Home | Philosophy Essays | Philosophy Links | Linguistics Essays | Linguistics Links | About me | Journal
In this paper I propose to analyse the registers found in two text types. One is the "compilation essay"; that is, an essay published in a collection or Schriftfest, usually compiled around a particular theme. This genre has much in common with the more familiar one of the academic journal article, and in fact many items in compilation volumes are reprinted from journals. Such volumes contain the works of respected writers in the field; unlike journals, it is very rare for a hitherto unpublished author to appear, and as such they are often used as models for aspiring writers. I shall look here at one fairly typical compilation, Meanings and Prototypes: Studies in Linguistic Categorization (1992), and in particular at Dirk Geeraerts' essay, "The lexicographical treatment of prototypical polysemy" (1992:195-210).
The second text type is what I shall refer to as "informal academic writing". This genre consists of texts written by academics for a more general, though still well-informed audience. I have coined this term to distinguish the genre from "popular science", where the author is writing for a mass market, and few assumptions of knowledge are made. Informal academic writing, on the other hand, is mainly a way for academics to present their ideas to readers both within and outside their particular disciplines; the reader is assumed to be well-educated, and may be familiar with the writers' discipline or, more commonly, a related discipline. Thus we find sociolinguistics for sociologists, psychology for linguists, linguistics for philosophers, and so on. Although articles in magazines such as New Scientist fall to an extent into this genre, I shall here focus on books.
The writer in this genre is placed in the rather difficult position where she or he must meet the intellectual standards of their own discipline, present their subject matter in a way that is easily comprehensible to readers from other disciplines (or no particular discipline) and, most importantly, write in a way that arouses and maintains the interest of a reader who has little or no professional interest. Good examples of this genre in the field of linguistics are Noam Chomsky's (1985) Knowledge of Language, Derek Bickerton's (1995) Language and Human Behaviour, and the main work I shall discuss, George Lakoff and Mark Johnson's Metaphors We Live By (1980).
In order to compare these two genres, I shall adopt for the most part Halliday's register analysis based on field, tenor and mode (Halliday, 1964, 1978, Halliday & Hasan, 1985), while making occasional use of Swales' (1990) concept of "move". After comparing the genres, I shall draw some brief conclusions concerning the consequences of these register-types for the teaching of academic writing, with particular reference to the problems of non-native students. First, however, a few general comments on the terms used are in order.
Register and genre are not interchangeable terms; neither do they refer simply to the two main schools of analysis represented by Halliday and Swales respectively. Halliday and Hasan define register as "a configuration of meanings that are typically associated with a particular situational configuration of field, mode, and tenor...[which must] include the expressions, the lexico-grammatical and phonological features, that typically accompany or REALISE these meanings" (1985:39). In contrast, genre, as defined by Swales, is "a set of communicative events, the members of which share some set of communicative purposes" (1990:58). Genre is thus higher in the semiotic hierarchy than register.
To clarify this point we may take Bex's (1996) Hallidayan analysis of two bills. Although the choices here regarding field, tenor and mode are not exactly wide, Bex notes enough variation to be able to claim that "each text can only realise its own register" (1996:104). He goes on to point out the need for "a higher level of analysis which groups texts into types and demonstrates what they have in common both in terms of their particular linguistic selections and in terms of their shared social purpose" (1996:104/5). It is this "higher level" which is best termed "genre".
The basic concepts of register analysis - field, tenor and mode - are themselves not entirely clear-cut. For the purposes of this essay, I shall adopt Halliday's definition of field as the way "registers are classified according to the nature of the whole event of which the language activity forms a part" (1964:90). This may be clarified with reference to Gregory and Carroll's definition: "Field of discourse is the consequence of the user's purposive role, what his language is 'about', what experience he is verbalizing, 'what is going on' through language" (1978:7). For tenor I shall use Gregory and Carroll's definition as the "relationship the user has with his audience" (1978:8).
Mode is slightly more problematic. Halliday's initial definition of mode as "the medium or mode of the language activity" (1964:91) is somewhat circular. He later clarifies this as "the symbolic organisation of the text, the status that it has, and its function in the context, including the channel" (1985:12). Mode is perhaps best seen as the physical medium of communication, along with the choices this provides, and the limitations it imposes. Since this also tends to play a key role in determining the choices available in field and tenor, I shall start with mode.
In terms of mode, the compilation Meanings and Prototypes is printed text, with no pictures, but a number of tables and diagrams, and a variety of typefaces, including Greek letters and symbols. Such features generally have the effect of reinforcing the "scientific" nature of a text, which can be seen as particularly important in a discipline such as linguistics which is still struggling for recognition as a "real" science.
The most important feature of mode, however, is permitted length. Just as an international telephone conversation usually imposes limits which would not apply in face-to-face communication, so does the type of published text impose similar limits. While a compilation of essays may be anything from 100 pages upwards, the individual contributions tend to be limited, irrespective of the length of the book. An essay may be slightly longer than a journal article, but not much; the average length of contributions to Meanings and Prototypes is only 20 pages, which is fairly typical.
Limitation of length means that authors need to use a fairly terse style if they are to say everything they want; they need to get to the point quickly and stick to it. This is reflected in the opening paragraph of Geeraerts' essay (p.195):
There have hitherto been two main directions in the application of the prototypical conception of semantic structures to the problems of lexicography. One the one hand, Wierzbicka (1985a, 1987a, 1987c) has presented a method of semantic analysis which replaces traditional dictionary definitions by exhaustive, in-depth analysis of the prototypes underlying natural-language concepts. On the other hand, I myself have argued that the prototypical model of lexical polysemy is an indispensable aspect of any theory trying to classify the diverse forms of semantic information that may be incorporated in lexicographical reference works (Geeraerts 1985a, 1987a).
To use Swales' (1990) concept of move, this follows the first two moves of a typical three-move academic introduction (Bhatia, 1993:83): establishing a territory (first two sentences) and establishing a niche by counter-claiming (last sentence). It could even be argued that the final reference to the author's own works goes some way to fulfilling move three (occupying the niche). The point here is that the moves are accomplished in only 94 words, aided by the simple cohesive device on the one/other hand.
If the text Meanings and Prototypes is part of a "whole event" (Halliday, 1964:90), what is this event? Academic writing is best characterised in terms of an on-going dialogue on a variety of overlapping topics. Texts are in (imaginary) dialogue not only with their readers, but with previous texts (Bex, 1996:53). Meanings and Prototypes is a contribution to the discussion of prototype theory, which is in turn part of larger discussions of theory in the disciplines of linguistics, philosophy and cognitive science, while Geeraerts' "The lexicographical treatment of prototypical polysemy" is both a contribution to Meanings and Prototypes, and a contribution to these broader discussions. The writer therefore needs to say where they stand in these discussions by clearly indicating the field of their text.
The inclusion of Geeraerts' essay in this compilation itself defines the essay's field in broad terms. Geeraerts himself then needs to identify the particular field, and this is achieved by the title, and also by the first paragraph, with its familiar "Swalesian" move-structure. The first sentence defines the field in general terms:
There have hitherto been two main directions in the application of the prototypical conception of semantic structures to the problems of lexicography.
In Gregory and Carroll's terms, this is "what his language is 'about'". As for the other main dimension of field, "'what is going on' through language", this is indicated by "On the other hand, I myself have argued that ...". We can thus summarise the field of Geeraerts' essay as:
What we may term "sub-fields" are indicated by subtitles; by scanning the text the reader can see at a glance how the field is further narrowed and are further managed by the opening sentences of each section, for example in the conclusion:
There are a number of points that may be retained from the foregoing discussion. Basically, there is no reason to identify the classical conception of categorization with an allegedly lexicographical model of definitional structures. (1996:209)
The first sentence marks the transition from sub-field (the linearization problem) to overall field (lexicography and categorization) while the second indicates the main contribution of the sub-field. Field, then, is more than simply the general subject-matter of the text - we can think of a set of nested fields (both in text and context), each with its own rhetorical structure.
Let us for a moment imagine that instead of writing an essay, Geeraerts is trying to convey the same information to a group of interested students in the university coffee bar. To accomplish the transition just mentioned in the first two sentences of the conclusion, he might say something like this:
OK, I've gone on about this linearization problem for quite a while, but I think there are a couple of really interesting points here, and the main one is that we don't need to identify the classical conception of categorization with this so-called lexicographical model of definitional structures.
Tenor is thus influenced by mode; certain linguistic expressions are more to be expected in face-to-face communication than in printed text, for example. A reader stands in a different relationship to a writer than a listener does to a speaker, just as the reader of a journal article ("colleague") stands in a different relationship to the writer than does the reader of a mass-circulation newspaper ("member of the public"). Tenor is similarly influenced by field, academic subjects tending to presuppose a rough equality of knowledge.
In written academic discussions there is a default tenor which is generally termed "formal/impersonal". However, there are different types of formality (e.g. tea at Buckingham Palace) and impersonality (e.g. a gas bill). We can characterise the default features of academic tenor as:
As we will see later, all these have their exceptions, but first I shall look at the Geeraerts essay as a fairly typical example of this tenor.
An impression of objectivity is given by a number of well-known stylistic devices. In the conclusion (pp. 209-10), for example, the first person pronoun does not occur once. Although the conclusion of a paper is where we look for the writer's opinions, they are here expressed with typical obliqueness, using the following "stock" phrases:
there is no reason to ...
An examination of ... reveals ...
Rather, ...
It follows from this observation that ...
... may be envisaged
... had better ...
... should be rejected as misguided
it would seem that ...
This contrasts somewhat with the introduction, where it is common for the writer to state their position clearly; they are introducing not only the text but also themselves (e.g. "On the other hand, I myself have argued that ..." (p.195)).
Reader anonymity is a feature of all published material. However, in journal articles and compilation essays the writer can at least assume that the reader is part of an expert group, which usually means that there is less need for definition and explanation. Geeraerts' piece is notably lacking in such; despite his use of a large number of highly specialised terms, the first explanation of a term occurs eleven pages into the essay.
Politeness is not a universal feature of academic writing; it varies between disciplines and individual writers. Acknowledged experts can afford more of a "knockabout" style (Chomsky and Bickerton being two notorious examples) but in general it is not considered good form to insult ones colleagues. This (slightly strained) politeness can be seen in Geeraerts, who frequently defines his position by contrasting it with that of Anna Wierzbicka. A typical example is the following:
Should prototypical meanings indeed receive the full definitions of 'gargantuan length' (1985:37) defended and admirably illustrated by her? From a theoretical point of view, she is undoubtedly right ...this is a position that I have elsewhere defended at great length myself (1985b), and which I see no reason to take theoretical issue with.
In other words, "Wierzbicka's got the theory right but the practice wrong."
Acknowledgement of others' ideas is, of course, a hallmark of academic writing; the more "academic" a text is, the more it will bristle with citation. Similarly, caution in expressing ones own ideas is to be expected, though again it is a mark of academic "privilege" to be able to make bolder claims. Finally, we may note that recently there has been a slight relaxation of standards in terms of linguistic "correctness"; for example Geeraerts, like many other writers, is quite happy to start a sentence with a co-ordinate conjunction.
Metaphors We Live By is a 242-page, paper-bound book, with no illustrations, but, like the previous text, a number of diagrams and a variety of typefaces (notably italics and small capitals, which have almost become a trademark of cognitive linguistics). It contains back-cover "blurb", consisting of the usual advertising/summary, a quotation from the book, favourable reviews, and brief information about the authors. We should not read too much into this opportunity to describe/sell the book, but it could be argued that it relieves the authors of some of the need to explain their work in the title and introduction.
Again, however, the main factor is length. While the introduction to an essay or article must briefly tell the reader what the field is, and place it in the broader field of the compilation or journal, the writer of a book primarily needs to convince the reader that it is worth the investment of time (and usually money) to continue. Metaphors We Live By follows a three-move structure which is somewhat different from the classic academic one:
[1] Metaphor is for most people a device of the poetic imagine and the rhetorical flourish ... Moreover, metaphor is typically viewed as characteristic of language alone rather than thought or action. [2] For this reason, most people think they can get along perfectly well without metaphor. [3] We have found, on the contrary, that metaphor is pervasive in everyday life ....
Move 1 defines the field as metaphor and gives the popular view. Move 2 gives an undesirable consequence of this view, implying the existence of an alternative view (establishing a territory), and move 3 presents this alternative (occupying the territory). In a sense this is the opposite to the classic advertising strategy: 1. Most people think this; 2. They are right; 3. You should think like most people (and buy our product). What Lakoff and Johnson seem to be adopting is the alternative strategy: 1. Most people think this; 2. They are wrong; 3. You should not think like most people (and therefore should read our book). While here we are obviously getting into the realm of functional tenor (Gregory and Carroll, 1978:8), it is worth noting how again this is influenced by mode.
An interesting feature of Lakoff and Johnson's book is that it often indicates field by giving examples (in contrast to Geeraerts, who does not give examples of prototypical polysemy until well into the essay). Thus the introduction has an extended discussion of the metaphor argument is war (1980:4-5) as a way of showing what the authors are talking about ("To give some idea of what it would mean for a concept to be metaphorical ..."(p.4), "This is an example of what it means ..." (p.5)). Thus the subject-matter is defined, explained and exemplified at the same time.
Metaphors we Live By differs in style from what we tend to think of as "academic writing". Looking at the features of the "default tenor", the first difference is that in this genre the need to maintain an appearance of objectivity through an "impersonal" style is less marked. This is shown clearly in the conclusion. Instead of the expected "Conclusions", "Findings" or "Discussion" section, we have an "Afterword" (p. 239) which is decidedly personal in tenor:
... the opportunity to explore our ideas ...
... having worked out all the consequences we could think of ...
... what stands out most in our minds ...
We continually find it important that ...
Note the difference here between the personal "we" (meaning the two authors) and the impersonal academic "we" found in phrases like "we can conclude that ...".
In contrast, reader anonymity is more real in this genre than in compilations or journals; the writer can only guess at the background of the reader. This, as we have seen, means that there can be fewer assumptions of knowledge, and more explanation and exemplification is required. To compensate for this anonymity a further use of "we" includes the reader (and indeed everyone); the first page alone contains eleven instances of (non-author) "we" and six instances of "our". In addition, the language is generally closer to spoken English: contractions, phrasal verbs and idioms are common. Italics are frequently used to substitute for spoken stress (e.g. "many of the things we do in arguing ..."(p.4) "this is the ordinary way of having an argument"(p.5)).
As for (iii) - politeness - and (v) - caution - the same standards generally apply, though perhaps with a little more lee-way. "We have found, on the contrary, that ..." (p.3) is typical of the slightly more confident statement of views in this genre. Feature (iv) - acknowledgement - applies as well, but there is generally much less quotation and citation, partly because the reader is less likely to be familiar with the works cited or be prepared to look them up, and partly, perhaps, because it detracts from the "spoken" style. When references are given, they tend to be embedded into the sentence more.
Finally, as regards observation of "correct" linguistic forms, book writers again have a little more freedom; Metaphors We Live By, while containing no obviously ungrammatical sentences, has sentences which would be classed as "incorrect" by the standards of an academic writing textbook.
Overall, then, we can summarise the tenor of this genre as more personal, less formal and much closer to spoken English than that of compilation essays and journal articles.
A crucial part of syllabus and materials design is selection of "typical" texts, both for reading practice and as models for writing, and abstraction of principles by which they can be decoded and/or reproduced. It should be clear from the above discussion that "academic writing" is not one genre but several. Which academic writing we adopt as a model, then, is a matter of some importance.
In selecting texts, both the skill practised and the linguistic and academic level of the students are important. If the skill to be practised is reading and the students' level of English is typical of non-native undergraduates, "informal" academic texts (whether books, or articles from magazines such as New Scientist or Scientific American) are a much better place to start than either journal articles and essays, or (as is commonly the case) popular magazines such as Time or Newsweek. Interdisciplinary texts have the dual advantage of comprehensible language and comprehensible ideas, while still enabling teachers to select material relevant to the students' disciplines. Similarly, a language support teacher helping a student research a term paper might do well to advise them to start with such texts, rather than go straight to a citations index. Essay compilations would be the next step, followed perhaps by journals (for the really intrepid student).
In terms of writing skills, the genre to be taken as a model is dependent on the students' level. While postgraduate students writing a dissertation need to follow the conventions of academic writing more closely, there are good grounds for taking informal academic writing as a model for undergraduates. Imitation of journal articles often results in a garbled mixture of "academese" and informal (and often inaccurate) English, when what is required is clear, simple English which nevertheless respects the more general principles of academic writing, such as objectivity, acknowledgement of sources, logical organisation, and so forth. The exceptions to this rule concern citation (generally speaking, the more, the better) and, perhaps, the introduction and conclusion sections. In this case, rather than have first-year undergraduates wade through obscure journal articles or essays, it might be better for teachers to write models themselves, or to use good examples of student writing.
These observations apply just as much to textbook-writing as they do to individual texts studied in class. While accuracy in grammar and punctuation is laudable, one does get the impression that authors of such standard textbooks as Writing Academic English (Oshima & Hogue, 1985) are expecting a little too much of students. Why, we might ask, should a non-native undergraduate writer take such pains to distinguish between subordinate and co-ordinate conjunctions when respected linguists do not? Similarly, it is surely preferable for a student to use "but" effectively than to use "notwithstanding" incorrectly. Taking informal academic writing as a model may provide a happy medium in register between the personal "letters to the editor" style favoured in general ELT textbooks and an ideal of academic writing which students at this level cannot possibly approach.
While researching and writing this essay, I was struck by a couple of points. The first, of course, was the existence of what I have dubbed the "informal academic style", which had caught my attention some time ago, but which I hadn't bothered to analyse thoroughly. Secondly, it occurred to me that much of the advice I had been giving students was, if not exactly wrong, at least unhelpful. When researching, I tend as a matter of course to go first to compilations of essays - if nothing else it's much easier than ploughing through a lot of journals. But what works for a native-speaker postgraduate may not work for a non-native undergraduate. Finally, whatever register we advise our students to adopt, it's important that their end-product is consistent. Sudden changes of tenor, as I hope this afterword shows, can really screw up the coherence of your text.
Bex, T. (1996) Variety in Written English: Texts in Society: Societies in Text. London: Routledge.
Bhatia, V.K. (1993) Analysing Genre: Language Use in Professional Settings. London: Longman.
Bickerton, D. (1995) Language and Human Behaviour. Seattle: University of Washington Press.
Blue, G.M. (ed.)(1993) Language, Learning and Success: Studying Through English. London: Macmillan.
Chomsky, N. (1965) Aspects of the Theory of Syntax. Cambridge, Mass.: MIT Press.
- (1986) Knowledge of Language: Its Nature, Origin and Use. New York: Praeger.
Coulthard, M. (ed.)(1994) Advances in Written Text Analysis. London: Routledge.
De Beaugrande, R. (1993) "'Register' in discourse Studies: a Concept in Search of a Theory." In Mohsen Ghadessy (ed.) Register Analysis: Theory and Practice. London: Pinter.
Geeraerts, D. (1990) "The lexicographical treatment of prototypical polysemy." In Savas L. Tsohatzidis (ed.) Meanings and Prototypes: Studies in Linguistic Categorization. London: Routledge.
Gregory, M. and Carroll, S. (1978) Language and Situation: Language Varieties and their Social Contexts. London: Routledge.
Halliday, M.A.K., McIntosh, A. and Strevens, P. (1964) The Linguistic Sciences and Language Teaching. London: Longmans.
Halliday, M.A.K. (1978) Language as Social Semiotic. London: Edward Arnold.
Halliday, M.A.K. & Hasan, R. (1985) Language, Context and Text: Aspects of Language in a Social-Semiotic Perspective. Victoria: Deakin University Press.
Hunston, S. (1994) "Evaluation and organisation in a sample of academic written discourse." In Malcolm Coulthard (ed.) Advances in Written Text Analysis. London: Routledge.
Oshima, A. and Hogue, A. (1981). Writing Academic English (2nd ed.).Menlo Park, California: Addison-Wesley.
Kaku, M. and Thompson, J. (1997) Beyond Einstein: the Cosmic Quest for the theory of the Universe. Oxford: OUP.
Lakoff, G. and Johnson, M. (1980) Metaphors We Live By. Chicago: The University of Chicago Press.
Lakoff, G. (1987) Women, Fire and Dangerous Things: What Categories Reveal About the Mind. Chicago: University of Chicago Press.
Mauranen, A. (1993) Cultural Differences in Academic Rhetoric. Frankfurt: Peter Lang.
Swales, J. (1990) Genre Analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press.
White, R. and McGovern, D. (1994) Writing. Hemel Hempstead: Prentice Hall.
Wierzbicka, A.(1990) " 'Prototypes save': on the uses and abuses of the notion of 'prototype' in linguistics and related fields." In Savas L. Tsohatzidis (ed.) Meanings and Prototypes: Studies in Linguistic Categorization. London: Routledge.
Home | Philosophy Essays | Philosophy Links | Linguistics Essays | Linguistics Links | About me : | Journal | Journal | Guestbook