LINGUISTIC CLASS-INDICATORS IN PRESENT-DAY ENGLISH

Author(s): Alan S. C. Ross

Source: Neuphilologische Mitteilungen , 1954, Vol. 55, No. 1 (1954), pp. 20-56
Published by: Modern Language Society

Stable URL: https://www.jstor.org/stable/43341716

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

is collaborating with JSTOR to digitize, preserve and extend access to Neuphilologische

Mitteilungen

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 20 Alan S . C. Ross

able to an extent which mak

beyond the limits of their orig
Since the recent war, two d
which should have a bearing on
first place, attention is being d
purely linguistic and literary
Recent articles in Ylioppilasleh
first hand-books on style ha
Secondly, German has now b
as the first foreign language (Sw
official languages), and a long
and cultural influence has thus
It will be interesting to observ
Norman Denison

LINGUISTIC CLASS-INDICATORS IN
PRESENT -DAY ENGLISH2

To-day, in 1953, the English class-system i

tripartite - there exist an upper, a middle and
It is solely by its language that the upper class is
off from the others. In times past (e. g. in the V
Edwardian periods) this was not the case. But, to-d

1 See e.g. 'Tyylitaito muotivitsausko?' by Matti Hako in

Issue No. 36, 1952, p. 8 - apparently inspired by Martti Rapóla,
'Äidinkielen tyylitaidon opetuksesta', Suomalainen Suomi 7, 1951,
p. 310 - and the unsigned article in Ylioppilaslehti 40, 1952, p. 8.
2 Phonetic notation . Length is indicated by a suffixed colon
([ka:d] card ), stress by a prefixed (closing) inverted comma
(increase is ['inkrijs] as a noun, [in'krijs] as a verb). In rare cases,
a difference between primary and secondary stress is indicated
by two, as against one, (closing) inverted commas. Symbols : -
[ rj] as singer, [0] as thought , [Ö] as breather, [š] as fisher, [ž] as
leisure , [j] as y acht, [ij] as beat , [e] as bet, [ei] as paid, [a?] as cat,
ļeaļ as the re, [o] as hot, [d:] as loid , [ou] as bone , |uwļ as boot, [a]
as cut, |a:] as turn , jai J as ride, [au] as found , (ai] as boil, [a] as
china , - and the rest self-evident.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 21

of the upper class is, for instance, not necessarily be

ucated, cleaner or richer than someone not of this clas
in general, is he likely to play a greater part in public af
supported by other trades or professions,1 or engage in
pursuits or pastimes than his fellow of another class. Th
it is true, still a few minor points of life which may s
demarcate the upper class,2 but they are only minor on

1 It may however be doubted how far the Navy and th

lomatic Service will in practice (in contradistinction to th
»democratised», even if there should be a succession of Labour
Governments; foreigners seem to expect English diplomats to be
of the upper class.
2 In this article I use the terms upper class (abbreviated : U),
correct , proper , legitimate , appropriate (sometimes also possible)
and similar expressions (including some containing the word should)
to designate usages of the upper class; their antonyms (non-U,
incorrect , not proper , not legitimate , etc.) to designate usages which
are not upper class. These terms are, of course, used factually and
not in reprobation (indeed I may at this juncture emphasise a point
which is doubtless obvious, namely that this whole article is purely
factual). Normal means common both to U and non-U. I often use
expressions such as U -speaker to denote a member of the upper
class and, also, gentleman , pl. gentlemen (for brevity, in respect of
either sex - pl. gentlefolk is no longer U). Class-distinction is
very dear to the heart of the upper class and talk about it is hedged
with taboo. Hence, as in sexual matters, a large number of circum-
locutions is used. Forty years ago, as I understand, U-speakers
made use of lady and gentleman without self-consciousness; the
antonym of gentleman was often cad or bounder . To-day, save by
older people, these terms can hardly be used to indicate class-
distinction, for they sound either pedantic or facetious (you cad,
Sir!). Lady and gentleman have, of course, senses quite uncon-
nected with class-distinction, but, to-day, the use of these words
in the sense 'man' and 'woman' between U-speakers has almost
entirely vanished save when prefixed with old (There's an old
lady to see you is different from There's an old woman to see
you , for the former implies that the person is U, the latter that she
is very non-U). She's a nice lady is non-U, He's a nice gentleman
even more so (man, woman or girl being the U-use here) - land-
ladies sometimes call negroes dark gentlemen or, occasionally,

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 22 Alan S. C. Ross

games of real tennis and pique

one's cards2 engraved (not prin
and, in some cases, a dislike o

gentlemen of colour. But U-speak

to their servants as do servants
U-speakers sometimes express
as X - is of good family or, alternatively, X - is rather
ill-bred, but they would perhaps be more likely to make use of
circumlocutions such as I should have thought X - was per -
fectly alright (or even : quite O.K.); the antonymical form might
be as vague as X - is obviously awful. In Oxford, in the late
twenties, the term not respectable was sometimes used in this sense
and intellectual Oxford of the period liked to use the German
expression rein / unrein ['an'rain] - this can of course only be
used to U-speakers moderately acquainted with German and these
are rare (a state of affairs in contradistinction to that obtaining
some seventy years ago). U-speakers might stigmatise expressions
or habits as non-U by calling them low (more rarely, vulgar ); at least
one Public School (Bradfield) uses pleb in this sense. The phrase not
done , once perhaps a U-phrase, seems to be dying out , save in schools.
At the beginning of the century many slang phrases were used to
designate non-gentlemen e.g. not off the top shelf , not out of the top
drawer , not one of us, not quite the clean potato, L. M. C. [ = Lower
Middle Class], a bit hairy in the heel , showing the cloven hoof , show-
ing [or similar introduction] the mark of the beast. At this distance
of time it is hard to decide which, if any, of these expressions were
U (he's a bad bred 'un and risen from the ranks, I suppose? certainly
were) and which belonged rather to the extensive class of non-
U-speakers trying to become U. As a boy I heard not quite agent,
not (quite) the gentleman used by non-U-speakers; to-day, well-
connected is definitely non-U. One of our great border families is
probably idiosyncratic. At one time common was used by U-speakers,
though it is no longer (save of horses); the abbreviated very comm
was apparently used by female U-speakers; he's a commoner is an arch
U-use (implying that the person is very far from having a seat
in the House of Lords).
1 But solo whist (or solo as its devotees call it) is non-U, though
much »lower» games (e.g. pontoon, nap and even slippery sam)
are not necessarily so. Whist used to be a U-game but is, to-day,
almost entirely confined to whist-drives which are non-U (they
stand up to deal, my dear I).
2 The normal U-word is card (though this is ambiguous

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 23

inventions such as the telephone, the cinema and the w

are still perhaps marks of the upper class.1 Again, whe
gentlemen often become amorous or maudlin or v
public, but they never become truculent.
In the present article I am concerned with the lin
demarcation of the upper class. This subject has been b
investigated, though it is much discussed, in an uns
manner, by members of the upper class. The late P
H. C. Wyld wrote a short article on the subject2. He w
equipped for the task, for he was both a gentleman and
logist. To-day, his views are perhaps a little old-fashio
instance, the dictum »No gentleman goes on a bus», attr
him, is one which most gentlemen have to neglect.
Both the written and the spoken language of the upp
serve to demarcate it, but the former to only a very s
tent. A piece of mathematics or a novel written by a
of the upper class is not likely to differ in any way f
written by a member of another class, except in so fa
novel contains conversation. In writing, it is, in fa
modes of address, postal addresses and habits of begin
ending letters that serve to demarcate the class.
Before proceeding to the detail of the present study

with ( playing)-card ). Carte de visite was apparently U

to-day seem unbearably old-fashioned. Calling-card and
card are non-U; the latter term is, in any case, an unfortun
because of the non-U slang phrase He's left his visiting-
dog) - foreigners would do well to beware of »idiomatic» se
such as The Picts left their visiting-card in the Pentla
(said, in a public lecture, meaning that the name Pict is
in the first element of Pentland ). Earlier, slang usages exis
to shoot a pasteboard 'to leave a card', but I do not know
they were U.
1 Certainly many U-speakers hunt - but hunting ha
long been something that the nouveau riche knows he shou
order to be U; many farmers hunt too. So, to-day, hunt
ipso facto a class-indicator.
2 The best English (S. P. E. Tract, No. XXXIX); cf., als
Chapman, Oxford English (S. P. E. Tract, No. XXXVII).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 24 Alan S. C. Ross

emphasise that I am here con

serve to demarcate the upper c
worse nor Dunkirk9 s fall (said
Buckinghamshire town by a
tainly not U, but sentences of
scope of this article; not using
speaker is U. In fact, in seekin
demarcation relevant to this st
the one hand, gentlemen and, o
not gentlemen, might at first
appear, as such. Thus, habits of
ses find no place here. The sam
is only where a definite diale
Standard - as, for instance, the use of the phoneme [a]
instead of [ae] (e. g. in cat) in Northern Standard, that it will
require notice. I may also note here that the U-demarcation is
of two types: - (1) a certain U-feature has a different, non-U
counterpart as non-U wealthy / U rich ; (2) a certain feature is
confined to U-speech and it lias a counterpart which is not
confined to non-U speech e. g. the pronunciations of girl as
(? [gjrfj)» [g®l], [ge9U are U, but many (perhaps most male)
U-speakers, like all non-U-speakers, use the pronunciation

I. The Written Language

The following points may be considered: -

(1) Names on envelopes, etc.
(2) Beginnings of letters.
(3) Names on cards.
(4) Postal addresses on envelopes, etc., at the heads of
letters, and on cards.
(5) Letter-endings.
Of these points the first three are mutually linked an
the second - beginnings of letters - is linked with the
spoken language; for, in general, a person known to the writer is

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 25

written to and spoken to in the same mode of address.

therefore be convenient to treat all modes of address to
though this means taking the spoken modes out of pla
Modes of address , particularly those used for the no
have always been a bugbear to the non-U. It is, for i
non-U to speak of an earl as The Earl of P - ; he sh
spoken of and to as Lord P - and also so addressed a
beginning of a letter if an introduction between him a
speaker / writer has been effected1. If the acquaint
close, P - should be used instead of Lord P - . Letters
to baronets and knights (known) should begin Dear Sir A -
X - 2, if the acquaintance is slight, Dear Sir A - , if it is
not slight. In speaking to one, only Sir A - is possible. In
speaking of one, Sir A - should not be used unless the
acquaintance is fairly close, Sir A - X - or X - being
correct. If the acquaintance is slight or non-existent, the use
of Sir A - in speaking of a baronet or knight is non-U and
»snobbish3» as attempting to raise the social tone of the
1 Where necessary 1 distinguish cases in which introduction has
or has not been effected by putting known or unknown in round
brackets ( ).
2 A - , B - , C - , etc. are christian names (the initials being
written A., B.t C., etc.); X - is a surname.
8 »Snobs» are of two kinds; true snobs (Thackeray's kind) and
inverted snobs . Both kinds respect a person the more the better
bred he is. True snobs indicate this in their behaviour to, and
conversation to and about persons of good family, though they do
not usually admit this. In their conversation about (but not in
their behaviour or conversation to) such persons, inverted snobs
indicate that they respect a person the less the better bred he is.
One would expect to find a third category : those who really do
respect a person the less the better bred he is, and indicate it. But
this third category does not appear to exist. Nearly all English
people are snobs of one of the two kinds (in this respect England
differs from Finland and Iceland and resembles Spain and pre-War
Hungary). And, just as it is impossible to find someone exactly
half male and half female, so it is impossible to find an Englishman
in whom true and inverted snobbery exactly balance.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
26 Alan S . C. Ross

speaker. Letters to ambassado

Excellency and the envelope s
P - Ambassador . In speech,
addressed as Commander , a Lie
In concluding this section it
letters to noblemen of very hi
the etiquette-books1 need no
Thus an (unknown) Duke add
necessarily conclude that his
might be a left-wing gentlem
On envelopes, gentlemen put E
who are, or who might wish
whether in fact armigerous or
oneself e. g. neither on a car
stamped-and-addressed envelo
has merely A - B. X - or A. B. X. - without prefix).
Knowledge of at least one initial of the recipient's name is,
of course, a prerequisite for addressing him with Esq. If the
writer has not this minimum knowledge (and cannot, or is too
lazy to obtain it) he will be in a quandary. In these circum-
stances I myself use the Greek letter O (as O . Smith , Esq.) but
this is probably idiosyncratic. But to address someone as
Smith,» Esq. is not so much non-U as definitely rude.2 Gentlemen
usually address non-U males as Mr.; in internal circulation
(e. g. in Government offices) gentlemen may address each other
in this way. School-boys at their preparatory school (and
younger boys) should be addressed as Master; at their public
school, merely as A. B. X -, (without prefix or suffix). The
non-U usually adress all adult males as Mr., but tradespeople
have copied the use of Esq. from their customers. Those gentle-
men who are inverted snobs dislike Esq., but, since they know

1 It is of course very non-U actually to consult these.

1 I may note here that many U-speakers omit the Esq . on
cheques.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 27

that to address someone as Mr. is non-U, they avoid th

and address all adult males without prefix or suffix (l
correct mode of address for public-school boys). Intell
of any class, often begin letters, even where the acquai
slight, with Dear A - X - . The correct modes of ad
for commoner-females offer fewer difficulties. On en
wives should be addressed as Mrs. A - B. X - , or Mrs.
A. B. X - , where the husband is A - B. X - . Widows
revert to their own initials. Two (or more sisters) may be
addressed collectively as The Misses X - ; Miss X - cor-
rectly designates the elder (or eldest) sister, Miss A - X -,
or Miss A. X - another sister.

Postal addresses . It is non-U to place the name of a house in

inverted commas (as "Fairmeads") or to write the number in
full, either without or (especially) with, inverted commas (as
Two , - worse "Two", - St. Patrick's Avenue). The names
of many houses are themselves non-U; the ideal U-address
is P - Q - , R - where P - is a placename, Q - a describer
and R - the name (or abbreviation) of a county as Shinwell
Hall , Salop.1 But, to-day, few gentlemen can maintain this
standard and they often live in houses with non-U names
such as Fairmeads or El Nido.

Letter-endings. The U rules for ending letters are very strict;

failure to observe them usually implies non-U-ness, sometimes
only youth. In general, the endings of letters are conditioned
by their beginnings. Thus a beginning (Dear) Sir 2 requires the
ending Yours faithfully , unless the writer hopes to meet the
1 Here I may note a curious indicator. In speaking, it is, in
general, non-U to use the whole name of such a house as in I'm
going to Shinwell Hall (the U-sentence would be I'm going io Shin-
well) - this obtains whether the house belongs to the speaker
(or his relatives) or not.
2 Whether the writer is U or not, this is the normal beginning of
all business letters to unknowns; the variant Sir is correctly used
to government officials, Sire (or Your Majesty) to kings; My Dear
Sir is felt as American.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
28 Alan S . C. Ross

recipient when Yours very tr

who begin letters with Dear Mr. X - sign them Yours
sincerely or Yours very sincerely ; perversely, the latter ending is
less cordial than the former. People who know each other really
well will begin Dear A - or Dear X - (males only) and
sign Yours ever . The ending Yours is often used even by gentle-
men if they are in doubt as to which ending is appropriate. Of
rarer endings Yours , etc . (used in writing letters to newspapers)
and endings such as I am Your Majesty's most humble and
obedient servant (in formal letters to kings) may be mentioned.
The name after the letter-ending offers little scope for
comment. Letters are perhaps most usually signed in such forms
as A ~ X -, A B. X - , A. B - X - (the choice
between the two last depending upon which christian
the writer is normally called by). If the writer is unk
(or not well-known) to the recipient, the latter cannot
whether the former is plain Mr. (if male), Miss , Mrs
something else (if female); it is therefore usual for the wri
inform the recipient if he is other than plain Mr. (if m
other than Miss (if female). In handwritten letters, a usua
of doing this is to sign as, for instance, ( Professor ) A
X - ; in typewritten letters ( Professor A - B. X - ) m
typed below the handwritten signature A - B. X - . I h
seen long titles (e. g. Dowager Countess of) appended as
footnotes to the signature. In concluding this section I may
mention that people sometimes sign themselves (or enter
their names in lists, etc.) with the surname only; this usage
is very non-U, the reason for its non-U-ness lying in the fact
that the correct signature of peers is of this form (e. g. the
Earl of P - signs himself just P - -).1

1 The correct form of postcards differs slightly from that of

letters, for both the beginning ( Dear A - , etc.) and the ending
( Yours sincerely , etc.) are omitted. Some U-speakers feel it wrong
to sign a postcard to a friend by anything save the bare initials
(A. or A. B . X .).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 29

No discussion of English modes of address could be c

without a detailed commentary on R. W. Chapman's e
Names , designations & appellations , published in 194
author states (p. 231) that the work is »an attempt to
the modern use, in good society in this country, of
names and designations, spoken or written, in the
third person2». Chapman does not specifically deal wi
usages but, since his enumeration is intended as ex
it may be assumed that, essentially, usages diverge
those given by him are non-U, except in so far as I d
them below.

It remains, therefore, to comment on certain points men-

tioned by Chapman where the usage of 1953 differs from that
of a rather old-fashioned person writing some years earlier.
1 arrange the commentary by the pages of his book, either
citing passages therefrom in inverted commas, or (where this
would be too lengthy) indicating in square brackets [»»] the
point under discussion.
p. 233 , note 1. [»I am told that some children call their mo-
thers by their Christian names»]. In the thirties it was indeed
the practice for many U (and doubtless some non-U) children
to call their mothers (and also their fathers) by the Christian
name. In very recent years there seems, however, to have been
a reversion to earlier types of nomenclature; at all events, in
1943, a young U girl said to me »They must be fairly aged for
they belong to the period in which parents allowed their child-
ren to call them by their Christian names».
p. 233 . [»' Pater* and * Mater * are obsolescent, perhaps obso-
lete»]. These designations were apparently once perfectly U;
so also, I understand, were The Mother and The Governor (in
the third person); the first three are to be found in the mouths

1 S. P. E. Tract No. XLVII.

2 By the »second person» he means speaking to a person, by the
»third person», speaking of one.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
30 Alan S. C. Ross

of U-speakers in Mrs. Henr

at a later date (c. 1910?), Pat
non-U-speakers and so becam
Chapman says, they appear to
p. 234. [»The problem of a fa
solved in various ways accor
Often the Christian names a
method is for the speaker to c
of the name he uses for his
mother Mummy , he might c
not know whether the curious
U, Scots or neither.
p. 235 . [»'My (sainted) Au
pression must surely be qui
kers (of either sex) often use
almost exactly the same force
p. 235 ff. [»The family : thir
use of your before a noun of r
e.g. your mother , your fathe
is difficult to imagine circum
or your sister would be used
Your Aunt Jane , preferred b
old-fashioned, Aunt Jane bein
use of this your rather more
as in your auntie [jar'aenti]

1 Mrs. Henry Wood (undoubt

dramatic novel East Lynne) w
Johnny Ludlow is a series of s
she owned, The Argosy , from 1
are to the six-volume edition of 1908). The stories, which are
narrated by Johnny Ludlow, are written in a colloquial style
and contain many conversations. These portray the language of
the squires of nearly a century ago, which is, essentially, the
parent of the U-English of to-day. The differences between these
two kinds of U-English are numerous and Johnny Ludlow is a
valuable fundgrube for the student of the subject.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 31

tainly non-U). U-speakers would certainly not make

sentence such as (to a child) »If you dont stop it, I'll as
father to give you a good smack-bottom when he co
pp. 237-8. [»Spouse : third person»]. The mode in w
speaker refers to his spouse is markedly distinct as be
and non-U speakers. A U-speaker, naming his wife to a
normally says My wife (or uses her christian name); to
a very non-U person he says Mrs . X. - 1. Chapman says
(p. 237) of a U-speaker referring to the hearer's wife [»'Your
wife' may be over-familiar if I do not know Jones [sc. the
hearer] very well»]. He advocates the use, then, of Mrs. Jones.
Actually, I think that, of recent years, there has been a con-
siderable increase in the use of Your wife , Your husband by
U-speakers, even in cases where the acquaintance is of the
slightest. Rank and seniority appear to have little to do with
conditioning the use of the more formal Mrs. X - ; still,
there must be very few people who could speak to a king of
Your wife. Non-U speakers do not in general make use of
my I your wife I husband, preferring Mr. / Mrs. X -, Some-
times remarkable phraseologies appear in non-U use : I have
heard A -'s daddy used by a speaker with reference to the
hearer's husband (the child A - not being present).
p. 238. [»'What does Weston think of the weather?', Mr.
Knightley asked Mrs. Weston 2. But I should be chary of follow-
ing this precedent».] I agree with Chapman. There is, however,
rather a similar case, not mentioned by Chapman (doubtless
because it is a very minor one) where surnames may be used.
School-boys and young men normally refer to each other by
their surnames, so parents of a boy, talking to one of his

1 There are still a few U households so old-fashioned as to use

The Master and The Mistress to servants (if the servants are them-
selves old-fashioned enough to adopt the usage they may drop the
article in replying e.g. Master / Mistress says). Your mistress was
certainly once U, but is now dead.
2 [»In Emma».]

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 32 Alan S. C. Ross

acquaintances, often use the

they do not know his Christi
ance may call the son by his
until a boy gets older (c. 16
deliberately ascertain his frie
be able to refer to them corre
the late twenties the use of the surname in these circumstances
was a known gaucherie and must therefore have been fairly
usual.1

p. 239. [»'Mr. William' is hardly said except by or to a

servant».] Master William , too, is often used of adults in these
circumstances, either because the servant has been with the
family a long time, or, sometimes, because the usage has be-
come part of the family tradition so that each fresh servant
adopts it 2.
p. 240. [»'Sir' is, of course, very often used between intim-
ates with a slightly jocular or affectionate intention; one may
say 'Good morning, Sir' to almost any intimate. 'My dear Sir,
I am very glad to see you'. But 'My dear Sir' usually conveys
a mild remonstrance».] These usages are, I think, obsolescent
among U-speakers and young U-speakers are inclined to dislike
them very much. In my experience people who use them are
either non-U (very often, commercial travellers) or, if U, are
elderly academics.
p. 241. [»'My dear Lady', rather than ' My dear Madam* ,
corresponds to the ' My dear Sir ' of familiar speech».] My dear

1 In connection with surnames, I may mention a habit not

noted by Chapman viz. the abbreviation of the surnames of close
friends. It was apparently U and was certainly thriving in the
nineties; at a much earlier period it appears in Johnny Ludlow
where the young Todhetley is often called Tod. The custom is now
obsolescent, save perhaps in the case of hyphenated surnames
(X-Y may be called X) and between close women-friends (e.g.
a Miss Robinson might be called Robbie ).
2 Servants frequently distinguish the wives of the sons of the
family as Mrs. William , Mrs. Thomas , etc.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 33

Lady was certainly once a U-usage (also, possibly, De

but it is now completely dating. It is however still po
imagine circumstances where My dear Madam might b
imate e.g. as between a solicitor and a female client
p. 241 . [»The use of 'Sir* by young men to their se
general is not easily defined, and the practice varies».
certainly true; my own use is to reserve Sir for men
age and / or great distinction. The War of 1939-45
predecessor of 1914-8, has brought about an enor
crease in the use of Sir because of Service rules. Chapm
[»Young women . . . are not expected to say sin] -
many do by reason of their having been in one of the
Services.

p. 241 . »[But is there any alternative [sc. to Miss] if one is

addressing a telephone operator or a barmaid?».] Yes, there is :
silence, perhaps the most favourite of all U-usages to-day.
Indeed it is remarkable how easy it is (save when engaged in
activities such as bridge or poker) to avoid the use of any
appellation at all.1 This has become increasingly the practice of
shyer gentlemen. The use of Miss in the circumstances men-
tioned by Chapman (and particularly to waitresses) is definitely
non-U.

p. 242 . [»'Master Smith* . . . imputes impudence or sharp

practice».] Obsolescent (or non-U?).
p. 243 . [»Christian names».] On this matter, as might be
expected, Chapman has a point of view out of date even by the
early thirties. I can only just remember the time, in the very
early twenties, when a typical boy-and-girl conversation might
1 This U-habit of silence has had a curious corollary. Most
nations say something when drinking (as Skdl! in Swedish or
Egészségére! in Hungarian) but, until 1939, English U-speakers
normally said nothing. Since then, however, the Service habit of
saying something has become almost universal and most U-speakers
therefore feel it churlish to say nothing; repressing a shudder,
they probably say Cheers! (though hardly God bless ! which,
though also frequent in the Services, seems non-U).
3 - Mittellungen

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 34 Alan S . C. Ross

have run : - »He. May I call you

If you like. He. Er - what is
that time the use of Christian n
continually increasing. In the th
for a member of a partie car
unknown to the other three to
name alone (or, often, just as John
prefix). In the War the use of C
further; in Government offices
man at the head of a large secti
their Christian names, while
pp. 245-6. [»since most people
names».] Chapman's statemen
gentlemen.
p. 248 . [»Use of surnames by women»]. In the third person,
it is now very usual for women to use the surname only of men
(e.g. of their husbands' friends); for men, or women, to use the
surnames only of women in this way is less common, though in
some circles (e.g. university ones) it is quite accepted. In the
second person, the use of the bare surname without Christian
name or prefix is rarer still. For a woman so to call a man is still
either foreign, bohemian, or intellectual-left. In general, women
call other women by the bare surname only in institutions for
women (e.g. in girl's schools, women's colleges, hospitals, and,
no doubt, in womens' prisons).
p. 250. [»Dukes : third person»]. I may add that dukes, if fairly
well-known to the speaker may appropriately be referred to by
Christian name and title e.g. George Birmingham , meaning
George , etc., Duke of Birmingham.
p. 251. [»A facetious use»]. His Lordship, in facetious use, is
definitely non-U and, often, inverted snob. There is a somewhat
similar non-U expression : young master (as in Young master's
making himself quite at home!) used of a young man considered
»la-di-da» 1.

1 For this word see p. 44, below.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 35

pp. 251 , 255. [»Abbreviations»]. Honourable and R

are abbreviated either as Honble., Revi or as Hon., R
usages are quite U, though the former is the more t>ld-f
p. 262. [»Naming of famous women»]. Garbo is ce
possible, but The Garbo (not mentioned by Chapman
well be more usual. The article is sometimes prefix
iously e.g. The Smith of a woman-professor, though I
sure how far this usage is U.
p. 265. [»Some people say 'Miss Austen'».] In my expe
to say Miss Austen instead of Jane Austen is either pr
pseudo-intellectual l.

II. The Spoken Language2

1) Intonation. There are certainly types of intonation which

are recognisably non-U, but they cannot be dealt with here
for they require specialised phonetic and / or acoustic treatment
2) Stress. In a few cases, a difference of stress will serve to
demarcate a pronunciation as between U and fton-U. Thus
yésterdáy (level stress) is non-U as against U yésterday ; or
again, U ['tsmparili] / non-U [tempa'raerili] temporarily , U
['fo:midabal] / non-U [fo'midabal], U ['intrestirç] / non-U [inta
' resti*;]. ['viena] is old-fashioned U for normal [vťcna] Vienna

1 In concluding this commentary on Chapman's work I cannot

refrain from citing the following (p. 253) : - »Undergraduates
call each other what they choose, but never I suppose anything
more formal than 'Smith', except on delivery of a challenge or the
like» [my italics). I feel that Chapman has slipped a century or s
here, for the duel has long ceased to be a feature of undergraduate
life!
2 The pamphlet The B. B . C.'s recommendations ¡or pronouncing
doubtful words [ed. R. Bridges] (S. F. E. Tract, No. XXX II) is
useful here, though some of the pronunciations advocated are not
so much U as pedantic (e.g. »aerated ày-erated» for ['¿oreitid],
»casualty cázewalty, not cázhewalty» (['kaežalti] is normal),
»Celtic . . . initial consonant . . . s9 not /c»), while others are, to-
day, definitely outré (e.g. »culinary kéw-» for ['kalinri]).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 36 Alan S . C. Ross

confessor is normally [kan'fesd

In some cases two stress-varian
cáke or sponge-cake2 (to-day
word, substituting sponge for
3) Phonemes. There are very f
materialisation of an entire ph
non-U pronunciations. The p
mal [ai], [au] - as in [reid] ri
ever, of common occurrence a
[ei] - Shakespeare's version of
a well-marked kind of non-
indicated by the epithet refain
me on the other side of the
often not a concomitant of re
has been regarded as a feature
experience, this feature hardly
cat - is, of course, a well-know
and is thus ipso facto non-U
/-phoneme - [ťij] 5 instead o
among non-U-speakers. Amo
pronunciation» of wh of which
to women. In some cases, too, t

1 Successor as fsaksesd] among this same class of speakers, as

against normal jsak'sesa], is possibly due to direct analogy with
confessor.
2 Sponge-bag does not appear to show this variation, spónge-
bag being normal; the difference is, no doubt, to be ascribed to the
differing compositional ellipses (sponge-cake 'a cake [which is] a
sponge', sponge-bag 'a bag [for holding] a sponge').
3 R. E. Zachrisson, The English pronunciation at Shakespeare* s
time ( Skrifta utgivna av K. Humānistiskā Vetenskaps-Samfundet i
Uppsala XXII. 6), p. 55.
4 Cf. the non-U character Mr. Starky in R. G. Woodthorpe's
Public-School murder , who is given by his author sentences such as
There he was , like a med thing , dencing round me in the pessage
(1932 ed., p. 25)*
6 With a ť reminiscent of that found in Armenian and Georgian.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 37

a phoneme may serve as an indicator. Thus many (but

U-speakers, but no non-U-speakers, say [git], [dzēst
instead of [get], [džast], [kaetš] for get , just adv., catch
otherwise materialising their e-, a-, «-phonemes as [i]1
In some cases, U-speakers prefer long vowels, in other
(against the converse for non-U) e.g. U [spuwn] : non-U
spoon but U [rum], [brum] : non-U [ruwm], [bruwm
broom. In U-speech hoof is usually [huf] with plural [h
[huwvz]. Some U-speakers use [go:n] instead of [gon
4) Phoneme-groups.
(i) L-groups.
(a) There is U-loss of I in [gof], [reif], ['soda] as against non-U
[golf], [rself], ['solda] ['soulda] golf , Ralph , solder. The old-
fashioned U-pronunciations of falcon , Malvern were also 7-less
(['forkan], [*mo:van]) but it is doubtful how far these survive
to-day.
(b) (Re)solue as -[soulv] is non-U (U : [solv]).
(c) Real , ideal as [rijl], [ai'dijl] are non-U ( really ['rijli]
being especially frequent) as against U [rial], [ať dial], ['riali].
(d) fault is non-U [folt] as against U [fo:lt]; similarly,
also, Balkans , Baltic , halt , malt, salt , vault.
(ii) » R»-groups .
(a) U-pronunciation has [a:] against non-U [o:] 3 in
Berkeley 4, Berkshire , clerk , Derby.

1 At one time this [i] evidently encroached further; for instance,

I believe that ['kiniist] is still old-fashioned U for ['kemist] chemist
(cf., no doubt, the tendency for old-fashioned chemists ' shops to
use an old spelling Chymist on their windows).
2 My impression is that [do:g] dog occurred in U-slang of about
the nineties, as in You look no end of a dawg ! (also in adj. dawgy
'smart'?).
8 The non-U pronunciation is clearly from the spelling; cf. also
p. 50, below.
4 Since it is definitely non-U to pronounce Berkeley as ['barkli],
U-speakers get a frisson if they have to enunciate the surnames
Birkley , Burkly (correctly pronounced ['ba:kli]) for, if a U-hearer

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 38 Alan S . C. Ross

( b ) As to intervocalic r : ['sa
U ['sirap]); ['starap] stirrup m
normal pronunciation is everyw
(c) Some U-speakers attest
|a:], mostly in the final positi
['farri], lion [la:n]) x. So, to su
perfect homonyms.
( d ) On the other side of the
have simple [o:] in boor Boer
( feather-)boa , boa(-constrictor ); m
Boer I boa are thus sets of homo
whether U- or non-U, would pr
homonymically with born (thou
normally have [bo:n]).
(iii) ME. au + nasal.
(a) To-day, U-pronunciation normally has [a:] in words
such as dance , demand , grant , [se] being non-U. I am inclined to
think that, in some words (demandi), U-pronunciations with
[ae] may have survived into living memory. To-day aunt is U
[a:nt] and ant is U [sent] but some non-U speakers reverse the
phoneme-roles, calling aunts ants and ants aunts.2
(b) [a:] certainly existed as a U-pronunciation in haunch ,
launch , staunch 3 until comparatively recent times 4, but it is
doubtful how far it survives; most U-speakers have [o:] here.
(iv) [a]
(a) U-speakers normally have [a] before the nasal in
accomplish, Brompton, Conduit 5 ( Street ) and constable as against

does not appreciate the spelling of the names (rare ones), they may
be suspected of using a non-U pronunciation.
1 Also via [va:].
2 But, until fairly recently, to make this reversal was possible U.
8 To stanch bleeding is hardly in spoken use to-day.
4 The one-time U-pronunciation [la:ndri] laundry (normally
[lo:ndri]) seems to me even more old-fashioned.
6 non-U ['kondjuit]. Increasingly, the old .U-pronunciations of
place-names are tending everywhere to be replaced by spelling-

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 39

non-U [d] (['konstabal], etc.). Some other words (e.g

once had U [a] too, but, in these words, the use see
escent1. [a] before nasal instead of [a] - indeed [a]
general - is felt by the non-U to be extremely U
anxious to »change their voices» (cf. p. 47, below) are in
fall into a »snob-change» of type Leeds [Veri] for norm
worry 2 and enlarge their [a]-domain; I have even hea
dog 3 from this kind of speaker 4.
(b) Some non-U-speakcrs use [o] for [a] in none an
In origin this is probably a dialect form, springing f
Industrial Midlands5, a great breeding-ground of non-
(V) fo].
(a) [in] for [irj] in verbal forms (hunting shootin' and
fishin* ) was undoubtedly once a U-indicator and it still
survives among a few very elderly U-speakers; among younger
ones, it seems, to-day, to be altogether dead.

pronunciations of non-U origin. Thus, to-day, few railway-booking-

offices understand ['aksita] for Uttoxeter ([ju'toksita] is prevailing).
Cirencester is interesting; the true U-pronunciation is ['sisita] or,
occasionally, ['sisista] (U-pronunciation influenced by spelling?);
the townsfolk, however, call it ['sairansesta] from the spelling.
1 In other words [a] is universal e.g. comfortable ; [bam] for
normal [bom] bomb is (old-fashioned?) Army use; in any case, no
doubt homonymity with bum has acted as a prophylactic against
the dissemination of the pronunciation (cf. normal [gra:s] grass
but [ses] ass to avoid homonymity with [a:s] arse - a point to
which I called attention TPS 1934, p. 99).
2 Discussed by m ečasopis pro moderni filologii XXXII, 1949, 38.
3 But ['havei] hovel appears to be U (though ['hoval] is also
possible).
4 In the summer of 1951 the B. B. C. caused some amusement
by habitually pronouncing the name of the cricketer Compton as
['kamptan].
5 J. Wright, The English dialect grammar , Index s. vv. none ,
one records »non» from Staffordshire, Derbyshire, Lincolnshire,
Rutland, Leicestershire, Northamptonshire, Warwickshire and
Shropshire; »won» from each of these counties (and from others
as well).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
40 Alan S. C. Ross

(i b ) [n] in length [lene], stren

as against U [ ìerje ], [strafe].
(vi) Vowel + voiceless spirant
(a) Some U-speakers have a lo
of this type. Thus, basing th
Historische grammatik der eng
in o//, crofter , loss , lost , frost , b
loft , froth , cos/, clotĶ cough , tr
(//055, wasp and also dross, i?os
(b) On the other hand, long
can sometimes be non-U; it is s
against U [a'laestik]. Probably in
tainly in emasculate , masculine ,
has no significance for class-dif
(vii) non-U use frequently has
['Iekša] 3 as against U ['piktša],

I conclude this section with a miscellaneous list of words.

The entry of a pronunciation in both the U and the non-U
columns indicates a clear dichotomy between the classes; the
placing of a U-entry in round brackets ( ) means that only
some U-speakers use the pronunciation (often, that it is old-
fashioned) while other U-speakers use that given in the non-U
column.

1 non-U-speakers have often ļ'oftanļ as against U ['oifanj (also

i'Dfan]?).
1 [to:s] exists as old-fashioned U.
• No doubt, also, fracture , stricture as ['fraekša], ['strikša],
though I do not remember having heard these.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 41

U non-U

acknowledge [ak'nalidž] [ak'noulidž] 1

Catholic (['kaielik]) 2 ['kaeelik]
either ['aiöa] [*ij öa]
extraordinary [ak' stro: dinri] [ 'ekstra
foreñead ['foridļ ['fo:hed]
geyser ['geiza] [*gijza]
handkerchief ['henkatšif] ['henka tšijf], ['henkatšijv]
hotel ([ou'tel]) [hou'tel]
humour (['juwma]) ['hjuwma]
mass ([ma:s]) 2 [maes]
medicine ['medsan] ['medisan]
a nought [a / no:t] [sen / o:t] 3
tortoise ['tortas] ['toitois], [*to:toiz] 4
vase [va:z] [vo:z], [veiz]
venison ['venzan] ['venizan]
W. ['dabalju] ['dabiju] 5
waistcoat ['weskat] ['weistkout] 6
(5) Morphology . Oddly enough, this important philologi
sphere produces hardly any marks of class-difference; for
1 But 1 have not hitherto observed the non-U pronunciation
['noulidž].
2 These forms are, in the main, confined to Catholic U-speakers
(cf. p. 36).
3 For 'zero* we have U nought as against non-U ought.
4 Porpoise is certainly U ['po:pas], possibly non-U ['poipoiz];
turquoise is U ['ta:kwa:z] but I am not sure of the non-U form
(['taikwoiz]?).
5 W. either 'the letter W or 'W.C.' (a frequent non-U expression
for 'lavatory') - hence, no doubt, the non-U childrens' word
dub(by 'lavatory'. (In this connection, I may mention a U-expres-
sion: Let me show you the geography of the house (meaning,
essentially, the lavatory); the non-U have many euphemisms
(e.g. the temple of health)).
• The old U-form [weskit] is now obsolescent; the word's
phonological companion wainscot had U ['wenskat] but this seems
for the most part to have been replaced by ['weinskat] in all classes.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
42 Alan S. C. Ross

such as childer (dialect) for chil

for taken 1 fall outside the sco
plained on p. 24. I have obser
omission of the genitival ' s in
in 1 bought it at Woolworth );
speakers. (2) Pret. sg. ate which
(6) Syntax . This is always a
philological sphere. Syntactic
very rare. Northern Standard
(1) the use of while for normal
(2) inversions, such as He* s be
phrases with prepositions ar
school 2, She's on holiday and, less
(6) Vocabulary z.
Article 'chamber-poť is non-
vives, U-speakers use ['džeri]
Bath. To take a bath is non-U
Civil : this word is used by
behaviour of a non-U person in
the difference between U and
tainly very civ i I.
Coach 'char-a-banc* is non-U

1 Wake has three possible past p

but, apparently, these variants
classes. Indeed all Englishmen hesitate when asked what the
»correct» past participle is.
2 In any case, boarding-school is little used by U-speakers, for,
to most of them, there is no other kind of school.
3 Naturally, care must here be taken to distinguish words and
phrases which are U / non-U dividers from those which are dividers
of other kinds. Thus the (well-known) phrase Animals always
know , don't they? (to my astonishment still alive in 1951) does not
divide those who are non-U from those who are U so much as
those who are fools from those who are not fools.
4 But the (recent?) verb to pot (trans.), used of babies, is surely
non-U?.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 43

itself is. Those U-speakers who are forced, by pe

them call them buses , thereby causing great c
coach runs into the country, a bus within a town).
non-U corsets / U stays.
Counterpane , bedspread , coverlet . Of these three
I think that the first is U, the second obsolete, the t
Cruet. The sentence Pass the cruet , please is v
cruets are in themselves non-U. In gentlemens* hous
ideally, separate containers - salt-cellars , peppe
tors , - grinders , -mills) and mustard-pots , so that the
ding U-expression will be I wonder if you could
( pepper , mustard ), please ?, or the like. Vinegar
constituent of many cruets but many uses of v
poured on fish or bacon-and-eggs) are definitely non
Crust or crumb?, used when cutting bread is (old-
non-U.

Cultivated in They're cultivated people is non-U and

so also is cultured . There is really no U-equivalent (some U-
speakers use civilised in this sense).
Cup . How is your c u p? is a non-U equivalent of Have
some more tea?, or the like. Possible negative non-U answers
are Vm doing nicely , thank you and (Quite) sufficient , thank you .
There is a well-known non-U affirmative answer : I don't mind
if I do (but this is U at Johnny Ludlow v. 365) 1.
Cycle is non-U against U bike , bicycle (whether verb or
noun); non-U motorcycle / U motorbike , motorbicycle is perhaps
less pronouncedly so.
Dinner . U-speakers eat lunch in the middle of the day
(luncheon is old-fashioned U) and dinner in the evening; if a
U-speaker feels that what he is eating is a travesty of his dinner,
he may appropriately call it supper . Non-U-speakers (also
U-children and U-dogs), on the other hand, have their dinner
in the middle of the day. Evening meal is non-U.
1 Cf., here, the facetious non-U expression Any more for any
more?

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 44 Alan S. C. Ross

Dress-suit. This is a non-U wor

answer the question What shall I
following ways : - 1) Dinner jac
fashioned); 3) Black tie; 4) Tails ;
German smoking , Nos. 4-5 Germ
dress is often used on invitation
currency among U-speakers (in an
guous); a sentence Shall we wear
possible, the appropriate expression b
Excuse my glove . This expression
is (?old-fashioned) non-U; male U-
their glove in order to shake han
Greatcoat (also topcoat?) are rath
coat being normal. Burberry 2
genre, macintosh or mac being no
Greens 'vegetables' is non-U.
Home : non-U They* ve a lovely
nice house .

Horse-riding is non-U against U riding. From the non-U

point of view the expression is reasonable, for to the non-U
there are other kinds of riding (cf. non-U to go for a motor-
ride I U to go for a d r i v e in a motor-car). But bicycle-ride
is normal.
Ill in I was very i 1 1 on the boat is non-U against U sick .
Jack. At cards, jack is non-U against U knave , save in
jackpot (at poker). My son, Pilot-officer A. W. P. Ross,
kindly calls my attention to the following passage from C.
Dickens, Great Expectations (ed. of 1861, i, 126) : - »He calls
the knaves, Jacks, this boy! said Estella with disdain».
La-di-da is an expression with which the non-U stigmatise
a U habit, speech-habit or person.

1 For women, a semi [sc. evening-frock] is non-U.

2 This use of Burberry no doubt arose because, even before
1914 (when U-speakers were richer than non-U-speakers), this
was a good and expensive kind of macintosh.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 45

Lounge is a name given by the non-U to a room i

houses; for U-speakers, hall or dining-room might wel
nearest equivalent (but all speakers, of course, speak
lounge of a hotel).
non-U menial / U mad.
A matter of business is non-U (as in Say you've come
him on a matter of business ).
Mention : If you don't mind my mentioning it is
Mirror (save in compounds such as driving -, shaving-
is non-U against U looking-glass.
non-U note-paper / U writing-paper x.
Pardon! is used by the non-U in three main way
1) if the hearer does not hear the speaker properly;
apology (e.g. on brushing by someone in a passage; 3
hiccupping or belching. The normal U-correspondences a
curt viz. 1) What?, 2) Sorry!, 3) [Silence], though, in th
two cases, U-parents and U-governesses are always t
make children say something »politer»; - What did you
I'm frightfully sorry are certainly possible. For Case 3)
other non-U possibilities e.g. Manners /, Beg pardon /, Pard
Phrases containing pardon are often answered with
by the non-U.
To pass a ( nasty ) remark . He passed the remark th
is non-U.

Pleased to meet you! This is a very frequent non-U response

to the greeting How ďyou do? lJ-speakers normally just repeat
the greeting; to reply to the greeting (e.g. with Quite well , thank
you) is non-U.
Posh 'smart' is essentially non-U but, recently, it has gained
ground among school-boys of all classes.
non-U preserve / U jam . But conserve is legitimate, though
rare (I imagine there to be some culinary distinction between,
for instance, conserve of quinces and mere quince jam).
1 This distinction (as well as some others e.g. non-U perfume /
U scent) is noted by N. Mitford, The pursuit of love (1945 ed., p. 31).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
46 Alan S . C. Ross

non-U radio / U wireless (but

Rude meaning 'indecent* is
U-correspondent.
Non-U servielte / U table-napk
all the linguistic class-indicator
Sit in He's sitting an exam . is
hardly U unless used very tech
Study in He's studying for
(U : working for).
Teacher is essentially non-U,
by the U to indicate a non-U
master , mistress with prefixe
Non-U children often refer to their teachers without article
(as, Teacher says . . .).
non-U toilet-paper / U lavatory-paper .
non-U wealthy / U rich.

Before concluding with some general remarks, there are

two points which may appropriately receive mention here.
First, slang . There seems no doubt that, in the nineties and
at least up to 1914, U-speakers (particularly young ones) were
rather addicted to slang. To-day, however, U-speakers use it
little and regard much use of it as non-U - save, of course,
in special circumstances (e.g. in the case of young boys at
school). American slang is especially deprecated (save, perhaps,
for 0. K.). The ultimate War, like the penultimate one, brought
a flood of slang into the Services, some of it of a very vivid kind
as, for instance, R. A. F. slang He tore me off a strip 'he repri-
manded me severely', I was shot down in flames ' I was comple-
tely overwhelmed in the argument'. Since the War, there has
been an unfortunate tendency for non-Service personnel to
use Service slang and it is clear that Service personnel regard
such use as in very poor taste. Nevertheless, the expressions
l' ve had iti (meaning, essentially, 'I have not had iť) and

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 47

Thaťs a bad show , have become very frequent among

ses of speakers 1.
Secondly, changing one's voice 2. In England
just as much as in the England of many years
question »Can a non-U speaker become a U-speaker
noticeably of paramount importance for many English
for some of their wives). The answer is that an adult
attain complete success. Moreover, it must be rememb
in these matters, U-speakers have cora ânoveiv , so
single pronunciation, word or phrase will suffice to b
apparent U-speaker as originally non-U (for U-speaker
selves never make »mistakes»). Under these circum
efforts to change voice are surely better abandone
fact, they continue in full force and in all stťata of so
the whole, the effect is deleterious. Thus, to take
example: in village schools, any natural dialect th
left to the children will have superimposed upon it th
of the primary school-teacher (a class of people entire
so that the children leave school speaking a mixture w
nothing to recommend it. In concluding this paragrap
mention that there is one method of effecting change
provided the speaker is young enough. This is, to s
first to a preparatory school, then to a good public-sch

1 There are, too, things for which the Service-slang wor

be used for no other exists (as, for instance, blood-chit 'do
signed by a non-R. A. F. passenger before a flight, which
the R. A. F. in the event that they damage or destroy h
2 This phrase is my own coinage (of many years ago
of no other expression.
3 Logically, the converse question »Can a U-speaker b
non-U-speaker?» should also arise, but, in practice it se

line here. At all events I have only come across one case of it
(in Leeds).
4 To-day similar arrangements can be made for girls; the older
approved method was, of course, a U-governess.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 48 Alan S . C. Ross

method is one that has been app

and, at the moment, it is alm
interesting to speculate upon
arise when the day comes when
able to afford to educate his children at these kinds of schools
(this day has already dawned).

If we consider the wider implications of the linguistic class-

indication discussed above, two points immediately arise: the
linguistic class-indicators are almost all philologically trivial
and, apparently, almost all of a very ephemeral nature.
The triviality is apparent at a glance; the deepest change
involved is only a differring phonetic materialisation of a
phoneme e.g. [aeu] for [au] of house . There is nothing here even
of the philological calibre of Finnish alternations of type gen.pl.
opettajain ~ opettajien (to opettaja 'teacher') - an alter-
nation which not only does not imply class-difference but,
apparently, implies nothing at all.
With regard to the ephemerality. In passing, I have already
called attention to two examples viz. 1) that Shakespeare -
and Queen Elizabeth 1 too - certainly spoke »retained» Eng-
lish in that they pronounced ride as [reid] (p. 36) and 2) that
Johnny Ludlow, a century ago, used 1 don't mind if I do quite
naturally (p. 43). I am convinced that a thorough historical study
of the class-indicators discussed above would reveal many present-
day U-features as non-U at an earlier period, and vice versa.
Such a study could be carried out. To take one example : the two
pronunciations of haunch - [ha:nts] and [ho:ntš] - both
develop from ME. haunch ( < OFr. hanche) along the lines
c. 1650 c. 1750

[haerntš]

[hauntš]

- [ha:ntš]

1 Luick, op. cit . §§ 520 - 1, 557 ff

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 49

If, therefore, we regard [harntš] as more U than [ho:ntš] (p

above) then, if the distinction is not ephemeral, it should f
low that, earlier, [hae:ntš] was more U than [harntš].
points could all be investigated with the aid of the »gr
marians», that large body of writers, English and foreign, w
from the mid-sixteenth century onwards, poured out a str
of works setting out the »correct» pronunciation of English
Englishmen and foreigners. The investigation would, howe
be very laborious and, also, rather difficult; in general, I m
relegate it to my (very few) colleagues who are specialists in
grammarians, and confine myself here to an illustrative st
of one grammarian.
I select J. Walker's Critical pronouncing dictionary a
expositor of the English language , first published in 1791
through his chapter »Principles of English pronunciation»
give an almost exhaustive list of the passages in which
would appear to be differentiating between U- and no
speech x. I give his paragraph-numbers and the little comm
tary that is needed 2.

1 I have eschewed citations such as the two following. Ȥ 1

There is an incorrect pronunciation of this letter [sc. u] wh
ends a syllable, not under the accent which prevails, not only
among the vulgar, but is sometimes found in better company;
and that is giving the u an obscure sound, which confounds it
with vowels of a very different kind. Thus we not unfrequently
hear singular , regular , and particular , pronounced as if written
sing-e-lar, reg-e-lar , and par-tick-e-lar; but nothing tends more to
tarnish and vulgarize the pronunciation than this short and obscure
sound of the unaccented u». »§ 346 buoy, pronounced as if written
bwoy , but too often exactly like boy. This, however, is an impropri-
ety which ought to be avoided by correct speakers». Such views are
typical of many of the grammarians and are due to their trying
to make the pronunciation fit the spelling. In such cases, of course,
care must be taken not to interpret words such as incorrect (§ 179),
impropriety (§ 346) as meaning 'non-U', 'non-U-feature'.
2 Frequently the only commentary needed is my own pro-
nunciation of the relevant word (which may be taken as normal)
e.g. as in »N ['oudjos] odious ».
4 - Mitteilungen

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 50 Alan S. C. Ross

Ȥ 92. When the a is preceded

it is, in polite pronunciation, so
sound like e, so that card , cart
like ke-ard , ke-art , ghe-ard, re-g
Ȥ 100. Service and servant ar
order of speakers, as if writte
among the better sort, we som
though this pronunciation of t
upon as a mark of the lowest
Derby and Berkeley , still ret
Darby and Barkeley : but even
into the common sound, nearly
keley ».
Ȥ 101 ... in the words, England , yes , and pretty , where the
e is heard like short i. Vulgar speakers are guilty of the same
irregularity in engine , as if written ingine ; but this cannot be
too carefully avoided».
Ȥ 104 . . . nothing is so vulgar and childish as to hear
swivel and heaven pronounced with the e distinctly, or novel
and chicken with the e suppressed».
Ȥ 101 [On obleege / oblige ] ... we not unfrequently hear it
now pronounced with the broad English i1, in those circles
where, a few years ago, it would have been an infallible mark
of vulgarity».
Ȥ 144. housewife , pronounced huzzwiff . . . Midwife is some-
times shortened in the same manner by the vulgar 2».
Ȥ 170 . . . and, as it would be gross to a degree to sound the
a in castle , mask and plant, like the a in palm , psalm , &c. so it
would be equally exceptionable to pronounce the o in moss ,
dross , and frost , as if written mauwse [sic], drawse and frawst h.
Ȥ 210. Plait , a fold of cloth, is regular, and ought to be

1 i.e. as in ride.
* But cf. N. [mid'wifari] midwifery .
8 Cf. p. 40 above.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 51

pronounced like plate , a dish; pronouncing it so as to rh

meat is a vulgarism» l.
Ȥ 218. There is a corrupt pronunciation of this diph
[sc. au] among the vulgar, which is, giving the au in d
sauce , saucer and saucy , the sound of the Italian a, an
as if written darter , sarce , sarcer, and sarcy; but this
ciation cannot be too carefully avoided. Au in sausag
sounded by the vulgar with short a, as if written sass
in this, as in the other words, au ought to sound awe ».
Ȥ 224. The familiar assent ay for yes , is a combinatio
long Italian a in the last syllable of papa , and the firs
of e. If we give the a the sound of that letter in ball ,
degenerates into a coarse, rustic pronunciation. Thoug
House of Commons, where this word is made a noun, we
frequently, but not correctly, hear it so pronounced, in the
phrase the ayes have it.»
Ȥ 238. Leant , the past time and participle of to lean , is
grown vulgar : the regular form leaned is preferable ^>.
Ȥ 241. The word great is sometimes pronounced as if written
greet , generally by people of education, and almost universally
in Ireland; but this is contrary to the fixed and settled practice
in England. That this is an affected pronunciation, will be
perceived in a moment by pronouncing this word in the phrase,
Alexander the great ; for those who pronounce the word greet , in
other cases will generally in this rhyme it with fate . It is true
the ee is the regular sound of this diphthong; but this slender
sound of e has, in all probability, given way to that of a as
deeper and more expressive of the epithet great».

1 To-day the normal usage is plait [plset] v. and sb. (as plaits
[of hair]) and pleat [plijt] v. and sb. ('fold of cloth'); see further
NED. s. vv. Plait sb., Pleat sb. (here - in a volume published
in 1907 - NED. still records [pleit] as a possible pronunciation
of plait).
2 See p. 55, note.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
52 Alan S. C. Ross

Ȥ 247. Cheesecake , sometim

breech , 6ri7cA, I look upon as vu
Ȥ 252. Either and neither are
and nigh-ther . . . Analogy, ho
the diphthong the sound of lon
breather , one who breathes. Th
rick always gave to these word
true one. We sometimes, indeed
words sounded like slender a, as
but this pronunciation must be
Ȥ 253. Ei has the sound of lon
rhyming with white and right
rhyming with eight and weight
able speakers».
Ȥ 267. Ew is sometimes prono
chewy but this is gross and v
rhyme with new, view, & c.».1
Ȥ 268 . . . the word ewe ...
yew . . . There is a vulgar pr
written yoe9 rhyming with doe ,
Ȥ 293 . . . the vulgar . . . ins
say o-jus and te-jus»2.
Ȥ 294. Indian . . . ought, th
usage, to be pronounced as if
In-je-an b.
Ȥ 299. The general and alm
diphthong [sc. oi] is that of a in w

1 Chaw v. was regarded by NED

lished in 1889 (s.v.). To-day the
does (the dog was chawing a bone9 this lane's got completely
chawed up with track-tyres would be two possible sentences),
it is perfectly U (chaw-bacon sb. 'a bumpkin' is surely obsolete
to-day - the last occurrence in NED. (s. v.) is from 1880).
2 N ['oudjds], ['tijdjas].
* Is ['indžan) the »respectable usage» intended?

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 53

This double sound is very distinguishable in boil , toil

joint , point , anoint &c. which sound ought to be
preserved, as there is a very prevalent practice am
vulgar of dropping the e, and pronouncing these w
written bile , tile9 spile , &c.»
Ȥ 300. I remember, very early in life, to have hear
pronounced as if written quine by some respectable sp
but this is now justly banished as the grossest vul
Ȥ 309. Soot is vulgarly pronounced so as to rhyme with b
&c. but ought to have its long regular sound, rhyming wit
Ȥ311. Moor , a black man, is regular in polite pronu
and like more in vulgar».
Ȥ 317. To wound is sometimes pronounced so as to
with found ; but this is directly contrary to the best
Ȥ 327. The vulgar . . . pronounce the o obscurely, an
times as if followed by r, as winder , feller , for win
fellow ; but this is almost too despicable for notice».
Ȥ 333. In Mantua , the town of Italy, both vowels ar
distinctly. The same may be observed of the habit
but in mantuamaker vulgarity has sunk the a and
mantumaker . The same vulgarity at first, but now s
by universal custom, has sunk both letters in victuals
compounds victualling and victualler , pronounced vitt
ling and vittlen.
Ȥ 341. Guild , in Guildhall , is, by the lower people of L
pronounced so as to rhyme with child ; but this is
opposite to the best usage».
Ȥ 374. The vulgar drop this letter [ sc . d] in ordin
extraordinary making them oťnary and extťoťnary ; b
a gross abbreviation; the best pronunciation is suf
short, which is ord*nary and extrorďnary; the first i
and the last in four syllables h.
1 To-day or'nary is an American dialect form well-known to the
English from light fiction; extr* or'nary is a form which might well
belong to the speech of a drunk gentleman (cf. also p. 41, above).

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
54 Alan S . C. Ross

Ȥ 393. Drought (dryness) is vu

Ȥ 404. L ... is sometimes sup
suppression is become vulgar. In
times suppressed, and the word
far from being the most correct
Ȥ 410 . . . our best speakers
the participial ing , so as to rhym
Ȥ 462 . . . the words [sc. tutor ,
pronounced as if written tewtor
neither, tshootur , tshoomult , tsh
toomour, as they are often pro
Ȥ 415 . . . pronouncing it [sc
[sc. w] is sometimes dropped in
if written awkard ; but this pro

The ephemeral nature of our p

class-indicators is very clear f
Walker. Nearly all the points
hundred and sixty years ago
class-significance, in that one
to-day no longer known in an
Only one of Walker's class-indic
§ 410) has been recognised by
even that one I have regarded
than my own. In three cases of
U alternative is chosen by Walk

1 N : as recommended by Walker
2 Viz.: § 252 - to-day either , n
non-U ['ijöa], ['nijöa]; §309 - soot is U [sut] / non-U [suwt]
([sat] is, to-day, American); § 100 - Walker accepts ['da:bi],
['ba:kli] Derby , Berkeley as »polite usage», although he does not
regard [*da:bi], [ba:kli] as vulgar, but, to-day, the former are

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 Linguistic Class-indicators 55

Among European languages, English is, surely, t

most suited to the study of linguistic class-distinc
not really know how far such a thing may exist in
In Finnish, I have the impression that no phenomen
sort exist. In German, there may well have been so
comparable; certainly, I recall that, in good Potsdam
of the late twenties, the expression kiiss ' die hand (on
duction to a female) was definitely frowned on -
society has vanished »swã hão nõ wsëre». In Russian
distinction between the two plurals of ofitser 'of
ofitsery and ofitsera - is certainly one of class. The
to be remarkably little literature on the subject save p
(rather naturally) by Russians and / or as concerns Ru
It is to be hoped that more studies of linguistic class-d
in the European languages will one day be forthcom
However, the general concept of a certain varian
language appertaining to a certain section of its speak
old women, or children) is one very well-known to an
gists and it is, no doubt, in the African jungle and am

definitely non-U, the latter U (p. 37 above). As to furth

To-day there would not be anything non-U in making M
homonyms (p. 38 above), though to keep them separate
imate; the former procedure is non-U to Walker (§ 311). In § 327,
Walker stigmatises the pronunciation ['fela] fellow as »almost too
despicable for notice»; to me, however, ['fela] is old-fashioned U,
though I use ['falou] myself. In § 238, Walker states that leant »is
grown vulgar» (as against leaned); to-day leant/leaned is hardly an
indicator though most U-speakers prefer leant . (For chew/chaw
§ 267, see my note ad loc.).
1 Cf., particularly, Y. Zhirmunskij, National* nyi yazyk i
sotsial'nye dialekty (and Primechaniya).
2 In this context I must emphasise something that I said in
the first Vilém Mathesius Memorial Lecture (Prague, 1948),
pp. 8 - 9, viz. that studies of this kind should only be undertaken
by native speakers.

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms
 56 Tautio F. Muslanoja

Red Indians1 that we shall find the gene

linguistic indicators of our English class-d
suitable point at which to end this artic
reached that awkward terrain where Lingu
Anthropology - and the anthropologists
appreciably active here.
Birmingham
Alan S. C. Ross

TWO LEXICAL NOTES

at random - c y p ri ne

At random. - The use of the OF. noun randon 2 'rapidity',

'impetuosity', 'abundance' in adverbial phrases like a grant
randon has been recorded since the twelfth century. The earliest
instance quoted by Godefroy (Dictionnaire de V ancienne langue
française , VI, 590-1) is from Raimberťs Ogier: Li sans vermaus
en cort a grant randon . There are several quotations of this
type in Godefroy, but his only quotation without the attribute •
grant is remarkably late (1622): Les uns [des faucons] volent
de poing et prennent a randon (c'est a dire de force , cum Ímpetu ),
les autres volent haut (E. Binet, Merveilles de Nature , ed. of
1622, p. 44).

1 Cf., particularly, E. Sapir, Abnormal types of speech in Nootka.

2 There is a variance of opinion concerning the etymology of
this word. E. Gamillscheg (Etymologisches Wörterbuch der franz.
Sprache , Heidelberg, 1928, p. 740) believes that it comes from the
unattested Franconian rando 'a shield', 'the edge of a shield'.
O. Bloch and W. von Wartburg (Dictionnaire étymologique de la
langue française , 2nd ed., Paris, 1950, p. 506) think that randon
is a derivative of the OF. verb randir 'to run fast', which goes
back to an unattested noun rant (cf. brandir , brant). Bloch and
von Wartburg compare a randon (a random) to expressions like
à reculons .

This content downloaded from

82.30.53.45 on Sun, 07 Feb 2021 23:31:01 UTC
All use subject to https://about.jstor.org/terms