A comparison between Plain language movements, style guides and practice

CLARITY

Frequently, though we talk about transparency,

We proliferate opacity
When what we need is clarity.
Nowadays, there's an ever-growing tendency
To obfuscate with much prolixity,
When what we need is clarity.
You wrote something long; that is wrong, it will not do.
Keep it plain and short and the message will get through.
Just write with ...
Clarity means abandoning obscurity
And preferring more simplicity.
Write English as it ought to be.
Yes, what we need is clarity.

(Dave Skinner, Translation Service

European Commission 1998)

Maria Sundin
English Department
C-essay: Linguistics
Supervisor: Britt Erman
HT-99
Contents
Introduction..............................................................................................................................................3
Aim and scope.........................................................................................................................................3
Material....................................................................................................................................................4
Method.....................................................................................................................................................6
Result and discussion..............................................................................................................................7
I. The plain language movement................................................................................................7
II. The style guides.....................................................................................................................11
III. The texts................................................................................................................................12
Conclusion..........................................................................................................................................17
References............................................................................................................................................19
Appendix................................................................................................................................................20
Introduction
Plain Language is now a worldwide activity. In Europe countries like the United Kingdom and Sweden
have been practising plain language for almost 30 years while other countries like France and Spain
recently have started to revise their legal and government documents in a clearer style. In the EU the
translators primarily from the United Kingdom have started a campaign for plain language in EU-
st
documents. This will probably lead the way into the 21 century.

What is plain language?

Plain language can obviously be described in many different ways, but most advocates of Plain
Language seem to agree on it being ”the writing and setting out of essential information in a way that
gives a co-operative, motivated person a good chance of understanding the document at first reading,
and in the same sense that the writer meant it to be understood” (Cutts 1995:3). It often means using
”appropriate structure and layout to help them navigate through the document” (Cutts 1995: 3).
Professor Robert Eagleson, Sydney, Australia, states that plain English is a clear and
straightforward way of expressing oneself, using only the words necessary. This language avoids
“obscurity, inflated vocabulary and convoluted sentence construction”. Still it is not baby talk. “Writers
of plain English let their audience concentrate on the message instead of being distracted by
complicated language. They make sure that their audience understands the message easily"
(Ehrenberg-Sundin1993).
It is therefore important to bear in mind that Plain English is not in any way an absolute or static
concept. It may alter over time and according to who the audience is. This is why Plain Language has
many general guidelines, since it is unnecessary to be specific if the rules change according to
audience. Readers´ prior knowledge and expectations change according to what group they belong to,
according to Cutts (1995: 4). It is simply a way of adjusting the text to its readers. However, quite a few
of the guidelines are rather specific as will be shown in this essay.

Aim and scope

Earlier research
Joseph M. Williams at the University of Chicago has stated that “Those who must write clear, readable
sentences now have access to about 90% of everything they need to know to write those sentences”
(Steinberg 1991: 61). That knowledge consists of three principles: “Express crucial actions as verbs;
locate the participants of those actions in the subjects of the verbs, and arrange information in those
sentences so that older, more familiar precedes newer, less familiar” (Steinberg 1991: 59).
This corresponds very well with what all the style guides in this study recommend for plain language.
At the same time other researchers, such as Thomas N. Huckin, Elizabeth H. Curtin, and Debra
Graham, Carnegie-Mellon at the University of Pittsburg have found proof of violation of many other
guidelines in actual plain language texts (i.e. texts which have been proof-read in plainness). For
example they found that the “guideline to ´Use the active voice /.../ is violated frequently in the
documents /.../ looked at – including the ten plain English ones. Likewise, the prescription to “Write
short sentences” is often ignored, as are other prescriptions like ´Avoid nouns created from verbs´ and
´Unstring noun strings.” (Steinberg 1991: 79):

Aim
My aim is therefore to investigate whether the plain language texts I have looked at also violate the
guidelines they are supposed to follow, and if so I want to discuss the possible reasons why. Perhaps
the maxims are oversimplified and perhaps the writers consider other contextual and rhetorical factors
involved, not just syntactic ones when they make their language choices.
The general aim with this essay is to investigate the similarities and differences between plain
language in the English speaking world and Sweden. In order to fulfil this aim my study is threefold. I
intend to:
• compare the plain language movements in the English speaking world and Sweden,
• compare prominent English and Swedish plain language style guides, and
• analyse English and Swedish plain language texts.

In analysing the plain language texts my main focus is on the following issues:
• to investigate whether the texts written by plain language experts live up to the style guides (and if
they do not, try to discuss some possible reasons why), and
• to find out whether the English and Swedish texts apply the rules in different ways since the
languages do have some typical differences in information structure.

Material
I have studied
• six books, most of them style guides, from the USA, the UK and Sweden: Bailey, Edward P. Jr:
Plain English at Work (1996), Cutts, Martin: The Plain English Guide (1995), Ehrenberg-Sundin,
Barbro et al.: Att skriva bättre i jobbet (English translation: To write better at work) (1999), Sundin,
Håkan: Tänk först – skriv sedan (English translation: Think first – write afterwards ) (1998) and
Wydick, Richard C.: Plain English for lawyers (1994).
• one booklet from the European Commission: Fight the Fog How to Write Clearly (1997)
• three homepages of plain language organisations from the UK and Sweden: Plain English
Campaign, Plain Language Commission and Klarspråksgruppen.

The texts I have chosen to analyse are:

• 294 sentences taken from the newly written government bill on a Swedish law, Inkomstskattelagen
1999 (English translation: income-tax law), overlooked by the Swedish language experts at
Regeringskansliet (English translation: the Swedish Cabinet Office),
• 195 sentences taken from the EU leaflet När har vi euron i våra fickor? intended for the public in
Swedish (1997),
• 358 sentences taken from 6 English leaflets about legal advice to the public by The Lord
Chancellor´s Department (1992), all marked with the Crystal Mark by the Plain Language
Campaign, and
• 176 sentences taken from the EU leaflet When will the ´euro´ be in our pockets? intended for the
public in English (1997).

Abbreviations used in this paper

I will refer to the texts by using the following abbreviations:
• Proposition 1999/2000:2 Inkomstskattelagen (referred to as Sw 1),
• När har vi euron i våra fickor? (referred to as Sw 2),
• the Lord Chancellor´s Department leaflets (referred to as En 1), and
• When will the ´euro´ be in our pockets? (referred to as En 2).

Limitations of the material

In order to get a general knowledge about the plain language movements in different countries I have
studied both more critical books (both informative books on plain language by serious writers and
critical studies of plain language texts by researchers) and the introduction of the style guides where
the writer draws a background. A possible limitation is that the style guides are written in a very
positive and selling way and might overemphasise the importance of the country’s own movement in
particular. The greatest limitation here concerns Sweden where all facts have been taken from
Swedish plain language sources.
The style guides have been picked partly at random since I did not know much about the most
recent books at the start of my survey. Later on I have found out that at least two of them are central
works in their particular countries; that goes for Plain English Guide by Martin Cutts in the UK and
Skriv bättre i jobbet by Barbro Ehrenberg-Sundin in Sweden. Both writers are leading front tables in
the plain language movement in their countries. Many other style guides in this study have apparently
been greatly influenced by them. Some state it openly, like How to Write Clearly by Fight the Fog, but
others remain silent about their inspiration. Not knowing who is being influenced by whom is of course
a limitation. It is interesting to find out how much the style guides have influenced each other – this
might be the result of copying style recommendations without careful and critical consideration.
The texts also have limitations. The greatest limitation concerns the text types. The English leaflets
are information leaflets whereas only one of the Swedish texts is. Sw 1 is a law greatly influenced by
the special legal language within the legal tradition. You can expect the texts to differ in many respects
solely because they belong to different text types. I would have preferred a Swedish plain language
information leaflet in order to reduce that difference.
Another difference between the texts concerns the reader orientation. All four texts are intended for
the public, but the language is adjusted on different levels in the different texts; Sw 1 being the least
reader oriented and En 1 the most reader oriented. This has implications for the style, and,
consequently, one can expect the writers of the most reader oriented texts to have made more radical
language choices out of consideration for their readers than the less reader oriented, for example by
choosing shorter sentences, lighter themes, more active voice sentences etc.

Method and analysis

Method
In order to make the comparison between the movements I have simply made a survey of the facts
available and focussed on history, starting points, activities and impact on society in general. I have
tried to pinpoint similarities and differences.
In my comparison between the different style guides I have focussed on what guidelines they all
have in common. In order to find these I have written down every piece of advice I could find. Then I
arranged them in different units based on what linguistic area they belonged to. Then I picked out what
could be found in all the style guides, i.e. advice that was common for all of them. If there are any
exceptions from this rule of organisation, I will comment on it in the discussion.
In the texts I decided to look at syntax and voice on sentence level. The factors, which all are
supposed to make the sentences clearer and the information they convey more accessible for the
reader, that I intend to analyse are:

• the average sentence length,

• if the active voice, not the passive, is used, and
• if there is old information at the beginning and new or complex information at the end in the
passive voice sentences.

Most style guides advocate active verbs, not passive at the same time as they advocate putting old
information at the beginning of the sentence and new information at the end. I suspect that these two
pieces of advice do not correspond, because if you want to save new information for the end you
frequently have to make the verb passive.

Analysis
Sentence length
When studying sentence length I only analysed full sentences starting with a capital letter and ending
with a full stop. Semicolons and colons have been treated as marks within the sentence.

Voice
When studying the voice my intention was to analyse only full sentences containing a finite verb. I
have only analysed the main clause of all these sentences, i. e. one per sentence, but I soon realised
however that I could not exclude all sentences without a finite verb. En 2 contains as many as 17
sentences without a finite verb, as in ex (1). Since most of them are understood, I counted them as full
sentences.

(1) Technical scenario for introduction of the euro and timetable for changeover to the single
currency in 1999 finalized (end of the process scheduled for 2002). (En 2)

Quite a number of the Swedish sentences in both Sw 1 and Sw 2 did not contain a finite verb, but on
the other hand, they did not even contain the past particle. Instead they used nominalizations to
express what the verb normally would have done, as in ex (2). I therefore excluded them from the
study. Perhaps this use of nominalizations is a strategy typical of Swedish.

(2) Fastställande av den tekniska planen för införandet av euron och tidsplanen för
övergången till den gemensamma valutan som inleds 1999 (hela processen skall vara
avslutad 2002). (Sw 2)

This might be an example of another passive voice strategy. Instead of making the verb into its
passive form, Swedish writers make use of nominalizations. Sw 2 contained several examples of
nominalizations without a finite verb at all. This might of course be a limitation of the study.

Information structure of the passive voice sentences

I have analysed the themes of the passive voice sentences in order to find out if they conveyed the
reasons for choosing the passive instead of the active voice. I wanted to investigate whether the
sentences contained light themes, invisible agents, fronting for emphasis and finally linking to other
sentences preceding the actual themes. Each of the units will be analysed and discussed separately.

Result and discussion

The Plain language movement – a comparison between the English
speaking world and Sweden

The United Kingdom

Already in the fourteenth century there were linguistically interested people like Chaucer speaking
favourably of plain English. In 1604 the first dictionary in English explained that ”hard vsuall English
wordes, borrowed from the Hebrew, Greeke, Latine, or French. &c. With the interpretation thereof by
plaine English words, gathered for the benefit & helpe of Ladies, Gentlewomen, or any other vnskilfull
persons.”(Cutts 1995: 5). From the seventeenth century, Protestants, especially Quakers, favoured a
simple style which they called plain language. (Cutts 1995: 5)
In the 1970s the plain English movement started in earnest by consumer groups using the mass
media to ”ridicule examples of obscurity in legal documents and government forms” (Cutts 1995: 6).
Chrissie Maher and Martin Cutts started the Plain English Campaign in 1979 as a grassroots
movement to fight ”gobbledygook - legalese, small print and bureaucratic language”
(http://www.plainenglish.co.uk/press.html). They obtained a broad initiative by doing so and ”as a
means of rallying forces in the community” they made change rapidly. (Steinberg 1991: 32). Maher
and Cutts made the Plain English Campaign a growing and successful consultant´s company, which
still exists today. The Plain English Campaign puts a seal of approval on public information living up to
plain English standard – the Crystal Mark – and it also gives an annual award for good documents and
hands out Golden Bull awards to ”organisations that are guilty of using impenetrable jargon”
(http://www.plainenglish.co.uk/press.html).
Insurance companies took up Plain English as a means to successful information. The first plain
English insurance policy was produced by General Accident Fire and Life Assurance Corporation
appearing in 1979 (Steinberg 1991: 31-32).
But it was not until the 1980s that the plain English movement´s impact on British society started.
The British government issued a White Paper (a policy statement) in 1982 ordering plain English for
the first time. The departments were to count their forms, throw away the ones not needed and rewrite
the rest in plain English. They were also supposed to report the progress annually directly to the prime
minister (Cutts 1995: 6).
Since then plain English has been on the move. Not only the government ministries have changed
their way of writing, local governments have also been influenced. English town halls for example have
plain English committees who vet documents before they are printed (Cutts 1995: 6). Many new
organisations committed to work for plain English were born in the 1980s: Clarity, The Plain Language
Committee, Fight the Fog etc. Many of them have been very successful in their work and received
qualified backing from the government. One reason for this might be the great savings they have
reported over the years; ”Both commercial and government enterprises in both countries have
reported extensive economic benefits, so much so that writing divisions are now being seen as
important components in improving productivity and efficiency within organisations” (Steinberg 1991:
30).
As Mark Adler, the founder of Clarity, the solicitors´ own plain language organisation, says:
- Clarity was founded in 1983 and we have used Plain English in all our documents ever since.
Then people would ridicule us by calling us cranks and accuse us for taking silly risks. Now we
see a complete change.

The United States

The plain language movement in the United States also developed from the consumer movement but
a decade earlier than in the United Kingdom, i.e. in the 1960s. The movement grew stronger in the
1970s as it did in the United Kingdom. Just like the English movement American language conscious
groups, such as teachers for example, tried to ridicule gobbeldygook. The Public Doublespeak
Committee was formed by the American Council of Teachers of English and they also gave a George
Orwell Award ”for honesty and clarity in public language” (Cutts 1995: 6). In 1974 American insurance
companies introduced a new policy influenced by plain language (Steinberg 1991: 31-32).
 In 1978 President Carter issued the famous Executive Order which was designed to ”make Federal
regulations cleared, less burdensome, and more cost effective´” (Steinberg 1991: 20). It is interesting
to notice that one argument for plain language was the economical benefit. At that moment the order
had no impact on business and industry (Steinberg 1991: 21). Instead the plain language movement
had its greatest impact on state legislation supposed to regulate consumer contracts; citizens should
be able to understand the contracts they were about to enter into (Steinberg 1991: 22).
In the 1990s president Clinton wrote a presidential memorandum on plain language where he directs
all the executive departments and agencies to “by October 1, 1998, use plain language in all new
documents” and orders that “by January 1, 2002, all such documents created prior to October 1, 1998
must also be in plain language.” (http://www.plainlanguage.gov/cites/memo.htm). On his initiative The
Plain Language Action Network (PLAN) was created. It is a government-wide group working to
improve communications from the federal government to the public.

Sweden
th
In the 16 century the Swedish king Gustav Vasa issued an order to the authorities to express
themselves in a way that the subjects understood; the sheriffs should use clear Swedish and not
th
German, Danish or Latin. In the 17 century the king Karl XI´s proclamation to use a clear language
resulted in clearly written laws. Karl XII spoke in favour of writing clear Swedish instead of French,
th
which was the fashion. At the beginning of the 20 century authorities tried to break the old fashion of
official jargon of writing long and complicated sentences.
In the 1950s the language professor Erik Wellander wrote an official report about the language in
government reports and suggested changes: to write in clear Swedish, to sift the material and to plan
the texts so that the reader would get the relevant information at the right stage. The texts should be
written in a way a reader with normal apprehension and with a knowledge of the subject should be
able to understand. In the 1960s the Cabinet Office declared that constitutional texts should be
brought closer to the common language’s words and syntax. The thought behind this was to influence
authorities by changing the laws, which serve as a pattern for other government and local authority
texts. The plain Swedish movement had started but changes were slow.
In 1976 the first language expert was employed at the government to develop methods for
improvements. In 1978 a new educational programme was installed at Stockholm university,
”Språkkonsultlinjen”: a two-year-long education to supply Swedish authorities with well-educated
persons to help with the improvements of the official language. Today there are about 180 qualified
Swedish Language Consultants.
In 1980 plain Swedish was written in a law, the administration law (1986:223), where it is stated that
authorities should express themselves in a way that is clear and easy to understand. Since 1980 there
is a team of language experts in the Government Offices who revise texts and encourage all
government officials in the ten ministries and the committees to use plain Swedish. The language
experts are placed in the Ministry of Justice, where they work together with legal advisors at the
Division for Legal and Linguistic Draft Revision.
 In 1994, the Swedish government set up the Plain Swedish Group in order to promote clear and
simple language in official documents and to encourage government agencies all over Sweden to start
plain language projects.
(http://www.justitie.regeringen.se/klarsprak/english/english.htm)

The European Union

The plain language movement has influenced the texts written or translated in the European Union
too. Plain language are now supposed to characterise language in standard form consumer contracts,
for example in the EC Council directive 93/13 where unfair terms are required to be removed (Cutts
1995: 7).
The English translators at the European Commission, with Dave Skinner as chairman, founded Fight
the Fog during the English chairmanship in 1998 (Emma Wagner 1999). They wanted the translators
to set an example ”using language which is as clear, simple, and accessible as possible, out of
courtesy to our readers and consideration for the image of the Commission”
(http://europa.eu.int/comm/sdt/en/stygd/index.htm). The translators were worried that unclear
language was damaging the EU´s image with the general public and because it was difficult to provide
good translations if the original texts were badly written (Emma Wagner 1999).
The European Commission´s Translation Service wrote their own English Style Guide How to write
clearly (which is based on Martin Cutt´s book The Plain English Guide) especially for Commission
staff. They argue for plain language since ”more and more is written in English by (and for) non-native
speakers or by native speakers who are beginning to lose touch with their language as a result of
working in a multilingual environment”.
The next plan for Fight the Fog is a clear writing award for Commission staff, “to encourage good
practice” (Emma Wagner 1999). The training element of their campaign has been taken over by
professional teachers as part of the Commission´s official training programme. Their biggest concern
is “dealing with authors who are not natural writers, or are not native speakers of English, and who
have been educated in very diverse national systems, some of which discourage concision and clarity
because it seems ´uneducated´” (Emma Wagner 1999). Fight the Fog have plenty of unofficial co-
operation too. It is a grass-roots campaign with no budget and no public relations element. The
translators are working from within the organisation of the Committee and combining the campaign
with other full-time jobs (Emma Wagner 1999).
There is still no international co-ordination of style guides for translators in the European Union. The
head of the Swedish Translators´ Commission, Kenneth Larsson, suggests in his official report
Svenskan i EU that a central co-ordination of the Swedish translation service must be organised
(SOU1998: 10). An international co-ordination is also needed. He is worried that other traditions, more
influenced by officialese (mostly the Roman rhetorical texts), may influence even the Swedish texts
which are said to be more influenced by plain language than many others.
Discussion of the similarities and differences
One is immediately struck by the similarities between the different countries. Sweden and the UK
seem to share many historical experiences in the fight for a clearer official language. However, there
are primarily two differences between the English speaking world and Sweden, which are interesting
to point at, namely testing and economic benefit.
Testing has been used by most organisations following plain English programmes both in the UK,
the United States and Australia, but not in Sweden to the same extent. In the English speaking
countries it has been ”undertaken at all stages in the development of a document, from the initial
planning of policy to the final draft and on to the finished version in use” (Steinberg 1991: 37). Many of
them stress the economic benefit they get from the change to plain language, something, which is
regularly tested too.

The style guides

The following advice is found in all the style guides of the study:
1) Length
• be short but concise; omit surplus words
2) Words
• use words the readers are likely to understand, i.e. familiar words
• use concrete words; avoid abstractions
3) Syntax
• over the whole document, make the average sentence length 15-20 words (one says 25 at the
most)
• name the agent
• put old information at the beginning and new or complex information at the end
• avoid gaps between the subject, the verb and the object
4) Tone
• use base verbs; not nominalizations
• use the active voice; not the passive
• avoid sexist language
• involve the reader by using ”you” or other personal pronouns
5) Planning
• organise the material in a way that helps the reader to grasp the important information early
and to navigate through the document easily
6) Layout
• use headings
• use bullet and other indented lists
All the style guides have a reader perspective, i.e. the texts should always be written with the reader in
mind. After having decided on whom the readers are the writer then can decide on words and syntax.
To no surprise all guides also recommended to use accurate punctuation.

Discussion of the similarities and differences

The similarities struck me at once and I was quite surprised noticing that they were so many. There
were no differences at all between the American and the British ones. It was almost as if they had
influenced each other from beginning to end. Of course they varied individually depending on who the
intended reader was – for example one style guide was intended for lawyers and contained many
examples from that field. Furthermore, there were hardly any differences between the style guides
from the English speaking world and the Swedish ones. However, I did find some differences.
For example the use of non-sexist language is something which is stressed in the English style
guides and cannot be found in the Swedish ones. Swedish style guides recommended no exact
sentence length, but advised variation between long and short sentences.
However, most of the advice given in all the English style guides corresponds extremely well with
the Swedish ones. It is almost like the very same method is practised over and over again. I was quite
surprised in realising this. I thought the style guides would stress different things since the languages
do show some differences in information structure.

The texts
I have chosen to study three different aspects of the style guides here. These are: sentence length,
voice and information structure of the themes.

Sentence length
Most of the Swedish sentences begin with a capital and end with a full stop. The Swedish sentences
are shorter in general. Most of the English sentences begin with a capital letter, but some, including
dotted enumeration, begin with a small letter (here the Swedish text begin with a capital letter) and all
are ended by a full stop, a semicolon or a colon. The English tradition seems to be dividing main
clauses by semicolons and colons more often whereas the Swedish tradition seems to be completely
new sentences. Semicolons are very rare in the Swedish texts.
This does not mean that all the English sentences are longer than the Swedish ones. Instead En 1
contains the shortest average sentence length of all the texts. This might be a sign of conscious
language choice in line with the plain language style guides. Sw 1 also normally contains short
sentences, which might imply the same. The EU-texts (Sw 2 and En 2) both contain longer sentences
in comparison.
Table 1. shows the average sentence length for the four different texts.
Sw 1 Sw 2 En 1 En 2
17.4 19 17.2 22.4
Table 1. The average length of the sentences (by number of words).

According to the style guides the average sentence length should be 15-20 words. The only text that
violates this is En 2.

Voice
According to Cutts every writer should prefer “the active voice unless there´s a good reason for using
the passive. Certainly the active tends to make the writing tighter, more personal, and introduces
action earlier in sentences. The passive tends to do the reverse /.../ “ (Cutts 1995: 48).

Sw 1 Sw 2 En 1 En 2
Active (158) 54% (162) 83% (324) 91% (149) 85%
Passive (136) 46% (33) 17% (34) 9% (27) 15%
Total 294 195 358 176
Table 2. The number of sentences in the active and the passive voice
respectively.

Table 2. shows that the active voice is by far the most favoured in all the texts, but the least so in Sw
1. The reason for this might be that it is a legal text impersonal in its tone. It is very difficult to address
a reader and the norm is probably not to do so at all. The text often refers to other texts and it seems
to be the easiest way out to use the passive voice then; instead of making a document the subject of
the sentence the passive voice is preferred.
The English texts have far more active sentences than the Swedish ones; En 1 containing most
active voice sentences in comparison. This might be a conscious choice of the writer of the text; one
can of course only speculate about the reason. It is interesting to note that this text also is the most
reader-oriented text of them all.

Information structure in sentences with passive voice

Why choose the passive voice at all? Most of the style guides accept passive constructions if the
writer has a good reason for doing so. The major reason that many of the style guides propose is to
make the agent invisible, but I also think there are several other just as justified reasons. I will here
look at four possible reasons that I find relevant in these texts. They are: light themes, invisible agents,
fronting for emphasis and linking to preceding sentences.

Light themes
Sentences both in English and Swedish are usually arranged so that it is easy for the reader to
process through the text. It is easier if the verb is not kept too far back. A light theme saves most of the
information in the rhematic part of the sentence, which leads to end weight. End weight is much
favoured both in English and Swedish. To create this easy access to the information, writers use
different strategies to make the themes light and move most of the new information into the rhematic
part of the sentence (Bülow-Möller 1989: 154).
Many of the style guides seem to be aware of this and a frequent piece of advice is to use the
passive if for example “you want to position old or known information at the start of the sentence, so
you can put new or surprising information at the end” (Fight the Fog). Of course this strategy also
leads to light themes, as we can see table 3.

Sw 1 3.3
Sw 2 4.3
En 1 4.5
En 2 6
Table 3. Average length of themes (by number of words).

As a comparison I counted all the themes in two of the texts – Sw2 and En 2 – and found the following
information. In average the English text contains 5.3 words per theme and the Swedish text 4.4 words.
This might imply that the passive voice does not in itself lead to light themes – the difference between
the themes in the passive sentences and the themes in the active ones is not significant. According to
Melin and Lange (Melin 1995: 160) the average length of themes in Swedish newspapers is 3.4 in
cultural articles and 3.9 in editorials. The themes of Sw 1 must then be regarded as very short since
the legal language must be regarded as even more formal than newspaper editorials and therefore
should be expected to contain longer themes. Perhaps this is a conscious choice of the language
experts who checked it.
A conclusion seems to be that light themes might be a reason for choosing the passive voice in Sw
1, but not so in the other three texts.

The great majority of the themes are subjects and most of them nouns, the rest are adverbials. There
is an interesting difference between the languages here. English favours subjects more than Swedish.
In my study I found that the two English texts contained between 89% and 91% subjects in theme
position whereas the two Swedish texts contained between 70% and 76% subjects. On the other
hand, Swedish has more possibilities for moving the constituents in the themes and thus favours
adverbials more than English. The Swedish texts contained between 24% and 30% adverbials
whereas the English texts contained between 9% and 11% adverbials. This can be seen in table 4.

Subjects Adverbials Total

Sw 1 (94) 70% (40) 30% 134
Sw 2 (25) 76% (8) 24% 33
En 1 (30) 91% (3) 9% 33
En 2 (24) 89% (3) 11% 27
Table 4. The different types of themes by number of tokens
and in percentages.
Invisible agent
Another reason for choosing certain themes is to make the agent invisible. In this case the writer may
leave out the subject and transform the sentence into the passive. (Bülow-Möller 1989: 156) This
might be the reason for sentences like ex. (3) and (4):

(3) The name ´euro´ adopted for the single currency. (En 2)

(4) Författningskommentaren är indelad i kapitel”. (Sw 1)

This is probably a predominant reason why Sw 1 has such a high score of passive voice sentences.

Fronting for emphasis

Still another reason for choosing a theme might be to have the point of interest fronted for emphasis.
This reason might be why Sw 2 has fronted the adverbial on the expense of the subject as in the
English equivalent. Swedish seems to have more possibilities of choosing adverbials as themes than
English (something, which will be discussed further on).

(5) När det gäller de övriga konvergenskriterierna har situationen förbättrats betydligt de
senaste åren, särskilt avseende inflationen, som nu ligger på en historiskt sett rekordlåg nivå
(2,7 % för hela unionen 1996), och de långfristiga räntorna. (Sw 2)

(6) The situation as regards the other convergence criteria has improved considerably in
recent years, especially for inflation, which has hit record lows (2.7% for the Union as a
whole in 1996), and for long-term interest rates. (En 2)

Linking the sentences together

Another reason for choosing certain themes is of course to link the sentence to the one before (or
earlier in the text). This is something which probably makes the text more fluent and therefore more
accessible to the reader. Surprisingly enough this is not something, which a majority of the style
guides advocate.
Tables 5-8 show that a majority of the themes in the passive voice sentences are linked to
sentences that preceded them in the text.

Subject Adverbial Total Percentage

Noun Anticip. It Pronoun

Linked 44 0 9 15 68 51%
Unlinked 41 2 0 25 65 49%
Total 85 2 9 40 134
Table 5. Themes in Sw 1.
 Subject Adverbial Total Percentage
Noun Pronoun

Linked 16 1 2 19 58%
Unlinked 8 0 6 14 0%
Total 24 1 8 33
Table 6. Themes in Sw 2.

Subject Adverbial Total Percentage

Noun Pronoun
Linked 17 8 1 26 79%
Unlinked 5 0 2 7 21%
Total 22 8 3 33
Table 7. Themes in En 1.

Subject Adverbial Total Percentage

Noun Pronoun
Linked 18 0 2 20 74%
Unlinked 6 0 1 7 0%
Total 24 0 3 27
Table 8. Themes in En 2.

The English texts show a remarkably higher score of linked themes compared to the Swedish ones;
there is a difference of more than 20%. Again En 1 reaches the highest score of linked themes;
something which might imply a conscious choice of the writer in order to make the text more
accessible to the reader.
The Swedish texts only have a slight majority of linked themes. Sw 1 shows a considerably high
amount of unlinked themes compared to the other texts; a reason for this may again be the text type.
The legal text is not as reader oriented as the other texts. It also contains heavily packed information,
i.e. it introduces new information in almost every sentence, something which might make linking more
difficult.
Different languages do not have the same possibilities for moving elements about in the sentence in
order to make use of those strategies. (Bülow-Möller 1989: 159). English and Swedish use almost the
same strategies, but Swedish seems to have somewhat more possibilities for moving the constituents.
This can be seen in the high frequency of adverbs in the Swedish themes where the English themes
favours nouns and pronouns. Another difference between the languages might be the linking of
themes. Perhaps linking is more frequent and important in English than in Swedish.

Conclusion
My aim for this study was threefold. First I wanted to make a general comparison of the plain language
movements of the USA, the UK and Sweden and of their style guides. Then I wanted to be more
specific and investigate whether the plain language texts and the EU-texts violated the guide lines
common of all the countries. I was especially interested in the texts written or checked by plain
language experts of course.
I found mostly similarities in my comparison both of the plain language movements in the different
countries and of the style guides. This was very surprising since the countries do have different
historical backgrounds and different languages with different information structure. Some differences
have been found and discussed, but the predominant conclusion is that the similarities prevail.
The most difficult question is the last one: do the texts violate the style guides? Or, as Cutts puts it
“Is the style appropriate for the audience, with a good average sentence length /and/ plenty of active
voice verbs”? I would say that the four texts in the study do not violate the style guides in any striking
way, and have, thus, chosen an appropriate style according to Cutts. The average sentence length is
within the recommended limits and there are plenty of active verbs.
On the other hand, the passive voice is quite common in all the texts (with some individual variation,
probably mainly because of the different text types). And the reason for this is that it helps the writers
use an information structure of the themes which probably helps the readers in their turn to navigate
more easily through the text.
A hypothesis for this study was that, since most style guides advocate active verbs, not passive, at
the same time as they advocate putting old information at the start of the sentence and new and
surprising information at the end, I would find a clash here. I suspected that these two pieces of advice
did not correspond, because if you want to keep new information for the end you often have to make
the verb passive. I might have been able to prove this hypothesis in this study; linking is definitively a
stronger incentive for the passive voice than the other reasons that I investigated in the study: light
themes, invisible agents and fronting for emphasis.
This study has hence shown that one major reason for using the passive voice is to link
the theme with rhemes preceding it. Perhaps this linking strategy is a more important tool than using
the active voice, in order to help the reader through the text. I think this is something many of the style
guides fail to mention.
An interesting find was that the number of linked themes were strikingly more frequent
in the English texts. This might be the result of a conscious choice of information strategy by the
writers but it might also be a sign of a special quality favoured in English, i. e. a difference between
English and Swedish. This is not possible to tell from a small study like this one, but it would be
interesting to investigate further. If English texts in general are better linked than Swedish texts, this
might explain why the English style guides do not stress linking as an important writing device. But on
the other hand, it does not explain why Swedish style guides do not stress it more.
I have not found other studies on this topic, except for the American study at the University of
Pittsburg.
The material I used had great limitations especially concerning the different text types. If I were to do
this again I would compare texts that all were informative, intended for the public and adjusted to them
written by plain language experts. As complementary texts I would use the same text types intended
for the public but not written by plain language experts. In that way I would see more clearly how
consciously the plain language experts worked with different language choices in order to make the
texts more reader oriented. It would also become clearer whether they violated the style guides or not.
Nevertheless, the result I have found is interesting and it would be even more interesting to
investigate it further. There are probably other violations of other guidelines and there are probably
interesting reasons why. I believe critical studies of this kind are important for the development of plain
language and for all the plain language experts working and participating in the practical language
work in offices all over the world. Only copying style recommendations without careful and critical
consideration is dangerous, and it does not make texts better.
This may even become dangerous, as automatic computer programmes obviously are influenced by
guidelines advocating active voice on behalf of passive voice. The grammar check in Word 97 corrects
“ungrammatical sentences”, i.e. sentences in the passive voice, to “grammatically correct sentences”,
i.e. sentences in the active voice. From my essay a sentence such as The first plain English insurance
policy was produced by General Accident Fire and Life Assurance Corporation was corrected into
General Accident Fire and Life Assurance Corporation produced the first plain English insurance
policy. Had I not been so stubbornly determined to keep the passive voice I would probably have
yielded to the computer expertise and changed the sentence, which would have resulted in lost linking
and emphasis.
References
Adler, Mark, founder of Clarity (telephone interview on 8 November 1999)

Bailey, Edward P. Jr: Plain English at Work (Oxford University Press, New York, Oxford, 1996)

Bulow-Möller: The Textlinguistic Omnibus: A Survey of methods for analysis (Handelshojskolens

Forlag, Köpenhamn, 1989)

Cutts Martin: The Plain English Guide (Oxford University Press, Oxford, New York, 1995)

Cutts Martin, founder of Plain Language Committee (e-mail interview on 23 November 1999)

Ehrenberg-Sundin, Barbro m.fl: Att skriva bättre i jobbet (Norstedts Juridik AB, Stockholm, 1999)

Ehrenberg-Sundin, Barbro: “Kampen mot gobbledygook” (language column in Klaraposten,

Regeringskansliet, 1993)

English Style Guide: English units of the Commission's Translation Service

http://europa.eu.int/comm/sdt/en/stygd/index.htm

Fight the Fog

http://europa.eu.int/comm/sdt/en/ftfog/index.htm

How to Write Clearly by Fight the Fog (European Commission, Luxenbourg, 1997)

Klarspråksgruppen in Sweden: Plain language work in Sweden

http://www.justitie.regeringen.se/klarsprak/english/english.htm

Melin , Lars och Lange, Sven: Att analysera text (Studentlitteratur, Lund, 1995)

Plain English Campaign

http://www.plainenglish.co.uk/press.html

Plain Language Commission

http://www.plainlanguage.demon.co.uk/sitein.htm

President Clinton´s memorandum on plain language

http://www.plainlanguage.gov/cites/memo.htm

Sundin, Håkan och Ehrenberg-Sundin, Barbro: Tänk först – skriv sedan. Effektivare rapporter och PM
(Effektiv Kommunikation AB, Stockholm, 1998)

Steinberg , Erwin R.(ed.): Plain Language. Principles and Practice (Wayne State University Press,
Detroit, 1991)

Wagner, Emma, one of the founders of Fight the Fog (e-mail interview on 16 November 1999)

Wydick, Richard C.: Plain English for lawyers (Carolina Academic Press, Durham, North Carolina,
1994)

”Speketh so pleyne at this time, I yow preye

That we may understonde what ye seye.”
th
(Chaucer 14 century)