Communication barriers in public discourse Document Design 4(1), 22–41
© 2003 by John Benjamins Publishing Co.

23

Inger Askehave and Karen Korning Zethsen Communication barriers in public discourse
The patient package insert

Keywords: Public communication, asymmetrical communication, target group, genre, patient package inserts The production of expert-to-layperson documents in private and public companies is an area of research that is gradually gaining ground in research communities throughout the world. This article discusses the nature of public communication written by experts or semiexperts and aimed at a large and heterogeneous audience, often potentially the entire population of a country. The article analyzes common barriers to this kind of communication, and considers topics such as the implications of very broad target groups, expertto-layperson communication and the legislative introduction of mandatory genres as opposed to genres which have evolved naturally over time. The case of the patient package insert is outlined as a prototypical example of ‘public communication’ su¬ering from the problems brought about by the existence of these common barriers.

Introduction
Public documents have long been considered problematic and have attracted the interest of document designers and discourse analysts, not to mention entire movements such as the Plain English Movement. Numerous organizations, committees, and boards have been set up at national and international level to criticize and ultimately improve the

readability of public documents. The health industry, for example, has been heavily criticized for its “hard-to-understand directions in tiny print” (Bresler, 2000) (the so-called ‘patient package inserts’) and the European Commission, consumers’ associations in member countries, and discourse analysts have gone to great lengths to provide suggestions for improving these directions (Davis, 1999; Eckkrammer, 1997; Lind, 1997; Kitching, 1990; Sless & Wiseman, 1997). However, despite all the good intentions, complex and ambiguous texts still abound in public communication.1 One of the reasons for this unfortunate tendency may be that discussions and research have focused on ways to improve speci¼c public documents (such as the patient package insert), whereas less has been done to examine whether the nature of public communication itself may be the real cause of the problem. The aim of this article is, therefore, to present a theoretical discussion of the concept of public communication, and then, for illustrative purposes, turn to an analysis of the patient package insert as an instance of public communication. The article is divided into two main parts. In the ¼rst part we provide a tentative de¼nition of public communication and introduce the basic characteristics of this particular type of communication. While this de¼nition serves to broaden our understanding of public communication in general, it also brings to light three speci¼c characteristics which we suggest constitute signi¼cant communication barriers in public document writing,

24

Inger Askehave and Karen Korning Zethsen

namely (1) a broad target group, (2) an asymmetrical sender-receiver relationship, and (3) mandatory genres. In the second part, we turn to the patient package inserts as a prototypical example of public communication, and illustrate how the common barriers suggested above manifest themselves in the case of these inserts.

f. Genre-mediated communication g. Regulatory a~nity h. Functional genres re a. Like mass communication, public communication is designed to reach an extremely large part of the population. The size of the audience varies of course from document to document. For example in the case of tax forms, the potential receivers are all Danes with an annual income exceeding 20,000 DKK, whereas the number of people receiving a parking ¼ne is considerably lower (though it should be kept in mind that all Danes with a driving license constitute a potential target group). re b. As a natural consequence of large target groups (potentially the entire adult population in a country), the audience in public communication is extremely heterogeneous. For example, young and old, well-educated and uneducated people, experts and nonexperts all form part of the potential target group. re c. The sender-receiver relationship in public communication is often extremely asymmetrical. Generally speaking, the public documents convey information of a rather complex nature (often referring to economic or legal matters) which cannot be considered common knowledge to the general public. The result is that the sender is the expert within the specialized domain and the receiver is more often than not the novice. re d. Another consequence of the large target group is the often impersonal and anonymous nature of the communication where the receiver is unknown to the reader. We may, however, ¼nd pseudopersonal letters within public communication where for example a standard letter from a pension fund is personalized by means of a salutation at the beginning of the letter.

Public communication
The characteristics of public communication In order to describe the communication barriers within public communication, we obviously need to address the nature of the communication taking place and de¼ne the characteristics of public communication. Public communication occurs when a company or an organization communicates with the general public. The genres used in the communication situation vary considerably — standard letters (from local government, insurance companies, pension funds, unemployment funds, etc.), forms (e.g., insurance policies, tax forms), and instructions (e.g., package inserts, manuals for operating machinery), to name but a few. It would be fair to say that to a certain extent public communication corresponds to mass communication where professional communicators use the mass media to disseminate messages to large-scale audiences. However, de¼nitions of mass communication are not su~cient in themselves for capturing the essence of public communication.2 In our de¼nition of the characteristics of public communication we propose the following 8 parameters where (a), (b), (c), (d), and (e) overlap to a certain degree with mass communication and (f), (g), and (h) can be regarded as extensions of the usual de¼nitions of mass communication: a. b. c. d. e. Large-scale distribution and reception Heterogeneous audience Asymmetrical sender-receiver relationship Impersonal and anonymous interaction Sender-controlled interaction

Communication barriers in public discourse

25

re e. The communication ½ow within public communication is, if not one-sided as in mass communication, then at least sender-controlled. In most cases interaction between sender and receiver is practically nonexistent; the sender communicates with the receiver and not vice versa. However, an interesting feature, which distinguishes public communication from mass communication, is the sendercontrolled feedback, which is required in connection with some kinds of public document. For example, taxpayers ¼ll out their tax forms and return them to the tax o~ces. Nevertheless, the interaction is still controlled by the tax o~ce because the taxpayers are not supposed to compose a new document themselves, but are merely asked to ¼ll out or sign the forms before returning them to the original sender. re f. As suggested above, mass communication is disseminated through the so-called mass media such as radio, television, magazines, books, and newspapers. However, this is generally not the case when it comes to public communication. The channel or medium through which the communication is transmitted is often, though far from always, the genre itself (a letter, a form, an instruction manual, etc.). We refer to this phenomenon as genre-mediated communication. re g. Genres within public communication often come into being as a result of legislation. For example, in Denmark we have a law specifying that to become entitled to unemployment bene¼ts, you need to become a member of the unemployment bene¼t fund. This process usually involves a number of informational lea½ets and forms, which need to be read, ¼lled out, and returned to the fund. As we shall see in Sections 2.2.3 and 3.3.3 this regulatory a~nity means that many public documents are designed by bureaucrats who focus more on the information they are obliged to convey in order to meet legal demands than on trying to meet the needs of the audience.

re h. The main di¬erence between public and mass communication lies, however, within the use of the information received. The purpose of public communication is ‘functional’ in the sense that the receiver (the general public) is not only informed about something but is also often supposed to use the document to perform some kind of action. For example, on the basis of a parking ¼ne notice, the receiver is supposed to pay a ¼ne. On the basis of the tax form, the receiver is asked to pay taxes, and with the instruction lea½et or manual the receiver learns how to perform an action, such as operating a machine or taking medication. This functionality is not only an important di¬erence between public and mass communication but is also the very parameter that justi¼es a thorough communicative investigation of public documents. Public documents are not ‘indi¬erent’ texts, they are prescriptive texts and ought to be understandable to the public who reads them and acts on the basis of them. Having discussed the concept of public communication in general we shall now turn to the speci¼c parameters which we suggest may constitute common barriers to successful public communication. Public communication, such as standard letters, tax return formulas, health campaigns, parking ¼nes, insurance policies, information on investments or holidays earned, and many more, has since its invention constantly been under ¼re. With regular intervals public communication debates ½are up in newspapers and on television, but though public language has of course to a certain extent changed with the times, no one would claim that it has improved drastically during, say, the past 20 years. But what exactly is so problematic about public communication? Most people do possess satisfactory basic communication skills that are employed successfully in their private lives; what goes wrong when they communicate in their o~cial capacities? In the following we will attempt to outline some of the main factors that we think are relevant in an answer to these questions, namely:

26

Inger Askehave and Karen Korning Zethsen

A. Nature of target group B. Asymmetrical sender-receiver relationship C. Mandatory genres Common barriers to public communication 1. Nature of target group When writing to a well-known, limited target group it is common to visualize your audience — in the most private kinds of communication, for example when writing to a friend, most people would probably agree that sometimes they feel that they are talking to the intended receiver. In the case of public communication the target group is often so broad that it may be extremely di~cult for the writer to employ the visualization technique and writers may therefore try to create a more limited target group for whom they are writing: “It is often noted in the communication literature that communicators use audiences other than the primary, actual one” (McQuail in Windahl & Signitzer 1992, p. 131). The problem is that this, arti¼cially created, audience (the so-called secondary audience, ibid., p. 131) often consists of people close to the writer — spouse, colleagues, or neighbours. This is often unfortunate for though these people are of course members of the public they are not necessarily representative of the population as a whole. The following quote is about mass communicators, but no doubt applies to public communicators as well:
Since most mass communicators belong to the middle class and associate with other middle-class people, you could assume that communication will have a middle-class bias. This tendency may have negative consequences for the effectiveness of communication in situations where the communicator-receiver relationship is heterophilous (ibid., p. 129).

trating on parts of the audience whose needs and interests are known”. Other approaches may be based on the public communicators’ more or less haphazard or imaginary visualizations of their audience. The communicator may be negatively inclined in relation to his or her audience (disillusioned for some reason) and give up in advance — “they won’t understand it anyway so why bother”. Conversely, the communicator may be positively inclined and may convince him- or herself that people are more knowledgeable than is the case. Apart from the various audience visualization strategies there is always the risk of not concentrating on any audience: the communication becomes sender-oriented instead of receiver-oriented, which may be even worse as focus will then be completely on the needs and knowledge of the sender, who in this kind of communication is very likely not representative of the audience (see the following section on the asymmetrical sender-receiver relationship of public communication). Finally, for the public communicator a solution to the broad nature of the audience could be to write for the lowest common denominator. This would ensure the best possible understanding of the documents, but has other problems: it may be considered patronizing by some parts of the audience and it may take up more space for example. To sum up, many public communicators probably rely on one of the following strategies: Possible strategies for a public communicator: 1. strategy based on the creation of a secondary audience – of personal friends, family, and colleagues – of a fraction of the audience (other than friends and family) whose needs and interests happen to be known 2. strategy based on disillusion 3. strategy based on optimism 4. strategy based on lowest common denominator If we consider our previous discussion and subsequent de¼nition of public communication, in which we hold that public documents are highly functional and ought to be understandable to the public who reads and acts on them, then choice 4, lowest common denominator, de¼nitely seems the best

On page 131 of the same book, McQuail mentions four consequences of insu~cient audience contact — a situation that may be relevant in the discussion of the problems of an extremely broad audience. One of the consequences supports the above claim, namely “the process of concen-

Communication barriers in public discourse

27

choice available (see further discussion in connection with our case study in the section called ‘The target group of the patient package insert’). 2. Asymmetrical sender-receiver relationship One of the important characteristics of public communication is the asymmetrical relation that exists between communicators and audiences in terms of knowledge and expertise. The communicators are often experts in the subject matter of the message (e.g., taxation, pension schemes), whereas the audience consists of a large heterogeneous group often referred to as laypeople or nonexperts, whose knowledge of the subject matter in question is generally rather limited. Studies show that this asymmetry between the interactants impedes the realization of successful communication (see, for example, Janssen & Neutelings, 2001). To explain why the asymmetrical relation between communicators and audiences often poses communication problems we should like to turn to the concept of discourse communities. The way people communicate varies considerably depending on such questions as who they are, where they are, what they do, and whom they are talking to. Researchers have tried to systematize and account for these variations by suggesting that individuals are socialized to speci¼c discursive practices according to their membership in di¬erent discourse communities, which in turn make use of particular kinds of discourse:
Discourse communities are socio-rhetorical networks that form in order to work toward sets of common goals. One of the characteristics that established members of these discourse communities possess is familiarity with the particular genres that are used in the communicative furtherance of those sets of goals. In consequence, genres are the properties of discourse communities; that is to say, genres belong to discourse communities, not individuals, other kinds of grouping, or to wider speech communities. (Swales, 1990, p. 9)

within the community. For example, when drawing up a legal document for clients with the purpose of providing for their children after the clients have passed away, lawyers do not have to ‘reinvent’ a particular text suitable for this purpose but merely choose an already established genre, the will, and use the appropriate linguistic conventions that apply to this particular type of document. Furthermore, they use specialist terms like ‘devise’ and ‘bequeath’, which may require a de¼nition or paraphrase outside the discourse community, but which are quite appropriate here since the meaning of specialist terms tends to be universally understood and accepted by the other members of the discourse community. Thus the advantages of rhetorical strategies especially designed for particular communities are obvious: they ensure e~ciency, economy, and precision when communicating with peers. However, there are also disadvantages associated with being trained within a particular discourse community, not least when the need to communicate with outsiders arises. Public servants in tax departments, for example, are bound to know more about taxation than the general public they address in their documents. However, due to the close relationship between expert knowledge and expert ‘language’, communication barriers often appear when the public servants have to write documents for taxpayers whose linguistic and factual competencies within the ¼eld of taxation are considerably poorer. This may be due to the fact that the communicators are unaware of the need to use a di¬erent communication strategy when addressing the general public simply because public servants are not professional writers and public institutions seldom employ language specialists who can ‘translate’ specialist discourse. Nevertheless, when people uncritically transfer the conventions of one discourse community to other discourse communities (with no membership overlap) comprehension problems are bound to appear.3 3. Mandatory genres As appeared in the previous section, discourse communities not only produce and consume discourse, they also prefer

The fact that communities generate and own particular discourses helps promote quick and e¬ective communication

28

Inger Askehave and Karen Korning Zethsen

certain kinds of discourse (Killingsworth & Gilbertson, 1992, p. 165). Following Swales (1990) we may refer to these kinds of discourse as ‘genres’. According to Orlikowski and Yates (1994, p. 542) genres are “socially recognized types of communicative actions … that are habitually enacted by members of a community to realize particular social purposes”. Genres are dynamic rhetorical forms that constantly change in response to the needs of individual users. This means that genres may change in terms of content and/or form and some genres fade out while new ones emerge (Berkenkotter & Huckin, 1995, p. 6). Typically a new genre emerges because members of a particular discourse community discover a need to realize new social purposes. For example, a few years ago teachers in primary and secondary schools in Denmark found a need to inform parents about the ongoing activities in class. For this particular purpose a particular type of newsletter emerged which has gradually found its form and has become an important genre (or tool) in teacher — parent communication. However, a genre may also be introduced by external mandate (Orlikowski & Yates, 1994). For example, EU or national legislation may require the introduction of certain documents in member countries. This in itself is not problematic as long as the new document is developed in more or less the same way as genres invoked by tradition — adopting the rhetorical forms appropriate for its purpose and audience. Unfortunately, genres required by law usually place severe constraints on the people producing the documents. For example, public communicators are not only asked to adopt a particular genre but are also required to follow particular guidelines concerned with format and other aspects speci¼ed in a directive. And as the legislators who draw up the guidelines have been trained within a specialist discourse community, the guidelines will often result in a scienti¼c, formal, and objective style in spite of the fact that these rhetorical forms are inappropriate in the context in which the documents are to be used. The result is that the public communicators produce a document that adheres to the rules and communication patterns stipulated in a legal directive, and the main concern (and covert purpose) of the

communicators becomes that of observing the legal requirements. This in turn makes it very problematic for the mandatory genre to appear as a useful or purposeful genre in its proper environment. The tone and style of the document preserve a social identi¼cation with the community of experts, which means that the document fails to achieve its proper goal in the public context (for example, to inform the general public about pension schemes). In other words, the use of mandatory genres in public communication may also be regarded as a constraint on communication and an important reason why communication may fail. Miller, 1981 (in Killingsworth & Ste¬ens, 1989) goes so far as to suggest that genres cannot be legislated but must evolve from traditional paradigms for social action. She bases this claim on her study of environmental impact statements where three di¬erent traditions were ¼ghting for attention — they all wanted to leave their linguistic mark on the genre. She claimed that “the imperfect fusion of scienti¼c, legal and administrative elements prevented interpretation of the documents as meaningful rhetorical action” (Miller, 1984, p.164). However, as we shall see further on in this article, we believe mandatory genres may be just as e~cient for realizing social purpose as naturally evolved genres provided of course that they have been designed to meet the needs of their audience.

‘The PPI of the Future’
For illustrative purposes we shall now move to a prototypical example of public communication: the patient package insert. Patient package inserts are written to be read by, potentially, the entire Danish population, and many of the di~culties connected with PPIs are presumably to some extent related to this fact. In this respect PPIs share the same fate as public communication in general, and this a~nity with public communication is further brought home by the fact that the PPI genre came into being as the result of legislation (as will be discussed below) — in other words the

Communication barriers in public discourse

29

genre has been designed by bureaucrats and has in its present form not been allowed to evolve over time. How “The PPI of the Future” came into being — Methodological considerations Here we shall draw on the results of a research project that we designed in 1999 when it came to our attention that patient package inserts were considered a huge problem by the Danish medical sector. As the result of harmonized EU legislation, package inserts became a legal requirement in Denmark in 1995 and the law speci¼es that the package inserts must be written in clear and understandable language as they are meant to help and protect the consumer. The problem was that none of the players in the Danish medical industry (including the medical companies who are responsible for writing the inserts) were satis¼ed with the resulting package inserts, but at the same time they were not able to pinpoint exactly what was wrong apart from the rather vague conclusion that the package inserts were simply too di~cult for the consumers to understand. In this part of the article, we shall use our general discussion of public communication characteristics to suggest possible reasons why inserts are ‘di~cult for the consumers to understand’. Realizing that user-friendliness, or rather the lack of user-friendliness, in PPIs constituted a serious problem acknowledged by all relevant parties, we wanted to get to know the medical industry and to try to understand all the factors that play a role in connection with the PPI. This was done in the following way: – Individual meetings with the employees who are in charge of writing the PPIs at four major medical companies, and subsequent contacts with some of the employees. – Meeting with the Danish Medicines Agency (part of the Danish Ministry of Health), who are in charge of approving certain PPIs and who lay down the standard terminology and wording to be used in PPIs. – Talk and discussion at a meeting of a special committee on PPIs with representatives from more than ten major

Danish medical companies at the Danish Association of Medical Companies (LIF), and subsequent contact with LIF. – Numerous telephone conversations with other players in the ¼eld, such as the Danish Consumer Agency, a pharmacist who experiences the problems with PPIs on a day-to-day basis and who often writes about the problems in the media, and others. – Reading of relevant legislation in order to understand the legal intentions, requirements, and approval procedures; reading of relevant scienti¼c as well as popular articles. – Completion of linguistic pilot analyses of a number of PPIs. Once we were in possession of enough background knowledge to be able to draw up a project targeted at improving the Danish PPI we wrote to the Danish Ministry of Health, which agreed to sponsor our project. We then moved on to more speci¼c questions that needed an answer. The main aim of our research project was to ¼nd out whether the Danish inserts comply with the requirements of Council Directive 92/27/EEC which states that the inserts must be written in “clear and understandable terms” (article 8) — in other words, whether the package inserts are in fact as userfriendly as the law prescribes. We chose to divide the project into three interlinked activities: A. To establish whether the criticism of the present PPIs is legitimate. On the basis of our background studies we obviously expected to ¼nd problems, but activity A also included pinpointing the criticisms, that is, ¼nding out exactly where the problems were instead of merely repeating numerous other sources which generally just claim that the PPIs are di~cult to read and understand. B. To come up with some of the basic reasons why inserts may be di~cult to read and understand for some people. C. To give examples of linguistic pitfalls in package inserts and to make speci¼c recommendations for their improvement.

30

Inger Askehave and Karen Korning Zethsen

In this paper all three elements are discussed, though speci¼c recommendations are left out. In order to answer the questions inherent in A, B, and C we decided on the following methods: We aimed at answering Question A partly by analyzing a corpus of Danish PPIs and letting the texts speak for themselves and partly by ¼nding out what consumers think by means of a questionnaire study. In order to answer Question B, which is very broad indeed, we had to draw on many di¬erent kinds of methods to obtain the relevant information. On the one hand it involved linguistic analyses of a large number of Danish PPIs in order to establish which linguistic features (and in a few cases nonlinguistic features) had a negative impact on the user-friendliness of PPIs. On the other hand we had to analyze legislative requirements, the educational background of the writers of PPIs and the relevant employees at the approving authorities, the stages and conditions pertaining to the creation of a PPI, translational aspects, the relationship between the medical companies and the approving authorities, and more. Do we have a problem? — The results of our questionnaire4 When designing the questionnaire we took great care to ensure that the questions would actually provide answers to the questions of the project. We came up with a simple (both as regards layout and language), informal (and user-friendly, we hope) two-page questionnaire (see Askehave & Zethsen, 2000a, p. 8–9) that addressed the customer directly. To ensure that the questionnaire study would be statistically signi¼cant, we had 1000 questionnaires distributed among 10 geographically diverse pharmacies (¼ve city pharmacies and ¼ve village pharmacies). The pharmacies were chosen at random within a given area. All 10 pharmacies were contacted in writing followed by a telephone call and were provided with a set of instructions aimed at ensuring neutral assistance when the questionnaires were handed out. The customers were not supposed to take the questionnaire home and if someone declined to participate, the questionnaire was to be o¬ered to the customer next in line.

We received 688 completed questionnaires, a response rate of almost 70%. The pharmacies were again contacted by phone and interviewed about customer reactions, etc. Surprisingly, only a very few customers declined to participate in the study and most customers were generally very interested in the subject. If we take into account the number of people reported as having turned down the questionnaire we arrive at a more realistic response rate of approximately 65%, all in all a very satisfactory result compared to the average response rate of questionnaire studies. That the PPI is being read, not least when new medication is involved,5 was proved beyond doubt. Of the respondents, 81% answered that they always or often read the PPI. Whether the package insert is also understood is more doubtful. According to the questionnaires, 18% think that they themselves need more user-friendly PPIs and 53% think that other people need more user-friendly PPIs. This probably means that more than half the respondents can think of one or more persons who would not understand the PPIs as they are now. A very interesting result is that 72% of the respondents who were employed within the health sector6 say that the PPIs ought to be more user-friendly. 1. Weaknesses of the questionnaire study Naturally, questionnaire studies also su¬er from methodological ½aws. In the present case it goes without saying that we did not reach the weakest members of society. Presumably a large group of elderly people do not collect their own medication and if you are a functional illiterate you would most likely do anything to avoid a questionnaire.7 Consequently the questionnaires showed, not surprisingly, that the study as a whole had too few elderly participants compared to the population as a whole and that the participants were far better educated than the national average. The di¬erences as regards age and education were, however, so pronounced8 that they cannot be explained by the above quali¼cations alone. No doubt the ¼ve city pharmacies, which were all located in city centers, had generated far too many young and well-educated respondents (they had a considerably higher

Communication barriers in public discourse

31

response rate than the village pharmacies, sometimes approaching 100%). This probably also explains why relatively fewer respondents than expected personally felt the need for improved PPIs, while more than half the respondents (and 72% of those employed within the health sector) found that other people needed more user-friendly PPIs. The 9% employed within the health sector obviously did not personally feel the need for more user-friendly PPIs — a fact which is of course connected to their status as more or less experts and should clearly not be attributed to userfriendliness in the PPIs (especially not when seen in relation to the above-mentioned 72%). Furthermore, there is a wellknown tendency to show yourself in a slightly better light than warranted even in anonymous studies like the present, a fact which might be o¬set by the tendency in some respondents to please the researcher, that is, to try to provide the answers which they expect are desired. Finally, as one pharmacist (and quite a few respondents) pointed out, not all people are the right persons to assess for themselves whether they have understood the PPI. In connection with the question of “whether a respondent had ever taken his medication in the wrong way due to misunderstanding the PPI”, several respondents wrote that “this is not necessarily something you are aware of yourself”. All in all the above-mentioned factors allow us to conclude that the number of answers which point at di~culties concerning the PPIs must be considered an absolute minimum for the country as a whole. Not one factor seriously points to the opposite. On the basis of the above we can thus conclude that the present PPIs are perhaps userfriendly enough if you ask the well-adjusted, well-educated Dane (who is not employed in the health sector). If, however, we take a look at the weakest third (roughly speaking) of the population we get a di¬erent picture. And it is this third which should be taken into consideration when assessing the user-friendliness of PPIs. As brie½y mentioned above, our methodological considerations on the best possible ways of answering the main questions of our project resulted in a decision to make a

quantitatively signi¼cant questionnaire study as well as detailed linguistic analyses of a large number of Danish PPIs. As we shall see below the linguistic analyses also turned out to involve comparative studies of a number of English and Danish PPIs as a translational aspect proved quite important. Time permitting, we would have liked to supplement our studies with qualitative interviews with patients as well as doctors and pharmacists. Very recently a report made by the British Consumers’ Association has come to our attention. The report is called “Patient information lea½ets: sick notes?” and is concerned with the user-friendliness of British PPIs. The report was completed and published in June 2000, precisely the same month and year as “The PPI of the future.” The two reports do not have exactly the same aims, nor do they cover exactly the same subject, but they do deal with very many of the same aspects of PPIs and always with almost identical results. The British report contains the following conclusions relevant to the present section:
However, Consumers’ Association research has found that patient information leaflets do a very poor job of helping patients take their medicines safely and effectively. Indeed, in their current form many patient information leaflets may be causing unnecessary confusion and alarm for those patients who read them. (Consumers’ Association, 2000, p. 2)

What is very interesting is the fact that the British report is based on other methods than those employed in our own study. The British report chose a qualitative approach and is thus based on a series of one-to-one interviews followed by a series of focus group discussions with patients, doctors, and pharmacists. Twelve in-depth interviews and six focus group sessions with a total of 36 patients were held to establish patient views about PPIs. Twelve general practitioners participated in the research as well as an unspeci¼ed number of pharmacists. A total of 21 PPIs were reviewed as part of the research. Just as is the case with quantitative studies, qualitative research does have inherent problems of which the most crucial must be the risk of interviewer in½uence and the

32

Inger Askehave and Karen Korning Zethsen

desire on the part of the interviewees to please the interviewer; a close second, it goes without saying, is the fact that one could question the statistical signi¼cance of such research. Particularly in the case of focus groups one could fear the risk of a tendency to provide the best anecdote or the like. This being said, we ¼nd it very signi¼cant and of great interest that two simultaneous research projects (which had not been in contact with each other) based on di¬erent methods have come up with such similar results on a large number of issues. In our view the reports supplement each other and in the following we shall refer to the British report whenever we ¼nd it useful. PPIs as an example of public communication Why PPIs are so hard to understand should be answered at the micro- as well as the macrolevel. To provide an answer at the microlevel requires detailed linguistic9 analyses in order to pinpoint where exactly the texts go wrong. To provide an answer at the macro level — to try to explain why the texts go wrong — it is necessary to analyze the situational context of the PPIs and to return to the discussions concerning the concepts of target groups, asymmetrical sender-receiver relationships and mandatory genres. The following is a simpli¼ed version of what takes place in the creation of a PPI: When a medical product is developed, a product summary of the contents, e¬ects, side e¬ects, and other details on the product is written in English by the head o~ce. This document is written by medical experts and is meant for other experts. Then a package insert is written (still in English) based on the product summary (PS) — the law requires a close relationship between these two texts in the name of consumer protection, but the package insert is of course meant for nonexperts. Finally, the English package insert is translated into Danish. Together with all relevant documentation on the product, the English as well as all the national PPIs are submitted to the approving authorities.10

During our research we came to the conclusion that at the macrolevel the poor quality of the Danish PPIs could be attributed to the following two processes, which we have called: Intergeneric translation — i.e., translation between genres Interlinguistic translation — i.e., translation between languages The intergeneric translation, which takes place in our case study when the PS is converted into the PPI, means that the target group is changed, the relationship between sender and receiver becomes asymmetrical and the rules of a new mandatory genre apply. The target group of the PPI as opposed to that of the PS will be brie½y discussed in the following section. The PPI as a new mandatory genre will be dealt with in the section ‘The patient package insert as a mandatory genre’ and the sender-receiver relationship will be discussed and illustrated with examples from our linguistic analyses of PPIs in section ‘The expert language in patient package inserts’. Interlinguistic translation naturally refers to the traditional translation between two languages, which takes place when the English PPI is translated into Danish. This will be brie½y commented on in footnote 13. 1. The target group of the patient package insert Above we discussed the problems connected with the very broad target groups of public communication and drew up a list of strategies available to a public communicator. The target group of the PPI is potentially the entire Danish population and must be characterized as very broad indeed; therefore the visualization of a target group may well be a very substantial problem to the writers. If we take a look at the strategies available, the ¼rst is ‘a strategy based on the creation of a secondary audience’. This secondary audience may consist of the writer’s friends, family, and colleagues. In light of the fact that the writing of the PPI, the translation of the PPI as well as the commenting on and correction of the PPI, and ¼nally the approval of the PPI are all carried

Communication barriers in public discourse

33

out by medical experts (mainly with an educational background as pharmacists) it must be considered very likely that friends, family, and colleagues have a higher level of education (and in the case of colleagues even within the same ¼eld) than the average Dane. Another possible secondary audience would be a fraction of the audience (other than friends and family) whose needs and interests the writer happens to know. If the writer is conscious of the fact that his level of knowledge of this particular ¼eld is higher than the average person and if the writer is able to draw upon experience with a certain group of laypeople, for example, the strategy may work. The danger is, of course, that such a group may perhaps still be too well-educated in general and will thus not cover the great majority of the population. If, however, the secondary audience happens to belong to what could be classi¼ed as the lowest possible denominator, the strategy may be much more successful. The strategies of ‘disillusion’ and ‘optimism’, respectively, are not to be recommended. Rather, we consider them a (quite realistic) description of what takes place when the writer ¼nds consideration of the heterogeneous target group impossible to deal with in a satisfactory way. The ¼nal strategy of ‘lowest common denominator’ is probably the best choice considering the functionality and overall aim of public communication. If the writer constantly asks whether something can be simpli¼ed or explained, there seems to be the best chance of ensuring the highest percentage of audience comprehension (and thus ability to act). However, the strategy of lowest common denominator is not without problems. It may be di~cult to decide where to draw the line of simpli¼cation and explanation, and it may be a great challenge to the writer (especially one with no linguistic background) to explain complex issues (for which he normally uses precise, specialist terms) in a simple way. Also, some people might ¼nd the very simple language patronizing and perhaps the text may even su¬er from status loss due to the removal of features connected with an expert genre.

2. The patient package insert as a mandatory genre In the section ‘Mandatory genres’ on public communication we commented on the use of mandatory genres in the public sphere and alluded to some of the problems caused by such genres. In this section we shall take a closer look at the PPI as a mandatory genre and account for some of the reasons why the PPIs have been unable to behave the way genres are supposed to — namely as “communicative vehicles for the achievement of goals” (Swales, 1990, p. 46). When the Danish Government adopted Directive 92/27/EEC, the medical companies were not only asked to provide consumers with information about their medicines but were also asked to present the information in a speci¼c way, that is to say, to adopt the genre conventions for the PPI laid down by an EU Directive. Thus the PPI can be characterized as a genre invoked by external mandate. Mandatory genres are not ‘unfortunate’ by de¼nition. They may be just as useful for conducting a community’s activities as genres instantiated through tradition. However, in order for them to be just as e¬ective as ‘traditional’ genres, the discourse community has to ensure that the rhetorical forms of the genre are appropriate for whatever purpose the genre is intended to achieve. However, in the case of PPIs there is a signi¼cant discrepancy between the rhetorical forms and the social goals that the PPIs are supposed to achieve. The rhetorical forms of the PPI have to some extent been predetermined by the EU. In Directive 92/27 EEC, the EU lays down strict guidelines as to the type of information which must be included in the PPI as well as the order of information:
The package leaflet [PPI] shall be drawn up in accordance with the summary of the product characteristics [PS]; it shall include in the following order: (a) for the identification of the medicinal product: – the name of the medicinal product…, – a full statement of the active ingredients…, – the pharmaceutical form, …, – … (Council directive 92/27/EEC of 31 March 1992, article 7)

34

Inger Askehave and Karen Korning Zethsen

As appears from the above quotation, the EU directive suggests the PPI be based on an already existing genre, namely the Summary of Product Characteristics (PS). The PS is a scienti¼c report used within a specialist discourse community to provide extensive product information about the medicine, notably by documenting the results obtained while testing the product prior to marketing. The main purpose of the PS is to provide doctors and other health care professionals with essential information about the medicine to ensure that the medicine is used correctly, e¬ectively, and safely. However, the document also serves to provide the approving authorities with information about the drug in order for them to authorize the marketing of the product. The relation between the PPI and the PS is basically a good idea. The legislators want to make sure the consumers get the same information about the drug as the authorities who approve the medicine for marketing and the doctors who prescribe the medicine. However, nonexperts are simply not equipped to receive the information in the same form. Studies show (Consumers’ Association, 2000; Askehave & Zethsen, 2000a) that consumers, doctors, and pharmacists alike ¼nd that the call for correspondence between the PS and the PPI, in terms of content and order of information, results in package inserts which are unable to meet the needs of their primary target group, the consumers. The EU Directive 92/27/EC advocates the following order of information in PPIs:
– Identification of product (name, ingredients, presentation, class, name and address of the manufacturer, and other details about the marketing authorization holder). A list of information the patient should be aware of prior to taking the product (contraindications, precautions for use, interactions, special warnings). Instruction for the proper use (dosage, method of administration, overdose, missing doses, withdrawal effects). Undesirable effects. Expiry date. The date on which the package leaflet was last revised.

–

– – – –

However, this restriction on the sequence of information is problematic considering the fact that the PPI and the PS are di¬erent genres with di¬erent purposes. According to Martin (1993) and Swales (1990) a genre is organized in terms of stages or moves where each move helps realize the purpose of the texts. When advocating identical move structures in PSs and PPIs, the legislators seem to ignore the fact that the PS is a kind of scienti¼c report which usually requires the following information structure: establishing the territory, providing evidence, and drawing conclusions. The PPI, however, is a di¬erent genre. It is a kind of manual, where the relevant and most important information should come ¼rst — the journalistic principle of ‘the inverted pyramid’ should dominate in the order of information. In other words, the conclusions should come ¼rst because this information is the most important to the readers. Analyses support the view that the sequence of information prescribed by the EU Directive is in fact not patient-focused (Consumers’ Association, 2000; Askehave & Zethsen, 2000a). Considering the purpose of the insert, namely to provide information to patients about the medicine to help them take their medication e¬ectively and safely, it seems unfortunate that the insert begins with a long list of information relating to the identi¼cation of the product. For the e¬ective and safe use of the medicine, patients require information about the conditions the medication is used to treat, how it should be taken and stored and, not least, side e¬ects. The result of following the guidelines set up by the EU is that the information which patients ¼nd relevant is provided in the last sections of the insert and this violates the genre conventions of manuals. Considering the fact that the PPI is a manual for correct use, it is interesting to note that the genre conventions proposed by the EU in terms of contents contribute to the ine¬ectiveness of the PPI as a manual. Studies show (Consumers’ Association, 2000; Askehave & Zethsen, 2000a) that patients often complain about the sheer amount and type of information presented in the PPI. For one thing consumers are generally overwhelmed by the enormous

Communication barriers in public discourse

35

amount of information provided, not least the long list of side e¬ects which leaves them feeling worried instead of reassured and con¼dent. Furthermore, consumers explicitly ask for other types of information which they would like to ¼nd in a PPI. For example, many consumers would like information about the particular condition for which they are taking the drug or to have access to information about aftercare services (Consumers’ Association, 2000, p. 16) In other words, consumers generally believe that the information they ¼nd relevant only constitutes an insigni¼cant part of the total information provided — they feel that the insert contains too much super½uous information. However, due to legal constraints, medical companies cannot meet the needs of their audience as that would violate the genre conventions laid down by the EU directive. 3. The expert language in patient package inserts (asymmetrical sender-receiver relationship) As suggested in section the ‘Asymmetrical sender-receiver relationship’, communicators within the public sphere tend to be specialists in the subject matter of their documents, and the same goes for the communicators of PPIs. PPIs are written by pharmacists from the departments of regulatory a¬airs in the medical companies. They are experts within the ¼eld of medicine and, as such, their linguistic behavior is greatly in½uenced by the discourse conventions of their expert discourse community whereas, presumably, their knowledge of successful public communication is almost nonexistent. From the very beginning of our PPI project we suggested that the lack of user-friendliness in PPIs may be attributed to the choice of words and expressions used in these particular documents. Consequently, we decided to take a closer look at the linguistic features of the Danish inserts (approximately 40 Danish patient package inserts from 10 di¬erent medical companies). It turned out that the following lexical and syntactic features were very dominant:11

Lexis – Medical jargon – “False friends” – O~cialese – Inconsistent terminology Syntax – Passive and impersonal style – Information-packed sentences Several studies show that not only do these features accentuate a scienti¼c and objective style (i.e., signify expert language); they also make the text less accessible to the general reader (Killingsworth & Gilbertson, 1992; Killingsworth & Ste¬ens, 1989; McKenna & Graham, 2000; Fluck, 1985). The aim of the analyses was to point out some of the rhetorical strategies that may hamper the readability of the inserts, and to provide recommendations for their improvement. The following sections present some of our main ¼ndings. LEXIS Medical jargon One of the key characteristics of expert language is the use of terms speci¼c to particular disciplines. Expert terms or medical jargon are closely related to the ¼eld of discourse and are normally used by experts only. From our analysis of Danish PPIs it appears that expert terminology is extremely widespread: Inteferon alfa-2b for subkutan anvendelse LT: Inteferon alfa-2b for subcutaneous use12 Kontraindikationer LT: Contraindications Anafylaktisk shock LT: Anaphylactic shock Peroralt LT: Per orally

36

Inger Askehave and Karen Korning Zethsen

Antidiabetikum LT: Antidiabeticum Symptomgivende gastro-esofageal re½ux LT: Symptom-giving gastro-esophagal re½ux It should be noted that Danish, in contrast with English, usually has two terms for the same medical concept: an expert term, which is Latin-based and generally not understood by the average Dane, and a lay term for everyday use. Everyday medical expressions of Latin origin are much more common in English than in Danish, as indeed in the vocabulary in general. Danes still mostly use old, simple, and immediately understandable Scandinavian words when talking about a medical subject (e.g., ‘livmoder’, ‘mavesår’, and ‘stivkrampe’). In English-speaking countries, however, the Latin expressions have in many cases merged with common language to such a degree that no Anglo-Saxon alternative exists — at least not one which is used in a medical context (e.g., ‘uterus’ (livmoder), ‘ulcer’ (mavesår) and ‘tetanus’ (stivkrampe)). Most of these Latin words and expressions are known by the common Englishman, but are either incomprehensible or sound very formal to the ordinary Dane. To meet the needs of nonexpert readers, some inserts provide an explanation of the expert term in brackets. Hypoglykæmi (dvs. for lavt blodsukker) LT: Hypoglycemia (i.e., too low blood sugar) Respirationsstop (ude af stand til at trække vejret) LT: Respiration stop (unable to breathe) However, in most inserts the writers prefer their ‘own’ expert terms instead of using plain Danish alternatives understandable to the general public. False friends Another phenomenon common to the terminology in PPIs is the use of ‘false friends’: words and expressions which are used both in everyday situations and in special contexts,

such as a medical one, where they have developed a specialized meaning that to the ordinary Dane constitutes an unusual and nontransparent use of the word. It is thus an additional source of confusion that the word is well-known to the reader, but does not make sense in its PPI context: Lokal ødemdannelse LT: Local edema creation. Administrere (= give) LT: Administer (in everyday Danish ‘administrere’ only means ‘manage’, as in ‘manage a business’) Behandlingssvigt LT: Treatment failure O~cialese In public documents we often ¼nd linguistic remnants of o~cialese, for example, rather long or complicated words and expressions which increase the level of formality and make the text more di~cult to digest (e.g., “in accordance with”, “in the event that”, “owing to the fact that”). In the Danish PPIs the choice of complex and formal solutions is also extremely common: “Tilstrækkelig” instead of “nok” LT: “Su~cient” instead of “enough” “I kombination med “ instead of “med” LT: “In combination with” instead of “with” “Forårsaget af” instead of “fordi” LT: “Caused by” instead of “because” Inconsistent terminology Inconsistent terminology is also very common in the PPIs. The average Danish reader, who does not possess the expert knowledge needed to judge whether terms are synonymous or not, may become quite confused when confronted with two or three di¬erent terms for the same thing:

Communication barriers in public discourse

37

Synonyms: Blodets glucosekoncentration LT: The glucose concentration of the blood Blodsukkeret LT: The blood sugar Synonyms: Hjælpesto¬er LT: Ancillary substances Øvrige indholdssto¬er LT: Additional content substances SYNTAX In the case of the syntactic features in the inserts, the lack of user-friendliness is of a somewhat di¬erent nature. Whereas expert terminology may be completely unknown to the consumer and may present an immediate obstacle to comprehension, most syntactic and grammatical features in inserts are not unknown to the reader. The nonexpert does encounter such structures as nominalizations and passive voice in less specialized texts. However, the frequency with which such complex structures appear makes the texts less accessible to the ordinary reader. Passive and impersonal style An extremely passive and impersonal style is very common in PPIs. This phenomenon is also a relic from the specialist discourse community where deagentive measures such as the passive voice and nominalizations, allow the expert to provide condensed, impersonal information for other experts who are trained to perceive and consequently talk about the physical world in terms of concepts rather than actors performing actions. A too extensive use of deagentive measures is very detrimental to lay comprehension, and especially in cases which require the patient to act (drink the solution, press the pump or ask his doctor), active forms are called for.

Passive voice Der skal udvises forsigtighed ved samtidig indtagelse af alkohol, da alkoholpromillen kan øges LT: Carefulness must be exercised in connection with simultaneous consumption of alcohol as the alcohol content of the blood can be increased Når en spray½aske påbegyndes, trykkes sædvanligvis 6–7 gange på pumpen LT: When a spray bottle is initiated, the pump is usually pressed 6–7 times These examples also show that personal pronouns are generally avoided in the Danish inserts. And, if personal pronouns are used, there is a strong tendency to use the very polite pronoun “De” (nowadays mainly reserved for the elderly or very formal situations), corresponding to the French “Vous.” Nominalizations The passive style of Danish inserts is also achieved through the extensive use of nouns instead of verbs: Forsigtighed tilrådes ved bilkørsel og maskinbetjening LT: Carefulness is recommended in connection with car driving and machine operation Tilrådeligheden af at køre bil skal overvejes under sådanne omstændigheder LT: The advisability of driving a car should be considered under such circumstances Svimmelhed forårsaget af for lavt blodtrykt fremkaldt ved for hurtige skift fra liggende til stående stilling LT: Dizziness caused by too low blood pressure provoked by too quick shifts from lying to standing position Information-packed sentences Generally, PPIs are rather long and in addition, the writers often try to cram as much information as possible into one

38

Inger Askehave and Karen Korning Zethsen

sentence. The result is lengthy, in½ated sentences. However, the longer and more information-packed a sentence, the harder it is for the readers to understand the essentials of the message. xx anvendes til symptomatisk behandling af problemer opstået som følge af godartet forstørrelse af blærehalskirtlen (benign prostatahyperplasi) i perioden inden operation LT: xx is used for symptomatic treatment of problems arisen as a consequence of benign enlargement of the prostate [in the original a Danish lay term is used] (benign prostate hyperplasia) in the period before operation Hos patienter med tidligere moderate til svære symptomer på sæsonbetinget allergisk rinit kan forebyggende behandling med xx påbegyndes op til ¼re uger før den forventede start af pollensæsonen LT: In patients with former moderate to considerable symptoms of seasonal allergic rhinitis preventive treatment with xx can be initiated up to four weeks before the expected start of the pollen season In contrast with sentences containing super½uous words, the information-packed sentences are complex because they contain too much relevant information. Moreover, sentences often become long and complex because the writer adds extra information to the nouns by means of pre- and postmodi¼ers. The result is heavy noun phrases which, when unpacked, would form sentences in their own right. Note also that the head of a noun phrase is often realized by nominalization, which again makes it even more di~cult for the reader to extract the relevant information: Symptomatisk (premodifer) behandling (head) af problemer (postmodi¼er) LT: Symptomatic (premodifer) treatment (head) of problems (postmodi¼er) Godartet (premodi¼er) forstørrelse (head) af blærehalskirtlen (postmodi¼er) LT: Benign (premodi¼er) enlargement (head) of the prostate (postmodi¼er)

Summary
We can safely conclude that Danish inserts su¬er from a direct transfer of expert language both in terms of terminology and syntactic structure, which suggests that the asymmetrical relationship between sender and receiver is not taken su~ciently into consideration. It is not a question of avoiding complex information, o~cialese, or a passive style altogether. However, when all these features are crammed into one paragraph or even one sentence, the PPI becomes di~cult to read and understand. In short, many Danish PPIs are so heavy and formal that the average reader has to spend considerable time and energy decoding the text in order to ¼nally deduce the relevant message. Ironically enough, both European and Danish authorities are aware of the unfortunate language in public documents. Guidelines on language use in o~cial legal documents from the Danish Ministry of Justice speci¼cally warn against the use of the above linguistic features in o~cial documents from the Danish State (Hansen & Lund, 1988). Moreover, the EU Readability Guidelines — designed to ensure that labels and package inserts are user-friendly — also address the readability problems and provide recommendations for improving inserts. However, according to the study by the British Consumers’ Association (2000, p. 26) the guidelines are often “either too vague or do not go far enough to ensure that manufacturers will produce good-quality patient information lea½ets.”13

Conclusion
In the ¼rst part of this article we discussed the nature of communication written by experts or semiexperts and aimed at a large and heterogeneous audience, often potentially the entire population of a country. We compared public communication to mass communication and concluded that there are strong a~nities as well as di¬erences (genremediated, regulatory a~nity, highly functional). We analyzed the common barriers to this kind of communication

Communication barriers in public discourse

39

and found out that the extremely broad nature of the target groups, the mandatory nature of the genres and the asymmetrical sender-receiver relationships are three highly problematic areas. In the second part of the article we outlined our project on patient package inserts as a case in point. Our questionnaire study established that the PPI is indeed considered a di~cult document that does not comply with the legislative intention of user-friendliness. On the basis of the results of our project we analyzed the suggested problematic areas of heterogeneous target group, mandatory genre, and asymmetrical sender-receiver relationship and our case study con¼rms the results of the theoretical discussion in the ¼rst part of the paper. As appears from the above sections, the reasons why most of the present Danish PPIs are not successful are multiple, but the three problem areas de¼ned are de¼nitely at the heart of the matter. The medical expert status of those making the PPIs (and in addition their lack of language and communication skills) is the most harmful factor in relation to the user-friendliness of the package insert. The medical experts cannot free themselves from their own background knowledge and linguistic habits (and pos-

sibly expert status) and fail to reach a realistic perception of their audience (which in the case of public communication can indeed be very di~cult). It is further more quite evident that though the PPI was introduced in order to inform and protect the patient, it su¬ers from the fact that it is a mandatory genre with a framework laid down by legal and medical experts and thus did not develop naturally in response to patient needs. Mandatory genres need not be unsuccessful by de¼nition, but we need more professional communicators at all levels of the process. Experts have typically, and quite naturally, not been trained in the rhetorical ways and means of language and consequently would not know, for example, how to raise or lower the level of formality in a text. We are convinced that a large number of di¬erent genres belonging to the sphere of public communication share the same general problems and we hope that our attempt at de¼ning “public communication” and our discussions of inherent problems may be helpful in studies of a similar nature. User-friendliness is a very important concept in public communication in a democratic society and therefore functionality ought to be at the core of all discussions of public documents.

Notes
1. When using the term ‘public communication’ we include major organizations such as banks, insurance companies, trade unions, and pension funds, which to a large extent rely on written documents (frequently of great length) when communicating with their customers. 2. For de¼nitions of mass communication see, for example, Bell (1996, 5th ed.), and McQuail (2000, 4th ed.). 3. Apart from lacking the linguistic skills of ‘translating’ expert language into nonexpert language, experts may also be reluctant and directly opposed to using everyday language because they fear loss of status and power when ‘popularizing’ their language. 4. The questionnaire study generated much

more information on various aspects of patient views on PPIs. The results cited in this section are only those relevant for present purposes; some of the other insights will be reported on in subsequent sections. For a more detailed description and discussion of the questionnaire study (design, results, and validity) see Askehave & Zethsen, 2000a. 5. After the questionnaire the respondents were invited to comment on any aspects of the subject they liked and 93 respondents chose to do so. 6. This amounted to 9% of the respondents. 7. The interviews with the participating pharmacies after completion of the study revealed, for example, that one pharmacy did not hand out a questionnaire to a customer who they knew had reading di~culties. 8. For example, twice as many respondents had

been to college than in the population in general. 9. And of course nonlinguistic analyses having to do with layout and design. 10. Basically there are two ways in which a product — including the PPI — can be approved: either locally by the national authorities or centrally by the European Medicines Evaluation Agency (EMEA) in cooperation with the Commission. By way of illustration we will focus on the European approval procedure, which is used increasingly and which, according to the medical companies, causes the most trouble. 11. See also Askehave & Zethsen, 2000b. 12. Whenever a Danish word or sentence is mentioned we provide a ‘literal translation’ (LT). The purpose of this very direct translation is to give the reader a clearer idea of the contents of

40

Inger Askehave and Karen Korning Zethsen

the Danish version. The translation is not necessarily syntactically or idiomatically correct in English. 13. When we analyzed our corpus of approximately 40 Danish PPIs it quickly became obvious that some of the problems were of translational origin: the PPIs contained a number of unfortunate features which we felt quite sure could be traced back to the English originals. This prompted us to make a number of contrastive analyses between the English and Danish source texts and target texts and we found a great number of translational ‘errors’ seen in relation to the target group and the given objective of user-friendliness. On the basis of our contrastive analyses we were able to conclude that without exception the Danish PPIs were more formal and dominated by expert syntax and terms than the English originals (which, as we have seen, have been subject to criticism themselves for being too formal because of the intergeneric relation with the PS among other reasons). The pharmacists who had translated the PPIs into Danish had simply taken in the actual messages of the original but not the style, and had reverted to their own habitual expert style when rendering these messages in the PPIs. These defects are rarely ameliorated by the approving authorities as the national PPIs are checked by medical experts (who are of course familiar with the expert style) who focus on the factual contents. The raised level of formality together with the downright errors or unidiomatic expressions originating from the translation process must be said to be signi¼cant detriments to the user-friendliness of the Danish PPIs. For a more detailed discussion of the translational aspects (including the more speci¼c analyses) see Askehave & Zethsen, 2000a and 2002.

References
Askehave, I. & Zethsen, K. K. (2000a). The Patient Package Insert of the Future. Report for the

Danish Ministry of Health [Danish and English version]. Aarhus: The Aarhus School of Business. Askehave, I. & Zethsen, K. K. (2000b). Medical Texts Made Simple — Dream or Reality? Hermes, 25, pp. 63–74. Askehave, I. & Zethsen, K. K. (2001). Intergeneric and Inter-linguistic Translation of Patient Package Inserts. In: Mayer, Felix (ed.). Language for Special Purposes: Perspectives for the New Millennium (pp. 882–887). Gunter Narr Verlag: Tübingen. Askehave, I. & Zethsen, K. K. (2002). Translating for laymen. Perspectives: Studies in Translatology, Vol. 10(1), 15–29. Bell, A. (Ed. 5)(1996). The Language of News Media. Oxford: Blackwell Publishers Ltd. Berkenkotter, C. & Huckin, T. N. (1995). Genre Knowledge in Disciplinary Communication: Cognition/Culture/Power. Hillsdale: Lawrence Erlbaum Associates Publishers. Bresler, K. (2000). Getting Rid of Gobbledygook. Pharmaceutical Executive, October 2000, 1–3. Consumers’ Association (2000). Patient Information Leaflets: sick notes? Report June 2000. Council Directive 92/27/EEC of 31 March 1992. Davis, J. J. (1999). Imprecise Frequency Descriptors and the Miscomprehension of Prescription Drug Advertising: Public Policy and Regulatory Implications. Journal of Technical Writing, 29, 133–152. Eckkrammer, E. M. (1998). Das Dilemma mit dem Beipackzettel: Ein italienischdeutscher Vergleich der (fach)sprachlichen Verunsicherungsfaktoren. In: P. Cordin, M. Iliescu and H. Siller-Runggaldier. Parallela 6: Italiano e Tedesco in contatto e a confronto/ Italienisch und Deutsch im Kontakt und im Vergleich. Atti del 7° Incontro italo-austriaco dei linguisti. Trient: 345–370. Fluck, H-R. (Ed. 3) (1985). Fachsprachen. Francke Verlag: Tübingen. Hamburger, A. (1988). O~ciel sprogbrug. In: E. Hansen & J. Lund (Eds.). Sproget her og nu.

København, Gyldendal, 146–156. Janssen, D. & Neutelings, R. (Eds.)(2001). Reading and Writing Public Documents. Amsterdam/Philadelphia: John Benjamins Publishing Company. Killingsworth, J. & Ste¬ens, D. (1989). E¬ectiveness in the environmental impact statement. Written Communication, 6, 155– 180. Killingsworth, J. M. & Gilbertson M. K. (1992). Signs, Genres and Communities in Technical Communication. New York: Baywood Publishing Company. Kitching, J. B. (1990) Patient Information Lea½ets — the State of the Art. Journal of the Royal Society of Medicine, Vol. 83, May 1990, 298–300. Lind, M. (1997). ‘Bedre information på indlægssedler efterlyses’. Farmaceuten, 16, 464–467. Martin, J. R. (1993). ‘Genre and literacy: modelling context in educational Linguistics’. Annual Review of Applied Linguistics, 13, 141–72. McQuail, D. (2000 4ed). MacQuail’s Mass Communication Theory. London: Sage. Miller, C. R. (1984). ‘Genre as Social Action’, Quarterly Journal of Speech, 70, 151–167. McKenna, B. & Graham, P. (2000). ‘Technocratic Discourse: A Primer’. Journal of Technical Writing and Communication, 30(3), 219–247. Orlikowski, W. J. & Yates, J. (1994) ‘Genre Repertoire: Examining the Structuring of Communicative Practices in Organizations’. Administrative Science Quarterly, 39–54. Sless, D. & Wiseman, R. (1997). Writing about medicines for people — usability guidelines for consumer medicine information. Canberra: Communication Research Institute of Australia. Swales, J. M. (1990). Genre analysis: English in academic and research settings. Cambridge: Cambridge University Press. Windahl, S. & Signitzer, B. (1992). Using Communication Theory. London: Sage.

Communication barriers in public discourse

41

about the authors
Inger Askehave received her Ph.D in Linguistics from the Aarhus School of Business in 1998 and is currently Senior Lecturer in the Department of English at the Aarhus School of Business. Her research interests include genre theory, stylistics, business communication, and New Age rhetoric. Contact: ia@asb.dk Karen Korning Zethsen received her Ph.D in Linguistics from the Aarhus School of Business in 1998 and is currently Senior Lecturer in the Department of English at the Aarhus School of Business. Her research interests include lexical semantics, stylistics, corpus linguistics, and translation theory. Contact: kkz@asb.dk

Copyright of Document Design is the property of John Benjamins Publishing Co. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.