the middle of May the ports were icebound. Shipping
therefore had to be compressed to a seven-month season, starting with
‘first open water’, which generally occurred during the first or second
week in May.
Poles are awkward to stow and not every ship is suitably constructed
to carry them, particularly under decks. In earlier days, when the
amounts were comparatively small, poles for the ESB were generally
shipped as deck cargo after other cargo, usually sawn timber, had been
stowed in the holds. Another difficulty was that poles bulk much more
than an equal weight of sawn timber. The normal shipping unit for
timber in the Baltic trade is the ‘standard’, (165 cubic feet of sawn
timber). In the case of poles, however, one can only get about two-thirds
of this amount of actual timber into the same volume, so that only ten
average-sized poles, totalling about 110 cubic feet, go to make up a
standard. As time went on and the EsB’s requirements became known in
Baltic shipping circles, ships specially suited to the pole-carrying business
were offered. As these could carry larger quantities both in the holds
and on deck they could be chartered at more reasonable rates.
In the initial stages the easiest terms to obtain were ‘liner’ terms,
whereby a ship on a scheduled run between ports would carry a quantity
of poles — usually on deck — as part of a general assortment of cargo. For
small quantities this was quite suitable and the most economic method.
The first shipment of poles from Finland for the Rural Electrification

87
The Quiet Revolution

Scheme was carried on this basis by the S.S. Ashbel Hubbard in July
1946. Even in 1947 when the first big consignment of poles (116,000)
was for shipment, the major portion (70,000) was contracted on liner
terms with the Head Bell Group (30,000) and Irish Shipping (40,000).
The remainder was divided on a time charter basis between seven Finnish
and Swedish ships, which between them contracted for thirteen voyages.
As Finland is a land of lakes, water was used to a great extent for the
transport of poles from the forests down to the ports. On Lake Saimaa,
north of the port of Hamina and close to the Russian border, large rafts
made up of thousands of poles were regularly towed up to eighty miles
on their way to the port. These usually started off as small rafts in the
upper reaches of the lake, being added to as they progressed down the
lake and finally finishing up with many thousands of poles. The final
count of one of these rafts, destined for REO was 5,401 poles with a total
volume of 62,125 cubic feet of timber. This was regarded locally as a
world record.
Even with chartered shipping, all was not plain sailing. Delays were
inevitable, owing to the uncertainty of loading and discharging times and
time at sea. Further delays were caused by disputes at the loading ports,
mostly regarding the time allowed for loading and the facilities offered,
as some of these ports were tiny and very short of proper harbour
installations. These delays in turn caused problems at the Irish end, as
uncertainty of arrival dates made it very difficult to organise berthing
and discharge facilities.
Once berthed, there was every incentive to unload and turn-around
the ships as quickly as possible. The chartering agreements (‘charter
parties’) included stipulations as to the time allowed for unloading and
there was a heavy financial penalty, demurrage, payable for delaying
the vessel’s scheduled departure. Dublin port in the immediate post-war
years was experiencing a boom. Container traffic was only in the early
stages of development and the quaysides were congested with merchan-
dise of every type awaiting customs clearance and collection. In order to
speed up unioading operations a system was devised for directly dis-
charging the REO poles on to transport vehicles which operated a shuttle
service to the EsB’s own pole-stacking areas which were well away from
the congested quaysides. This required a high degree of organisation and
synchronisation but it was so successful in achieving a high rate of
discharge that it was adopted as standard at all three port depots.
Inevitably, ships sometimes arrived at congested ports and were unable
to unload immediately, thereby putting REO in danger of incurring
demurrage charges. Usually, however, extra effort on the part of the
dockers recovered the position and over the total shipping campaign

88
Shipping the Poles

involving hundreds of voyages, demurrage payment was almost


unknown. On the other hand, sizeable refunds were frequently obtained
as the dockers, time and time again, discharged the ships well under the
time allowed for in the charter party. In the case of one large ship, the
Maria Lisa Nurminen, the standard discharge time allowed in the charter
was eight working days. On two occasions the cargo of over 10,000 poles
was discharged in five days and on a third occasion 10,460 poles were
unloaded and stacked in four days. It is no surprise to find that the
captain conveyed to the owners his preference for trading with Irish
ports. Sometimes the speed of discharge caused a little embarrassment,
as when the Master of the S.S. Airisto, discharging 11,326 poles at
Limerick, had to alter plans to fit in both a trip to Killarney and the
Dublin Horse Show during the allowed eight-day discharging period: he
had to sacrifice Killarney and rush back to take his ship out three days
earlier then he had anticipated. Limerick became a very popular port
with the owners of that vessel.
In August 1947 while one ship was discharging its poles in Dublin a
second pole ship arrived with no berthage space available — and none
likely to become available for several days. With heavy demurrage
charges looming, the problem was considered by the reo staff. One
suggestion was to drop the poles over the side of the vessel into Alexandra
Basin and construct a raft of the poles similar to those on the Finnish
lakes. Information on these was available as, on one of his visits, Neil
O’Donoghue had made meticulous sketches showing the construction of
such rafts. The operation, however, was one hitherto outside Irish
experience. The Ess often, of necessity, recruited young engineers fresh
from college and threw them quickly in at the deep end. Here, at the
very beginning of operations, the opportunity occurred to do this almost
literally. The phone rang at headquarters. A young engineer was in the
front hall reporting for duty. The reaction of the Engineer-in-Charge
was immediate. He later recounted the incident: ‘I dispatched my deputy,
P. J. Dowling, to the front hall with explicit instructions. ‘““Can you
swim?’’, ‘‘Yes, fairly well’, “Right! Go down immediately to Alexandra
Basin and take charge of the construction of the pole raft.” He com-
plained to me later, “On my first day on rural electrification, I didn’t
even get past the front hall of the Ess. Instead I found myself hanging on
grimly to the side of a ship contemplating the possibility of a watery
grave as poles in their hundreds whistled past my head, to land with a
smack in the murky waters of the Basin’’’. In the event the operation
was successful. A large raft of some thousands of poles, based on Neil
O’Donoghue’s sketches, was built in the best Scandinavian fashion by
ESB pole-field staff. The methods used for slinging, tying and towing

89
The Quiet Revolution

were all taken from O’Donoghue’s notebook. The completed raft was
towed down the Liffey by the Port and Docks tugs Norway and Poolbeg,
brought around with some little difficulty into the estuary of the River
Tolka and moored at the back of the pole-field, whence transfer to the
stacking area was comparatively easy. The only really anxious moment
was when it was necessary to ‘put about’ almost 180 degrees in order to
get up the Tolka. An ebbing tide and a westerly wind conspired to send
the raft across the Irish Sea in the direction of Liverpool. However, the
planning had been well-done. The tide changed and a combination of
flowing tide and herculean efforts on the part of the tugs persuaded the
raft to change course and enter the calm waters of the Tolka.
By the middle of November 1947, of the 114,000 Scandinavian poles
purchased, 95,000 had been delivered. Two ships had been chartered to
lift the remaining 19,000 poles before the ice set in. Cables received on
20 and 24 November advised that both ships had ‘touched ground’ on
their way to the loading ports. A later advice was that one, the Scand-
navic, had broken up, fortunately without any loss of life. The 19,000
poles had consequently to be overwintered in Finland.
It was a tragic coincidence that on the very day in August 1947 that his
sketches were being used to construct the EsB pole raft in Alexandra
Basin, Neil O’Donoghue should have met a sudden and untimely death.
He was followed in the job by Peter Conroy, an accountant who brought
to the business of chartering a keen bargaining ability and an intense
interest in all things nautical. Conroy was responsible for the shipment
of the greater part of all the poles used on rural electrification — over
one million. He revelled in the cut and thrust of offer and counter offer.
He rapidly built up a first rate intelligence on the availability and going
rate of the different ships in the pole trade and became recognised by
shipowners in the Baltic trade as a client who was difficult to hoodwink.
He was a good storyteller and many of the incidents quoted here are
taken from his contributions to REO News and the ESB Journal.
One vessel deserves special mention, the venerable S.S. Satakunta,
which year after year carried poles from Finnish ports to the Irish pole
depots and became the best known ship of the pole fleet. A vessel of
2,207 gross tons, she was built in 1898 by W. Gray & Co. Ltd. at West
Hartlepool and christened Everest. In 1939 she was sold to the Satak-
unman Laiva O/Y and registered under her new name at Pori, Finland.
Her dimensions made her very suitable as a pole carrier. To quote from
a letter to the editor of the Evening Mail on 13 October 1950 — ‘....she
is, after over half a century of seagoing service, a great credit to the
workmanship of the men who built her’.

90
ome ee

Above: The heavily laden MV Make had two inches to spare in crossing the bar to
Limerick docks in July 1959. Below: Erecting the first pole at Kilsallaghan, Co.
Dublin. On the extreme right is P. J. Dowling and next to him is W. F. Roe.

ad

OT
The Quiet Revolution

As more and larger vessels suitable for pole transport were located,
and as the masters got more experience of the Irish ports and discharging
performance, the amounts carried per ship increased, thereby keeping
costs down.
Sometimes it was touch and go whether the large, heavily loaded
vessels would clear the harbour bar. In August 1953, the Airisto was due
in Limerick at the time of the August bank holiday. She was no stranger
to Irish ports, having discharged 10,500 poles in Cork a couple of months
earlier. This time she was carrying 11,326 poles and was fully down on
her marks. Her estimated time of arrival would, however, coincide with
neap-tide and it was feared that her draught would exceed the require-
ments of Limerick Port for this condition. The shipowners were alerted
to the tidal conditions predicted, and it was suggested that if there were
any doubts the vessel could be diverted to Cork. However her Master,
Captain Ingerttila sent a message from Dover that the ship was now
sailing on a lesser draught than on the earlier part of the voyage and that
he would pick up the Limerick pilot at the mouth of the Shannon. The
good weather would also be a help.
Arriving at Kerry Head the ship was on an even keel. Ballast tanks
had been pumped, bunkers were at a minimum, and some of the crew
even jested that drinking water was scarce. The vessel was now drawing
16’6” — almost precisely the minimum clearance required by the har-
bour authorities for the expected tide. The pilot was taken aboard for
the forty-mile trip up the estuary. The calm conditions held and at 6pm
on August Monday the Airisto was safely berthed at Limerick Docks.
Despite the absence of shoreside cranes (which necessitated the use of
the ship’s own winches), the ship was discharged in five days. Captain
Ingerttila returned hurriedly from his planned tour of the southwest. His
ship was ready to take out on Saturday afternoon instead of the following
Tuesday as he had expected.
Even this virtuoso performance was overshadowed when, in July 1959,
news came through that the M.V. Make had scraped into Limerick
Docks. The verb describing the vessel’s arrival appears to have been well
chosen, for, with a record cargo of 13,000 poles, the vessel was drawing
18’ of water. The clearance at high tide on that night went the vessel
arrived was predicted as 18'2". Normally a vessel can gain some extra
clearance by emptying the ballast tanks. In the case of the Maké however,
with over 50% of the cargo above decks, thé emptying of the tanks
would cause stability problems. Peter Conroy recalled that his main task
on that occasion was to calm the fears of the harbourmaster until the
vessel had finally come to rest at the quayside. The unloading of the
13,000 poles was completed in five days, which made some amends for

wpe
Shipping the Poles

the strain on the harbourmaster’s nerves. Limerick Port was experiencing


its busiest spell for over three years and berths were at a premium.
In September 1954 the S.S. Karen carrying 6,800 poles from Hamina
to Limerick encountered hurricane conditions going through the Ska-
gerrack. 2,400 of her poles were deck cargo. The vessel, a Danish ship,
had been chartered for three voyages. On this third trip the trouble
started as the ship left the Baltic, heading north through her home waters
of the Kattegat. A few extracts from her log give some idea of the
hazards of the sea, which are such an everyday reality to the mariner
and so readily ignored or forgotten by the rest of us.

September 15 Left Copenhagen (after bunkering). Draught 14’8” fwd.


and 15’6” aft. Fresh S.W. wind and passed the Skaw at midnight.
Sept. 16 Similar weather, heavy seas. Steaming only 5 knots and shipping
heavy water over decks. Crew overhauling deck lashings. Wind increas-
ing to gale force. Had to heave ship to.
Sept. 17 Wind WSW hurricane strength. Vessel hove to, shipping heavy
water over cargo. At 6 am shipped heavy sea shifting deck cargo and
some poles washed overboard. Deck lashings renewed. At 8 am a heavy
sea caused further shifting of deck cargo. Vessel took heavy list to port
and went athwart the sea losing steering way. A further heavy sea caused
additional shifting of deck cargo and some more poles washed overboard.
Port tank pumped out to reduce list and vessel put about. Proceeded to
Fredrickshaven and moored there.

Over the next six days the deck cargo was unloaded, repairs carried
out and deck cargo reloaded. The vessel bunkered and left Fredricks-
haven on 23 September. Her troubles, however, were not yet over.

Sept. 24, 25, 26, 27 and 28 Fresh S.W. wind, strengthening S.W. wind.
Gale force. Hove to. Vessel labouring heavily.
Sept. 29 Whole gale. Decks full. Vessel labouring heavily and decided to
proceed under Duncansby Head for shelter.
Sept. 30 Weather improving, wind veered to east. Proceeded on passage
and passed through Pentland Skerries.
Oct. 4 Arrived Shannon Estuary 6 am. Anchored awaiting pilot. Docked
Limerick 5.30 pm.

93
The Quiet Revolution

The ship had taken twenty-five days to make the journey and had run
the gauntlet of two storms of hurricane and whole gale force. It delivered
6,600 poles, enough for a further six rural areas. It is a fair assumption
that the crews that erected these poles or the consumers who eventually
enjoyed the electricity were completely unaware of the hardships and
dangers undergone by those who transported them from the Baltic.

THE DONEGAL COASTERS

While large ships were busy carrying poles in bulk from the Baltic to
the three Irish port depots, small coasters were at the same time trans-
porting the complete package of materials for a number of rural areas
from the same depots to a multitude of tiny ports, piers and jetties on
the western and north-western seaboard, in most cases right up to the
doorsteps of the areas to be serviced. Not that the coasters were left out
of the Baltic runs. While their capacity was small — generally varying
from about three thousand poles for the larger coasters down to one
thousand or even five hundred on the small vessels — this in many cases
was an advantage. It enabled them to slip into the smaller ports and lift
small parcels of poles which would not warrant a larger vessel. Even
when they had to make three or four voyages as against one voyage of
a larger vessel, they were still competitive. One Dutch shipping company,
Gruno Shipping, which operated a very large fleet of small coasters
carried the bulk of the EsB pole shipments from the Baltic to Dublin,
Cork and Limerick in the 1953 shipping season using two small ships,
M.V. Martien and M.V. Thérése.
In 1954, when the final cargo of the season had arrived in mid-
December, twenty-four trips of about nine days in each direction had
been made to lift 99,000 poles at five Finnish and two Swedish loading
ports as far north as Toppila in the Gulf of Bothnia and as far east as
Hamina on the Gulf of Finland, about twenty kilometres from the
Russian border. There was one cargo of 10,460 poles and three of over
6,800 but the balance was carried by the small ships, averaging under
3,000 poles per trip.
Again in 1956 the Invotra fleet of coasters, also Dutch, shipped 60,000
poles — 40% of that year’s quota — to the Irish ports. In that year a record
total of 152,000 poles valued at over £700,000 was shipped from the
Baltic; 60,500 to Dublin, 56,000 to Limerick and 35,000 to Cork.
The versatility of small coasters was further demonstrated when in the
early summer of 1954 it was necessary to purchase six thousand creosoted
poles from Sweden in order to keep up supplies of the more popular
sizes until the creosoting of the new raw poles commenced in October.

94
Above: A skinning crew in Cork
— they skinned the bark from
the imported poles prior to creo-
soting. Left: Delivery of poles by
coaster. Below: One _ horse
power!

2D
The Quiet Revolution

The first cargo of three thousand poles was loaded at Gothenburg for
Dublin, both of which are large well-equipped ports. The second cargo,
however, which was also of 3,000 poles was loaded on the Dutch coaster
M.V. Muphrid N at the Swedish inland port of Otterbacken situated on
Lake Vanern which meant the vessel had to be small enough to traverse
the Gotha Canal. This cargo was brought direct to the ports of Galway
and Westport, an achievement that would have been impossible with
the larger ships.

The Electrification of County Donegal O The story of the coasters is,


however, so bound up with the story of the electrification of County
Donegal that it would not be possible to recount one without the other.
Situated as it is in the north-western corner of Ireland, bounded to the
west and north by the Atlantic Ocean and to the east and south east by
the border with Northern Ireland, Donegal’s only direct land connection
with the rest of the Republic is through a very narrow throat at Bally-
shannon in the extreme south of the county. All traffic from the Republic
not passing through Northern Ireland must cross the River Erne at
Ballyshannon bridge, then go north on the one road to Donegal Town.
Here the road divides, one branch heading north towards Letterkenny,
the other west towards Killybegs, Ardara and Glenties. Recently, with
the help of EEc funds, the road has been considerably improved but in
the late forties and early fifties it was narrow and twisting, most unsuit-
able for heavy commercial vehicles.
The physical isolation of the county from the rest of the State and the
poor communications caused it to fall very much behind in infrastructural
development. In so far as connection with the national electricity grid
was concerned, Donegal had perforce been left behind in the large
expansion of the thirties, when practically every town of any size in the
rest of the State had got EsB supply. At the end of World War II, the
only towns in County Donegal serviced by the EsB were Letterkenny,
Lifford and Convoy, all fed from the Northern Ireland grid by means of
a cross-border link at Strabane. The total number of electricity con-
sumers so supplied was 750. All other towns and villages, some of
considerable importance such as Ballyshannon, Bundoran, Buncrana,
Donegal Town, Ballybofey, Stranorlar and Killybegs, were dependent
on local electricity undertakings which, while mostly adequate to supply
light domestic requirements, could not cater for the large motive power
loads demanded by industry. The economic development of the county
was therefore being held back, not alone by its geographical isolation
and poor communications, but by an inadequate public electricity supply.

96
Shipping the Poles

The difficulty was the remoteness of County Donegal from the existing
grid. Sligo Town, 41 miles south of Donegal Town, was the nearest point
of EsB supply, but at that time even supply to Sligo itself was not very
secure, dependent as it was on a single 38kV ‘tail’ fed from the main grid
station at Carrick-on-Shannon, 34 miles further away to the south-east.
Without major and prohibitively expensive development the extension
of EsB supply into most of Donegal did not appear feasible.
The breakthrough came with approval by the government in April
1945 of the hydro-electric development of the Lower River Erne with
generating stations at Ballyshannon and Cliff which meant that a large
and secure source of electricity would shortly be available within the
county. The 38kV line was extended from Sligo to Ballyshannon as an
advance link to service the construction sites while plans were developed
to run heavier transmission lines to feeding points on the national grid.
Immediate advantage was taken of the connection with Sligo to change
the town of Ballyshannon over to EsB supply and by March 1946 this had
been completed. Supply was extended to the adjoining seaside resort of
Bundoran in 1947. The government had given a directive that in order
to get the Rural Electrification Scheme off the ground in as widespread
a manner as possible, one rural area should be commenced in each
county to start with. The arrival of the EsB supply at Ballyshannon made
it feasible to comply with this directive. Rossnowlagh rural area adjacent
to the town was commenced in August 1948.
With the completion in 1950 of the 123 miles of 110kV line from
Ballyshannon to Dublin, Donegal’s link with the national grid was
secure. From Ballyshannon northwards, firstly to Donegal Town and
thence through the Barnesmore Gap to Letterkenny, a 38kV trunk line
was extended capable of meeting all immediate demands. By 1950,
therefore, all was ready to commence the intensive electrification of the
county.
In September 1950 work began on the area stretching west from
Donegal Town to Mountcharles, thence on to Dunkineely, Killybegs
and Kilcar. The north-east saw the crews arriving in the Raphoe area in
September 1951. By 1955 four separate crews were busily erecting rural
networks in different parts of the county and by the end of 1956 supply
had been extended to twenty-eight rural areas with about twelve thou-
sand new electricity consumers. By 1960 the 38kV line had been run in
a complete loop around the county, increasing the security of supply
immensely, forty-five areas had been completed and over twenty thou-
sand new consumers connected.
In the very early stages of the development the weakness of the road
communication between Donegal and the rest of the State was a problem.

oT
The Quiet Revolution

The delivery of construction materials by road from the three main port
depots of Dublin, Cork and Limerick was difficult enough even in the
case of the rest of the country, hampered as REO was by the still serious
shortage of heavy-duty road transport. The available fleets were very
hard-pressed to meet the ever-increasing construction rates being
attained in the field. In the case of Donegal, the position was near-
critical. The road distances involved were immense. To Dungloe for
example, the road mileage from the nearest pole depot, Dublin, was 213
miles, while from Limerick it was 220 miles. Nor was rail an alternative.
Many of the Donegal areas then being selected for development had no
rail facilities or only narrow-gauge lines, some of which were then in the
process of being dismantled. There was also the problem of cross-border
transport, always at prohibitive rates owing to the double and treble
handling involved.
The experience gained in the use of small coasters in the Baltic trade
now came into play. The possibility of using these to transport the large
quantities of material required to areas on the west coast, particularly in
Donegal, was examined. For a county with such a long coastline — four
hundred miles, extending from the shores of Lough Foyle to the coast of
County Leitrim — Donegal is not endowed with many large ports. It has,
however, a large number of small harbours, jetties and piers, some of
which were built as Famine Relief works or by the old Congested
Districts Board, many of which had not seen even the smallest cargo
boat since the CD Board had ceased its activities earlier in this century.
There were difficulties in getting in and out and in most cases a vessel of
any size would have to rest on her bottom at low tide.
There were, however, great advantages in cost and speed in the
complete furnishing of an area with materials in one fell swoop, as
against slow piecemeal delivery over the very poor roads of the time.
REO was lucky in being in contact with coastal shipping companies which
had wide experience of slipping in and out of tiny ports in all parts of
Europe. One of these was the Dutch Gruno Shipping Company, which
carried the bulk of the 1953 Baltic pole shipments. The story of the first
coastal cargo shipped by this company is worth recounting in some detail.
On Tuesday 6 October 1953, one of the company’s vessels, the M.V.
Marwit left Cork for the port of Ramelton in Lough Swilly with a cargo
described in the ship’s manifest as ‘Materials in boxes, crates, coils,
bundles and loose as per vouchers and 1,329 electric light poles’. The
cargo was more elegantly and romantically described by a scribe of the
time: “The angels smiled because they knew that the cargo of the Marwit
was nothing so prosaic. With that knowledge which is denied to men
they knew that the little boat was laden with Christmas presents, with

98
Shipping the Poles

presents of electric light and power and the comfort and prosperity that
goes with them for the good people of Ramelton and Rathmullen in
County Donegal.”!
So anxious were all concerned to transport the maximum amount, that
the vessel was right down on her marks, drawing 10’9” of water. Captain
Baarscheers, a veteran of countless forays into nominally inaccessible
ports, was only slightly worried by the knowledge that the maximum
clearance over the bar at Ramelton was officially advised as being only
10’ and that would be available only at high water of spring tide due at
6.00 pm on the following Thursday. He thus had two problems: firstly to
arrive at the harbour bar not later than high water, and secondly to nurse
his ship over the bar. Proceeding via the Irish Sea the vessel rounded
Malin Head into Lough Swilly in good time to pick up the pilot at Inch
Island. To quote from the contemporary report: ‘Pilot Brown held her
back to a steady two knots for the last three miles of her journey as the
spring tide rose beneath her. Precisely at 5.45 pm she crossed the bar
without incident and tied up at Ramelton quay where about 500 people
had assembled to welcome her.”
The unloading was carried out immediately and speedily using the
ship’s own winches operated by the ship’s engineer and first mate, neither
of whom spoke English, assisted by an unloading gang of local men who
knew no Dutch. However, with shouts and internationally understood
signs the unloading was completed speedily if somewhat noisily with just
one minor casualty as the 6’7” Ess chargehand Jack Newton was knocked
into the water by a sling of poles to the loud cheers of the audience on
the quayside. As the local Garda Sergeant expressed it “There hasn’t
been such excitement in Ramelton since the Civil War!’
The next morning the Marwit again crossed the bar in the morning
mist, this time en route for her next contract in the Algerian sunshine.
A month later the M.V. Martha, also of the Gruno line, arrived in
Ramelton with a second load. These operations were somewhat unusual
in shipping circles as the ESB were suppliers of cargo, charterers, receiv-
ers, port agents, brokers, ships’ agents and stevedores, all at the same
time.
In the next few years practically every usable harbour, jetty and pier
on the Donegal seaboard and even a private, landing stage at the Capu-
chin monastery on the Ards Peninsula near Creeslough were used to
land cargoes. As most of the piers were prohibited landing places in
closed ports, except to fishermen, the permission of the Revenue
Commissioners and the approval of the Department of Industry and
Commerce had to be obtained in advance. For each shipment, the
cargoes were loaded mainly at one of the three port depots and occa-

99
The Quiet Revolution

sionally at Galway, Sligo, Ballina and Westport, where some overseas


consignments had been discharged. On one occasion a five-port, seven-
point operation was carried out by the M.V. Flevo, loading at Dublin,
discharging and reloading at Cork, discharging at Galway, Cleggan and
Newport, and returning to Cork to load a cargo for Dublin. It is sad to
record that this ship was lost with all hands in the North Sea a few years
later when it was not, however, on REO charter.
The coasters used were not all Dutch, of course. The Limerick Steam-
ship Company vessel M.V. Galtee was constantly engaged in coastal
deliveries. One voyage of this vessel worth recording took place in
October 1954. Actually there were two ships involved, the second being
another Irish ship, the M.V. George Emilie of Arklow. The George
Emilie loaded at Dublin for a two-port discharge, first at Bunbeg and
then at Letterkenny. The Galtee loaded at Cork for a three-port discharge
— Dublin, Letterkenny and Burtonport.
Letterkenny port at the time had only one berth and was at the head
of the narrow five-mile navigation of the Swilly River with its outlet into
the southernmost reach of Lough Swilly. The timing called for the
George Emilie to have discharged at Letterkenny and have cleared the
Swilly River in time to allow the Galtee to enter with her cargo. Despite
the most severe storms of the year, both ships kept to time and the
discharged George Emilie cleared the mouth of the river on schedule,
passing on its way down the Lough the loaded-down Galtee at its
moorings awaiting the rising tide. However, just as the Agent and the
EsB staff had collected the local pilot for the river and set out from
Letterkenny for the anchorage, puffs of smoke were seen in the distance
ascending the horizon in line with the ascending tide in the river. To
quote from the contemporary report:

As we drove along the bank of the river towards the Lough, a small
steamboat flying the Union Jack was observed in the mouth of the river,
the funnel spitting and spluttering steam and smoke in an effort to keep
the vessel afloat as there was not yet enough water in the river to keep the
heavily laden boat off bottom. The skipper had evidently decided to enter
the river while daylight lasted and inch his way up on the rising tide with
his cargo of coal. It was a small vessel and could make port on a draught
several feet less than the Galtee. As there is only one berth in Letterkenny,
a midnight conference was held aboard the Galtee after a perilous voyage
to the anchorage in a rowing boat. It was decided to miss Letterkenny and
sail for Ramelton where discharge was completed on to about 60 feet of
pier, the remaining quayside having fallen into the water sometime earlier
in the year. The Galtee flying the tricolour was the biggest ship ever to go
up the River Lennon from Lough Swilly to Ramelton. Discharge at the

100
Shipping the Poles

third port, Burtonport, was uneventful but on leaving, the Galtee ran into
a continuation of the severe gales and had to shelter in Killybegs for
several days.°

In contrast to these conditions were some balmy days in April 1955


when the Galtee again arrived in Lough Swilly bound for Ramelton with
a cargo of poles for the Portsalon area some fifteen miles to the north.
Because conditions were so perfect the captain decided to land the poles
on to Portsalon pier right in the heart of the rural area where they were
required. This pier is situated only five miles from the mouth of Lough
Swilly on the Fanad peninsula and, being subject to the prevalent
Atlantic swells, was hardly ever used. The weather held long enough to
enable full discharge,

much to the disappointment of the people of Ramelton, 15 miles down the


Lough who were awaiting the arrival of the ship as scheduled. However,
when the ship returned a week later, this time for a scheduled discharge
at Portsalon, heavy swells prevented this and the ship had to proceed to
Ramelton where the people ... turned out in force on Low Sunday evening
to give the Galtee a rousing welcome and ... an assurance to the Captain
that they were pressing the County Council to have the pier repairs
completed before his next visit.*

Not all shipping was to such sheltered waters as Lough Swilly. At


Teelin, Downings, Burtonport, Bunbeg and Malin Head, even at
Donegal Pier itself, the turbulent Atlantic frequently made itself felt.
When the island of Arranmore came up for development, delivery of
the required poles and other materials presented a problem as there was
no suitable pier for the discharge of even a small coaster and the prospect
of ferrying materials out piecemeal in the half-deckers that normally
serviced the island was a daunting one, both in regard to time and to
cost. Finally, ‘after much persuasion of the shipowners’ it was agreed to
beach the Dutch M.V. Whitsun loaded with five hundred poles, cable,
transformers and other equipment, including a delivery van and three
compressor-tractors, on a selected shallow-water section of the fore-
shore. Close to the beaching site were several small, partly covered stags
of rocks, very dangerous in stormy weather. The ship was beached on
the island on Sunday afternoon, and when the tide receded at 8.00 am
on Monday morning unloading began. Everything went well until the
evening of the third day when a north-easterly gale hit the area, raised
a heavy swell on the spring tide running at the time, dragged the anchor,
smashed the moorings and almost tossed the ship onto the island only

101
The Quiet Revolution

fifty yards off and then lifted it within a few feet of the rock. Only the
skill and heroism of the ship’s captain and the Arranmore lifeboat crew
saved the ship from disaster. At the height of the storm the captain’s
wife and child were taken off. The Derry Journal ran a leading article in
its issue of 11 April entitled ‘Hats off to them’.
Some captains were not as adventurous. At Portsalon the captain of
one Irish coaster refused to make a return trip when the locals told him
that a coaster of the size of his vessel had not been seen there for more
than thirty-six years. The narrow pool of tidal water which served as the
port of Letterkenny on the River Swilly was also refused by several
owners after a few exciting trips there with quite small boats. For the
record, navigation conditions have been improved immensely since then,
allowing much larger vessels to ply regularly to the port.
Away from the exposed Atlantic coast, in the relatively calm waters
of Lough Foyle, battle was joined between two vessels for the first
occupancy of a single berth pier. The story can best be recounted in the
words of Peter Conroy who as honorary Commodore of the REO coaster
fleet could not be regarded as a completely neutral observer:

In the earlier case, we had a race for a single berth port [Letterkenny]
between our chartered ship, the M.V. Galtee and the S.S. Mulcair carrying
coal out of a British port. We lost that race, perhaps because we didn’t
know all the tricks of the trade, but we learned something from the losing
and on Friday, 26th August, of this year [1955], we found ourselves once
again in competition. This time our candidate was the Dutch motor vessel
Port Talbot (350 tons) and the opposition was the S.S. Ronaic with 500
tons of coal from Maryport. The goal was the single berth pier at Moville,
known locally as the ‘Famine Pier’.
It was a hot, rather oppressive day that 26th August and large crowds of
holidaymakers and daytrippers were assembled at vantage points on the
western shores of Lough Foyle. The main attraction from their point of
view was the approaching exercise of the NATo forces due to take place in
the Lough. The local people went about their tasks in their usual calm
mood but the righteous strained their eyes towards the mist-enshrouded
Magilligan Point and Innishowen Head at the mouth of the Lough, where
the big battle craft would appear as soon as visibility permitted. A small
handful of people also watched anxiously towards the points, and the
question uppermost in their minds was would the Port Talbot make it
before Ronaic because our land forces were mobilized at Moville ready to
unload the poles and other materials for the construction of Muff Rural
Area.
Under the heat of the August sun the mist lifted, and round the headlands
came the first of the great ships. A stir of excitement ran through the
watching crowds as Portuguese, Netherlands and Norwegian naval units

102
Shipping the Poles

moved into the Lough to join the British in occupation there. The full
panoply of war was apparent: air and submarine escorts fussed hither and
thither, all anxious to show their paces and join in the mock battle further
down the Lough.
As the mist lifted further and the battle fleets stretched themselves out
after the narrow entry into the Lough, the tenseness among the small
handful watching the lesser struggle increased: for there surely, sandwiched
in among their bigger sisters, were the two coasters, dwarfed down by
comparison to the size of fishing smacks.
There was a general heave to as pilots came aboard, and then the watching
few noticed that the Port Talbot seemed to be doing a lot of apparently
aimless dodging about, but the more shrewd observers knew that this was
not aimless; this was tactics and the race would be to the tactician.
Seamanship was at a premium now, and the Dutch captains of coastal
vessels are masters of seamanship. The expert observers relaxed their
tension and even the landlubbers could see the pattern of manoeuvre.
Suddenly the Dutch vessel was away to a quick start and was well in the
navigation channel. It was almost like a good coursing greyhound getting
away quickly from the slips, and the first ‘turn’ and the initiative were with
the Port Talbot. The small group of spectators exchanged quiet grins: the
race was in hands, and in the cool of the August evening the Port Talbot
was at Moville Pier, and the Ronaic, ploughing through its stern wave,
must lie ‘off’.
The big battleships sailed on down the Lough and the greater battle was
joined but at Moville Pier the land crews were already discharging the
Port Talbot to the accompaniment of zooming aircraft, the fire of heavy
guns, and the brilliant light of target flares and markers of the NATO
exercise.
The captain of the Ronaic looked a little disappointed when he came
ashore from his anchored position off the pier. He had been beaten in the
race, and, of course, no one particularly likes that, especially when time
and money are involved. He seemed surprised that unloading had com-
menced at that late hour (it was now Spm) and would finish next day. Such
a thing, he said, could only happen in the ‘Free State’.
Late the following night when the Ronaio had moved into the vacated
berth and the Port Talbot was ready to sail, the two captains forgathered
with our officials in the local hotel and the story was told of the earlier race
we had lost. The toast was ‘revenge for the Swilly’.°

While Donegal tended to occupy the centre of the coastal shipping


stage, down along the west coast the little ships sought out the small
harbours bringing with them a new way of life in the form of poles,
transformers and drums of cable and indeed wakening up many of these
ports from a long slumber. The list is almost endless: Malin Head,
Moville, Buncrana, Culdaff, Letterkenny, Ramelton, Portsalon, Down-

103
The Quiet Revolution

ings, Ards, Bunbeg, Burtonport, Teelin and Donegal Pier. Further down
the coast was the port of Sligo and, in County Mayo, Westport, Newport,
Ballina and Belmullet. In Belmullet the M.V. Whitsun arrived with poles
for Geesala rural area and provided considerable excitement for the
community, particularly the children who were seeing the first ever ‘big’
ship in their port. In the current progress report for the area it was
solemnly recorded that ‘during the discharge, the captain received
numerous local businessmen and Church dignitaries aboard the vessel.’
Cleggan in County Galway, Portmagee and Valentia Island in County
Kerry, Castletownbere and Beare Island in Bantry Bay also saw the
coasters delivering not only the materials but the machinery and trans-
port required to carry out the works. The M.V. Whitsun had to be
partially beached alongside the stone jetty at Beare Island to unload its
cargo which included three tractors.
All in all, the coastwise shipping operation was a success. It made it
possible to construct electricity networks even in remote areas to reason-
able time and cost targets by delivering the materials in bulk right on the
spot where and when needed. The enthusiasm and commitment shown
by staff involved, both of the EsB and the shipping companies, was
remarkable. Calculated risks were taken by many captains in approach-
ing ports without adequate, up-to-date charts. Many of the ports had
long been silted up and were fringed with foul ground of one kind or
another, not to mention the always turbulent Atlantic ocean. The REO
land crews, usually of local young men, engaged in the loading and
unloading operations, appeared to be almost obsessed with demonstrat-
ing how quickly they could load and discharge and they would obviously
take it as a defeat if a vessel missed a tide because of delay on their part.
Despite the apparent frenzy of the loading and unloading, only one
accident (not fatal) was recorded against the activity on its five years of
operation.

104
CHAPTER NINE

‘Sing the Peasantry and then . . . J

The People — 1946

THE POPULATION TO BE SERVED BY THE SCHEME was about 1.75 million, of


which about 35,000 (or about 2 per cent) lived in small villages. The
remaining 98% lived in the open countryside, sometimes in small clusters
of houses, but generally in scattered dwellings. It was estimated that
about 80% of the total target population depended on farming for a
livelihood. The remaining 20% was made up of rural workers not in
agriculture, town workers living in the country, government officials,
shopkeepers, publicans, pensioners, clergy, schoolteachers. To this
non-farming group the coming of electricity meant an opportunity not
so much to increase productivity as to improve their amenities and
standard of living. Generally they had comparatively steady incomes.
Many were already familiar with the benefits of electricity and were
more than anxious to avail of these without delay. It was from the ranks
of these non-farming rural dwellers as well as from some of the more
progressive farmers that the instigators and organisers of the local rural
electrification committees very often emerged.
The success or failure of the scheme, however, would depend on how
it was received by the 80% of the rural population who depended on
farming for their livelihood. From the very conception of the Shannon
Scheme, its eventual extension to include the farms of Ireland was
envisaged by its planners. It was realised that the prime objective — the
transformation of the new State into a modern socially and economically
progressive community — would be only partially achieved without the
inclusion of the farming sector. Unless the quality of farm life was
improved and the low productivity of the majority of farms lifted, the
main export from the country would continue to be people, mostly the
young and active and usually from a farming background.
The experience of the US, Canada and many countries of western
Europe over the thirties confirmed that the coming of electricity to rural
areas had sparked off a transformation of attitudes, living standards and

105
The Quiet Revolution

economic productivity. It was realised, however, that the problems of


rural electrification in Ireland would be unique in many aspects and that
the approaches that were successful in other countries would not
necessarily be so in the Irish context.
To begin with, the time of launching the scheme — immediately after
World War II - was hardly propitious. Even in times of prosperity, the
initiation of such a large-scale project would have been an ambitious
venture. In the Ireland of the late 1940’s and early 1950’s with its
depressed and slow moving economy, it was an immense act of faith.
In 1946, Eire, formerly the Irish Free State and later, in 1949, to
become the Republic of Ireland, was still a very young political entity.
It had had a difficult birth in 1922 and in the intervening twenty-four
years had been buffeted by a series of economic and political storms
which had left its economy in a depressed condition at the end of World
War II. An agricultural country with a home market of less than three
million people, it had traditionally depended on exports of agricultural
produce, almost entirely to Britain. In the decade immediately preceding
independence, which included the period of World War I, there had
been an unlimited market in Britain for Irish produce at very high prices.
Because of the war, competition from other suppliers was almost non-
existent, and unfortunately too many Irish exporters took advantage of
this to ship food of very poor quality. Shortly after the end of the war
the bonanza ceased and prices collapsed. To quote one writer, ‘When
the war was over and supplies from Denmark and overseas were again
available, the British consumer had only too clear a memory of bad eggs
and worse butter from Ireland.”
The prospect facing the government of the new State was a daunting
one. On top of the collapse of the agricultural export market, the country
had been badly shattered by the conflicts of the War of Independence
and was to endure ever more from the succeeding Civil War and the
period of severe civil unrest which followed. By 1924, however, the task
of rebuilding was well under way. One of the first aims of the new
National Government was to recover the export markets which had been
lost. Regulations were swiftly introduced to ensure standards of clean-
liness and purity in the packaging and marketing of pork, bacon, eggs,
and dairy produce to such effect that by 1929 the quality image of these
products had been restored on the British market and prices paid had
risen to the level of the competitors. Penetration of other markets was,
however, an almost impossible task and was to remain so until Ireland
was admitted to the EEc in the early seventies.
1929 saw the beginning of the great world depression with its conse-
quent detrimental effect on world trade. In Ireland this was followed by

106
One of the back-breaking chores of the farm wife in pre-
electrification days.

107
The Quiet Revolution

the ‘economic war’ with Great Britain in 1932 when penal duties on Irish
produce were imposed by Britain in order to recover the amount of the
land annuities being withheld by the de Valera government. Yet again
the British market for Irish agricultural produce collapsed. The total
value of exports dropped from £36 million in 1931 to £19 million in 1933
and cattle exports from £18 million to £9 million. Ireland retaliated with
duties on British goods. These stimulated the small industrial sector and
employment in industry and services rose to meet the limited home
market. This, however, was offset by the fall in the income of the farming
sector as their costs rose and markets and prices toppled. In 1935 with
the ‘coal-cattle pact’, and in 1938 the Anglo-Irish agreement, there was
a recovery in quantity and price of agricultural produce on the British
market, but this was not sufficient to compensate Irish farmers for losses
in the early part of the decade.
In 1939 World War II broke out. This time there were not the bonanza
prices of 1914-18 as a very deliberate cheap food policy and strong
control by the British government held down prices in this, the only
export market available to the Irish farmer. British and Northern Ireland
farmers were encouraged towards maximum food production as part of
the war effort and were compensated for low prices by heavy government
subsidies. No such subsidies were available to farmers in neutral Eire.
As the war progresed exports dropped to half the pre-war level while
imports plunged to less than one-third of the 1938 figure. Imports of
fertiliser and animal feeding stuffs and of wheat for flour milling dried
up almost completely so that the country was compelled to feed itself
from its own resources. Compulsory tillage was introduced and the area
devoted to growing wheat rose to almost three times the pre-war figure.
The country survived on a diet of home-produced meat, potatoes, eggs,
butter and bread milled from home-grown wheat, which, while monot-
onous, ensured that it was still one of the best fed communities in the
world during this difficult period.
By the end of the war the Irish economy, especially that of the farming
sector, had suffered immeasurably. Import prices had risen by 122%
between 1939 and 1945 while export prices only rose by 89%. As a
compensation there was a large rise in foreign assets. This rise was not,
however, the result of booming exports as in the 1914-18 period, but of
falling imports as both foreign commodities and shipping became
unavailable. Agricultural production stagnated as, in the absence of
fertilisers and animal feeding stuffs, output was only maintained by
drawing down the fertility of the soil.
An examination of the structure of the post-war farming in Ireland
will also help in understanding the particular problems that made the

108
The People

task of rural electrification in Ireland in 1946 so challenging. The total


acreage available for farming in the state (excluding woods, mountain,
bogs, marshes etc.) is about twelve million acres of which at the time
about seven and a half million or 60% were in pasture, 2 million devoted
to hay and two and a half million (22%) in tillage. The number of
agricultural holdings of over one acre amounted to about 320,000, giving
an average size of holding of thirty-seven acres. However, there was a
very wide variation from average both as to farm size and quality of land.
Table 2 quoted in the 1944 ‘Report on Rural Electrification’, shows the
counties in which most of the acreage is in large holdings (predominantly
the eastern countries), in medium-sized holdings (predominantly the
southern counties) and in small holdings, the western group. Table 3
gives the break-down of farm size in the same three groups of counties.
The large number of small holdings even in the first two groups is
noteworthy. In the eastern group, 70% of holdings were under fifty
acres, while in the southern group the figure was 62%. Thus, even in the
most fertile parts of the country the majority of the farmers were of
comparatively modest means. In England, by contrast, the average size
of farm was about twice that size.

When one turns to the western group however, the gap between the two
‘Irelands’ is clearly seen. Historical, political, geographical, and geolog-
ical influences had resulted not only in smaller farms (88% under fifty
acres and 68% under thirty acres in 1946) but in poorer land, a less
favourable climate and consequently lower average farm incomes than
in the remainder of the country.

A survey of twenty farms of sizes ranging from ten to one hundred


acres carried out in mid-Roscommon in 1945-46 by R. O’Connor gives
aver good picture of the typical western dry-stock farm of the period:

Meadow and pasture occupied 80% of the available acreage while tillage
took up 18%, the remainder being unproductive. The average value of
total output per farm was £416 which inclyded a figure of £94 as value of
the produce consumed in the home. This latter included milk, butter,
buttermilk, bacon, poultry, crops and turf.
Expenses excluding labour amounted to £96 leaving an average total
labour income of £320 or 77% of total output. When the cost of hired
labour per farm of £46 was deducted, the total family income per farm was
£274 or just £5 per week. The survey showed that the average total ‘family
labour units’ per farm was 1.87 (one labour unit represents the equivalent
of an adult worker fully engaged for 52 weeks), giving an average annual
family labour income per unit of £146.8 or £2.16.0 per week.

109
The Quiet Revolution

TABLE 2
Size-Distribution of Farms in the Various Counties

Aggregate agricultural acreage in each size-group as a percentage of the total


agricultural acreage of each county at 1 June 1931.
(Arranged from Table 61, Statistical Abstract 1940, Dept of Industry and Commerce) -

Group County Up to30 30-50 50-100 100-200 Over 200


Acres acres acres acres acres

Countiesinwhich Kildare , 8 16 26 41
most of the acreage Meath 13 10 14 22 41
is in large farms. Dublin 15 10 20 26 29
Wicklow 10 10 24 28 28
Westmeath 16 15 21 20 28
Offaly 14 15 ZS 21 2
Louth 26 13 20 18 23
Average 14 12 19 23 32

Countiesinwhich Kerry 19 21 35 18 7
most of the acreage Limerick 13 7 34 25 11
is in medium sized Cork 10 14 34 29 13
farms Wexford 11 14 31 28 16
Clare 19 22 30 17 12
Kilkenny 10 14 30 28 18
Tipperary 11 15 29 25 20
Carlow 12 13 29 27 19
Waterford 8 9 26 33 24
Laois 15 14 25 26 20
Average 13 16 ot 25 15

Counties in which Mayo a3 22 12 6 7


most of the acreage Leitrim 50 26 17 5 2
is in small farms Monaghan 46 24 19 4 4
Sligo 45 22 16 9 8
Cavan 44 26 19 7 4
Roscommon 41 23 16 10 10
Donegal 35 20 25 14 8
Longford 34 23 21 12 10
Galway 30 26 21 il 12
Average 41 24 18 9 8

110
The People

TABLE 3
Number of Agricultural Holdings Exceeding One Acre
1 June 1945
(Arranged from Table 61, Statistical Abstract 1946, Dept of Industry and Commerce)
Up to30 = 30-50 ~—-50-100 ~—-100-200 200 acres
acres acres acres acres and over Total

Kildare a 2S 778 823 692 417 5,835


Meath 4,437) 11,791" 1,270 867 542 8,907
Dublin 3,063 477 506 346 144 4,563
Wicklow 2,028 818 1,200 711 280 5,037
Westmeath 3,83) S850 15230 554 336 7,536
Offaly 3,278 1,576 1,466 663 294 CPT
Louth 3,493 "Bes 538 252 114 5,110
Total Zone) Jom sU33q0 4,085 062127 44 238
(52.5%) (17.5%) (16.0%) (9%) (5%)

Kerry 8,814 4,041 3,987 1,327 440 18,609


Limerick 292,070) 3,080 13220 244 12,345
Cork D0 0 ol. 846 821 21651
Wexford O5a0 eel 128 82,290 e250. 314 8,943
Clare G203" 95,0867 83,125 963 361 14,393
Kilkenny DO90™ 1-040 Perl 010 nist 19 338 7,109
Tipperary 05002) 31578 89,099 be L825 578 16,137
Carlow 1,490 614 855 440 133 35502
Waterford 2,107 830 ~=—-:1,339 955 314 5,545
Laois 37518 715322" = 1,346: 718 248 6,952
Total AQAA 2522717) 29.051 913,009 © 93,(91> mel 21802
(41%) (21%) (24%) (11%) (3%)

Mayo 23,824 4,520 1,678 388 219 30,629


Leitrim 7550 9.210 842 156 24 11,182
Monaghan 7,770 2,024 1,002 180 29 11,005
Sligo 8,659 2,205 935 242 98 123139
Cavan 9,910 3,015 1,400 239 50 14,614
Roscommon 11,253 3,740 1,456 391 98 16,938
Donegal 16,286 3,465 2,638 1,055 384 23,828
Longford 4,040 1,366 750 228 ig) 6,463
Galway 1538599 7,095.5 3,287 808 347 27,397
Total 105,551 29,640 13,988 3,687 1,328 154,194
(68%) (20%) (9%) (2%) (1%)
ee Ween eS Hite a ee
Total 178,246 62,649 50,652 21,441 7,246 320,234
(All counties) (56%) (20%) (16%) (6%) (2%)

111
The Quiet Revolution

If the value of the produce consumed in the farmers’ homes was excluded,
the annual cash remuneration per unit of family labour would be £69 on
the under 50 acre farms; £128 on the over 50 acre farms, and £96 on all
farms. Calculated at a weekly rate this would amount to 27/- per unit and
the small farms (under 50 acres); 49/- on larger farms and 37/- on all
farms.”

240,000 of the 320,000 holdings in the country at the time were of fifty
acres or less — indeed 178,000 of them were thirty acres or less (see
Table 3). It should be noted that the land in the part of County
Roscommon covered by the O’Connor survey is by no means the least
fertile and if this survey were taken as indicative of the situation of the
small farmer in the country as a whole, then a very large proportion of
the potential consumers of the Rural Electrification Scheme were
involved in 1946 in a subsistence type of economy.

Dr J. J. Scully in a study carried out much later on farming in the west


of Ireland contrasts this farming for subsistence with commercial farm-
ing. He defined the commercial farm as market-orientated, its output
influenced not alone by the relative productivity of inputs but also by the
relative profitability of alternative farm products and farming systems.
Commercial farms by this definition there were in plenty, mostly in the
eastern and southern parts of the country and mostly in the larger
acreages. The owners of these farms, geared as they were to market
demands, welcomed electricity as a means of increasing efficiency and
productivity. In the low-income or subsistence farm on the other hand

farm operations are usually performed in a traditional manner with the


same inputs being combined in the same way to produce the same products
from one year to the next... .
The low income segment is normally static or maybe contracting. It is the
segment which suffers the greatest depletion in manpower because of
migration to other sectors with the progress of economic growth.*

This was the case in the majority of farms in the poorer districts, but
Scully’s description also applied to a very large number in the more
prosperous parts. It was with these householders that the local rural
electrification committees and EsB Area Organisers had to work hardest
to persuade them of the benefits electricity could bring.

There were many factors inhibiting change. Clearly, many small-farm


families were living in near poverty. Following the failure of the potato
crop and the Great Famine of a century earlier, Irish farming had moved

112
The People

away from the intensive tillage systems and towards a heavy dependence
on livestock rearing. The switch to low-intensity cattle production was
understandable in the context of a depleted population on the land and
poor markets for other produce, but this was not the kind of enterprise
to generate high farm incomes and indeed it was badly suited to the small
farm, particularly on the poor soils of the west.
There were also sociological obstacles. The Land War was still fresh
in the folk memory. Land ownership rights had been dearly won and in
too many cases security of possession was regarded as more important
than progressive agricultural development. At the 1951 census 73% of
the country’s 199,000 male farmers were over forty-five years of age and
almost a quarter of these were unmarried. Even where there were
potential heirs working on the farm there was a general reluctance to
yield decision-making or hand over ownership to the younger person.
Educational levels were low: less than 5% of farmers had gone beyond
the post-primary school. Agricultural advisory services were still com-
paratively underdeveloped. Perhaps most importantly the great rural
movements, such as Macra na Feirme, Macra na Tuaithe, Muintir na
Tire and the Irish Countrywomen’s Association, which were later to do
so much for adult education and for social and economic development
had not yet emerged to their full strength.

With this rather depressing combination of circumstances rural emi-


gration was inevitable. Sons and daughters sought a better life elsewhere,
leaving, in many cases, the least resourceful and least educated to run
the farm — a tragic merry-go-round which almost ensured that the
pattern would be repeated from generation to generation. The hard-won
political independence of 1922 had not in itself, as many of its founders
had hoped and expected, put an end to this delibitating haemorrhage of
emigration which, generation after generation for over a century, had
been draining the countryside, mostly from the farming community and
especially from the areas of poor land and small farms.

By 1948 the government had become so concerned that it set up a


Commission on Emigration.and Other Population Problems to study the
situation and make recommendations. The Commission identified the
stagnation in agricultural production and the decline in agricultural
employment as one of the prime causes of emigration. It reported that
it would be necessary to increase agricultural output by fifty per cent in
order to maintain the existing population on the land, but highlighted
the immensity of this task by pointing out that there had been no growth
in the volume of this output over the previous half century. If the system

115
The Quiet Revolution

of agriculture were based on the intensive use of land, the Commission


believed that not only could the agricultural population be maintained
at its then level, but that it could be increased.

Turning to living conditions the Commission stated:

In rural areas, many houses are quite unsuitable as regards design, accom-
modation and comfort, if judged by modern urban standards. The contrast
between rural and modern urban housing is so marked that it must have
a considerable effect on the decision of a person in a rural area whether to
emigrate or to marry and settle down in his locality...
Poor material standards of life are to be found in many parts of the
country, but particularly in rural areas where the exacting demands of
agricultural activities are aggravated, in a great many cases by the inade-
quacy of such amenities for houses and farms as power, light, water and
sanitation. Much drudgery is caused by out-moded methods of day to day
working and living.‘

With regard to the Rural Electrification Scheme, the Commission


observed:
The national scheme of rural electrification should, in due course, consid-
erably affect the lives of those in rural areas. The provision of electricity
for power, light, heating and cooking, not alone in dwelling houses, but
also in out-offices and farmyards should revolutionise many habits of rural
life and in particular improve the lot of the housewife. It should enable
many farmers to introduce improvements and modern amenities compar-
able with those of urban centres. It could ease the way for the return of
the craftsman to rural areas, providing employment for him both in
agriculture and in cottage industries.

Despite such endorsements of the scheme there remained considerable


inertia to be overcome. Many farmers, particularly those outside the
dairying areas,’ with small acreages and uncertain incomes remained to
be convinced that rural electrification could ease their lot. Not all could
see in electricity the promised release from the many time-consuming
tasks and the consequent opportunities to devote extra time to more
productive activities. Indeed it was questionable if, at the time, profitable
markets could be secured for such increased production. Not all husbands
could see a necessity to ease the traditional drudgery of the farm
housewife. To many, electricity first appeared merely as an expensive,
if admittedly a greatly superior, alternative to the traditional oil lamp or
candle. Their forebears had successfully survived without it and they
could see no great advantage in hurrying to involve themselves in this
new expense.

114
Above: The electricity arrives at Rosses Point, Co. Sligo, November,
1947. Below: The new electric pump — a source of wonder and of
benefit to all in the home and on the farm.
The Quiet Revolution

That there was good reason for this cautious approach and that it was
not confined to electricity was pointed up by Professor James Meenan in
his minority report as member of the Commission on Emigration and
Other Population Problems:

The average age of male farmers was 55 years in 1946. The average farmer
would thus have been 23 years old when the First World War broke out in
1914. Since then, he has lived through two world wars, the Anglo-Irish
War and a Civil War; two collapses in prices including a major depression;
four devaluations of currency, the ‘Economic War’ and all the business of
reconstruction after 1918, 1922, 1938 and 1945.
It is not perhaps a matter for wonder if farmers to-day take short views
and are reluctant to embark on improvements that, however excellent
they may be in theory, are only hostages to the fortunes of prices and to
the political and economic events that control prices (p.377).

It is necessary to stress that the reluctance to change on the part of so


many was based on more than suspicion of innovation or of the motives
of the innovators. There was a positive side — holding on to a traditional
way of community life that had been shaped by history and tempered in
the fires of many holocausts. It was simple, sometimes austere, but
basically wholesome. Hard physical work, neighbourly co-operation and
a strong, simple religious faith were its characteristics. There was a
symbolism in the fire on the hearth which was never allowed to go out.
The lack of mobility outside one’s own area contributed to a strong
consciousness of community and the sense of belonging.
Despite the absence of wealth, it could be said that these poor rural
communities were better adjusted to their straitened circumstances than
their more affluent successors were to theirs. Incomes were low, but so
were expectations. To quote John Healy: ‘The rhythm was slow and
fixed; the wants were small and if you didn’t have it you went without.
In those days you wanted for little enough on small holdings like
these... .”° The more significant aspects of the cash economy had not yet
permeated people’s lives, especially outside the dairying regions where
regular cheques for farm produce were not in existence. In a sense, the
coming of rural electrification with its fixed charges and regular billing
was a definite wedge in the change to a more money-conscious rural
society.
In this context, the experience of Jim Wolahan, Rural Area Clerk in
Dundalk District in 1955, is worth recounting as he told it to the writer.

In July [1955] I was transferred from Oldcastle to Killanny rural area on


the Louth-Monaghan border. As I had only a bicycle for transport I

116
The People

enquired locally for a digs near the Area Office at Essexford. I was directed
to a cottage occupied by the man of the house, his wife, two children and
a ‘granny’. It was spotlessly clean and they were willing to take me. When
it came to fixing terms, the woman was nonplussed. She had never kept a
lodger before. What had I been paying in Oldcastle? £3 per week. Well
Oldcastle had amenities not available here — no toilet and as yet no
electricity — how about 50/- or say 7/- per day. If I went home for week-
ends she would only charge 35/- for that week. Later she decided that even
this was too much. She pointed out that she had her own eggs and brown
bread and apart from buying meat for my dinner had very little expense.
Despite my protestations, she later decided that 25/- was sufficient as it
transpired that I was now going home every week-end.
When the time came to transfer to a new area in south Monaghan, about
15 miles away, my personal transport had progressed from the bicycle to
an old VW ‘Beetle’. I was asked if I would remain on in the house and
‘commute’ to work and I agreed. However, the woman of the house
pointed out that as I now had my lunch out, she would not have to buy any
more meat for me. Her expenses were now almost nil and she proposed
15/- per week as my new rate. After some argument she compromised at
£1!
One day granny announced ‘I am going to look after your shoes’. She
wanted to have some little function in my welfare which did not cut across
the role of the younger woman. Every morning I found my shoes shining.
There was not much place for money in that house.

Change was undoubtedly necessary if the rural economy was not to


stagnate and its population dwindle to crisis point. Most rural people
were aware of this, but the kind of change, its extent and its ‘knock-on’
effects gave cause for deliberation. Indeed writers such as Hugh Brody’
would hold that, in the event, many of the recent social and economic
changes have left these communities all the poorer. It must be left to
future social historians to debate how much of a certain quality of life
was lost with the efforts to eliminate the harsher elements of rural living
in Ireland.

117
CHAPTER TEN

Moving In

IN THE EARLY STUDIES on the organisation of the scheme it was rec-


ommended that the rural parish, with an average area of twenty-five to
thirty square miles and containing five hundred premises, would be a
suitable geographical unit on which to base a rural electrification area.
There were about eight hundred rural parishes in the country, not all,
unfortunately, conforming to the average in either size or population.
To quote W. F. Roe, ‘the original parishes were formed by monks sent
out from the monasteries and it depended on the energy or stupidity of
the monks as to what size a parish would be. We chopped the parishes
around somewhat, but tried to keep them as the nucleus of each Rural
Area. We found the GAA idea was a great help, because most of their
teams were parish teams. . . their loyalty was owed to the parish and in
this way we had rivalry between the parishes as to which should get the
electricity first.”!
Owing to the time-span involved — it would take at least ten years to
complete the scheme — it was perhaps inevitable that there would be this
keen competition between parishes to get to the head of the queue. It
was important that an easily understood and acceptable set of criteria to
determine the order of development should be established. Otherwise
there would undoubtedly be widespread pressure, political and other-
wise, to have areas selected out of turn.
It was also essential from the point of view of the government and the
EsB that with the large capital expenditure involved the work should be
carried out in as orderly and economical a manner as possible. The
government, in its approval of the scheme in August 1943, had stressed
that priority should be given to the most remunerative areas with the
stipulation that initially one area must be developed in each county. For
reasons of efficiency and control, the work was decentralised within the
framework of the ten EsB ‘country’ administrative Districts and thus,
after the initial selection of one area in each county, the contest for
priority was in fact between areas within each Ess District.

118
Moving in

The term ‘most remunerative areas’ referred to the ratio between the
capital cost of supplying the consumers and the annual fixed charge
revenue yielded, which (with the help of the government subsidy) was
intended to cover the annual fixed costs of the networks. When the
selection process had been in operation for only a short time it was
pointed out that proximity to existing electricity networks gave an area
an unfair advantage even where the percentage acceptance in a more
remote area was higher. The criteria were consequently modified so as
to give the percentage of householders accepting supply equal weight
with the percentage return on capital. It was therefore of the greatest
importance in securing early selection that the maximum number of
householders would agree to take supply. A large proportion of the
farming community had yet to be persuaded that the advantages of
electrification outweighed the costs involved. Many small farmers were
still operating in a subsistence economy. The prospect of committing
themselves to paying a regular fixed charge in summer as well as winter
irrespective of their electricity consumption and of their financial situa-
tion was difficult to accept. In addition there was a widespread suspicion
that electricity was dangerous.
On the other hand, in most areas a considerable number of house-
holders were more than willing to sign up for electricity. As well as the
more progressive farmers there were the clergy, teachers, shopkeepers,
public employees and other rural dwellers who were more aware of the
benefits of electrification, both to themselves and to their community as
a whole. It was from these people that the initial pressure for electricity
came and in most cases the early selection of an area depended on their
success in persuading their more reluctant neighbours to sign up. The
focus of activity in this field was the local Rural Electrification
Committee.
Roe had quickly realised that his task would involve not alone moti-
vation of the REO staff but of the rural community itself. If the scheme
were to be launched without the involvement and commitment of the
rural community it would fail. It would involve a change in ways and
attitudes which would not be achieved by REo agents alone. A desire for
change and improvement in the prevailing standard of living, and for
availing of the coming of Rural Electrification to achieve this, must be
fostered from within the community. Roe’s involvement with Muintir na
Tire had made him sensitive to community reaction and thinking. Some
lengthy discussions with Canon John Hayes (founder of Muintir na Tire)
convinced him that the way to motivate the consumers in the rural areas
was by getting them to participate in the work. Thus was born the idea
of the parish Rural Electrification Committee.

119
The Quiet Revolution

The make-up of these local committees varied. Sometimes they


evolved from the local guild of Muintir na Tire, sometimes from Macra
na Feirme, the local co-operative society or the parish council; in many
cases some local people simply came together and formed an ad hoc
group. Every effort was made to avoid identifying the committee with
any political party or denomination.

THE PRELIMINARY CANVASS

The main function of the committee was to carry out the preliminary
assessment of demand for electricity in the area and submit this to the
ESB in the form of a memorial signed by all householders prepared to
install it. Where the response was highest an official canvass was then
carried out by an REO Area Organiser who measured up the premises,
determined the fixed charges and got the householders to sign official
application forms. He was also in a position to explain the benefits and
answer queries from householders some of whom had only the most
rudimentary idea of electricity at that stage.
Naturally the local committee was motivated to obtain a high per-
centage of acceptances so as to ensure that its area qualified for early
selection. There were two criticial stages in this regard. The first was
when the original memorial was being prepared, as on this depended the
promptitude with which the area was canvassed officially by reo. The
response varied from ‘negative’ through ‘doubtful’ to ‘enthusiastic’.
Writing from west of the Shannon, Ora E. C. Kilroe reflects the typical
reaction of the younger farm housewife:

Two young men descend on us like a heavenly visitation. They are a


deputation from the local Committee and want to know if we will take the
electric! I can hardly believe my ears. What a boon and solace it would be
and how miraculously it would lighten the unrelenting drudgery of a
farmhouse. I walk about all day in a dream of immersion heaters, electric
irons, churns, incubators and husband happily contemplates oatcrushers.
We hear only one in the District has refused it, a hardy old dame of 85
who says indeed she will not have it and be setting fire to the thatch.’

A rather touching letter quoted in REO News in November 1951 is


indicative of the latter approach, particularly among the older genera-
tion. It was written to a friend of the editor by an old retired servant of
the family who then lived in a small cottage between Monasterevin and
Rathangan:

120
Above: The ESB crew is welcomed and entertained by a grateful
community. Below: Battery radio no longer required!

121
The Quiet Revolution

_..the electric light is at last installed in far-off Clonmoyle. Kate H— could


not stick the bright light at first and lit an oil lamp. I did laugh at her. They
are all very grand until it comes to the paying for it and then they may get
a surprise... .

The old lady flatly declined the offer of her former employers to pay for
the installation of light in her cottage and pay all the electricity accounts:
she simply preferred not to have electricity.
A former Area Clerk recalled an elderly couple living on their farm in
County Monaghan who would not take supply despite repeated visits
from members of the local committee. Their only child, a daughter, had
married, and her husband had moved in and was running the farm.
However, the old man would not relinquish control up to the day he
died. As expected, he left the place to his daughter but on condition that
there would be a room provided for his widow. This request was of
course honoured by the young couple who immediately looked for
electricity supply. The most convenient point of connection was at the
back of the house, outside the widow’s room. She, however, would not
allow electric wires to be connected to any wall forming part of her
room. Connection was made to another part of the house wired for
electricity — all except the widow’s room. She continued to use an oil
lamp up to the day of her death in fulfillment of what she interpreted as
her late husband’s wishes.
Not all the older people were so reluctant. In the Looscaun area near
Woodford in County Galway in 1953 a seventy-eight year old bachelor,
Dan Bonfield of Gurteeny, took supply into his tiny two-roomed house,
possibly the smallest premises connected up to that time. His total
installation was one light, so that he could play cards with his neighbours.
He was very fond of a game and with the natural dimming of his eyes at
that age he felt that an oil lamp or candle was insufficient if he was to
keep a reasonable eye on his fellow-players.

THE OFFICIAL CANVASS

The second and more vital stage was when the REO canvass commenced
and the householder was advised of the actual fixed charge on his or her
premises and asked to sign an official application form. At this stage
many who had signed the original memorial fell out, usually to the
dismay and chagrin of the local committee, as a high percentage of
refusals generally meant relegation to a low position in the league table.
The efforts of the committees and the methods employed to try to
secure a high sign-up were varied. In many cases a further bout of intense

122
Moving in

canvassing was involved. Other methods adopted were more spectacular.


The committee of the Knockbridge, Louth and Tallanstown areas of
County Louth, for example, persuaded local businessmen to donate
advertising space in the Dundalk Democrat for the purpose of exhorting
the faint-hearted to sign up. The examples given show the vigorous style
considered necessary to achieve results. (The second example shows the
ability of the donor to interlace his own message with that of the
committee.)
It is not possible to describe the workings of the local committees
without referring to the role of the local clergy. For generations the local
priest in rural Ireland was looked to by his congregation for leadership
and advice. What was more natural than that the chairman of the local
Electrification Committee should be the parish priest or curate (and for
that matter that the secretary should be the local schoolteacher)? In
general the clergy recognised the value of electrification and took a
positive attitude in encouraging their parishioners to avail of the scheme.
Indeed, in many cases the local PP personally took part in a round up of
refusals among his flock when the canvass was not going too well. Phil
Casley, one of the Area Organisers attached to the Galway District
recalled that in the southern part of the county he was asked to accom-
pany the parish priest on such a mission. It was obvious, however, during
the exercise, that many of the householders who signed up did so because
of their traditional reluctance to refuse a direct request from the PP.
Nevertheless, one parishioner successfully avoided the issue by disap-
pearing into the horse’s stable with a large forkful of hay when he
recognised the purpose of the approaching pair. By the time they had
reached the door of the stable, all that was to be seen inside was a horse
contentedly munching from the pile of hay. Of the wary householder
there was no sign, nor did he subsequently reappear in any other part of
the farmyard. The visit had to be written off as a failure. Subsequently
the ESB man ascertained what he had at the time suspected, that a pile
of hay can be excellent camouflage, given a co-operative horse.
Unfortunately many of the agreements obtained in this way did not
last the course, especially in the early years. When the area was selected
for development and the construction crews moved in, the moment of
truth arrived for many who were faced with the choice of wiring up their
houses and getting the supply connected or of backing down and opting
out of the scheme. In some areas at this stage, as many as one-third
backed down. The reasons given were many and varied, but for the vast
majority it was a feeling that when it came to the crunch electricity was
a luxury they could not afford. In the case of such people a former Ao
recalled:

123
RURAL ELECTRICITY - SIDE “LIGHTS” FROM HISTORY -
1 — 878 B.C.: The Firefly and Lantern-Beetle catchers of Abyssinia condemned the
introduction of resin wood “‘splinters” for lighting. “Spliriters’” were TOO DEAR &
TOO DANGEROUS !
134 B.C.: “Splinter” makers of Sweden condemned the introduction of rush
“dips” - TOO DEAR & TOO DANGEROUS !
310 A.D.: The “dippers” - of Paris cried aloud against the use of tallow candles.
These were - TOO DEAR & TOO DANGEROUS !
1660 A.D.: The Candlemakers’ Guild of London protested against the introduction
of the “‘convex oil lamp” - TOO DEAR & TOO DANGEROUS !
1819 A.D.: The oil lamp makers of Quito were “afire” against the introduction of
coal gas - TOO DEAR & TREACHEROUSLY DANGEROUS !
1878 A.D.: The Coal Gas Company of Tibet (we’re told) “fumed” against the
introduction of electricity - TOO DEAR & TOO DANGEROUS - !!
Householders of Tallanstown, Louth, Knockbridge, you are not living in Tibet and
surely you do not want to be 70 years behind the times. Think then of something e!se to
say against Electricity, for it is neither DEAR NOR-DANGEROUS. This is our last
earnest appeal and your last chance. Now or never. Sign your acceptance form without
hesitation.
(This space has been generously donated to your Committee by Messrs. Macardle, Moore
& Co., Dundalk Brewery)

KNOCKBRIDGE - LOUTH - TALLANSTOWN


RURAL ELECTRICITY
Deanna (Dunbin) writes : “I jilted the cad. He broke his promise to take Electricity”
More power to you Deanna. The mean rat was almost guilty of BREACH OF
PROMISE. Don’t worry dear: we have five of the handsomest - if shyest - young
bachelors on our Committee. THEY*LL say it with ELECTRICITY if

AAA

HERE TAM YOUR PETER E?


PMY KNOCK EDP OUTSBY THE oe
BUT TAS WE BOTH GIVE YOU DE Liren TD
PEEASE SUPPORT US DAY nN Deter
YOU WILL KNOW
Wiehe Tee Wesel
If you buy your clothes at
26, Clanbrassil St., Dundalk.

Advertisements for the rural scheme published in the Dundalk Democrat.

124
Moving in

Some organisers felt misgivings, especially in a house with old folk, to see
them torn between a desire ‘to help the priest’ and their genuine fear of
tying a new ‘Ground Rent’ around their necks. However, these misgivings
quickly dissolved when one called on the same people after they had been
‘switched-in’. Their obvious delight with electricity had completely over-
come their fears. It was amusing to hear them comment about cobwebs
which they had never known existed.*

A secondary reason in the case of the older people was that it was
somehow dangerous. The Evening Herald (7 February 1953) carried a
story from the Annies district of Scotshouse near Clones where an old
man paid his customary weekly visit to his sister who had just been
connected up. He watched with fascination as she operated the new
electric iron and boiled water in the new electric kettle, but stubbornly
refused to drink the tea made from the latter, as he believed that the
water was electrified. Willy nilly, she had to make fresh tea from water
boiled in a traditional kettle on the turf fire.
As times improved, as the value of electricity to their neighbours
became manifest and as their original fears receded all of these house-
holders almost without exception eventually applied for connection.
Not all the clergy were in favour of the scheme. Here and there one
found a parish priest who had fixed ideas about how his parishioners
should live — as their forebears had done — and who suspected that
electricity was only the first step in a radical and possibly unwelcomé
change in their lifestyle. Others considered that it involved the people in
unwarranted expense. In a parish not very far from Cork City, the parish
priest was opposed for this last reason. If a large number of the pari-
shoners found it difficult to pay their church dues, as appeared to be the
case, how could they afford this new expense? On the other hand, his
curate, a young man, was so committed that he was accused by his
superior of preaching rural electrification on Sundays instead of the
Gospel. However, the work went ahead and towards its completion the
parishoners were anxious to have a formal ‘switch on’ ceremony in the
local hall. It was decided for obvious reasons to hold this while the old
man was away on holiday. In the event, he returned unexpectedly as
proceedings were getting under way, no doubt having been alerted to
the goings- on by a loyal parishoner. He strode on to the platform where
the various members of the committee were assembled. Facing the
audience he declaimed ‘Ye have it now and I can tell ye it’s going to be
dear, and mighty dear, but ye would have it and ye can now go and pay
for it!’. Even in this case, not many months had passed before the
reverend gentleman himself had been converted to electricity. In another

125
The Quiet Revolution

parish the PP was adamant that he would not take the supply until the
canvasser slyly let drop the information that the bishop was very keen
on the scheme. Standing up and striding to the table he thumped it,
exclaiming ‘Well, what’s good enough for Jim R- is good enough for
me. Ill take it.’
In general, however, clergymen were very aware of the opportunity
offered by electricity and the necessity to grasp it. Many worked very
hard with the local committee and with the REo crew when it arrived, to
ensure that in their locality as many householders as possible would take
supply and avail of it to improve their living standards. In the final report
on Ballivor Area in County Meath the RAE refers to the help given by Fr
Kiernan, the parish priest:

From the first day of arrival he gave the construction crew every assistance,
got accommodation for them and on the spiritual side with the help of the
Area Organiser Barney McEneaney arranged a special retreat at times
most suitable for their working hours. If we ever decide to have a patron
priest of rural electrification he is readymade.

One of the veterans of those early days, Noel McCabe, in recalling the
event to the writer added:

To ensure the maximum attendance Fr Kiernan visited the local hostelries


and ‘requested’ the owners to close down for an hour each evening while
the retreat was in progress. However, being a just man he also ‘requested’
the local Sergeant not to visit the pubs for a corresponding period after
closing time!

With the arrival of the area crew in the parish the prospect of electri-
fication at last became a reality and a new flurry of activity began.
Suitable premises had to be secured for the area office, as also had a
stores and outdoor storage compound. Digs had to be provided for the
travelling members of the crew. House wiring contractors had to be
located and householders encouraged to commence wiring their prem-
ises. This last was the almost irrevocable commitment and many of the
faint-hearted fell out at this stage.
The Area Organiser and the field design engineer usually arrived some
weeks ahead of the rest of the crew to confirm the final ‘acceptance’
position and ensure that the pegging out of the lines got well ahead of
the construction gangs. Wayleave notices had to be served on landowners
to cover every pole and house service, and any disputes arising had to be

126
Moving in

cleared up before work could begin. This latter task was usually carried
out by the Area Organiser who had to call on every householder in the
area. Sometimes there were problems of identification, as in the Clonark
area in County Roscommon, when the Area Organiser found one group
of eight householders all of whom were named McManus. A similar
problem arose near Glengariff in the Castletownbere area where eight
householders connected to the same transformer in the townland of
Firkdale were all found to be O’Sullivans. For the record, the eight first
names were Johanna, Humphrey, Edward, Mrs W., Patrick J., Patrick
E., Michael J. and Michael P. Og.
This first visit of the organiser had many objectives. He ascertained if
the householder was accepting supply and, if the answer was negative,
usually employed his persuasive powers to try to effect a change of mind;
he urged him or her to get the house wiring under way and advised on
provision and position of sockets; he served wayleave notices to cover
any poles on the householder’s land and negotiated changes in proposed
pole positions where necessary to avoid obstruction to farming activities.
At a later date he would return to check on the progress of the wiring
and to demonstrate and possibly sell some items of electrical equipment,
but this first programme of visits gave him a valuable overview of the
total scene and of what problems were likely to arise during the construc-
tion period.
There was frequently a long period of time between the original
canvass and the eventual arrival of the construction crew. In many cases
the Rural Electrification Committee, which had been so active in those
early days, had been dissolved, its main object now achieved. In other
cases, however, the members were still active and concerned to ensure
that all went well. These afforded tremendous help to the crew in using
their influence to prevent householders from backsliding, in locating
lodgings and office accommodation,’ recruiting local labour, organising
the house wiring and resolving many of the problems that inevitably
arose on a project having had such impact on people’s property and way
of life. Even when the formal committee no longer existed, an informal
group, including the local clergy, schoolteachers and prominent local
citizens, frequently emerged and provided very valuable support and
liaison with the community in general.
The newcomers were welcomed. Even in areas where the lodging of
visitors was a rarity there was seldom any difficulty in obtaining accom-
modation for the ten to twenty people who formed the skilled nucleus of
the team. For many crew members of urban background the experience
of rural life and rural hospitality thus gained provided a new and

127
The Quiet Revolution

enriching dimension in their lives and the years spent on ‘rural’ were
looked back on with considerable nostalgia.
This mixing of the visiting members of the crew with the local com-
munity and the recruitment of local men for the duration of the job did
much to ensure the acceptance of the scheme as an integral part of
community development, and many small farmers had their meagre
incomes supplemented by earnings on the scheme. Indeed for many
farmers’ sons it was their first opportunity to earn money in their own
right.° The result was an unparalleled degree of goodwill and co-opera-
ation. Despite the necessary encroachment on the land and property of
almost every householder, disputes were a rarity and where they arose
were usually solved on the spot.
The Esp staff on their part responded to the friendly welcome by
entering into the local scene with enthusiasm. Jim Wolahan, a former
Area Clerk recalled his stay with a farm family in their house in County
Monaghan and how, on returning to his digs at night, when the family
had gone early to bed, he would find his supper laid out on the kitchen
table. His first duty, however, was to take a feeding bottle of milk to a
piglet — the delicate runt of the litter — which was cosily ensconced in a
canvass bag hammock beside the fire. Only when the bonham had been
fed and tucked in for the night did he commence his own supper. In
being entrusted with this important chore he felt he had been completely
accepted as one of the family.
The people of Blackwater, Co. Wexford, must have felt that the Area
Organiser Phil Cox was one of their own as they saw him leading the
parish priest’s donkey and cart down the village street laden with a table
and press to equip the newly opened area office, which their energetic
committee had miraculously located. The same man became even more
integrated into the community when he married a local girl. The newly
arrived RAE in Ballivor, Co. Meath, Angus Ryan, looked down the
village street and anxiously enquired of the local Area Organiser ‘Who
is the mot?’ Some months later he led her to the altar. Over the course
of the scheme, the number of Ess staff who first met their future wives
while on rural electrification work ran into very large figures. The
Kilkenny Journal of 14 September 1950 referring to the departure of the
crew from the completed Gowran area said ‘Gowran will miss them;
these boys were drafted in here total strangers less than six months ago
and promptly won the friendship and confidence of all and the hearts of
some’.
In the evening the pubs and dance-halls were enlivened by the new-
comers who brought to the local scene their own particular style. Some
had earned their place in the travelling crew by virtue of their fabulous

128
Moving in

strength or physique — they could fling poles around like matchsticks —


and delighted in demonstrating their porter-drinking capacity. Others
were more interested in the local women. Many had musical or dramatic
talent and proved a great asset to the local societies. The advent of
electricity to the parish hall was frequently the occasion for a celebratory
concert or play. One crew operating in the west Cork area staged many
such concerts and plays from its own resources in aid of local worthy
causes.
While the visiting crews were generally well behaved, there was the
odd brawl. Often this was over local girls, who to quote an old ‘rural’
hand ‘were always glad to welcome new blood’. It had, he explained,
been the same with the army during the war and the visiting rural
electrification crew were regarded as worthy successors. The local boys
employed for the ‘duration’ also made their presence felt in the village
on pay night and inevitably there were clashes. An RAE in County Sligo
was horrified on opening a copy of the Champion to see the headlines of
a case in the local court: ‘Ess Men Convicted of Assault’. With visions
of his visiting team being run out of the district he made enquiries to find
that the men in question were natives of the locality employed as
short-term casual labourers on the scheme. When asked for their occu-
pation for the purposes of the charge, they had obviously felt that ‘EsB
men’ gave them a cachet superior to that of ‘farmer’s son’.
As work in the area came to an end there was a phasing out of the
local temporary staff. There was a certain loss of face in being among
the first to be let go and reaction was on occasion dramatic. In an area
in County Tipperary one such aggrieved worker expressed his displeasure
by coming in to the village in the evening with a shotgun, which he
discharged at intervals while loudly calling for the supervisor Martin
Whyte to show himself. The latter recalled the incident thirty-five years
later: ‘I had been feeling tired after a hard day but when I heard the
shots the tiredness dropped off me and I ran in fear of my life’.
Another turbulent stalwart - a member of the tree-cutting gang —
raised his axe and, to the accompaniment of some lurid language, buried
it almost to the hilt in a convenient tree-trunk when told he had got the
‘chop’. The supervisor had taken the precaution of waiting until he was
on the opposite bank of a river before imparting the unwelcome news.
Such incidents were, however, the exceptions. Usually the parting was
made with good humour and expressions of goodwill on both sides. In
some cases one or two of the most promising local lads were invited to
travel with the crew to the next area as ‘semi-skilled’ workers, which was
often the first step on the ladder of promotion.

129
The Quiet Revolution

The easy integration of the visiting crew members with the local
community did much to soften the traditional reluctance to change and
the misgivings that might otherwise have existed. In the Lispole area in
County Kerry the roving reporter of The Kerryman wrote of casually
dropping in on the ‘switch-on’ and referred to the high opinion the local
people had of the construction crew, particularly of the business-like
manner in which they tackled their job and the excellent friendly relations
that existed between them and the residents of the locality.
Perhaps the tribute paid by the parish priest in Carnaross, County
Meath, said it all. At Mass the Sunday before the last of the crew
departed he publicly wished them farewell: ‘I was very moved to see the
way in which these nice people came amongst us, did their job with
speed and efficiency that opened our eyes after our experiences with
other official bodies, behaved quietly and decently and left without any
fuss or display’.

Canon John Hayes,


founder of Muintir na Tire.

130
CHAPTER ELEVEN

Problems

BACKSLIDERS

IF ONE WERE TO ASK a Rural Area Engineer or a Rural Area Organiser


who worked during the first decade of the Rural Electrification Scheme
what caused him the greatest problems, the answer would probably be
backsliders, those householders who had signed application forms during
the official canvass of the area but who, when the construction crew
arrived, refused connection or had still not been connected when the
area was completed. The design and costing of the electricity network
for each area were carried out on the basis of the signed application
forms. The estimated return on capital for the area, which, together with
the percentage of applications determined the position of the area in the
selection table, depended on the estimated fixed charge revenue from
the acceptances. There were thus many undesirable consequences to
backsliding. The withdrawal of a householder at this late stage generally
meant loss of expected revenue without a corresponding saving in capital
expenditure. The loss of even one household in a group frequently
resulted in others moving into the ‘uneconomic’ category and so further
reduced the number of houses connected. If, as frequently happened,
backsliders appeared in large numbers, the economic return from the
area could fall drastically. When this occurred, householders in areas
still awaiting selection and perhaps only slightly lower in the table could
feel aggrieved and make accusations of unfair treatment. An appreciable
number of areas with heavy backsliding could significantly alter the
overall economies of the scheme at any particular time and call into
question many of the assumptions on which it was based. Backsliding
was consequently a very serious problem, which had to be tackled and
solved unless the future of rural electrification was to be put in jeopardy.
To the young Rural Area Engineer, this situation was as unexpected
as it was difficult of solution. While courses in the engineering schools

131
The Quiet Revolution

had no doubt covered adequately the technological problems of electric-


ity distribution, associated problems based on human nature or human
reactions to innovation were unlikely to have been envisaged, still less
treated. A totally new dimension was thus added to his job — the
development and exercise of his human relations skills. This applied
even more so to the Area Organiser who was in the front line in this
battle, calling, as his job required, on every householder in the area and
arguing, persuading, cajoling him or, less often, her into reconsidering
their withdrawal.
It was first necessary to pinpoint the cause or causes of the backsliding.
Going back on one’s word had never been a characteristic of the Irish
rural dweller: the commerce of fairs and markets depended on the
tradition that a man’s word was his bond. Yet the phenomenon of
backsliding was widespread in the first decade or so of the scheme. The
situation improved with time but in the late forties and early fifties many
Rural Area Engineers were aghast when the time came to pack up and
move on, to find that they had connected only a fraction of the number
of consumers expected and budgeted for.
In Easky, Co. Sligo, for example, at the end of the construction period
in May 1948 only 210 out of an expected 518 had been connected
(although by the following December this had risen to 345, which was
still only two-thirds of the original number who had signed up). In Tara
Rural Area, Co. Meath, in 1950 the final report of the RAE recorded
seventy-seven backsliders out of an original acceptance of 451 house-
holders. In this case twenty new consumers took supply. The report for
Dromohane, Co. Cork, recorded fifty-seven backsliders, mostly cottages
and small farms, out of an original figure of 320. It also notes that twenty
householders who had not signed up originally had now joined the
scheme and that these were the larger type of farmer. The RAE in Grange,
Co. Sligo, noted seventy-three backsliders out of an original figure of
460 and (here is a clue) identified them as mostly old people or small
farmers who had originally signed ‘in a burst of enthusiasm’. In this area,
however, the score was balanced somewhat by fifty-six extra consumers.
On the other hand, considerable gains in the number of consumers
connected were achieved in some areas. In the Dundrum area in County
Tipperary, while twenty-six householders out of the original 340 backslid,
sixty-five new consumers were obtained. The closing report refers to one
backslider who ‘had built his tiny bungalow on a raft which he rolled
away out of the area leaving behind him only a number on the map, a
signed application form and a most indignant Area Organiser’. In many
areas the number of ‘new’ consumers connected up outweighed the
backsliders, although in most of these cases an intensive redesign of the

132
Whadd ya mean, you've changed your mind ?

133
The Quiet Revolution

networks was required to minimise the extra capital required. Even


when there was heavy backsliding it was frequently expected that this
would only be temporary. The report from Nobber, Co. Meath, where
at the completion of construction work there were ninety backsliders,
observes that ‘in many of the cases, only the head of the house had a
serious objection and was being subjected to considerable pressure from
other members of the family’.
In areas adjoining large towns and cities it was to be expected that
backsliding would be minimal and this was the case in Monkstown in
Cork where the consumers connected increased from 502 to 676, giving
a return on capital of 11.4% (‘the highest in the country’). In Cloughran
adjoining Dublin Airport the figure went from 165 to 294. However,
equally dramatic increases were found in unexpected locations. In the
Achill area, while there were seventy-seven ‘completely recalcitrant’
backsliders, which caused three others to become uneconomic, ten
houses that had become vacant and one which had been demolished by
lightning, 244 new consumers were obtained to give a final figure of 804,
‘due’, in the words of the final report, ‘to the unrelenting efforts of the
Area Organiser, Phil McGovern, who maintained his efforts where men
of lesser courage and industry would have admitted defeat’. In Ballina-
cargy, Co. Westmeath there were ten backsliders out of 182 original
acceptances but fifty-four additional acceptances were secured. The
backslider figure ‘would have been much higher but for the efforts of
Rev Fr Kearney, CC, in bringing in the most recalcitrant [that word
again!] of the refusals’. He was tireless in his efforts and his assistance is
summarised in the final report, ‘I have no hesitation in acclaiming him
as the best Ao of all time’.
West Donegal, one of the least prosperous parts of the country in 1952
might be forgiven if the number of backsliders was higher than normal.
However, to quote from the closing report for the Kilcar area: ‘The
original total acceptances was 223 and the final total, 323... . There was
only one backslider and even this one had been connected up since the
job was completed. An amazing characteristic of the area was the
enthusiastic way in which the householders took to the idea of rural
electrification although the vast majority of them had little idea of it
beforehand. This is particularly surprising when we learn that farming as
such is almost non-existent in the locality, there being an overall average
of approximately 6 acres of arable land per holding... .’ This contrasts
with the situation in the opposite side of the county: in the comparatively
prosperous area of Manorcunningham there was a net loss of 137
consumers, with a reduction in fixed charge revenue of over £700 per
annum. The figure included 132 backsliders, rendering a further thirty-

134
Problems

three houses ‘uneconomic’; fourteen houses had become vacant since


the canvass and forty-two new consumers were obtained.
One simple reason for backsliding stands out: an over-enthusiastic
canvass by the local committee. The greatest proportion by far of
backsliders came from the categories of small farmers and agricultural
labourer. These households very often operated on a semi-subsistence
economy where regular payments of cash for services were not a tradition
and where income was either very low, very irregular, or both. Many
then regarded electricity as an unwarranted luxury and recoiled from
the concept of paying out a regular fixed sum irrespective of the use they
made of the service, which in many cases would be confined to winter
illumination. The problem facing the local committees was that in many
areas these reluctant householders comprised a high proportion of the
total and unless they were persuaded to sign the relevant application
forms the chances of early selection of the area for electrification were
poor. It is little wonder then that pressures of all sorts were exerted to
bring the acceptance figure up and in very many cases it was a case of
signing ‘to bring the electricity to the parish’. When, with the arrival of
the construction crew, the moment of truth also arrived, necessitating
an investment in house wiring, many of the reluctant signatories decided
to opt out. Much time and effort by Ess staff and by local helpers were
put into trying to recover the situation.
Not all backsliders fell into the above category. There were many
reasons for refusal or delay in accepting supply; in the case of public
buildings it might be a simple case of bureaucratic delay in wiring the
building. In this respect the Kilkenny Journal of 14 September 1950
takes the Government itself to task. The Gowran correspondent writes:

Those in high places have not spared their energies in exhorting the
country people to avail themselves of the wonderful benefits of Rural
Electrification and urging them to make no delay in installing electricity,
yet the only two institutions in Gowran which come directly under the
Government — the Railway Station and the Garda Barracks — are left
without the ‘wonderful benefits’.
I don’t know what conditions prevail in either system but certainly the
example is not very inspiring and speaks poorly for the sincerity of the
Government push behind Rural Electrification.

In the case of the 132 backsliders in Manorcunningham an analysis


was given by the RAE. Included were twenty-seven farms, seventy-nine
tenants’ houses, eighteen small owner-dwellings, two churches, one
school, three shops, and one hall. Of the twenty-seven farmers, two

135
The Quiet Revolution

refused supply, ‘so as to avoid poles on their land’, but the remainder
gave no decisive reason. Of the seventy-nine tenants, thirty-seven
refused because the landlords would not wire the houses and the remain-
der were occupied by labourers who felt they could not meet the expense.
This latter reaction was echoed in 1953 in the Kildangan area in County
Kildare, which included in its total of 509 premises 196 labourers cottages
and a further one hundred or more houses of similar or smaller size. In
spite of the fact that Kildare County Council wired the cottages, to quote
the closing report, ‘most of the tenants were not at all enthusiastic about
supply and the construction crew were faced with the appalling total of
106 backsliders’. The report goes on: ‘It was quite clear that the area has
been subjected to a considerable amount of local pressure prior to and
during the canvass and as many of the residents are only in part-time or
seasonal employment it is not surprising that the results were as bad as
they actually proved to be’.
In Rathoe Area, Co. Carlow, in 1951, where there were sixty-eight
backsliders against a gain of sixteen extra consumers, a detailed analysis
of the former was carried out by the RAE. The results were interesting:

No reason given 28)


Occupier in poor financial state 19
Houses had fallen into bad repair 8
Houses had become vacant since canvass 6
Occupier had died since canvass =
Board of Works refused to wire Garda barracks 2
‘Economic’ consumers had become ‘uneconomic’
due to ‘backsliding’ 7

Total: 68

Another survey of the unconnected householders in an area in the


midlands that had been developed for some time gave some interesting
results:

Waiting to see how charge would work


out with neighbours 70%
Rendered uneconomic due to neighbours
refusing connection 20%
Originally refusing because of illness, etc. 10%

Action taken to try to reduce or eliminate backsliding included per-


sonal visits, usually by the Area Organiser, mostly alone but sometimes

136
Problems

accompanied by members of the Committee, public meetings, demon-


strations of equipment, threats to withdraw from the area or to take
down parts of the network. Intensive canvassing of householders who
up to then had refused to join the scheme often had the result of getting
new consumers to offset the losses. In Bekan, Co. Mayo, an area where
the average farm holding was fifteen acres, it appeared at first as if
upwards of a hundred householders were going to drop out. A series of
personal visits by the Area Organiser, however, reduced the figure to
twenty, and in addition 138 extra consumers were gained.
Sometimes the local clergy helped. The RAe in his final report on the
Kildalkey area in County Meath speaks in glowing terms of Fr Kiernan,
the helpful parish priest of Ballivor who also had care of Kildalkey, and
attributes a gain of sixty-two new acceptances to his efforts. In the
Castlegregory area in west Kerry, on the other hand, an appeal was
made to the local clergy, but, to quote the final Area Report, ‘though
they exhorted their people to accept supply, they themselves became
backsliders’. ‘Apart from this,’ the report goes on to say, ‘the position
was not too bad’(!).
The extra consumers obtained as a result of hard work by Reo staff
during the construction period did much to offset the backsliding. An
analysis made of the position at 31 March 1952, when 180 areas had been
completed, showed that the numbers actually connected slightly
exceeded the original number of ‘economic acceptances’ for these
areas — 56,818 connected as against 56,332 original economic accept-
ances. This represented, however, only 60%.-of all premises (93,970) in
these areas and indicates the large number of unconnected premises still
remaining which would have to be tackled under the various ‘Post
Development’ campaigns.

THE ‘GROUND RENT’ CONTROVERSY

The decision in the report of 1944 to recommend a two-part electricity


tariff for rural electrification was not taken without full consideration of
all other options. The main factor in favour of this approach is that the
annual cost of providing electricity can be divided logically into two
parts. One is the annual cost of the distribution network itself which is
made up of the annual capital charges, operation and maintenance costs,
meter reading and billing costs and administrative costs. Once the
network has been built and consumers supplied these costs are incurred
irrespective of the amount of electricity used by the consumers. They
are ‘fixed’ to this extent, but of course must in time respond to changes
in the costs of the various components. The second part is the generation

7]
The Quiet Revolution

and bulk transmission costs which are best expressed in the form of costs
per kWh or per ‘unit’ of electricity, as they depend in the main on the
actual amount of electricity generated and transmitted in bulk to the
main feeding points.
For a developed and stable pattern of electricity usage the total costs
may, without undue difficulty, be charged for solely on a unit basis
without a fixed charge component. For a scheme such as the Irish rural
electrification scheme where the pattern of use would have to be built up
slowly over the years, such an approach could lead to financial disaster
by actually inhibiting the development of the use of electricity from the
start. In this regard the 1944 report quoted the experience of Sweden,
one of the pioneers in rural electrification in the twenties. Much of the
distribution was left to autonomous local rural distribution societies
which in an effort to meet the immediate expressed desires of their
members failed to make proper provision for the long term.
It was the study of such cases which convinced the Ess of the necessity
of using a two-part electricity tariff system. Without this system it is
doubtful if the rural electrification scheme would ever have got off the
ground. The fixed portion of the charge based on the floor area of the
house and out-offices quickly became known as the ‘ground rent’, and
among rural dwellers and rural politicians was for many years to be the
most hotly debated aspect of the scheme. To consumers who had been
brought up to pay for their light and heat more or less as they used it, it
appeared unreasonable to require them to commit themselves to paying
a substantial annual sum merely for the provision of supply. They had
no objection to paying the unit charge for the electricity actually con-
sumed and indeed would not have objected, in the short run at least, to
a higher unit rate if this would eliminate the fixed charge, feeling no
doubt that this would allow them more control of their expenditure. The
fact that the annual fixed charge (usually from £6 to £9) was divided into
six equal two-monthly instalments made things worse. Rural consumers
foresaw themselves paying the fixed charge in summer, when they were
using little if any electricity.
The farming community had only recently won its freedom from the
clutches of rack-renting landlords. Now its members saw themselves
faced for the indefinite future with a new series of regular and substantial
payments to the EsB based on the size of their houses and outbuildings
and irrespective of how much electricity they used. There was only one
term they could apply, ‘ground rent’. It made no difference how often
or how emphatically the Ess officials explained that it was a charge to
meet the annual network costs, and that these costs did not depend on
the amount of current used and were best met in this manner. Having so

138
Problems

met these costs the actual costs of current itself could be kept low. It
could not therefore by any criterion be called a ground rent. It was all to
no avail. The term stuck, particularly in the west.
At times, indeed, this ‘ground rent’ was directly identified with the
land rent paid to the old landlords and proposals to increase it brought
similar reaction. Towards the end of 1960 the EsB, against a background
of rapidly rising costs introduced a 10% increase in fixed charges, the
second 10% increase since 1946. The reaction country-wide was vigor-
ous, with protest after protest being made. That it was a strongly
emotional issue was shown by the report in the Western People of 11
March 1961. A mass meeting of five hundred farmers in Crossmolina
passed a resolution, which was sent to the Chairman of the EsB by the
Honorary Secretary of the National Farmers’ Association, North Mayo
Regional Executive:

The farmers of North Mayo protest against the raising of the ground rent
by the EsB and the increased charge for electricity. The increase in the
ground rent has brought the small farmer from Donegal to Kerry back to
the days of the Clanricardes and the Boycotts. It is unchristian that the
lowest wage earners in Europe (see Farm Survey 1957) should have to pay
for the most expensive electricity west of the Iron Curtain. The £3,000 a
year man in the city with a seven-roomed house has a cheaper ground rent
and cheaper electricity than the cottage in Erris or Connemara. To rob the
poor to pay the rich seems to be the new policy of your Board. It is not
surprising that 252 of our homes have disappeared from the Western
Seaboard. What the crowbar brigade failed to do the Ess is helping to
accomplish. We, the farmers of North Mayo pledge ourselves to use every
means in our power to fight this injustice.

A similar resolution was forwarded to the Taoiseach.


In view of the fact that one of the primary objects of the Rural
Electrification Scheme was to prevent rural homes disappearing and that
Irish electricity was by no means the most expensive, the Chairman
considered that the meeting had over-reacted somewhat to the situation.
In a very detailed reply, made the following points.

The scheme was heavily subsidised by both the taxpayer and the [urban]
electricity consumer.
Escalating costs had made increases in charges unavoidable. These had
been kept as low as possible. The average fixed charge increase was only
3d. per week.

139
The Quiet Revolution

A reduction from eighty to sixty units, which had been simultaneously


introduced in the size of the initial high-priced blocks of units, went far
to mitigate the increase in the fixed charge. For consumers using, say,
between ten and twenty units a week the average overall increase would
be less than 1d. a week, and even for consumers using fifty units a week
the increase would only amount to about 23d. on average.

These increases scarcely warranted comparing the EsB with Clanricarde


or Captain Boycott, but the reactions of these western farmers did
indicate their emotional identification of the EsB ‘ground rent’ with the
exploitation of their forebears by landlords and their agents. Their
determination to resist collectively echoed the heady days of the Land
League and the Plan of Campaign.
Even the efforts of the EsB to emphasise the term ‘fixed charge’ instead
of ‘ground rent’ backfired: if it was a ‘fixed’ charge it could never be
altered. A letter from the Secretary of Muintir na Tire Branch, Glinsk,
Ballymoe, Co. Galway, dated 11 January 1957 (following the first 10%
increase in the fixed charge) reads:

On behalf of the Glinsk Guild of M. na Tire, I am requested to write to


you as regards the increase in the ground rent. In this area we were led to
believe that the units may be raised but that the ground rent would never
be increased. Therefore if you insist on the increase, we request that the
whole area be switched off.

There was a rather more lengthy protest by another gentleman, the


central paragraph of which deserves quotation:

Now Sir, the users that you have got into your net or the Farming Geese,
as I might call them, have a few old manure bags to hold the feathers and
I'm blowed if they will let you take the goose. They are looking for a
reduction in the G R charges and unless they get it they will get out the
meters. The EsB will get no revenue and the poles and wire may remain a
memorial to the EsB. It would not pay them to take them down. Think it
over — a half loaf is better than no bread. My rates have gone up £10
within the last 15 years. Add £8.9s. EsB charges and the loaf 1s.1d., butter
4s.4d. You can call it government by the people or whatever you like, but
the people won’t stand for it much longer. They have borne the thing too
long... .

Similar protests arrived from Ica guilds and NFA branches, all to the
effect that it was understood that the fixed charge meant just that and
would never be increased. Some claimed that a guarantee was given. It

140
Problems

is worth pointing out again in these days of galloping inflation, that the
amount in question averaged about 3d. or slightly over 1p per house per
week.
There was another problem regarding the basing of the fixed charge
on the floor area of the dwelling house and, at a lower rate, of the out-
offices. Where estates had been broken up by the Land Commission,
one of the incoming owners usually got the original large house and
out-offices with his land allocation, while the other owners had new,
smaller houses and out-offices built, more matched to the size of their
land divisions. While the farm size and general income and electricity
requirements might be the same, the occupant of the large house had to
pay a higher fixed charge than his neighbour. A typical case from County
Meath was raised by the nra in 1958. One farmer had fifty-eight acres,
as part of a five hundred acre farm which had been divided. He had got
the original house and out-offices for which his two-monthly Ess fixed
charge was £2.12s.6d. One of his neighbours had 150 acres, but because
of a smaller house his fixed charge was only £1.7s.3d. Another neighbour
with fifty-eight acres was paying only £1.3s.0d.
Another anomaly created by fixing the charge on the floor area was in
the case of Protestant clergymen in rural parishes. They normally had to
live with their families in huge rectories and manses, inherited from
more spacious days. Their salaries were extremely low and their elec-
tricity demand was usually quite modest. Nevertheless, they had very
high fixed charges, much higher than most of their neighbours.
Unfortunately it was not possible to devise a formula which would
resolve such anomalies without creating even more. However, as infla-
tion took hold and the cost of the unit escalated, and as electricity
consumption increased, the contribution made by the fixed charge to the
total electricity bill steadily got smaller and eventually ceased to be a
major bone of contention.

141
CHAPTER TWELVE

Switching On

IN EVERY AREA, the high point of the scheme was the ‘switch-on’ of the
first consumers. Later, when up to one hundred areas per year were
being connected, this had become almost a routine and merited only a
few paragraphs in the local paper. In the early years, however, the
connection of a remote parish captured the imagination of the media
and very full coverage was given, not only in local but in the national
newspapers and indeed on the radio. REO, appreciating the value of such
publicity (particularly in the persuasion of ‘doubtfuls’ and ‘refusals’ to
take supply) co-operated enthusiastically with the local people in mount-
ing a ceremony which would do justice to this important milestone in the
life of the parish. Prior negotiations with the local authority had usually
resulted in street lights being approved and erected. Traders had installed
shop-window lighting and all the houses in the village had their internal
wiring completed, with all bulbs installed ready for the great moment.
The typical scenario was a gathering of all the inhabitants in the village
hall. On the stage was mounted a large switch, around which gathered
leaders of the community: the local clergy, TDs, county councillors and
one or two senior representatives of the Ess. If the occasion was consid-
ered to be of sufficient importance, a government minister was present.
It was usually arranged that the switch on the stage simultaneously
switched on all the village houses and street lights, as well as the lights
in the hall. Where this was not feasible, a signal to strategically placed
operators achieved the same result. In the hall itself a number of portable
high-powered lighting battens was usually installed for the occasion to
reinforce the conventional installation. Speeches were made, paying
tribute to the organising committee and to the EsB staff. The historic
importance of the occasion was noted and hopes were expressed that the
coming of the light was symbolic of the dawn of a new era of enlighten-
ment and prosperity for the community. Almost invariably the opening
verses of the Gospel of St John were quoted ‘... and the light shineth in

142
Switching On

the darkness and the darkness hath not overcome it...’. A blessing was
given, frequently, especially in the northern counties, jointly by the
clergy of the different denominations, and the guest of honour was then
invited to press the switch. One could anticipate a gasp of delight and a
thunderous applause from the assembled crowd as the brilliant lighting
took over from the existing candles and paraffin lamps. The gloomy hall
was lighted as if by the noonday sun. Outside, the new public lights lit
up the village street, while the brightly-lit shop windows added to the
general gaiety. Far out into the countryside, bright pinpoints of light
commenced to twinkle like stars against the blackness of the night as the
rural ‘spurs’ were switched on. Heroes of the hour were the members of
the EsB work crews, many of them local lads who had put in long hours
and extra effort to ensure that supply was on time. Frequently the local
committee invited the whole crew as guests to a meal and a dance.
Each ‘switch-on’ ceremony had its own particular flavour. The Dun-
garvan Leader of 28 August 1948 reports on the switching on of Ballyduff:

Ballyduff was ‘lit-up’ in every sense on the night of Friday 20 August when
the great ‘switch-on’ ceremony of Rural Electrification was performed.
The ceremony was timed to start at 9.30pm and for some hours before
that it was obvious that the people of Ballyduff and surrounding districts
intended to make it a great night and one to be remembered. People
arrived on foot and by car from all parts of the district and from Lismore,
Fermoy, Tallow, Conra and Araglin. The greatest possible interest was
taken in this, the culmination of many months of organisation by the
Ballyduff Guild of Muintir na Tire and hard work by the staff of the Ess.
One old resident stated that he never saw a night of more spontaneous
enjoyment. Everyone was happy and realised what a great night it was for
the district when light and power, which is going to revolutionise country
life, was made available.

The switch-on in Bansha was reported in the pages of REO News in


June 1948:

Bansha 24 May: Rev Fr J. M. Hayes, P.P. Bansha, switched in the village


of Bansha from a platform on the Main Street. In spite of heavy rain,
there was a large attendance of people from the two parishes of Bansha
and Knockmoyler, accompanied by the Bansha Fife and Drum Band.
Afterwards the local Guild of Muintir na Tire entertained the REO con-
struction crew to supper in the schoolhouse. Members of the crew
contributed to the musical programme which followed. Speaking at the
ceremony, Fr Hayes emphasised that ‘... rural electrification ... is more

143
The Quiet Revolution

than an amenity — it is a revolution which will sweep away inferiority


complexes’.

Again, the same issue of REO News reported the switching on of


Abbeyshrule:

Abbeyshrule 27 May: As no suitable hall was available, the local committee


provided a large marquee for the occasion. Over 1,000 people were present
to see the Minister for Justice, General Sean MacEoin, switch on. A
demonstration [of electrical appliances and their use] followed to what has
been our largest audience to date. The subsequent supper was prepared
by members of the Irish Countrywomen’s Association in a temporary
electric kitchen erected in our demonstration van.

The Longford News of 29 May 1948 reported the proceedings at length,


especially the speech of General MacEoin:

It is now 25 years since the idea of the Shannon Scheme was first mooted
by a youngster who is today a Minister — Mr McGilligan. There were
misgivings by well-intentioned people at the time and it was even called a
‘White Elephant’. To-day, we know the difference. We could do with five
or six such white elephants. Abbeyshrule to-day is the envy of all
Ireland... .’ Preparing to switch on for the first time, the Minister said
amidst a hushed silence ‘may God give the Light of Heaven as I am about
to give you light now’.
All around became instantly illuminated and the applause lasted several
minutes.
An EsB Official then demonstrated a number of machines... . His talk and
illustrations were followed with keen interest by a crowd of several
hundreds — there were a few idiotic interruptions by a local wit — or half
wit — and an occasional echo of ‘the loud laugh that ...’. It was a pity that
the demonstrator hadn’t got an electric chair handy!

In a lengthy article headed ‘Kilmessan’s Night of Jubilation’, the


Meath Chronicle of 10 December 1949 gave full coverage to four separate
functions held on the same evening to mark the coming of electricity to
the locality. The first was in the Temperance Hall where Liam Cosgrave,
Parliamentary Secretary to the Taoiseach officially switched on the
supply:

The hall was lighted by candles, lanterns and paraffin lamps and truth to
tell it was far from ineffective. When Mr Cosgrave touched the switch
which illuminated the hall with the new current, there was an outburst of
applause which could certainly have been heard on Tara.

144
Above: The ‘switch-on’ — the large metal-clad switch used for such
events was ‘liberally bespattered with rust spots from frequent
former applications of holy water’. Below: Switching on the
100,000th consumer, March 1954 — the Ballinamult Creamery, Co.
Waterford, floodlit for the occasion.

rciry FERNS

145
The Quiet Revolution

Simultaneously, the floodlight system installed outside operated on the


hall, church, parochial house and the factory building and the whole scene
was one to be remembered. An illuminated address commemorating the
event was presented to the Minister for Industry and Commerce through
Mr Cosgrave... .
The party then left for the newly opened Mineral Water Factory and after
an inspection proceeded to the church for Benediction. Later, over one
hundred guests sat down to a dinner in the Station Hotel where various
speeches were made and toasts drunk to the bright future which would
follow the coming of electricity to the village and its surrounding rural
area.

Local merchants were among the first to appreciate and avail of the
new power. In Tulsk, Co. Roscommon, one of the first of the newly
developed fluorescent lights made its ‘rural’ bow in a small thatched
public-house on the night of the switch-on in summer 1948. In September
1950 ice-cream was on sale in the village of Templetouhy, Co. Tipperary,
one hour after the first consumer had been connected.
In the December 1951 issue of REO News there is a long poem of six
stanzas entitled ‘The Lighting of Scariff by a member of the crew, Oscar
Hannon, who sang it to the air of “The Mountains of Mourne’ at the
switch-on celebrations. At least one stanza deserves quotation as illus-
trating the unsophisticated joy of such occasions:

Oh Mary now Scariff’s a wonderful sight


With the people all getting the ‘new fashioned light’
After watching and waiting for many a year
At least they’ve consented to send it out here.
An badly ’twas needed as you must well know
For in Scariff at night-time wherever you go
There are dangerous angles all over the place
Where many a craythur was stretched on his face.

Such ‘poems’ were not unusual. At the switch-on of Killanny,


Co. Louth, a similar tribute to “The Boys from the Es’ was rendered by
a local schoolteacher, the last stanza of which went:

Transformers they of rural life —


A blessing on their head
They give you heat, they give your power,
They banish darkness dread
They'll ‘fluoresce’ the old land
From Finn to lovely Lee
They know their job and do it well
The boys from the Ess.

146
Switching On

In some areas local pride was not content with a single ceremony. In
the Ballinlough area on the Roscommon/Mayo border there were three
separate switch-ons, the first was held in Ballinlough itself as befitted the
importance of the principal centre. This was quickly followed by a second
in Cloonfad and a third in Garranlahan, each occasion being marked by
a supper and dance. Neither were food and drink stinted. In Drumlish,
‘the local committee provided an excellent supper which included fresh
salmon and champagne, followed by an excellent musical programme’.
Champagne was also produced at the switch-on banquet in Woodford,
Co. Galway, where afterwards two celebration dances were held simul-
taneously — one the rather sedate official function in the Parochial Hall,
and the second a more boisterous affair organised by the crew members
themselves in the commercial ballroom to which a half-barrel of Guinness
had been rolled up the village street, the gift of a grateful publican.
In many northern areas where the proportion of Catholics and Prot-
estants approached something like parity, the blessing and switching on
of the supply were carried out jointly by the pastors of both denomina-
tions. In the predominantly Catholic south, however, the task was
generally given to the parish priest, with the clergy of other denomina-
tions attending as guests of honour. In a midlands village, obviously not
notable for its ecumenical leadership, the elderly parish priest was
reputed to have settled an old score with the local C. of I. rector during
the blessing ceremony. Having duly sprinkled the switch with holy water,
he turned and also sprinkled the audience, with an extra special douche
for the rector, who, denied a place on the platform by the parish priest,
had been placed by the committee in a seat of honour in the middle of
the front row.
The engineer in charge of the Clifden area recollected the switch-on
ceremony in the local hall. In the centre of the stage was the ubiquitous
large metal-clad switch which had given trojan service at many a
switch-on and whose cover was ‘liberally bespattered with rust spots
from frequent former applications of Holy Water’. On the platform were
both the parish priest and the C. of I. rector, but as representative of the
majority faith the task of blessing and switching-on fell to the priest. All
went well with the ceremony until the moment came to pull the switch.
Amid a hush of expectancy, he operated the lever. Nothing happened.
There was a shocked silence for some seconds and then a loud whisper
came from the back of the hall: ‘Let Rev Mr Fairbrother have a go!’ As
the tension broke and the audience exploded into laughter the reverend
gentleman obliged by taking a bow. Immediately afterwards the hall
blazed into light amid tremendous applause: a minor electrical fault had
been rectified by an alert electrician backstage.

147
The Quiet Revolution

There were some very special switch-ons to mark special milestones.


In February 1952 the fifth anniversary of the connection of the first rural
consumer was celebrated when the 55,000th consumer, the parish hall in
Kilsaran, Co. Louth, was switched on by the parish priest, Fr McEvoy,
in the presence of a platform party which included Frank Aiken, then
Minister for External Affairs, and the esp Chairman, R. F. Browne. An
exhibition of electrical equipment was mounted by REO and local traders.
This included, for the first time, a television set as County Louth was on
the fringe of the service area of the BBC’s Northern Ireland transmitter.
It is recorded in the area report, however, that ‘unfortunately, or perhaps
fortunately, it was a bad night for reception or he [the TV dealer] would
have stolen the show’.
Even this event was eclipsed by the switching on of the 100,000th rural
consumer, the Ballinamult Creamery of the Knockmeal Co-Operative
Society in County Waterford, by the Minister for Industry and Com-
merce, Sea4n Lemass, on 1 March 1954. On this occasion, as well as a
large contingent of EsB chiefs, the attendance included all the local TDs,
senators, county councillors and other public representatives and — to
quote REO News - ‘all the residents within miles of Ballinamult’.
Television continued to push itself to the fore as ‘with the co-operation
of Pye (Ireland) Ltd. the complete proceedings were televised with
receivers placed to show the ceremony to overflow crowds both inside
and outside the building’.
In May of the same year (1954) a very special ceremony took place in
Rosmuc, Co. Galway, when the Taoiseach, Eamonn de Valera, attended
the switch-on of the area which took place outside the former cottage of
his old friend and comrade-in-arms, P. H. Pearse. It was an open-air
occasion and a huge throng was present to watch the Taoiseach press the
switch, which he hoped was symbolic of the better future for the country
for which Pearse had dreamed, worked and died. It is notable that this
was the only occasion on which de Valera as Taoiseach participated in
such a ceremony.
Perhaps the most original approach to a switch-on was in Mountgordon
in County Galway. At the request of Fr Hennelly, PP, a film of the
event was made by the members of the camera club in Castlebar. Shots
were also taken of the area crew at work during various stages of
construction. The edited film was sent to America to play an important
part in an appeal for funds for the building of a new church at Park.
Many hundreds of other switch-ons took place which did not make
national headlines, but which were just as momentous for the areas
concerned.
A continually recurring question, raised especially by members of the
older generation, was whether the electricity now lighting up their

148
Switching On

locality really came from the Shannon Scheme. When assured that this
was substantially so, although of course there were now other stations
also contributing, their pride and joy were obvious. For them the
Shannon Scheme had been one of the first great manifestations of an
independent nation and on this night they were at last partaking of its
fruits in their own community.
The construction crew also shared in the general euphoria of the
occasion as an extract from an article in the final issue of REO News
points out:

Not all will perhaps admit it, but there were very few Rural Area Engineers
who did not feel a glow of pride and satisfaction as they surveyed their
first group of consumers after they had been switched on. This pride was
also evident in the faces of the crew as they gathered, as was general in
those early days, at the local church, hall or school and watched a local
celebrity press a switch and listened to the gasps of delight as night was
made into day; or as they strolled down a village street under the light of
the new street lamps; or listened to the purr of a newly-installed electric
motor carrying out its task of relieving some of the toil and drudgery of
the country side.
Then, the difficulties and squabbles receded into the background as the
worthwhile nature of their work came to the fore. Theirs was the satisfac-
tion of knowing that in their own way they were leaving this part of the
world a better place than they had found it.

149
CHAPTER THIRTEEN

Threading the Way

THE FIRST POLE OF THE RURAL ELECTRIFICATION SCHEME was erected at


Kilsallaghan, Co. Dublin, on 5 November 1946. It was a dull autumn
day. There was very little ceremony. Present were W. F. Roe, P. J.
Dowling and a small gathering of REO staff. It was to be a long journey
to the erection of the millionth pole and the connection to the national
electricity network of well over 400,000 rural households in 792 areas.
Government instructions were that one area in each county should be
developed initially, so as to spread the benefits as widely as possible.
This created some problems, as some counties were far better organised
than others to initiate the preliminary canvassing necessary in order to
have areas considered for selection.
The second area to be selected was Patrickswell, Co. Limerick, and
work commenced in February 1947. This was followed in May by
Inniscarra (Co. Cork), Tinryland (Co. Carlow), Multyfarnham
(Co. Westmeath) and Easky (Co. Sligo). In July, Pollroan
(Co. Kilkenny) was commenced, followed in September by Ballymacel-
ligot (Co. Kerry), Ardrahan (Co. Galway) and Kilbride (Co. Wicklow).
In his monthly progress report for September 1947 the Engineer-in-
Charge informed the Board that the first area — Kilsallaghan, Co. Dublin
— was now completed and that construction was proceeding in nine other
areas. In three further areas survey was well advanced and was comm-
encing in seven others. Counties in which no area had yet been selected
were Cavan, Clare, Donegal, Monaghan, Offaly, Roscommon and Long-
ford. The year 1948 saw work starting in the remaining counties:
January: Kill (Co. Kildare), Murrisk (Co. Mayo)
February: Ballyduff (Co. Waterford), Bansha (Co. Tipperary), Carrig-
allen (Co. Leitrim), Clonaslee (Co. Laois)
March: Abbeyshrule (Co. Longford), Lusmagh (Co. Offaly)
April: Quin (Co. Clare), Termonfeckin (Co. Louth)
May: Mantua (Co. Roscommon)

150
Threading the Way

June: Shelbourne (Co. Wexford)


August: Julianstown (Co. Meath), Rossnowlagh (Co. Donegal),
Tydavnet (Co. Monaghan)
December: Gowna (Co. Cavan)
In most Districts, the first areas selected were used as training grounds
for crews destined for work in later areas. For example, in Sligo District
the structure of the crew that commenced work in Easky in May 1947
was developed in such a way that it could later split into two trained
crews, one of which moved to Carrigallen in Leitrim and the other to
Mantua in Co. Roscommon. As work proceeded in each area, the design
and pegging of pole positions and other preliminary work went ahead in
the area next on the list so that on changeover from a completed to a
new area, the crew was able to swing into action with the minimum
interruption in work flow.
A comparison between 1947/48 and 1948/49 shows the rapid acceler-
ation in production achieved in the course of one year:
1947/48 1948/49
Miles of line built in year 760 1,720
Number of poles erected 15,986 32,002
Number of consumers
connected for year MPP 9355

By 31 March 1949 the Ess was able to report to the government that,
despite the many difficulties in obtaining materials, the scheme was well
under way. 49,000 poles had been erected and conductor strung on 2,490
miles of line. 11,500 rural households had been connected to the national
electricity network in thirty-seven completed rural areas. Owing to
escalation in material prices, the costs had been higher than originally
anticipated and in the esp Annual Report for the year 1948/49 the
adequacy of the 50% capital subsidy was questioned: ‘... to allow the
Scheme to proceed the Board obtained an undertaking that any defi-
ciency during the first two years working would be covered by an
additional subsidy from the Transition Development Fund. Negotiations
were in progress to ascertain the additional sum required. ’ This emphasis
on the inadequacy of the subsidy was repeated in the 1950 and 1951
Annual Reports. Nevertheless, there was no slackening of pace. Indeed,
for the years ending 31 March 1951 to 1953, forty-nine areas were
completed in each year. For those areas at the bottom of the queue,
however, this was not fast enough. Pressure grew through their local
representatives and in turn from the government to the Esp for an
acceleration in the pace of working. In June 1953 it was decided to

151
The Quiet Revolution

increase the rate of development by 50%. Additional material was


ordered, extra staff recruited and trained and the design and survey of
new areas undertaken.
In its Annual Report for the year 1953/54, the Board was able to
record that development had now reached the rate of seventy-five areas
per year. It further indicated that a new target rate of one hundred areas
a year had been set for the year 1955/56. Ninety-nine areas were in fact
completed. It would be of interest to pause at this point, 31 March 1956,
to review the progress in the first nine years. 499,000 poles had been
erected, 43,000 kilometres of line strung and 163,000 consumers con-
nected in 463 completed areas. Thus, after a difficult start, over one half
of the proposed scheme had been completed. A further cause for
satisfaction was that, after a slow start, the connection rate in developed
areas now averaged 67% of all premises, which was almost up to the
White Paper target of 69%. The capital subsidy had been withdrawn by
the Interparty Government in October 1954, but this was not allowed to
retard the pace of development.
Electric lighting as a substitute for the oil lamp was the first and most
widespread use. REO staff vigorously advocated the installation of power
sockets both in the house and in the out-offices to make it possible to
reap the benefits of electrification in other ways. In the early years,
however, this was a very slow process, especially in the case of the less
affluent and less progressive. Major changes in life style could not be
expected to take place overnight. In order to get some idea of progress
in this field, REo conducted a survey during 1953 in a number of areas
that had been connected for some time. Results were encouraging. As
was to be expected, most progress had taken place in the home. Almost
80% had mains radio (the saving on battery-charging costs frequently
went a long way to offsetting the fixed charge), 64% had electric irons,
44% had electric kettles and 25% had installed electric cookers. Washing
machines and electric wash-boilers tied at 17%. Progress in installing
piped water, one of the greatest benefits hoped for with the coming of
electricity, had, however, been disappointing — only 11% had water
pumps installed.
Out in the farmyard, development had been slow. While 25% had
installed lights in the yard and in the out-offices to extend the winter
working day, the number of power sockets installed and the out-dwellings
was very small. About 9% of farmers were using electric motors for such
chores as root-pulping and the same percentage had installed infra-red
lamps for pig and chicken brooding. The electrification of milking was
still in its infancy, with only 6% using electricity for this purpose.
In the case of the first seven areas, which had been connected in 1947

152
Threading the Way

and which thus had had most time to adapt to the new resource, it was
encouraging to note that the average annual electricity consumption per
consumer, which in 1948/49 had been 509 units, had reached 1,145 by
1953/54. This average figure, however, concealed a wide variation
between areas, ranging from 1,611 units per consumer for Patrickswell,
Co. Limerick, to 625 for Ballymacelligot, Co. Kerry.
The peak of construction activity was reached in the years 1954/55 to
1956/57. For the year ended 31 March 1955, seventy-five rural areas were
completed and 27,316 new consumers connected. In the following year,
this performance was overshadowed when ninety-nine areas were com-
pleted and 34,257 new consumers added. In 1956/57 the number of areas
recorded as completed was eighty, but the new consumers added reached
the record total of 34,627. At this stage, almost 200,000 consumers out
of the initial target of 280,000 had been connected at a total cost of
£20,661,000. The rate of capital expenditure had also reached a high
peak: for the two-year period of 1955/56 and 1956/57 it had exceeded
£7.8m and was over 10% of total national capital expenditure. With the
end of the initial phase now in sight, it was necessary to commence a
controlled tapering down of activity to relieve the high demand on scarce
national capital resources and to avoid a serious problem of staff redun-
dancy, which a sudden stop would create. Thus, the number of construc-
tion crews, which had reached forty in the peak years, was gradually
reduced and key personnel were absorbed into other activities. The
number of new areas completed showed a corresponding phasing down,
as seen in table 4.
TABLE 4

New Areas Completed and Consumers Connected


No. of No. of
Year new areas No. to Consumers No. to
ending completed Date connected Date

1958 60 603 PS 215,243


1959 45 648 14,997 230,240
1960 40 688 13,458 243,698
1961 40 728 11,570 255,268
1962 26 754 9,315 264,783
1963 19 Fie) 8,829! 213,012
1964 12 785, 10,820? 284,432
1965 if 792 tiie 296 ,205
1 (including 3,800 ‘post development’)
? (including 7,692 ‘post development’)
> (including 10,776 ‘post development’)

153
The Quiet Revolution

The 792 areas developed under the initial phase covered an area of
23,400 square miles or 86.5% of the country. If the areas of poor land
were deducted (i.e. land with a Poor Law Valuation of 5 shillings an
acre or less, into which the extension of the scheme was not contemplated
in the White Paper) it could be claimed that 98% of the remainder had
been covered at this stage.
It will be seen from the table, however, that a new phase had com-
menced in 1963 — the ‘post development’ phase. This involved the
sending back of crews into already developed areas to connect house-
holders who for various reasons had not accepted supply or who had
been regarded as ‘uneconomic’ during the initial development of their
areas, but who were now pressing for connection. The Electricity (Sup-
ply) Amendment Act 1962 enabled this to be done by permitting the EsB
to extend the benefits of the scheme to areas hitherto classified as
‘uneconomic’ and on improved terms to unconnected houses in devel-
oped areas. Thus, connections under the original phase diminished and
the number of ‘post development’ connections rose as the geographical
spread of the scheme approached completion. By the end of the year
1964/65, at which point supply had been extended to all 792 areas,
296,000 rural consumers had been connected, of which 23,000 had been
supplied under ‘post development’.
While this signalled the completion of the scheme as envisaged in the
White Paper (i.e. 280,000 consumers or 69% of the total dwellings), it
was by no means the end of rural electrification. When the White Paper
was published towards the end of the war years, the target figure of
280,000 or 69% of houses connected could be considered acceptable. By
the 1960s, however, the 31% thereby excluded were in no way prepared
to accept a future without electricity. After the stagnation of the fifties
the economy was beginning to pick up and the benefits of electricity were
now becoming manifest even to the most conservative. Intense pressure
grew for a fresh look at the conditions for subsidised connection of
electricity supply.
A ‘planned post-development’ (ppp) scheme was agreed between the
ESB and the government whereby every householder in the 792 areas
who was still without electricity would be given a further opportunity to
obtain supply on subsidised terms. Between 1965 and 1971, 50,000 extra
consumers were connected under this scheme, so that by 31 March 1971
the number of ‘rural’ consumers stood at 346,000.
Even at this stage, however, when close on 90% of all rural homes
had now been supplied under the scheme, there was no respite for the
ESB. The clamour from the remaining householders, mostly located in
isolated pockets, grew in intensity and the pressure by their local rep-

154
Threading the Way

resentatives on the government and by the government on the ESB was


irresistable. A further post-development scheme was agreed under which
the Board undertook to offer every unconnected householder in the 792
developed areas a last chance of obtaining a subsidised connection. Any
new connections after this would be on a strictly commercial basis,
without any help from government subsidy (See chapter 20 for a more
detailed account of these schemes.) About 60,000 consumers were
connected under this ppp scheme. The Board in its 1975/76 Annual
Report recorded that by 31 March 1976 the total number of consumers
connected under the Rural Electrification Scheme was 405,890, or
around 98%/99% of all rural dwellings. Only a few thousand houses
situated in the most isolated regions and, consequently, most expensive
to service, now remained without supply. The initial concept of a 69%
connection as being all the economy could bear had, with the introduc-
tion of the various post-development schemes, long been thrown out the
window. Now electricity supply, even to the most isolated dwelling,
began to be portrayed as the democratic right of every citizen. While
most of the still unconnected householders were scattered in ones or
twos in remote situations all over the country, in some locations they
were sufficiently grouped to form strong pressure groups. Two examples
were the Black Valley, a very isolated and scattered community of about
forty houses south of the Gap of Dunloe in the Macgillycuddy Reeks
area of County Kerry, and Ballycroy, an almost equally isolated group
in western Mayo. Strong and articulate representations from such groups
persuaded the government to introduce special legislation at the end of
1976 to enable supply to be extended on a subsidised basis to about 1,600
remote dwellings in similar situations. This legislation effectively brought
to an end the long series of extensions to the original Electricity (Supply)
(Amendment) Act of 1944 which had had the effect of transforming the
limited scheme originally envisaged into effectively the complete elec-
trification of the country.

1D)
CHAPTER FOURTEEN

Spreading the Message

IN ORDER THAT THE SCHEME should be a success, both economically


and socially, it was essential that the new consumers should not only be
shown how to make the best use of the new power but should be actively
encouraged to do so. This involved REO in a widespread programme of
education, demonstration and promotion of the many ways in which
electricity could help in the home and on the farm. To the majority,
electrification meant the coming of the ‘light’ and this was of course the
first and prime application in all households; but it had to be demon-
strated that this was not all, and that there were many other ways in
which electricity could be used to advantage.
Tremendous benefits would be reaped at low cost by the provision of
an outside light to enable yard work to be carried out during the hours
of darkness. This could be extended comparatively cheaply and with
great effectiveness to the lighting of the farm sheds and byres, thereby
lengthening the farmer’s working day. Even this took a long time to be
widely adopted.
In the house, the transition from soft but poor illumination by the
traditional oil lamp to the brighter but harder electric light was frequently
difficult to accept. All too often a low-wattage bulb was installed in the
kitchen — the family living-room — both to minimise the contrast with the
old oil lamp and to economise in the use of electricity which still had the
image of an expensive alternative. In an effort to combat this reluctance,
REO gave a free 100 watt bulb to every newly connected consumer for
installation in the kitchen. It was hoped that after a short time the family
would become used to a good level of illumination and would be reluctant
to settle for less. This was effective only to some degree. In many cases
the transition was too traumatic and lower-wattage bulbs, usually 40 or
60 watt, were quickly substituted. Even these were much brighter than
the oil lamp. As late as 1958 in the Mount Delvin Area in County Galway
the number of householders who substituted the smaller bulb was so

156
Spreading the Message

large — about 60% of all — as to prompt the area staff to carry out a
mini-survey as to the reasons. They were told that the 100 watt bulb
was — too hard on the eyes, made the people feel sick or that the bright
light tended to put out the fire. This reflected the widely held belief that
any bright light, especially the sun, if allowed to shine on the traditional
turf fire would tend to extinguish it. This was of course because it masked
the glow otherwise seen in the comparative gloom of the hearthside.
Another but seldom admitted cause of the dissatisfaction with the
high-wattage bulb was that it was merciless in revealing dirt, dust,
cobwebs, defects in the furniture, all crying out for higher standards of
housekeeping.
REO News of March 1959 quotes the Rural Area Engineer in Carrig-
atogher Area in County Tipperary, who found that if he waited for a
few weeks after connection before giving out the free 100 watt bulb, very
few went back to a lower wattage. By then they had got used to the 60
watt level of lighting and so the jump to 100 watt was not so great.
The promotion of high-wattage bulbs for lighting was of course only
scratching the surface of the development problem. Because of the cost
involved, householders tended to confine the internal wiring of their
premises to the minimum and the provision of adequate sockets was a
very low priority.
In the early years the average cost of installation by rural wiring
contractors was from £1 to £2 per point, but even this was considered
beyond their means by many. In the Ballycastle area in north Mayo, the
parish priest of Lackan, Fr Murphy, organised a bulk-buying co-opera-
tive which provided the wiring materials for about twelve shillings per
point, the actual installation being a do-it-yourself job. Indeed much of
the wiring in the more remote areas was on this basis or else was carried
out by the local handyman. The circuits wired, however, were of the
most basic type and did not require any great knowledge of electricity.
Provided standard wiring materials were used it was difficult to make a
mistake. Even so, requests for aid arrived regularly and a letter from a
County Sligo consumer quoted in REO News (September 1957) ran:
‘Dear Sir, Please send me some booklets and leaflets on how to wire
houses and farmyards for the electric current, also send me some leaflets
on how to treat a person for electric shock, write soon, yours truly. . .’.
In the February 1954 issue came a story — surely apocryphal — from the
Kingdom of Kerry where a ‘contractor’ extending the existing wiring in
a rural dwelling was not quite sure if the earth connection was connected
correctly and for safety tested it on the cat — ‘the unfortunate animal
entered the feline Valhalla by the shortest route’. The circuit fuse was

157
The Quiet Revolution

then removed and, to make doubly sure, a cork was screwed into the
fuse base pending the quick arrival of an Ess electrician.
REO promoted the use of electricity on two main fronts — domestic and
farmyard. Much effort went into persuading householders to fit an
adequate number of socket outlets. It was a difficult task and even in
1964, surveys showed that the average number of outlets at two and a
half per premises. By 1974, however, following a decade of relative
affluence, the average had risen to five and a half.
In a typical rural area the promotion of the use of electricity usually
consisted of a series of stages as outlined below.

Preliminary Contact 0 During the official canvass the Area Organiser


who called at each house was able, with the help of suitable leaflets, to
outline many of the benefits electricity could bring.

Direct Approach 0 When the area was selected for development, the
Area Organiser again moved in, visiting every house to check up on
potential consumers. He raised the question of adequate wiring and went
on to discuss suitable appliances. This was the first discussion at which
it was clear that supply could be given and that the householder would
take it.

Display 0 Where possible, the Area Organiser secured a small show-


room or shop window incorporated in the vicinity of the area office,
where prospective consumers could view the appliances. Sometimes, as
in Blackwater, Co. Wexford, the opportunity of a vacant shop was
eagerly grasped for this purpose.

Good example UO The Area Engineer tried to have his ‘switch-on’ and to
get consumers connected as early as possible in the construction period
and the newly-delivered appliances immediately put to work. Those yet
to be connected, or who had not yet decided to purchase appliances,
were thus provided with practical examples of their benefits.

Demonstration 1) At a convenient time during the construction period,


local halls were hired for a demonstration week. This was a joint effort
between. the Area Organiser and a demonstrator from headquarters,
who came with a large van fitted out with a comparative range of
appliances and demonstration equipment. These vans were constructed
so that in addition to carrying equipment for indoor demonstrations,
they could also act as open-air demonstration theatres at country shows
or other such functions. During the demonstration week the exhibition

158
Right: The direct
approach. Below:
Spreading the
message.

Parott tk
COMMER Cle
pa Ni
puRINe gal DG a3

159
The Quiet Revolution

was open during the afternoon, while at night a comprehensive pro-


gramme, including films of domestic and farmyard electrical appliances,
was staged, usually to a packed house. This programme also included
practical electric cooking by demonstrators from the local District head-
quarters, the cooked dishes being shared out among the audience. At
other times during the week, special visits were made by the Area
Organiser and the visiting demonstrator to individual consumers — black-
smiths, small industries, etc. — who had special requirements outside
the normal experience of the Area Organiser.
To combat its high-cost image the actual amount of electricity used in
the various applications was measured by a meter and the cost computed.
Having explained the system of charging for the various blocks of units
consumed (in 1951, for example, the first eighty units per bimensal
period cost 2.8d. per unit; the next 280, 1.3d. per unit and any units
above 360 per period cost 1.05d. per unit), the demonstrator then
indicated the amount of work a unit could do. It would light the 100 watt
bulb for ten hours, do the ironing for a week, wash the clothes for a
month (people were invited to bring bundles of soiled clothes), cook all
meals for one person for one day and boil 18 pints of water. A member
of the audience was invited to check the meter before and after to verify
the claimed figures. In Kilvine near Claremorris in County Mayo a
schoolboy checked that boiling a full six-pint kettle used 0.3 units. When
then asked by the demonstrator how many kettlefuls a unit could boil,
he reflected a moment and in a loud proud voice proclaimed ‘three and
a sup’. He then went on to calculate that this would provide eighty cups
of tea.
These practical demonstrations did much to break down prejudices
regarding the cost of electricity. For farmyard applications the demon-
strators stressed the almost unbelievably low cost of electric motive
power. The 4hp motor used only one unit for four hours working, in
which time it could perform better than the strongest worker. Using
electric motors one unit would grind one hundredweight of meal, pump
1,000 gallons of water from the average well and milk one cow twice a
day for ten days. At one venue the operation of a root pulper by the tiny
hp motor was being demonstrated to a cautious farmer (two tons pulped
for one unit), when the farmer’s young son was observed trying to
persuade his father to invest and simultaneously urging the REO man to
intensify his efforts when Dad appeared to hesitate. Evidently his interest
sprang from the fact that he was responsible for pulping on the farm.
Demonstrations such as these were useful but the real task was to get
consumers actually to purchase and install applicances. The success of a
demonstration was measured by the sales which resulted.

160
The mob ile showrooms

161
The Quiet Revolution

Post-Development O In many areas there were active electrical retailers


~ very often the same men who had wired the houses — who canvassed
for business with the full co-operation of the rzEo staff. The first appliance
to be installed was generally a mains radio, as the saving on the
cost of battery-charging often went a long way to offsetting the fixed
charge. The Esp did not enter into this field, as it considered that it was
best left to the local dealer who was in a position to offer the specialist
service required. It did, however, continue to remain active in the sale
of domestic and farmyard appliances, as it was felt that the private sector
was not equipped to provide the drive or the geographical spread
necessary to ensure sufficiently rapid development. A fleet of well-
equipped sales-vans was established, each van fitted out as a miniature
display. The salesmen were usually drawn from the ranks of Area
Organisers who had shown a flair for this work. An important appliance
in their stock was the electric water pump and they were trained to
answer most of the queries regarding the provision of water on tap. They
also had a stock of domestic equipment which they could demonstrate
in the actual farmhouse kitchen and leave for a few days trial if requested.
The same applied to the farmyard appliances — electric motors, grain
grinders, oat rollers etc. As the milking machine market was served by
specialist firms, REO did not sell these machines, but availed of every
suitable opportunity to demonstrate the merits of machine milking,
particularly at agricultural shows throughout the country.
The numbers of the different appliances sold by the vans in the six-
month period April-September 1953 and 1954 indicate how the market
took off. The order of preference of appliances is shown in brackets.

1953 1954

Irons 160 (2) 479 (1)


Kettles 134 (4) 474 (2)
Cookers 21 (1) 41983)
Washing machines 111 (5) 164 (5)
Wash boilers 67 (6) 86 (6)
Fires SIC) eto ata
Vacuum cleaners 28 (9) 44 (11)
Refrigerators 13. (ie) eon)
Pumps 150 2(3) es sOmta)
Grain grinders 20 (11) 51 (9)
Motors 35, (8) 61 (8)
Welders 25; (10) ~= 305710)

162
Spreading the Message

The mobile salesmen worked closely with the local EsB office and
showrooms, generally using it as a base for restocking and frequently for
completing sales initiated in the farmyard or farmhouse. The local staff
co-operated both in the actual selling and in the after-sales service. On
the 23 July 1960 a wedding anniversary present of a washing machine,
complete with appropriate greeting card and tied up in pink ribbon, was
delivered punctually by the Kells staff to a very delighted woman out the
country. It was from her husband then working in Nicaragua, who had
conducted the whole operation by correspondence. A farm family near
Kenmare, Co. Kerry, purchased an electric cooker which they asked to
have installed in time for a ‘Station’ being held in the house a few days
later’. This was arranged by the Area Electrician with two days to spare.
The family was so pleased that he was invited to the function. To
continue the story in his own words, ‘the dear old lady had forgotten
everything I had told her about the controls due to the excitement. She
came to me in a state of desperation — would I ever show her again.
Too late, I considered, she was beyond understanding anything I said by
this time so I donned an apron, secured pots, pans, eggs, bread for
toasting, etc. and I cooked the breakfast for twenty people, including
two priests, clerks, farmers with big appetites dnd even for the people of
the house (it was a Friday morning — hence the eggs).’
In the dairying areas the creamery was a great meeting place. Each
morning the carts with their churns lined up and gossip was exchanged
while the milk was taken in. This was duly noted by the reo staff and
many early morning creamery demonstrations were held in the open air
with gratifying sales results. A typical example recorded was at Upper-
church, Co. Tipperary, where from 7am to 9am on a sunny summer
morning in 1957 a comprehensive demonstration of equipment was
given, including electric cooking. A breakfast of sausages and tea was
served by the ao. In Lislynn area on the Cavan/Meath border what was
described as an ‘Intimate open-air demonstration’ was held for a small
group of women — who were invited to bring along their washing! While
the washing was being done, the merits of water on tap, electric water-
heating and of course washing machines were discussed.
On occasions, an opportunity for extra publicity for the rural scheme
was perceived, as in the case of An Ras Tailteann, the big national
multi-stage cycle race organised in 1961 by the National Cycling Asso-
ciation of Ireland. One of the large rural demonstration vans, decorated
with a suitable slogan, was employed as a mobile laundry for the
competitors’ clothes, using domestic electric boilers, washing machines
and tumbler dryers. The thousand-mile race was run in eight stages from
Friday to Friday, commencing and finishing in Dublin. The overnight

163
The Quiet Revolution

stops were at Navan, Castlebar, Tuam, Castleisland, Killarney, Clon-


akilty and Wexford. The EsB van accompanied the competitors over the
whole route and at each of the stops it was set up, opened out and up to
five hundred items of soiled clothing — team jerseys, singlets, shorts,
socks, caps'etc. — were washed and dried in full view of the large crowds
and were ready to hand back to the competitors next morning. The
slogan on the van was ‘Electricity Leads in the Race for Cleanliness’.
REO produced an extensive range of literature in the form of leaflets
and booklets. This covered every phase of development, Rates of
Charge, Farmyard Wiring, What a Unit Can Do, How Units Can Help,
Domestic and Farm Appliances, Kitchen Planning, Water Pumping,
Water Heating, Electricity for Poultry, The Electric Grain Dryer, Ani-
mal Food Cooking, Soil Warming. As a back-up to all these efforts, a
continuous series of display advertisements was inserted in all fifty-two
local papers, in the farming weeklies, in six rural and about thirty other
weekly or monthly magazines. While these mentioned domestic equip-
ment, they mostly laid stress on farming applications (Fig. 1), frequently
quoting actual farmers’ experiences. Domestic applications were in the
main covered by weekly advertisements in the four daily and four
evening papers. These were aimed at urban as well as rural domestic
consumers, as by the end of the fifties the difference between the two in
the use of electricity for domestic purposes was beginning to disappear.
Indeed, the arrival of the sixties marked a major break-through for
the promotional efforts of the REO staff. On the broad economic front
the stagnation of the forties and fifties was yielding to a more dynamic
phase. In rural homes the REO promotional drive was now being carried
out in the context of an improving economic situation and a growing
receptivity to innovation. This was reflected in a growing increase in the
use of electricity and a greater readiness to invest in electrical appliances,
particularly in the domestic area.
The gap between urban and rural living standards began to close
rapidly. More new houses were built and old ones modernised. The
traditional dark rural kitchen became transformed, and the drudgery of
rural housekeeping was relieved by the installation of hot and cold
running water and labour-saving appliances. Table 5 traces the growth
of appliance ownership among rural electricity consumers. By 1958, after
a decade of rural electrification during which the national economy
remained in low gear, the only domestic appliances that had been widely
accepted were the low-cost iron and electric kettle. By 1968, however,
electric cooking had moved in, as had the refrigerator, which made it
possible to store perishable food and reduce the frequency of journeys
to the local shopping town. Television, one of the greatest contributors

164
An early-morning creamery demonstration —— and a breakfast of
sausages and tea served by the ao.

165
The Quiet Revolution

to relieving the isolation of the countryside, had been eagerly adopted


and after a further decade would be in almost every house. The high
level of ownership of washing machines (59% by 1979) indicates higher
living standards with less toil. In the early seventies home freezing
cabinets appeared, reaching 18% ownership by the end of the decade.
They permitted a higher quality and greater variety of diet. Their most
popular use was in the long-term storage of home-killed meat.

TABLE 5

Domestic Applicance Ownership among Rural Electricity Consumers


(Percentage ownership rounded off to nearest 1%)

Percentage of Consumers
Appliance 1958 1964 1966 1968 1973 1979

Main Cooking
Electric Pe ae lee yl aly >| BAS
Bottled gas 4 14 18 24 33 34
Solid fuel S42 OOS ieee Se)
Oil 10 ee 3 Oa a ee 4, 6
Electric water heater > 8 8 Same.
Electric blanket 1 Ay DS
Foodmixer — 2 SNARE Sees
Hairdryer 1 — 7NA 17 4
Electric iron 6S 9s S2 ae Cpe SOMES,
Electric kettle 39564] S503 eee) Seo
Refrigerator 4) 28) 4G 19 e446
Freezer —- — — — 4 = #18
Television — 23 43 55 NA 84
Toaster 3 4 ay INWAN Ih — P)
Vacuum cleaner De 10) 2b Gee eS
Washing machine Tl 19'5 26s 159
Source: EsB Time — Series Surveys, May 1981
NA: Not available

It is of interest to examine the growth of electricity consumption over


the same period. Coincident with the increase in electrical appliance
Ownership was an increase in unit (kWh) consumption. Table 6 shows
the distribution of this consumption. In 1964, 67% of rural consumers
used less than 800 units for the year (21% used less than 200 units), while
in 1979, 68% used over 800 units (27% used over 4,000 units). These

166
Spreading the Message

figures are, however, national averages. As with many other consump-


tion patterns in rural Ireland, the gap between east and west shows up
in further analysis. A study by Michael A. Poole of Queen’s University
Belfast for the year 1965/66* shows a definite falling off in electricity
expenditure per rural household as one moves from the south-east
towards the north-west, the average expenditure east of a line from
Dundalk to Tralee reaching double that of the area west of this. Another
interesting indication from Poole’s study was that variation in electricity
expenditure was more closely correlated with local bog availability
(perhaps a source of alternative fuel) than with income.

TABLE 6
Annual Electricity Consumption Per Household Among Rural Electricity
Consumers
(Rounded off to nearest 1%)

1964 1966 1968 1973 1979


Electricity consumption Percentage of consumers

Up to 200 Units (kWh) 2 ee Al | 8


200 — 400 kWh 23 Tepe 4 ato 4
400 — 800 kWh 23 eee ee) 2 ae) 8
800 — 1200 kWh Oe lil ie Aye dip
1200-2000 kWh HQ ie Wek ity ifs
2000 — 4000 kWh es | eels) RY 8)
Over 4000 Units 5 Soe Oe 2a,
Source: EsB Time-Series Surveys, May 1981

167
CHAPTER FIFTEEN

Exhibitions and Competitions

THE RDS SHOWS

IN 1946, the year the Rural Electrification Scheme was launched, the
Royal Dublin Society had been 215 years in existence. It had been
founded in 1731 for improving ‘Husbandry, Manufactures and other
Useful Arts and Sciences’. In its later years much of its good work was
done through its Shows. The first livestock show and exhibition of
agricultural implements by the Society was held in April 1831 in the yard
adjoining Leinster House and the first Spring Show was held in Balls-
bridge in April 1881. From then on it was held almost every year,
interrupted only by the two World Wars and the War of Independence,
and it grew steadily in status and appeal, especially to the farming
community.
The Spring Show was first attenuated and finally suspended completely
during the 1939-45 war and was resumed in 1946. Attendances boomed
as farmers and their families, isolated and limited in mobility during the
war, were now anxious to catch up with the latest developments in
agriculture and agricultural equipment.
The value of the Show as a vehicle for communicating with the rural
community was quickly recognised by the Rural Electrification Office
and it approached the rps for suitable facilities to mount an exhibit. The
RDS in its turn, recognising the potential contribution of electrification to
the improvement of Irish farming and farming life, enthusiastically
welcomed REO and co-operated whole-heartedly in providing suitable
exhibition space.
From 1947, when the first Rural Electrification stand took its bow, up
to the time of writing, the Rural Electrification Exhibit has been a
regular feature of the Rps Spring Show. For a few years, exhibits were
also mounted at the Horse Show held in August. The Spring Show
attendance, however, was far more representative of the wider rural

168
Exhibitions and Competitions

community, and it was to this Show that most attention and resources
were dedicated.
In the late forties, when the pay-back from farming was low and the
value of electricity and electrical aids to the majority of rural dwellers
had yet to be proven, the emphasis was on the introduction of the
benefits of electricity on the farm in a simple and inexpensive way to
unsophisticated and inexperienced audiences. Year by year, as markets
improved and as farming became more specialised and market-orien-
tated, the REO exhibits changed accordingly. The electrical appliances
and equipment shown on the stands were more and more designed to
meet specific farming problems and were subjected to critical appraisal
from increasingly knowledgeable farmers and farm wives. The changing
exhibits reflected in a striking way the social, economic and technological
revolution taking place in rural Ireland.
The aim of the early exhibits was to demonstrate the many ways in
which electricity could be of value, not alone in the home but out in the
farmyard. Great emphasis was put on the contribution which could be
made by a small and inexpensive electric motor in relieving the farmer
and the farm housekeeper of many of the time-consuming chores which
beset the ordinary rural homestead. The low cost of its use was drama-
tised in many ways to an audience pre-conditioned to believe that
electricity was an expensive commodity to be used with great circum-
spection. In particular the ¢hp single phase electric motor costing about
£10 was highlighted as being capable of doing a host of jobs around the
farmyard such as root-pulping, sawing firewood, operating a grindstone
or bench drill, water-pumping or, in the farm kitchen or dairy, churning
or separating, all at a cost of one farthing’s worth of electricity per hour,
twenty-four hours a day if need be.
In order to convince the sceptics recourse was had to many dramatic
demonstrations of this tiny machine. The most successful, in the 1950
Show, was the coupling of a small generator to a converted bicycle so
that a volunteer from the audience could compete against the }hp motor
over two minutes by pedalling as furiously as possible for that period.
The pedaller’s efforts were recorded on a dial in the form of ‘watt
minutes’. At the end of the period most volunteers were exhausted and
chagrined to see the little motor calmly exceeding their best efforts and
prepared to go on doing so indefinitely. One competitor of herculean
physique, by exerting a tremendous effort, to the cheers and encourage-
ment of the large audience, actually succeeded in pushing the pointer up
to slightly better than the motor before almost collapsing from exhaus-
tion. He was unique, however, and his nearest rivals were left way

169
The Quiet Revolution

behind. None of those who had a go, nor indeed their audience, needed
further convincing of the capabilities of the small motor.
From the first exhibit onwards, the value of running water on the farm
and in the home was stressed as was the comparative ease and low cost
with which this could be done. The farmers’ penchant for doing as much
as possible with their own hands was recognised, and in the 1949 Show
three systems for obtaining running water at the kitchen sink were
shown, all of which were capable of installation by the average handy-
man. The cheapest and most basic was a small centrifugal pump operated
by the ubiquitous 4hp motor, discharging through a 2” pipe into the
kitchen sink. Not even a tap was needed, as a cord switch over the sink
started and stopped the motor. A more ambitious installation, which
could provide the foundation for a hot water system, was a self-contained
centrifugal pump feeding a roof tank and controlled by a float switch.
Finally there was the 35-gallon capacity pressure-storage pump unit
which required no roof tank. This was shown feeding a complete house
system including an 11-gallon water-heater and milk-cooler.
As time went on the water systems shown on the stand got more
elaborate and more specialised. Emphasis was put on the self-contained
pressure storage system which provided excellent pressure (20 to 40 lbs
per square inch was the most common range), and which could be
installed with the minimum of structural work. To feed the pressure-
storage system a range of pumps was shown to meet the various well
conditions and water requirements. For shallow wells, the simple cen-
trifugal pump or the piston pump, usually belt-driven by a thp motor;
for deeper wells, the centrifugal jet pump; and for the deepest, the piston
operated deep well pump and, later, the submersible pump, capable of
insertion down a narrow bore hole to an almost unlimited depth.
Various other aspects of the provision of running water were high-
lighted. The grants available from the Departments of Agriculture and
Local Government got full publicity. In order to achieve impact, com-
petitions were held, as in the case of the giant tap delivering around
4,000 gallons per hour, where competitors were invited to estimate the
amount delivered over the Show period. ‘Gimmicks’ were employed to
excite interest — the Magic Tap from which water poured continuously
without any apparent connection to supply, and the Mystic Kettle,
apparently suspended from a rope without any electrical connection but
which boiled continuously.
Group water schemes were also given publicity, the most elaborate
exhibit being the huge model of the Kilcornan Group Scheme shown at
the 1962 Show.

170
: : Me os

EsB pavilion at the Munster Agricultural Show, Cork 1952, before


the onrush of the crowds.

it
The Quiet Revolution

Water-heating also got its share of promotion as various appliances


for heating water in the home and in the dairy were shown. Year after
year the message was repeated: ‘Water in the home for cleanliness and
comfort; water in the farmyard and on the farm for clean conditions,
less drudgery, more profit; water, hot and cold, in the milking parlour
and the dairy for hygiene and quality milk’.
In the early post-war years milking machines were not yet universal in
the dairying counties, and one of the greatest sources of drudgery was
the twice-daily seven-days-a-week milking chore. Most farmers with
large dairy herds looked to rural electrification to provide a solution to
the problem, and electric milking was featured in increasingly sophisti-
cated forms in successive rural electrification exhibits. Owners of smaller
herds tended to be more cautious. In 1948 a Canadian portable
milking machine for the small herd owner was shown costing £54.6s.4d.
for the single-bucket and £80 for the two-bucket model. Following
assertions that ‘cows would kick at the machine’, REO’s Publicity Officer,
Paddy Ennis, offered to milk two cows which had never been machine-
milked before, simultaneously, with the two-bucket version, in any
location where electricity was available. The challenge was taken up by
a North Cork farmer and down on the farm, before a large audience, the
successful milking of the two cows was achieved in four minutes. The
cows were completely unruffled but the operator was under some ten-
sion. It was only when the applause had died down that he admitted that
he had never milked a cow in his life before, either by hand or by
machine. As it turned out this was the most telling point of the whole
demonstration: neither the cows nor the operator required previous
experience!
A visitor from India to the stand had never seen a milking machine.
He tested the vacuum pull on the cups and asked if it could be made
stronger. He was shown how this could be arranged and departed well
satisfied remarking that the original fifteen inches of vacuum would not
be a strong enough pull for use on a buffalo!
The opening of the RDs’s new Simmonscourt extension to Spring Show
exhibitors in 1957 provided REO with an opportunity to stage a more
elaborate exhibit. The rps offered a spacious site in a choice location.
Full advantage was taken and a full size model farmyard was built which
was to become known generally as the ‘Ess Simmonscourt Farm’. On
the ‘farm’ there was a full display — usually of working examples, which
the farmer could try out to his own satisfaction — of the various electrical
aids to farming: grain handling, featuring augers and elevators; grain
drying, showing at various times different types of dryers with suggested
loading and unloading arrangements; provender milling with working

12
Exhibitions and Competitions

examples of hammer mills, plate mills, stone grinders, oat rollers and
food-mixers with practical examples of mounting and feeding arrange-
ments; milking machines from the portable machine to in-stall milking
and the milking parlour with full-scale layout. On some occasions a herd
of pedigree milch cows was exhibited which were milked twice daily.
Manufacturers of farm buildings and suppliers of equipment co-operated
gladly in providing the ever-changing framework for the exhibit and
reaped the benefits of the resulting publicity. The layout and equipment
were changed regularly to keep pace with developing technology. Close
co-operation was maintained with the Department of Agriculture, An
Foras Taluntais, Macra na Feirme and other authorities to ensure authen-
ticity. On at least one occasion the layout was designed by members of
Macra na Feirme, whose members staffed the stand, to answer the
multitude of queries on the farming practices illustrated. In 1961 there
was a pig-rearing display which was a replica of a carefully costed project,
carried out in Colaiste Charman, Gorey.
On the occasion of the 1969 Show an elaborate ‘living exhibit’ was
staged which attracted great attention. To quote the Irish Farmers’
Journal:

Among the talking points of the Spring Show, the dairy layout in the
Simmonscourt Extension is surely a front runner. Two aspects of it should
be underlined. It is on the Ess stand, it is designed and staffed by the
Department of Agriculture and it was built by the Irish Sugar Co. In this
way it is an indication of what can be done when State and Semi-State
resources are rationally pooled. It is also a living exhibit in a show that has
too many dead ones. A herd of Friesian cows from Warrenstown (more
co-operation) can be seen doing what cows do in kennel conditions, and
there is actual milking carried out to show how quality bonuses may be
earned. In short there is a good index on offer and farmers are given a fair
chance to assess it. They are not left with a taste of it and told to use their
imagination for the rest.

This particular exhibit was filmed by rTx and featured in their ‘On the
Land’ television programme.
Advantage was also taken of the extra space in the Simmoncourt
extension to show, at the 1957 Show, the first version of the IcA/EsB Farm
Kitchen, designed by Eleanor Butler and staffed by members of the 1ca
as well as the EsB. The application of electricity to horticulture also made
its bow. Electrical heating, soil warming, soil sterilisation, plant irradia-
tion, bench warming, micro climate and mist propagation were demon-
strated under actual working conditions. As the technology of the
horticultural industry developed this was reflected in the horticultural

173
The Quiet Revolution

exhibit, especially in the environmental control area with plant irradia-


tion and ‘growing rooms’ in which the total environment of plants was
precisely regulated.
In 1960 came a radical development. The Farm Kitchen had aroused
great interest among rural housewives at previous shows, but of necessity
it had initially been housed in an unprepossessing prefabricated building.
Now it was decided to build the Farmhouse of the Sixties. As the
economy began to take up after the stagnation of the fifties, new houses
were being built in ever-increasing numbers in the rural areas. REO in
conjunction with the Agricultural Institute and its architect, Paddy Tuite,
decided to exhibit a carefully designed prototype farmhouse which would
break completely from tradition and incorporate the best aspects of
design, not alone in the kitchen but throughout the house as a whole. It
would be specially designed to cope with farm life. It recognised that the
normal method of access was from the back, directly from the farmyard,
and that the front door was little used except for special visitors. Great
attention was therefore paid to the back entrance, directly off which was
a parking place for muddy boots and a shower room, wash basin and
toilet. The adjacent kitchen was large enough to serve as the normal
living area for the farmer and the family, and was of course equipped
with a wide range of electrical aids. Adjoining was a utility room where
chores such as laundry and ironing could be carried out without inter-
fering with the activities in the kitchen. Here also were drying facilities
for the wet outdoor clothes which are an occupational hazard of farming.
The Farmhouse was an instant success. The /rish Farmers’ Journal had
this to say:

The EsB stand in the Simmonscourt Extension has, for the past few years,
been one of the most ambitious at the Spring Show. This year it is more
ambitious than ever. The farmhouse, so long a focus of interest for rural
housewives, has been completely re-designed with the co-operation of the
Agricultural Institute and is now a fully furnished and equipped three-
bedroomed house. Anyone thinking of building a new house ought to see
It.

A small indication of rising standards was the installation and complete


acceptance in 1961 of a wash-hand basin in the principal bedroom in
addition to the normal fittings in the bathroom. Some years earlier this
would have been derided as a pretentious luxury. In this year also a TV
set made its appearance in the living room.
The deep-freeze cabinet first appeared on an exhibit in 1960 when
pupils of St Martha’s College, Navan, were invited to give a display of

174
Right: The EsB/ica farm
kitchen at the 1957
Spring Show. Below:
Interested onlookers at
grooming demonstration.

175
The Quiet Revolution

cheese making and processing of broiler chickens for the table. The
processed birds were then deep frozen for market. The following year
the girls were again invited back to the stand to demonstrate the proc-
essing and storage of broiler turkeys. Later in the decade the deep-freeze
appeared as part of the kitchen equipment. The economic and dietary
advantages of being able to store a whole animal carcass in prime
condition had begun to appeal to more and more farm families. Thus in
1972 at the EsB stand we find Mr Schwer of the Meat Research Section
of An Foras Talintais demonstrating to rapt audiences how to reduce
quarters of beef to cuts suitable for the freezer.
In 1973 Ireland entered the EEc and the markets of Europe were open
to the Irish farmer. The stagnation and lethargy of the fifties were being
left behind as the industry took off into a new prosperity. This was in
turn reflected in the exhibits at the EsB stand. Specific requirements for
achieving quality and quantity production were identified. Instead of
simply promoting hot water in the dairy, the stand now exhibited ‘A
dual-purpose water-heater manufactured by an Irish firm to provide
water for udder-washing and the once a month (170°F) hot wash re-
commended by An Foras Taltntais for dairy farmers using chemical
cleaning’. ‘Wet mix’ and ‘dry’ feed systems for large pig enterprises,
slurry disposal installations and water treatment for the removal of iron
and excess minerals were now among the items listed on the stand.
An awareness of the necessity for environmental protection was indi-
cated by an exhibit dealing with research into more effective effluent
control. Electricity’s contribution to land reclamation was shown by
exhibits dealing with pumped drainage. (Two million acres of swamp
land could benefit thus.) The range of electrical farming equipment
requiring three-phase supply was reflected in the showing of single-phase
to three-phase converters. The build-up of native industry was illustrated
by the fact that these converters and indeed much of the equipment on
show were now of Irish manufacture.
In addition to being prominently featured at the rps, rural electrifi-
cation was also promoted vigorously at all the main agricultural shows
throughout the country. After the Spring Show the most elaborate
exhibit was at the Munster Agricultural Show held in Cork in early
summer, but other important venues such as Limerick, Tralee and Athy
were also given special attention. Very frequently modifications of Spring
Show exhibits were staged, in many cases following up a theme such as
water supply, grain handling, processing of feed etc. At the smaller
shows the mobile demonstration vehicles were the backbone of the REO
exhibit, sometimes backed up with additional features. Many of the
patrons of the country shows would have already been to the Spring

176
Exhibitions and Competitions

Show and thus it was quite common that a discussion on some application
of electricity to farming initiated at Ballsbridge might be continued with
REO Staff on ‘home ground’.

THE NATIONAL WHOLEMEAL BREADMAKING COMPETITIONS

The concept of national competitions for the baking of traditional


Irish home-made wholemeal bread arose from an exercise in co-opera-
tion between the National Ploughing Association and REo on the occasion
of the World Ploughing Championships held in Killarney in 1952. On
this occasion the NPA requested assistance from REO in providing elec-
tricity to power various trade exhibits on a site remote from any mains
electricity network. REO responded by lending and operating a number
of its mobile generators. Apart from helping out the NPA, it saw in the
huge and mainly rural attendance an excellent opportunity to demon-
strate what electrification had to offer.
The co-operation continued and, as the rural networks grew, it became
a practical proposition to extend mains electricity for the event rather
than rely on mobile generators. The availability of mains electricity
enabled the development of many side attractions and exhibitions,
which, when added to the ploughing competitions themselves, helped to
make for an enjoyable day out for all the family. It also provided lighting
of the exhibition site thus eliminating many of the security and other
difficulties previously experienced as darkness fell.
REO felt that such a popular rural occasion warranted something special
in the way of promotions. The concept of a competition which would
provide an attraction for women was mooted and so grew the Annual
National Wholemeal Breadmaking Championship Finals which were to
form an important side event to the ploughing competitions. There was
a widespread belief that for the best home-made wholemeal bread a
pot-oven was necessary. REO hoped to demonstrate that an electric oven
could produce wholemeal bread as good and as tasty as the best from
the pot-oven, with far less trouble and in a much more consistent manner.
It was, therefore, a condition of the competition that all entries should
be baked in an electric oven.
Preliminary competitions on a county basis were first run off by the
various ESB Districts. The county champions, as well as receiving their
county prize, had their expenses paid to attend the national finals. A
large marquee was erected near the ploughing area and electricity laid
on. A row of identical electric cookers was set up to bake the entries
which were then evaluated by a panel of judges. The excitement height-

177
The Quiet Revolution

ened in the late stages of the competition as a packed marquee eagerly


awaited the announcement of the winner.
There were some difficulties in holding such a competition in winter
in the fields. One of the most abiding memories of REO staff is of constant
rain, frequent gales, at least one snow-blizzard and almost always deep,
sticky mud, as the fields on which the crowds circulated were almost as
badly ploughed up by tens of thousands of gumbooted feet as were the
ploughing competition areas. Sometimes it was a major effort to reach
the bread-making competition marquee, isolated as it was in a sea of
mud.
Such was the experience in the case of the first National Finals, which
were held in connection with the NPA National Championships at
Tramore in 1958. On the day before the event the newly erected marquee
was torn to shreds by a ferocious gale and it took a tremendous effort by
the local EsB staff to restore it to something approaching serviceable
condition in time for the competition. There was a rather ambitious
concept of actually grinding wheat in a small electric home grinder and
using the resulting whole flour in the competition. Objections were
raised by many competitors, as they were not used to this type of
whole-flour and the idea was not repeated. The National Champion who
emerged was Mrs Elizabeth Gorey of Burnchurch, Co. Kilkenny.
The following year there were two classes of competitors, senior and
student. This resulted in a huge increase in entries, especially from
students of technical schools. The senior winner of the finals, which were
held in Burnchurch, was Miss Mary Turner of Carrick-on-Shannon,
Co. Leitrim, while the champion student was 16-year-old Mary Kilgan-
non of Gleneaskey, Co. Sligo. A feature of the competition that was to
continue was the involvement of the electrical trade and the farming
press in the provision of prizes. The judges were Mrs M. Curtis, Ica,
Miss E. M. Bonfil, Department of Education and Mona Fitzpatrick,
Chief Demonstrator, Ess. These three ladies were to continue as judges
in most of the ensuing competitions, a tribute to the integrity and
dedication they brought to their task.
The senior competition was to continue without a break for over a
decade — one of the longest-running joint promotions in which REO was
involved — with venues at Oakpark, Killarney, Donea, Athenry, Danes-
fort, Enniskerry (which was mounted in a howling blizzard), Wellington
Bridge, Tullow, Mallow, and the final competition was at Rockwell
College in 1969. The student or junior section was run as an adjunct to
the senior competition up to 1963, when the concept of a separately run
schools competition was developed. This had the enthusiastic approval
of the Department of Education, which arranged for the preliminary

178
Exhibitions and Competitions

competitions to take place in the various technical and secondary schools.


The National Finals of this competition were mounted on a dramatic
scale under the joint banner of the EsB and the electrical manufacturing
firm of GEC Ireland, in the Gaiety Theatre, Dublin, City Theatre,
Limerick, Cork Opera House and in the new Abbey Theatre, Dublin.
This competition was terminated in 1970 after a very successful six-year
run.

179
CHAPTER SIXTEEN

The Voluntary Organisations

THE TASK OF REO in extending electricity and developing its use would
have been much more difficult and would have proceeded much more
slowly without the sustained help and co-operation received from the
voluntary rural organisations. Four main bodies deserve special mention:
Muintir na Tire, Macra na Feirme, Macra na Tuaithe and the Irish
Countrywomen’s Association. The story of rural electrification would be
incomplete without some account of these organisations and in particular
the close co-operation over the years between them and REO in achieving
what after all was a common objective. In a happy ‘chicken and egg’
situation the coming of electricity made it possible to develop attractive
community meeting places with heating, lighting and facilities for drama,
cine-projection, public address and similar aids, which in turn increased
the effectiveness of the organisations and helped their growth and their
contribution to the rural electrification effort.

MUINTIR NA TIRE

The name of Muintir na Tire (People of the Land) will always be


coupled with that of its founder Rev. John Hayes, who launched this
organisation for community development on a parish basis in May 1931.
The growth and progress of the movement have been charted by Rev.
Jerome Toner, OP, in his book Rural Ireland, Some of its Problems' and
by Stephen Rynne in his biography of Father Hayes.” Father Hayes was
committed to raising the whole standard of rural life. In a broadcast on
25 July 1945, he said: “We realised that the land problem was not merely
economic; the merely economic solves no problem. It was more than
economic. It was the whole life of the people that needed stimulus.’ It
was a happy coincidence that the man appointed to head the Rural
Electrification Organisation, W. F. Roe, had since 1938 been an active
and prominent member of Muintir na Tire dedicated to the ideals of its
founder.

180
The Voluntary Organisations

Both men saw in rural electrification one of the most effective means
of providing the stimulus required to overcome rural stagnation. Speak-
ing on 24 May 1948 at the switching on of electricity for the first time in
Bansha, Co. Tipperary, of which he was then parish priest, Fr Hayes
made this point. Having referred to the more obvious advantages of
rural electrification he went on:
. . . but there is a still greater advantage and that is the social aspect. No
material instrument can do so much to uplift the rural people and give
them a status in accordance with the important position they hold in the
nation. Rural people who supply the fundamental necessities for the whole
nation should at least have an equal right to the amenities of the nation....
It is more than an amenity, it is a revolution which will sweep away
inferiority complexes.*

From the very start there was a very high degree of co-operation between
Muintir and REO. The parish guilds were of the utmost assistance in
initiating and organising the preliminary canvass to establish the priority
for the official EsB canvass. In this the guild members made strenuous
efforts to sell the advantages of electrification to often reluctant house-
holders so as to achieve a high sign-up and a consequently high ranking
on the list for official canvass and selection. In his book Fr Toner adverts
to these efforts. Under the chapter heading ‘Muintir na Tire 1944-1949’
he writes:
The reports for these years show clearly that the war is over. There is a
significant remark of ‘Contacted REO’, which means that the guild
approached the Rural Electrification Office for the purpose of having its
area considered under the terms of this Scheme . . . . The work of Muintir
na Tire in disposing farmers to avail themselves of this Scheme is of great
importance. In the lectures and free debates arranged by parish guilds,
much of the prejudice against innovations is broken down and farmers are
encouraged by the confidence of others in the Scheme. Indeed, Muintir na
Tire would find it hard to develop rural Ireland without electricity. If the
Scheme is a success, and that seems probable, it will bring about a complete
change in Irish rural life. It will raise the general standard of living in
country homes; it will speed up and increase agricultural production; it
will induce capital to establish rural industries and it will brighten the
social life of the parish. One may hope, therefore, that it will go far
towards solving the problem of emigration; people leave rural Ireland
because the land is not supporting its own. ‘Contacted REO’ then in a parish
report . . . represents a fundamental work in the programme of the guilds.

The ‘Rural Weeks’ O In January 1947 W. F. Roe received a letter from


his friend Fr Hayes inviting him to give an illustrated talk on rural

181
The Quiet Revolution

electrification and mount an exhibition of electrical appliances at a


Muintir na Tire ‘Rural Week’ in St McCartan’s Diocesan College,
Monaghan, in the following summer. (The Rural Week — more exactly
a three-day rural week-end — was Muintir’s big annual event and had
been inaugurated in 1934 at St Joseph’s College Roscrea.) Thereafter
REO participated in many Rural Weeks where it sought to get the message
of rural electrification across. In the following year at Carlow the main
REO theme was on how to go about getting one’s parish listed for
consideration under the scheme. In 1952 at Summerhill College, Sligo,
there were demonstrations in electric cooking. A film “Running Water’
was also shown each day to promote the concept of water on tap in the
house and farmyard. In 1956 at the Killarney Rural Week the emphasis
of the REO exhibition was on power tools suitable for farmers and for
small community projects which might be set up by the local guild of
Muintir. The main power-tool manufacturers co-operated to mount a
comprehensive exhibition and working demonstration of equipment
which included drilling, sawing, welding, brazing, ceramics and small
stained-glass work.
This close co-operation in the Muintir na Tire Rural Weeks continued
until the abandonment of the concept in 1970 in favour of an annual
national conference. It had been of mutual benefit, providing as it did an
extra attraction in the Rural Week programme for Muintir and providing
also a most valuable shop window to REO in which to display its wares
and promote its message.

MACRA NA FEIRME AND MACRA NA TUAITHE

The war of 1939-45 highlighted the importance of the agricultural


industry in Ireland and the pressing need to move away from the
Cinderella image it had developed. In particular it was vital that proper
agricultural education should be provided for those young people des-
tined to work the land. The usual practice was that post-primary and
higher education was reserved for the sons and daughters of farmers who
would move out into other occupations. The son inheriting the farm
usually received no formal education beyond that of the primary school.
The small number of agricultural advisers and rural science teachers
then existing were well aware of the necessity to provide a proper
educational service in this area and did all they could in the face of
limited resources, holding winter agricultural classes and rural science
classes in the vocational schools.
In order to preserve the contact set up between pupils and teachers at
these winter classes, the idea of young farmers’ clubs was born and in

182
The Voluntary Organisations

1942 and 1943 the earliest such clubs, known as farmers discussion
groups, were formed. These led in 1944 to the formation of a national
organisation of young farmers’ clubs, with a central executive. In Sep-
tember 1947 a national headquarters was opened by Sean T. O’Kelly,
President of Ireland, in Athy, and in December the title Macra na
Feirme was officially adopted. In 1948 a fortnightly paper entitled Young
Farmers Journal was launched, which was destined to be the forerunner
of the very successful Irish Farmers Journal.
At this period the question of an organisation to speak for all farmers
on economic issues was raised. It was decided Macra na Feirme would
concentrate on the educational, cultural and social interests only and in
1955 a separate organisation, the National Farmers’ Association, later
to be re-named the Irish Farmers’ Association, was founded to look after
the economic interests of farmers.
From the very start Macra na Feirme gave very strong support to the
concept of rural electrification. Like Muintir na Tire, it helped to organise
local canvasses and to support REO in the official canvass and selection
process. There was a great deal of dual membership and in many cases
the two organisations worked closely together to try to ensure early
selection of their parish or area. On the odd occasion a degree of rivalry
broke out, as when a contest developed in 1948 between the parish of
Cahir and that of Bansha as to which would be the first area to be
developed in County Tipperary. The prime mover in Bansha was Muintir
na Tire, led by its founder and parish priest of Bansha, Fr John Hayes.
In Cahir, Macra was the main organiser and when Bansha was selected
on the basis of a better economic return, although Cahir had a better
sign-up, a deputation from Cahir called on W. F. Roe to protest against
the alleged favouritism.
Despite the occasional controversy, relations between Macra na
Feirme and REO continued on a most co-operative basis. REO staff
attended with films and demonstrations at the various Macra functions,
gave lectures on the applications of electricity at their winter programmes
and co-operated in organising competitions, while Macra members in
their turn strongly supported REO’s efforts to extend and develop the
Scheme. On at least one occasion the REO stand at the rps Spring Show
was designed and staffed by members of Macra.
In 1951 the idea of a junior Macra to cater for the twelve to eighteen
age group was born. The main thrust came from a number of rural
science teachers in vocational schools, but it immediately received strong
backing not only from Macra na Feirme, which gave it particular organ-
isational support in its early days, but also from Muintir na Tire and the
Irish Countrywomen’s Association. The new organisation was given the

183
The Quiet Revolution

title of Macra na Tuaithe. It was broadly based on the same concept as


the American 4H Clubs with strong emphasis on the principle of learning
by doing. The members of the new organisation were the young people
who were to remain on the land. They were bright, forward-looking and
eager to experiment with new methods. The promotion of Macra na
Tuaithe throughout the country was helped immensely in 1958 by a grant
of £30,000 from the American Kellog Foundation, to be used for this
purpose over a five-year period. In order to continue the financing of the
organisation when the Kellog Foundation grant ran out in 1963, the
National Youth Foundation was established, to which Irish industries
and commercial concerns were invited to contribute an annual sum.
Over 200 firms responded positively. The EsB became a member of the
Foundation from its inception and also took part in a scheme of national
awards, which had been set up to encourage initiative and effort among
its members. In addition to its membership of the Foundation, the EsB
also agreed to sponsor an annual Home Improvement Award for Macra
members. This award was for the planning and carrying out of an interior
decoration scheme, and the EsB was joined in its sponsorship by the firm
of Wallpapers Ltd. In the same year the EsB also agreed to contribute
towards the costs of an annual Macra Leadership Course, involving fifty
participants, at the Ica residential college at An Grianan. Inspired by
the enthusiasm of its founders and supported by both its industrial
sponsors and the State, through the Department of Education, Macra
na Tuaithe flourished. In 1959 there were 125 clubs, in 1963, 231 clubs,
and by 1966 there were 253 clubs dispersed throughout the whole
twenty-six counties.

The Citizenship Award UO In 1967 the Home Improvement Award


was discontinued owing to difficulties associated with its supervision and
in 1968 a new competition was sponsored by the EsB entitled Know Your
Area. As originally proposed, this was a club project with four prizes of
£50 each for the winning club in each of the four provinces, the four
provincial winners competing for the national prize of £100 at the annual
summer gathering. In 1969 the Jrish Farmers Journal joined in the
project providing additional welcome finance and it was decided to
award, in addition to the cash prize, a perpetual trophy and replica for
retention by the winners. For this revised competition now known as the
Know Your Area Citizenship Award, the provincial basis was revised.
County Donegal was included in the Connacht group and Cavan and
Monaghan were included with Leinster, so giving three geographical
divisions. The perpetual trophy took the form of a carving in cherrywood
and bog oak.*

184
The Voluntary Organisations

The Citizenship Award competition was based on Macra’s Citizenship


Programme and involved the collection, organisation and recording of
data on one or more aspects of life in each club’s area. Director Michael
Cleary speaking in 1970 expressed the purpose of the Citizenship Pro-
gramme as ‘to help young people acquire the skills necessary to maintain
and develop a true democratic society. Democracy implies that people
must be informed citizens. Our project trains young people in the
methodology of studying their society and, even more important, in
doing something about the problems in their society once these have
been identified.’
The Programme and Award developed a whole new concept as to
what the organisation should do. It had originally been founded to
promote, principally at least, rural science and home economics as
extra-curricular activities from the local vocational school. Arising out
of the Citizenship Programme came a new concept: a more active
outgoing role in local society. From knowing one’s area came the
transition to action. In the early stages of the competition the main
emphasis was on investigation of the local scene and possibly criticism of
what the investigation turned up. The more active clubs were not,
however, content with just this; they initiated action programmes to
correct or improve the situation as they had found it. In later years it
was an essential part of the competition that the club should become
involved in some type of action.
From 1971 on, the Award was sponsored by the Ess alone. It unearthed
a formidable array of talent among the youth of rural Ireland. From the
very beginning, the standard of entry was high and the standard of the
winners and runners-up approached that of professionals. The winner of
the Award in 1969, its first year, was the Ballyfin, Co. Laois, club. The
runners-up were Grange, Co Sligo and New Inn, Co Tipperary. The
sponsorship of the Award by the Ess continued even after the official
termination of the Rural Electrification Scheme and the Citizenship
Programme still plays a central part in the life of the clubs. The variety
of projects is endless: local effects of pollution, history of parish, relat-
ionship between nutrition and health, local implications of EEc member-
ship, formation of mussel-farming co-operative, tidying up of local
cemetery, study of old people’s problems, promotion of active interest
in sport.
In 1981 the title of the organisation was changed from Macra na
Tuaithe to Fordige or the National Youth Foundation, to reflect the fact
that it was no longer an exclusively rural body. Clubs had now been set
up in urban locations and it is interesting to note that in that year the
Award was won by an urban club, The Cool Kids in Ballymun, Dublin,

185
The Quiet Revolution

in its third year of existence. The project, Operation Access, was chosen
in the Year of the Disabled to help the public realise some of the
difficulties facing disabled people. Having acquired a number of wheel-
chairs, members invited both shoppers and shopkeepers in the local
shopping centre to sit in the wheelchairs and try to do some of the things
normally taken for granted such as shopping and making telephone calls.
The findings shattered the complacency of the community and brought
about a greater understanding of the problems of the disabled. Improve-
ments were demanded in many areas. An immediate result was that the
local supermarket widened the space at the check-outs to allow wheel-
chairs to pass through.
The finals of the competition and the presentation of the Award were
always the highlight of the year’s activities and took place at the organ-
isation’s summer gathering, usually at Gormanston College. An indica-
tion of its prestige was the fact that on various occasions the Award was
presented to the winning team by the President of Ireland, Erskine
Childers, and the two religious primates, Cardinal Conway and Arch-
bishop Simms.
No one project so vividly illustrates the social and educational progress
of rural Ireland in one generation as the Citizenship Programme. From
the stagnation of the late 1940s to the dynamism of the youth of the late
1970s was a tremendous leap in which the provision of amenities made
possible by rural electrification played a vital part. Without these, the
task of the voluntary bodies such as Fordéige would have been extremely
difficult, if not impossible.

THE IRISH COUNTRYWOMEN’S ASSOCIATION

Sir Horace Curzon Plunkett, father of the co-operative movement in


Ireland, writing of his initial difficulties admitted to making a couple of
fundamental mistakes:

But we [the founders] failed to realise fully two things: the first was the
enormous difficulty of establishing a new social organisation in a country
where there is none already existing; it is like digging foundations in sand.
The second was the importance of women’s work. We had put ‘better
business’ first, ‘better farming’ next, and ‘better living’ was to follow as a
result of these two, but, as things are to-day, we shall not get very far
either with better business or better farming unless we can show some
better living as a bait and thus stimulate the desire for more. And this the
women can do best; in fact they alone can do it.°

186
The Voluntary Organisations

Plunkett’s experience was not lost on the strategists of REo. The ability
and motivation of the women to achieve the ‘better living’ referred to by
Plunkett and to spur on the often reluctant male partner to accept and
utilise the new electricity for this purpose was manifest. In developing
co-operation with the Irish Countrywomen’s Association, REO was push-
ing at an open door: it was precisely to achieve this raising of the standard
of living in rural Ireland that the Ica was founded in 1910 under the name
of the Organisation of United Irishwomen. The organisation had its
roots in Plunkett’s great co-operative movement which had up to then
been male-dominated, but, to quote Ellice Pilkington, one of the foun-
ders of the United Irishwomen:

I was recommended by a relative of my own, a member of the Irish


Agricultural Organisation Society to attend the Annual General Meeting
of that Society in 1909 .... Many women who are now ‘United Irish-
women’, but who were then entirely unconnected with the Agricultural
Organisation Society, were also present. Mr George Russell (AE) at that
meeting read a paper on rural civilisation and in it pointed out to us quite
clearly that our place was ready for us if we were ready to step into it.°

On 15 June 1910, Mrs Harold Lett, Vice-President of Wexford Farmers’


Union formed the first branch of the United Irishwomen at Bree, Co.
Wexford, a non-sectarian, non-political organisation affiliated to the
1aos. This was followed by a visit to Paddy the Cope in Dungloe,
Co. Donegal, in December, resulting in the formation of a branch there
of two hundred members. Four more branches followed in Wexford and
then one in Kilkee, Co. Clare. The Society of United Irishwomen had
been successfully launched.’
It was stressed from the beginning that the Society was strictly non-
political and non-sectarian and intended to include all levels of society.
Nevertheless, and inevitably, the title United Irishwomen appeared to
many to imply a political connection and it was decided to change it to
the Irish Countrywomen’s Association, or in Irish Bantracht na Tuaithe.
By 1969 there were 22,000 members in 877 local guilds throughout the
country.
The purpose of the new organisation was to organise women in every
district in Ireland for community service and to improve the country’s
domestic and reconstruct its social life. To quote Plunkett once more:

... they come in to complete our work where we believe it to be at once


most important and most incomplete . . . and as one of those responsible
for this imperfect movement, I hope they may realize the most cherished
and least fulfilled ambitions of its male leaders, the evolution of a healthy

187
The Quiet Revolution

and progressive community life. To women’s influence I look for the


brightening of the social sky in the grey dawn of peasant Ireland.*

Dr Muriel Gahan in a lecture given in 1969 takes up the theme:

‘The evolution of a healthy and progressive community life’ — what a task


in the year 1910. No wonder the men had despaired of their better farming
and better business alone bringing it about. Nothing but a woman’s
dauntless spirit, a cause to which she was utterly committed, and a
programme dealing with needs as she met them, could have made any of
those pioneer countrywomen confront the rural problems of the day. With
prejudice of every kind against them, these women did confront them
under their chosen banner ‘Deeds, not Words’ . . . Countrywomen’s
present day needs, with which among others, our Association concerns
itself, for more piped water and labour saving houses, home advisers,
adult education and training, have developed from those basic needs met
by the first United Irishwomen, District Nurses, adequate food and cloth-
ing, with milk as a priority.”

The advent of rural electrification brought nearer the prospect of


piped water and labour saving houses to Irish countrywomen and the
Association threw itself into a full scale campaign to promote these.
Piped water was regarded as the top priority, as with it would come the
end of the drudgery of hauling buckets of water from the well, a task too
often relegated to the rural housewife. With it also would come a
transformation of the sanitary facilities which even in the late 1940s
were, in the case of nineteen rural homes in every twenty, of an extremely
basic nature.'°
Thus it was that at the ica Annual Fair at the Mansion House Dublin
on 8 and 9 November 1950, a special display, with water as the central
theme was mounted by the EsB Rural Electrification Office. The
centre-piece, designed to make a strong initial impact, was a huge
aluminium water tap out of which water was gushing at the rate of 2,500
gallons per hour. To supplement this was shown a full size replica of a
corner of a cottage kitchen with sink and hot and cold water. James
Dillon, Minister for Agriculture, was photographed in the act of turning
on the huge tap. Later he spoke of the importance of running water and
pointed out the assistance available from his Department. His concluding
remarks were highly applauded: ‘Young ladies of the country; make it
known that there will be no more marriages until there is hot and cold
water on tap in the kitchen’.

188
Dr Muriel Gahan, champion of the spread of rural electrification, who was
dedicated to the improvement of the lot of the Irish country woman. She is
holding an AiB community service award, 1974.

189
The Quiet Revolution

The ESB/ICA Kitchen 0 Apart from water, the importance of good


design and layout in the achievement of a labour-saving home was
continually stressed by the ca. In the kitchen particularly, where the
rural housewife spent so much of her time and did so much of her work,
a well-designed layout could do much to eliminate traditional drudgery.
In 1956 the Minister for Agriculture, again James Dillon, requested the
IcA to set up a typical farm kitchen of the future as part of the Depart-
ment’s exhibit at the rps Spring Show. Once again the EsB joined in and
the 1cA Farm Kitchen was born. It showed how a typical farmhouse
kitchen could be renovated to become a comfortable living and working
centre for the whole family. The following year, as the opening of the
Simmonscourt extension made more space available for the EsB exhibit,
again the EsB and Ica, with architect Eleanor Butler as consultant, joined
in designing and furnishing a modern labour-saving farm kitchen. Such
was the interest shown that it was decided that the kitchen should be
made available for exhibition throughout the country, and in March
1958 the EsB/IcA Mobile Farm Kitchen took to the road.
An itinerary was devised so that the kitchen could be exhibited at
strategic locations throughout the country. It was open to the public for
a certain period each day, but at night was reserved for members of the
local guilds of the 1cA, Muintir na Tire and Macra na Feirme, who
received lectures from kitchen planning experts of the Ica and from
technical instructors who showed how the kitchen was a conversion of a
traditional rural kitchen and how it could be built up over a period, long
or short, on the do-it yourself principle. There were also demonstrations
of electric cooking and general kitchen craft. The exhibit was built on a
special trailer chassis, which had to conform to road traffic requirements
when in transit. On reaching its selected site, however, it opened out to
form a spacious kitchen which provided for a food preparation and
serving centre, ample cupboard space, sink area, drop-out ironing board,
and, in addition to an electric cooker, a hearth to retain the traditional
atmosphere of the old fashioned kitchen. Off the kitchen proper was a
bathroom and laundry area.
For many years the mobile kitchen worked its way up and down the
country being seen and discussed by young and old alike. Special essay
competitions were organised for school children on what aspects of the
kitchen impressed them and small prizes awarded. Sometimes it provided
a community service as when on a cold night in December 1958 in
Dunfanaghy, Co. Donegal, sandwiches and tea were prepared and
served in the kitchen to 120 participants by a ladies committee on the
occasion of a ‘twenty-five’ card drive in the local hall in aid of the school
building fund. Some weeks later when the kitchen was making its last

190
The Voluntary Organisations

stop in connection with the Ica winter programme in County Donegal,


it again came to the rescue in Derrybeg, when the water supply to the
local hotel failed, knocking out the solid fuel water-heating and cooking
system on a day when a large funeral had resulted in a heavy demand for
luncheons. With the help of the mobile kitchen parked nearby, the EsB
man in charge, Colm Browne, arranged immediate alternative cooking
facilities for the hotel staff ensuring an excellent luncheon to all of the
customers and earning the plaudits of the local community.
The kitchen did not as a rule attend special functions such as shows
where it would be only required for one day, but on the occasion of the
Annual General Meeting of the Wexford Federation of the Ica, held in
Coolattin on 11 June 1959, and coinciding with the Coolattin Fete, an
exception was made. The opening ceremony was performed by ‘Biddy
Mulligan’ (the late actor Jimmy O’Dea), who was met by Olive Countess
Fitzwilliam, on whose estate the Fete was held, at the door of the
kitchen. Over 2,000 visitors passed through during the remainder of the
day.
In these and in many other ways the mobile kitchen was kept to the
fore in the media ensuring that wherever it went it drew large crowds.
On the fairday in Tullamore in May 1959 a farmer called in to the kitchen
which was on exhibition and offered £1,000 in cash for it as it stood!
There is little doubt that the subsequent raising of the standard of Irish
rural kitchens as pleasant and efficient places in which to live and work
was due in great part to the EsB/Ica kitchen in its various forms, mobile
and static, reinforcing as it did in a most effective way the continuing
educational efforts of the Association.

The An Griandn Scholarships 0 In 1953 the Kellog Foundation of


America made a gift to the Ica of a large country mansion in Termon-
feckin, Co. Louth, for use as a residential college. Apart from the main
house there was a small cottage adjoining the farmyard and the idea
grew that this could be refurbished and made to resemble a modernised
farmhouse, incorporating the design of the mobile kitchen as far as
possible, and used as a training unit for young rural housewives of the
future. The EsB agreed to provide the electrical equipment and to award
fifty-four scholarships annually from 1954 to cover six-week courses for
nine girls at a time, while the Ica provided the training. The candidates
were from the vocational school courses in the various counties."’
The scholarships were confined to girls who had been successful in
their courses at the vocational or techincal school and who were most
likely to return to their homes after the course, having acquired new
skills as homemakers. Selection was carried out by the Department of

191
The Quiet Revolution

Education. Emphasis was on the practical use of electrical aids and the
course covered cooking and household accounts (with special emphasis
on budgeting), laundry and home management, and poultry, dairy and
gardening.
A further set of twenty-seven annual scholarships was sponsored by
the Ess for senior ICA members, again in homemaking, with emphasis on
the use of electrical aids to relieve drudgery and improve the standard
of household management. In the farmyard a building was reconstructed
and converted into a cafeteria and folk museum with a working replica
of the EsB/IcA Farm Kitchen as the focus. Here also the EsB sponsored
specialised courses such as the ‘farm guesthouse’ course for farm
housewives wishing to join the Bord Failte scheme.
The ICA was very interested in the concept of promoting wholemeal
bread baked from home-milled wheat as an economical and nourishing
item in the diet of the rural community, and a feature of many demon-
strations with the Farm Kitchen was the production of genuine stone-
ground wholemeal flour ground in the “Barngold’ home-grinding mill
and the baking of the wholemeal bread in the electric oven at a total cost
of about 2d. per pound of finished bread.
In May 1959 a ‘Rural Family Week’ was organised by the Ica in
Rathmines Town Hall in co-operation with the Home Economics Section
of the Food & Agricultural Organisation (FAO) at which the role of
electricity was given prominence. A paper entitled ‘Electricity and the
Rural Family’ was read by P. J. Dowling, Engineer-in-Charge, REO.
While the Ess, the Ica and the Departments of Agriculture and Local
Government were each working to promote the extension of piped water
supply to rural homes, progress was painfully slow in view of the huge
numbers still to be served. The 1ca decided that a much more intense
and unified campaign was necessary if worthwhile progress was to be
achieved. It therefore called a meeting of all interested parties on 24
August 1960 at its headquarters in St Stephen’s Green to launch a
national campaign to develop rural water supplies. The Ica notes in the
Farmers Gazette of 3 September 1960 announcing the campaign stress
that ‘every Irish countrywoman must become a preacher in this cam-
paign. If each and every one of us lifts her voice and demands water for
rural homes, the present scandal will be remedied, for it is a scandal that
only 12 per cent of rural homes, have a piped water supply.’ (See chapter
17 for a more detailed description of the campaign.)

The Drimoleague Kitchen 0 The Area Organiser for the Cork County
Federation of the 1cA, Mrs Kathleen Gleeson, had seen in Holland the
work of teaching rural housewives new and better ways of housekeeping

192
The Voluntary Organisations

carried out by a similar local organisation in the kitchen of a house


belonging to a selected resident of the area. In the summer of 1959 the
idea was adopted in Drimoleague, Co. Cork, in the house of Mr and
Mrs Humphrey O’Leary, two years married, who were in the course of
remodelling their kitchen. The Ess agreed to sponsor the project and
supply the electrical equipment. Firms in Cork also gave asssistance by
providing tiles, paint, pots and pans etc. while the Cork Vocational
Education Committee agreed to provide a weekly class in domestic
economy. A committee representative of the various groups co-operating
was set up to run the project. Sixty local women applied for enrolment
for classes held in the environment of a well-designed and well-equipped
kitchen. The project operated succesfully for the agreed period of two
years after which it was terminated to prevent its becoming a burden on
the very co-operative owners. At the winding up ceremony the chairman
of the operating committee was enthusiastic about its value: “The exam-
ple of kitchen planning has been followed in the houses of the district
while its good influences on the social side should not be omitted — the
local men and women met and discussed their problems having a cup of
tea between lectures in the Electric Farm Kitchen.’
In 1961, a similar project was launched at Castlemaine, County Kerry,
in the home of a member of the local tca Guild, Mrs Timothy O’Brien.
Classes in domestic economy were held twice a week from 1 March to
the end of June with the co-operation of the County Kerry Vocational
Education Committee. An average of about forty, mostly IcA members,
attended. In 1963 members of Macra na Feirme and Macra na Tuaithe
were involved as the ESB commenced a series of weekly demonstrations
in the use of electricity in and around the home. The afternoons were
for Macra na Tuaithe (twenty-nine boys and six girls attended), while
the evenings were for the seniors with an average attendance of fifteen
men. After the agreed period of two years, the project was terminated.
By the completion of the Rural Electrification Scheme in the mid-
seventies the ESB/ICA partnership could look back with satisfaction on
the progress achieved towards ‘better living’ for the rural community.
Electricity was available in almost every home, the standard of rural
housing and housekeeping had improved and water on tap was well on
the way as a reality for all. However, the task is not yet finished and
while television has largely superseded the meetings and gatherings in
the local hall as a source of entertainment and information, the close
co-operation still continues in many areas particularly in the sponsorship
of courses in An Grianan.

193
CHAPTER SEVENTEEN

Water on Tap

Part of the drudgery of the woman’s work on the farm was attributed to the
farmhouses. The long low thatched building, with the well a field or two
away, and the candle in the window at nightfall — all that might be some-
thing for homesick Yankees to enthuse over, but it was a heartbreak for the
unfortunate woman who had to keep it clean: who had to cook on its
back-breaking hearth, while watching her children blinding their eyes trying
to get their lessons done by candle or oil lamp light: and it was she who had
to tramp the muddy path to the well, and carry back every bucket of water
needed for her work.'

EVEN THE MOST ARDENT ADVOCATE Of electrification on the REO staff


would have to admit that running water should take priority in a rural
home over the other benefits of electrification; but of course it was the
coming of electricity that made it possible for the first time to install
water on tap in the majority of rural dwellings at a reasonable cost.
Electricity is but a means to an end, the end in this case being the raising
of living standards, the elimination of drudgery in the farmhouse, the
farmyard, and on the farm, and the increasing of productivity and,
consequently, farm income. What better first step towards these objec-
tives than the installation of an electric pump.
Running water could do much to raise living standards. A footnote in
the Report of the Commission on Emigration and Other Population
Problems refers to the absence of water on tap or of proper sanitation in
the Rural Ireland of the late 1940s:

In the 1946 Census Report on the matter it was stated that in rural areas
in the twenty-six counties only 9 per cent of households have a piped water
supply, 44 per cent get their water from private wells, 26 per cent from
pumps, 12 per cent from streams and the remaining 9 per cent from other
sources. Half of the farm dwellings get their water from private wells but
only 5 per cent have a piped supply laid on to the buildings and these are
all on farms over £200 Poor Law Valuation. Only a little more than half

194
Water on Tap

the number of farm dwellings with a piped water supply laid on have a
fixed bath. As regards sanitary facilities, four out of every five farm
dwellings in the twenty six counties have no special facilities such as flush
lavatory, chemical closet, privy or dry closet, and of the very large
dwellings, (P.L.V. £200 and over), about one seventh have no special
facilities. Of the small number with facilities (one out of every five), only
about a quarter have indoor lavatories. Thus, only one out of every 20
farm dwellings has an indoor lavatory.’

Drudgery was an inevitable outcome of the situation. In over 90% of


homes every drop of water used in the house had to be carried over the
threshold, most of it from a considerable distance, usually by the farmer’s
wife or children. Every drop of hot water had to be heated over the fire.
Most Irish farm housewives set high standards of cleanliness, but under
such circumstances the achievement of this involved an inordinate
amount of time and effort. To quote from John Healy: ‘on churning day
she’d [Grandma] scald everything with iron kettles of hot water. Iwould
run to the well several times that day and the last can of water went for
the washing up afterwards, when everything had been scalded again.”
When account is taken of the time spent on the multitude of other chores
around the farmyard, traditionally the responsibility of the woman of
the house, it will be seen that the lot of the average farmer’s wife was a
hard one. It is no cause for wonder that so many farmers’ daughters
opted for urban life and urban husbands, rather than marry farmers and
inherit the life of unremitting drudgery that had been the lot of their
mothers.
Just as running water in the house would emancipate the farmer’s wife
from so much drudgery, so out in the farmyard and in the fields it could
do the same for the farmer. For the dairy farmer in particular it meant
a higher standard of hygiene and subsequently higher prices for his milk.
For his cows a continuous supply of drinking water could mean higher
milk yields and for dry stock, better weight gain, not to speak of the
time saved in watering stock on so many farms. It was natural then that
from the very outset REO put the greatest emphasis on the provision of
running water on the farm as one of the first fruits of electrification. At
every lecture, demonstration or other promotion of the Rural Electrifi-
cation Scheme water pumps and water systems were given pride of place.
Very quickly a pump advisory service was set up, which developed into
a comprehensive water advisory service which gave advice and assistance
on all aspects of running water installation from the location of water
sources to the selection, installation, and maintenance of pumps and
water systems. This advice and assistance was made freely available, not

195
The Quiet Revolution

only to individual householders, but to group organisations, local author-


ities and government Departments. Thirty-five years after its foundation
the service, now part of the Agricultural Advisory Unit of the Ess, is still
contributing to the national drive for the installation of running water in
every home in the State.
REO was conscious that preaching the advantages of electric water
pumping was futile unless the actual appliances were made easily avail-
able and co-ordinated efforts made to install them in consumers’ prem-
ises. From the outset, therefore, even in the immediate post-war years
when supplies were scarce, a policy of stocking pumps of a size suitable
to the requirements of rural consumers was followed. The emphasis was
on reliability and simplicity of installation, plain equipment with no frills.
At the rural electrification stand at the Spring Show of 1948, three
systems for obtaining water at the kitchen sink were shown. All could be
installed by the average handyman. The first and most basic exhibit was
designed to take the mystery out of installing a water supply. It consisted
of a small centrifugal pump mounted over a ‘well’ and driven by a belt
from a fractional hp motor. The output from the pump was brought in
over the kitchen sink by means of a 3” pipe and discharged directly from
the pipe into the sink. Not even a tap was involved. A cord switch over
the sink switched the motor and pump on and off as water was required.
The ultimate objective was, however, to have hot and cold water on tap
and the second example showed how a self-contained electric pump
could feed a roof tank being switched on and off automatically by means
of a float switch as the tank filled and emptied. This could form the basis
for a combined hot and cold water supply by adding an electric water
heater. The third exhibit was a pressure-storage system with a tank of
35-gallon capacity which was completely self contained, operating at a
pressure of 20 — 40 lbs per square inch, which required no roof tank and
indeed required the minimum of structural work to install. This type of
system was to prove by far the most popular. Over the years tens of
thousands were installed using different types and sizes of pumping and
storage units to meet the customers’ requirements.
At the 1950 Spring Show one of the REO stands was given over
completely to water supply and full publicity was given to the Department
of Agriculture grants that were available. Both shallow-well and deep-
well pumping systems were shown actually working. In addition, two
film strips were shown continuously, one on How to Bring Water Supply
to the Farm and the second on Hot Water on the Farm.
As the Rural Electrification Scheme progressed no opportunity was
missed to promote the message of running water in the home and on the
farm. Members of the Irish Countrywomen’s Association were very

196
et : ge as . ~ aS
gee: sasemmnn S

Above: Carrying water h ome — ESB advice below is ‘let it run’.


Below: James Dillon, Minister for Agriculture, turning on the giant
REO tap at the cA Annual Fair, Mansion House, Dublin, November,
1950.

197
The Quiet Revolution

powerful allies. Time and time again pressure from the woman of the
house was the deciding factor in persuading the man of the house to
invest in a water system. Once the first steps had been taken the rest
followed quickly. Starting with a sink and cold water tap in the kitchen
a hot water installation soon followed. A flush lavatory was then installed
and a bathroom. Supply was quickly extended to taps in the farmyard
and, in places where streams were scarce or unreliable, out to the fields
to ensure a constant and reliable supply of drinking water for the stock.
The importance of efficient servicing (diagnosis of problems, repair of
faults etc.) was recognised at an early stage and training courses for ESB
electricians in maintenance and repair of pumps were instituted. A
comprehensive series of articles on pump trouble-shooting was published
in REO News and specialists in all aspects of water pumps made regular
visits to local centres.
In pre-war years resources for the provision of public piped water
supplies were of necessity confined almost exclusively to urban areas.
The emphasis was on the improvement in public health conditions and
priority lists were drawn up, which naturally concentrated firstly on the
more densely populated urban locations as presenting a greater potential
public health hazard than the sparsely populated rural areas. Just before
the 1939-45 war, however, a start was made in supply to rural areas on
a regional basis in the Rosses, Co. Donegal, a thickly populated area
with poor local water sources. The war put an end to such expansion and
in the immediate post-war years when capital for public investment was
scarce, roads, housing and electrification itself received priority over
public water supply in rural areas. By 1950, however, there was again
movement and a system of contributions from the central government to
the cost of local authority water supply schemes was introduced. In the
same year a system of grants for private installations on farms was
initiated by the Department of Agriculture, followed in 1952 by a similar
grant system for private houses by the Department of Local Government.
The Department of Agriculture scheme provided for a grant of 50%
of the cost (up to a maximum grant of £100) of providing water suitable
for domestic use and piping it to kitchen sink and water tap in farm
dwellings. In initiating the scheme the Minister for Agriculture, James
Dillon, appealed particularly to the wives and wives-to-be of farmers ‘to
employ the maximum diplomatic pressure, which connubial propriety,
or the privilege of bethrothal may properly allow, to persuade the head
of the household to provide this indispensable amenity of a piped water
supply’. These grants paved the way for the installation of private
schemes but, despite the forceful appeals of the Minister, progress was
painfully slow in the early years. In the year 1950/51, only 150 water

198
Water on Tap

pumps and pumping systems were sold to rural consumers by the Ess. In
1953/54 the total sales for the year were about eight-hundred pumps (of
which the EsB share was 350), a pitifully small figure in view of the strong
promotional effort. A sub leader in the Irish Independent on 19 May
1954, reporting that in the third year of the grants being made available
only eight hundred grants had been paid, gloomily calculates that ‘unless
this figure can be improved on, it will be nearly 500 years before every
farm has water on tap in the kitchen’.
However, due to the combination of grants, the support of the vol-
untary rural organisations and intense promotion, particularly by REo,
the rate of connection rose, albeit slowly. Sometimes it was helped along
by a little judicious blackmail on the part of the woman of the house. In
the Oldcastle area the Area Organiser’s notes for April 1955 record that
one farmer’s wife in his area had informed her husband that she would
carry no more water in buckets if he did not sink a well before the end
of June. 1956 saw 1,100 pumps sold by the Ess and by 1959 a Local
Government committee reported that 12% of rural dwellers now had
piped water supplies, 9% from private sources and a further 3% served
by public supply. The total still amounted to only about 50,000 houses,
while at this stage electricity had been extended to almost 250,000
dwellings. To quote from an article in The Manchester Guardian on 28
April 1959 ‘It is still possible [in Ireland] to see a man balancing two
buckets of water from the pump as he hastens home to be in time for
some favourite television act. His house may have a vacuum cleaner,
but not a tap.”
Whatever the reason for the slow progress in ‘rural aquafication’, it
could not be attributed to lack of effort by REO. In almost every issue of
REO News, at every rural electrification demonstration, in REO news-
paper advertisements, the message of water on tap was promulgated.
When in 1955 the problem of locating suitable sources of water was
identified as a major problem, arrangements were made between REO
and the Geological Survey Office to provide a water location service.
On any particular farm or in any particular locality the Gso would indicate
on a six-inch map the most likely water sources and give an estimate of
the depth at which water would be found and the amount available. This
arrangement was of mutual benefit, as the borehole results on site were
fed back to the Geological Survey thereby enabling them to update their
records. A still further advance in the finding of water came with the
advent of borehole-contractors prepared to operate on the basis of ‘no
water, no fee’.
Field demonstrations of the actual installation of a pumped water
system on a farm were also part of the REO promotional effort. Attend-

199
The Quiet Revolution

ances at these by local farmers were high — up to 100 in some cases — and
here they could see for themselves how simply and comparatively cheaply
a system could be installed.

In the years following the war suitable pumps and pumping systems
were, like most equipment, difficult to secure. The most suitable unit for
the average small farm was a shallow well pressure-storage system
operating at 20 to 40 lbs per square inch, which in 1948 sold for £32. At
first these were obtained from the Everite Company in the USA, but
devaluation of sterling and the dollar shortage in 1949 caused REO to look
elsewhere. After trying out a large number of alternatives, a German
pressure-storage system using a piston pump, the LeOwe ‘“Wasserknecht’
(later called ‘Waterpak’), was selected and many thousands of these
were installed throughout the country over the next couple of decades.

Shallow well pumps, however, depending on suction alone, would


only lift water from a depth of about twenty-five feet and pumps suitable
for deeper wells were also required in increasing quantities. One of the
most versatile of these was the jet pump introduced by REO in 1951 as
part of the American Everite range. This fed some of its output back
down the well under pressure. At the bottom of the well a special
arrangement utilised this fast-flowing jet of water to force fresh well
water to the surface. In their early stages of development, jet pumps
were able to pump from depths of around eighty feet. They were simple
in construction having a minimum of moving parts. For extreme depths
and very high output, submersible pumps, tailored to fit bores of down
to four inches in diameter and which could be installed at the bottom of
boreholes, were available. Depending on the size of the pump and
motor, these were of almost unlimited capacity and could operate from
great depths. Hence they were very suitable for supplying large schemes
from deep boreholes.

In 1953 Irish-manufactured purips became available when the firm of


Unidare entered the domestic pump field with a unit based on the
centrifugal jet principle. This was initially available as a shallow well or
deep well automatic pressure-storage system with a maximum operating
depth of eighty feet which met the requirements of most wells bored at
the time. As the demand for bigger pumps, higher operating pressures,
and deeper wells grew over the 1960s larger units were developed,
culminating with capacities of up to 2,500 gallons per hour, operating
pressures of up to 100 Ibs per square inch and lifting from depths of
almost four hundred feet.

200
Water on Tap

PIPING
Another obstacle to the installation of running water in the post-war
years was the scarcity of piping. Although this was somewhat outside
the normal field of electrical merchandise, REO took steps to locate and
stock supplies of steel tubing so that work on installations was not held
up. The jointing of this tubing, which was available only in comparatively
short lenths, was tedious and, since it involved the cutting of threads,
required a certain degree of skill which was rather off-putting to the
average householder. Two developments, however, helped to clear this
bottleneck and contributed immensely to the spread of rural water
supplies.
In 1953 the firm of Unidare in Finglas, Dublin, commenced production
of their “Hydrodare’ plastic water pipe. In contrast to the steel piping,
this was cheap, strong, light and corrosion-proof. It was reasonably
flexible and could be supplied in coils of considerable length, which
reduced the problem of jointing — in any case a comparatively simple
process — and it lent itself to merchanised methods of laying. In 1958 the
Dutch company of Wavin opened a small factory in Drumcondra,
Dublin, to manufacture rigid PVC water piping in 20-foot lenghts. Again
the piping was extremely light, strong and corrosion-proof; it could be
easily jointed and was available in a wide variety of bores. In 1962 the
company moved to a much larger premises in Balbriggan, Co. Dublin,
and in 1971 commenced manufacture of flexible piping. The ready
availability of these two Irish-made products removed one of the biggest
obstacles to the widespread extension of piped water in rural Ireland.

THE GROUP SCHEMES


It was obvious from the start that the extension of piped water on the
basis of an individual pump to each farm would be a slow process. In
addition, in many parts of the country reliable sources of water were
difficult and expensive to develop, prohibitively so if for only one user.
While the government and the local authorities were dedicated to pro-
viding public water supplies on as extensive a scale as possible, their
immediate priorities were the construction of new and the improvement
of existing supplies in the expanding urban areas and in some of the
more thickly populated rural communities. It would take many years
before they could direct their resources towards the more sparsely
populated country areas. This situation led to the development of the
Group Water Scheme, under which a number of adjoining householders
co-operated in developing a common water source.
The first record in REO of this kind of co-operation between neighbours
was in 1956 in the Kilfinny Rural Area of County Limerick. James

201
The Quiet Revolution

Cronin, an engineer in the EsB, whose parents farmed in the area,


suggested that his father and two neighbours should locate a reliable
common source of water and install a common pump to supply water to
their farms. Drinking water had to be carried a long distance. Water for
farmyard use had to be carted during the summer droughts and in the
winter, when the cows were in their ‘pounds’, the only supply available
in the farmyards was rainwater collected in large concrete tanks. Each
farmer had a few wells on the farm, but these invariably failed for three
months each summer, fed as they were by surface water from nearby
drains. Full co-operation was given by REO in selecting and providing a
suitable pumping system and advising on the layout of the supply. A
diviner was employed who located a suitable source and after some
initial reluctance, an agreement was drawn up and signed by the three
parties.
In 1957 a big breakthrough came at Manor Kilbride, in Co. Wicklow,
where a young curate, Rev. Joseph Collins, was anxious to organise a
communal water supply for thirteen families, including five farming
families at Old Court where the only water available was from a county
council tap half a mile from the centre of the group. To quote his own
words:

. .somebody mentioned that the Department of Agriculture was giving


grants for the installation of private water supplies and the idea was born
that by combining a number of these grants it might be possible to extend
the Co. Council pipeline down the road. The Department of Agriculture’s
Inspector was approached and agreed that while in theory the scheme was
practical, in actual fact it was not, because there were only five farmers
among the thirteen families.
However, he referred us to the Housing Section of the Department of
Local Government which gives grants to every householder — farmer or
otherwise — for the installation of a piped private water supply. A
telephone call to the Department of Local Government confirmed this and
to the further question of whether it would be possible for a number of
householders to have these grants and install a joint water supply came
the cautious reply that it had never been done before, that prima facie
there was nothing against it but that further enquiries would have to be
made in the Department before a definite reply could be given. In due
course came the official reply that such a scheme would have the blessing
of the Department.*

Fr Collins also got agreement to a most important principle: contri-


butions of labour to the scheme could be evaluated and included as costs
for the purpose of the grant. The result was that by the end of August

202
Water on Tap

1958 every one of the thirteen houses had a sink and piped water on tap
in the kitchen. The cost per house was £17 plus the Local Government
grant of £40 plus their labour contribution. The low costs were of course
due in no small measure to the county council allowing the use of its
water source and its quite considerable existing pipeline, and to the fact
that the only paid labour was that of the carpenter and plumber.
Once the ice had been broken, as it were, the idea of group water
schemes took off. In Manor Kilbride itself the first effort was quickly
followed in 1959 by a more ambitious scheme involving forty-four houses
and nearly nine miles of piping. At this stage the Local Government
grant had been increased to £50 per house and in addition, the Wicklow
county council gave supplementary grants of a similar amount.
Group schemes were welcomed by the central government and by the
local authorities. Because of the voluntary labour content they were
usually much cheaper than public supplies and of course they did not tie
up resources needed for urgent work elsewhere. Provided that the
standard of construction, pipe sizes, hardware etc., met the official
specification, these schemes could be incorporated into public regional
schemes at a later date. Meanwhile the householders would have enjoyed
the benefits of piped water. Indeed the sense of doing something for the
community’s benefit was in itself a strong motivator. Desmond Roche of
the Dept. of Local Government refers to the

spirit of community development which has been their moral mainspring.


Their adherence is cemented by the understandable rivalry between new
voluntary and old statutory organisations. There have been signs of a
tendency for the group scheme movement to turn its collective back on
local authorities and all their works and to seek a kind of frontier freedom
of individualism and self-reliance. The movement has been compared
somewhat irreverently to the evangelical revolt from the established
church.°

From 1956, when the first group scheme involving three neighbouring
farmers in County Limerick had been installed, the EsB had been pro-
moting this type of co-operation with a limited amount of success. Now,
with the green light given by the authorities, the concept of group water
supplies quickly took root in the rural community. Because of the
knowledge and experience which had been acquired by REO the govern-
ment requested its assistance in setting up the project. REO reacted
enthusiastically. It developed a very close liaison with the various govern-
ment Departments and local authorities. Its water advisory service,
which had been operational on a modest scale as far back as 1947, was

203
The Quiet Revolution

strengthened. It gave authoritative advice to consumers on grants, water


sources, layout of systems and the type and size of pumps, piping,
pressure tanks etc. best suited to cope with particular situations. In
addition, using the Installation Contracts organisation in the various ESB
Districts, REo frequently tendered for and obtained the contracts for the
installation of the pumping and control systems for group schemes of
different sizes. In this area, as with much of its installation work, the ESB
saw its role as that of a trail blazer. From the earliest days of the Rural
Electrification Scheme it had been involved in the installation of pumps
and pumping systems and in the solution of farming water supply
problems. It had developed expertise in both pumping hardware and
control systems and had, over the years, kept in close touch with
developments. In particular it had identified pump types and models
which were especially suitable for Irish conditions. It welcomed the
opportunity, whenever its tenders were successful, to introduce the most
suitable equipment and to demonstrate proper installation techniques
which could be observed and followed by other contractors. The intro-
duction of each new model was accompanied by an intensive training of
service electricians in its maintenance and repair and by the stocking of
spare parts so as to ensure a competent after-sales service. It was hoped
that if REO set high standards in the beginning other individuals and firms
in the business would follow suit.
A typical group installation was that at Ballyduagh near Cashel where
in 1958 six farmers dependent for all their water on a single well decided
to install a pressure-storage system at the well head, piping the water
from there to their dwellings and farmyards. The farms were widely
separated, requiring 6,000 yards of piping, but by doing most of the
heavy work themselves and by availing of the grants, the farmers kept
net cost to about £110 each. This particular installation got wide publicity
in the local press and set off a chain reaction of group schemes in the
region.
As group water schemes were an ideal example of community devel-
opment by co-operation it was little wonder that Muintir na Tire fostered
them enthusiastically. A very well publicised scheme, which set a head-
line for many subsequent schemes, was the Kilally and Ballinrish scheme
sponsored by the Kilworth, Co. Cork, guild of Muintir. The value of the
labour provided by the thirty participants was estimated at £1,200. The
total cost was £3,500 and the average contribution after payment of
grants was less than £17 per house, shared according to ability to pay.
The group scheme idea grew quickly and became widely accepted as
a meansof providing a large section of the rural community with piped
water at a reasonable cost.’ From groups of a few adjoining farmers, the

204
Water on Tap

schemes grew to embrace whole communities. A case in point was the


parish of Killeigh near Tullamore where in 1965 the parish priest,
encouraged by the success of an earlier scheme for ninety-five houses in
Geashill now developed plans for supplying the other houses in his
parish. REO was asked to quote for the pumps and equipment. With the
closest co-operation between the parties involved, the parish priest, the
parish committee, the consulting engineer, Local Government engi-
neers, REO and of course the parishioners themselves, a group water
scheme was developed to bring piped water to over four hundred houses.
It was thus as large as, indeed larger than, many public water supply
schemes, and more advanced in design. It had a main reservoir of
100,000 gallons capacity, six multi-stage centrifugal pumps with stainless
steel impellers, guide vanes and shafts and a sophisticated electrical
control system to ensure reliable and economic operation. The capacity
of the well was phenomenal. Seven years later when over five hundred
houses had been connected, it was calculated that over 100,000,000
gallons had been supplied from the one well without a single break.
Over the fifties, however, the ICA was not at all satisfied with the
progress in providing water on tap. By 1960 when almost a quarter of a
million rural homes had electricity, only about 50,000 of these had yet
installed piped water supplies. On the 14 August 1960 a meeting was
called by the Association at its headquarters in 23 St Stephen’s Green,
Dublin, with the object of launching a special campaign to develop rural
water supplies. In inviting interested parties to attend, the Association
had cast its net far and wide. The minutes record the attendance of
Deputy Paudge Brennan, Wicklow County Council; Patrick J. Dowling,
Engineer-in-Charge, Rural Electrification; Tom Finegan, NFA; Michael
Owen Fogarty, Macra na Feirme; J. J. Byrne, Agricultural Institute;
Noel Manahan, Unidare Ltd; P. J. Meghen, Muintir na Tire and
Limerick County Manager; C. O’Neill, Farmers’ Gazette; Larry Sheedy,
Trish Farmers’ Journal; Joan Crady, Muriel Gahan, Nan Minahan,
Phyllis O’Connell, B. Smith and D. Tomlin of the ica. Aine Barrington
was in the Chair.
The meeting set up a joint committee in the Campaign for Rural
Water Supplies which over the next couple of years, in addition to widely
publicising the need for water on tap, did much to co-ordinate the efforts
of all parties concerned. It provided a ferum for discussion on such
topics as the impact of regional schemes on the rates, interim payments
of grant instalments, block payments of grants in the case of Group
schemes and grants for non-vested county council cottages.
The main stated thrust of the Committee’s efforts, however, was ‘to
publicise the need for water on tap and the urgency of installing it by

205
The Quiet Revolution

whatever means is found to be the most practical and economical’. It


mounted conferences and exhibitions on Rural Water Supplies at An
Grianan and in the Mansion House in Dublin. In all cases the fullest
co-operation was given by REO, particularly in mounting the exhibitions.
The Minister for Local Government indicated his support by attending
the An Grianan conference and was backed up by Desmond Roche of
his Department, who gave a comprehensive survey of the existing
position. Fr Collins, the veteran of the Kilbride schemes, spoke of his
experiences of communal ‘aquafication’ to an audience of over two
hundred, most of them women. He finished his lecture with a rousing
challenge:

Many men in rural Ireland are not yet fully aware of the advantages of
piped water supply. Very often they will get a water supply with the
advantage of the stock in mind rather than the housewife, and very often
if the stock don’t want water, the housewife has to do without it also. This
is just where you take over. Let your menfolk know you will not tolerate
such treatment. Tell them you don’t want to end up in your old age with
a bad heart got from dragging buckets of water over long distances.
Proclaim to them the gospel of rural water. Try to get them to form local
groups and when they form such a group, get them to co-operate one
hundred per cent in the carrying out of the work. Rural Electrification was
the rural social achievement of the 1950s; let Rural Aquafication be the
rural social achievement of the 1960s. Tell your menfolk what has been
done in Kilbride and many areas of Ireland and say to them: ‘Go and do
thou in like manner’.

In its work the Committee had full and enthusiastic backing from the
farming press. The /rish Farmers Journal in particular issued a number
of supplements with contributions from experts on various aspects of
rural water supply. Some of these were very extensive, as for example
the twelve-page supplement issued to coincide with the An Grianan
conference.
At the Turn of the Tap exhibition held in conjunction with the water
conference in April 1961 a competition was held at the REO stand with a
prize of a complete set of bathroom equipment, and was continued at
the REO stand at the Spring Show which followed a few weeks later.
Competitors, who were required to be rural residents, were asked to list
in order of preference, twelve suggested benefits which water on tap
would bring. Almost 8,000 entries were received and the competition
was won by Mr F. Tomany from County Monaghan. The order of
benefits as seen by the majority of competitors was as follows:

206
Water on Tap

Saves time and labour spent in carrying buckets of water from the well
Ensures that water is available at all times at the turn of a tap
Enables a bathroom and toilet to be installed
Improves personal hygiene
Provides a safeguard in the case of fire
Makes washday easier for the woman of the house
Eases the task of preparing food, especially vegetables, and makes it
more hygienic
Makes possible the installation of a hot water system
Makes automatic watering for all stock possible
With hot water, it ensures hygienic and easy cleansing of dairy utensils
Ensures healthier cattle and higher milk yields
Simplifies the problem of cleaning stock buildings.

It is worthy of note that the first eight placings referred to the standard
of living in the home. The advantages out on the farm were relegated to
the lower positions. This result left no doubt about the women’s priori-
ties, but it also carried some suggestion that farming was still regarded
more as a way of life than as a competitive business. (This changed
somewhat in the seventies. The September 1975 issue of Prospect, the
ESB marketing magazine, quotes the example of Cloonminda,
Co. Galway, where the cow population had increased by 230% and dry
stock by 100% since the group water scheme became operational in
1972.)
All this activity had the desired effect of boosting demand for water
on tap over the sixties. The estimated figure for electric water pumps in
rural consumers’ premises on 31 March 1961 was 25,000, of which 8,000
were supplied by the Ess. This gave a figure of 9% of all rural electricity
consumers with water pumps. Figures given in a paper presented to the
Institution of Water Engineers by Bernard J. Tighe, senior engineering
inspector, Department of Local Government,* show that by 1966 grants
for private installations were running at about 6,000 per annum.By
September 1966, 173 group schemes had been completed serving 2,018
houses; 153 serving 2,579 houses, were in progress and 1,189 further
schemes were pending. The costs to that date were:

Reservoir schemes: 49 schemes (906 houses), average cost £166 per


house.
Automatic pressure installations: 49 schemes (355 houses), average cost
£169 per house.
Extensions from mains: 75 schemes (757 houses), average cost £117 per
house.

207
The Quiet Revolution

At that stage group water schemes were making water on tap available
to houses at the rate of 2,000 per annum.
The regional schemes were of a much larger size. An analysis of
eighteen such schemes serving 11,957 houses gave an average of 1,110
houses per scheme with a range of from 185 to 2,509. The cost per house
averaged £344 with a range of from £212 to £520. Thus the group schemes
cost about one half of the regional schemes per house supplied. However,
the costs are not strictly comparable. The regional schemes included
expensive headworks for water treatment and the ensuring of a good
margin for increased demand not usually provided in group schemes and
had larger pipe sizes to permit expansion. They did not, however, extend
supply to the houses but provided a supply point on the public roadway
to which householders could make a connection at their own expense.
On the other hand, the costs quoted for a group scheme included the
provision of a kitchen sink in each house, complete with cold water tap.
It was intended that in the course of time most group schemes would be
incorporated in regional schemes thereby ensuring a more reliable supply
source and a more uniform and carefully controlled quality of water.

HOT WATER ON TAP IN THE HOUSE AND IN THE DAIRY

REO’S promotion of water did not end with the provision of the cold
water tap and the kitchen sink. The benefits of hot water on tap were
constantly stressed. The traditional farmhouse building, however, did
not lend itself easily to the roof tanks required for the conventional hot
water system. A number of inexpensive electric water heaters were
therefore developed to provide, in conjunction with a pressure-storage
pumping system, hot water on tap at low capital and running costs.
These included a self-contained cistern water heater which could supply
a number of taps and which did not require a roof tank. There was also
available a very basic and inexpensive non-pressure water heater with a
single outlet, mainly for use at the kitchen sink.
With the increasing emphasis on high standards and the very large
increase in milk production which followed Ireland’s entry to the EEC,
the availability of adequate hot water on tap in the dairy became of great
importance. As herd sizes grew the number of farmers using milking
machines increased, as did the demand for hot water to cleanse the
equipment. A survey commissioned by the Ess in 1972 indicated that of
the 40,000 dairy farmers having ten milch cows or more (usually the
figure at which machine milking was found necessary), only about 6,000
had any form of water-heating installed in the dairy. The remaining

208
Mrs Hetherington of Errill did not have water on tap in her own home until the
Errill group scheme started.

209
The Quiet Revolution

34,000 were dependent on hot water carried from the house or even
brought back from the creamery in the milk cans. An intensive campaign
for the installation of dairy water-heating was launched and water heaters
specifically designed for dairy requirements were developed and manu-
factured by Irish firms.
In many houses also the provision of a bathroom, now coming to be
regarded as a necessity following the installation of water on tap, pre-
sented problems. In the larger houses an existing room was frequently
available for conversion, but in the case of the smaller houses it usually
had to be built on as an addition to the house. This presented a problem
as the do-it-yourself ability of the average rural householder did not
stretch to the design and construction of additions to the house. The cost
of employing a professional contractor was frequently beyond people’s
resources. Towards the end of 1959 the rural area engineer in the
Kilmyshal area on the Carlow-Wexford border, Joseph McBride, with
the help of a co-operative builder drew up a simple design for a combined
bathroom and toilet which was published with full construction details
in REO News and which undoubtedly helped many rural families to
install their first bathroom at a reasonable cost. The estimated cost on a
do-it-yourself effort was £226 including septic tank. At the time the
grants from the government and from the local authority generally
covered up to two-thirds of the cost.
Throughout the sixties and seventies the campaign for the installation
of running water continued with the full backing of the rural voluntary
organisations, central and local government and not least REO, which
regarded its mission incomplete unless water on tap complemented the
extension of electric power. In 1959 an EsB survey showed that about
106,000 or 46% of rural homes supplied with electricity had water on
tap. By the beginning of 1980 a similar survey (Table 7) showed that of
425,000 rural electricity consumers 336,000 or 79% now had water on
tap available through individual schemes, group schemes or the large
regional schemes.
The remaining premises, including about 10,000 houses without elec-
tricity supply, thus amounted to about 100,000. About 70% of these
were located in the ‘west’, either in Connacht itself or in Clare, west
Cork, west Limerick or Kerry in the south, or in Donegal, Cavan,
Monaghan or Longford (Fig. 2). As part of an overall programme for
the stimulation of agricultural growth, the EEC in 1981 approved an
allocation of money for the development of rural water supplies in these
counties. This enabled a very generous scale of grants to be approved by
the government, amounting to 90% or a maximum of £750 per house in

210
Water on Tap

Gaeltacht areas and 80% or a maximum of £600 per house in non-


Gaeltacht areas with special grants available for island communities.
REO Water Advisory Service, though now part of the larger Agricul-
tural Advisory Unit of the Ess, still plays a big part in the organisation
of schemes and of promotional propaganda in the provincial media. It
supplies information on available grants, how to estimate water require-
ments, water testing and treatment, pump selection and installation,
electricity supply to the pumphouse, sizes of piping etc. These features
are strongly supported by advertisements from well drillers, pump sup-
pliers and group waterworks contractors. A comprehensive booklet
Rural Water Supply — How to go about it which gives information on
almost every aspect of the problem was published by the Unit in early
1982 and supplies made available to the various authorities and the
public. It can thus be seen that the thrust by the Ess for rural aquafication
to accompany rural electrification, which was commenced in the late
1940s, continues as strongly as ever, as does its co-operation with the
central and local government authorities to ensure that within the short-
est possible period water on tap in the rural home will be as universal as
electricity is at present.

Citas:

Practical demonstration in pump installation by the area organiser.

211
The Quiet Revolution

TABLE 7
Rural Water Supplies in Ireland at Beginning of 1980

ESB No. of rural


District electricity consumers With water on tap

No. %
Athlone 37,795 D3,523 67
Galway 39,974 28,781 72
Sligo 67,058 46,941 70
Tralee 35,997 23,798 70
Portlaoise 37,597 32,709 87
Dundalk 47,148 38,661 82
Limerick 48,312 37,200 TE
Dublin South 5,165 4,752 92
Dublin North-west 16,331 16,331 100
Waterford 42,927 37,776 88
Cork 48 ,999 43,609 89

Total 425 303 335,881 79

Source of Household Water Supply

Piped Water
%
Public water mains 33.9
Group water schemes* 15.4
Own electric pump 28.6
Miscellaneous 1s"

79.4%

Other Sources
Roadside tap or pump 4.7
Well 19

20.6%

(esB Consumer Survey, 16 September 1980)


*From information supplied by Department of the Environ-
ment at end of 1981, about 50% of Group Water Scheme
consumers are supplied from extensions to public mains and
50% from own sources, the latter systems almost always involv-
ing electric pumps.
Note: The 100,000th consumer in a group water scheme was
connected in October 1982.

212
CHAPTER EIGHTEEN

Electricity on the Farm

WHEN IT CAME TO DEVELOPING the use of electricity in agriculture, the


small size of the Irish farm, the limited cash resources of the average
farmer and the unsatisfactory markets for his produce were inhibiting
factors. A comparison with Britain made in 1956 (Table 8) showed that,
counting only farms of over five acres, 46% of British farms were over
fifty acres, as against only 28% in Ireland.
The cheap food policy of Great Britain, until the late 1960s the only
export market of any consequence open to the Irish farmer, held prices
down. The British farmer was compensated by various grants and sub-
sidies which were not of course available to Irish farmers. In contrast
also to the British situation of considerable paid labour was the fact that
in Ireland most of the farms were family units with very little paid help.
Mechanisation in the farmyard would in such cases appear to save the
farmer little money and such investment as could be made out of meagre
resources was generally in land improvement or possibly in the purchase
of stock or a tractor. There was of course the odd farmer who was quick
to appreciate the advantages of farmyard mechanisation, and who was
in a financial position to avail of these; but in general up to the mid-
1960s the development of electricity in the Irish farmyard was much
slower than in the farmhouse.
From the mid-sixties on, with the prospect of the opening up of
European and other foreign markets, a change was discernible in the
style of Irish farming. The mixed farm, with its ‘little bit of everything’
began to give way to the specialised farm with one or two strong
enterprises, dairying, sheep, tillage, poultry, pigs etc., with a far greater
output than hitherto. This style and intensity of operation, displaying as
it did a more scientific and commercial approach than formerly, placed
far more emphasis on labour saving and productivity and consequently
on electrification. There were thus two distinct phases of farming style
and activity encountered by REO with the watershed occurring somewhere

213
The Quiet Revolution

around the middle 1960s. These two periods presented different chal-
lenges and problems.

TABLE 8
Number and percentage of farms of over 5 acres in each size
group in Great Britain and Ireland 1956.

Size Great Britain Treland


(Acres) No. % of Total No. % of Total

5-15 91,000 Bes: 60,776 21.0


15-50 99 ,000 28.0 147,986 51.0
50-100 69,900 20.0 51,755 17.9
100-200 50,160 14.6 21,928 7.6
Over 200 41,420 11.6 7,249 2
Source: Paper, ‘Rural Electrification in Ireland’ given to 1956 Annual Conference of
British Electrical Development Association by P J Dowling, BE, BSc, ARCSc I.

THE EARLY YEARS


— 1946-65

Irish farming at the end of World War II was not in a healthy state.
While the agricultural advisory services and rural vocational schools
made trojan efforts to raise farming standards they had to contend with
a serious lack of resources and a considerable apathy among the farming
population. In every county, however, a small number were prepared to
experiment and it was chiefly among these that REO concentrated its
promotion and development. One of the most promising areas was in
grain processing, conveying, drying, ventilated storage, grinding and
rolling. It was considered that a ‘grow your own, process your own, feed
your own’ policy with regard to feeding stuffs would reduce considerably
the feeding costs of livestock and permit profitable expansion. With the
help of electricity, the handling, drying and processing of home-grown
grain could be carried out at a reasonable cost and without cutting in
unduly on the farmers’ time.

Grain Milling and Grinding 0 Among the first farmyard appliances


offered in 1947 was a 3hp plate mill of Canadian manufacture with a
built-in motor which had a take-off pulley for driving other appliances,
such as rootpulper, small circular saw, grindstone or milking machine.
It cost £45 when introduced and had an output of up to 3 cwt per hour,
depending on the fineness or coarseness of the grind, for three units of
electricity. It was at the height of its popularity in the late 1940s and

214
Electricity on the Farm

early 1950s when increased costs and dollar difficulties caused REO to
look for a cheaper substitute. Nevertheless, the Canadian Woods grinder
continued to give good service. In 1980 a farmer in East Donegal
reported that his Woods grinder was still working happily. After thirty
years he had found that replacement plates were still available from the
ESB.
The substitute for the Woods grinder was the German ‘Bauerngold’
(literally ‘farmers gold’ and rapidly anglicised to ‘Barngold’) which,
instead of steel plates, had grindstones of a quartz-concrete mixture.
Powered as it was by a Lhp motor, it had only one-third the output of the
Woods and so took three times as long to produce the same amount. As
the minimum amount of attention was necessary, this caused no problem:
the farmer could leave the machine running and attend to other tasks.
The amount of electricity used per hundredweight of meal ground was
about the same as for the larger machine, about one unit.
This illustrates a message which REO was constantly preaching to
farmers. For the average or small farmer, there was no need to invest in
large machines where electricity was concerned. Taking the grinder as
an example, if the farmer’s requirements were, say, 6 cwt per day of
ground meal, the electricity cost was much the same whether it was
produced by a 3hp machine in two hours or by a lhp machine in six
hours. Provided hoppers and bins were of adequate size, the smaller
machine could operate without attention and, if provided with a simple
pressure switch, could be made to switch itself off at the end of the
operation. Grinding by tractor, on the other hand, required constant
attention and so a large machine was desirable to reduce the time taken
to a minimum. The same applied to pumping. A ¢hp motor could pump
250 gallons per hour from the average well continuously without any
supervision. This added up to 1,000 gallons for four hours for one unit
or 6,000 gallons in twenty-four hours for four units of electricity at a cost
of less than 6d. (23p). In the case of the Barngold mill, the message was
well taken, as, despite its small capacity, it rapidly became a best seller.
Thousands were installed throughout the length and breadth of the
country.
For some purposes, such as poultry-feeding, a hammer mill was
considered more suitable, as it broke up the husk of the grain more
effectively than did the stone or plate mill. The grain, having been
broken up by the hammers, had to pass through a screen or sieve which
could be as fine or as coarse as required. Thus, a far more precise control
of the product could be achieved. Imported hammer mills were rather
expensive, but in 1954 an Irish hammer mill was developed by Patrick J.
Tobin who had a small engineering works in a truly rural setting at

PAIS)
The Quiet Revolution

Edermine near Enniscorthy, Co. Wexford. This was a good example of


the feed-back effect of rural electrification and the opportunities it
afforded to native enterprise. The mill was powered by a 3hp motor and
had an output of from 2 cwt to over 12 cwt per hour, depending on the
fineness of the screen and the type of grain being ground. The Tobin
mill proved a most popular machine among Irish farmers. At a later
stage, a similar, but smaller, hammer mill of 13hp was produced by AcEc
of Waterford.

Grain Drying O Grain drying by electricity on the farm was proving


quite successful in Great Britain, but the average Irish farm needed a
smaller dryer. For the medium to small grain growers, particularly those
who wished to store and feed their own grain a series of small tray-type
grain dryers was developed by reo. These were on the do-it-yourself
basis, REO providing the fan-heater units and complete plans for the
building of the dryer. They were of three sizes, with heater units of 5kW,
10kW and 15kW, which were capable of a moisture extraction of 6%
from 3 ton, 1 ton and 13 tons of grain, respectively, in 10 hours. While
some hundreds were installed, their low throughput and labour inten-
siveness inhibited their popularity, particularly in later years, when the
acreage of barley grown had increased greatly.

Grain Conditioning O For short-term storage, pending delivery to the


grain merchants, a system of grain conditioning was advocated which
involved the quick cooling of the grain (i.e. within 10 days), to 10°C by
ventilating it with cold night air. This system ideally studied the producer
in the 30-ton range, since a single 12hp fan would achieve this target.
Investigations carried out by An Foras Taluntas indicated that green
wheat with an initial moisture content of 21% could be stored without
deterioration for up to six weeks by this method.

Grain Handling U For handling grain, a very simple screw or ‘auger’


grain elevator powered by a fractional hp motor was very popular,
eliminating as it did much of the back-breaking shovelling associated
with grain movement.

Equipment Prices 1 Over the thirty years of the scheme, there were of
course many price changes, usually upwards. In order to give some idea,
however, of prices, those ruling for the most popular items in August
1960 (about half-way through the scheme) are given, including motor
and starter where these were separate from the appliance:

216
Above: Grain milling and grinding — the Woods 3HP grinder with
auxiliary take-off pulley. Below: Safe from crushing by ‘mother’ in
their electrically heated creep.
The Quiet Revolution

Barngold mill £49.00.0


Barngold oat-roller £60.00.0
ACEC hammer mill £59.10.0
‘Tobin’ hammer mill £81.00.0
Bentall automatic
hammer mill £129.07.0
Mayrath auger elevator £28.10.0 Herendineot
£30.00.0 tags
£35.00.0

SkW ~~ £35.10.0
Grain drying
10kW £46.12.0
fan heaters
1SkW £65.15.0

For the larger farms more sophisticated equipment was available, limited
only by the depth of the farmers purse and the economics of its adoption.
The larger the throughput and the wage bill, the stronger the case for
mechanisation and automation. In the early years of the scheme, how-
ever, these farmers were very much in the minority. Once the EEC
markets opened up in the 1970s, a considerable number of farmers, or
co-operatives of farmers, engaged in specialist production, which
involved much larger units than those described.

The Farm Workshop U1 Deputy Jim Hughes, one of the TDs who was
an enthusiastic advocate of rural electrification at its very inception
confided to P. J. Dowling, when the latter was Engineer-in-Charge, that
of all the possible benefits of the scheme, the one which most attracted
him initially as a farmer was the prospect of an electric grindstone to
keep his tools sharpened.
Rural electrification coincided with increasing mechanisation on the
farm. New and second-hand tractors flooded in from Britain after the
war and electricity helped farmers to carry out much of their own
maintenance and repair. One of the most popular appliances was the
electric drill with its many attachments and one of the larger drill
manufacturing companies conducted a week-long factory seminar for
REO salesmen. Another versaatile appliance was the single-phase electric
welder, which sold for about £50-£100, depending on size. With a drill,
a welder and a few simple hand tools, the farmer could not alone carry
out extensive machinery repairs, but also fabricate many items of his
own design for use on the farm or around the farmyard.
The rural blacksmith also received a new lease of life just when he
needed it. Horses were disappearing as mechanisation increased, but

218
Electricity on the Farm

with an electric fan for his fire, an electric welder and a small range of
powertools, the blacksmith was now into a new and fast-expanding
business in the repair of farm machinery.
It was hoped that the thp or $hp motor (selling for about £10 and using
only one unit in 3 to 4 hours), combined with a self-assembly speed
reduction unit (costing about £8), would prove a popular and inexpensive
introduction for the smaller farmers to the cheap motive power now
available for mechanisation of many farmhouse and farmyard tasks. The
slogan was “Whenever there is a handle to be turned, electricity can do
it better and more cheaply’. In particular, it was hoped that it would be
applied on a large scale to such regular chores as churning in the
farmhouse and root-pulping in the farmyard, both well within the capa-
bilities of the small motor.
However, while some consumers were receptive, hopes for general
acceptance of even such a modest level of mechanisation were not
realised. The only really wide-scale use of the fractional hp motor in the
early years was for water-pumping, generally as an integral part of a
pressure-storage system.

REGENERATION — THE SIXTIES

By the end of 1958 there were signs that a fresh breeze was preparing
to blow away the economic doldrums of the fifties, demonstrated by
contrasting extracts from the Jrish Review and Annual Supplement pub-
lished by The Irish Times for the years 1957 and 1958. In the article on
the economy for 1957 we read that ‘Production continues to languish,
emigration to flourish. . . . Emigration is attributed . . . to economic
backwardness and our economic backwardness is due to lack of produc-
tivity ... the cattle trade is faced with imminent disaster, as more
and more areas in Great Britain are closed against beasts that are not
free from tuberculosis’. Later we read of a decline in the export of eggs
and poultry.
In contrast, in the issue for 1958, the Economic Survey opens thus:

A bewildering succession of new developments marked the year 1958 in


the Irish economy. New events and new ideas crowded the scene and a
new spirit, very different from the dejection of 1956 and most of 1957
motivated the actors . . . agriculture, industry, external trade, finance and
the Government sector were all affected in varying degrees by the quick-
ening tempo of economic activity and the recovery of confidence in future
prospects.

219
The Quiet Revolution

Much of the stimulus for this change was provided by the publication by
the government in mid-November 1958 of a White Paper, ‘A Programme
for Economic Expansion’, based on a study by T. K. Whitaker, Secretary
Department of Finance, entitled ‘Economic Development’. The pro-
gramme emphasised that increased agricultural production would for the
most part be for sale abroad and must, therefore, be competititve. “State
assistance should concentrate less on price supports and guarantee than
on measures designed to bring about increased agricultural productivity,
that is lower production costs per unit of output. It is the only way that
will get increased agricultural production and achieve incomes on a
sound and permanent basis.’ The programme went on to analyse the
problems of the different sectors of agriculture and outline a strategy for
dealing with these.
While the major advances in the rural economy were still some way
into the future, it could be said the 1958 marked the beginning of a
change in attitude which resulted in the upsurge in agricultural produc-
tivity and incomes, which was such a strong feature of the late 1960s and
particularly the 1970s following Ireland’s accession to the EEC. This was
reflected in the growing use of electricity as an aid to production. On the
other hand there is little doubt that without electricity Insh farmers
would have found it impossible fully to exploit the opportunities provided
by the opening up of the Common Market.
Much of this new dynamic in Irish agriculture was provided by the
foundation, also in 1958, of An Foras Taltintais (The Agricultural
Institute). The Institute was set up following an agreement between the
governments of Ireland and the United States and financed out of
Counterpart Special Account money provided by the United States
under the Marshall Aid Programme ‘to review, facilitate, encourage,
assist, co-ordinate, promote and undertake agricultural research, includ-
ing horticulture, forestry and bee-keeping’.
Prior to the setting up of the Institute, facilities and funding for
agricultural research were very limited. Research into and development
of the application of electricity in agriculture were almost non-existent.
It was obvious to REO from experience in other countries that electricity
had much to offer agriculture, but it was equally obvious that electrical
applications which were successful elsewhere were not going to be simply
and easily transferred. In the 1950s a limited amount of work was done
by REO on certain applications of electricity to agriculture and horticulture
with the enthusiastic help of the Agricultural and Horticultural Depart-
ments of University College, Dublin. These mostly involved the testing
out under Irish conditions of techniques developed as a result of basic
research elsewhere. The Agricultural Department carried out a pro-

220
Electricity on the Farm

gramme of experiments in environmental control for stock-rearing, crop


drying and barn hay drying by electricity. A parallel horticultural pro-
gramme also involved environmental control. This included potato
sprouting, soil warming, plant irradiation, supplementary lighting, grow-
ing rooms for tomato plants, automatic watering and mist propagation.
Resources for basic research or for the widening of the scope of the
experiments were, however, very limited.
The setting up of The Agricultural Institute, with its vastly greater
resources, permitted a much wider research programme into the appli-
cations of electricity to agriculture and horticulture. REO worked in close
co-operation with the Institute in the appropriate programmes. The
research work of the Institute was not of course confined merely to
agricultural technology. It also embraced agricultural economics, farm
management, marketing, farm buildings and rural sociology. The results
of its wide spectrum of research were transmitted to the farming sector
through the County Committees of Agriculture and the local agricultural
advisers. Before long a new and more enlightened approach to farming
and its problems was discernible. This was helped by the educational
work of the voluntary organisations and by the increasing availability of
second-level education. One of the great changes in the case of many
small farmers was the transition from the old system of mixed farming
into more profitable specialisation patterns which suited their particular
circumstances — dairying, pig breeding, poultry production, etc. As the
pattern changed and with increasing emphasis on efficiency and produc-
tivity, there was an increasing usage of electric power. As discussed
elsewhere, one of the big problems facing the Ess in the late 1960s and
1970s was reinforcement of the rural electricity networks to meet this
rapidly increasing demand.

Milk Production 0 One of the greatest contributions of rural electrifi-


cation to agricultural production has been in the dairying industry. From
1970 to 1979 milk production increased by 56%, from 640 million to
1,000 million gallons. About 85% of this went into manufacturing,
mainly butter and cheese of high quality for which markets had opened
up in the EEC. Butter exports increased from £10.6 million in 1970 to
£198 million in 1979 and cheese exports from £6.2 million to £85 million.
In the same period the agricultural work force dropped by 22% from
276,000 (26% of total at work) to 214,000 (18% of total).
This increase in milk production was due not only to higher yields per
cow (the number of cows had increased in the same period by only
23% — from 1,713,000 to 2,208,000), but also to the higher productivity

joel
The Quiet Revolution

in the farmyard and in the dairy by a much smaller work force. Well-
designed herd handling and milking methods using milking machines
permitted one-person milking of even very large herds. Electrically
refrigerated milk coolers and bulk milk tanks permitted quick cooling,
longer storage periods, higher quality milk and more cost-effective
collection methods. Electric pumps ensured adequate water supply for
drinking, pre-cooling and washing, while specially designed electric dairy
water heaters with automatic time-switch control provided adequate hot
water for personal cleanliness, udder washing and cleaning of milking
equipment, thus enabling the producer to meet the exacting hygienic
standards required by modern markets. Specially designed electric
pumps enabled the high-pressure hosing down of byres, milking parlours
and yards, also helping to keep up high standards of hygiene required on
the modern dairy farm.
The number of milking machines grew from about 1,000 in 1946 and
10,000 in 1960 to an estimated 60,000 in 1979. The number of milk
coolers, including refrigerated bulk milk tanks, grew from about 30,000
in 1970 to 46,000 in 1979, the increase being mainly in the refrigerated
bulk tanks.

Pig and Poultry Production 0 Modern living styles and marketing


requirements have ensured the demise of the traditional picture of the
pig as ‘the gentleman who pays the rent’ in practically every Irish
farmstead. The tendency is more and more towards a concentration of
pig production amongst a smaller number of producers. In 1970 and 1980
the number of pigs produced was about the same, at about 1.1 million,
but the number of producers had dropped from 24,000 to about 10,000.
Some of these are very large producers — an article in The Irish Times
in March 1982 quotes twelve large units in County Cavan with some
10,000 pigs or more each. With such large concentrations of animals,
precise control of the environment is vital for optimum growth and
prevention of disease. Electric fans for ventilation, electric floor heating
and infra-red lamps in creeps and farrowing pens, electric pumps for
slurry disposal and for washing down, all contribute to efficient produc-
tion and hygienic conditions.
Poultry production has followed the same lines as that of pigs, larger
concentrations of birds requiring similar control of environment. With
such large numbers, milling and mixing of feed can be most economically
carried out on the premises and again the availability of electricity
permits efficient production with the minimim of attention. For the very
large units, mechanised feeding equipment has been developed, which _

mee
Above: Effectiveness of electric power for cleaning at an early
morning demonstration. Below: Hot water for washing the udder
and for stimulating the ‘let down’ of milk preparatory to the use of
the electric milking machine.

223
The Quiet Revolution

greatly increases the number of animals or birds that can be looked after
by one person.

Waste Management and Cleaning in the Farmyard U The intensified


production methods on the farm and in the farmyard have brought with
them their own sets of problems, particularly in the area of waste
disposal. In the traditional small farm, with straw bedding for compar-
atively few animals, the disposal of waste first to the farmyard dung heap
and then on to the land was a manageable task for the farmer and was
a system which had provided an ecological balance over thousands of
years.
New and more intensive methods of rearing, feeding and housing
stock necessitated by larger animal numbers and a smaller work-force,
increased areas under concrete and increased quantities of silage have
given rise to new problems of effluent disposal. The effluent waste in
dairy herds from milking sheds, parlours and adjacent areas consists of
solids, soiled water, detergents, seepage from dungsteads and silage pits
and all waste water that has been in contact with animal manures in
calving boxes, wintering sheds etc.
Intensive strawless housing means that in most cases to-day, animal
manure is slurry, a semi-liquid material. The amounts involved are large.
A herd of a hundred cows or the same number of 500-kilo beef cattle
will produce 900 gallons of slurry per day. Two problems immediately
arise — the removal of the slurry from the farmyard and buildings and
its disposal in such a way as not to cause pollution elsewhere. The
fertiliser value of this slurry can be quite high; for example at 1980 prices
its use on a two-cut silage could save up to £25 per acre in replacement
or artificial fertilisers. On the other hand, its excessive use, particularly
in drumlin country, where it slides off with rain into the rivers and lakes,
can be most harmful to fish life.
Electrically operated units are available to deal with all kinds of slurry
pumping whether it is from underground to overhead tanks; from storage
tank to tanker/spreader or direct to the land by sprinkler systems. Some
pumps are equipped with cutting mechanisms for dealing with foreign
matter (straw, twigs etc.) in the main storage tank. The power required
varies but many units are installed with motors as low as 3 to Shp.

Grain Storage (] With the more intensive stock rearing the amount of
barley grown for home feeding increased from 335,000 acres in 1968 to
645,000 acres in 1978.' While the small tray-type electric dryers devel-
oped during the 1950s and early 1960s had not been very successful,
chiefly because of the amount of handling involved, a more promising

224
In the farm workshop the electric grindstone is vital for keeping tools sharp.

fps)
The Quiet Revolution

method of storing comparatively large quantities was developed by An


Foras Talintais in co-operation with the EsB. This system was based on
research carried out by the British Electricity Council and, in contrast to
previous systems using both motive power and heat, used motive power
only to blow cold air through a holding bin. It was much more economical
than those involving heat. It enabled barley to be held safely for home
use for up to six months. Because of its cheapness and simplicity it was
widely adopted by even the smaller farmers. By 1976 about 10,000
installations of this type of drying/storage unit were in operation.

ELECTRICITY IN HORTICULTURE

Before the advent of An Foras Taltntais a notable amount of research


and experimentation was carried out by Professor Edward J. Clarke and
Dr Joseph V. Morgan of the horticultural department of University
College Dublin in conjunction with REO. The earliest experiments were
in the area of electrical soil warming for hotbeds and bench warming in
the glasshouses for the earlier production of plants from seed and
propagation from cuttings. In the case of the latter the addition of
artificial misting using high pressure pumps and electronic control gave
very positive results and the system was quickly adopted by commercial
growers.
Experiments on supplementary lighting for the earlier production of
tomato plants were also very successful, and were followed by the
development of growing houses for commercial production of plants.
These gave, in a very small insulated house, a completely controlled
environment with lighting, heating and watering controlled so as to
provide ideal growing conditions with the minimim of energy input. This
technique was also quickly adopted by commercial growers to produce
strong early plants and thereby early tomatoes, which earned the highest
prices. With the establishment of the Kinsealy Research Centre of An
Foras Taluntais, research in these areas continued on a more intensive
scale, again with close co-operation from REO. Automatic watering and
ventilation were also subjects of research and experiment which helped
the Irish tomato industry to expand rapidly in the 1960s and early 1970s.
The oil crisis of 1973 and the equally serious but less dramatic increase
in oil prices in the late seventies have, however, had a disastrous effect
on this energy intensive industry. In 1973 the total cost’ of the energy
inputs per acre was under £4,000. In 1978/79 this cost had risen to £14,000
and in the 1979/80 season it approached £20,000.”
At present the industry is vigorously seeking methods of reducing its
high energy costs. Two possible fields of investigation in which electricity

226
Electricity on the Farm

might help are in the use of heat pumps and of extract heat from
electricity generating stations.

Heat Pumps UO) By upgrading heat extracted from the outside environ-
ment, a river or a lake, a heat pump can supply as heat up to four or five
times as much energy as is required to drive it. High capital costs,
however, at present inhibit the use of heat pumps for overall glasshouse
heating. Nevertheless, experiments in Kinsealy indicate that a combi-
nation of root zone warming (possibly by heat pump) and lower air
temperatures in the glasshouse might be successful in reducing costs
without undue loss of crop.

Extract Heat from Electricity Generating Stations UL) While not strictly in
the ambit of rural electrification, a short mention of an experiment in
Lanesboro, Co. Longford, which uses heat extracted from the turf-fired
generation station is relevant. By means of a modification of the boiler
plant in the station, steam is extracted at a pressure and temperature
higher than that at which it is normally condensed. This steam then heats
water in a heat exchanger to a temperature of about 90°C similar to the
temperature of water in a normal glasshouse heating system. In Lanes-
boro, the hot water is supplied to a pair of glasshouses of one acre each,
growing commercial tomato crops. The costs (1981) of the heat supplied
came to about 60% of the cost of providing an equivalent supply of heat
from an oil fired boiler. With future increases in oil costs the percentage
saving to the growers using the extract heat is expected to increase. The
total heat supply being provided is 3.7 MW of which the loss of electricity
production is 0.53 MW. Allowing for the fuel needed to replace this
electricity at another station, the coefficient of performance (ratio of
energy supplied to energy consumed) for the scheme is approximately
2.4 to 1.
The progress of this experiment is being monitored very closely by its
sponsors which include the government, the Bank of Ireland and the
ESB, as well as by the horticultural industry as a whole. Experience will
show whether a solution to the present high energy costs in glasshouse
cropping may lie in this direction.

Potato Sprouting by Electricity 0 The practice of sprouting seed potatoes


by natural light goes back many decades if not centuries. Sprouting by
artificial illumination was first tried in Holland about 1946. It rapidly
spread to the neighbouring countries of Europe. The idea was taken up
in Britain in 1954 and in Ireland in 1961.

oi
The Quiet Revolution

Seed potatoes which are exposed to adequate light and protected from
frost during the winter develop short sturdy sprouts which ensure a fast
growth when planted resulting in heavier yields of both early and main
crop potatoes. The planting can be more flexible to take advantage of
better weather conditions and the more advanced crops suffer less from
potato blight. Properly sprouted seed can be used in mechanised planters
without risk of damage, crop growth is more even and there are few
missed spaces — as often happened when unsprouted seeds failed to
germinate.
Using fluorescent electric lamps to provide the lighting allowed the
use of existing well-insulated frost-free houses, rooms or cellars instead
of the expensive glass-walled houses required for sprouting by natural
light which normally required some heating in frosty weather. Increases
in yields of 50% to 75% were reported. While the technique is now well
established it was again co-operation between REO and ucD which enabled
it to be successfully launched. REO acquired, stocked and sold the
requisite fittings and gave instructions on site on their proper use.
This chapter had endeavoured to trace, mainly be examples, the
development of electricity as an aid to Irish agriculture and horticulture.
The lack of adequate research, experimental and advisory services and
of adequate markets was in the early years of rural electrification an
inhibiting factor. In recent years the farming picture has changed dra-
matically. The more scientific and commercial approach now existing
within the industry and the better marketing opportunities have been
reflected in the better use of electricity as an aid to agricultural produc-
tivity thus realising the hopes of the sponsors of the Shannon Scheme in
1925 and the Rural Electrification Scheme in 1946.

228
CHAPTER NINETEEN

Rural Industries

IT WAS HOPED that in addition to providing the means to a better standard


of living and increasing agricultural productivity, rural electrification
would enable industry to come to the smaller centres of population. This
applied particularly to the smaller industries and indeed to the individual
craftworker. The Report of the Commission on Emigration echoed this
hope: ‘It [rural electrification] could ease the way for the return of the
craftsman to rural areas, providing employment for him both in agricul-
ture and in cottage industries.”
Possibly the first small industry in rural Ireland to result from rural
electrification was in the village of Bansha. In March 1948, work started
on the electrification of Bansha Area, the first in County Tipperary.
Simultaneously, work also commenced in the village on the conversion
of an old mill to a jam factory. The concept was that all the fruit grown
by the farmers of the parish would have an outlet in the factory which
would thus, in addition to providing employment in the village, provide
an incentive to grow more and better fruit. Under the leadership of Fr
Hayes the local guild of Muintir na Tire formed a limited company with
most of the parishoners as shareholders. Unfortunately for a number of
reasons the undertaking did not survive. The next small industry
recorded as having started up (also in 1948) was a weaving mill in
Kilsallaghan, Co. Dublin, followed in early 1949 by a flax scutching mill
in Tydavnet, Co. Monaghan.
Slowly, as rural electrification spread through the country, small.
industries emerged. Very often these were an extension of existing
businesses. For instance the local garage or the local blacksmith installed
a welder, lathe, grinder, drills and perhaps a power hacksaw and devel-
oped as a general engineering works. The small engineering business of
Patrick Tobin, Edermine, Enniscorthy, was a case in point. When
electricity became available, Mr. Tobin installed an electric welder and
became a sizeable manufacturer of farm gates, cattle crushes and similar

e229
The Quiet Revolution

agricultural equipment. In 1954, with the help and encouragement of


REO, he developed a hammer-mill which could be sold at an attractive
price. Over the next decade the number of ‘Tobin’ mills sold to Irish
farmers reached four figures. Examples like this began to emerge on an
increasing scale as rural electrification progressed, but the big boost to
rural industries came with the availability of grants and other aid from
two state organisations — Gaeltarra Eireann and the Industrial Devel-
opment Authority.

GAELTARRA EIREANN

Gaeltarra Eireann was established in 1928 to improve the organisation,


conduct and development of the rural industries in the Gaeltacht regions,
i.e. regions where the Irish language is still spoken as the vernacular. In
this it was the successor of the Congested Districts Board established in
1891, which among other things was charged with the task of generating
economic activity and alleviating the extreme poverty then prevalent in
the west of Ireland where most Gaeltacht regions are situated.
The overall jurisdiction of the organisation covered parts of counties
Donegal, Mayo, Galway, Meath, Kerry, Cork and Waterford, with a
population (1979) of about 77,000. The Gaeltacht areas consist generally
of scattered, isolated and physically remote rural areas, with no industrial
tradition. Poor drainage exacerbates the problems of the lime-deficient,
windswept soil. No local wealth-base exists to fuel enterprise in either
industry or trade. In terms of overall natural resources these were the
poorest areas in the country traditionally exhibiting the highest unem-
ployment and emigration rates in the state.
From its foundation, Gaeltarra Eireann was empowered to establish
rural industries. The advent of rural electrification opened up new
possibilities in this regard, even in the most remote localities. Because
of their remoteness, most of the Gaeltacht areas were well down in the
queue for electricity supply. In the early fifties, however, the names of
these areas appear with increasing frequency in REO reports. In 1965 the
powers of Gaeltarra were extended to allow the formation of joint
ventures with private industry and to grant-aid new projects. In 1971 the
limits on these were broadened to allow for the exercise of full-scale
industrial development programmes. These new powers, coinciding with
the now general availability of electricity throughout the Gaeltacht areas,
heralded an upsurge in the development of industries in these regions.
One of the most impressive achievements of Gaeltarra Eireann was
the establishment of a modern industrial estate in the Gweedore district
on the west coast of Donegal. This thickly populated locality, devoid of

230
Above: Electricity provided the impetus for agriculture-based indus-
tries as well as for new manufactures. Below: Giving advice on farm
equipment from the mobile showrooms.

231
The Quiet Revolution

natural resources apart from the sea and the bogs, had a long tradition
of emigration and migration. The availability of electricity in 1954 opened
up new prospects, which were eagerly seized by Gaeltarra and by the
people. There is now a large industrial complex, which gives good
employment to about six hundred local people and which includes the
manufacture of sophisticated communications equipment as well as
radiators, packaging materials, food processing and worsted yarn. In
contrast to the small pole-mounted 5kVA single-phase rural transformers
of 1954, a large substation of 7,400kVA supplied from the 38,000 volt
grid is now required to meet the electricity demand. Gweedore is now
a thriving community with hundreds of new houses with modern amen-
ities. Many of these have been built by former emigrants returning to
work and rear their children in the place of their birth. The population,
which was declining steadily over the fifties and sixties, rose by 14%
from 1971 to 1979.
Not all the industrial achievements of Gaeltarra Eireann, and its
successor Udards na Gaeltachta, have been on such a large scale. The
size of industry initiated directly by the Udards or encouraged by the
provisions of equity or grants, varies from the enterprise employing
several hundred to that employing only one or two, perhaps a husband
and wife. The cumulative effect, however, has been striking. By 1979
the number of subsidiary, associate and larger grant-aided companies
totalled eighty-six with a total annual sales of £36 million. The total
factory area amounted to 170,000 square metres and the number of
persons employed in that year in industries sponsored by the Udaras was
4,500, a very significant contribution in the context of a total population
of 77,000.
The sophistication of some of these modern industries is impressive.
In addition to the indigenous type of industry such as fish processing,
vegetable freezing, boat building, spinning and weaving, there are com-
puter services, manufacture of optical lenses, hardened glass and glass
fibre, quartz crystal, medical products, sports garments, fitted kitchens,
air conditioners, industrial processings and engineering, photographic
and printing services.
The availability of local employment in industries such as these and
the amenities made possible by electricity have helped in a major way in
the rebirth of hitherto depressed and isolated areas. In no sphere is this
transformation more manifest than in housing. The picturesque thatched
cottages have well-nigh disappeared and have been replaced by bunga-
lows in the modern style. They are frequently criticised (and some with
justification) as being too suburban in appearance and out of sympathy
with their environment. Be that as it may, they are comfortable, easily

232
Rural Industries

run and equipped with essential amenities including bathrooms, modern


sanitation and in many cases central heating. Judicious landscaping and
planting of trees and shrubs will help blend them with their surroundings.

Industrial Development Authority 0 The small industries programme


was launched in 1967 by the Industrial Development Authority as a pilot
scheme to develop a programme of support and advisory services for
small manufacturing firms in Ireland. It was rapidly extended to cover
the whole country outside the Gaeltacht areas (already catered for by
Gaeltarra Eireann). Other 1pa programmes promoted the setting up of
medium and large industries and, while some of these were situated in
rural locations, it could not be said that many owed their existence to
rural electrification. The objective of the small industries programme,
however, was to aid existing small firms and to generate new small
industries and provide employment opportunities especially in small
towns and villages which would be unlikely to attract other types of
development. Rural electrification had opened up practically every rural
village and hamlet in the country to aid under this programme. Between
1 April 1967 and 31 December 1977 the programme provided grants to
thirteen hundred small firms, spread throughout the length and breadth
of the State amounting to £11.7 million out of a total commitment of
£23.3 million. The total employment on the latter date was 16,800. This
impressive growth was stimulated not only by the grants available but by
the readily accessible advisory service provided by the IDA and, partic-
ularly in the case of the western counties, by the initiative of the county
development officers and the county development teams.”
Again, as in the case of the Gaeltarra industries, the variety of
enterprises is prodigious, demonstrating the versatility of the rural entre-
preneurs. They cover all aspects of engineering, from precision work to
machinery repair, textiles, clothing, knitwear, boatbuilding, furniture,
plastic products, coachbuilding, craft products, electrical, agricultural
and mining equipment, fish processing, pottery, plastic products, steel
saws, moulding equipment, marine engineering, glass crystal, hardwood
veneers, liquid crystal displays and fibreglass products. A couple of
examples are given in some detail.

1. Moffett Engineering Ltd., Clontibret, Co. Monaghan U) This industry


is located in a completely rural setting about seven miles north-east of
Castleblaney and four miles from the border with Northern Ireland. Its
managing director is Carol Moffett whose grandfather in his time oper-
ated the local smithy. Her father, Cecil Moffett, who inherited the

233
The Quiet Revolution

business, recognised the opportunities in agricultural engineering and


expanded into the manufacture of ploughs, sprayers and similar farming
items. In 1954 rural electrification came to Clontibret and enabled him
to widen his range of product. In REO News October 1958 mention is
made of a 40hp grain mill in County Monaghan manufactured by Moffett
of Clontibret. He required a good steel-cutting saw but as those on the
market were very expensive he designed and manufactured his own.
This was so successful that he commenced the commercial manufacture
of steel-cutting friction saws, which were marketed under the trade name
‘Monacut’. A tractor-mounted hedgecutting saw was added to the range
of products and also marketed successfully.
In 1972 when Carol, aged 19, was a student of languages at Trinity
College, Dublin, her father died and she, being the eldest, had no option
but to return home and take over the business which employed three
people. She recognised that if it was to prosper under modern conditions,
a more sophisticated engineering input was needed. Up to this, the main
operations had been shearing, bending and welding of steel in various
forms. Now she added hydraulic and electric drives and controls and
embarked on a much wider product range. With the help of the county
development team and a grant from the IDA a new factory premises of
24,000 square feet was built at some little distance from the original
forge and workshop. Thirty-three local people are now employed in the
manufacture of a wide range of engineering equipment including the
Moffett hydraulic ‘Steelmaster’ punching and shearing machine and the
most ambitious development, the Moffett ‘Multicast’ production system,
which is a very sophisticated automatic plant selling at £70,000, for the
production of a wide range of precast concrete products such as post and
panel fencing, reinforced panels, insulation panels, roadside fencing,
security fencing, slatted flooring for cattle houses and vineyard poles.
For her markets Ms Moffett has looked to the continents of Europe,
Africa, South America and Australia, travelling widely in search of
customers rather than waiting for the customers to seek her out, and
availing to the full of the advice and assistance provided by Céras
Tractala Teo. Her pace of working normally requires a fourteen hour
day.
In 1981 she sold two large Multicast systems to a Swedish company
against strong competition. The purchaser had seen her first Irish instal-
lation in Daingean, County Offaly. Before the deal could be completed
the machine had to meet the very strict requirements of the Swedish
Bureau of Standards. Moffett Engineering Ltd. had to guarantee that
the machine, which had to be delivered (in nine large trailer loads)
erected and handed over in working order would turn out a product

234
Rural Industries

continuously every six minutes. A factor which clinched this particular


order was that the Moffett machine could produce units of up to six
metres in length, whereas the largest generally available on the market
at the time could only handle four metre units. Later came the Moffett
‘Maxigrip’, a clamp for mounting on fork-lifts and delivery trucks for the
loading and unloading of blocks, bricks, paving-slabs etc.; the ‘Multicore’
system for the production of concrete pipes and manhole rings; and the
‘Mixveyor’, a conveyor which extends the off-loading ability of a
ready-mix concrete truck.?

2. National By-Products Ltd., Castleblake, Rosgreen, Cashel UO In com-


plete contrast to the first example is an industry founded, again in a
completely rural setting, by the Ronan brothers near Rosgreen, Cashel,
a village probably better known for its proximity to the famous Vincent
O’Brien stables. In 1958, following the coming of rural electrification,
the brothers Louis and John Ronan took on the development of a
large-scale piggery. Most of the feed came from swill and offal from
neighbouring towns and meat-processing plants. As the brothers
acquired a greater knowledge of the procesing of offal a second product
was developed, meat and bone meal. This grew rapidly in importance as
did a third product, tallow. By 1961 there were thirty-five employees in
the processing plant, ten on the pig farm and an office staff in Clonmel
some eight miles away.
The growth in electricity demand reflects the growth of the enterprise
itself. In 1958 the electricity supply made available through the rural
electrification scheme was via a single-phase line erected primarily to
supply the requirements of the farmers in the vicinity. In 1960, by adding
a third wire, this was converted to three-phase to supply a SOkVA
substation at the Ronan pig farm. In 1965 a 200kVA transformer was
installed, in 1968 a 400kVA; and by 1974 a 640kVA was required. (As
a rough guide to the reader, this last would supply a maximum of about
750 horsepower.) The enterprise and its power demands continued to
grow and in 1982 a special heavy power line had to be built direct from
the EsB grid station at Cashel to meet the still growing power require-
ments, which had now approached the 1,000kVA level, and to provide
for likely new growth in the future. The plant which had started off in a
modest way in 1959 now employs about eighty people in Castleblake
alone. A skin and hide business and the accounting department in
Clonmel give further employment.
The main product from the Castleblake plant is ‘Premier’ meat and
bone meal, which has a very high protein value and is used widely as a
supplement to cereals such as barley to produce a balanced animal

233
The Quiet Revolution

ration. The company has also entered the pet-food market both directly,
by preparing and canning its own product, and indirectly, by exporting
in frozen block form lungs and liver to UK pet-food firms for further
processing and marketing under their own brand-names.
The firm is constantly improving and up-dating its production methods
to improve quality and reduce energy demands. It has, for example,
recently installed modern electro-mechanical plant for removal of water
from offal replacingto a great extent the older oil-hungry steam process.
It is currently investigating the latest developments in the processing of
certain animal constituents for human medication. The existence of the
rural electrification networks will ensure that adequate power will be
available for any proposed or envisaged expansion in the years to come.

Many more such examples could be given to illustrate the potential


commercial and manufacturing talent which existed in rural Ireland
awaiting only the advent of electric power and the stimulus of the
financial aid, advice and encouragement provided by the official devel-
opment bodies. In the development of rural industries, electrification
has contributed on two fronts — by providing the power for the indus-
tries themselves and by making possible a rural living standard which
competes successfully with that of the cities for the requisite work force.
Over the sixties and seventies a large number of workers gladly returned
with their families from the industrial cities of Britain to take up employ-
ment with new industries in their home localities — frequently at a
reduced level of pay. In 1969, Lucey and Kaldor published an analysis*
of the impact of industrialisation on two predominantly rural com-
munities in western Ireland, Tubbercurry, Co. Sligo and Scarriff,
Co. Clare. The study showed that two-thirds of the Tubbercurry employ-
ees and over half of those in Scarriff were rural residents mostly living
on farms. Most of them continued to do farm work in their off-time.
There had been an increase in population of 231 in Tubbercurry and 318
in Scarriff over what it would have otherwise have been, due to immi-
gration and non-emigration.
The impact of this dispersed industrialisation is particularly evident in
counties Galway, Mayo, Kerry and Donegal. In counties Sligo and
Clare, apart from the two small centres of Tubbercurry and Scarriff,
industry has tended to be more centralised in Sligo town and Shannon
industrial estates. Even in these cases, however, a large percentage of
the work-force lives in the surrounding rural area where the essential
amenities of living are also available with the added bonus of a country
environment.

236
CHAPTER TWENTY

Financing —
the Post-Development Phase

PLANNED POST DEVELOPMENT

AT THE END OF 1960 with over a quarter of a million consumers connected


the completion of the original development scheme was in sight and the
government requested a review of the position and suggestions as to
future developments. In March 1961 this review was sent to the
government.
The review stated that by the end of 1961/62 all areas qualifying under
the current arrangements would have been developed and would be
completed by the Board under the agreed subsidy arrangements. Supply
would then have been extended to 280,000 premises in 775 areas leaving
approximately 100,000 premises unconnected in these areas. This cor-
responded to the implementation of the scheme originally outlined in
the White Paper of 1944. The capital investment in the scheme would
then have reached £313m of which only £64m would have been obtained
by way of capital subsidy. The overall effect on the Board’s finances at
the above stage would be an annual burden of the order of £1m.
There would then remain seventeen areas containing about 6,000
premises which would not have qualified for development because of
low return. In addition to these there would be approximately 100,000
premises remaining unconnected in the developed areas. The estimated
capital cost of connecting all the above would be about £10m. From the
Board’s experience it was estimated that about 5,000 to 6,000 of these
householders would seek to be connected each year for the following
15-20 years at a cost of £500,000 to £600,000 in each year.
However, the magnitude of the current annual deficit on Rural Elec-
trification was such a serious handicap that the Board could not contem-
plate adding to it by carrying out further development without adequate
subsidy. After ‘a full and sympathetic consideration of the factors
involved’ the review concluded that this would mean the provision by

ou
The Quiet Revolution

the government of the capital free of cost (i.e. 100% subsidy). In the
absence of an adequate subsidy the Board could undertake further
development only on the basis of setting aside for rural development a
portion of any revenue surplus it might earn in a particular year.

THE INTERDEPARTMENTAL COMMITTEE

Having considered both the clamour from householders who had


‘missed the bus’ in the initial development and the financial position of
the Board, the government decided in July 1961 to set up an Interde-
partmental Committee to consider the position which would arise on the
completion early in 1962 of the current scheme of Rural Electrification
and to submit a report and recommendations.
The committee consisted of Dermot O’Riordan (Secretary of Depart-
ment) and Niall A. O’Brien, Department of Transport and Power,
Charles J. Byrne, Department of Finance, Sean MagFhloinn, Roinn na
Gaeltachta, Desmond Roche, Department of Local Government, P. J.
Brennan, Department of Agriculture, C. A. Barry, Department of
Industry and Commerce, Patrick J. Dowling and John B. O’Donoghue,
ESB. Patrick Moriarty and Michael Shiel, EsB, attended some of the
meetings and Matthew J. Brophy, Department of Transport and Power
acted as secretary. The committee submitted its report on 15 January
1962.
The report recommended that the initial development of the seventeen
outstanding areas should be completed and that an extra grant of £90,000
should be made to cover the cost of main feeder lines in these areas
bringing the total subsidy to £300,000 out of the estimated £500,000
required. This work should be done as a logical completion of the initial
development scheme. It was estimated that some 2,500 to 3,000 dwellings
out of the 6,000 in these areas could then be connected at standard rates.
It was estimated that there would be some 112,000 unconnected
premises on completion of the seventeen outstanding areas. The com-
mittee recommended that the 50% subsidy should be continued and
connection offered on the basis of a 7.5% minimum return, which was
only slightly higher than the 7.3% currently required for post develop-
ment connections.
The committee recommended the retention of special service charges,
as a safeguard against the Board’s being compelled to supply even the
most remote and uneconomic premises.
It was estimated that under the above conditions supply could be
extended to all but 20,000 or so of the remaining 112,000 premises at

238
The Post-Development Phase

reasonable charges — from basic fixed charge up to basic plus an ssc of


not more than 100% of basic (i.e. twice the basic in all).
For the remaining 20,000 premises the committee considered that
electricity supply would be too costly and recommended that a special
grant of £10 per premises be made available to these to enable bottled
gas installations to be made free of charge.
As the committee saw it, the ultimate position would be as follows:

Electricity
Number of premises con-
nected at normal fixed
charges. (This included
268,000 existing inclusive of
about 2,000 in the 17 areas yet
to be developed plus 54,000
new connections.) 322,000 (82%)
Number of premises con-
nected with special service
charges. (This included 14,000
existing plus 30,600 new
connections. ) 44,600 (11%)

Total connected to
electricity 366,600 (93%)

Bottled Gas
Total premises with subsi-
dised bottled gas _instal-
lations 20,000 (5%)

Unconnected
Total premises not con-
nected for electricity and
not qualifying for bottled
gas subsidy (i.e. not involv-
ing an ssc exceeding 100%
of Basic FC). (It was assumed
that many if not all of this cat-
egory would install bottled gas
at their own expense.) 7,400 (2%)

239
The Quiet Revolution

A minority report was submitted by the Department of Finance and


two reservations were registered by the EsB.

MINORITY REPORT

The substance of the minority report of the Department of Finance


was that originally it was never contemplated that every house in the
country would receive a supply of electricity under the scheme. The
White Paper had assumed a 69% connection. In fact by 31 March 1962,
75% of rural dwellings would have been connected. In the 775 areas
already developed some 108,000 houses remained unconnected. Of these
100,000 had refused to take supply when it was offered to them on
conditions which would not be improved on under the extension of the
scheme proposed in the Report. Of that 100,000 some 50,000 had been
quoted standard rates without special service charges. There was in fact
no pressing demand for electricity from these areas or from the seventeen
undeveloped areas.
The Department considered that the State had more than done its part
in assisting rural electrification. While the objective was the provision of
a desirable amenity and of a means of increasing agricultural production,
in fact on most farms electricity contributed more to comfort than to
output. It was essential, now that we had applied to join the Common
Market, that a growing proportion of the country’s resources should be
channelled into productive development.
Neither was the Department in favour of the bottled gas subsidy
scheme. The reservations by the EsB were:

The Ess did not consider that rural consumers would accept bottled gas
as a permanent and reasonable alternative to electricity. It also consid-
ered that, except in the most remote areas, the subsidisation of bottled
gas would have an adverse effect on the Board’s business.
Secondly, the recommendation that the subsidy should continue at 50%
would mean that the special service charges would apply to about half
(56,000) of the remaining houses. This would add seriously to the
resentment against the Board which these charges engendered. Cur-
rently, only about 14,000 consumers out of 280,000 were paying special
service charges.
On the other hand, the Board’s proposal made in March 1961 to the
Minister involving 100% subsidy would, if accepted, permit the connect-
ion of virtually all remaining houses at normal rates of charge and would,

240
A brighter outlook for schoolchildren in the Black Valley with the advent of
electricity.

241
The Quiet Revolution

in addition, allow an easement or abolition of the special service charges


for 6,000 out of the 14,000 currently affected.

SUBSIDY IS INCREASED

The EsB’s arguments obviously made some impression on the govern-


ment. The Electricity (Supply) (Amendment) Act 1962 which was sub-
sequently passed made provision for the continued subsidy of rural
electrification (from this point on mostly post-development work) at the
rate of 75% of capital cost up to £75 per house. In return, the Board
agreed to extend supply at a minimum return of 4.5% provided the
return on the basic fixed charge was not below-2.9%.
As the initial development came to its close, the transition to planned
post development (PPD) was effected smoothly. Gangs went back to the
developed areas in the order of original development where this was
possible, preceded by Area Organisers who carried out a canvass of the
unconnected consumers. Of the 10,800 new rural connections in 1963/64,
7,700 were post development consumers. By the end of 1965, these
connections were running at the rate of over 10,000 per annum.
However, 1965 brought national financial problems which extended
into 1966 and had the effect of decelerating the pace of the ppp scheme.
A large adverse balance of payments developed in the first half of 1965
and in July the government introduced a corrective package, which
included among other measures a reduction in the public capital pro-
gramme already fixed for that year. By 1966 the number of new connect-
ions had dropped to well under one half that of previous years.

PRIORITY CONSUMERS AND SYSTEM IMPROVEMENTS

At this stage it is necessary to go back some years to follow the


build-up of two other demands on the available rural electrification
capital resources, ‘priority’ consumers and ‘system improvements’ (i.e.
increasing the capacity of the networks), both of which had a retarding
effect on the progress of the planned post-development scheme.
In the early years of the Rural Electrification Scheme the economy of
the country was, if not stagnant, growing very slowly compared with that
of other western European countries. In rural areas and in the agricul-
tural sector this was particularly so. Marshall Aid was of great help in
the post-war years in launching the large public capital programme
(which included rural electrification) aimed at rebuilding and improving
the infrastructure. This was also helped by a plentiful supply of labour.

242
The Post-Development Phase

There was also, in the immediate post-war years, a large accumulation


of external reserves, which, however, were run down by the mid-fifties.
This rebuilding was not accompanied by any appreciable growth in
agricultural productivity (except perhaps in some of the richer eastern
and southern areas). Consequently the average rural dweller’s standard
of living, already low by European standards, fell even further behind.
Emigration was rife, especially in the poorer agricultural districts. There
was an overdependence on grazing and (as pointed out in the Department
of Finance minority report), little inclination or indeed incentive to use
the newly available electricity for other than amenity purposes. The
result was a slow growth in the use of electricity in the home, in its
utilisation for farming activities and in the development of rural based
industries. There was not much demand for increasing the capacity of
the networks to meet new or growing loads. The EsB rural crews could
concentrate their efforts on the planned aspects of the scheme.

EXPANSION IN ECONOMY BRINGS PROBLEMS

Following on the Whitaker report in 1958 and the first programme for
economic expansion, a change became evident in rural Ireland. Slowly
at first and then more rapidly, the economic pace quickened. More and
more use was made of electricity in the home and on the farm; small
industries were set up in rural areas; hotels and guest houses opened
their doors to the revived tourist industry; new houses were built. All
these made new demands on the rural electrification resources. By the
mid-sixties a very large proportion of ‘rural’ capital was being allocated
to the reinforcement of networks to cater for increased electricity
demand and extending new supply to these consumers, who were con-
sidered ‘priority’ consumers as they formed an important part of the
national economy, providing employment and earning valuable foreign
currency.
When capital had been more freely available the Ess had been able to
cope with the demand from these applicants and also to keep to its
programme of providing supply under the ppp scheme. There were tens
of thousands of householders who for various reasons had not accepted
supply under the initial development but who now were exerting strong
political pressure for connection. As available capital became scarce the
reinforcement of the networks to keep pace with growth in demand and
the provision of new supply to the priority consumers left very little
residue of capital for other connections. Again and again throughout the
country PPD programmes were postponed. Promises of supply given in

243
The Quiet Revolution

good faith were constantly being broken as the pressure for priority work
pushed back the connection of these premises.
At the beginning of 1965/66 the EsB’s proposed rural capital budget
was £2.3m made up as follows: system improvements £340,000, new
‘priority’ business £500,000, planned post development £1,460,000. As
a part of the reduction by the government of the public capital pro-
gramme, the capital available for rural electrification for the year was
reduced to £1.05m. Since the demands for network reinforcement and
priority new connections continued perforce at the same rate or indeed
slightly higher than in previous years, the full impact of the reduction
fell on the post development programme. At the time the axe fell, there
were about 3,000 prospective consumers who had their houses wired up,
to whom a definite commitment had been made and who now were being
pushed back in the programme for possibly another year. Again the
government and the EsB were subjected to strong and widespread
criticism.
A meeting was held in the Department of Transport and Power
attended by the Chairman of the ess and the Engineer-in-Charge of
rural electrification. A further £300,000 in capital expenditure was
authorised, enabling the most pressing commitments to be met. Agree-
ment was also reached on the ‘priority list’ which now included new
tenancies by existing consumers, workshops, garages, industries,
churches, schools, hotels and guest-houses.

SPECIAL SERVICE CHARGES AGAIN AN ISSUE

The question of high special service charges continued to preoccupy


the government. Obviously the political pressures to eliminate these
were intense and the government was most anxious to relieve the
situation. It was estimated that the authorised capital expenditure of
£42m would be reached in August 1968 and a new Bill increasing the
legislative limit of rural electrification expenditure was due to be intro-
duced in the Dail in the new year. This gave an opportunity to review
the situation, as any alternative subsidy arrangements which might
relieve the position could be incorporated in this Bill and, if agreed, be
given immediate implementation.

SPECIAL SERVICE CHARGES REDUCED

The EsB carried out a survey of all unsupplied rural premises and of
the degree to which special service charges would apply under the current
subsidy arrangements. It found that for 12,000 consumers the level of
special charges would not be unreasonable (up to 50% of basic charge),

244
The Post-Development Phase

while for 23,500 others they would range from ‘high’ to ‘extremely high’.
It developed a package proposal for government approval: if the govern-
ment, while maintaining the existing 75% capital subsidy, were to extend
the cash limits per house from £75 to £150 the EsB could reduce the
minimum return required on the capital investment to 4.5% for all
premises. This would reduce very considerably the level of special service
charges, particularly for very remote consumers.
The government refused to consider extending the subsidy limit.
However, to the Board’s consternation, it was instructed to implement
the other half of its proposed package — the reduction of the minimum
required return to 4.5% in all cases, even to existing consumers. An
accompanying undertaking from the government ‘to review the situation
at the end of two years’ was accepted by the Board with a pinch of salt.
The good news was quickly conveyed to the rural community in a
government press release on 20 May 1968. To the EsB, however, the
refusal to extent the subsidy limits to balance the concessions now being
given was yet another nail in the coffin of the original subsidy concept.
The loss on the rural revenue account for the previous year had been
£1.7 million and the accumulated loss had now reached over £11 million.
The latest directive would greatly increase the cross-subsidy required
from the urban electricity consumers. The Board had little option,
however, but to accept, however reluctantly, the directive given.
The easing of conditions was like the bursting of a dam. There was a
flood of applications for connection. Householders who up to now had
considered the terms too stringent, clamoured for supply. As well as
those who had held back during the initial development, a new category
of premises started to grow in importance, reflecting the improved
economic status of the rural-dweller, the newly built house. A new house
might be built by a young couple setting up home or it could be a
replacement for an existing house. Where the householders had been
electricity consumers in their previous residence, or where one of the
couple getting married was the son or daughter of an existing consumer,
they were regarded as having prior claim to those who had failed to avail
of previous opportunities to obtain supply.
Thus, another grouping was added to the already formidable list of
‘priority’ cases. From the mid-sixties on, these made more and more
inroads into the available capital. It was soon realised that unless the
capital budget was increased, connections under the ppp scheme,
demands for which were also increasing, would have to be drastically
reduced. However, the government’s capital budgeting procedure,
operating as it did on an annual basis did not easily allow budget increases
once the estimates for the year had been agreed. On the other hand, the

245
The Quiet Revolution

ESB was expected to respond to demands at short notice, particularly in


the case of priority consumers. Furthermore, capital costs of connection
had commenced to increase rapidly. This increase in costs coupled with
a fixed annual budget meant fewer connections in the non-priority
categories. Loud protests came from those householders whose connec-
tioin had to be postponed. The Ess found itself, not for the first time,
caught in a pincer squeeze. It was coming under strong political pressure
to speed up its connections and was meeting equally strong resistance
from the government to spending more than allowed for in the annual
budget.
By 1969 the rate of new house building in rural areas was heading for
4,500 per year — all in the ‘priority’ category. Supply to these and the
improvement of supply to existing consumers threatened to absorb
almost all the capital allocation for rural electrification. Because of the
financial situation, this had been held down to £2 million per annum
instead of the £3 million required. The cut of £1 million meant in effect
that only £0.2 million would be available for ppp work with a drastically
reduced level of connections (only about five hundred instead of three
thousand per annum). To make matters worse, this reduction was in the
context of a greatly increased demand following the introduction of the
more attractive conditions for connection.
In trying to complete the ppp scheme the EsB was aiming at a receding
target. The resurgence in the economy in the sixties had extended to the
rural areas. A faster growth of real income per capita, reduced emigra-
tion, earlier marriages and an expansion in agricultural output, in house
building and in tourism all contributed to the demand for more electricity
and for connections to new houses, rural industries, commercial under-
takings, hotels and guest-houses. It was obvious that unless a completely
new approach was developed the ppp scheme would perish from capital
starvation.

THE 1971 PACKAGE

In June 1970 James J. Kelly was appointed Ess chief executive. He


was a man with little time for pussyfooting or for clawing his way through
a web of intangibles, variables and uncertainties. He liked to face
problems squarely and seek positive solutions. In one of his first major
policy submissions to the Board, he outlined the problem.
Under the initial Rural Electrification Scheme, completed in 1963/64,
792 areas had been supplied with electricity and in these 273,600 premises
had been connected at a cost of £32.2 million of which £6.9 million had
been provided as subsidy by the government. Up to 31 March 1970, the

246
The Post-Development Phase

planned post development scheme initiated in 1961 had been extended


to 505 areas. The extra premises connected under this Scheme and under
the ‘priority’ umbrella amounted to about 78,000. The cost of this work
and of necessary system improvement work was £14.25 million of which
£7.1 million had been provided as subsidy. Thus at 31 March 1970
351,600 rural consumers had been connected at a total cost of £463
million of which £14 million (or 30%) had been provided as subsidy.
The subsidy, of course, was entirely insufficient to avoid an operating
loss on the scheme. This was expected to be £2.2 million per annum for
the financial year 1969/70. If work continued to proceed at the current
rate and on the current basis, it was estimated that by the end of 1974/75
the loss would reach £3.36 million per annum.
Looking to the future, it had been indicated that the maximum subsidy
available over the five years 31 March 1970 to 31 March 1975 would be
at the rate of £1.8 million per annum or £9 million over the five-year
period. The government had also indicated that 31 March 1975 was the
date set by it for the termination of rural subsidy. From that time on,
new connections would have to be made on the EsB’s normal commercial
basis, which in the case of rural dwellings could involve substantial
capital contributions from the householder concerned.
A change in the whole approach to rural electrification was required
if the scheme was ever to reach completion. The constraints resulting
from the linking of annual capital expenditure to the annual amount of
subsidy agreed by the government were having serious effects.

Completion date was a receding target. The capital available for planned
post development was reducing while the number of potential connect-
ions was increasing.
It was no longer possible to plan a rural programme.
The national objective of rural supply was not being met but was being
pushed aside under current arrangements.
The Board’s reputation for service was under severe and justifiable
criticism. This criticism was mounting.
There was an inherent injustice to many potential consumers under the
current arrangements.

The situation could not be allowed to continue and it was proposed


that the Board itself should now take the initiative by setting a firm
programme which did not depend on the current subsidy allocation.
Having set the programme, the Board should then seek discussions with

247
The Quiet Revolution

the government to achieve the best subsidy arrangements which the


financial situation would allow. The proposal was approved by the Board
and welcomed by the government which was of course under severe
political pressure to achieve the completion of the scheme as quickly as
possible.
From the Board’s point of view, it marked the abandonment of the
social service approach to the problem with progress dependent on the
arbitrary annual allocation of capital. Instead it would now treat it as an
electricity supply problem, to be assisted by a contribution from the
government up to 1975, but leaving the Ess free to plan it as a single
operational exercise. The new approach would accelerate the growing
deficit on rural activities, but, as pointed out by the Board’s Chairman
in a letter to the Minister for Transport and Power (July 1970), the
Board found itself in a position where it had little option but to decide
on a scheme which would dispose of the problem finally. The proposals
did not mean that the EsB believed that a reduced state contribution over
the five-year period represented a fair distribution of the burden of cost,
but rather that the Board was so anxious to finish the job that it was
prepared to shoulder a greater share of the cost.
The Chairman also pointed out that the original philosophy of the
Rural Electrification Scheme was that it should involve no loss to the
ESB. The concept had long since been forgotten. It was abandoned in the
first instance when the government in 1955 withdrew the subsidy, and
the loss which had been steadily growing had increased again in recent
years by the reduction of the special service charges. Therefore if the
Board was to tackle the problem on the lines indicated there were a
number of corollaries which should be accepted by the government.
These included the Esp’s right to retain special service charges, to incur
the extra losses involved and recoup them by tariff adjustment and
cross-subsidisation. It was also necessary that there would be no hold-up
on the necessary tariff adjustments and that there would be an end to
the stop-go policies on the capital involved which had so bedevilled the
programme in recent years. The total estimated cost of the five-year
scheme, allowing for inflation, was £21 million of which system improve-
ments at £12.5 million was the major item. It was estimated that 17,000
PPD consumers and 19,000 ‘priority’ consumers would be connected at
final costs of £3.5 million and £5 million respectively.
The government agreed to contribute £10 million as a final subsidy
contribution after which it no longer would be responsible for subsidising
rural electrification. Due to financial stringency this subsidy would not
be paid on the basis of work done as heretofore, but in instalments up
to 1977/78.

248
The Post-Development Phase

The main parameters of the new scheme were as follows:

The unconnected houses in those areas in which ppp had been completed
would get one further opportunity of availing of supply under the
scheme. This would be publicised by advertisements in local papers.
In ppp areas in which work was in progress or in prospect all unconnected
householders would be visited by EsB canvassers.
Each area would be deemed ‘closed’ as work was completed except for
certain categories which would continue to be connected on a subsidised
basis up to March 1975.' Any other applicant for supply in ‘closed’ areas
would be quoted terms based on principles similar to those quoted to
non-rural consumers. In these cases, if the quotations were accepted,
supply would be extended with the minimum possible delay as was the
practice in urban locations.

In this way, the residents of every unconnected private or farming


residence in every rural area in the country would have one final
opportunity to avail of supply on subsidised terms after which this would
no longer be available.
In 1971, the separate rural revenue account was discontinued by
agreement with the government in anticipation of the Rural Electrifi-
cation Scheme ending in 1975. The loss shown in the rural revenue
account for that year was £3,254,800 and the accumulated loss to date
£19 million.
As so often in the past, events overcame the predictions on which the
new programme was based. The prospect of early membership of the
EEC brought with it a further quickening of the economic pace in rural
Ireland. In 1970 farmers achieved record expansion in output. There was
a great swing from milk to beef. 1971 was a good year for cattle but was
eclipsed by 1972. In the midsummer of that year, six months before
Ireland’s official entry into the EEC, there was a huge rise in cattle prices
from the low base to which they had hitherto been tied by British cheap
food policy. By this time also milk had come back onto the scene as the
number of milch cows was expanded. By 1973 agriculture was booming
as Ireland anticipated the benefits of EEC membership, wider markets
and higher prices.
This boom was directly reflected in the rush of householders to get
electricity connection. In contrast to the position at the commencement
of the fifties their economic state and social outlook was now such that
electricity was regarded as an essential amenity and aid to production.
There was also the knowledge that if they missed this last chance of

249
The Quiet Revolution

connection under the subsidised scheme, they would have to pay much
more for connection later. By the summer of 1974 an updating of the
1971-75 scheme in the light of actual experience indicated that the
original estimate of 36,00 consumers to be connected was far short of
the mark. It was now estimated that 57,200 would require supply and
that the programme would now overspill into 1975/76. This high connect-
ion figure would be achieved at the expense of the planned system
improvements programme which had fallen very far behind schedule as
resources had to be diverted to the now more pressing task of new
connections. (The system improvements backlog was to be taken up in
the succeeding years and financed completely from the Board’s own
resources.) Not all new connections were of domestic premises. In
1973/74 for example, 180 applications for supply to new industries in
rural locations were processed costing £214,000. None of these industries
had been envisaged a couple of years before.
By 1976 it could be said that at last the Rural Electrification Scheme
had been completed. The original scheme had been designed to connect
280,000 rural premises at a cost based on pre-war prices of £14 million.
In 1976 the Minister for Transport and Power was able to report in the
Dail that 420,000 houses had been connected to the rural electrification
networks at a total cost of about £80 million of which £28 million
represented State subsidy.

POSTSCRIPT

There was, however, a postscript to the Scheme. In the original White


Paper it was accepted that 14% of all rural dwellings, or about 56,000,
were so remote as to be outside the scope of any practical electrification
scheme. It was also thought at the time that the pressure from these
remote dwellings for inclusion would not be very great. Over the years,
however, as their isolation lessened through better communications and
as their perceived entitlement to a better standard of living grew, the
pressures by these householders for inclusion in the scheme became
irresistible. Year by year the number of houses excluded was whittled
down and by 1976 when the subsidised scheme was scheduled to end, it
was a bare 1-2%, the 4,000 to 8,000 (an exact count was difficult) most
remote premises in the country.
In 1976 the Minister for Transport and Power introduced the Electric-
ity (Supply) (Amendment) Bill 1976. In his introduction he pointed out
that in the previous thirty years some 420,000 houses had been connected
to the rural electricity networks at a total cost of about £80 million, of
which some £28 million represented state subsidy. This represented some

250
The Post-Development Phase

98% to 99% of all rural houses. There were still anomalies. The Minister
instanced the plight of the Black Valley (Co. Kerry) and Ballycroy
(Co. Mayo) as being probably the worst. These were isolated com-
munities which, while offered supply under the 1971 to 1975 phase, were
not in a position to accept owing to the very high costs involved and very
high capital contributions required.
The Bill proposed that householders in these circumstances, i.e. who
had refused supply under the final phase of the scheme because of the
requirement to pay capital contributions, would be given a further and
final opportunity to obtain supply on subsidised terms but, this time,
without having to pay a capital contribution. The government in effect
would pay the capital contribution. The householder would be liable for
the annual special service charge. In taking steps to eliminate this
anomaly, the government felt it was going as far as it could reasonably
go.
The above circumstances were not confined to the householders in the
Black Valley and the householders in Ballycroy. The exact number was
not known at the time but a quick EsB estimate had given a figure of 800
to 900 householders. This was not definitive as there were also individual
houses in this category here and there throughout the country and the
final total might well be more than the Ess estimate.
Following the passage of the Act a detailed survey was carried out and
a total of 1,580 premises established as qualifying for subsidy, almost all
in the poorer western counties. The total capital cost was estimated as
a result of the survey at £3.5 million of which, under the terms of the
Act, the EsB would provide £1.6 million and the State £1.9 million.
Even this was not the end of the story. A more prosperous and more
progressive Ireland was reflected in the steadily growing consumption of
electricity over the 1970s. In 1950 the average annual electricity con-
sumption per rural domestic consumer was 570 units; in 1960 it was 910
units and in 1970 average annual consumption had reached 1,830 units.
Up to the 1970’s the rural electricity networks as originally constructed
had mostly been adequate to cater for the demand. In the cases where
reinforcement was necessary due to rapidly growing local demand, it was
generally possible to fit this into the construction programme without
much difficulty. The general upsurge in consumption over the 1970s as
the connection figure approached 100%, coupled with a greater demand
per consumer, threw a great strain on much of the network and neces-
sitated the introduction of a programme of large-scale reinforcement.
However, the clamour for new connections was so loud and so pressing
that the EsB concentrated on the connection of new consumers at the
expense of the reinforcement programme. With the achievement of 98-

Zl
The Quiet Revolution

99% connection by 1976 it was now possible to allocate more attention


to the reinforcement (or ‘system improvements’) programme.
The building of new houses in rural areas went on unabated over the
late seventies rising to more than 10,000 per year. Even though the
subsidised electrification schemes were now at an end (with the exception
of the limited 1976 scheme), the owners of these newly-built houses
expected electricity to be made available as soon as the house was
finished albeit on non-subsidised terms of supply usually involving a
substantial capital contribution by the householder. (An instalment
scheme for the payment of such contributions was introduced by the EsB
in 1976.) Thus, although the Rural Electrification Scheme as such had
come to an end, the period 1976 to 1980 was one of great activity in
three different areas of rural electricity supply: system improvements,
extension of supply to the houses qualifying under the 1976 legislation
and extension of supply on a non-subsidised basis to the 10,000 houses
being built each year in rural locations.
The first consumer under the Rural Electrification Scheme had been
switched on in the village of Oldtown, Co. Dublin, on 15 January 1947.
The objective of the scheme at that time was to connect 280,000 con-
sumers or 69% of the then estimated 402,000 rural premises. By 31
March 1980 the number of electricity consumers in rural locations had
reached the figure of 468,000 and, owing to the building of new houses,
was growing at a rate something in excess of 10,000 per annum. Because
of inevitable overlapping, it is not possible to give exact figures for the
different phases of rural electrification, but in round figures, progress
was as follows:

New
Consumers
Connected
1st phase 1946-1963 280,000
1st phase post-development 1963-1971 60,000
2nd_ phase _ post-development 1971-1976
(including 1,600 consumers under 1976 Act) 62,000
Unsubsidised rural connections to 31 March 66,000
1980

Total at 31 March 1980 468 ,000

The total capital expenditure in connecting this figure, including network


reinforcement as demand grew, came to £109,355,000, of which
£27,900,000 had been provided by government subsidy.

252
CHAPTER TWENTY-ONE

A Tale of Two Parishes

EVERY ONE Of the eight hundred or so rural parishes has its own story of
progress and of the contribution made by rural electrification to its
development. To document these stories fully would require many
books. I have therefore taken the story of just one small area as a
microcosm of the larger canvas — two parishes in west Donegal, an area
endowed with striking natural beauty but with few natural resources, to
which rural electrification was extended in 1952.
As one travels westward from the town of Donegal, past the busy
fishing port of Killybegs, to the two westernmost parishes of Kilcar and
Glencolmcille, the land gets poorer and more mountainous, and the
landscape progressively more beautiful. The heavily indented and rocky
coastline includes some of the highest sea cliffs in Europe, while inland
are spectacular vistas of mountain, moorland and glen. The glens
between the barren mountains and parts of the sea coast where the land
is somewhat less hostile, have been settled for thousands of years: the
court cairns in the shadow of Slieve League and in the glen itself testify
to the presence of human beings some five thousand years ago.' In more
recent centuries the indigenous inhabitants of the glens and sea coasts
were supplemented by the dispossessed tenants of more fertile lands to
the east resulting in a very large population in a very restricted area with
few natural resources except turf for the fire, rough grazing for sheep on
the rainswept mountain pastures and the plentiful fish in the surrounding
waters. It was on these resources that the large population eked out
subsistence, spinning, dyeing, weaving and knitting the wool, and fishing
from small boats close to the shore. Families were large and, while the
harsh environment bred a hardy and resourceful people, it was not
possible to feed many mouths from the tiny holdings of poor land or
from the proceeds of the cottage industries.
Emigration was inevitable — often by the most active and adventur-
ous young people who frequently rose to positions of great eminence in

253
The Quiet Revolution

their newly adopted lands. If the emigrants prospered, a very significant


addition was made to the resources of the family remaining home by
way of what were officially known as ‘emigrants remittances’, the peri-
odic cheque from the son or daughter abroad. This was indeed the only
sizeable source of cash income for many of the poorest families in the
region.
The high rate of emigration in the first half of the present century gave
great cause for concern. In the 1950s a young curate in the parish of
Glencolmcille, Fr James McDyer, became so distressed at the draining
of the lifeblood from the parish that he initiated strenuous efforts to
develop local projects to provide employment and so help to keep the
young people at home. His efforts and achievements in the parish itself
have passed into history. He awakened in the nation as a whole a
consciousness of the danger of losing a valuable part of its cultural
heritage if parishes such as Glencolmcille were allowed to die.

Fr McDyer saw in rural electrification a means of halting and even


reversing the outward flow of people. Electricity could help in the
development of local industries, including the tourist industry; it could
brighten up the homes of the people and dispel the mental and physical
gloom; it could ensure an attractive social life and through television
open up new ideas and concepts.
When in 1950 the rural power lines started to snake westwards from
Donegal Town a new hope was thereby awakened for the future of these
west Donegal parishes. Through rock, bog and swamp to Mountcharles
and Dunkineely, Killybegs and into the parish of Kilcar the lines were
driven, extending the new power to hundreds of rural and village
dwellers. By the summer of 1952 the parish of Kilcar had electricity and
the Glen River boundary of the parish of Glencolmcille was reached.
Across the river to the village of Carrick and three miles south to the
tiny harbour of Teelin the lines stretched so that by Christmas the
inhabitants on the east side of Glencolmcille parish were enjoying the
benefits of electric power. Now, however, there was a full stop in the
westward thrust. Beyond Carrick was a barren five-mile tract of moor-
land and beyond this the glen itself, the onetime abode of the saint from
whom the parish drew its name. Even on the basis of the liberal terms
made possible by the government subsidy, the number of householders
in the glen itself and in the neighbouring glens who were prepared at this
stage to avail of supply was too small. The construction crews were
withdrawn to bring supply to other areas in the county and it appeared
that it would be many years before they would return.

254
ren, SAVIN

Fr James McDyer with colleague at Glencolmcille.

255
The Quiet Revolution

Fr McDyer recalled:
we ran into fairly determined opposition by the few. These were mainly
ageing householders who had no families . . . However it would be difficult
to blame them. Their attitude sprang from their traditionally disenfran-
chised existence and their isolation from involvement in progress. After
all, were they not to fear another alien force from outside. They had good
reason to suspect such forces, and progress. . . . However I was deter-
mined on this occasion that I would not be thwarted because there could
be no worthwhile progress without electricity.
I asked the Electricity Supply Board to send down one of their best men
to address a general meeting. They did better. They sent down Colm
Browne who was a native of the district. The meeting in the community
hall was packed, which was a good omen. Colm really excelled himself
and I was sure I could feel the vibrations of assent from the audience.”
Fr McDyer then set his parish council to- recovering the situation.
Evening after evening they called into the small houses explaining,
persuading and cajoling the householders into signing the required
application forms so as to secure the numbers and revenue required to
ensure electrification. The benefits to the individual and to the locality
as a whole were teased out over and over again. So were the objections
that electricity would be too dear and too dangerous in a thatched
cottage. The economic state of the majority of householders was indeed
such as to give them pause before committing themselves to paying a
fixed charge of from £6 to £8 a year. The cash income of the majority
was small, for some perhaps the sale of a few sheep, cattle or wool; for
others payment for piecework knitting or weaving or the occasional
money order from the son or daughter abroad. While the people were
self-sufficient to a great extent, certain commodities had to be purchased
and there was no money to spare for unnecessary luxuries. What if they
found it difficult or impossible to meet these new bills? Many preferred
to forego the promised benefits rather than risk the humiliation of
disconnection at some future date.
However, the dedication and persuasive powers of Fr McDyer and his
small band achieved success. Sufficient signatures were obtained and in
December 1954 an historic and significant little procession paraded in
period costume down the village street of Glencolmcille. In the front
was a child carrying a rush light, followed in turn by other children
carrying other means of illumination — a tallow candle, a wax candle, a
single wick oil lamp, a double wick lamp, an incandescent lamp and
finally an electric bulb. At the local church prayers were said; a blessing
was given, a switch was thrown and electrification of the parish of
Glencolmcille was at last complete.

256
Two Parishes

It had not been easy. The engineer’s report on the area reflects some
of the construction difficulties, especially through the many bogs trav-
ersed by the lines.

Of 900 poles erected, over 700 had to be fitted with stays to ensure stability.
One pole, 42 ft in height . . . kept sinking for a fortnight and eventually
decided to say put at a depth of 103 ft. . . Servicing also was very difficult,
the houses being long, low and thatched, with very poor walls and no
chimneys. . . of the total number which had originally signed application
forms, nineteen dropped out. Of these five houses had become vacant,
eight householders were now building new homes and in two cases family
bereavement had enforced a change of view. However, sixty-eight new
consumers were gained and the annual fixed charge revenue rose from
£1,072 to £1,370. Capital costs rose more steeply however due to construc-
tion difficulties.

Finally the report speaks very highly of Rev Fr McDyer, CC, and his
parish council for the considerable help they gave in keeping down
backsliding.
One of the problems recalled by Fr McDyer was the difficulty in
recruiting local labour for the scheme because of the fear of the small
farmers of losing the dole.

I wrote a trenchant letter to the late Bill Norton who was Minister for
Social Welfare in the government of the time... . On the following
Monday upwards of thirty extra workers were directed by the Labour
Exchange to report for work. It was well for me that my action was never
known because my popularity would have plummeted.’

In 1983, some twenty-nine years later, a visitor to the parishes of Kilcar


and Glencolmcille who had not seen the district since those early fifties
would have noted tremendous changes.
Local industries in the two parishes and the development of the fishing
port of Killybegs ensured that employment was available to any young
person who wished to remain and settle down. No longer was the
emigrant ship the inevitable fate of many young people. Many thatched
cabins had been replaced by modern bungalows with electricity, running
water and modern sanitation. New houses were constantly being built as
young people, now assured of a living locally, got married and set up
their homes. People from outside the parish, from outside the county
and indeed from outside the country itself moved in, attracted by the
beauty of the locality now supplemented by the availability of modern
living amenities.

254,
The Quiet Revolution

The growth of local industries has been promising. Tourism in this


remote and beautiful region has been given a boost by the hotel, guest-
house and ‘bed and breakfast’ accommodation made possible by the
ready availability of electricity and water. Television, which is installed
in almost every house, has banished any sense of remoteness as has the
wide ownership of motor cars made possible by the wages now coming
into most houses.

THE TWEED INDUSTRY

Donegal tweed from the two parishes has long been famous for its
quality, texture and colour. The industry had been under the auspices of
the Congested Districts Board as far back as the 1890s. From 1928 it was
organised and encouraged by Gaeltarra Eireann, and up to the late 1960s
a successful business based on the handwoven cloth was operated. Skilled
handweavers were employed at the factory at Kilcar, at marts in various
locations and in their own homes. The coming of rural electrification in
the early 1950s improved conditions immensely, particularly by providing
good lighting for the work during the long winter evenings. However,
with the advent of free trade in the early 1970s, it was obvious that new
and higher levels of productivity, quality control and design would have
to be achieved to place Donegal tweed in the high fashion category on
the export market at keenly competitive prices. It was necessary to
modernise and mechanise and introduce sophisticated design and mar-
keting techniques. The challenge was taken up with vigour and the result
in a modern industrial complex in Kilcar under the aegis of Udards na
Gaeltachta (the successor to Gaeltarra Eireann) employing over seventy
people in the tweed factory and another 130 in the ancillary dyeing and
spinning plants with an estimated sales figure of £13m for the year 1980.
While most of the workers are drawn from the two parishes, it has been
necessary to recruit from as far away as Ardara to fill some vacancies.

FISH PROCESSING

Fishing has long been traditional in the locality, as has fish processing.
Filleting and smoking of fish were encouraged by the Congested Districts
Board at Teelin and Killybegs from the end of the last century and by Sir
Burton Conygham at Rutland, Burtonport, as early as 1780. Killybegs
has long been a major fishing port and a large number of ancillary
industries has grown up since the coming of the rural electrification
networks ensured power for expansion. The growth of fishing in Killy-
begs has provided many jobs for the people of the neighbouring parishes

258
Two Parishes

of Kilcar and Glencolmcille. More recently, however, smaller but no less


successful fish-processing industries have been developed in the parishes
themselves.
At the tiny port of Teelin, from which small fishing boats operate, a
local man, Mr Jack Gallagher, operates a small but very modern enter-
prise which purchases, processes and markets the catch from the small
local fishing boats as well as a proportion from the larger boats operating
from Killybegs. Starting with a very small single-phase electricity demand
from the local rural electrification network the requirements of the
enterprise grew rapidly. By the late 1970s, electric power requirements
had risen to over SOhp involving numerous three-phase motors. This
demand was met by installing a converter which delivered three-phase
supply from the single-phase branch. By 1981, however, demand had
risen to about 150hp with blast freezers, cold rooms, smoking plant and
electronically controlled vacuum-packing machinery. The original
single-phase branch line erected in 1952 to supply the modest domestic
needs of the village had finally to be converted to three-phase to meet
the demands.
The plant, which between processing staff and fishermen ensured
employment — part-time and full-time — for about fifty local people, is
equipped to process and produce fish, mostly herring, mackerel and cod,
in any form the market requires — fresh, frozen, wet salted, dry-salted,
smoked, brined (roll mops) — and ship them in barrels, boxes or indiv-
idual vacuum packs to markets in Britain, the Continent of Europe and
the West Indies. Mr Gallagher told the author of a visit from a buyer for
the famous Harrods of London Food Halls who, having tasted a smoked
herring in a Paris restaurant, tracked the fish to its source — the tiny
processing plant in Teelin — with a view to securing supplies. Unfortu-
nately for Harrods all available supplies had already been contracted
for, mostly to French buyers!
Six miles up the Glen River in the townland of Meenaneary on the
boundary between the two parishes is the fish processing plant of Earagail
Eisc Teo., which formerly was the vegetable packing factory of the
Errigal Co-operative Society, one of Fr McDyer’s early experiments. In
mid-1981 the plant, under the direction of Mr Jarlath Morris and still in
the course of expansion, already employed thirty-five local people and
has targeted for a total of eighty permanent jobs plus a further thirty
seasonal (August to November) jobs.
As in the case of the Teelin plant, the Meenaneary enterprise covers
a wide variety of fish processing. Mackerel, which has become the biggest
volume landing, is the most promising product. It is generally marketed
in frozen form — ‘round frozen’, frozen fillets and ‘headed and gutted’ —

209
The Quiet Revolution

for markets in France, Germany and Africa. Herring, on the other hand,
is generally salt-cured or smoked and marketed as roll mops and salad
fillets to Germany and Holland. Crab, which in early days was frequently
thrown overboard when caught, is now a big money earner in the form
of whole crab, crab meat, fresh pasteurised vacuum packed bodies and
crab claws. Markets are in Sweden (two-thirds of output) and France
(one-third). Cod, which generally has a short season (February and
March), is usually processed by salting. Good markets exist in
Mediterranean and West Indian countries for this salted cold-water fish
to replenish salt lost by perspiration.
Unlike most fish-processing plants, that at Meenaneary is not sited at
a fishing harbour. It is fifteen miles from Killybegs and six from Teelin.
However, Jarlath Morris maintained that the fact that the staff, almost
all local including many small landowners, had such a proprietorial
interest in the business and such local pride that the productivity of the
plant and the quality of the product more than compensated for the extra
cost of haulage.
A feature of both the fish-processing plants is the large cold stores
where the product can be held for long periods to take advantage of
market trends. The growth of the Meenaneary plant can be measured by
the growth of the freezing and cold-store requirements
1971 — 45hp blast freezer plus 10hp cold store installed
1975 — 80hp blast freezer plus 20hp cold store added
1981 — Three 100hp blast freezers plus 40/50hp cold store (with capacity of 800 tons)
added.
The total requirements of over 500hp are supplied from a rural three-
phase line originally erected in 1952 as a single-phase branch to service
the then modest electricity demands of the domestic consumers in the
locality.
The picture thus emerging from a survey of progress in the two most
westerly parishes in County Donegal is that the extension of rural
electrification in the 1950s heralded the transformation of a whole rural
society of about 2,500 souls. From being a backwater the area has
developed into a well-balanced forward-looking community with excel-
lent housing and all the necessary services for comfortable living in a
pleasant rural environment. Local manufacturing and processing indus-
tries supplement the traditional sheep rearing and fishing to ensure a
high level of employment within the parishes themselves. No industry is
yet so big as to isolate the workers from their rural background. They
continue to live where their roots are instead of disappearing into the
anonymity of a big city, thereby enriching not only their own lives, but
the life of the community as a whole.

260
Epilogue

OUR CHRONICLE OF THE PROGRESS Of the Rural Electrification Scheme


in the preceding chapters has led us from the late forties to the early
eighties. This period of little over a generation has.seen an immense
change in Irish rural society. It would be appropriate in the closing pages
of our story to review this change and the contribution made by rural
electrification. Better housing, water on tap and modern aids made
possible by electricity have greatly enhanced living standards in rural
homes, which are no longer so isolated. A more informed and progressive
society now obtains as a result of widespread secondary education, the
social and cultural activities of the voluntary organisations and the better
communications available through telephone, television and radio. (One
Area Organiser estimated that in many of the rural areas he canvassed
in the late forties only about one family in six had a radio, battery
operated and consequently used very sparingly.) The rural community,
which now more than ever includes many non-farming families, has
developed into a strong and articulate element playing a full part in the
life of the nation.
Out on the farm, especially where a new and more enlightened
generation has succeeded in wresting control, there is a new and more
business-like approach. Electricity, now installed on practically every
holding, is playing its part in the application of modern technology. On
many farms it is helping to improve quality and output and hold down
production costs in the context of a steadily declining work-force.
Improved marketing techniques are being employed to develop existing
markets and open up new outlets for increased production. Much,
however, remains to be done. Irish agriculture is still the most under-
developed and undercapitalised in the EEC. Much potentially productive
land still stagnates under the control of elderly tradition-bound owners.
One authority’ has estimated that, with proper development, agriculture
could provide an extra 75,000 jobs over a fifteen-year period by reducing
the drift from the land, increasing input and increasing the volume of
goods for processing.

261
The Quiet Revolution

Rural industries are helping to compensate for the declining employ-


ment levels in agriculture. Even in the case of large industries centred in
the towns the contribution of rural electrification has been considerable.
Much of the labour force for these is drawn from the surrounding rural
hinterland where the amenities of living now available, combined with
a rural environment permit an attractive life style which in turn discour-
ages permanent migration to the towns. It could be claimed that many
foreign industrialists have been attracted to set up in Ireland because
their executives and work forces could live in pleasant rural surroundings
while still enjoying most of the amenities of modern living.
After a long period of decline the rural population is now rising
steadily. This reversal is particularly welcome in the west, where the
population had continued to drop even through the sixties when the rest
of the country was experiencing a revival in growth. Between 1971 and
1979 the population of Connacht increased by 7.1% while that of the
three Ulster counties in the Republic rose by 9.1%. In County Donegal
the increase was 12.5%.
Utopia is of course far from being reached: at the time of writing there
are immense problems, economic and social, looming. The progress of
the people of rural Ireland however, in the period under review, has
been by any standards remarkable. The contribution to this progress by
the Rural Electrification Scheme has been a major one and has fully
justified the faith of its architects and that of the founders of the Shannon
Scheme in the infant years of the Irish Free State.

262
The old and the new. Above: Milking in the early twentieth century;
below: A modern electrically powered rotating milking parlour.
APPENDIX ONE

The White Paper

REPORT ON RURAL ELECTRIFICATION

IN PREPARING THE REPORT, Thomas McLaughlin and his team of assistants, P. J.


Dowling and A. J. McManus, drew freely on the proceedings on various World
Power Conferences, the reports and publications issued by the Rural Electrifi-
cation Administration of the United States, The Hydro Electric Commission of
Ontario, and the British Electrical Development Association.
It first set out the dimensions of the problem.

Of the 221 cities and towns totalling 1.15 millions in population, 193 had EsB
supply, 26 were supplied by local undertakings and 2, with a total population of
1,077, were without supply.
Villages of 200-500 inhabitants comprised 215, of which 104 had Ess supply, 28
had supply from local undertakings and 83 (total population of 23,690) were
without supply.
There were 386 villages of under 200 inhabitants (total population 47,600) of
which 80 (total population 11,200), had EsB supply; 13 (total population 1,700)
had supply from local undertakings and 293 (total population 34,700) were
without supply.
Finally, there was a scattered rural population of 1.7 million persons of which
only 18,000 enjoyed the benefits of public electricity supply.

Part I of the body of the document examined the design, construction and
likely costs of a practical rural electrification scheme. It surveyed the existing
distribution network and visualised the rural supply system as a natural extension
of this. However, it saw that growth would have to be planned on a complete
area basis rather than as a ‘ribbon’ form of development. The line design
provided for was such as to permit of the lines being strengthened in a simple
and economic manner as the load grew.
Four trial areas were selected to represent the three main categories of
farming, milk production, tillage and cattle grazing. These areas, which were (i)
near Thurles, Co. Tipperary, (ii) around the village of Glanworth, Co. Cork,

264
The White paper

(iii) between Ferns and Enniscorthy, Co. Wexford and (iv) around Ardee,
Co. Louth, were also selected to represent areas where the farms were predom-
inantly of large, medium and small size.
Complete detailed engineering designs, working plans and unit costs for the
various Operations were worked out using a cheaper, simpler and lighter design
than for networks designed to serve areas of concentrated population. The total
cost of supplying the estimated 402,750 rural premises was then worked out and
came to a first approximation of £17 million at pre-war prices, or an average of
£42 per premises. Finally, this part of the report emphasised the magnitude of
the proposed scheme. On the basis of the average of 5.37 dwellings per mile of
supply line given by the models, this would entail 75,000 miles of line. The
highest rate of construction hitherto reached at the peak of the Shannon Scheme
by the contractors was 650 miles in one year and the highest reached by the EsB
itself in the subsequent years was 380 miles per annum. This highlighted the
necessity for a completely new type of construction organisation if the rural
scheme was to be completed within a reasonable time.
Part II of the report consisted of a review of rural electrification development in
other countries. In many European countries the farming community lived as a
general rule in large clusters of farmhouses or villages with a consequent high
density of dwellings in the area to be served with electricity. Supply to these was,
from a technical and economic viewpoint, the same problem as the supply to
rural villages and small towns, a problem which had already been solved in
Ireland. The Irish farmers, however, generally lived in well-scattered locations,
a situation which might best be compared to the United States or Canada.
Progress in rural electrification in the u.s. had been impressive. In California it
was claimed that 84% of farms had electricity supply and New England, New
Jersey, Massachusetts and Connecticut had achieved figures of 93%, 83% and
83% respectively. However, it was emphasised that the U.s. was such a vast area
and conditions so divergent in the various states that any discussion on rural
electrification as applied to the country as a whole could be of little value.
It was in Canada, in the organisation of electricity supply to rural consumers
in the province of Ontario, that an almost exact parallel to the problem under
investigation was found. This development was therefore examined in detail and
reported on at length. The salient features of the Ontario development were as
follows:

The farmers lived in scattered dwellings as in Ireland.


The supply authority was a statutory Board akin to the Ess. Members of the
Board were appointed by the government and finances provided by way of
advances from the State.
Supply was given from a national electricity network.
The State electricity board organised, built and managed the local rural network
and sold direct to the consumer. Development by way of supply to communal
rural groups had been tried and abandoned.

265
The Quiet Revolution

In order to bring the burden of the capital expenditure within reasonable


economic limits, it was found necessary for the government to subsidise devel-
opment to the extent of 50% of the capital cost.

In no other country or province studied was there such a close parallel with
conditions in Ireland. Furthermore, in no other case was the problem of subsidy
dealt with so clearly and cleanly. A large number of subsidy systems and other
means of giving financial assistance operating in other countries had been
examined but all were in some way over-complicated or, to quote one of the
authors, ‘messy’.
When the Irish report was being prepared it was not possible, because of the
war, to visit and examine systems in other countries. There was almost complete
dependence on published reports and other such literature, and in this regard
the well-produced Annual Reports of the Hydro Electric Commission of Ontario
were of the greatest assistance.
Part III of the report dealt with organisation, management and service. Organ-
isation was approached from two main considerations, technical and
administrative.
Technical considerations required that supply must be extended in the first
instance from the existing 10,000 volt system. The focal points from which the
rural lines would radiate would therefore be the 38,000 volt/10,000 volt trans-
former stations. This would be done by means of 10kV, three-phase ‘backbone’
lines. Thus the country would be divided up into a number of large rural supply
areas. Each of these would form a unit in itself, but the networks within it would
be built up by a gradual process, construction necessarily extending outwards
from the existing 10kV system. Within these large rural supply areas, any smaller
areas could be taken for construction but it was sensibly enjoined that areas
closest to the existing 10kV system must be given priority in development.
The administrative aspect involved consideration as to what form the rural
organisation should take. Should administration be undertaken by co-operative
associations of rural dwellers (as in Sweden and in many parts of the U.s.) or by
the Electricity Supply Board itself? The report gave examples from Sweden of
the pitfalls involved in getting small co-operative supply organisations to keep
proper books and introduce proper rates of charge. It pointed out that in Ontario
where rural networks were developed in the manner and on the national scale
contemplated for Ireland, the sale of electricity to communal groups was tried
and abandoned in favour of direct sale to individual consumers by the State
electricity board. On the other hand the report demonstrated how neatly and
economically the administration of the rural scheme in Ireland, if carried out by
the EsB would mesh in with the existing ‘District’ and ‘Area’ organisation.
Finally, Part III took up the question as to how best the electricity could be
used to improve farming productivity. It emphasised the great value placed on
rural electrification internationally as a means of improving social conditions in
rural areas, lessening the burden of drudgery on farmers and their families,

266
The White paper

brightening their lives and reducing the great gap existing between the amenities
of city and country.
Part IV of the report dealt with the financial aspect of extending the electricity
network into the rural areas. Up to now, every extension of supply by the EsB
was subject to rigid commercial evaluation and was not undertaken unless it
produced an adequate financial return. This would be patently impossible in the
case of widespread rural electrification. This part of the report examined the
capital involved, the annual costs arising from the capital expenditure and the
possible methods of charging the consumer in order to recover these costs. It
came down on the side of having a uniform basis of charge for supply to all rural
areas. It also accepted that this charge should be in two parts: a ‘fixed’ charge
which was based on the annual costs of the rural distribution network and a
‘unit’ charge for the actual energy consumed based on the cost of generation and
transmission. This ‘two part’ method of charging was to be one of the most
misunderstood and hotly challenged aspects of the scheme.
Having examined the many options on which the fixed charge could be based
— valuation of holding, acreage of arable land, the size of the electrical installation
among others — and the problems that each system would give rise to, the report
finally recommended the floor area of the dwelling-house and out-offices as the
basis for charge, very much on the lines of the 1936 Schedule of Rates for Small
Villages and Isolated Rural Consumers. While, like all the others, it was not free
of anomalies, it had some relation to the quantity of electricity likely to be used
and also, in some measure, to the householder’s relative ability to carry the
financial burden.
The report then went on to consider the scale of the fixed annual charges per
consumer on the principle that on the average for all areas, the sum of the fixed
annual charges in a rural supply area should not be less than the fixed annual
costs for the area. These costs, which included interest on the borrowed capital,
sinking fund for repayment, depreciation, operation and maintenance of net-
works, had been estimated to amount to approximately 12% of the capital cost
of the network for the area.
As a first approach to the problem of fixing a scale for the tariff, an estimate
was made of the results of applying the scale in the existing rural tariff to the
four small-scale trial areas for which plans and estimates of capital costs had
been worked out. It was concluded that with the scale of fixed charges in this
tariff, the necessary return of 12% on the capital investment would hardly be
achieved in practice from the sum of the fixed charges and that it would be
unwise to calculate on a better return than 9.7%.
In the examination it became apparent that even a return of this order was
dependent on the exclusion of individual premises where the capital cost of
connection was excessive in relation to its fixed charge revenue. It was calculated
that a limit of 14 times the annual fixed charge (or a ‘criterion ratio’ of 14),
would allow 86% of the dwellings in the trial areas to be connected and give a
return of about 10.8%. However, it was thought probable that some 20% of
these would choose not to take supply leaving 80% of 86% (or 69%) of the total

267
The Quiet Revolution

dwellings in the areas to be connected. The cost per dwelling of connecting these
would be higher than if the whole 86% took supply while the average revenue
per dwelling need not necessarily be any better. Calculations showed that the
total fixed charges desirable might prove to give only 9.7% of the capital cost
and therefore to enable this 69% to be connected to the criterion ratio would
have to be raised to 15.6 (i.e. all premises costing 15.6 times their annual fixed
charge or less would be included).
This 9.7% return on capital expenditure was of course short of the estimated
12% required to break even, but it was considered the best compromise. Raising
the fixed charges by about 25% would theoretically bring about a break-even
position, but in practice it would probably mean that fewer householders would
opt for supply. As it was, under the existing rural tariff, the farming consumer
with out-offices would pay on average about 75% more in fixed charges than for
a correspondingly sized dwelling in a town or large village. Increasing this
differential still further was not considered advisable. Another method of theo-
retically breaking even would be to lower the criterion ratio, but if this course
was followed, a ratio of about 8.5 would be required to break even, which would
mean only about a 30% connection — certainly not ‘rural electrification’ on the
extensive scale hoped for.
Having considered these options, the report maintained that the conclusion
already arrived at must stand, i.e. that, consistent with a reasonable degree of
development in the trial areas, viz. 69% of the dwellings connected, the per-
centage return on the capital investment would not be greater than 9.7% per
annum. Consequently, a deficiency would result (involving some form of
subsidisation).
In conclusion, the report referred to the success of rural electrification in other
countries and stressed that the development of electricity supply to serve the
rural and farming community in Ireland must, as in these other countries, be a
gradual process. The problems involved in engineering organisation, manage-
ment and finance (and their solution) had been clarified by the experience in
other countries. Development in Ireland could proceed on much smoother lines
and this should lead to speedier progress.
There were, however, many problems and difficulties peculiar to the supply
of the rural areas. They had been set out in the report. The primary problems
were of a financial nature — the high degree of capital investment called for and
the relatively low monetary return that the investment could yield. In other
countries this had required State subsidies. Before any progress could be made
in the work of organisation and planning of development on a national scale in
Ireland, the fundamental question of finance had to be clarified. The extent of
the necessary capital called for was set out in the report.

268
APPENDIX TWO

Definition of ‘Rural’

BECAUSE OF THE LARGE AMOUNTS OF MONEY involved in the government


subsidy, it was important that from the very outset a clear definition of ‘rural’
should be agreed between the EsB and the government. As initially agreed,
‘rural’ included all areas situated outside the following categories.

(i) Places legally defined as County Boroughs, Boroughs, Municipal Boroughs,


Urban Districts or Towns.
(ii) All towns and villages of over 250 population.
(iii) All towns and villages listed in the census returns 1936 (i.e. collections of
twenty houses and over) already supplied by the Ess.
(iv) Areas occupied by the following categories of consumers:
(a) consumers on the fringes of networks supplying areas such as (i), (ii) or
(iii) above which might be connected LT’ to such networks;
(b) loads which were situated in the immediate proximity of areas such as
(i), (ii) or (iii) above and which might be considered as owing their
position to the proximity of such an area;
(c) allisolated loads over 100kVA maximum demand.

The general approach in the above definitions was that towns and villages of
above 250 population were not intrinsically rural. Similarly, isolated loads of
over 100kVA were considered as industrial and did not come within the definition
of ‘rural’ electrification.
By the spring of 1949, however, following the rapid take-off of rural electri-
fication, it had become obvious that many towns and villages of over 250
population could find themselves in the position of seeing EsB supply in the rural
areas surrounding them while still being without electricity themselves or depend-
ent on a usually inadequate local supply. Pressure mounted therefore to have
them included in the scheme. The inclusion of such centres of population would
in nearly every case improve the economies of supply to the adjacent rural areas.
Such inclusion, on the other hand, would have only a marginal effect on the
finances of the scheme as a whole as the population and fixed charge revenue
(of the order of £30,000 per annum) in the villages and towns in question was

269
The Quiet Revolution

small in comparison with the population and expected fixed charge revenue (of
the order of £1.5 million) covered by the whole rural scheme.
In early summer of that year, agreement was reached with the government on
a re-definition in which ‘rural’ areas were all locations other than the following.

(i) Towns and villages listed in the census returns (i.e. collections of twenty
houses and over) which were already supplied by the Ess.
(ii) The following towns not supplied by the EsB
Boyle, Co. Roscommon — Population 2,093
Buncrana, Co. Donegal — Population 2,295.
(iii) Areas occupied by the following categories of consumers:
(a) consumers on the fringes of networks supplying areas under (i) and (ii)
who might be connected at low tension to such networks;
(b) loads which were situated in the immediate proximity of areas under
(i) and (ii) and which might be considered as owing their position to the
proximity of such an area, e.g. institutions, housing schemes.

The principal effect of the alteration in definition was to include within the
scope of the scheme sixty-two towns and villages not then supplied by the EsB.
These fell mainly into the 250-500 population group with forty-seven villages in
this group. There were eleven in the five hundred to one thousand population
group and four in the one to two thousand group.

270
APPENDIX THREE

The Fluctuating Minimum Return

ONE OF THE MOST PUZZLING FINANCIAL ASPECTS Of the scheme to the lay
person was the frequency with which the minimum return required from indiv-
idual consumers changed. This was the percentage return that the standard
annual fixed charge on the premises gave on the capital cost of supply.
The original minimum return of 5.7% has been pitched, in so far as this could
be done in the presence of so many uncertainties, so as to achieve the coverage
recommended in the White Paper (i.e. supply at standard rates could be offered
to 86% of premises). This would remain valid provided only that there was no
change in the original relationship between capital costs and fixed charges. A
rise in costs without a corresponding rise in the level of standard fixed charges
would mean larger numbers of premises failing to reach the 5.7% return required
and consequently reduce the number to which supply could be offered at standard
rates.
From the very first years of the scheme this problem was highlighted by a
succession of rises in costs of material and labour. There was a strong reluctance
on the part of the government and indeed of the Ess itself to match each
successive rise in costs with a simultaneous increase in fixed charge rates. (Indeed
the term ‘fixed charge’ itself, as related elsewhere, was interpreted by many as
meaning that once set it could never be changed). The first increase in fixed
charge rates (10%) was not effected until 1956 by which time over £16 million
had been invested in the scheme and capital costs had risen to 150% of those
obtaining in 1946.
In order to preserve the originally planned coverage and thus the spirit of the
scheme, each substantial rise in capital costs was, therefore, followed by a
reduction in the minimum return required. This of course had a depressing effect
on the overall economics of the scheme, but it was hoped that in the long run
suitable adjustments in the capital subsidy would be obtained in compensation.
Table 9 shows the variations in the minimum return required from 1946 to 1970.
It will be noted that upward revisions in fixed charge levels (e.g. 13 September
1956 and 22 November 1960) allowed a corresponding raising of the minimum
return, but in general the trend is downwards as fixed charge increases failed to
keep up with the escalation in capital costs. The downward trend in the minimum
return resulted, as a matter of course, in a corresponding downward trend in the
return from the scheme as a whole leading to a heavy annual ‘loss’ on the scheme

Aa
The Quiet Revolution

which reached £3.2m in the year 1970/71, the last year in which separate iuiele
accounts appeared in the EsB Annual Report. In effect, of course, this ‘loss
represented the amount of cross-subsidy from the EsB’s urban consumers.

TABLE 9
Minimum Return Required

Areas under
Construction Post-Development Remarks

Farm & Farm & Criterion ration of 14


domestic others domestic others referred 46 sac While

Paper. When Scheme


commenced costs had
increased 50% & FC
by 20%.
14 x © =17.5 new
120

CR = 5.7% giving
expected return of 7.76

1946 Original scheme 5.7% 7.76% — — Householders


returning less than
5.7% (or 174 to 1) not
accepted

Dec. 1947 Original scheme 5.7% 7.6% 7.16% 7.76% Above now accepted at
7.76% of return

ibsyppal Syl Rise in costs 5.0% 7.16% 5.0% * 7.76% To preserve coverage
*PD return relaxed

1219152 Rise in costs 4.2% 7.16% 4.2% 7.16% Ditto

2.4.54 Directors raise PD 4.2% 7.16% 6.0% 7.16%


return

13.9.56 Rates of Charge 4.6% 8.5% 6.6% 8.5% To preserve coverage


revision

13.10.56 Directors’ decision 4.6% 8.5% 9.0% * 9.0% Capital Contribution to


bring up to 9%; no ssc
allowed

11.2.58 Pressure from 4.6% 8.5% 6.6% * 85% Capital Contribution


government, Director’s abolished
decision

25.2.58 Rise in capital costs 4.0% 8.5% 6.6% 8.5% To preserve coverage

272
The Fluctuating Minimum Return
26.4.60 Directors’ decision 4.25% 8.5% 6.6% 85%

22.11.60 ROCrevision 4.70% 9.0% 7.30% 9.00% To preserve coverage

15.6.62 Increased subsidy 4.70% 9.0% 5.00% 7.40%

19.10.62 Increase in costs 4.40% 8.50% 4.70%* 7.00% *With sliding scale

25.6.63 Increase in costs 4.20% 8.10% 4.50%* 6.70% *With sliding scale
(Last of original areas completed in 1965 — Minimum Return now only applies
to PD)

Farm &
domestic others
21.5.68 Directive from government 4.5% 6.7%

15.11.68 |=ROC revision 4.9% 7.2% To preserve coverage

23.1.70 ROC revision 5.2% 7.7% Ditto

CRITERION RATIO AND MINIMUM RETURN

It was estimated in the White Paper that the annual fixed charge revenue
would have to reach 12% of the gross capital cost for the scheme to break even.
It was also held that it would be unwise to calculate on a better overall return
than 9.7% by applying the existing ‘rural’ fixed charge tariff and even this
depended on the exclusion of premises where the ratio of capital cost to annual
fixed charge revenue was excessive (p. 36 of White Paper). It was estimated that
a limit of fourteen times the annual fixed charge (or a criterion ratio of fourteen)
would allow 86% of dwellings in trial areas to be offered connection and give a
return of 10.8%. However, it was probable that 20% would not take supply,
leaving 80% of 86% (or 69%) to be connected. All the above calculations were
on the basis of pre-war costs.
When the scheme eventually commenced costs had increased by 50% while
the increase in the fixed charge tariffs permitted by the government was only
20%. Therefore, in order to achieve the original coverage, the criterion ratio
had to be increased to 17.5 (i.e. 14 x = ), corresponding to a minimum return
of 5.7%. On the same basis, the expected original overall return of 9.7% was
reduced to 7.76% (i.e. 9.7% x ),
150

213
APPENDIX FOUR

Fixed and Unit Charges

WHILE THE GOVERNMENT HAD STIPULATED in its original approval of the Report on
Rural Electrification in August 1943 that current rural rates were to apply subject
to any war increases generally applicable, certain changes on the then existing
rural tariff were proposed by the EsB in January 1946 and approved. While
keeping the total fixed charge in an average area constant, the out-office fixed
charge was reduced by about 33%, the loss in revenue being made good by a
12.8% increase in the fixed charge for dwellings. The increase in fixed charge
caused by the inclusion of the out-offices was spread more evenly (as a percent-
age) between the different dwelling-size groups.
It was also proposed that the scale of fixed charges set out in the report, which
were based on 1939 level of capital costs, be increased by 50% to bring them to
the 1946 level of costs, but the government would not agree to an increase of
more than 20% (letter from Minister, 2 July 1946).
With regard to the unit charges for consumers on the two-part tariff the
Engineer-in-Charge REO in a submission to the Board dated 11 July 1946 wrote
as follows.

The unit charges (proposed) are not the same in the existing rural tariff, the
published report and the report of 30 January 1946.
The difference is shown below:

Size of Block Price per


per 2 months Unit
Existing Rural Tariff 60 units 1.75d
1939 plus 40% 300 units 1.05d
over 360 units 0.70d
Published Report plus 40% 100 units 2.8d
over 100 units 1.05d
Tariff proposed in Report 80 units 2.5d
of 30th January 1946 over 80 units 1.0d

It will be remembered that there was no special rate for Motive Power for farmyard
purposes in the published Report tariff or the proposed tariff. This was the reason
for increasing both the size and price of the first block of units. In view of the
Minister’s letter requesting ‘that the existing rural tariff rates be made available to

274
Fixed and Unit Changes

rural consumers’, it is recommended that the proposed rates of 2.5d and 1d as set
out above be adopted as roughly equivalent. Further, instead of offering the existing
special water-heating rate of 0.7d and 0.46d, it is recommended that an 0.75d rate
be offered on the proposed rate for all units over, say, 360 units per period.

The proposed unit rate would then be as follows.

For first 80 units (per two-monthly period) 2.5d per unit


For next 280 units 1.0d per unit
All units over 360 0.75d per unit

This was adopted and incorporated into the 1946 ‘Rural Rates of Charge’ leaflet.
In June 1951, due to an escalation in the price of coal (British coal at £4 per ton
was difficult to obtain and American coal at £8 per ton had to be substituted),
0.3d. per unit was added to all unit charges.

215
APPENDIX FIVE

Things Chiefly Technical

THE EXISTING SYSTEM

AT THE COMMENCEMENT OF THE SCHEME there was already afairly widespread


electricity network in the State supplying 250,000 urban consumers. A small
110kV main transmission grid linked the main cities of Dublin, Cork, Limerick
and Waterford, with extensions to Dundalk and Carrick-on-Shannon. A second-
ary transmission system of 38kV lines joined these and the other main load
centres in a more widespread pattern.
As the 38kV sub-transmission system was already widespread and could be
further extended at a reasonable cost, it was decided that there would be no
necessity to adopt a higher voltage than the already standard 10kV for rural
distribution. Where long distances or high loads caused feeding problems, these
could be overcome by providing extra 38k V/10kV supply points from the existing
38kV grid or on extensions of it. By standardising on 10kV as the distribution
voltage, it would also be possible to incorporate many of the existing 10kV
‘urban’ feeders into the rural supply system. In the event, it was found possible,
in the case of some very long 10kV feeders where excessive voltage drop was the
problem (e.g. in west Cork, west Kerry, Connemara, west Mayo and west
Donegal), to delay investment in new feeding points by using 10kV boosting
transformers at points along the line. Where the line loading subsequently caught
up with the original capacity, new 38k V/10kV injection points were subsequently
provided to reduce losses and provide a more secure feed with the original
network acting as standby.

BASIC DECISIONS

As the resources of the State were severely limited, it was vital that the
maximum coverage should be achieved for the money available, but without
departing from acceptable standards of supply. Two issues of importance had to
be resolved at the outset: firstly, should supply normally available be three-phase
or single-phase, and secondly, what assumptions should be made regarding the
demand to be catered for? On these decisions would depend to a great extent
the ultimate cost and success of the scheme.

276
Things Chiefly Technical

THREE-PHASE VERSUS SINGLE-PHASE

The big advantage of three-phase supply from the farmer’s point of view was
that three-phase electric motors were simpler, cheaper and more easily obtained,
especially in the larger sizes, than single-phase. In the farmhouse, of course
three-phase supply had no advantages over single-phase. Many Continental
farmers and most British farmers had three-phase supply for motive power and
the more progressive Irish farmers were well aware of this fact. However, the
average British farm was far larger and had a higher potential electricity demand,
while on the Continent the farmhouses and farmyards were often concentrated
into sizeable villages requiring three-phase supply in any case (as indeed did
similarly sized villages in Ireland).
In Ireland, for historical and other reasons, most farms were small and farmers
almost invariably lived on their holdings, which resulted in a very dispersed
pattern of load. Furthermore, at the time, the general level of farming activity
and standard of living was very low. It could therefore be expected that in the
initial years the average demand and consequently the revenue forthcoming
from this dispersed pattern would also be low. An average annual consumption
of eight hundred to one thousand units per consumer was assumed for the
purposes of the original design. The problem was to devise a system that would
meet adequately the expected initial demand and require the minimum initial
capital investment, but would be capable of expansion easily and at a reasonable
cost to meet growing demand.
The solution was the ‘dual three-phase and single-phase, phase/phase’ distrib-
ution system.' In this, a three-phase three-wire 10kV line ran through the centre
of each area of about 25 square miles. This joined two 38kV stations, could be
fed from either source as required and thus formed a secure supply ‘backbone’.
Branching out from this, single-phase 10kV spur lines connected between two
phases, penetrated into all parts of the area, feeding at intervals small single-
phase 10kv/LT pole-mounting transformers serving groups of houses or even
individual houses.
In most cases these single-phase lines were simply convertible to three-phase
by the addition of an extra conductor, while the transformers could easily be
replaced by larger sizes as the load grew. The length of a spur or branch could
be anything from a few hundred metres to 20km with thirty to forty transformers
in some extremely sparsely populated districts. The maximum load in these last
was unlikely to exceed 200-250kVA in the initial years. In the case of large
groups, the LT lines were so designed that they could easily be converted to
10kV. This would enable a group of houses initially fed by one transformer to be
‘split’ later into two or more transformer groups at the minimum cost to meet
growth in load.
The decision turned out to be a sound one. Initial investment was kept at a
minimum and subsequent system reinforcement could be paced to keep up with
growing demand and growing revenue. As far as motive power was concerned
single-phase motors of up to 3hp could normally be supplied from the single-

277
The Quiet Revolution

phase system. They were easily available and while they were somewhat more
expensive than their three-phase counterparts, the price difference was only a
small fraction of the saving in the supply network costs. 3hp motors were
adequate, and indeed still are, for almost every job on the average Irish farm.
In the comparatively few cases where larger motors were necessary, up to about
40hp could still be supplied single-phase although these motors were compara-
tively dear and normally required some local strengthening of the supply arrange-
ments. For requirements above this, three phase was normally made available
under a special contract. In recent years, static and rotary converters have made
it possible to supply comparatively large three-phase loads, up to 100hp from the
single phase system at a fraction of the cost of converting the lines to three phase.
The cost saving is particularly high where three-phase supply is required for
reasons other than size of load, (e.g. many specialised electrical machines with
a comparatively low power demand are available only in three phase form). In
one example, the cost of line and transformer conversion would have been in the
region of £8,000, whereas the problem was solved by the installation of a
converter at a cost of £400. (See Appendix 6 — Three-phase versus single-
phase.)

LOAD ASSUMPTIONS

Here again it was important to steer a course between keeping the capital cost
to a minimum and meeting adequately the power demands of the consumers.
Studies were made on the development of electricity consumption in rural areas
of other countries; agricultural experts were consulted, as were numerous
potential consumers. As a result, for the purpose of arriving at the design load,
an empirical relationship was developed between the size of house and out-
offices, as expressed by the Fixed Charge, and the expected electricity demand.
This came to 0.4kW of demand per £1 of bimensal fixed charge (or per £6 of
annual fixed charge). For very small dwellings a minimum of 0.25kW was taken
while even in these cases, if the application form showed that the immediate
installation of any substantial appliance was intended, a minimum of 1kW was
taken. In the case of a large number of consumers, eight or more, being fed from
the one transformer, it was sufficient to add up to the individual loadings to
determine the overall load to be catered for. When a smaller number of houses
was involved, the aggregate demand was multiplied by a factor ranging from 1.1
for seven houses up to 4.0 for a single house with its own individual transformer
thus:
No. of consumers ee 3 4 ae 7 8
Factor ay PoaPrey ln Ube sy Ey al al

Thus the design took the fullest advantage of ‘diversity’ — the fact that many
electricity demands are non-coincident—to avoid unnecessary capital investment.
This formula served the EsB well in the initial development stage by permitting
the rapid connection of a very large number of consumers at the lowest possible
cost while at the same time providing an acceptable standard of electricity supply.

278
Things Chiefly Technical

ELECTRICAL DESIGN, 1OKV

The backbone of the rural system was to be three-phase 10kV lines running
between 38kV/10kV stations. Ideally there would be two such lines connecting
adjacent stations and these would be so routed as to feed through the centres of
four to six rural areas of 65 to 90 square kilometres (25 to 35 square miles) each.
With the load assumptions described earlier, the 10kV rural distribution
network was designed so as to slot in with the minimum disturbances to the
existing 38kV and 10kV system. The design provided for the two backbone lines
to unite at some distance, about one kilometre, before entering the station
compound. This reduced the number of 10kV cubicle outlets necessary at the
station. A typical 38kV/10kV station would have one urban outlet for the
adjoining town and three rural outlets, each rural line bifurcating to provide two
rural backbone lines which again united and terminated at an adjacent station.
Halfway along each backbone line a ‘normally open point’ was provided,
generally an air break switch locked in the open position thus ensuring that each
half was under the control of its normal feeding station while allowing alternative
feeding in an emergency. With the average loading conditions, it was desirable
to have the 38kV/10kV stations not more than 40-50km apart so as to keep the
voltage drop to a maximum of 5% at the normally open point 20-25km out.
However, as could be expected, local geography and other circumstances
frequently made this target impossible to attain.
The conductor originally chosen was 25mm” hard-drawn copper for the back-
bone and main spur lines. The longer single-phase spur lines were also designed
to be in 25mm/ copper and short spurs or branches off the longer spurs feeding
three or four transformers could be erected in the cheaper 16mm?’ galvanised
steel. Because of the danger of corrosion from the salt-laden atmosphere, steel
conductor was not erected within 16km (10 miles) of the sea.

VOLTAGE DROP

The maximum allowable variation from the nominal 220 volts at the consumer’s
meter was + or — 9%. This gave greater elbow room in the case of spurs at the
feeding end of the backbone line (where the 10kV voltage drop was very little)
and considerably less for spurs taking off near the normally-open point where
the drop on the H.T. might be up to the allowable limit of 5%. The situation
was, however, helped considerably by the provision of alternative tappings
(off-load), usually + or — 5% on the HT side of the rural transformers. Thus,
transformers near the feeding station could be set to provide a ratio of 10,500
volts HT to 220 volts LT while at the remote end the tapping could be 9,500 volts
HT to 200 volts LT.
In the feeding stations the provision of on-load tap changing facilities in the
38kV/10kV transformers was of great advantage as they followed the setting up

279
The Quiet Revolution

of the output 10kV voltage where the lengths of rural lines were abnormally
long, and also the provision of voltage compensation for load increases.
Even in 1983, after an average of twenty seven years of service, the 10kV
system as designed was still proving adequate to cater for the greater part of the
rural demand. This was achieved by continuously providing extra 38kV/10kV
injection points as the load grew to levels far exceeding the original design level.
Between 1946 and 1981 the number of these injection points on the 10kV
network increased from ninety-seven to about 350.

FIELD DESIGN

The typical rural area contained about five hundred dwellings. The main
design instrument was the Ordnance Survey 6 inch map. The initial canvass of
the area would have pinpointed and indexed every house on this map and also
every house built since the map was made. Other canvass documents indicated
for each house its size, fixed charge, whether an ‘acceptance’ or ‘refusal’ and, if
an ‘acceptance’, its likely electricity demand.
Working on this information the designer divided the houses to be supplied
into groups, each of which could be supplied conveniently at low tension from
one transformer. He then determined the transformer size from the load data
and selected the transformer locations so that the minimum amount of low
tension network was required. The 10kV spur lines feeding the transformers and
connecting with the backbone line were then tentatively sketched in so as to
minimise the length required. Meeting all these requirements necessitated con-
siderable juggling, but as the designer gained experience, he was able to produce
an efficient preliminary design very quickly. He then walked over the ground,
frequently finding that he had to alter the route or possibly the transformer
location to avoid obstacles not apparent from the map. At this stage he also
fixed pole positions, decided spans and pole sizes and generally prepared for the
construction stage.
As each spur was tapped off from two phases of the three-phase backbone
line, care had to be taken to balance up as far as possible the load on each phase
in the area as a whole and to indicate the tappings to be used clearly on the
design documents.

CONFIGURATION

For three-phase lines, an equilateral triangular formation was used with phase
rotation clockwise looking out from the feeding point. As each half of a backbone
line was normally fed from its adjacent 38kV/10kV station, transposition was
required at the normally open point to permit back-feeding. At a later stage,
horizontal formation for three-phase lines was adopted as this permitted easier
conversion from single phase and allowed shorter and cheaper poles to be used
for new three-phase lines.

280
Things Chiefly Technical

ELECTRICAL DESIGN LOW VOLTAGE

In the case of a large group of houses with an initially low electricity demand,
capital was saved by using one transformer where eventually two or more would
be required. In this case, provision was made for later conversion of portion of
the LT to 10kV by building it initially to the 10kV specification regarding pole
size, clearances, conductor size, tension etc. The only difference was the use of
LT insulators instead of 10kV. The erection subsequently of a second or third
transformer to meet growth of load was very simple and the conversion of the
portion of the LT line involved to 10kV was merely a matter of replacing the
insulators. In the case of other low voltage lines, feeding houses close together,
shorter poles, shorter spans and lower mechanical tensions in the conductor
were employed than in the convertible sections. In this latter case the conductors
were generally in vertical formation with the neutral on top and fused phase wire
at bottom. Until the changeover to aluminium, the copper size normally used in
LT groups was 25mm? with 50mm? used occasionally*for heavy loads. When
aluminium conductor came into use, 50mm? and 25mm?’ sizes with steel core
(sca) were adopted especially where conversion to 10kV was likely. For lower
mechanical tensions, similar sizes but without the steel core were used.

HOUSE SERVICES

It was always desirable to have the aerial from the service pole to the house
as short as possible to achieve maximum ground clearance, avoid clashing of
wires and minimise the pull on the house structure. The conductor used was
usually 16mm? or 10mm* copper or 16mm” sca. The connection at the house end
was frequently to a chimney bracket in the case of one-storey houses or to
insulators mounted on fishtail stalks set into the house structure where sufficient
ground clearance could be achieved. From this point, insulated conductor clipped
to the house structure was run to the meter position. Houses with mud walls
were still to be found here and there. They required special treatment as their
structure could not bear any of the normal fixing arrangements.

MECHANICAL DESIGN

In developing the mechanical design of the network optimum balance was


sought between safety, operational demands, continuity of supply, ground
clearance under different weather conditions and always the overriding question
of cost and availability of materials. A fundamental parameter was the mechan-
ical tension to which the conductor would be pulled as on this would depend the
type and strength and hence the cost of the structures used. The experience
gained by the Ess in line design, building and operation in the previous twenty
years was of immense value in this regard. The quality (99.9% purity) and
characteristics of hard-drawn copper had not changed over the war years and the

281
The Quiet Revolution

German (Verband Deutscher Elektrotechniker — VDE) standard’ which had


been applied in the pre-war years was continued.
The selected working tensions were:

12 kg per sq mm for 50mm’ copper or a total end pull of 600 kg per conductor
19 kg per sq mm for 25mm? copper or a total end pull of 475 kg per conductor
24 kg per sq mm for 16mm? galvanised steel or a total end pull of 384 kg per
conductor.

Using these parameters, sag charts for field use were produced showing the
correct sag under various combinations of span and erection temperature to
achieve the working tension for each size of conductor.
In the field, ‘sagging’ was carried out by the chargehand selecting a span of
about average length in the middle of the stretch of line and nailing a lath
horizontally on each of the adjoining poles at a distance down from the point of
suspension corresponding to the correct sag. By sighting between the laths it was
simple to determine the required sag had been achieved. During sagging oper-
ations the conductor was suspended on free-running pulleys fixed to the cross-
arms. The operation involved first pulling up the conductor tightly until it was
slightly more taut than the sag required. This was to ‘kill’ initial elasticity. The
tension was then slackened until the conductor dropped to the proper sag when
it was transferred to the insulators and bound off. At the maximum loading
conditions designed for, a factor of safety of 2 was allowed.

CHANGE FROM COPPER TO STEEL-CORED ALUMINIUM

By early 1947 copper was becoming prohibitively expensive. The price of


aluminium on the other hand was remaining stable and the question of substi-
tution was urgently examined. The tensile strength of pure aluminium is very
low but a successful combination of a high tensile steel core to take the mechanical
tension, overlain with strands of pure aluminium which provided the main
electrical path, had been developed and was in use in many undertakings at this
time. Canada had been a pioneer in this field and medium low-voltage lines of
steel-cored aluminium conductor (or sca as it had come to be known) had been
tried out successfully as early as 1921 by the Hydro Electric Commission of
Ontario. Some of these lines had been taken down and examined after twenty-
five years and had been found to be in very good condition. The location,
however, was far. away from the sea and there were widespread doubts about the
resistance of the combination of aluminium and steel to electrolytic corrosion
when exposed to salt-laden moist winds in coastal areas. Furthermore, at many
points the aluminium would still have to be connected to copper terminals or
conductor. There would thus have to be bi-metal connectors used at these points
and there were again fears of corrosion caused by electrolytic action between
dissimilar current-carrying metals in moist conditions.

282
Things Chiefly Technical

The general message was, however, that if certain precautions were taken,
sca should prove satisfactory. The decision was taken to change over and with
the arrival of the first consignment of sca in 1948 the use of copper for rural
distribution lines was discontinued. For the 10kV lines, two sizes of the new sca
conductor were adopted. To replace the 25mm’ copper, No. 1 AWG (aluminium
area) SCA comprising six aluminium strands and one steel strand was selected.
This had a total cross sectional area of just under 50mm?’ (actually 49.48mm/’)
and a conductivity equivalent to 26.67mm/’ copper. The ultimate tensile strength
of the conductor was 1,575 kg. It was known by its international code name of
‘Robin’. (All sca conductor was given code names of birds. Each name stood for
a particular size of conductor and was accepted and understood internationally. )
At 1950 prices the gross cost per kilometre of a 10kV line erected in Robin was
£300 for three phase and £235 for single phase as against £360 and £280 for
copper. To replace the 16mm’ galvanized steel conductor, No. 4 AWG (alumin-
ium area) SCA was selected. This again had one steel and six aluminium strands.
The total area of conductor was just under 25mm? (24.71) and its conductivity
equivalent to 13.3mm? copper. The ultimate tensile strength was 830kg. This
was used on the ordinary 10kV single phase spur lines and on lightly loaded LT
lines. The code name for this conductor was ‘Swan’. The gross cost of a 10kV
single-phase line erected using this conductor was £175 per kilometre at 1950
prices, as against £210 per kilometre for copper. Finally, for aerial house service,
No. 6 AWG, again one steel and six aluminium strands, was used to replace the
16mm? and 10mm’ copper conductor. The code name was ‘Turkey’.
Sag-charts for use in the field were made out but while there was only one set
for copper, which was used for both erection and maintenance work, in the case
of sca there were separate erection and maintenance charts. The erection chart
required the conductors to be pulled tightly to produce 5% to 10% (depending
on length of span) less sag than the design figure. This allowed for stretch in the
conductor which gradually eased out to its permanent sag. The maintenance
chart took this stretching into account and was used for regulating sca lines
which had been in service for some time.
Using 10-metre poles the maximum span achievable for a 10kV three-phase
‘Robin’ line was 90 metres; with 11-metre poles, this could be extended to 98
metres. For single-phase 10kV line in ‘Swan’ using 11-metre poles, the maximum
span was 132 metres if a 168 cm crossarm was used.

SCA VIBRATION PROBLEMS

Aluminium wire is far more brittle than hard-drawn copper and on the earlier
Canadian and American sca lines vibration failures frequently occurred near the
insulator binding. This happened on open flat countryside where low winds could
set up sympathetic vibrations in the tightly stretched conductor similar to the
vibrations in a violin string. The solution was to damp the oscillation by using
armour rods at the suspension points. After some years experience it was found
that Irish conditions did not lead to this type of vibration failure and the use of

283
The Quiet Revolution

armour rods was discontinued. Aluminium tape was used, however, at the
insulator bindings to avoid abrasion and chafing of the soft aluminium strands.
The binding wire itself was of medium hard aluminium.

ELECTRICAL PROTECTION OF RURAL NETWORKS

10kV three-phase lines leaving 38kV/10kV stations were originally designed


to have one of two types of electrical protection, oil circuit breakers with definite
inverse time limit overload relay (RIS), which were generally omitted in small
stations (under 5 MVA), or high capacity fuse holders with drop-out fuses,
normally loaded with 60 amp fuse links. This was not completely satisfactory
owing to the length of time required to restore supply even in the case of
transient faults. The subsequent introduction of automatic reclosers at main
feeding points effected a great improvement in continuity of supply. Two sizes
were adopted, set to cut out at 100 to 200 amps according to the loading of the
line involved. Reclosers have been found most effective in quickly restoring
supply in the case of transient faults caused by lightning or blown debris in a
storm while ensuring safe disconnection in the case of a more permanent fault.*
Where spurs take off from the backbone line they are protected at the take-off
point by 15 amp HT fuses in drop-out fuse holders. Transformers erected directly
on the backbone line are similarly fused. There is no further HT protection on
the single phase spur lines but on the longer spurs, solid isolating links in standard
HT fuse holders are sometimes erected at an intermediate point to facilitate
operation and maintenance.
The small number of fuses in series is due to difficulty experienced in achieving
selectivity. HT fuses of less than 15 amps have now been dispensed with due to
their tendency to blow on very short-time transient faults. Automatic reclosers
at the feeding station ensure that in the case of a long duration short-circuit fault,
the fuse protecting the faulty section will have blown before the recloser goes
through its third cycle, thus isolating this section while maintaining supply to the
rest of the network. This keeps area outages to a minimum.
From 1961 onwards, small rural single-phase transformers were fitted with
low-cost internal HT fuses. These, by blowing at an early stage in the event of
a winding fault, minimise the possibility of a 10kV to LT fault developing which,
by superimposing 10kV on the LT supply could lead to a dangerous situation at
the house service. They also help to identify a faulty transformer on a long
single-phase spur fused only at the take-off point.

EARTH FAULTS

Earth faults are generally caused by a conductor breaking and making contact
with the ground or an earthed item of equipment or by a breakdown to earth in
the insulation of bushings, insulators etc. Under normal conditions, the capaci-
tances to earth of the three phases are equal and balanced. Under the condition
of one phase to earth, the current flowing back to the system via the earthed

284
Things Chiefly Technical

point is the out of balance capacitance current of the two ‘healthy’ phases. On
the rural distribution system the 10kV network is not earthed and the earth fault
currents are too low to blow the line fuse. At the feeding station, however, the
neutral point of the 10kV side of the station 38k V/10kV transformer is connected
to earth by means of a potential transformer. In the case of an earth fault this
detects a potential between neutral and earth and operates an alarm to alert the
operator on duty. Supply will be interrupted to the minimum extent while the
task of locating the fault goes ahead. If the fault is due to a broken line in a
hazardous situation, supply to the section will be immediately disconnected
pending repairs. This system ensures that under earth fault conditions there is
the minimum interruption in the area as a whole.

LT PROTECTION

There are usually two fuses in series on the LT, one at the transformers and
the other, the service fuse, in the consumer’s premises. At 33kVA single phase
transformer requires a 200 amp fuse while a 15kVA will have a 125 amp. The
service fuse at the consumer’s premises varies from 30 to 60 amps, depending on
the size of the house installation.

EARTHING OF LT

The tank of each transformer is connected to earth at the transformer pole


with a maximum value of earth resistance of 20 ohms. For network earthing, the
neutral point on the LT side of three-phase transformers is connected to earth,
while for single-phase supply the lagging phase is earthed (thereby becoming the
neutral). Where possible, this neutral earth is on the first LT pole from the
transformer so as to avoid interference with the transformer tank earth. A
second neutral earth is provided at the end of the LT run and if there are ten or
more LT poles in the group, extra earths are provided. To minimise ‘step-
voltage’ in the ground surrounding the earth pole in the case of a fault, the earth
conductor is enclosed in a PVC tube (in pre-PVC days it was wrapped in a
vaseline-impregnated tape) from one foot above ground-level to a foot and a
half below.
The earth resistance for the neutral of the LT group should not be greater
than 100 ohms. In clayey soils the earth electrode is made up of a number of 0.5
inch diameter galvanised steel rods screwed into each other and driven vertically
into the ground. Sometimes in the early stages a galvanised steel plate was
buried, but in general the steel rods were found to give a lower resistance and
involved far less labour. Where rock near the surface prevents the driving of
rods a galvanised earth strip is buried in a trench. A cheaper but not so effective
method is loosely to loop the earth lead around the pole butt about one foot
below ground level. It is usually necessary to instal a number of these to achieve
the required protection.

285
The Quiet Revolution

EARTHING AT CONSUMERS PREMISES

Owing to the absence of metallic water pipes it was not possible to use the
water mains as an appliance earth as in most towns at the time. At each
consumer’s premises either an earth road or buried earth plate was installed to
which the frames of electrical appliances were connected. However, the high
resistance of this earth connection combined with that of the network neutral to
earth inhibited correct fuse-blowing in the event of the appliance frame becoming
alive.
Earth leakage trip switches were advocated as protection to ensure circuit
isolation in the event of a fault. However, the immediate post-war voltage-
operated models were unreliable and unpopular with consumers who objected
to paying for this (apparently unnecessary) piece of equipment. To improve the
situation, ‘neutralising’, or using the supply neutral to ‘earth’ consumers equip-
ment, was introduced in the early fifties and gave positive fuse blowing in the
event of an appliance earth fault. Successful operation depended on the conti-
nuity of the neutral path back to the transformer and required special measures
to ensure this, such as additional neutral earths on the system network itself,
compression connectors on the network neutral and duplicate screw connections
at the neutral blocks. In recent years the introduction of current-operated earth
leakage trip switches has greatly improved the standard of electrical protection
available in the consumer’s premises. The provision of this last type of protection
is now required by the Ess in all new premises.

PERFORMANCE MONITORING AND SYSTEM IMPROVEMENT

The particular characteristics of the rural system are its widespread and
dispersed nature with very long lengths of line (almost 100,000km — 60,000 miles
—in all) and its huge number of small transformers (about 100,000). Monitoring
the performance of this amount of plant and equipment is a formidable task
which is now being take over by computer through the development of a
distribution management system. This has two main components

Network Digitising UO This involves storing in the computer memory geograph-


ical and electrical data on 66,000km (41,000 miles) of 10kV network so that
network details in any section are instantly ascertainable.

Transformer Load Management UO By identifying the consumers fed from each


of the 100,000 rural transformers and associating this information with the meter
reading and revenue billing process, the computer can provide continuous
statistical information regarding the electricity demand on each transformer.
Regular computer print-outs list every transformer where demand tends to
exceed capacity and so ensure timely replacement. This same information,
combined with that provided by the network digitising process permits instant

286
Things Chiefly Technical

assessment of the network feeding conditions in any group and indicates if and
where network reinforcement is necessary.

LOOKING BACK

At the time of writing it is thirty-seven years since the first pole was erected at
the end of 1946. It is of interest to look back at the performance over this period
of the rural network which now supplies almost half a million electricity con-
sumers. In general it would appear that the technical decisions taken were sound.
The 10kV three-phase and single-phase lines as designed and constructed are
still adequate in the great majority of areas. Load growth has been accommo-
dated in most economic fashion by the provision of extra 38kV/10kV injection
points and by the use of 10kV boosters. Lines and transformers have in most
cases outlived their “book’ life.
A number of severe storms and blizzards tested the design and construction
standards at an early stage. The first big test came in December 1954 when
32,000km (20,000 miles) of line supplying 126,000 consumers had been erected.
The blizzard tore across the centre of the country from Brittas and Kells in the
east to Clifden in the west. Wet snow consolidated on the lines to a thickness of
over four inches accompanied by high velocity winds. Poles, conductor and
headgear were broken or twisted, services were pulled out of houses, some
chimneys were pulled down. Over 10,000 consumers lost supply. The rural line
crews rose to the occasion and all but the most isolated consumers had supply
restored within a few days, the majority within hours.
After storms such as this the performance of the networks was analysed. The
analysis identified areas of weakness where improvements could be made at
reasonable additional cost. With regard to the networks generally, however, it
was a question of balance between two requirements, keeping costs to a minimum
and achieving the highest possible reliability. A factor in the equation was the
speed with which supply could be restored in areas suffering damage from these
rare but intense storms, and it was found that increasing the resources allotted
to the repair organisation so as to give a faster response would give a better
return on investment and a higher overall standard of service than a corres-
ponding investment in the networks. The enthusiasm and commitment of the
line crews to restoring supply, often in the most atrocious weather was, and still
is, an essential ingredient of such service. Without this, even the best organisation
and equipment available would be ineffective. ,
As the networks have grown, so also has the availability of alternative supply
paths to feed the backbone lines. When coupled with the provision of automatic
reclosers to restore supply rapidly in the case of transient faults, the result is a
high standard of continuity on most of the 10kV network, a standard not far
short of that available in the major urban locations. Similarly, the voltage
regulation is being continually improved by the provision of extra injection
points and other reinforcement. Owing to the large distances and sparse popu-
lation, some areas on the western seaboard have still problems in continuity.

287
The Quiet Revolution

Lightning storms are also prevalent in these areas. Extra reclosers and lightning
arrestors have been installed at key points. As the demand increases and industry
hopefully develops in these areas, extra expenditure will be justified in extending
the 38kV system to provide loop feeding.
While it is accepted that the general design of the rural networks is satisfactory,
with hindsight some of the problems encountered in service might have been
avoided. Using Steel-Cored Aluminium conductor near the exposed west coast
proved to be an error. Climatic conditions along this coastline resulted in
considerably more electrolytic action between the two metals than had been
anticipated or indeed hitherto experienced in other countries. The resulting
corrosion required the early replacement of sca line with the well-tried but more
expensive copper. The non-tension bi-metal connectors which were used in the
initial stages were also subject to heavy corrosion, particularly in moist, salt-
laden conditions, resulting in radio interference, heavy voltage drop or total loss
of supply. It could also be said with hindsight that erecting such small trans-
formers as 25kVA and 3kVA was a mistake as in a very short time, with load
growth, thousands had to be replaced by larger units at considerable cost to the
ESB and inconvenience to consumers. If Sk VA units had been erected originally
instead of these, the difference in cost would have been small and the amount of
replacement minimised.
In cases such as these it was a question of learning from experience. There
were few precedents to guide the Irish pioneers. The verdict today must be that
the job was well done; that the Irish rural electricity network is giving a very
high standard of service and good return to the nation for the resources invested
in its creation.

THE RURAL ‘BIBLE’

It was obvious that in a scheme requiring the erection of up to 10,000km (6,000


miles) of electrical distribution lines and the servicing of up to 34,000 houses a
year, speed and accuracy of field design and effective construction techniques
would be essential. The field engineers would for the most part be very
young — energetic and enthusiastic, most probably, but lacking in experience.
A comprehensive manual covering the practical aspects of the work and which
would complement the technological training of the engineering school was
urgently required.
No such manual existed. One of the first tasks of the newly-formed Technical
Division under Harold Montgomery and his assistant J. F. Bourke was to
assemble all the available practical knowledge on the design and construction of
distribution networks in rural locations. From this a field manual was prepared
for use by Rural Area Engineers. The authors drew freely not alone on ESB
experience and practice but also that of the Rural Electrification Administration
of the United States, the Hydro Electric Commission of Ontario and the British
Electrical Development Association. The resulting manual, Design of Rural
Networks, quickly became an indispensable part of every Rural Area Engineer’s

288
Things Chiefly Technical

equipment and earned for itself the title of ‘the rural bible’. Several decades
later W. F. Roe referred to it as ‘one of the keystones of the Rural job,..... it
is a matter of interest that this ‘Bible’ was widely used throughout Europe. I
remember getting requests for it from countries as far apart as Iceland and
Turkey.”
The manual was produced in loose-leaf form so that it could be kept up to
date in technological developments, in the materials situation and in current
costs and prices. It was divided into seven main sections.

Preliminary Survey was based on experience in the field by mature surveyors. It


gave valuable hints to the beginner in dealing with the various problems and
pitfalls which could be encountered.
Electrical Design covered general layout, load and voltage calculations, design
economics, conductor and transformer sizes, electrical protection and earthing.
Mechanical Design gave tables of pole sizes, spans and sags for different sizes of
conductor and their appropriate tensions enabling quick selection to suit the
terrain.
Selection of Materials was one of the largest sections containing as it did tables
covering pole and headgear size, spans, conductor spacing, ground clearance,
stay size, house servicing, metering etc.
Estimating Costs were broken down under various headings enabling the engineer
to give a quick and accurate estimate for any proposed section of work. As
material and labour costs escalated rapidly, this portion of the manual required
constant revision.
Public Relations included relations with government departments and local
authorities. It also covered the EsB’s statutory powers and its obligations to the
general public and to landowners, wayleaves, compensation for damage, inter-
ference with post-office lines, roads, railways, airfields, sports fields, schools etc.
Working Drawings. A comprehensive act of these was provided at the end of the
manual and was constantly revised and added to as available materials and
technologies changed in the course of the scheme.

As one of the main objects of the manual was to eliminate as far as possible
the use of design formulae in the field, there was a comprehensive range of tables
and charts whereby the correct pole size, span, spacing etc. could be read off for
almost any set of circumstances. These speeded up the work immensely while
also giving confidence to the young engineer in the design and layout of the
networks. Following the decision in 1947 to change from copper to aluminium
for practically all overhead conductor it was necessary to revise radically the
design parameters. This involved a major revision in that part of the manual
dealing with spans, tensions and clearances and necessitated the preparation of
a large number of additional drawings covering the new types of hardwear
involved. Even now, thirty-five years after its first appearance, the manual could
still be regarded as an up-to-date guide to rural distribution network construction.
The performance of the rural lines over the period is a tribute to its effectiveness.

289
APPENDIX SIX

Three-Phase Versus Single-Phase

WHILE THE NORMAL ELECTRICITY SUPPLY available in urban areas is three-


phase alternating current, the stringent economies of rural electrification, partic-
ularly with a sparse and dispersed pattern of dwellings, as in Ireland, demanded
single-phase supply. As described in Appendix 6 the result was a three-phase
backbone and single-phase distribution system which ensured single-phase supply
to the maximum number of consumers at the minimum cost. For the first thirty
years of the scheme the absence of a generally available three-phase supply was
accepted with little question, mainly because the power requirements of the
great majority of farmers were confined to pumps, milking machines and small
grinders and hammer mills requiring not more than 3hp, which could easily be
powered with single-phase motors.
In the case of farms — comparatively few in those early days — with large power
requirements, three-phase supply could generally be extended on acceptable
terms because of the considerable extra revenue involved. In the early seventies,
however, with the economic acceleration of agriculture following Ireland’s
accession to the EEC, the power requirements of a growing number of farmers
began to increase. Larger feed-processing units, larger milking machines, aux-
iliary equipment for milk-cooling and storage raised the question of the adequacy
of single-phase supply. More and more frequently the allegation was made that
the lack of three-phase supply was inhibiting expansion of the agricultural sector.
In order to assess the position, an extensive study of the role of three-phase
electricity supply in Irish agriculture was carried out in early 1976 by the Ess.!
The object of the study was to establish in what sectors of agriculture, if any, the
extension of three-phase supply would facilitate expansion and contribute
towards higher productivity on Irish farms. The study took into consideration
the overall economics of extending three-phase supply and of the alternatives
(e.g. extra tractor power) in the case of farms with the greatest scope for
increasing productivity and profitability. It also set out to derive information on
levels of single-phase supply desirable to identify operations to which the
application of electricity would be of benefit to agriculture, and sought to involve
the agricultural industry to a greater extent in the EsB’s forward planning to meet
the needs of the industry in the years ahead.
In addition to intensive desk and field research, discussions were held on the
project with the Irish Farmers’ Association, the Irish Creamery Milk Suppliers’

290
Three-Phase v. Single-Phase

Association, the Department of Agriculture, the Agricultural Department of


University College, Dublin, the Pigs and Bacon Commission, economists, engi-
neers, and agricultural specialists in the research centres of An Foras Taluntais.
In addition discussions were held with the staffs of the British National Agricul-
tural Centre and the Electric Farm Centre in Stoneleigh, Warwickshire, and
finally with suppliers of equipment and machines. The main findings of the study
were as follows.

While lack of three-phase electricity was not inhibiting expansion in the agricul-
tural sector, areas were identified where it could make a positive contribution to
increased productivity on the farm.
Of the farmers interviewed who had single-phase supply, not one identified lack
of three-phase as a hindrance to expansion or increased productivity.
All farmers would like three-phase supply, for reasons of convenience, provided
it was free.
The main requirement for major motive power was for home compounding of
stock feed and disposal of slurry.
For pig enterprises over a certain level, the return on investment, taking into
account contributions for three-phase electricity supply, as well as costs for
storage and milling and mixing equipment, could be very attractive.
In the case of slurry disposal the investigators were not convinced that the power
required for the various stages of the process was such as to require a three-
phase supply.
Forced ventilation for pig housing and large scale poultry producers could be
catered for adequately by single-phase electricity supply.

For the average progressive farmer it was not so much a question of supply
being inadequate as that some of the new plant coming on the market required
three-phase rather than single-phase supply. This applied particularly in the case
of motors. Although single-phase motors of up to 3hp were readily available on
the market, above this rating three-phase was the norm. Especially in the case
of the larger ratings three-phase motors were cheaper, gave better service and
were easier to maintain than single-phase. In addition, some specialised machin-
ery coming into use was equipped with purpose-made three-phase motors even
at ratings of less than 3hp. This applied in the case of small rural industries as
well as in farming, and also to auxiliary equipment associated with glasshouse
heating systems.
While for consumers in proximity to the three-phase backbone line the cost of
conversion of their supply to three-phase could be comparatively low, most
consumers were not in such a favourable location and quotations of severai
thousands of pounds for mains conversion were common, effectively putting
three-phase outside the means of all but the largest consumers. A breakthrough

291
The Quiet Revolution

came with the development of single- to three-phase converters, particularly


when in 1977 two Irish manufacturers put on the market models specifically
designed for Irish needs. These could be installed in the consumer’s premises
and could convert the available single-phase supply into three-phase which,
while not equalling the quality of the normal three-phase mains supply, was quite
adequate for most three-phase equipment. The big advantage was that the cost
was generally significantly less than mains conversion, usually less than half.
For some small industries the saving could be much greater. In the case of a
rapidly expanding fish processing plant in County Donegal, an investment of
£2,000 in 1978 in a converter deferred for several years the necessity to convert
a long single-phase spur line at an estimated cost at the time of over £10,000 and
thus enabled expansion to take place without an undue financial burden on the
enterprise. The cost of a group water supply for Knockbride in County Cavan in
1977 was reduced by £4,000 by the use of a converter, again as an alternative to
converting a single-phase mains supply to three-phase. Two principal types of
converter were employed — static and rotary. Static converters, which depend
mainly on capacitors, are generally more suitable for single motor applications
where the converter can be matched exactly to the motor. Rotary converters
incorporate an unloaded pilot motor with a single-phase input and three-phase
output and are more suitable for multi-motor applications. While the effective-
ness of the converters as a substitute for a conventional three-phase supply was
undeniable and the economics most attractive, particularly in the case of small
loads or as an interim stage in the case of growing loads, it must be said that in
the long term and for loads of any appreciable size the most satisfactory solution
was the provision of a three-phase mains supply.

292
APPENDIX SEVEN
Corrosion
WITH PURE ALUMINIUM CONDUCTOR the question of rust, which is such an
enemy of ferrous metals, does not arise. In the open, a layer of aluminium oxide
quickly forms on the surface, but formation of the layer inhibits further action
rendering the conductor virtually corrosion-proof even in moist, salt-laden
atmospheres. There are problems, however, when an aluminium conductor
carrying an electrical current comes in contact with a dissimilar metal such as
steel, brass or copper, in the presence of moisture.
As aluminium has a very low tensile strength it is not on its own suitable for
use in overhead lines subject to high mechanical tensions. For a long time this
inhibited the substitution of aluminium which was a cheaper metal than copper
and not as subject to wild market fluctuations. The development of a conductor
with a high-tensile steel core surrounded by aluminium strands — steel-cored
aluminium, quickly abbreviated to sca — opened up the whole field of overhead
electricity transmission and distribution to aluminium. It had only about one-
half the conductivity of copper so that twice the cross-sectional area were
required, but was also only about one-third of the weight so that for similar
current-carrying capacity, it was in fact lighter. For rust inhibition, the steel core
was galvanised but it was found that in the moist, salt-laden atmospheres such
as are found around the coast, corrosion quickly took place. When salt and
moisture penetrated between the outer aluminium strands, electrolytic action
was set up between the aluminium and the zinc coating of the steel core. The
zinc, being the more electro-positive of the two metals was quickly eaten away
exposing the steel. At this stage the action reversed and the aluminium,
electro-positive with regard to the steel, commenced to disintegrate into a
powder. As the current-carrying capacity of the aluminium was reduced, over-
heating occurred which accelerated the process until finally all the strands broke.
This type of corrosion was confined to a strip about seven miles wide inland from
the coast beyond which sca gave satisfactory service. For this strip some trials
were made of the rather expensive all-aluminium-alloy conductor which had
87% of the tensile strength of hard-drawn copper and consequently did not
require a steel core. Once again threre were corrosion problems and so in the
1960s it was decided to revert to hard-drawn copper lines for the coastal areas.
The other main corrosion problems occurred where it was necessary to join
aluminium conductor to copper such as with house services or transformer or
fuse terminals. To effect this, aluminium bi-metal connectors either ‘parallel
groove’ or ‘split bolt’ type were used. These had copper liners for the copper
conductor and again in coastal areas electrolytic action resulted in corrosion and
overheating leading to high resistance and failure. Tubular compression type
connectors were then introduced, filled with inhibiting grease or graphite, but
again similar trouble developed when in time the weather washed the grease off
at the exposed end of the connector. In this case, however, it was found that
bandaging the finished connection with a vaseline-impregnated tape (Denso
tape), so as to seal it off completely from the atmosphere gave reliable results.

293
APPENDIX EIGHT

Castletownbere — An Adventure in Rural


Electrification
(Extract from Final Report on Development of Castletownbere Rural
Area by Area Engineer Noel O’Driscoll February 1953)

WHILE BATTLING WITH THE ROCKS in the Schull-Ballydehob fastnesses, blasting our
way to our goal we were informed that our next area was Castletownbere. No
stick of exploding gelignite could produce more of a stunning effect in the Area
Office than when the news reached us. Then we knew that we were being
accepted and acknowledged as mountainy men, men of steel and gelignite,
capable of shaking still further the serenity of the West Cork mountains whose
calm had not been disturbed by the noise of men and clash of steel since
O’Sullivan Beara.
In November 1951 I left Ballydehob to visit Castletownbere area, the future
scene of our endeavours. Looking at the country between Glengarriff and
Castletownbere I wrote off the battle of the Schull area as a skirmish, as I felt
that the real battle was here. Here were crags, crevices, canyons, woods, bogs,
etc., which defied all exaggeration. W. Trueick, the pegging engineer, was very
much depressed at the thought of what lay ahead of him as we climbed up the
winding road from Glengarriff to the heights of Loughavaul and beyond again
to Coolieragh. However, when we topped the climb at Coolieragh the vista of
mountain and sea that met our eyes gave us a temporary respite from our morose
reflections.
Here was a scene that is hard to equal anywhere else in Ireland. Ahead of us
lay the country of O’Sullivan Beara. Away in the distance lay Beara Island like
a sleeping monster resting on the sea, protected on the northern side by the
massive bulk of Hungry Hill, and farther west by a ring of mountains whose
western slopes dip down into the Atlantic Ocean. Behind us we looked across
Bantry Bay at Bantry away in the distance sheltered by the bulk of Whiddy
Island. Nearer to us was Glengarriff with its myriad of islets and heavily wooded
hinterland, cosy and comfortable looking, secure in the shelter of its encircling
mountains.
On a cold November day in the weak wintry sunshine people do not stay long
admiring scenery from such a cold vantage point as Coolieragh, and so we
continued our journey westwards along by Adrigole, close to the Healy Pass,
skirting the foot of Hungry Hill with its silver streak waterfalls and finally we
arrived at the capital of the Beara Peninsula, Castletownbere.

294
Castletownbere

From this cursory survey of the countryside over which our backbone line was
to be erected, we saw that indeed there were going to be many difficulties. Even
at the very start of the line we experienced great difficulty, as the countryside
around Glengarriff is very heavily wooded, criss-crossed with rivers and streams
and massive bounders rear up their ugly heads to block every feasible path. I
called on Head Office in my agony and Mr McEnri arrived and between us we
designed a reasonably satisfactory line out of the tangle. Loughavaul was the
next pegging worry as here our line had to go right up the steep slope of a
mountain and over the top. This was carefully profiled and the difficulty solved
by using A-poles on 185 metre spans. This pegging was done in the cold January
of 1952 and it is not with happy memories that I recall those cold windswept
slopes where we had to use a shield of men on the windward side of the
instrument to prevent its blowing over. Hands were cold and handkerchiefs were
much in use and many a sheep was routed from his place of refuge behind a rock
to give place to our shivering community. The rest of the pegging which Mr
Trueick completed was normal by fantastic standards and I am sure Mr Trueick
will not recall his mountain sliding with pleasure, as it is not such a sport when
one does it on his buttocks. Wayleave difficulties were negligible as no one
denied us the doubtful pleasure of planting poles on the virgin rocks. Then, too,
as the average size of farm is about 8 acres of land divided into about 16 fields,
it would be difficult to put the poles anywhere else but by the fences.
Construction work was exceedingly difficult especially for the first 6 miles of
line. Here where no horse or tractor could travel the men had to drag, haul and
carry the poles from the roadside up to site on the side of the mountain. It was
pathetic to watch the men carrying their pole, now disappearing behind a rock,
lost to view for some time, and again reappearing on the slope of another rock
farther on, slipping and sliding, and it was not melody that drifted back to the
ears of the watcher. It was a great feat to deliver these poles on site without
injury to the men but what was still a greater feat was the getting of the
compressors up there. This was indeed a marvel of strength and ingenuity, for
these compressors were got into positions that even the mountain sheep would
declare impossible. Yet neither man nor compressor sustained injury. More
tricks were performed with blocks and tackle than would baffle the keenest
student of mechanics.
As it was in the end of January 1952 that work commenced, we went about the
job in a rather peculiar way. We started at the end of the backbone line and
worked back. This was the only sensible way to tackle the job, as any work in
the water-logged mountains around Glengarriff would have been impossible at
that time of year. The going was tough all the way but there was one very
decisive factor in our favour and that was the weather. The local people told us
that never in their memory was there such continual fine weather in that part of
the country. I will say that the good weather expedited the work considerably
and made working conditions reasonably pleasant. While construction work was
in progress three other extensions were approved — Glengarriff, Ardgroom and
Allihies. The route of the line to Allihies followed an old road which twisted its

295
The Quiet Revolution

tortuous way right over the top of the mountain. On the Allihies side of the
mountain the slope was steep and so the route had to be chosen very carefully.
It was a case of hopping from crag to crag and in all it took only 9 poles to come
from the top 900’ to sea level. Incidentally, there was no case of ‘uplift’ thanks
to keen eyes. This road over the mountain was never used by mechanical
transport before our arrival. It was more like a goat track than an actual road,
yet the truck carried the poles up there and the tractor brought up the compres-
sors. Usually the tractor had to help the truck up and this was done by driving
a heavy ground pin into the road and attaching blocks. Then the tractor pulled
downhill on the blocks and so pulled the truck uphill. I travelled on this road on
one occasion in the Supervisor’s van . . . Iwalked it ever after.
There are a considerable number of disused copper workings in this neigh-
bourhood and there seems to be considerable mineral wealth in the mountains
all round. Yet no one seems anxious to take the gold out of ‘them thar hills’. It
was mentioned that there was some move afoot to open up Allihies Mines again,
but when and by whom nobody knows. If emigration from that part of the
country is to be stopped, some kind of industry will have to be established there,
as otherwise the statement of the Land Commission agent when he saw us
putting up the poles, “a waste of good timber’, will be true.
Castletownbere was a boom town in the days when there was a garrison on
Beara Island and when the Allihies Mines were in production. Fishing was also
at its peak at that time and money flowed freely. Now there is nothing left but
names in the Labour Exchange. Let us hope that electricity will electrify them
into action again.
Construction work ceased in October last and the hills echoed the thunder of
our exploding gelignite for the last time. We used 2,950 Ibs of gelignite and 3,850
electric detonators. I wonder will that figure be beaten.
There will not be much post-development work, as practically all who were
desirous of supply are connected. At the end of October we left Castletownbere,
that cosmopolitan town where Irishmen, Frenchmen, Spaniards and Englishmen
rub shoulders, and headed eastwards to Cork, 94 miles away.

296
Notes

CHAPTER ONE

* On 14 December 1982 a belated tribute was paid by the Institution of Electrical Engineers
in the form of a lecture given at its London headquarters entitled ‘Nicholas Callan
— Neglected Electrical Pioneer’.

CHAPTER TWO

‘Extract from recommendations made by Messrs Siemens Schuckert on the proposed


Shannon Scheme and quoted in REO News June 1951.
> In fact it overran by six months because a long strike, some very bad weather and the
sinking of a ship containing materials.
> Quoted by P. J. McGilligan in the debate on Rural Electrification in Dail Eireann, 24
January 1945.

CHAPTER THREE

' This was raised to 173 times on the commencement of the scheme.
? In 1949/50, the amount voted, £325,000, covered the repayment of half of the advances
made in the calendar year 1948. In the 1950/51 estimates it was decided by government
that the requirements of the 1945 Act would be satisfied by charging to voted monies an
annuity designed to repay the subsidy moiety of the advances over the same term of years
(50) as the EsB was given to repay its moiety.

CHAPTER FOUR

' With depreciation and ‘other’ costs amounting to 7.08% of capital, a return of 8.5%
would leave 8.5% —7.08%, or 1.42%, available to meet interest and sinking fund charges
of 4.07%. The shortfall therefore would be 2.65%. This is 65% of the total interest and
sinking fund charges, requiring a capital subsidy of 65% in order to break even.
? This practice of lowering the minimum return in the face of rising capital costs was
implemented so as to preserve the ‘coverage’ and it was repeated again and again in the
course of the scheme. Ideally, rises in capital costs should have been followed immediately
by increases in the rates of fixed charge. For various reasons this was not allowed to
happen. Both in timing and in amount, increases in the fixed charge rates tended to lag
behind increases in capital costs so that the general trend was to lower the minimum return
required so as to preserve the number of premises to which supply could be offered without
special service charges or capital contributions. Thus, from a figure of 5.7% in 1946 the
minimum return required dropped to 4.0% in 1958. After a brief rise to 4.7% in 1960, it
dropped back to 4.2% for the final phases of the original development. Minimum returns
required from ‘post development’ consumers were however, almost invariably higher than
these. (See Appendix 3.)
3 In Ireland, Some Problems of a Developing Economy edited by A. A. Tait and J. A.

Hef|
Bristow, Dublin, Gill & Macmillan 1972, there is an analysis by Bristow of this cross-
subsidisation within the Ess. His conclusion was that in the year 1968/69, rural prices were
lower by over 20% and urban prices higher by almost 9% than they would have been in
the absence of cross-subsidisation at the then consumption levels of the time.

CHAPTER FIVE

1 14,3.45 by W. F. R’, REO News, March 1948.


2 The term REO is used generally throughout the book to denote this organisation. In
certain cases, however, the broader title EsB is used where this is considered more
appropriate.
3 The accent on youth was not confined to engineers. A random check in a typical area in
1956 gave the ages of the four principal officers as follows: Engineer — 25, Supervi-
sor — 25, Organiser — 25 and Clerk — 23.
‘ Apart from the Head Office secretarial staff, REO was an all-male preserve. This of course
reflected the culture of the times. To the best of the writer’s knowledge not a single female
application was received for a job on the rural scheme. In present times it is more than
likely that a proportion of the RAEs, AOs, Area Clerks and electricians would be female.

CHAPTER SEVEN

In the White Paper the estimated cost of supplying all rural premises in the country, about
400,000, was £17 million at pre-war prices using wooden poles. To supply the 280,000 (or
69% of all premises) provided for in the Report would cost £14 million on the same basis.
It was estimated that, using wooden poles, the cost in 1946 would have escalated by 50%
over pre-war costs to £21 million, while if concrete poles were used the estimated cost
would be £28 million.
2 The issue was so heavily oversubscribed that the initial 5s. share was dealing at 7s. on the
day after the issue.
> By 1982 the company had grown to a group of eight companies operating eleven factories
occupying an area of eighty-three acres in Finglas and employing over 1,200 workers.

CHAPTER EIGHT

' From an article by Peter Conroy in REO News, October 1953.


2 REO News, December 1953.
° REO News, November 1954. These were the same gales which had put the pole ship S.S.
Karen in peril in the Skagerrack (see page 93).
* REO News, May 1955.
> REO News, December 1955.

CHAPTER NINE

* James Meenan, The Irish Economy Since 1922, Liverpool University Press 1970.
2 R. O’Connor, ‘Financial Results on Twenty Farms in Mid-Roscommon in 1945-46’,
paper read before the Statistical and Social Inquiry Society of Ireland, 29 October 1948.
* John J. Scully Agriculture in the West of Ireland — A Study of the Low Farm Income
Problem, The Stationery Office Dublin 1955.

298
* Commission on Emigration and other Population Problems 1948-54, Report, (The
Stationery Office, Dublin 1955).
° In April 1948 the engineer in charge reported to the Board that there was four times as
many applications for electricity supply from the dairying areas as from the non-dairying
areas.
° John Healy, Nineteen Acres, Kennys, Galway 1978.
’ Hugh Brody, Inishkillane Change and Decline in the West of Ireland, Allen Lane, The
Penguin Press London 1973.

CHAPTER TEN

' Interview with the writer, July 1980.


? The Irish Press, 5 February 1949.
> It was claimed by traders that the sales of paint in rural areas shot up after the electricity
was switched on as the new light showed up grubbiness which hitherto had escaped notice
in the pervading gloom.
* Sometimes a little judicious blackmail spurred the committee to extra effort. In the
Blackwater area of County Wexford the search for a suitable office premises was proving
fruitless. RAE C. V. Conway rang the District Office in Waterford from the public telephone
in the local post office. In a loud clear voice he told his superiors of the problem and
suggested that he might skip Blackwater for the time being and move on to the next area
on the list. Quite by accident he chose a time when the post office was crowded with
customers and just happened to leave the door of the phone booth ajar during the
conversation. It appeared to be equally coincidental that an hour or so later he was told
that a suitable premises had been located by the committee.
* Stage comedians of the time, particularly in country districts, drew on the scheme for
much of their material: ‘I have a great job with the Ess — digging holes for poles’, or the
old chestnut, ‘Honour the Light Bigade — Oh what a charge they made!’

CHAPTER FOURTEEN

1 A traditional Irish house Mass dating from the period of anti-Catholic ‘Penal Laws’.
? Michael A. Poole, ‘Rural Domestic Electricity Expenditure in the Republic of Ireland’,
Irish Geography Volume VI No. 2, Dublin 1970, The Geographical Society of Ireland.

CHAPTER SIXTEEN

1 Jerome Toner, OP, Rural Ireland, Some of Its Problems, with comments by Sir Shane
Leslie and T. J. Kiernan. Dublin, Clonmore & Reynolds Ltd., London, Burns Oats &
Washbourne Ltd. 1955.
2 Stephen Rynne, Fr John Hayes, Founder of Muintir na Tire, The People of the Land,
Dublin, Clonmore& Reynolds Ltd. 1960.
3 REO News, July 1948.
4 Tt was about twenty-four inches high and had a central motif depicting An Claidheamh
Soluis, the Sword of Light, with two figures representing youth at work, The Sower, and
at play, Games. This beautiful carving was the work of John Haugh who at the time was
a teacher in Newry vocational school. He had been trained as a woodcarver at Glenstal
Abbey and after spending some years carrying out wood carving in Belgium had come to
Newry via Marino vocational school, Dublin.

299
5 Horace Plunkett, Ellice Pilkington and George Russell (AE), The United Irishwomen,
Their Place, Work and Ideals, Maunsel & Co. Ltd. Dublin 1911.
6 Ibid.
7 For a fuller description of the foundation of the Society and its work see The Irish co-
operative Movement, Its History and Development, by Patrick Bolger, Dublin IPA 1977.
Ea Opacit
9 Muriel Gahan, lecture ‘The United Irishwomen’ April 1969.
© Report of Commission on Emigration and Other Population Problems 1954 Footnote to
page 175.
1 Two participants were selected from each county including the North and South Ridings
of Co Tipperary.

CHAPTER SEVENTEEN

1 Mary Purcell, lecturing on ‘The Parish Plan Helps the Woman’ in the course of Ladies’
Day at Muintir na Tire Rural Week 1959.
2 John Healy, Nineteen Acres, Kennys Galway 1978 p. 12.
3 Desmond Roche, ‘Rural Water Supply’, Administration, Institute of Public Administra-
tion Dublin Winter 1960. :
* Paper, ‘Rural Aquafication’ read by Fr Joseph Collins at ica Rural Water Supplies
Conference held at An Grianan, 25 October 1960.
> Desmond Roche, Op. cit.
An estimate by the Department of Local Government in 1959 and quoted by D. Roche
(see footnote 3) indicates that the probable percentage of direct service by local Authority
Schemes would be about 50% leaving half the then unsupplied houses to a greater or lesser
extent dependent on their own initiative for a piped supply, helped of course by grants and
other forms of assistance.
7 Bernard J. Tighe, ‘Rural Water Supply in the Republic of Ireland’, Journal of the
Institution of Water Engineers, August 1966.

CHAPTER EIGHTEEN

" This however was mainly a substitution: the acreage of oats dropped accordingly.
* About 95% of this energy bill was for oil used in heating the growing area. Electricity
(accounting for about 5%) was used for heating and illumination of growing houses,
blowers, circulating pumps, control equipment, lighting etc.

CHAPTER NINETEEN

* Commission on Emigration and Other Population Problems 1948-1954, Report Dublin


Stationery Office 1955, paragraph 439.
* These teams carry out all the initial processing of small industry projects (up to fifty
employees) and also provide the after-care service. Priming grants are also available
through them from the western fund.
* In November 1983 Carol Moffett was selected for one of the ‘People of the Year’ awards
which were presented by an Taoiseach, Dr. Garret Fitzgerald.
* Denis I. F. Lucey and Donald R. Kaldor, Rural Industrialization — The Impact of
Industrialization on Two Rural Communities in Western Ireland, Geoffrey Chapman
London Dublin Melbourne 1969.

300
CHAPTER TWENTY

' These included industrial consumers, registered hotels and guesthouses on commercial
tariff, large intensive farming activities on commercial or industrial tariffs.

CHAPTER TWENTY-ONE

‘ ‘Tt may be at first hard to believe that it was the agricultural wealth of these valleys that
attracted the first farmers who came to Ireland 5,000 years ago.’ From Glencolmcille, a
guide to 5,000 years of history in stone, by Michael Herity, Elo Press Dublin, 2nd ed. 1980.
* James McDyer, Father McDyer of Glencolumbkille — An Autobiography Brandon Book
Publishers Ltd. Dingle Co Kerry 1982.
er Opi cit.

EPILOGUE

Professor Thomas F. Raftery, Head of Department of Agriculture, University College


Cork, quoted in The Irish Times, 16 February 1983.

APPENDIX TWO

' “Low tension’, not requiring erection of transformer.

APPENDIX FIVE

' This system is described in detail in a joint paper read at the 1981 CIRED* conference
in Brighton: ‘A Distribution System for Rural Zones with Low Consumption Rate’ by P.
Messager (France), K. Cronshaw (Great Britain), C. Cunningham (Ireland) and M. Deuse
(Belgium).
2 This assumes an empirical worst loading condition of 180d grammes per metre length
acting vertically downwards at —5°C where dis the diameter of the conductor in millimetres
and takes a maximum tension based on the ‘maximum permanent static tensile stress’
which the conductor can withstand for one year without breaking.
3 The operation sequence of a typical recloser is as follows.
On the occurrence of a short-circuit current exceeding the set figure the switch opens. It
recloses after 14 seconds and if the fault has disappeared, remains closed. If, however, the
fault persists, the switch immediately re-opens. It repeats this cycle twice and if the fault
still persists after the third reclosing, the switch locks out in the open position thereby
disconnecting the line involved pending investigation and repair of the fault.
4 Interview with the writer July 1980.
*Congrés International des Réseaux Electriques de Distribution organised by IEE Conference Service Savoy Place
London WC2R OBL.

APPENDIX SIX

1 The Role of Three-Phase Electricity Supply in Irish Agriculture, Agricultural Unit,


Industrial Development Section, Electricity Supply Board, July 1976.

301
clergy, role of 134,137,143, electrical design 279,281
Index 147,148,157 Electrical Engineers, institute
coal 26,275 of 923)
(T)=Table Collins, Rev. J. 202,206 electricity
community involvement 9,61, annual consumption 18,153,166-
Aberdare Electric Co. 80-1,83 see 65,135,137 7,251,277
also Unidare competitions 177-9 conductors 80-1,151,279,
A.C.E.C (Ireland) Ltd. 82-3,216 configuration 280 281,293
Agricultural Institute 174,205,221 Conroy, Peter 90,92,102-3 consumers of 21-2,28,40,48,58-
agriculture 10,16-17,64,112,152, Conway, Cardinal 186 9,63
153,261 ,277,290 copper wire 80-1,282-3 ‘priority’ cases 242,243,244,
acreage 109-11 Coras Trachtéla 234 245
depressionin 105-14,214,219 Cork 79,94,100,104,110(T),111 earth faults 284-5
andthe EEC 249-50 (T),176,179,193,230,276 line construction 48,66,151,152
importance of Castletownbere 294-6 poles 48,60,66,68,74-6,79,151,
electricity 8,23,72, Drimoleague kitchen 192-3 152
113,115-16,156,160,164, electrificationin 132,134, shipping of 87-104
181,188 ,213-28,240,261 150,264 price rises 32,40,81
see also farm equipment; water water systems 204,210 transformers 48,81-3
systems corrosion 293 single-phase 81,83,176,
Agriculture, Department of 170, Cosgrave, Liam 144 276-8,290-2
173,198 202,238,291 Costello, J.C. 23,38 three-phase 83,176,276-8,279,
Aiken, Frank 148 Crady,Joan 205 280,290-2
aluminium wire 80-1,282-3 creameries 163 Electricity (Supply) Act 1928 20
Ardnacrusha 16,26,42 creosoting 79 Electricity (Supply) (Amendment)
Cronin, James 201-2 Acts:
backsliding 62,64,131-7 Cuffe, Robert C. 58,68 1945 29,35,40,48,155
Baltic Trade 87-94 Curtis, Mrs. M. 178 1958 47 1962 154,242 1976 250-1
Bank of Ireland 227 Czechoslovakia 82 Electricity Supply Board (ESB)
Barrington, Aine 205 18,20-2,32
Barry,C.A. 238 dairying 8,208,221-2 annual reports 151,152,155
bathrooms 174,210 decentralisation 21,50,118 number of connections 22,40,
Belgium 82-3,86 demobilisation 63 48 ,66,97,137,151-5,238-9,
blacksmiths 218-19,229 demonstrations 72,144,158- 242,246 ,250,252
Bonfil,E.M. 178 60,162-4,169,172,196,199 repairs 22,198
bottled gas subsidy 239 Design of Rural Networks 39, Water Advisory Service 203-4,
Bourke, John F. 58,68,288 71,288-9 211
Boyle, Prof. 23 de Valera, Eamon 84,148 see also finance; Rural Electrifi-
breadmaking 177-8,192 Dickinson, E.W. 23 cation Scheme: water systems
Brennan, Paudge 205 Dillon, James 188,190,198 emigration 8,11,113,236,243,
Brennan, P.J. 238 Disabled, Yearof 186 253-4
Britain 26,82,84-5 227,236,259 districts 21,50,56,60 Emigration and Other Population
electrificationin 31,277,288, district engineers 50-2,60 Problems, Commission on 113-
291 domestic appliances 148,152, 16,194-5,229
and Irish agriculture 106-7,213 160-7,173, 174-5 Ennis, Patrick J. 58-9,172
British Electricity Council 226 Donegal 110(T),111(T),184,187, Eme river 97
Brody, Hugh 117 190,215,262 ESB Journal 90
Brophy, MatthewJ. 238 coasters 79,94-104 European Economic Community
Browne, Colm 191,256 electrification in 68,96-103,134- (EEC) 176,220,221,240,249
Browne, R.F. 148 §,150,151,270,276 exhibitions 168-77 see also
Butler, Eleanor 173,190 Glencolmcille 254-9 demonstrations
Byrne, CharlesJ. 238 industries 230-2,236,292
Byrne, J.J. 205 water schemes 198,210-11 farm equipment 152,160,162-4,
Dowling, Patrick J. 25,50,53-4, 169-73
Callan, Nicholas J. 14-15 89, 150,192,205 ,214,218,238,264 farming see agriculture
Canada 80,214-15,265-6,288 Drimoleague kitchen 192-3 farm workshops 8,218-19
Carlow 15,110(T),111(T),136,150, Dublin 76,82,110(T),111(T),179, field design 280
210 185,229 finance (of Rural Electrification
Castletownbere (Cork) 294-6 electrificationin 13,15,53,134, Scheme)
Cavan 68,76,110(T),111(T),150, 150,252,276 backsliding 62,64,131-7
151,184,210,222,292 portof 79,88,94,100,101 budget 1966 244
Childers, Erskine 186 deficits 42,44-7,246-7,249
Churchill, Winston 84 economic depression 16,106-7 depreciation 37,39
Citizenship Award 184-5 Economic Development (Whitaker) estimated costings 19-20,21,
Clare 110(T),111(T),150,187,210, 220,243 24-5 ,28,37-8 81 137-157 ,246,267
236 Edison, T.A. 15 rural account 38,54,60,249
Clarke, Prof. E.J. 226 Education, Department of 178,184 state subsidy 28-9 ,31-47,151,
Cleary, Michael 185 effluent see waste disposal 237,240,242 245,247,251 ,268

302
lossof 40-4,121 water schemes 196,205-6 Mangan, Dermot 78
see also fixed charge; planned Irish Creamery Milk Suppliers’ Matthews, R.B. 23
post-development; special Association 290-1 Mayo 96,99,110(T),111(T),160,
service charge; unit charge Irish Farmers’ Association 290 164,230
Finance, Department of 34,80, Trish Farmers’ Journal 173,174, electrificationin 134,137,139,
238,240,243 183,184,205
,206 147,150,155 157,236,251 ,276
Finegan, Tom 205 Irish Independent 18,199 Meath 110(T),111(T),163,164,
Finland 68,77-9,87-8,90 Irish manufacture, use of 45,82, 199,230
fisheries 29 86,176,200,215-16 electrificationin 132,134,137,
fish processing 258-60,292 Trish Times 13,18,219,222 141,144-5,151
Fitzpatrick, Mona 178 mechanical design 281-2
fixed charge 24,29,32,34,40,137- Joye, Annie 54-6 Meenan, Prof. James 116
41,267,271,273,274-5 Meghen,P.J. 205
Fogarty, MichaelO. 205 Kelly, James J. 246 milking machines 83,152,162,
Food and Agricultural Organisation Kerry 104,110(T),111(T),164,176, 172,208,222
(FAO) 192 182,193,210,230,236 Minahan, Nan 205
Foras Talintais, An 173,176,216, electrificationin 150,153, ‘minimum return’ 35-6,39-40,
220,226,291 155,157-8,163 ,251,276 47,271-3
Fordige see Macra na Tuaithe Kildare 76,110(T),111(T),136, Moffett Engineering 233-5
150 Monaghan 110(T),111(T),182,
Gaeltacht areas 40,172,230-3 Kilkenny 83,110(T),111(T),135, 184,210,229,233-4
Gaeltachta, Roinnna 238 150 electrificationin 68,117,
Gaeltarra Eireann 230-3,258 kitchens 190-3 150,151
Gahan, Dr. Muriel 188,205 Kottgen, Jean 83 Montgomery, Harold 58,288
Gallagher, Jack 259 Moracrete Ltd. 76
Galway 96,100,104,110(T), Land Commission 141 Morgan, Dr.J.V. 226
111(T),164,207,230,236 Lands, Department of 86 Moniarty, Patrick 238
electrificationin 140,148,150, Laois 110(T),111(T),150,185 Morris, Jarlath 259-60
156-7 Larkin, James Jr. 29-30 Morrissey, Daniel 48
GEC Ireland Ltd. 179 Leitrim 110(T),111(T),150,151, Muintir na Tire 53,113,143,180-
Geological Survey Office 199 178 2,183,190,204-5 ,229
Germany 31 Lemass, Seén 25-9,31,34,39,46, Munster Agricultural Show 176
Gleeson, Mrs. Kathleen 192 47,148
Glencolmcille (Donegal) 254-9 lighting 15,152,156-7 National By-Products Ltd 235-6
Government Information Bureau fluorescent 83,146 National Farmers’
Limerick 79,89,92-3,94,100, Association 46,
grain processing 214-15 110(T),111(T),176,179 140,183 ,220-5
grain storage 224-5 electrificationin 150,153,276 National Ploughing Association
grants (water systems) 170,198-9, water schemes 201-3,210 177
201-8 load assumptions 278 National Youth Foundation 184
Grimmet, H.W. 23 local organisations 10,180-93,221 Northern Ireland 96,108
‘ground rent’ see fixed charge Local Government, Department Norton, Bill 257
group water schemes 170,201-8 of 170,198 ,202,206,207 ,238 Norway 76-7
Gruno Shipping Co. 94,98-9 Longford 110(T),111(T),144,150,
210,227 O’Brien, Niall A. 238
hammer mills 215-16,218,229-30 Louth 110(T),111(T),116-17,146, O’Connell, Phyllis 205
Hannon, Oscar 146 148,150,265,276 O’Connor, Michael V. 73
Hayes, Canon John 53,180,181, Lucey, D.L.F. and Kaldor, O’Connor,R. 109,112
183,229 D.R. 236 O’Donoghue, Cornelius (Neil)
Healy, John 116,195 54,77,89-90
Higgins, J.M.F. 24 McBride, Joseph 210 O'Donoghue, JohnB. 238
Holland 82,86,94,227 McDonald, Patrick 57,79,82-3, O’Driscoll,Noel 294-6
Home Improvement Award 184 84,86 Offaly 110(T),111(T),150,205,
horticulture 72,173-4,220-1,226-8 McDyer,
Dr. James 254-9 234
Hughes, Jim 218 McEntee, Se4n 23,26 O’Kelly, Sean T. 183
hydro-electric schemes 16,29,97 MacEoin, GeneralSean 144 O’Neill, C. 205
see also Shannon Scheme McFaul,J.F.O’D. 24 O’Riordan, Dermot 238
McGilligan, Patrick 18,29,144
Industrial Development Authority McGovern, Phil 134 Parnell,C.S. 15
(IDA) 233-6 McLaughlin, Thomas A. 16,23, Pearse,P.H. 148
industry, rural 229-36,250, 24-5 ,50,54,264 Pickles, J.S. 23
258-60,262 McManus, Alphonsus J. 25,264 Pigeon House 26
Industry and Commerce, Depart- MacranaFeirme 113,173,180, pigs 176,222
ment of 26,28,99,238 182-6,190,193,205 Pigs and Bacon Commission 291
Irish Countrywomen’s Association Macra na Tuaithe 113,180,182-6, Pilkington, Ellice 187
(ICA) 113,140,144,173,180,183, 193 piping 201
186-193 Mag Fhloinn, Se4n 238 planned post-development schemes
AnGnandn 184,191-2 Manahan, Noel 205 40,45-7,153,154

303
interdepartmental materials 48,71,74-86 ‘system improvements’ 40,252,
committee 238-42 organisation of 48-66 286-7
Plunkett, Sir Horace C. 186,187-8 opposition to 18,29,115-16,
Poole, Michael A. 167 120-1,126 see also backsliding Tighe, Bernard J. 207
‘priority cases’ 242-6 pre-existing system 276 Tipperary 86,110(T),111(T),163,
private electrical development progress of 40,151-2,252 185 204,229,235
15,20,96 public relations 21-3,58-9,158- electrificationin 132,143,150,
Prospect 73,207 64,214 157,181 ,183,264
protection of lines 284-6 selection system 35-6,40,58, Tobin, P.J. 215-16,229-30
118,150-1 Tomlin, D. 205
Radio Telefis Eireann (RTE) 173 see also finance Toner, Jerome 180,181
radio transmitters 61 rural organisation engineers transport 57,60 see also shipping
Rantala, Viljo 78-9 (ROEs) 60 Transport and Power, Department
Ras Tailteann 163-4 Rural Water Supplies, Campaign of 238,244,248,250
REO News 67-73,90,143-4,146, for 205-6 Tuite, Paddy 174
148,149,157,198,199,210,234 Russell, George (AE) 187 tweedindustry 258
Revenue Commissioners 99 Rynne, Stephen 180
Roche, Desmond 203,206,238 Udards na Gaeltachta 232,258
Roe, W.F. 13,53,54,56-7,62,63, St McCartan’s Diocesan College Unidare 200,201,205 see also
77,84,150,183 ,289 182 Aberdare Electric Co.
and Muintir na Tire 53,180, St Martha’s College, Navan 174 unit charge 24,138,267,274-5
181-2 St Patrick’s College, Maynooth 14 United Irishwomen 187
and REO News 67-8 sales organisation 21,22,64,72, United States of America 82,
Ronan, Louisand John 235 162,163 see also demonstrations 200,265 ,288
Roscommon 109,110(T),111(T), Salovaara, OnniJ. 77-8 Kellog Foundation 184,191
1) sanitation 194 Marshall Aid 220,242
electrificationin 146,147,150, Scully, DrJ.J. 112 University College, Dublin 220-
151,270 Shannon Board of Control 53 1,226,228,291
Royal Dublin Society Shannon Scheme 15,16-22,20, urbanconsumers 22,31
(RDS) 168-76,183,190,196,206 23,29,50,54,57,149,228 262,265
tural, definition of 57,269-70 Sheedy, Larry 205 van Siemens, Werner 14
rural areas 21,26,56 Shiel, Michael 238 voltage drop 279-80
rural areaclerks 64-5 shipping 60,79,87-104 voluntary organisations 10,
rural areacrews 50,61, Siemens Schuckert 16,57 180-93,221
65-6,70,127-30,149 153 Simms, Archbishop 186
rural area engineers (RAEs) Sligo 100,110(T),111(T),178,182, war surplus 61,84
60-3 ,64,74,149,157,158 185,236 waste disposal 176,224
rural area office 52,56,61 electrificationin 97,132,150, Water Advisory Service 203-4,211
Mareas 70 151,157 Waterford 76,83,110(T),111(T),
rural area organisers Smith, B. 205 230
(RAOs) 63-4,131-5,136, Smyllie, R.M. 13 electrificationin 143,148,
159-60,162,163 social changes 8-9 ,29-30,115, 150,276
rural electrification, planning 117,186 ,266-7 water heating 171,208-12,227
for 21,23,24-6,29,31 special service charge 19,39-40, water systems 170,188,192,
White Paperon 28,31,34-5, 45 ,46,47,238,240,244-6,248 194-212
44,54 ,152,154,237,250,264-8, Spring Show (RDS) 168-76, grants 170,198-9 201-8
271,273 183,190,196,206 pumps 152,162,170,194,195-
rural electrification offices staff 50,56,57,61,62,65,67-70,151
(REOs) 52,56,57-60,104,177 state-sponsored bodies 9,18,20 Wavin Pipes Ltd 201
development sections 58,71-2 state subsidy 28,29,31-47,151, wayleaves 63,64
and Gaeltacht 230-3 237,240,242,245,247,251,268 ACs 110(T),111(T),134,
andfarmers 213-16,221,228 lossof 40-4,121 15
and local organisations 180-93 post-development 238-42 Wexford 110(T),111(T),151,158,
materials 57,71 steel-core aluminium (SCA) 282- 164,187,191,216,265
technical sections 58,71 4,288,294 Whitaker, T.K. 220
and water systems 194-212 storm damage 287 Wicklow 110(T),111(T),150,
Rural Electrification Scheme 9,21, Swan,J.W. 15 202-3
23,24,40,44-5 ,48-51,52,228 Sweden 76,82,94,138,234,266 wiring contractors 157-8
line construction 48,66,151,152 switching-on ceremonies 13,142- Wolahan,
Jim 116-17
local committees 119-23,127, 9,158,256 World WarII 26,35,74,96,108,
135 Switzerland 82 116,168,214

304
COO
ity

seen

You might also like