Hedging and Discourse: Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts [Reprint 2010 ed.] 9783110807332, 9783110155914

Citation preview

Hedging and Discourse

Research in Text Theory Untersuchungen 2ur Texttheorie Editor Jänos S. Petöfi, Macerata Advisory Board Irena Bellert, Montreal Antonio Garcia-Berrio, Madrid Maria-Elisabeth Conte, Pavia Teun A. van Dijk, Amsterdam Wolfgang U. Dressler, Wien Nils Erik Enkvist, Abo Robert E. Longacre, Dallas Roland Posner, Berlin Hannes Rieser, Bielefeld Hartmut Schroder, Frankfurt (Oder) Volume 24

W DE

G Walter de Gruyter · Berlin · New York 1997

Hedging and Discourse Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts Edited by Raija Markkanen and Hartmut Schröder

w DE

G Walter de Gruyter · Berlin · New York 1997

© Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability.

Library of Congress Cataloging-in-Publicafion-Data

Hedging and discourse : approaches to the analysis of a pragmatic phenomenon in academic texts / edited by Raija Markkanen and Hartmut Schröder. p. cm. — (Research in text theory = Untersuchungen zur Texttheorie ; v. 24) Collection of articles originally published 1991—1993. Includes bibliographical references (p. ) and index. ISBN 3-11-015591-5 (alk. paper) 1. Discourse analysis. 2. Pragmatics. 3. Language and culture. I. Markkanen, Raija. II. Schröder, Hartmut III. Series: Research in text theory ; v. 24. P302.H4 1997 401'.41-dc21 97-17338 CIP

Die Deutsche Bibliothek -

CIP-Einheitsaufnahrae

Hedging and discourse : approaches to the analysis of a pragmatic phenomenon in academic texts / ed. by Raija Markkanen and Hartmut Schröder. - Berlin ; New York : de Gruyter, 1997 (Research in text theory ; Vol. 24) ISBN 3-11-015591-5

ISSN 0179-4167 © Copyright 1997 by Walter de Gruyter & Co., D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of diis book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from die publisher. Printed in Germany Printing: Werner Hildebrand, Berlin Binding: Lüderitz & Bauer-GmbH, Berlin

Preface The concepts of 'hedge' and 'hedging' found their way from logic and semantics into the study of texts in the 1960s and have since been developed further within language pragmatics. These concepts seem to offer a way to fill an important gap in the study of discourse and text. In spite ofthat the concepts themselves still continue to be unclear and their use by linguists is far from uniform. This problem is apparent also in this volume: its aim is not to provide a systematic and unified introduction into this relatively new area of research. On the contrary, the volume presents several different views and approaches to 'hedge' and 'hedging', in order to give the reader an impression of the scope of theses concepts. It is very clear that no general agreement can be reached at the present stage of research in the area. However, the expanding use of these concepts requires an interim balance, at least in the form of an anthology, while waiting for a more conclusive treatment of the topic. The present volume is a collection of articles by authors who represent different languages and cultures as well as different perspectives to the topic. Most of the articles were written between 1991 and 1993 and reflect therefore the then prevailing state of discussion. As editors, we would like to thank the writers for their patience and cooperation. Because of various obstacles, the printing of the volume had to be postponed several times, and as a result some newer publications could not be taken into consideration. Therefore only the annotated bibliography includes publications up to 1995. We wish to thank Nancy Nyman, Jerry Schuchalter and Jessamyn Jackson for their native-speaker help in the editing of the articles. We also thank Stefan Krempl, Dagmar Zimmer, Elona Sagert and Angelika Hennecke for their invaluable assistance in the editorial work. Finally, our gratitude also goes to J.S. Petöfi for including this volume in the series Research in Text Theory.

R.M. and H.S.

Contents

Preface

V

I. Introduction Raija Markkanen and Hartmut Schröder Hedging: A Challenge for Pragmatics and Discourse Analysis

3

II. Hedging Strategies in Academic Discourse Paul G. Meyer Hedging Strategies in Written Academic Discourse: Strengthening the Argument by Weakening the Claim

21

John Skelton How to Tell the Truth in The British Medical Journal: Patterns of Judgement in the 19th and 20th Centuries

42

Vasili Namsaraev Hedging in Russian Academic Writing in Sociological Texts

64

III. Interactive Aspects of Hedging Avon Crismore and William J. Vande Kopple Hedges and Readers: Effects on Attitudes and Learning

83

Anna Mauranen Hedging in Language Revisers' Hands

115

Wolfram Wilss Hedging in Expert-Language Reviews

134

Contents

IV. Cross-cultural Aspects of Hedging William Grabe and Robert B. Kaplan On the Writing of Science and the Science of Writing: Hedging in Science Text and Elsewhere

151

Minna-Riitta Luukka and Raija Markkanen Impersonalization as a Form of Hedging

168

Tarja Nikula Interlanguage View on Hedging

188

Heinz Kreutz Some Observations on Hedging Phenomena and Modifying Devices as Regional Markers in the Speech of Young East Germans. .. 208 V. Hedging: The Concept, its Origins and a Bibliographical Guide Gudrun Clemen The Concept of Hedging: Origins, Approaches and Definitions

235

Hartmut Schröder and Dagmar Zimmer Hedging Research in Pragmatics: A Bibliographical Research Guide to Hedging

249

Name Index Subject Index

273 277

I. Introduction

RAIJA MARKKANEN and HARTMUT SCHRÖDER

Hedging: A Challenge for Pragmatics and Discourse Analysis

"If you hedge against something unpleasant or unwanted that might affect you, you do something which will protect you from it. If you hedge or hedge a problem or question you avoid answering the question or committing yourself to a particular action or decision." (Collins 1987) This description of the everyday meaning of the verb hedge suggests that as a linguistic term it might also refer to the choice of a certain kind of communicative strategy. This is actually the case in most uses of the term hedge in linguistic literature: it is possible to detect at least some affinity between its everyday meaning and the linguistic expressions referred to when using the term. However, there have been linguists who, like Skelton (1988), deplore the use of the term precisely because it reflects the connotations of the everyday usage of the word. Dictionaries of linguistics do not normally mention either the concept or the term 'hedge*. Exceptional in this respect are the Dictionary of Stylistic* and two German dictionaries of linguistics. Both German dictionaries use the term 'Heckenausdruck' and list it under the headings of semantics and lexicography. Their definitions for the term are the following: "Adjektivische oder adverbiale Wendung, durch die angegeben werden kann, in welchem Maße in einer gegebenen Sprache und dem dazugehörenden Kulturraum kategorisierende Aussagen im Sinne einer Repräsentativitätsskala möglich bzw. sinnvoll sind." (Bußmann 1990) and "Bezeichnung für Ausdrücke, die andeuten, in welchem Sinne bestimmte Exemplare einer bestimmten Kategorie zugeordnet werden. Aus der Tatsache, daß (jeweils relativ zu einem spezifischen kulturellen Hintergrund) manche Exemplare als bessere/typischere Beispiele einer Kategorie angesehen werden (...) ergibt sich ein Bedürfnis für solche Hecken." (Gippert 1993)

These definitions indicate that the formulation of concepts in everyday communication requires the use of hedges because concepts (e.g. 'bird') trigger prototypical images in people's minds, which makes it necessary to somehow mark their less prototypical representatives. Thus, if we mark a concept with a hedge, we do not refer to a prototypical representative of the class, but to a non-prototypical one. Therefore a sentence like A penguin is sort of bird is acceptable, but A raven is sort of a bird is absurd. Compared with the two German dictionaries, the Dictionary ofStylistics gives a somewhat different view. This dictionary, too, refers to the semantic origin of the concepts 'hedge' and 'hedging', but it classifies them as belonging to the

4

I. Introduction

fields of discourse analysis and speech act theory and defines them as "qualification and toning-down of utterances and statements (...) in order to reduce the riskiness of what one says". The motivation for their use is given as "mitigation of what may otherwise seem too forceful" and "politeness or respect to strangers and superiors". The use of hedge as a linguistic term goes back at least to the early 1970s, when G. Lakoff (1972) published his article Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts. Lakoff was not interested in the communicative value of the use of hedges but was concerned with the logical properties of words and phrases like rather, largely, in a manner of speaking, very, in their ability "to make things fuzzier or less fuzzy" (LakofF 1972, 195). Interestingly, however, Lakoff also briefly points out the possibility that hedges may "interact with felicity conditions for utterances and with rules of conversation" (Lakoff 1972, 213). In accordance with Lakoff's main concern, however, the term hedge has later been defined, for example by Brown/Levinson (1987, 145) as "a particle, word or phrase that modifies the degree of membership of a predicate or a noun phrase in a set; it says of that membership that it is partial or true only in certain respects, or that it is more true and complete than perhaps might be expected". This definition is interesting in that it includes in hedges both detensifiers and intensifiers, which was how LakofF also saw hedges. However, many users of the term limit it only to expressions which show that "the match between a piece of knowledge and a category is less than perfect" (Chafe 1986, 270). Since the early 1970s the concept of hedge has moved far from its origins, particularly since it has been adopted by pragmatists and discourse analysts. The term is no longer used only for expressions that modify the category membership of a predicate or noun phrase. As a matter of fact, Lakoff himself had already mentioned in his 1972 article Robin LakofFs observation that certain verbs and syntactic constructions convey hedged performatives (/ suppose/guess/think that Harry is coming; Won't you open the door?). The idea of hedged performatives became then one way of widening the concept of hedges. Fräser (1975) considered the effect that modals and semi-modals have on the illocutionary act denoted by a performative verb in performative sentences like I must advise you to remain quiet. Thus, for example, the modal must gets the speaker off the hook, relieves him/her from some of the responsibility. Fräser calls these cases 'hedged performatives', without naming the modals 'hedges'. In a later article (Fräser 1980), in which he discusses the mitigation of the harshness or hostility of the force of one's actions, Fräser seems to have adopted Lakoff's view of hedges and, accordingly, limits them to expressions like kind of, sort of. Other researchers, like House/Kasper (1981) and BlumKulka/Ohlstein (1984), have also discussed hedges as a means of modifying certain types of speech-acts, notably requests and apologies. In addition to the idea of hedged performatives, the concept was also widened in another way when hedges were taken to be modifiers of the speaker's commitment to the truth-value of a whole proposition, not just the category membership of a part of it. Thus, for example, Vande Kopple (1985) in his categorization of metadiscourse types considers the use of hedges as showing a

R. Markkanen and H. Schröder, Hedging

5

lack of full commitment to the prepositional content of an utterance. In other words, hedges (e.g. perhaps, seem, might, to a certain extent) are by him seen as modifying the truth-value of the whole proposition, not as making individual elements inside it more imprecise. This widening of the concept of hedge to contain the modification of commitment to the truth of propositions has led some researchers to think it necessary to distinguish between two types of hedges. This has been done for example by Prince/Frader/Bosk (1982) in their discussion of hedging in physician-physician discourse. They start from Lakoff's definition of hedges as devices that make things fuzzy, but add that there are at least two kinds of fuzziness. One is fuzziness within the prepositional content, the other fuzziness "in the relationship between the prepositional content and the speaker, that is the speaker's commitment to the truth of the proposition conveyed" (Prince/Frader/Bosk 1982, 85). Accordingly, there are two types of hedges: those that affect the truth-conditions of propositions, which Prince/Frader/Bosk call approximate™ (e.g. Hisfeet were sort of blue), and shields, which do not affect the truth-conditions but reflect the degree of the speaker's commitment to the truth-value of the whole proposition (e.g. I think hisfeet were blue). A similar distinction is drawn by Hübler (1983), who distinguishes between what he calls understatements and hedges, although both are devices used for expressing 'indetermination'. For example, a sentence like It's a bit cold in here is indeterminate. However, according to Hübler, there are two kinds of indetermination: phrastic and neustic. Phrastic indetermination concerns the prepositional content of a sentence, whereas the neustic type is connected with the claim to validity of the proposition a speaker makes. This distinction Hübler then carries over to distinguish between understatements, i.e. expressions of phrastic determination, and hedges, i.e. expressions of neustic indetermination. Thus, a sentence like It is a bit cold in here contains an understatement, while It is cold in Alaska, I suppose contains a hedge. Hübler's division thus greatly resembles that by Prince/Frader/Bosk, whose approximators correspond to Hübler's understatements and shields to his hedges. In both cases, it can be asked what end this division serves, what is gained by it besides making it easier for someone carrying out concrete analysis to limit his/her data. Hübler himself admits that both understatements and hedges perform the same function of expressing indetermination, of making sentences more acceptable to the hearer and thus increasing their chances of ratification. It is thus easy to agree with Skelton's (1988, 38) criticism of Prince/Frader/Bosk's division: it "seems to be sustainable only in the abstract: it looks more like a description of a property of text sentences than of language use". Furthest away from the original concept of hedge are those approaches in which hedges are treated as realizations of an interactional/communicative strategy called hedging. Thus, Markkanen/Schroder (1989; 1992), who discuss the role of hedges in scientific texts, see them as modifiers of the writer's responsibility for the truth value of the propositions expressed or as modifiers of the weightiness of the information given, or the attitude of the writer to the information. According to them, hedges can even be used to hide the writer's attitude. Markkanen/Schroder also suggest that hedges offer a possibility for

6

I. Introduction

textual manipulation in the sense that the reader is left in the dark as to who is reponsible for the truth value of what is being expressed (Markkanen/Schröder 1992). When this kind of purely functional starting point is adopted, there is no limit to the linguistic expressions that can be considered as hedges. Thus, Markkanen/Schröder consider as important devices for hedging - at least in scientific writing - the use of certain pronouns and avoidance of others, the use of impersonal expressions, the passive and other agentless constructions, in addition to the use of modal verbs, adverbs and particles, which are usually included in hedges. They also suggest that certain rhetorical and stylistic devices could be included. The difficulty with these functional definitions is that almost any linguistic item or expression can be interpreted as a hedge. Thus, for example, Darian (1995) claims that hedges can belong to any part of speech and lists as examples nouns, verbs, adjectives, adverbs, and even articles (One solution is...'). In this context it should be emphasized that no linguistic items are inherently hedgy but can acquire this quality depending on the communicative context or the co-text. This also means that no clear-cut lists of hedging expressions are possible. However, the advantage of functional definitions is that they make it possible to draw attention to an important aspect of communicative behaviour. As there are so many different ways that the concept hedge has been defined, some narrower than others, it is natural that there are other terms and concepts that come close to it, sometimes covering partly the same area of language use. The most important concept that cuts across the area of hedges is that of modality. Like definitions of hedge, those of modality also vary a lot in scope. According to Palmer (1986, 2), the notion of modality is vague and leaves open a number of possible definitions. A lot of the discussion on modality has concentrated on the modal auxiliaries (e.g. Palmer 1986) but modality has also been treated as "a central organizing principle in language" (Stubbs 1986, 4). If we consider Lyons' (1977, 797) definition of epistemic modality, "Any utterance in which the speaker explicitly qualifies his commitment to the truth of the proposition expressed by the sentence he utters, whether the qualification is made explicit in the verbal component (...) in the prosodic or paralinguistic component, is an epistemically modal or modalized utterance", we can see its affinity to many of the definitions of hedge, for example to Hubler's view of hedges as expressions of neustic indetermination, or to Prince/Frader/Bosk's shields. But it is not possible to include in Lyons' conception of epistemic modality hedges as defined by Lakoff and others who see them as modying parts of the proposition. However, even these hedges can be included within the realm of epistemic modality if we accept Stubbs' (1986, 5) view that "it is possible to indicate degrees of commitment to just three kinds of linguistic item: not only (1) to propositions but also (2) to illocutionary forces and (3) to individual lexical items". The last mentioned type, lexical commitment, is realized through items like so called, so to speak, quote unquote, which makes it possible to include also the Lakoffian type of hedges in modal expressions. Stubbs' view of modality allows even the inclusion of linguistic items and structures like logical and pragmatic connectors, past tense when used hypothetically, and passivization, which allows the deletion of the agent and therefore the avoid-

R. Markkanen and H. Schröder, Hedging

7

ance of commitment. This view of modality comes close to the functional, pragmatic definition of hedges referred to above (e.g. Markkanen/Schroder 1989; 1992). The concepts of modality and hedge thus overlap to a lesser or greater extent depending on their respective definitions. This connection is very clear in the case of modal verbs with epistemic meanings. When hedges are taken to be modifications of the commitment to the truth-value of propositions, for example the English modal auxiliary may is always listed as a typical example. Thus, in It may be true it is a hedge but also an expression of epistemic modality. Sometimes also the deontic meanings of modals allow interpretation as hedges. For example, in English the hypothetical would could be seen as a hedge because it makes an utterance non-categorical. Preisler (1986, 92) actually points out that "even when modal forms convey speaker-external meanings, these are often given interpersonal significance by the particular context in which they appear, usually as part of a tentativeness strategy". It seems possible to see the relationship between modality - mostly of the epistemic type - and hedges in two ways: either modality is the wider concept and includes hedges or the other way round, hedging is the umbrella term and epistemic modality a part of it. Another concept that cuts across the area of hedges - and epistemic modality - is evidentiality, again depending on how broadly hedge is understood. Chafe (1986, 271) defines evidentiality as "any linguistic expression of attitudes toward knowledge", i.e. assessment of its reliability. Knowledge, according to Chafe, has various modes: belief, induction, hearsay, and deduction, each of which is based on a different source. Most of the expressions that Chafe gives as examples of the realization of these different modes are expressions that have also been included in hedges by other linguists. Chafe himself uses the term hedge only of expressions that denote that "the match between a piece of knowledge and a category may be less than perfect" (Chafe 1986, 271), thus agreeing with LakofFs original idea of hedges. Vagueness is another concept close to hedging as it refers, among other things, to the use of expressions like about, sort of, i.e. expressions that denote the impreciseness of quantity, quality, or identity, which is very much like LakofFs "fuzziness" (cf. Channell 1990; 1994; Zuck/Zuck 1985). Similarly, von Hahn (1983, 99) uses the term 'Vagheit' (vagueness) when referring to the false assumption that the language in scientific texts is exact. Linguists who, like Chafe, use the term hedge in a more limited way need other concepts/terms to cover the rest of the phenomena included in hedges in the more extensive definitions. A further example of this is House/Kasper's (1981) use of the term 'hedge' only of adverbials with which the speaker avoids precise prepositional specification (e.g. kind of, sort of, somehow, rather). They then distinguish several other types of what they call modality markers, like understates, which are used to underrepresent the state of affairs denoted by the proposition (e.g. a little bit, not very much) and downtoners (e.g. possibly, perhaps), which are used to modulate the impact an utterance is likely to have. As to the motivation for the use of hedges, a lot of the discussion has concentrated on their use in spoken discourse, and the most frequently mentioned motivating factor is politeness, as defined by Brown/Levinson (1987). In their

8

I. Introduction

view, hedges are mainly used for negative politeness in face-saving, in which they are put to elaborate use. In positive politeness they figure only in expressions of extremes, like marvellous and appalling, which are typical of this form of politeness, 'safely vague' because they leave it to the addressee to figure out how to interpret them. In negative politeness strategies hedges are used, according to Brown/Levinson, to hedge on the illocutionary force of an utterance or on any of the four Gricean Maxims. In both cases the motivation for their use is the desire to save face, either the hearer's or the speaker's. Although Brown/Levinson (1987, 67) say that it is possible to distinguish between acts that primarily threaten the hearer's face and those that threaten the speaker's own face, they admit that the latter acts are also potential threats to the hearer. Thus, in their discussion of politeness and the ways to express it, it is the hearer's face-wants that get emphasized. It is, however, possible to turn up the other side of the coin and emphasize the importance of hedges for the speaker's own face. Their use may be motivated, for example, by the fear of being proved wrong later on. Being imprecise or mitigating one's commitment to the truth-value of a proposition or a claim makes it possible to say, if proved wrong, that the claim was only tentative or an approximation. This explanation is supported by Hübler's (1983) view that the reason for using hedges is to make sentences more acceptable to the hearer and thus increase their chances of ratification, which in turn is made necessary by the inherent negatability of sentences. According to Hübler, the function of hedges is to reduce the risk of negation. Thus, it can be claimed that, in all communication, while showing deference to the addressee, the speaker or writer also tries to protect him/herself from potential anger, contempt or other humiliation on the part of the addressee. Both desires are present in all communicative situations, but one may be stronger than the other in a particular case. House/Kasper (1981,157) seem to share this view when they say that "both these functions - one defensive and ego-oriented, the other protective or alter-oriented are fulfilled by politeness". Thus, in some situations the desire to protect oneself from the potential denial of one's claims may be greater than the desire to show deference to the addressee. The surer a speaker feels about his/her own position vis-a-vis the interlocutor, the less need there is for hedging for the purposes of self-protection. Related to the above issue are the claims made by some linguists that gender has an influence on the use of hedges, meaning that women hedge more than men. Thus, R. Lakoff (1975) lists hedges as one of the features typical of women's speech, as one way of sounding feminine and thus reflecting their position in society. Preisler's (1986) study of linguistic sex roles also shows that linguistic tentativeness signals are correlates of both sex and interactional role, which is either task-oriented or socio-emotional. It is the latter role that correlates with tentativeness. As can be expected, Preisler's data show that women lead in the use of hedges in general, whereas men lead in linguistic assertiveness. Preisler therefore concludes that sex is a more general determinant of speech than the interactional role, which means that "men and women have developed sex-specific speech patterns" (Preisler 1986, 288). However, there is no general agreement on this; other linguists (e.g. O'Barr/Atkins 1980) have

R. Markkanen and H. Schröder, Hedging

9

suggested that, rather than being a feature of women's speech only, the use of hedges is a marker of 'powerless language' in general, i.e. of the way language is used by people who are in an inferior position in society. Holmes' (1990) findings in her study of gender differences in the use of hedges and boosters disagree even more strongly with LakofPs and Preisler's claims: Holme's data show that hedges are actually frequently used by women "as positive politeness devices signalling solidarity with the addressee, rather than as devices for expressing uncertainty" (Holmes 1990, 202). Also, in a study on the use of metadiscourse by American and Finnish university students in argumentative writing Crismore/Markkanen/Steffensen (1993) found that Finns used more hedges than Americans, i.e. a cultural difference, and that Finnish men used them more frequently than women. The issue is thus more complex than has been assumed. There has also been a growing interest lately in hedging and the motivation for its use in academic/scientific writing. That hedges are actually used in scientific discourse, which is supposed to be above all rational and neutral, is connected with the fact that scientific discourse obeys the same mechanisms as ordinary everyday communication does, although it tries to hide this, more or less successfully, by using a code of its own. We assume here that science is not only content; that is, scientific texts are not only content-oriented and informative but also aim at convincing and influencing their audience. According to classical rhetoric, formulation of a scientific text is not merely built on pragma and on docere (instructing, informing) but also on dekctare (entertaining) and movere (moving, enchanting). Thus, in addition to 'going into the subject matter* (pragma), a text should also emphasize the reliability of the author (ethos) and also move the reader emotionally (pathos). The last two, ethos and pathos, are closely connected with the expressions used in the text, including hedging. The rhetorical style of a scientific text is not merely a decorative addition to an otherwise informative text; rather form and content are inseparable, as Stolze (1992, 232) suggests in the following quotation: Hermeneutisch gesehen sind die rhetorisch-stilistischen Mittel, von denen eine leserbeeinflussende Wirkung ausgeht, nur ein Teil der Form, die mit dem lediglich analytisch abzutrennenden Inhalt eine unlösbare Einheit bildet. Texte lassen sich zwar didaktisch in diese beiden Elemente aufteilen, aber weder der Inhalt noch die pure Form hätten für sich eine Substanz, welche jene Wirkung hervorbringen könnte.

It should also be emphasized here that the stylistic devices used, including hedges, acquire their meanings through the author-reader interaction, on the basis of the text and the commucation situation. We can thus come to the conclusion that a hedge is 'a textual phenomenon' and 'a virtual quality' of a text. Hedges are of course present at various levels of textualization but are not inherent characteristics of texts. As noted above, a text does not contain hedges per se but gets them through the author-reader interaction. This apparent subjectivity is, however, to some extent controlled by culture, since people who belong to a particular language community normally share socially determined aesthetic ideals through their shared educational background (cf. Spillner 1974, 67). But in an intercultural communication situ-

10

I. Introduction

ation, cultural differences may cause misunderstanding and communication breakdown, because the author and the reader do not share the same norms and expectations. As in other types of communication, in scientific writing politeness has been seen as the motivating factor for hedging. Myers (1989) claims that even in this type of discourse hedging is used for the sake of negative politeness, more specificially, to mark a claim "as being provisional, pending acceptance in the literature, acceptance by the community" (Myers 1989, 13). He even suggests that an unhedged claim, or what looks like a claim, is "probably not a statement of new knowledge" (Myers 1989, 13). Furthermore, according to Myers, hedges reflect a relation between the writer and the readers, rather than the degree of the probability of the statement. Although Myers (1989, 4) suggests that everyone, regardless of their position, must appear as humble servants of the discipline, it can still be assumed that the amount of hedging writers employ depends on such factors as their position within the scientific community, the potential readership, and even the writers' personalities, i.e. how sure or unsure they feel about their own position within the field (cf. Markkanen/Schroder 1989; 1992). Furthermore, it could be assumed that in academic writing the use of hedges varies according to the field the writer represents, i.e. that there are scientific fields in which hedging is more frequent than in others. It could be expected that the texts of fields like linguistics and philosophy, for example, would contain more hedging than the texts of natural sciences and technology because of the different bases of argumentation in these fields. Argumentation in philosophy is not based on bringing in experimental data and concrete evidence, as in natural sciences and technology. As Spillner (1983, 35) points out, in texts in which the use of experimental data and logical deduction are not so important, the style of writing becomes an essential element in achieving credibility. The convincingness of an argument in such texts depends on the use of linguistic devices, including hedges (Markkanen/Schroder 1989). However, the use of hedges is not alien to writers in science and technology either. Dubois (1987, 15) discusses the myth of perfect objectivity in science noting that "it has been shown that the scientist can unobtrusively inject his personal views into his communication" by using linguistic items that express uncertainty or impreciseness. Similarly, Butler (1990) shows the importance of modals in English biological and physical texts, in which they "serve the weighing of evidence and the careful drawing of conclusions from data, in the making of claims from evidence, and more particularly in making generalizations about what is possible in the behaviour of the universe, on the basis of observation of what actually happens" (Butler 1990, 139). In his article dealing with hypotheses in introductory science texts, Darian (1995, 101) also comes to the conclusion that "hedges are probably the clearest indicators of hypotheses" and gives an extreme example from a text in geology: (Although the steps in the creation of oil are still very poorly known,) the following (1) simplified theory is (2) rather (3) widely held and is supported by enough facts to be (4) at least (5) somewhat (6) near the truth. (Darian 1995, 102) These findings suggest that the differences in the use of hedges between texts in different fields are not so great as has been often assumed. It could even be

R. Markkanen and H. Schröder, Hedging

11

claimed that hedging in academic fields, regardless of the scientific paradigm, actually serves the struggle for objectivity, which the scientific community often considers a necessary characteristic of scientific writing. Prince/Fraser/Bosk (1982), who have analysed spoken discourse among physicians, suggest that their frequent use of hedges demonstrates a scholarly orderliness in the representation of knowledge. It was suggested above that the amount of hedging may depend on the individual language user's position vis-a-vis the audience and even on his/her personality. There is thus room for individual style in the use of hedges even in academic writing. A skillful writer may also use them for his/her own specific purposes. In his analysis of F.R. Leavis' literary-critical text "The Great Tradition", Simpson (1990) shows that, contrary to expectations, Leavis modalizes (i.e. hedges in our terminology) information that is not "risky" or controversial and leaves controversial information unmodalized. The effect of this is, as Simpson (1990, 91) points out, that "important questions concerning the canon of English literature can be glibly passed over and, at the same time, attention can be deflected toward issues that are, by contrast, less significant and more peripheral". The pattern of hedging (or modalization in Simpson's terminology) used by an individual writer becomes really significant when we think of the effect that it may have on readers, of how it directs the readers to evaluate the information. Thus, in the case of an influential writer like Leavis, it might even lead students "to challenge all sorts of commonly held assumptions about the canon of English literature" (Simpson 1990, 92). This also supports the view proposed above that hedging is not an inherent characteristic of a text but a product of writer-reader communication, i.e. the linguistic expressions used in hedging get their meaning through the response they produce in the readers. An interesting new challenge for research on hedging is provided by the new postmodern concept of science and its influence on discourse and, along with it, on hedging as a device of textualisation. Geist (1992) criticizes the research on language used in academic writing for its still continuing use of key words like economy, precision and explicitness, although words like complexity, uncertainty and ambiguity are currently in use elsewhere. Geist sees the situation as paradoxical: a lot of the research on scientific writing is still firmly based on the idea of specialization, which was problematized in scientific thinking already in the 1980s (Geist 1992, 234). According to Geist, the problem is that our understanding of reality and with it our concept of what constitutes a scientific approach have changed profoundly and found a completely new orientation. These new tendencies have also had an effect on scientific thinking, which is now characterized, among other things, by the problematization of the commonly accepted. The 'new discourse' that arises from the above tendencies seems to be in contrast with what has until now been considered as self-evident, particularly in the Anglo-American text conventions. In his analysis of the 'new discourse' in the humanities, Laermann (1991) comes to the conclusion that things like problem, argument and evidence are not valid anymore: Bis über die Schmerzgrenzen der Sinnlosigkeit entfernen sie sich von zwei elementaren Voraussetzungen traditioneller Wissenschaft: von Themen und

12

I. Introduction von Argumenten. Über lange Passagen hinweg verzichten sie zum einen darauf, sich nur einem Thema zu widmen. Statt dessen wenden sie sich eher Problemkomplexen zu, die kaum eindeutig zu umreißen sind. (...) Wenn der herrschende Betrieb (oft fragwürdig genug) Prämien für Interdisziplinarität verteilt, dann sind die neuen Unklaren darüber jeweils schon hinaus; denn sie bewegen sich in den Sphären einer Transdisziplinarität, die keinerlei Grenzen anerkennt. (...) Die innere Organisation undeutlicher Texte ist zum anderen gekennzeichnet durch einen weitgehenden Verzicht auf Argumentationen.

In the same vein Lepenies (1993, 18) writes of a fusion and mixture of genres. It would be interesting to find out whether the use of hedges and the attitude to hedging have changed as a result of these developments in postmodern science. It could even be assumed that hedging might play an essential role in the new discourse. As the above discussion suggests, hedging is an important interactional strategy both in spoken and written communication. Thus, to be effective communicators we should acquire a feel for its appropriate use in different communication situations. It is therefore no wonder that - in addition to the analysis of hedges and hedging for their own sake - there have been studies of a more applied kind. These studies aim at finding ways of making communication more effective through providing communicators in different areas of life with both knowledge about hedges and the opportunity to develop their skill in using them. This practical aim is evident in studies of hedging at least in the following areas: (i) (ii)

(iii) (iv) (v)

contrastive/cross-language analysis and comparative stylistics; translation studies: for example, studies concerned with target-language oriented translation, studies on the equivalence of effect between source and target texts, problems connected with text types in different cultures, and 'untranslatable' linguistic and cultural phenomena; studies on teaching writing in the mother tongue and die readability of texts; studies on foreign language teaching, particularly on teaching writing in a foreign language, which also includes contrastive rhetoric, i.e. the comparison of culture-bound text conventions; intercultural communication studies, e.g. studies on the reception of texts written in a foreign language.

As the above list indicates, studies on the use of hedges are particularly relevant for language teaching, both in the mother tongue and foreign languages. The difficulties experienced in the appropriate use of hedges may partly be due to the lack of teaching of this area of language use. Students are not taught to modulate their propositions, not even sensitized to the issue. The use of hedges in writing may even be discouraged, perhaps because many of the words and phrases used as hedges are seen as empty fillers. Judging by the guidebooks for good writing, these items may be commented on in passing but not systematically taught. Teaching the appropriate use of hedges, like other pragmatic phenomena can be very problematic for several reasons. One reason is that, as suggested above, hedges get their meaning through the contexts in which they occur. Another reason is that their use is often connected with the speakers/writers' values and beliefs, even their personalities, which makes teaching them a delicate matter. Therefore, some kind of awareness-raising may be the only possible method of teaching in this area of language use.

R. Markkanen and H. Schröder, Hedging

13

As using hedging expressions appropriately may cause problems in some communicative situations - like academic writing - even in the mother tongue, it is no wonder that it is problematic in a foreign language. This is because the rules of appropriateness vary across cultures. Consequently, to become effective communicators in a foreign language, learners have to acquire these rules, particularly since transfer from the mother tongue is possible also in this area. The skillful use of hedges, which requires subtlety and sophistication even in the mother tongue (Skelton 1988), is clearly part of a language user's pragmatic competence, lack of which may lead in foreign language use to mistakes that are more serious than, for example, grammatical errors. This is because pragmatic errors are not so Obviously erroneous' as faulty syntax; they only make the foreign language user sound, in the case of hedging, more impolite or agressive, more tentative or assertive than he/she intends to be, which then may even lead to a communicative failure, (cf. Thomas 1983) There are some interlanguage studies that suggest the kind of problems foreign language learners might have in the use of hedges. In a study on the interlanguage of German learners of English, Kasper (1979) suggests that a kind of modality reduction takes place in the foreign language learners' speech. This means that modality is present at an early stage of the learner's speech act planning but does not for some reason occur in its surface realization. This reduction, she claims, is "a consequence of low awareness of modality as a pragmatic category" (Kasper 1979, 274). Other researchers have found another, more concrete, type of reduction in foreign language learners' expression of modality: their repertoire of expressions is more limited when compared with that of the native speakers of the language. For example, Kärkkäinen (1990) found that Finnish learners of English use fewer expressions of epistemic modality than do native speakers of English. They have less variation in the expressions and stick to a few favourite ones like / think. Foreign language learners also seem to prefer the more explicit kind of modification to the implicit ones. According to Färch and Kasper (1989), this preference is due to the fact that the explicit expressions (e.g. / think) are longer and have their own propositional content and illocution. The above mentioned interlanguage studies are all concerned with spoken discourse, but of course problems with hedges occur also when writing in a foreign language, and they seem to be similar. Thus, according to the findings of Ventola/Mauranen (1990), Finns writing in English also showed the tendency to stick to a few 'safe' expressions of epistemic modality, had less variation in the expressions than did native speakers of English, i.e. they did not behave in a native-like manner in this respect. Clyne's (1991) study of German scholars writing in English confirms the assumption mentioned above that there may be transfer from the mother tongue and one's native culture in the use of hedges. According to Clyne, who has looked at the amount of hedging in academic texts, German writers hedge more both in their native language and in English than do native speakers of English. They even use double and triple hedging. In other words, they follow the norms of their native culture when writing in a foreign language. To sound native-like in a foreign language, a speaker or writer should, then, have a rich repertoire of hedging expressions at his/her disposal and use an appropriate amount of them.

14

I. Introduction

Another area in which the differences in the cultural conventions concerning hedging cause problems is translation. If the requirement is accepted that a good translation must not only preserve the meaning of the original but also produce in a reader an equivalent response to that produced by the original text, adjustments in hedging may be necessary. If the cultural norms have caused the writer of the original text to use a lot of hedging and the translator has not reduced their amount to suit the norms of the target culture, the result may be an irritatingly tentative, uncertain text. The opposite can of course also happen: the translated text may sound much too assertive to people used to more 'hedgy' texts. This is a real problem in the case of scientific texts, which must often be translated into a foreign language, most frequently into English these days. To be accepted into an international journal using English, the text should conform to the 'Anglo-American' cultural norms. The core of the translation problem is, however, that the translator should know the intentions of the original writer, i.e. how assertive or tentative s/he intended to be. This knowledge may be impossible to acquire, and the translator does not dare to make adjustments in hedging. In the light of the problems hedges cause for learning to write effectively, for foreign language learning and for translation and revision of texts, it is obvious that much more research is needed. What seems to be required are more studies on cross-cultural differences, on the effects of hedges on readers, more comparisons of hedging phenomena in texts of different academic fields, of different genres, and their treatment and role in translation. From the methodological point of view, we see the following as possible points of departure for future systematic research on hedges - we are thinking here in the first place of hedging in written communication, particularly in the academic/scientific field. Firstly, the starting point could be text as product, i.e. the focus could be on hedges as text-internal phenomena. Secondly, the starting point could be text as process, where the focus would be on the text-external factors that have an effect on text-production, including taking into consideration the potential readership. A third possibility would be to treat text as interaction, which would lead to studies on the reception of texts and their influence on readers. Naturally, it would also be interesting to combine all the three perspectives in one and the same analysis, which would give a more complete view of hedging. From the more practical point of view, i.e. for the purposes of intercultural communication, language teaching and translation, we see as relevant at least the following types of studies, in addition to the self-evident contrastive/crosslinguistic studies: (i) (ii) (iii) (iv)

comparisons of codified stylistic conventions and criteria for 'good style', particularly in the area of academic/scientific communication; examination of non-codified criteria for 'good style' and quality criteria for scientific texts; analysis of texts written in a foreign language and of translated texts; description and comparison of how texts are received by readers who have the same cultural background as the author and those who do not.

The relatively new area of research on writing in a foreign language uses the methods of text analysis but also tries to analyse the writing process itself. In

R. Markkanen and H. Schröder, Hedging

15

the latter approach, the influence of the conventions used in writing in the mother tongue are of great interest. It would be a relatively simple task to compare the codified stylistic conventions for example in guidebooks for writing and advice given by the editors of scientific journals in different languages. This approach is actually very close to contrastive rhetoric, the basic assumption of which is described by Martin (1992) as follows: "the rhetorical organization of Lj written texts is the result of the transfer of Lj rhetorical forms for any language". This strong assumption is later moderated (Martin 1992, 3): "Lj rhetorical norms and strategies, while undoubtedly influential for the shape of 1^ written texts, are but one factor influencing the text shape". Purves (1988, 13) seems to agree with this latter view when he suggests that, although the rhetorical devices used in textualisation are culture dependent, they are not necessarily dependent on the language of the writer. The conventions of the discipline in question may also have their role to play in scientific writing. What we hope to have accomplished with the above introductory discussion is to show the importance of the phenomenon that we have called 'hedge/ hedging' for different areas of language use. We readily admit the difficulty connected with the concept. As was shown above, its use originates in logic and semantics, but has lately been developed further in pragmatics and discourse analysis so far that it now extends to areas like metacommunication and to communication strategies like mitigation and politeness. Through this extension the concept has lost some of its clarity and sometimes seems to have reached a state of definitional chaos, as it overlaps with several other concepts. This problem concerns many other linguistic concepts and their definitions, beginning with the concept of 'language' itself. However, the concept 'hedge' has the advantage of being usable for phenomena for which no other suitable concept exists. All the contributions in this volume adhere to this extended sense of the concept, some extending it more than others, but none of them restrict its use to the narrow meaning given to it by its 'founding fathers'. The articles also demonstrate the various perspectives - some more theoretical, some more practical - from which hedges and hedging can be considered, thus suggesting directions for future research.

References Blum-Kulka, Shosana; House, Juliane and Kasper, Gabriele 1989 "Investigating cross-cultural pragmatics: An introductory overview", in Blum-Kulka, S., House, J. and Kasper, G. (eds.): Cross-cultural Pragmatics: Requests and Apologies. Norwood, N.J.: Ablex, 1-36. Blum-Kulka , Shoshana and Ohlstein, Elana 1984 "Requests and apologies: A cross-cultural study of speech act realization patterns", Applied Linguistics, Vol. 5/3, 196-213. Brown, Penelope and Levinson, Stephen 1987 Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bußmann, Hadumod 1990 Lexikon der Sprachwissenschaft. Stuttgart: Kröner.

16

I. Introduction

Butler, Christopher S. 1990 "Qualification in science: Modal meanings in scientific texts", in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse (Written Communication Annual 3). Newbury Park: Sage, 137-170. Chafe, Wallace 1986 "Evidentiality in English conversation and academic writing", in Chafe, W. and Nichols, J. (eds.): Evidentially: The Linguistic Coding of Epistemology Norwood, N.J.: Ablex, 261-272. Channel, Joanna 1990 "Precise and vague quantities in writing on economics", in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse (Written Communication Annual 3) Newbury Park: Sage, 95-117. Channel, Joanna 1994 Vague Language. Oxford: Oxford University Press. Clyne, Michael 1991 "The sociocultural dimension: The dilemma of the German-speaking scholar", in Schröder, H. (ed.): Subject-oriented Texts (Research in Text Theory 16). Berlin; New York: de Gruyter, 49-67. Collins 1987 Collins CobuildEnglish Language Dictionary. London: Collins Publishers Crismore, Avon; Markkanen, Raija and Steffensen, Margaret S. 1993 "Metadiscourse in persuasive writing: A study of texts written by American and Finnish unversity students", Written Communication, Vol. 10, No. 1,39-71. Darian, Steven 1995 "Hypotheses in Introductory Science Texts", //ML, Vol. XXXIII/2, May 1995, 83-108. Dubois, Betty Lou 1987 "Something in the order of around fourty-four: Imprecise numerical expressions in biomedical slide talks", Language in Society, 16/4, 527-541. Farch, Klaus and Kasper, Gabriele 1989 "Internal and external modification in interlanguage request realization", in Blum-Kulka, S., House, J. and Kasper, G. (eds.): Cross-cultrual Pragmatics: Requests and Apologies. Norwood, N.J.: Ablex, 221-247. Fräser, Bruce 1975 "Hedged performatives", in Cole, P. and Morgan, J.L. (eds.): Syntax and Semantics Speed Acts, Vol. 3. New York: Harcourt Brace and Jovanovich, 187-210. Fräser, Bruce 1980 "Conversational Mitigation",/o#m?/ of Pragmatics, 4:4, 341-350. Geist, Uwe 1992 "Allez-Hopp! Fachsprache und Paradigmenwechser, in Grindsted, Annette and Wagner, Johannes (eds.): Communication for Specific Purposes - Fachsprachliche Kommunikation. Tübingen: Narr, 243-266. Gippert, Jost 1993 "Heckenausdruck", in Melder-Lexikon Sprache. Edited by Helmut Glück. Stuttgart; Weimar: Metzler. Hahn, Walther von 1983 Fachkommunikation: Entwicklung, Linguistische Konzepte, Betriebliche Beispiele. Berlin; New York: de Gruyter. Halliday, Michael A.K. 1973 Explorations in the Functions of Language. London: Edward Arnold. Holmes, Janet 1982 "Expressing doubt and certainty in English", RELC Journal, 13/2, 9-28. Holmes, Janet. 1990 "Hedges and boosters in women's and men's speech", Language and Communication, Vol.10/1, 185-205.

R. Markkanen and H. Schröder, Hedging

17

House, Juliane and Kasper, Gabriele 1981 "Politeness markers in English and German", in Coulmas, F. (ed.): Conversational Routine. The Hague: Mouton, 157-185. Hübler, Axel 1983 Understatements and Hedges in English (Pragmatics and Beyond IV: 6) Amsterdam: John Benjamins. Kasper, Gabriele 1979 "Communication strategies: Modality reduction", The Interlanguage Studies Bulletin, Vol. 4/1. Utrecht, 266-283. Kärkkäinen, Elise 1990 Modality as a Strategy in Spoken Interaction. Unpublished licenciate thesis. Oulu: University of Oulu. Laermann, Klaus 1991 "Fiat Nox! ", in Die Zeit, No. 32, 42. Lakoff, George 1972 "Hedges: A study of meaning criteria and the logic of fuzzy concepts", in Peranteau, P., Levi, J. and Phares, G. (eds.): Papers from the Eighth Regional Meeting of Chicago Linguistic Society. Chicago: Chicago University Press, 183-228. Lakoff, Robin 1975 Language and Woman's Race. New York: Harper and Row. Leech, Geofitey N. 1983 Principles of Pragmatics. London: Longmans. Lepenies, Wolf 1993 Die Übersetzbarkeit der Kulturen - Ein europäisches Problem, eine Chance für Europa. Unpublished paper. Lyons, John 1977 Semantics. Cambridge: Cambridge University Press. Markkanen, Raija and Schröder, Hartmut 1989 "Hedging as a translation problem in scientific texts", in Lauren, C. and Nordman, M. (eds.): Special Languages: From Human Thinking to Thinking Machines. London: Multilingual Matters, 171-175. Markkanen, Raija and Schröder, Hartmut 1992 "Hedging and its linguistic realization in English, German and Finnish philosophical texts: A case study", in Nordman, M. (ed.): Fachsprachliche Miniaturen. Frankfurt/Main et al.: Peter Lang, 121-130. Martin, James 1992 Towards a Theory of Text for Contrastive Rhetoric. New York et al.: Peter Lang. Myers, Greg 1989 "The pragmatics of politeness in scientific articles", Applied Linguistics 10, 1-35. O'Barr, William and Atkins, Beryl 1980 "'Women's language' or 'powerless language'?", in McDonnell-Ginet.S.; Borker, R. and Furman, N. (eds.): Women and Language in Literature and Society. New York: Praeger, 93-110. Palmer, Frank R. 1986 Mood and Modality. Cambridge: Cambridge University Press. Preisler, Bent 1986 Linguistic sex Roles in Conversation: Social Variation in the Expressions ofTentativeness. Amsterdam: Mouton de Gruyter. Prince, Ellen F.; Frader, Joel and Bosk, Charles 1982 "On hedging in physician-physician discourse", in Di Pietro, R.J. (ed.): Linguistics and the Professions. Norwood, N.J.: Ablex, 83-97. Purves, Alan C 1988 "Introduction", in Purves, Alan C. (Hg): Writing Across Languages and Cultures. Issues in Contrastive Rhetoric (Written Communication Annual 2). Newbury Park: Sage, 9-21.

18

I. Introduction

Simpson, Paul 1990 "Modality in literary-critical discourse", in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse (Written Communication Annual 3). Newbury Park: Sage, 63-94. Skelton, John 1988 "The care and maintenance of hedges", ELT Journal, vol. 42/1, 37-43. Spillner, Bernd 1974 Linguistik und Literaturwissenschaft. Stilforschung, Rhetorik, Textlinguistik, Stuttgart: Kohlhammer. Spillner, Bernd 1983 "Methodische Aufgaben der Fachsprachforschung und ihre Konsequenzen für den Fremdsprachenunterricht", in Kelz, H.P. (ed.): Fachsprache l: Sprachanalyse und Vermittlungsmethoden. Bonn: Dümmler, 16-29. Stolze, Radegundis 1992 Hermeneutisches Übersetzen. Linguistische Kategorien des Verstehens und Formulierens beim Übersetzen. Tübingen: Narr. Stubbs, Michael 1986 "A matter of prolonged field work: Notes toward a modal grammar of English", Applied Linguistics, Vol.7/1, 1-25. Thomas, Jenny 1983 "Cross-cultural pragmatic failure", Applied Linguistics, Vol. 4, 91-112. Vande Kopple, William 1985 "Some exploratory discourse on metadiscourse", College Composition and Communication 36, 82-93. Ventola, Eija and Mauranen, Anna 1990 Tutkijat ja englanniksi kirjoittaminen. (= "Researchers and Writing in English") Helsinki: Yliopistopaino. Zuck, Joyce Gilmour and Zuck, Louis V. 1985 "Hedging in Newswriting", in Comu, A.-M., Vanparjis, J.; Delahaye, M. and Baten, L. (eds.) Beads or bracelet? How do we approach LSP?, Selected papers from the 5th European Symposium on LSP. Oxford: Oxford University Press, 172-180.

II. Hedging Strategies in Academic Discourse

PAUL GEORG MEYER

Hedging Strategies in Written Academic Discourse: Strengthening the Argument by Weakening the Claim 0. Introduction

The term 'hedging* is one of the slightly facetious, creative terminological innovations that we owe to the revolutionary spirit of American linguistics in the early seventies of this century. The topic was introduced into the debate not on its own account, but because certain expressions used as hedges (such as strictly speaking, technically speaking, regular, sort of) have a diagnostic value in analysing the semantics of certain linguistic expressions (Lakoff, G. 1972). As it often happens, the discussion of the notion has evolved far from its origin. In recent years, attention seems to have shifted away from the role of hedging in oral discourse, a topic prevailing in much of the literature on hedging in the eighties (Hübler 1983, Bradac/Street 1989/90, Herman 1989), to hedging in academic discourse (Butler 1990, Clyne 1991, Johnson 1992, Markkanen/ Schroder 1989, Myers 1988, Simpson 1990). There is a paradoxical difference between the two types of discourse with regard to the role of hedging in them. While in oral face-to-face communication hedging is regarded as a hallmark of'powerless' speech-style (Bradac/Street 1989/90), in written academic discourse it may serve to strengthen the argument. What is paradoxical about this difference is that the evaluation of hedging as signalling weakness or strength should be distributed in this way and not vice versa. After all, everyday face-to-face interaction typically serves to establish, modify and maintain interpersonal relationships rather than to discuss a topic or solve a problem exhaustively and efficiently. This should give more weight to politeness phenomena and should make hedging more desirable and acceptable. Academic discourse, on the other hand, at least as judged by its own proclaimed epistemological standards, should aim at producing statements that are falsifiable. Claims should be made as strong as possible. The stronger they are, the easier they are to falsify. Hedging typically weakens the strength of speech acts, making them less 'negatable' (Hübler 1983, 23)1. So why should academic discourse resort to hedging strategies? It will be one goal 1

Hübler actually uses a different terminology the details of which need not concern us here. His formulation can, however, be translated into Searlean terminology in the way I have phrased it. The term strength is used in the sense of Searle/Vanderveken 1985. Degree of strength in this sense is basically what distinguishes, e.g., an order from a suggestion, or a vow from a promise.

22

Π. Hedging Strategies in Academic Discourse

of this paper to offer an explanation for this apparent paradox by showing how the academic world has developed its own forms and standards of hedging in accordance with prevalent ideas about the nature of academic research and communication. It will also be seen that genuine hedging is much rarer in written academic discourse than might be expected from statistical counts of modal and similar expressions (cf. Butler 1990, 167). When we look at hedging phenomena in written academic texts, we find that most of the common devices of colloquial language do not only not occur, but are downright unacceptable in this register, cf. (1), which is the beginning of the 'JT section in the LOB corpus, with its modified version (Γ): (1)

Unfortunately the accuracy with which an impurity dependent physical or chemical property of sodium can be measured decreases with decreasing impurity concentration. To get over this difficulty Alcock has suggested that instead of measuring direct ly the concentration of oxygen in the flowing sodium its thermodynamic potential should be measured by a suitable galvanic cell incorporated in the circuit. |SJ01|Pl-22

(Γ)

I'm afraid the accuracy with which an impurity dependent physical or chemical property of sodium can be sort of measured decreases with decreasing impurity concentration, as you may call it. To get over this difficulty it was Alcock, I think, who suggested diat instead of measuring directly the concentration of oxygen in the flowing sodium maybe its, technically speaking, thermodynamic potential should be measured by a sort of galvanic cell sort of incorporated in the circuit.

(Γ) is clearly unacceptable as a written academic text. Instead, we find formulations like the following: (2a) (2b) (2c) (2d) (2e) (2f)

(2g)

This is a very much over-simplified example, but it may serve to emphasise the point that common criteria of adaptation often contradict each other. |SJ22|P1 This is an extremely difficult question to answer but, possibly rashly, I am inclined to say 'yes'. |SJ22|P52 This is a finding the authors do not mention, and it leads one to suspect that the school population in Geneva which they examined consisted of able children. |SJ24|P73 ...it is hard to see what explanation could be offered for this at present. |SJ25|P24 ...the comparative success of two dictatorships and the failure of the second republic in this century might be adduced in support of this view if one were concerned with political theory. |SJ27|P72 Although small changes from quarter to quarter as shown in the national income accounts must be treated with caution because of the 'statistical discrepancy', there is every reason to regard as correct the view held at the time in official circles that disposable income and consumption expenditure for goods as well as services was being well maintained despite the recession. |SJ43|P4 I hope that some of my findings may serve to illustrate...|SJ22|P66

The quotations in (2) represent, admittedly, extreme examples of hedging from academic texts in different fields of knowledge. But they serve to exemplify a wide range of strategies used and illustrate three points about academic hedgAll examples are from the 'J' section of the LOB corpus of present-day English. In what follows it will be referred to as 'the subcorpus' or 'LOBJ1. Examples are identified by the number of the text fragment (SJ01) and number(s) of sentence(s) within that fragment (Pl-2). For further information on the corpus see Leech 1987.1 wish to thank Dieter Mindt, Felicitas Tesch, Dietrich Lange, Hans Peters, and Edgar Werner Schneider for their friendly assistance in making the corpus accessible to me.

P.G. Meyer, Hedging Strategies in Written Academic Discourse

23

ing: The forms of hedging are manifold, hedgin g expressions tend to group themselves into elaborate formulae, and it is not easy to delimit a class of hedging expressions. But it should be noted that the examples are not representative or typical of written academic text. They were merely the most conspicuous examples found in a superficial scanning of concordances of modals and suspicious performative verbs in the corpus investigated. It also showed that elaborate explicit hedging formulae such as in (2) are relatively rare in the corpus. Assuming that the effect of hedging devices is face-saving (Myers 1988), either for the author or for the reader, we might explain this by the following argument: It is one of the secrets of face-saving strategies that they are the more effective, the less recognizable they are (Hübler 1983, 167). Academic discourse abounds in face-saving strategies which are extremely effective, just because they are not easy to recognize. This may partly be so because there are no unique lexical signals for them which would enhance recognition, partly because there are excellent other excuses to use them. This can also be shown by looking at the examples in (2). Let me take some of them as the starting point to discuss the different mechanisms and strategies of hedging. The discussion will be confined to those typical of written academic discourse. Each example highlights a different aspect, and some will give occasion to draw in further examples.

L The interlacing of strengthening and weakening devices (2a)

This is a very much over-simplified example, but it may serve to emphasise the point that common criteria of adaptation often contradict each other. |SJ22|P1

The point to be made in (2a), "common criteria of adaptation often contradict each other" (it is not important for us to understand this point to be able to analyse the hedging around it), is a statement, obviously a generalization. This generalization is said to be emphasised by an example given in the previous context. So what accounts for the hedging effect of the formulation? Examples are not, strictly speaking, hedges. And the author states explicitly that he wishes to strengthen ("emphasise") the point he is making. The example may then be seen as raising the strength of the speech act even further. It is no longer a mere statement, but a statement confirmed by an example. But examples also reduce "the writer's responsibility for the truth value of a proposition" (Markkanen/Schroder 1989, 172) by invoking facts that 'speak for themselves'. And this in turn counts as a typical hedging quality. This characteristic of hedges in academic discourse will be found in many cases: an inextricable contradictory interwovenness of strengthening and weakening of illocutionary force. The weakening effect is heightened in the present example by at least five further qualifications of the example given: (1) It is not claimed that the example actually "emphasises" the generalization. It is merely said that the example "may serve to emphasise" the generalization. By making the relationship between example and generalization more indirect, the exemplifying effect and thereby the strength of the generalization is further weakened. (2) The indirect

24

II. Hedging Strategies in Academic Discourse

relationship between example and generalization is still further weakened by the modal expression "may", leaving open the possibility that the example "may" not emphasise the point the author is trying to make. (3) The example, in turn, is qualified by a well-known politeness strategy ('be pessimistic', Brown/Levinson 1979, 320), as "simplified", and this qualification, in turn, is strengthened by two intensifiers, (4) the prefix over- and (5) the boosted booster very much. The author is warding off potential reader criticism in advance by, so to speak, humiliating himself, strongly depreciating the value of his example. So the source of the hedging effect is probably to be found here: By weakening the example that is meant to strengthen the generalization, the generalization is ultimately weakened. The weakening, however, has a paradoxical strengthening effect because it makes the statement, as it stands, impregnable. The question remains what use there could be for such weak but unassailable statements. But this is a question of epistemology, not linguistics. 2. Volitional hedges: inclination and desire as conditionsfor speech acts (2b)

This is an extremely difficult question to answer but, possibly rashly, I am inclined to say 'yes'. |SJ22|P52

Here we have another extreme case of hedging. The point is to find an answer to a question whose precise nature need not concern us here. The answer given is not a straightforward 'yes', but is multiply hedged on different levels. The first level is the speech-act level, where we find a performative verb (say) instead of a mere statement. The performative verb by itself is not a hedge, but it is used to build a hedge by the formula I'm inclined to say, which is an extremely cautious formulation, bordering on a near-refusal to perform the speech act at all. It looks like a very weak form of a certain type of hedged performative analysed by Fräser (1975, 202f.) in rather a roundabout way. According to him, expressing a desire to perform a speech act (I-wouldlike to suggest that...) may be taken as a hedged, indirect performance of that speech act because it is taken as indirectly asking the hearer for permission to perform the act, and, by some principle of efficiency, this permission is taken for granted unless the hearer voices some opposition to the speech act. This taken-for-grantedness, in turn, makes both the hearer's giving explicit permission and the speaker's explicit performance of the speech act redundant. Searle's analysis of performatives indicates a somewhat simpler solution. According to Searle (1989), it is a characteristic (in fact, the only general characteristic) of performative verbs that they denote an intentional action which may be performed simply by manifesting (not necessarily uttering) the intention to perform the action. In other words, performatives are like magic in that they allow an agent to bring about a state of affairs simply by manifesting the intention to bring about that very state of affairs. Thus, if you manifest your intention to suggest/claim/promise that p, you have in fact suggested/claimed/promised that p, and no further inferences are necessary. Searle's analysis leaves open the question what the difference could be between an explicit performative and a performative hedged by a verb expressing an intention. But it is clear that, even though expressing

P.G. Meyer, Hedging Strategies in Written Academic Discourse

25

an intention to perform a certain speech act is sufficient for performing that speech act, it is not necessary for doing so, and it remains an indirect way of performing the act, whereas an explicit performative is a clear case of a direct speech act. The above example now looks somewhat different from examples of expressed desire. To begin with, an inclination is even weaker than a desire. Manifesting an inclination implies less of a commitment then manifesting a desire or an intention. An inclination to perform a speech act just falls short of being sufficient for the performance of that act. But what is more important is that expressing an inclination presupposes a different kind of situation from a situation of expressing a desire. Expressing a desire typically presupposes some kind of constellation in which the fulfilment of the desire is in some way facethreatening to the hearer. Expressing an inclination presupposes a hearer who is in some way interested in my inclination: someone who wants to know about my inclination, or someone who has an interest in my performing a certain action I might be inclined to perform. In this case the potential face threat is to the speaker. The case of I'm inclined to say is thus completely different from the cases analysed by Fräser by way of asking permission to perform the act. It does make sense to ask permission for an act that is possibly face-threatening to the hearer. But it would be absurd to ask permission for a speech act that is possibly face-threatening to the speaker. On the other hand, inclinations certainly form a scale of volitional strength together with desires, wishes, and intentions. So we might still wish to analyse example (2b) along the same lines as other cases of volitional modality. This type of analysis is compatible with a Searlean explanation of volitionally hedged performatives, but not with Fraser's analysis. Fraser's analysis is excluded because different types of volition imply different burdens efface threat for speaker and hearer, and Fraser's analysis demands that the face threat is always on the hearer's side. Searle's analysis operates without assumptions about face threats or inferences on the basis of politeness considerations, and simply assumes that a certain strength of volition is sufficient for effectively performing a speech act. Now if a speech act, as we said above about the I'm inclined to say formula, falls short of this sufficient strength, then we may assume that some inference mechanism must be at work. The example is then comparable to cases of ability hedging (I can say..., see below) in that it is not sufficient (ability is never a sufficient condition for reality). I'm inclined to say thus means that given certain conditions the probability that the act will be performed is greater than the probability that some other act will be performed. The hedging formula leaves open the possibility that some condition might arise which precludes the performance of the speech act (in our case: an assertion) and suspends any commitment of the speaker to the conditions of satisfaction (in our case: the truth) of its prepositional content. But there is more hedging in (2b). Not only does the author stop short of actually performing the speech act in question, he or she also qualifies this would-be speech act by a parenthetical admission that the inclination to perform the act (or the act itself?) is "possibly rash". Again we find the politeness strategy of self-criticism and pessimism (Brown/Levins on 1979, 320), typically used in speech acts strongly face-threatening to the hearer. In Gricean terms we

26

II. Hedging Strategies in Academic Discourse

might say that the author is questioning whether the speech act in question fulfils the second maxim of quality: "Do not say that for which you lack adequate evidence" (Grice 1975, 46). A "rash" statement is most probably one that the speaker lacks adequate evidence for because there wasn't enough time to probe into it. The third layer of hedges around the "yes" in (2b) is provided by the excuse that the question to which it could serve as an answer is "extremely difficult to answer", thus asking the hearer for mercy upon and indulgence to an author faced with such a formidable task. The excuse also relates to the pessimistic evaluation of the speech act as "possibly rash". An "extremely difficult question to answer" naturally takes some time to answer, so that any answer provided within a few sentences might appear "rash".

3. Inherently weak performatives: the case of suspecting as mental state and speech act The next example contains fewer words of hedging, but it includes another strategy that seems important: (2c)

This is a finding the authors do not mention, and it leads one to suspect that the school population in Geneva which they examined consisted of able children. |SJ24|P73

The plain statement about the children examined in Geneva (by Inhelder and Piaget, by the way) is hedged by a performative verb (suspect) that has a weakening effect in itself. Note that we found performative verbs in both examples discussed so far (emphasise, say). But emphasise is actually the opposite of a hedge because it raises the strength of the speech act. We may say that similar to insist, the function of emphasise is not to bring about a particular illocutionary force or point, but to strengthen the force of an underlying illocutionary point which is brought about by other means: emphasise may not only enhance assertions, but also invitations, offers, and even directives, say is about the most neutral performative verb we can get; it is not bound to one specific illocutionary point. We may say that it is a performative verb of the propositional act (Searle 1969, 22ff), rather than of the illocutionary act. The performative verb we find in (2c), suspect, is not a full-fledged specified illocutionary verb either. It is one of those hybrid expressions that vacillate between speech act and mental state. Strictly speaking, suspecting is not the performance of a speech act, but the utterance of a mental state. But it may be used performatively in that when you say Ί suspect that...', you have uttered a suspicion, and in some, however weak, respect committed yourself to the truth of this suspicion. The verb may also be used in reported speech: (3)

Sherlock Holmes suspected that Moriarty perpetrated the crime.

(3) is truly ambiguous: It may refer to a private suspicion of the master detective, or report an utterance he made in front of others. The proof that (3) may be used to report an utterance independently of a mental state, is that (3), which is in the past tense, may be uttered while the suspicion is still lingering

P.G. Meyer, Hedging Strategies in Written Academic Discourse

27

and the speaker knows this. When (3) refers to the mental state, it would have to be transformed into the present tense in this situation. In any case, suspect as a mental state is less than a belief (which is the appropriate mental state for uttering a sincere assertion). Thus when you utter a suspicion, you perform a weak form of assertion, if not less than an assertion. Nevertheless, the illocutionary point is clearly assertive, because it is the truth of the propositional content and the speaker's sincerity that is at stake, although commitment is not as strong as in a typical assertion. The point of using a performative such as suspect is thus to reduce the author's commitment to the truth of the proposition. In (2c), this commitment is further reduced. The author states that he is "led to suspect", that is, again the author is seeking an excuse for the performance of the speech act. The suspicion uttered is not just a whim, but is founded in a "finding" that leads one to "suspect" a certain state of affairs. This strategy of shifting responsibility away from the author to "the facts" is typical of another form of hedging: speaking facts. In the present case the author pictures himself as the object of an objective situation that takes the role of agent: it leads (!) the author to suspect. 4. Hedging explanations: a case study 4.1. Offering as a weak subsidiary speech act (2d)

...it is hard to see what explanation could be offered for this at present. |SJ25|P24

(2d) is our first example without a formulation that could be taken as performative. It is not the performance of any of the speech acts referred to in it (offer, explanation}, but, cautiously hedged, a refusal to perform them. It could be said that it thus represents a more extreme form of hedging than hedged performatives: No particular speech act is performed, except that some hypothetical speech act is considered, discussed, or refused. This, by the way, is one of the most common uses of speech act expressions in the subcorpus. The speech act that is in the centre of attention here is an explanation, one of the most sacred speech acts in science. Going through the concordances of explain and explanation, we find that explicit explanations are put forward extremely cautiously in the subcorpus. Of the 74 occurrences of the two lexemes in LOB_J, 53 refer to scholarly explanations. Of these, 27 references are negative, such as in the above example, or completely hypothetical or sceptical, such as in the following examples: (4a) (4b)

If there are also cells sensitive to curvature of a line a similar mechanism would explain the sort of finding depicted in figure 1 (b). SJ25|P60 The explanation of these results is not easy to find. SJ38|P68

Only two occurrences could be classed as completely unhedged, one quasi-performative, the other announcing: (4d) (4e)

This explains why the richest party does not always win elections in Britain or America. SJ41|P77 The mechanisms of para-, ferro- and ferri-magnetism will then be explained and reference made to the temperature behaviour of the saturation magnetisation of certain ferrites. SJ70JP97

28

II. Hedging Strategies in Academic Discourse

Six references involve a presupposition3 that an explanation was performed elsewhere, such as in: (4f)

This explanation of FAE is based on sound physiological evidence and is so simple that it seems highly convincing. It does not, however, explain mere displacements in apparent spatial position occurring as a FAE: for this phenomenon, the Osgood and Heyer type of explanation appears reasonably plausible. SJ25JP65-66

18 of the 53 references to academically relevant explanations in the subcorpus are in some way performative,4 but they mostly involve hedges of differing strength. Since this is one of the largest figures for hedged performatives that we get in one speech act verb, the occurrences of explain can be conveniently used to exemplify the full range of hedging phenomena around performatives: apart from the more elaborate formulae that we have been discussing, we find several other devices. A satisfactory explanation may be conceptualized as a certain quantity of something that is said to be sufficient. One way of weakening the force of an explanation may then be to say that the 'explanatory quantity' is not quite sufficient. Thus in (4g)

It may be pointed out that in our model productivity makes a significant contribution to the explanation of the spread between earnings and wage rates, when all variables are expressed as are expressed as levels, but ceases to be a significant factor in our least squares computation in which variables are subjected to a first-difference transformation. SJ44|P45

what is offered is only "part" of the desired explanation. A similar quantitative conceptualization may be found in (4h)

The new pattern evolving, therefore, may roughly be explained in terms of a correlation between the height of the building and the income group and the degree of density of population in the parish. SJ27|P28

where a qualification ("roughly") is used that is also possible in speaking about quantities. The metaphor is, however, spatial rather than quantitative: the desired "ideal" explanation is conceptualized as having a certain outline which is traced only "roughly" by the explanation actually put forward. A quantitative interpretation is also possible for the "some" in (4i)

A complete understanding of this differential sexual response is obviously impossible, but a clearer examination of the method of questioning suggests some explanation and makes possible an assessment of the significance of this finding on the validity of the results on birth control methods. SJ31|P20

But we also find two hedging strategies that we have already encountered: First, the strength of the explanation is weakened by the inherently weak performative verb suggest. This verb signals a reduced illocutionary commitment on the side of the author. Second, it is not a personal agent that is said to perform this act, but again some fact. Just like in (2c) the author was led to suspect by some facts, in the present example it is facts that suggest something to the author. 3 4

Presupposing a speech act is another way of shifting responsibility, of reducing the 'negatability' (cf. Hübler 1983) of a claim. It will not be dealt with any further in this paper. See below for some qualification of this statement.

P.G. Meyer, Hedging Strategies in Written Academic Discourse

29

Another weak illocutionary force that is routinely associated with the speech act of explaining is Offer' (s. (2d) above). Although the overall effect of weakening illocutionary force is similar to the case of suggest, the mechanism is slightly different. The author, instead of imposing an explanation on the reader, is merely making an offer to the reader. In ordinary face-to-face interaction, an offer is slightly more face-threatening to the hearer than a suggestion because it involves a question by the speaker whether a certain commitment is wanted by the hearer and makes the commitment dependent on the hearer's assent. An offer thus demands a definite immediate reaction from the hearer, more strongly than a suggestion, which allows for an evasive answer. An offer that is answered evasively ('Well, let's see...') is no more felicitous than an offer that is turned down. In a suggestion, it is the weakness of the commitment that accounts for the hedging effect; an offer, on the other hand, may imply a fairly strong, though conditional, commitment, but it leaves options to the hearer, and this is what is crucial in our connection here. The reader who is offered an explanation may accept it or reject it. The author's commitment is not on the agenda at all. A suggested explanation is simply an explanation that is put forward with reduced commitment. It could be said that offering an explanation is an even more sophisticated form of hedging than suggesting because it involves less explicit self-disclosure from the author: In suggesting, the author has to commit him- or herself to low commitment; in offering, the author would be committed to the offer, but not to a specific degree of commitment to the validity of the explanation offered. Note now that an offer is never performed explicitly: in (2d) above, the offer is hypothetical, and in the following example, a hypothesis, not a person, is said to offer the explanation: (4j)

It [= a hypothesis mentioned several sentences before, P.G.M.] also provides a basic factor in the causation of phonetic change...and offers an explanation of many of its observed characteristics. |SJ32|P22

In (4j) the author includes himself in the group of addressees of an offer made by the hypothesis. He thus presents himself as being in a position to choose whether to accept the offer or not. This is the lowest degree of commitment that can be imagined: The author is neither committed to the truth nor to the falsehood, nor to a specific degree of commitment to either of them. It is only the principle of relevance that helps us to interpret this formulation: Why should anybody mention that a hypothesis offers an explanation for something, if this offer has no relevance to anybody? Mentioning an offer in a text can be relevant by way of accepting it, rejecting it, or discussing it. The offer in (4j) is neither rejected nor discussed at length. This would require special marking. We must therefore assume by default that it is, however provisionally, accepted.

4,2, Tools: acting but not speaking

Besides the facts there are other items in the research process to which the responsibility for explanations and other speech acts can be shifted:

30 (4k)

II. Hedging Strategies in Academic Discourse It is, I take it, hardly necessary for me to sum up my view that we are still far from having, either in theory or in statistical analysis, techniques which enable us to explain satisfactorily the main features of international trade. SJ46JP48

When the author speaks of "techniques enabling us to explain" something, we are placed at a different point in the chain of causation leading from the facts to our knowledge about the facts. Facts can only explain something to us when we see them. According to the causal theory of perception (Searle 1983, 63f), which underlies much of our everyday reasoning about perception, felicitous seeing involves a causal chain from the object seen to whatever we perceive. In the case of a scholarly investigation, this causal chain may be mediated by all kinds of instruments, methods, research activities, preliminary hypotheses, and complex cognitive processing. Anything in this chain may be said to enable us to see whatever is there to be seen. And when the facts can only be seen by means of some specific techniques, it is these techniques that are the most relevant link in this causal chain. This is how it can happen that techniques enable us to explain, because they enable us to see whatever seems explanatory to us. But note that the techniques do not themselves explain, as the facts may do, but they only enable us to explain. Though techniques may be represented as agents, they do not speak. This is the usual way of dealing with instruments in grammar: Instruments may replace the agent in physical activities (and the use of the verb enable in (4k) reminds us of the physical character of a prototypical action). But they are limited to such activities in which they figure as immediate instruments. They cannot turn up as metonymic agents of processes in which they could not figure as instruments. To make this more conceivable, imagine the following simplified situation: Suppose you want to find out something, and a possible way of finding it out is to open a certain door. In this case it is admissible to say: (5a)

A look behind this door will tell me what I want to know

or even (5b)

Opening this door will tell me what I want to know

or (5c)

What I'll find behind this door will tell me what I want to know.

But now suppose you need a certain key to open the said door. It would be odd to say: (5c)

*This key will tell me what I want to know,

although it is perfectly natural to phrase (5d)

This key will enable me to find out what I want to know.

We may thus see the close parallel between prototypical instruments in everyday situations and methods or techniques in the academic world. A different link of the same causal chain is involved in the following exampie: (41)

This knowledge explains the very high mortality rate in drowning. SJ16|P17

P.G. Meyer, Hedging Strategies in Written Academic Discourse

31

Knowledge is the last step in this causal chain, the result of clear and distinct perception. And of course it is only when we know about facts that they can explain anything to us. 4.3. Necessity as excuse Necessity is often invoked in everyday life to mitigate a face threat to the hearer, when we can think of no better way to legitimize our act or claim: (6a) (6b) (6c)

I simply have to have it! I could not help doing it! I have got to speak to you!

In the same way, problematic claims in academic discourse may be put forward with more than usual strength. One way of heightening the strength in this way is using a modal expression of deontic necessity. The idea behind this strategy is similar to the speaking world metaphor: Necessity is the most objective modality that leaves no option to the individual. It may be asked why this strategy is mentioned in an article on hedging phenomena. After all, necessity is a strengthening, not a weakening device: A claim put forward in the modality of necessity is much stronger than a modally unmarked claim. But then again we may say that the modality of necessity is a way of shifting responsibility and commitment away from the speaker, that is the function it has in the examples under (6). Modal necessity is common as a hedge with many performative verbs. Fräser (1975, 196) explains that it is natural with speech acts that are not welcomed by the hearer and by the speaker (/ must order you to leave is ridiculous when the hearer wants nothing more than being given leave and hypocritical when the speaker is most eager to get rid of the hearer). Let us, accordingly, look at some examples from the subcorpus. It is to be expected that deontic necessity is not particularly frequent in explanations since an explanation is usually welcomed by the reader. The examples that we do find thus have to be explained as special cases: (7a) (7b) (7c)

The intense psychological shock which is the immediate consequence of another's injured body has really to be explained on more than aesthetic grounds. |SJ52|P78 Indeed the only plausible explanation seemed to be one associated with the length of schooling. |SJ38|P24 The brusqueness of the refusal, the failure to take advantage of what was generally sensed in Versailles to be the beginning of the end, seems so out of keeping with Bismarck's desire to renew peace negotiations that the explanation must surely be sought in Bismarck's attitude to the earlier exchanges between Moltke and Trochu. |SJ57|P19

Examples (7a) through (7c) invariably reject alternative explanations as much as putting forward new ones. These alternative explanations were possibly the ones imputed to the reader. Rejecting a belief that the reader may entertain is certainly face-threatening to the reader and thus in need of some confirmation and hedging. Note how necessity alone is not left to convince the reader: The necessity hedges are in turn hedged by mitigating devices such as really (7a), plausible (7b), and surely and be sought (7c).

32

II. Hedging Strategies in Academic Discourse

Necessity is more at home in mathematical contexts: it is here that the rules of logic operate most inexorably, and it is an explicit goal to formulate theses in such a way that no other choice is left. But even here, although necessity can be spelt out without further hedging, it is only invoked when an undesirable step has to be defended: (7d)

...in order to obtain...we must invoke a further axiom. |SJ51|P70

Invoking a "further" axiom always looks undesirable because it seems to contradict Occam's razor: the fewer axioms are necessary to obtain a result, the better the solution. It is the conflict between a venerable epistemological principle and the peculiarities of the case under discussion that evokes the use of must here. Again the facts of the case play an active role; although they are not mentioned explicitly in the sentence, they implicitly act as the source of the necessity. As far as face-work is concerned, it may be said that violating the epistemological rules of simplicity is a threat to the author's positive face: the more elegant the solutions, the better the reputation. Deontic necessity may not only serve to save the author's positive face, but also the reader's negative one: (7e)

However, it must be re-emphasised that we have included hours worked in our computation, which are an indicator of the direct influence of demand on the spread. |SJ44|P47

Emphasis, whatever means may be applied by a speaker to bring it about, is an imposition on the hearer, especially when you emphasise the point that you are emphasizing by using an explicit performative formula. Under normal circumstances, just plainly saying what you want to say must suffice. Emphasizing your point may offend the hearer because it may look as if you regard the hearer as incapable of grasping what you are saying unless you emphasise it. It is for this reason, presumably, that we never find the plain performative */ emphasise... Explicit emphasis is always hedged: either in the roundabout way we saw above in (2a), or by using a volitional modality manifesting an intention to perform the speech act (I would like to emphasise..), or, like in this case, by claiming necessity for the speech act. The same can be said about the semantically close verb stress. Apologizing for the imposition on the hearer is all the more necessary in (7e) as the emphasis is, above all, a repetition ("re-emphasised"). Explicit repetitions never go without an apology in our subcorpus, a typical example being: (7f)

It is inevitable that I repeat some of the arguments...|SJ35|P24

Further speech act verbs that show some affinity to the modality of necessity as a hedging device are add (we must add), admit (it must be admitted), conclude (it must be concluded), regard (must be regarded as...), recall and remember (it must be recatted/remembered). Adding and recalling are probably regarded as impositions on the hearer similar to repetitions. Admissions are clearly face-threatening to the author; necessity in this case is what prevails against individual face wants. In conclusion processes, logical necessity is assumed to be at work, and this may be emphasised by an appropriate modal expression. Apart from that,

P.G. Meyer, Hedging Strategies in Written Academic Discourse

33

hedging by deontic must is not very common (32 instances in all) and seems to be a matter of personal style.5 One reason for this may also be that a more cautious variant of necessity is available in English: the modal should. Its use in academic texts is often nothing but an understatement of deontic necessity. It may thus be regarded as two hedges in one word: The modality of necessity serves as an apology towards the reader (hedge number one), and the choice of should instead of must brings about the understatement (hedge number two), should is used in many of the semantic contexts were we also found must in a hedging function. It accounts for further 10 instances of hedging by deontic necessity (not counting the occurrences where hedging is brought about by an epistemic meaning of should, signalling a certain degree of probability). Here are some examples: (7g) (7h) (7i) (7j) (7k)

...it should be borne in mind that small concentrations of hydrocarbons dissolved in liquid oxygen do not necessarily present a hazard. |SJ79|P11 It should be clearly appreciated that these equations are in no sense based on classical mechanics |SJ05|P32 ...it should be stressed that the probabilities refer to fully specified distributions |SJ18|P21 It should be emphasised that, while these paradoxes were formulated in terms of deterministic dynamics, they were not immediately dissipated by the advent either of quantum theory or...|SJ18|P48 It should be noted that C.F.O. do not produce forecast charts of the 500-200 mb thickness...|SJ10|P6

(7k) is special because in this context, should cannot be replaced by must. This is because (7k) implicates an instruction which includes the reader. It seems that noting is more of an imposition than remembering or bearing in mind. Using the verb note is a fairly powerful strategy of directing the reader's attention, so that must in combination with note imposes an obligation that seems too harsh to the polite author. Deontic necessity is even harsher then the plain imperative; note is one of the few imperatives that regularly occur in subject-oriented texts. That asking to note counts as a relatively strong imposition is also bome out by some collocations of note in the corpus, among which indications of importance and interest are prominent (it is interesting/important to note...): the author is trying to persuade the reader to note the argument by drawing attention to its importance.6 4.4. Ability to perform the speech act as preparatory condition (8a)

An extensive, though local, inundation can easily be explained... |SJ02|P42

It is a common strategy for an indirect speech act to mention a preparatory condition of ability. In order to perform a speech act felicitously, certain abili5

Four of the 80 text fragments in the subcorpus alone contain 15 of the 32 instances of hedging by must. It should also be noted that there are other means of expressing strict necessity, such as the adjective necessary, or the formula there is no denying. The negative counterpart of necessity is usually expressed by cannot or impossible. *> See also the semantic analysis of note and other mental verbs presented by Schneider (1988, 119f.); cf. also Meyer, P. G. (1997, § 5.4).

34

II. Hedging Strategies in Academic Discourse

ties beyond mere linguistic ones are always required. In particular, when the speaker undertakes a commitment to do certain things (such as in an offer or promise), ability preconditions are often mentioned (I can give you a lift downtown). Uses of the verb explain have some affinity to offers or promises because it is not guaranteed that whatever they claim as explanans actually explains the explanandum. So we might say that the verb explain cannot perform an explanation, but only offers or promises it. Explanations as acts do not have the quality ascribed to performative verbs by Searle (1989, 538): that the use of the verb describing the act in appropriate conditions guarantees the performance of the act it denotes. Given this, it would not be surprising if can might occur with explain in conditions strongly resembling an offer or a promise.7 What is further to be noted is that (8a) emphasises the ease of explanation. This looks like rather an ingenious combination of self-advertisement and face-work by displaying modesty: An offer may be played down and made more acceptable by emphasizing the ease with which it may be realized (It's absolutely no problem for me to do this for you}. With scientific explanations, there is something else involved: the easier the explanation, the better the theory and the scholar who is offering it. But now look at (8a) in a larger context: (8a')

With regard to the deluge , let me say that it is its world-wide occurrence which makes physical difficulties. An extensive, though local, inundation can easily be explained, but where did die water issue from and to where did it retreat to if there was enough to cover the whole surface? |SJ02|P41-42

The wider context of the text shows that the point of the can be explained formula is not a hedged explanation, nor a promise or offer of an explanation: It is really a statement about the ease of explanation of a certain phenomenon, in this case, a flood. In fact, the explanation is regarded as so easy that the author does not bother to give it. This example should teach us to be careful with classifying modals accompanying performatives as hedging. Expressions of ability, in particular, are often true statements of ability, not polite performatives; also look at the following: (8b)

Physiological occurrences are blind; as such they can be described without reference to anything else including the thing wanted, or the objective of the endeavour. |SJ54|P21

The author in (8b) is not setting out to describe anything, but is reasoning about the possibility of describing certain events. The following examples, involving other speech acts, are less misleading: (8c) (8d) (8e)

Summarizing the introduction, it can be stated that the anomalous electrical proper ties of films are principally due to their structural imperfections |SJ08|P2 In die case of the humus podzols which are found in general adjoining peat or certainly in the lowest position in the catena, it can be convincingly argued that development has been influenced by ground-water. |SJ03|P104 Around Tisbury itself die numerous Pordandian quarries show a variety of dips. Some apparendy can be ascribed to false bedding. |SJll|P62-3

7 Note in this connection die phraseological character of the collocation offer an explanation/

P.G. Meyer, Hedging Strategies in Written Academic Discourse (8f)

35

What can be said is that, after a number of gestures, some friendly, some ambiguous, towards the school board in its hour of crisis, Sidney became confirmed...|SJ39|P45

(8c) through (8f) are truly performative. And they do not merely offer to argue or state or ascribe: By stating the ability to perform the speech act, they perform it. This can be shown by the fact that it would not be grossly unfair to quote the authors of the above passages using the same speech act verbs that they use performatively: (8c') (8d') (8e') (8f)

Summarizing the introduction, S states that... In the case of the humus podzols which are found in general adjoining peat or certainly in the lowest position in the catena, S argued that... Around Tisbury itself the numerous Portlandian quarries show a variety of dips . Some are ascribed to false bedding by S. S says that...

Fräser (1975, 199ff.) distinguishes three categories of ability-based hedging in performatives: One occurs where the ability to perform the speech act is new (typically marked in the hedging formula by now). The speech act is so desirable that it is performed as soon as the speaker is able to perform it, and the circumstances are such that an announcement of the ability is taken as the performance of the act. This seems to be the case in (8c): You cannot summarize before you have reached the end of a passage, and a summary is always desirable; this is so self-evident that a "facilitating adverbial" such as now (Fräser 1975, 201) is unnecessary. The case of (8d) through (8f) is somewhat different: The ability to argue in the way indicated in (8d) or to ascribe whatever is ascribed in (8e), etc., is certainly not represented as new. And the other categories provided by Fräser do not fit either. The modal expressions are best explained as a defensive reaction to anticipated critical questions of the form: "What can be said about this case?", implying that the reader might question the ability of the author to provide an argument. But since this whole 'dialogue' is completely imaginary, the formula acquires a hedging quality. And indeed, it seems a reasonable strategy for anybody wanting to reduce the negatability of claims, to engage in an imaginary dialogue with the reader, anticipating imagined critical questions. A somewhat different case is represented by the following example: (8g)

I can therefore sympathize with (though I am sceptical of) the proposals by Bohm and de Broglie |SJ18|P3

The modal hedging in (8g) is not so much addressed to the reader as reader, but to the reader as competing researcher or author.8 It is the same use of can that can be observed when a request is fulfilled only partly. The author in (8g) is discussing other researchers' proposals. Obviously, he does not fully agree. The modal can is best regarded as a polite concession to colleagues, meaning: 'This is as much as I can do for you'. However, in the usual case, modal ability hedges around performatives are not mere politeness rituals. They remain true statements about ability which it also makes sense to negate and to question: 8 See Myers (1988, 5) for a discussion of the significance of this distinction.

36

II. Hedging Strategies in Academic Discourse

(8h) (8i)

Now because of collisions the value of the energy can not be precisely stated for a time much greater than t |SJ05|P13 The questions I wish to pursue, therefore, are: can we accept the methods of these four inquiries as satisfactory and dependable procedures for establishing the 'just' wage; and, if we can, how wide is the scope of their application? |SJ45|P32

Compare this with some other hedging formulae: (2gl) (2g2) (2g3) (7gl) (7g2)

?? I hope that some of my findings may not serve to illustrate... ?? I do not hope that some of my findings may serve to illustrate... ?? May some of my findings serve to illustrate...? ?? It should not/need not be borne in mind that small concentrations of hydrocarbons dissolved in liquid oxygen do not necessarily present a hazard. ?? Should we bear in mind that...?

The manipulated examples show that no real assertions are involved in the formulae used in (2g) and (7g). In some formulae, negation and questioning are admissible, but do not seem to have any effect whatsoever on illocutionary force: (7el) (7e2)

However, it need not be re-emphasised that... However, do we need to re-emphasise that...?

The use of the performative verb in (7e) is so formulaic that regardless whether you affirm, deny or question the speech act, you cannot help performing it.9

5. Hedging the certainty of findings: ability, possibility, probability

A further common formula to be considered in connection with ability-based hedging can be found in contexts where the occurrence of a phenomenon is to be reported, but the source of the knowledge can or need not be mentioned for some reason: (9a)

Lumps of a very dark and hard ferruginous sandstone, recalling a tropical laterite, can oho be found with ironstained Purbeck slabs in the surrounding arable fields. |SJ11|P72

The meaning of the formula in (9a) differs very little from an unhedged formulation (...are alsofound..). In the strict sense of Hübler (1983, 23), the modal is not to be interpreted as hedging, as weakening the strength of the assertion, but rather as understating the certainty of its truth. It is as if the writer wanted to save the face of possible other researchers that went over the mentioned fields and did not find the mentioned sandstone. But it should be clear that reducing the degree of certainty to mere possibility is also read as a weakening of the assertion. But again this formula should be approached with caution. There are various other uses of the phrase can be found: (9b)

9

The mean of the data can then be found from [formula]. |SJ19|P96

That (7el) is stylistically marked and less acceptable than (7e), is another matter.

P.G. Meyer, Hedging Strategies in Written Academic Discourse

37

This use of can is typical of method descriptions or hypothetical reasoning about method, such as is also found in the following: (9c) (9d)

Antibodies related to these systems can only be identified by a laboratory possessing a panel of red cells extensively 'genotyped' to cover them. |SJ13|P57 The economic return on such investments, in the form of faster-moving and less congested traffic, can be fairly precisely calculated. |SJ42|P57

It is highly questionable whether these cases should be regarded as hedging at all. can here indicates a true statement of ability, of what the method or the conditions of the experiment enable us to do. It could of course be argued that assigning the source of the ability to method or situation is a way of shifting responsibility. But in (9c) and (9d), the speech acts that are mentioned, identification and calculation, are simply not the point, but only their possibility. They can in no way be ascribed to the author of the passage in question. But this cannot always be seen from the formulations themselves but from the wider context. It is thus advisable to look very closely at occurrences of can when looking for hedging phenomena. In both these contexts, reporting a common finding and discussing findings, the difference between can and may is very subtle. As HJ. Meyer (1989, 130) points out, "may is always replaceable by can but not vice versa", the reason being that can is more general in meaning whereas one reading of may carries the additional implication of "altemativity", i.e., other modes of actions or other findings are possible, may is thus often found introducing a list of alternatives: (9e) (9f)

The wave equation describing the motion of electrons in a crystal in a perturbing field of force may be derived in a number of ways. |SJ05|P56 By means of appropriate manifolding, a number of such blocks may be assembled together either in series or in parallel, or a combination of both, according to requirements. |SJ07|P56

In the following example, can and may are not interchangeable: (9g)

...under some conditions the FAE may be determined by apparent size, and indeed can be interpreted within the framework of this hypothesis. [SJ25JP9

may in this example simply signals probability, whereas can refers to a real possibility. Both may be regarded as hedging, but for different reasons: may has a hedging effect simply by virtue of turning a modally unmarked statement into a statement of probability, can is a hedge because it modalizes a speech act verb in the way described in section 4.4. The above examples exemplify the use of coming-to-know verbs (verbs such i&find, observe, determine, detect etc.) in the reporting of findings. The use of such verbs in itself may be interpreted as hedging: Instead of simply stating what is the case, the process of investigation and fact-finding is invoked. This is usually done in formulae such as we found that..., it wasfound that... or X wasfound to...The hedging effect comes about because the facts are not represented as God's truth, but simply as 'what was found'; and humans are fallible. On the other hand, invoking the fact-finding process may also be interpreted as strengthening the basis for the claim. Formulae such as it was found that...

38

II. Hedging Strategies in Academic Discourse

indicate that the claim is not unfounded speculation, but the result of a methodically controlled research process. Additional modal hedging, such as shown in the above examples, is relatively infrequent in these contexts. Modals such as can and may are more likely to be found in truly hypothetical or methodological contexts, where they do not have a clear hedging function, as we have seen. There is, however, a farther use of coming-to-know verbs in explicit hedging formulae, such as the following: (9h) (9i)

As far as can be determined the 19 % of non-cooperators were not different in age or other environmental factor from the rest |SJ14|P8 But none of these names, as far as I can ascertain, had appeared in the language before this date. |SJ22|P54

In (9h) and (9i), the fact-finding process is problematized to a high degree: Knowledge of the facts is represented as necessarily fragmentary. By drawing attention to this condition (which in principle is true of all human knowledge), the strength of the assertion is considerably weakened. 6. Summary and discussion In the preceding pages, we have seen that a wide range of hedging strategies are employed in academic language. In particular we found: - Volitional modality may be used where a speech act is potentially facethreatening to the reader (I would like to suggest); but where the inclination is represented as sufficiently weak, the author's face may also be attended to: I am inclined to say... - avoids commitment to the truth of the proposition. - The author may use illocutionary forces that are inherently weak (suspect, suggest). Again the author's commitment to the truth of the proposition is reduced. - The commitment can be further reduced by pretending that the author is not putting forward the claim of his or her own accord, but that the facts actually speak for themselves, or compel the author to claim what s/he is claiming. Impersonal subjects of speech act verbs (This explains..) and deontic modal expressions of necessity (It must be concluded..) are the linguistic hallmarks of this strategy. - Related, though distinguishable, is the strategy of shirting responsibility to method: In the most explicit version of this, it is claimed that 'techniques enable us' to perform some desirable speech act; in many occurrences of can, the implicit source of the ability is method. However, in this area, it is most difficult to draw a line between hedging and true statements of ability. - Modal expressions of ability do occur as true hedges: First, they may be used to hedge a speech act verb, answering anticipated critical questions in an imaginary dialogue with the reader ('What can be said?'- What can be said is..). Second, expressions of ability may be tokens of politeness to the authors of other texts (We can agree that..). Third, in connection with com-

P.G. Meyer, Hedging Strategies in Written Academic Discourse

39

ing-to-know verbs, can and may may be used to further weaken the strength of an assertion {...can be found...). Fourth, they may be part of hedging formulae that question the completeness of the knowledge reported (As far as can be ascertained...). - Last but not least, the fact-finding process itself may be invoked in presenting results. This type reveals the double character of hedging in a most convincing manner: It both strengthens and weakens the argument. The weakening effect comes about because, after all, findings may be fallacious. But a finding is certainly better than an unfounded claim, and this accounts for the strengthening effect. We may thus see that strengthening and weakening are not only constantly interlaced in academic argumentation, but also in many cases indistinguishable. One further point to be emphasised is that usually several of the above strategies are combined to form typical formulae. A last question to be taken up again is what reasons there may be for using hedging strategies at all in academic discourse. We had provisionally assumed the function to be face-saving. This is basically Myers' (1988) explanation. He claims that hedging and other politeness phenomena in academic discourse are the result of following a rational strategy. This explanation is meant as an alternative to another one, saying that scholars are simply following conventions of academic discourse when they hedge face-threatening speech acts. I do not think that these explanations necessarily exclude each other; after all, politeness is in most cases conventional, and conventional rules may have a rational basis. But I also think that both explanations capture only half the truth. I would like to offer a different account which does not plainly contradict the others, but rather complements them. There is more than just face-saving that is involved in trying to make statements impregnable. There is an underlying, mostly implicit cognitive model of the scholarly process that is older and still stronger than the ideal of falsifiability quoted above. Let me try to sketch this very briefly.10 According to this model, it is first of all the task of research activities to provide a sound basis for impregnable statements. This basis is what is usually called 'knowledge'. The more certain the knowledge, the more valuable it is. And the more certain it is, the less hedging is needed to state it. It is only when research activities fail to provide a basis for unhedged statements that language comes in to compensate for the lack of certainty. There is thus an excellent excuse, or, to put it less sarcastically, a compelling reason, to use face-saving devices in academic discourse. The strength of the speech act, in the ideal case, reflects the certainty of knowledge. The second Gricean maxim of quality (see p. 26 above) finds a particularly strict application in academic discourse. It may be said that academic discourse was the model on which it was formulated as an ideal. Scholars primarily save their face, i.e. their reputation as scholars, by following this maxim; in doing so, they serve the goals of science in coming closer to the truth, and they may pretend that nothing else is involved: Officially it is not the personal face of a scholar that is at stake, but truth. We For a more elaborate statement of this model, see Meyer, P.G. 1997, § 2.

40

II. Hedging Strategies in Academic Discourse

may thus see that the different explanations for hedging in academic discourse do not exclude each other, but are true on different levels: What looks like a face-saving strategy on an interpersonal level, is nothing but the strict application of a rational maxim on the level of the logic of conversation, which on the level of genre has been codified as a set of discourse conventions, and on the epistemological level serves to further the goals of science. The latter, however, is the raison d' etre of the whole genre and accordingly has some claim to priority. References Bradac, James J. and Street, Richard L. 1989/ "Powerful and powerless styles of talk: a theoretical analysis of language and 1990 impression formation", Research on language in social interaction 23, 195-241. Brown, Penelope and Levinson, Stephen 1979 Politeness: some universds of language usage. Cambridge: Cambridge Univ. Press. Butler, Christopher S. 1990 "Qualifications in science: modal meanings in scientific texts", in: Nash, Walter (ed.): The writing scholar: studies in academic discourse (= Written communication annual. 3). London: Sage, 137-170. Clyne, Michael 1991 "The socio-cultural dimension: the dilemma of the German-speaking scholar", in Schröder, Hartmut (ed.): Subject-oriented texts: languages for special purposes and text theory (= Research in text theory. 16). Berlin: de Gruyter, 49-67. Fräser, Bruce 1975 "Hedged performatives", in Cole, Peter et al. (eds.): Speech acts (= Syntax and semantics 3). New York: Academic Press, 187-210. Garside, Roger et al. 1987 The computational analysis of English: a corpus-based approach. London: Longman. Grice, Paul 1975 "Logic and conversation", in Cole, Peter et al. (eds.): Speech acts (= Syntax and semantics. 3). New York: Academic Press, 41-58. Horman, Lawrence A. 1989 "The evaluative consequences of hedges, hesitations, and intensifiers: powerful and powerless speech-styles", Human communication research 15, 383-406. Hübler, Axel 1983 Understatements and hedges in English (= Pragmatics and beyond. 4, 6). Amsterdam: Benjamins. Johnson, Donna M. 1992 "Compliments and politeness in peer-review texts", Applied linguistics, 13, 51-71. Kolde, Gottfried 1989 "Probleme der Beschreibung von sogenannten Heckenausdrücken im allgemeinen einsprachigen Wörterbuch", in Hausmann, Franz Josef et al. (eds.): Wörterbücher: ein internationales Handbuch zur Lexikographie — Dictionaries: An international encyclopedia of lexicography = Dictionnaires:..., 1. Berlin: de Gruyter, 855-862. Lakoff, George 1972 "Hedges: a study in meaning criteria and the logic of fuzzy concepts", Papers from the Regional Meeting of the Chicago Linguistic Society, 8, 183-228. Lakoff, Robin Tolmach 1974 "The logic of politeness: or minding your p's and q's", Papers from the Regional Meeting of the Chicago Linguistic Society, 9, 292-305.

P.G. Meyer, Hedging Strategies in Written Academic Discourse

41

Lakoff, Robin Tobruch 1977 "Politeness, pragmatics and performatives", in Rogers, A. et al. (eds.): Proceedings of the Texas conference on performatives, presuppositions and implicatures. Washington: Center for Applied Linguistics, 79-106. Markkanen, Raija and Schröder, Hartmut 1989 "Hedging as a translation problem in scientific texts", in Lauren, Christer and Nordman, Marianne (eds.): Special language: from humans thinking to thinking machines. Clevedon: Multilingual matters, 171-179. Meyer, H.J. 1989 "Modality in science texts", Special language = Fachsprache, 11, 127-134. Meyer, P.G. 1997 Coming to know: Studies in the lexical semantics and pragmatics of academic English. Tübingen: Narr. Myers, Greg 1988 The pragmatics of politeness in scientific articles (= LAUD. Series A.203). Duisburg: LAUD. Schneider, Edgar Werner 1988 Variabilität, Polysemie und Unscharfe in der Wonbedeutung, 1-2 (= Linguistische Arbeiten, 196-197). Tübingen: Niemeyer. Schröder, Hartmut 1989 "Gesellschaftswissenschaftliche Fachtexte und interkulturelle Fachkommunikation5', Special language —Fachsprache, 11, 37^41. Searle, John Rogers 1969 Speech acts: an essay in the philosophy of language. Cambridge: Cambridge University Press. Searle, John Rogers 1983 Intentionality: an essay in the philosophy of mind. Cambridge: Cambridge University Press. Searle, John Rogers 1989 "How performatives work", Linguistics and philosophy, 12, 535-558. Searle, John R. and Vanderveken, Daniel 1985 Foundations of ittocutionary logic. Cambridge: Cambridge University Press. Simpson, Paul "Modality in literary-critical discourse", in Nash, Walter (ed.): The writing scholar: studies in academic discourse (= Written communication annual. 3). London: Sage, 63-94.

JOHN SKELTON

How to Tell the Truth in The British Medical Journal: Patterns of Judgement in the 19th and 20th Centuries

The aim of this paper is to look at the way hedging and related phenomena work in context, rather than in isolation. In it I attempt to do three things. Firstly I look at the theoretical problems which have beset the study of what is called hedging. I do so in the abstract, using text sentences. Secondly, as a prelude to considering the way in which hedging and related phenomena are differently deployed in academic medicine over time I discuss briefly some of the main features in the changing structure and function of articles in the British Medical Journal (BMJ) between 1853 and the present day. Thirdly, I look in detail at three BMJ articles drawn from a larger corpus, from 1853, 1883 and 1991, both to illustrate the difficulty in assigning pieces of text to the category "hedge" and related categories, and to contrast the manner in which these categories are handled.

1. The theoretical problems There are four theoretical problems which I wish to discuss. Two are a matter of terminology, two of procedure. I shall look at the terminological issues first. 1.1. Terminology The word "hedge", because it is an ordinary language term used technically rather than a fully technical term, carries with it a residue of its ordinary language meaning. These imported connotations have had serious consequences. "Hedging" as a label which has hindered the study of hedging as a phenomenon, because it has left us with a perception which is both negative and incomplete. If it is used as a technical term for something large and complex, as it often has been, the word hedging is an unfortunate choice. I shall argue that it is better used for a small and relatively well-defined part of a much bigger feature of languages.

J. Skelton, How to tell the Truth in The British Medical Journal

43

1.1.1. Hedging as a negative term The ordinary language meaning of "hedge" has to do with such matters as a mitigation of certainty, but it implies not only a willed mitigation, but obfuscation for dubious purposes. It is associated therefore with a kind of defensiveness, an evasiveness, a sliding out from under. Hedging in this view is the politician's craft. Of course the connotation of evasiveness is misleading. There are three different ways in which the force of propositions may be modulated, and all have been labelled "hedges" in the literature. All three may indeed be used evasively - or they may not. Firstly, what are known as hedges can offer an approximation of the truth. An evasive example: (1)

Father Why did you punch your little brother? Son I didn't. My hand sort q/~slipped

But the shiftiness is not of the essence. Lakoff's early example (Lakoff 1972), in the paper which offered the first real exploration of the hedging phenomenon, will be recalled. He wished to distinguish between sentences like: (2)

An eagle is a bird

and

(3)

A chicken is sort of a. bird

There is, however, nothing particularly suspicious about saying that a chicken is not as prototypically bird-like as an eagle. Quite the contrary: this is merely a judgement, possibly a carefully-chosen judgement, that a proposition is true only to a limited extent. In this sense sentence 3) is in fact more precise - by virtue of the approximator - than "A chicken is a bird". Secondly, what are known as hedges can express a denial of responsibility. One may state a proposition as a fact: (4)

The world is roun

Or one may use a hedge to distance oneself from it: (5)

// is said the world is round

Here the comment functions to deny the speaker's commitment to the proposition it governs. It acts to place disclaiming quotation marks around the proposition. Yet though one can easily imagine that the intention is to pass the buck, to sustain a guarded neutrality which will exonerate the speaker if the proposition is disproved, this need not be so: (6)

A: I'm going to Iceland this summer B: Oh, they say the scenery's just amazing

Here, B certainly denies responsibility for the truth value of the statement, but there is no suggestion of anything untoward. Thirdly, neutrality will often shade into doubt, and in practice it can often be difficult to distinguish this category from the preceding one. In the sense in which an expression of doubt may function to obscure what is in fact clear, "hedges" of this type can often be underhand, but need not be. Moreover, it

44

II. Hedging Strategies in Academic Discourse

can be equally difficult to place hedges of this type along an axis of doubt and certainty. Consider a sentence like: (7)

I'm not sure if that's a good idea

This might in fact be a polite way of saying "I'm certain it's a bad idea". Equally, it might display genuine uncertainty. But in fact either interpretation creates potential problems. If there is no doubt on the speaker's part, is the sentence a hedge at all? And if there is doubt, might the sentence not be functioning to offer a very precise statement about the extent of one's certainty? If so, is it a hedge? This last point is at the heart of the problem of negative connotation. Many people believe that to lack certainty shows weakness, indecision, muddle or vagueness. It does not. On the contrary, not to be doubtful is often unwise, and empowering people to express doubt where there is no certainty is, for example, an important pedagogic task. 1.1.2. Hedging as an incomplete term

The second terminological issue has to do with the narrow limitations of the ordinary language term "hedge", which is only concerned with mitigations of certainty. Yet it makes little intuitive sense to say that "I'm not sure" in sentence 7) is a hedge, but that the very large number of other possible comments which can replace it along the paradigmatic axis, and which are formally and func tionally so similar, are not hedges because they do not mitigate certainty. Would one really want to say that the confirmatory I'm sure the world is round and the mitigating I'm not sure the world is round were unrelated? This hardly seems a useful division of the cake.Similarly, consider the following. Which are hedges - in the narrow sense that they mitigate certainty? (8) (9) (10) (11) (12) (13) ΟΠ

I'm sure the world is round I believe die world is round I think the world is round I imagine the world is round I take it the world is round I guess the world is round

(14) (15) (16) (17) (18)

I'm I'm I'm I'm I'm

completely sure the world is round quite sure the world is round more or less sure the world is round pretty sure the world is round fairly sure the world is round

The internal semantics of almost all of these phrases permits an interpretation, without context, which may either reinforce certainty, mitigate it or merely state it. In almost all cases, it would take intonational information to disambiguate: to put it another way, all can be followed with equal felicity by "she said hesitantly" and "she said, confidently". Indeed, even within a real context of utterance a speaker may be unclear as to whether greater or lesser certainty was the goal.

J. Skelton, How to tell the Truth in The British Medical Journal

45

These sentences share an overriding similarity. It is hard to conceive of a language in which expressions of doubt, vagueness and lack of commitment are systematically realised entirely separately from expressions of their opposites: rather, they are systematically set in opposition to each other, so that the speaker is forced to make a choice as to the status of the proposition being uttered. This is true of tiny corners of the language (7 'm totally sure versus / 'm moderately sure) and of major features of it, such as an opposition between indicativelike and subjunctive-like moods. All languages, that is to say, present the user with a resource which s/he is bound to take advantage of, a resource which permits the speaker to say something and at the same time to comment on what it is they are saying. Choices about certainty, vagueness and commitment are made continually, and hedges therefore are best viewed as part of a larger phenomenon, the commentative potential of a language. Natural languages are reflective: they not only say things, they offer reflections on the status of what they say, and this happens all the time. As a means of marking propositions for attitude, indeed, the commentative nature of a language is pervasive, resembling more the intonational system than the small set of detachable units of words and phrases which "hedges" are often taken to be. The comments which may be made fall into two categories: they both pass judgement on the proposition they govern, on either its truth value or its emotional value. Thus / suspect the moon is made of green cheese is a comment about truth value, // is good to hear the moon is made of green cheese is a comment about value. The term "hedge" is best used, very narrowly, only for mitigations of responsibility and/or certainty to the truth value of a proposition. This breakdown gives a hierarchy as follows:

Comment

__

Truth-judgement

Value-judgement

Hedge

From this point, I shall use these terms in this way. Where I use the word "hedge" with the ambiguity which is reflected in the literature, I shall use the phrase "traditional hedge". (It must be stressed that "hedges", that is mitigators of certainty, are only one of the entities that occur at the lowest level of the hierarchy: words/phrases which intensify certainty, for example, are other obvious candidates for recognition here).

46

II. Hedging Strategies in Academic Discourse

1.2. Procedure 1.2.1. Subjectivity of decision How does one distinguish between proposition and comments? There are certainly strong associations between truth judgements, for instance, and particular formal characteristics (eg verbs of believing; passives; impersonal structures with it is and the like), but in general decisions are bound to be subjective, to involve a subjective decision about the function a piece of language serves. This poses problems however, since it involves the surrender of claims to external validity. The proof that something is past tense is either the morpheme {-ed} or one of a closed set of well-understood and well-described other changes. There is no such external, that is formal, yardstick to "prove" that something is a request, a suggestion - or a comment. There is then no formal, a priori definition of what a truth judgement is such that we can assign examples of language certainly to one of two mutually exclusive categories, truth judgement and not truth judgement. The same is true for other comment-types, including traditional hedges (and it is also true for such other functional labels as "suggestion", "request" and so forth, a difficulty which, in language teaching, was the soft underbelly of the purely functional syllabus). The sort of difficulty this causes can be briefly illustrated in academic text by looking at two things. Firstly, there is in a lot of academic writing a great deal of constructive ambiguity about the way in which the passive is used to report the work of others: it is not often clear, when we are told "Smith (1993) said that χ was the case", whether this represents a hedge on the truth value of x, or a mere reporting of a fact "(I confirm that) Smith (1993) said that χ was the case", or a piece of supportive evidence "(Like me), Smith (1993) said that χ was the case". Secondly, there are a set of descriptive phrases in academic text which are, to variable and uncertain degrees, both truthjudgmental and valuejudgmental. Thus: This doubtful/provocative/plausible/interesting/surprising/incredibk/extraordinary/amazing line of argument...Even with a context, it is by no means certain whether it is more appropriate to label particular instances as truth or value judgements.

1.2.2. The domain of comments Quantitative statements about the frequency with which any comment-type occurs are likely to falsify data unless some consideration is paid to what the comment governs. As items modulating the force of what follows, one senses a difference between My car is quite big and // is possible that you are wrong. Both of these truth-judgements govern propositions, but the first proposition ("My car is big") is modulated in a way which is trivial: the second is not. Intuitively, the distinction is between a judgement whose domain is less than a clause, and one whose domain is a clause or more. The former I shall call Minor Truth Judgements, the latter Major Truth Judgements, or simply Truth Judgements. Minor truth judgements are not of great significance, and I list them only for the first two articles studied below.

J. Skelton, How to tell the Truth in The British Medical Journal

47

Finally in this section, I should stress that there are areas I have no space to explore. The treatment of value judgements here is necessarily shallow; nor are the relationships explored between such things as intensifiers/de-intensifiers, "vagueifiers" and expressions of quantity such as "a lot/little".

2. Thefunction of truth-judgements in medical writing

The remainder of this paper, then, does two things. Firstly, it briefly describes the changing structure and function of articles in the BMJ as a whole. Secondly, it offers a detailed analysis of the way the commentative system is handled in extracts from three articles, from 1853, 1883 and 1991, so that the reader may see and assess the sort of decisions I have tried to make, particularly with reference to truth judgements. My emphasis is on the development of the truth-judgement system as an increasingly sophisticated resource for the forensic separation of fact and interpretation, and its deployment in pursuit of Judicious choice. The papers I consider here are drawn from a larger corpus. I do not, however, offer a quantitative characterisation of the way in which the role of the truth-judgement develops, though such a characterisation is certainly possible in principle, since to do so would require a greater degree of exposition than I have space for. 2.1. The BMJ article: a note on structure and function

The development of the research article as a genre has been explored in detail by Bazerman (1989) and Myers (1990), who were perhaps the first to bring the potential of this sort of diachronic study to the attention of linguists. Nothing I say contradicts these seminal studies. Of immediate relevance to this study is Atkinson's excellent recent paper on the development of the Edinburgh Medical Journal (Atkinson 1992). As regards hedging, and the study of truth, precision, certainty and uncertainty in medical texts there is relatively little. Kong et al look at the way probabilities are expressed (1986); Dubois (1987) considers precision, Prince et al (1982) discuss hedging in spoken physician-physician discourse and Salager-Meyer (forthcoming) offers a text-based, quantitative study of written language. In addition the present author looks at "comments" in academic articles including medicine, and this study builds to some extent on that, though the concept "comment" has changed to some degree (Skelton 1987). The BMJ has gone through many different forms since its inception. In the early years of the Journal much of what it published consisted of reflective, often beautifully written and balanced talks on subjects of contemporary interest, delivered to august bodies by the luminaries of the day. The printed versions of these talks display their status as oral text perfectly clearly, and in many cases they make fascinating and pleasant reading - a reminder, as so much of 19th century scientific writing is, that whatever has been gained by the devel-

48

Π. Hedging Strategies in Academic Discourse

opment of the modern scientific journal, much has been lost. A second major section of the Journal was to convey news. A third, of course, was the editorial. And the final major section in the early days is that labelled "Original Communications": for this section, practising doctors report on cases they have been involved in, treatments and procedures they have tried, or problems they have been presented with and which they may or may not have solved. A few of these cases have been previously read to learned bodies, most apparently have not. The category which is to contemporary eyes most obviously missing from this is the category which is the bread and butter of contemporary scientific enquiry, the research paper. Enquiry remains at the level of the anecdotal, the history of a particular case rather than the accumulation of a series of cases to arrive at general principles. The obvious reason for this is that the infrastructure to support research (and report on it; though the medical journal itself was in the end to provide part of such an infrastructure) did not exist. However, authors at the time consciously and deliberately selected for study the unusual rather than the general, while the sort of enquiry we would recognise today as medical research - general statements arising from the study of events in controlled conditions - developed slowly. At any rate: the four-way functional split is not, prima facie, very different in the contemporary BMJ. There are changes, developments and shifts of emphasis which tell their own story and reflect changing fashions in medicine, changes in the readership of the Journal, and changes in the history of ideas, but these lie outside the scope of this paper. The obvious difference is the addition of the "Papers" section (in fact the journal's title page abbreviates the full title, which is "Original Papers"), in which original research is presented. This research invariably has a quantitative basis and aims to make general statements. In addition, it closely follows the strict "IMRAD" format, of Introduction, Methods, Results and Discussion - the "A" of the acronym rather weakly refers to the "and" of the list. However, it is not quite the case that the Original Papers of the present day have replaced the Original Communications of the last century. Rather, the true inheritors of the individual case-reports which the Original Communications presented is the section labelled "Short Reports" which was not discontinued until very recent years. These short reports are very often (though not always) concerned with individual cases. What we see, therefore, over the last 140 years is a slow formalisation of the overall structure of papers which are deemed acceptable for publication, a parallel formalisation of the research techniques which they reflect: and a slow drift away from the unique and anecdotal towards the generally applicable and large-scale. The reader is invited at this stage to read the first two papers in the Appendix, from 1853 and 1883 respectively; they are typical representatives of the early days of the genre. Broadly speaking, in these two years - thirty years apart but very similar with respect to the type of paper which was published - the genre structure runs as follows:

J. Skelton, How to tell the Truth in The British Medical Journal (a) (b) (c)

49

Assert rarity and/or importance of case Give chronological account Draw conclusion

In the two 19th century papers contained in the Appendix, an appeal to rarity is contained in the title of the first ("Peculiar case of accidental uterine haemorrhage"), and reinforced at the end ("No example at all similar has occurred to me before..."). In the second, it is contained in the opening sentence ("The following is one of those rare cases..."). One of the things a research article must do is - to repeat Swales' phrase (Swales 1990) - "create a research space". The contemporary research paper does so, as is well established, by asserting the importance, centrality, or significance of the topic These papers may do this, but equally they stake their claim to validity and to space in the journal by asserting the opposite: that what they say is unusual rather than central. This reflects a profound difference in the way science is perceived. Novelty value in the sense that the topic is not just new but unusual - is more the province of the reporter than the scientist, and many of these 1 9th century reports are, in the modern sense, more newsworthy than new. This means, at the commentative level, that the uniqueness may be underscored by value judgements of surprise (as in "I was surprised to find that it had scarcely diminished" in the Harrinson paper). The underlying logical structure of the great majority of 19th century papers in the BMJ is straightforwardly chronological. For present day papers, distortions in chronological ordering are routine in support of the ostensibly logical organisation of the IMRAD pattern. This enables the modem author to gloss over the problems, false starts and hiccups which arose en route, as is well known (see eg Gilbert and Mulkay (1984) passim). There is therefore a great deal of temporal cohesive marking, as one would expect; and commentatively, a certain amount of personal doubt, judgement ana supposition is facilitated. This mode of exploration, introspection and doubt is the central tone of voice of the time. So it is today, one might say, with the exception that the typical hesitancy (more rarely, the slight professorial arrogance which I detect in the 1883 paper, though you may disagree) is depersonalised, pushed away from the author. Truth judgements are not today what they are here, introspective remarks about evanescent hypotheses which get shaped and changed by the march of time and the progress of the case. This is part of what we mean by saying that the scientific article before the modem day was personalised. This label is, however, less than the whole truth. The articles from 1853 and 1883 - the feature diminishes from about that date - are characterised by a strong sense of narrative; they are often dramatic, sometimes even moving. They run typically from a mise en scene ("I arrived at the hospital about 7pm and found...", Buchanan), through a baffling succession of events, to the concluding section, in which doubts are sometimes but not always resolved, often, sadly, as the result of a post-mortem. The structure is not unlike that of the detective story, though the 19th century doctor does not always find the guilty agent - Sherlock Holmes, of course, made his debut in 1887, and Conan Doyle was a doctor: it is tempting to speculate about the extent to which this format was in his blood.

50

II. Hedging Strategies in Academic Discourse

One final point here. The concluding section, which was to evolve eventually into the formal "Discussion" section of today, had a poorly established function. Often a deliberate review of what the case implies for medical practice in general is undertaken, so that to this extent the methodology is properly inductive, and clearly an antecedent of contemporary practice. Sometimes (as with Harrinson) the conclusion reflects the beginning, and reasserts the rarity of what has been presented. Sometimes (as with Buchanan) the ending of the text is purely the ending of the narrative sequence. Not infrequently the tone is rhetorical, the article rising to a peroration and even concluding on a note as much moralistic as scientific. This example is from 1853: A practical writer has lately said, "Whoever abuses the functions of an organ, accepts a blank bill upon the future. It may run for a longer or a shorter date but it must come due". Alas! bills, wherein Disease is the witness, and Death the acceptor, belong to an old currency of a world-wide circulation. That diey are frequendy self-drawn, is proved by sad experience; die present case affording a painful illustration. Mr S., although rarely intoxicate had been in the daily practice, during several years, of indulging in strong spirituous potations. He was, moreover, a free smoker...(Lowther 1853) 2.1.1. Truth judgements in two 19th century articles

I now consider the way that truth judgements are applied in the extracts from the 1853 and 1883 papers in the Appendix. I do this to give the reader a sense of the sort of judgements I am trying to make, but also to point a contrast with the contemporary (1991) paper which is analysed subsequently, and of which extracts are also contained in the Appendix. Both of these papers are strongly narrative in flavour, and precisely chronological in their account. Paper 1 (Harrinson 1853) Minor truth-judgements: having tokrabk health/extreme pain/indescribably distressing/slightly conscious/scarcely diminished/very short/well contracted

These all function like LakofPs "sort of a bird": the first and last two are true approximators, though their function is to make the phrase more rather than less in consonance with the facts. The middle two are intensifiers, evidently enough. The domain of these judgements is local; they govern nothing beyond the immediate noun phrase, and are listed below as minor judgements. As is typically the case with minor judgements, they have little textual power, and in this article as others, the number of tokens of this sort of judgement does not give a good picture of the extent to which a particular text is truth-judgmental. "Well contracted" (line 54) sounds like a value judgement though it is not. Line 26: wefelt justified in thinkingThis is certainly a truth judgement of some sort: whedier it is a hedge depends on whether one argues diat it emphasises certainty ("we were justified") or mitigates it. Line 27: the uterus must be the organ involved A deductive use of "must" - compare the use of "must have" in the following passage. Line 28: had probably occurred A straightforward hedge.

J. Skelton, How to tell the Truth in The British Medical Journal

51

Line 29: The indications appeared to be A classic hedge. Compare other uses of "appear" in this text, and contrast "life appeared to be all but extinguished" (line 22): this I would argue is neither a hedge nor a truth judgement of any sort, but a statement about the patient's appearance, the function being similar to that of a sentence like "His appearance was untidy and shifty". Line 40: I was surprised to find Certainly commentative in function, but clearly a value judgement, rather than a truth judgement. It exemplifies well, however, what it is that hedges, comments and other evaluations all share: they state a proposition - "it had scarcely diminished" - and express an attitude towards it. Line 45: nor did it apparently Probably a hedge, and probably implying that the patient was asked if she felt any easier - but it is hard to be sure. I count it as a hedge. Line 50: she apparently feü into a quiet sleep This looks more like a hedge than not, and is counted as such, since the force of it is that it appeared to be a quiet sleep - but appearances deceived. Line 56: I much regret that Clearly a value judgement. Line 58: undoubtedly A straightforward truth judgement. Line 59: probably nearly entire "Probably" is a straightforward hedge: "Probably nearly" is best accounted for as a "multiple hedge", in Salager-Meyer's phrase, and treated as a single entity (Salager-Meyer forthcoming).

In purely quantitative terms, this paper makes 8 (major) truth judgements, has 2 clauses containing major value judgements, and 7 minor truth judgements. Paper 2 (Buchanan 1883)

Line 25: Line 30: Line 32: Line 34: Line 35:

Line 39:

Minor truth judgements: almost livid/almost imperceptible/quite conscious/most certainly fortuitous/practically closed/practically closed/ All of these, it will be seen, function to render the proposition more, not less, precise. the direction in which the bullet must have entered The use of English "must have" for deductions about the past is always a marker of a truth judgement. failed to detect any evidence....Compare the example immediately following. an....examination failed to give any indication that.. precise, academic statement, like that immediately above, clearly a hedge in the sense that it mitigates certainty, but precisely the sort of scrupulous judgement which is appropriate. I, therefore, believed that the bullet had passed through... I could not find any evidence of its presence there The proposition ("it was present") is not quite on the surface of the text. This must be a hedge, however, as the writer is saying neither "it was not present" nor "there was no evidence that it was present". This type of formulation is more typical of more modern writing, and the 1991 paper in the Appendix has a number of analogous examples. stated my fear that... "I fear that" is more typically evaluative, probably, but its use to the family is a polite statement of empathy with their anxieties, and will have struck the family with the force of a truth-judgement. Note that, strictly, the writer is not here making a truth-judgement, but reporting on one made earlier...this way of putting it however serves the function of conveying his judgement to us.

There are therefore six major truth-judgements to this point in the paper, and six minor truth-judgements. There are a number of things about these truth-judgements which are both noticeable and typical. Firstly, they tend to appear in clusters during the course of the articles, but the site of these clusters is unpredictable - that is, they are not associated with a particular phase (the beginning, say) of the articles. This is in essence a result of the chronological ordering of events which the texts present. Judgements were potentially appropriate at any stage of the unfolding drama, and these judgements are presented as they are made, and in the order

52

II. Hedging Strategies in Academic Discourse

in which they are made. Secondly - though the number of truth judgements quoted above from just two articles is very small, the pattern is typical - there are at most two and usually single tokens of each truth-judgement type. Only "appeared/apparently" and "probably" occur more than once as major truthjudgements, and in total for 11 tokens there are 9 types. Thirdly, truth judgements (and value judgements) are strongly associated with the words "I", "we" and "my". There is overt arrogation of responsibility. Fourthly, there is a strong association of truthjudgements with deductive reasoning. Evidence is weighed and judgement is reached.. That is to say, judgements are associated with hypotheses in the past, hypotheses made at the time of the events the articles describe. Some of these hypotheses may have been resolved by time, others not. Finally, the overall thrust of the articles is deductive rather than inductive, A particular case is examined, and deductions are informed by what general truths are known. The instance of "must have" in the examples above demonstrates this fact with unusual clarity - "must have" is a marker in English for reasonable deductions made about the past. This is in contrast to contemporary methodology, where a host of particular cases are gathered together so that general principles may be induced. None of these things are true of contemporary Original Papers. Indeed, one could argue that the only thing about the truth-judgement system which has remained unchanged is its careful deployment to ensure that claims remain conservative to hedge, in other words. Only the essence is the same; the details are rather different. 2.1.2. Truthjudgements in a recent article

The contemporary paper in the Appendix (Tedder et al. 1991), which the reader might now care to turn to, was published in the BMJ on June 1st 1991.1 have chosen to look at this particular paper in detail because it exemplifies the important points very straightforwardly. The authors worked at the Middlesex School of Medicine at the time of the research. Eight credited authors is a large number, even by the standards of academic medicine, but not tremendously unusual (another paper in the same issue of the BMJ also credits eight authors). The paper is in standard IMRAD format, with the second section called "Materials and Methods". There are no truthjudgements whatever in the Materials and Methods section. That is, truths are merely stated, not commented on. Truth-judgements occur in all other sections, however; but the judgements which are made are of very different types. To take the Introduction first: Line 5

There is epidemiologicol evidence that sexual contact is a risk factor...which suggests that the agent responsible may be sexually transmissible Note first the appeal to evidence (the proposition that sexual contact is a risk factor is judged to be true as a result of evidence). This evidence is then characterised as suggestive - it permits a possible speculation. "Suggest" and "may" together function as a multiple hedge on the proposition "the agent is sexually transmissible". Line 21 Initial observations...are difficult to interpret An interesting refusal to offer a truth judgement

J. Skelton, How to tell the Truth in The British Medical Journal

53

Line 26 A recent retrospective...indicated that seroconversionfor C100 antibody...was rare...and sugested that...this antibody was unrelated to sexual lifestyles Ostensibly - though see below - the use of "indicated" and "suggested" are traditional hedges: they are counted as such. They permit the sentence and the proposition it contains to appear as speculation rather than fact. (In fact this study is mentioned to be knocked down: the authors' conclusions contradict it). Line 36 The results indicated that hepatitis C virus is likely to be sexually transmitted Another multiple hedge in the use of "indicated" and "likely".

What is most noticeable here is the distinction between evidential judgements - comments on truth-value which are founded on empirical evidence, and speculative judgements, where evidence is used as the basis for speculation, for the truth it suggests rather than the truth it contains. Equally evident is the creative ambiguity in the use of the words "indicate" and "suggest". These terms are allowed to carry both the connotation of tentativeness and at the same time, it appears, a degree of surreptitious certainty from the authors. The true level of commitment to the propositions the Introduction sets forward is thus concealed. The Results section is quite different in character. Here, with a single exception, truth judgements centre exclusively on significance and association. Thus, for example: Positivity was strongly associated...with male homosexuality Among heterosexual men and women there were significant associations with positivityfor antibody... Increasing age was associated with positivity...but the association did not read) significance... There was no association with...sexually transmitted disease

The Results section contains seven further uses of "association" and two of "significance". The exceptional truth judgement mentioned above also contains a further example of each of these words, as follows: The analysis of the interrelationship of risk factors...suggests that similar findings would have been produced if a multivariate analysis could have been performed, but because of the strong association between positivity for antibodies to HIV and hepatitis B virus core antigen both of these factors would probably not be significant simultaneously. [line 98 ff: my italics]

In other words, a speculation is offered about a hypothetical experiment which was not conducted because it could not be conducted. Apart from this example, "truth" has what is best defined as a technical meaning. Truth is what statistical calculations about associations and significance tell us. Only evidential judgement is permitted and, given the precision with which truth is defined, there is no room for uncertainty and the personal commitment or lack of it of the authors is neither here nor there. This is therefore in some respects the precise opposite of hedged language. The language is, however, still commentative: a comment is made on the truth-value of such propositions as, say, "Male homosexuals yield positive results". This, it will be seen, is simply another, less scientific and scholarly, way of saying that a null hypothesis ("There is no difference between the number of positive results from homosexual men compared with heterosexual men") is tested empirically.

54

Π. Hedging Strategies in Academic Discourse

Finally, the Discussion section combines both the evidential and the speculative. Medical articles often include, at the start of the Discussion section, a comparison of findings with those of similar studies. These studies are treated evidentially, and in this particular paper are described once more in terms of what significant associations they show. Speculative judgement, of which there is typically a very great deal in scientific writing in Discussion sections, comes later. Thus Line 177: This observation may imply that...A. multiple hedge, with the co-occurrence of "may" and "imply". Line 180: It has been suggested that heat inactivated or frozen stored serum may give fake positive reactions...Another multiple hedge, once more using die word "suggest", once more widi creative ambiguity so that it is not clear whether die authors are merely reporting a suggestion, or committing diemselves to it. Line 182: Serum from patients with chronic liver disease also may give false positive reactions Once more, diere is a degree of ambiguity here: to what extent is this a known and welldefined difficulty, to what extent is it a statement of a vague possibility? Line 213: This provides strong evidencefor sexualtransmission., .although the numbers...were too smattto allow a multivariate analysis Speculative radier dian evidential, in fact, despite the use of the word "evidence", because the point was not statistically demonstrable. Line 229: it seems that most patients...z classic hedge Line 232: The detection of virus...may imply that... straightforward hedge Line 239: It is important thatfurther studies...A. recommendation in a linguistic form typical of academic science; more a value judgement dian a truth judgement? Difficult to say, but on balance a judgement about value. Line 250: We have shown an unequivocal association Back once more to evidential judgement, once more dirough die word "associated". A move which offers a clear and precise statement of what has been shown or demonstrated is almost always present as part of die Discussion in academic medicine.

In addition to the above, there are a couple of truth judgements whose domain is matters relating to the whole study, or large parts of it, rather than a sentence in it. Thus, line 163 "Though these findings are entirely compatible with sexual transmission, some caution is required because the data...were obtained retrospectively". Similarly (line 168) "In our experience...sexual orientation is usually correctly reported...". 2.1.3. The articles in contrast The contrasts between the 20th century paper and those from the 19th century are very clear. Firstly, the guiding organisational principle in the 1991 paper is not chronological nor is it so in the etymological sense of this word; it does not follow the logic of time: rather it follows the logic developed over the course of the 20th Century for the reporting of scientific findings. The time references are therefore different, and where judgements are concerned, these are no longer allowed to appear as interim or provisional decisions. Secondly, the study is depersonalised, though there are several examples of the use of a first person pronoun, and though two examples, the instance just quoted and one quoted below, offer a deliberate, personalised commitment. The overall extent, however, to which there is personal, authorial commitment to truth-judgements is difficult to establish, and this fact is used to weave a curtain of creative

J. Skelton, How to tell the Truth in The British Medical Journal

55

ambiguity around many of the judgements which are in fact made. Thirdly, because of the distortions of chronology built into the new logical ordering, truth judgements are not evenly spread throughout the article. There are none in the Materials and Methods section, and while there are plenty of judgements in the Results section, there are almost no traditional hedges. This is a pattern which is almost invariable in academic medicine (and to the best of my knowledge academic writing in the "hard" sciences in general). Fourthly, the type of truth-judgement which is made now splits clearly into two types, evidential and speculative. Evidential truth-judgements are those judgements which are based exclusively on the rules of evidence for science. They are, in essence, judgements of statistical significance. They occur in the Results section, and also in the Discussion section, during the stage at which comparisons with similar studies are being made. They rarely or never appear elsewhere. Such truthjudgements are unhedged in the sense that they are free of any obvious attempt to abdicate responsibility; but at the same time they do of course permit the authors to hide behind the figures or, more precisely, the rules of the game insist that they do so. Speculative judgements occur only or almost only in the Introduction and Discussion sections. In the Introduction the picture is somewhat obscured by the fact that a portion of the Introduction, sometimes a very large portion, is taken up with reporting the findings of others, in a manner which is inherently ambiguous with respect to the degree of commitment of the present author. There is a further type of truth-judgement which appears in the Discussion section of academic medical papers and which is almost always quite explicitly present. This is the point at which a judgement is made about the truth-value of the whole study. In this paper this function is realised by (line 250): "We have shown an unequivocal association between C 100 antibody and male homosexuality...". This claim is, indeed, unusually forceful. Though there are, in the majority of studies, formal claims made of this nature, and introduced by the word "demonstrate" or, as here, "show", it is moderately unusual for the claim to be further strengthened by a minor judgement as it is here by "unequivocal", and very unusual for the claim to be made in the first person, as it is here, with the further strengthening which the overtly personal commitment implies. The fifth and final point to be made by way of contrast between contemporary and early articles is to do with the actual vocabulary of the papers. The 19th century papers show a high type-token ration - indeed, for the two papers here it approaches 1:1: in the 1991 paper, for the 12 evidential judgements there are 12 uses of "associate" and cognates, four of "significant" and cognates, and no piece of evidential judgement occurs without at least one of these terms. For the 10 speculative judgements identified above in the paper (disregarding the two reflections whose domain is the whole study), there are 4 tokens of suggest, 4 of may, 2 each of imply and indicate, and 1 each of likely, provide and seem. The vocabulary of truth judgements, in other words, is very restricted, and there is no overlap between evidential and speculative judgement.

56

II. Hedging Strategies in Academic Discourse

3. Conclusion

The three articles studied in detail in this paper offer a very stark contrast, and indeed are selected with that in mind. The real picture, naturally, is more confusing. For example, the relationship between the slow demise of the case study approach which typified early writing and which is replaced by the "Short Reports" of more recent times befove it too was abandoned, requires more study than I offer here. The most interesting cause of confusion in the overall picture, however, is the Second World War. During this period, as one might expect, the subject matter changes to include the study of wartime injuries; there is a sudden overwhelming need to report on and understand medical problems of a kind outside the range of most peacetime doctors, and a number of the papers of this era convey something ofthat urgency. In addition, however, one sees a desire for a more modem-looking methodology and organisational structure being thwarted by the exigencies of wartime. Very typical of these years is a paper by Pinnock and Wood (1943), both of whom were at the time Lieutenant-Colonels in the Royal Army Medical Corps. They begin their paper by bemoaning the poverty of information available and apologising for the lack of what in contemporary papers is indispensable, a discussion of previous research: As little information has been published concerning the effects of blast injury to the abdomen the following cases are recorded in order to supply additional data. Under conditions of active service it is not possible to do justice even to the very scanty literature.

The bulk of the paper then describes the effects on sailors of being in the water when two depth charges were exploded near them. Five survivors (four of whom subsequently died) were taken from the water, and it is these cases which are reported. The study of five cases indicates the contemporary (and at that time too, the preferred) strategy of observing the general rather than the unique, and there is an attempt to contain speculative judgement within the concluding, Discussion-like section, labelled "Commentary" which in particular gives a modern feel to the paper. On the other hand, the personal and the sense of the dramatic of an earlier age can still surface, perhaps under the shock and pressure of wartime. In the Commentary, therefore, alongside the objective and restricted vocabulary we are familiar with today and which concludes this quotation, comes an entirely different sense of engagement with the topic: That the general appearance [of the five patients] was so good misled us to the extent that we were loath to accept perforation as the cause of the rigidity [of the abdomen, experienced by all the patients except the survivor] (recollecting what we had read in the Army Medical Bulletin No. 5 of the great rigidity encountered in cases of blast-lung), and the opinion was formed that the condition was due to blast-lung, with probable actual damage to the abdominal muscles. Even 24 hours later, when there obvious deterioration in the condition of these patients and when operation was decided upon, one of our surgical colleagues remarked that he had not seen a patient with a perforated viscus of so many hours' standing with so good a facial aspect, and could not believe that such a catastrophe had occurred.

J. Skelton, How to tell the Truth in The British Medical Journal

57

These papers of the war years are quite atypical, for the obvious reasons mentioned: but they display themselves very clearly, perhaps for those same reasons, as being on the cusp of modernity, looking both backwards and forwards, and they fit very broadly into an overall pattern of development for truthjudgements. This pattern develops under pressure from the changing constraints, and in particular from the increasing constraints in favour of a modest uniformity of style. Truth-judgements are separated out into those which have a statistical basis and those which do not, as the basic methodology shifts from deductive to inductive. The different types of judgement are increasingly limited to specific areas of the text, evidential judgement to Results, speculative judgement to Introduction and, particularly, Discussion. Overt personal commitment to propositions is replaced by a set of complex but intuitively well-understood conventions which permit a considerable degree of creative ambiguity. The vocabulary of truth-judgements becomes increasingly restricted, stylised and formalised, to the extent that words associated with speculative judgement like "suggest", "indicate", "imply" now have almost the status of technical terms.

References Atkinson, Dwight 1992 "The Evolution of Medical Research Writing from 1735 to 1985: The Case of The Edinburgh Medical Journal", Applied Linguistics 13, 4, 337-374. Bazerman, Charles 1989 Shaping Written Knowledge. Madison, Wis: University of Wisconsin Press. Dubois, Betty Lou 1987 "Something of the Order of around 40 to 44: Imprecise Numerical Expressions in Biomedical Slide Talks", Language and Society 16, 525-541. Gilbert, G.M. and Mulkay, M. 1984 Opening Pandora's Box: a Sociological Analysis of Scientific Discourse. Cambridge: Cambridge University Press. Kong, A. et al. 1986 "How Medical Professionals Evaluate Expressions for Probabilities", New England Journal of Medicine 315, 740-744. Lakoff, George 1972 "Hedges: a Study in Meaning Criteria and the Logic of Fuzzy Concepts", Journal of Philosophical Logic 2, 458-508. Myers, Greg 1990 Writing Biology: Texts in the Social Construction of Science. Madison, Wis: University of Wisconsin Press. Prince, E. F. et al. 1982 "On Hedging in Physician-physician Discourse" in di Pietro J (ed):Linguistics and the Professions. Norwood, NJ: Ablex, 83-97. Salager-Meyer, Francoise (forthcom.) "Hedges and Textual Communicative Function in Medical English Written Discourse". Skelton, John 1987 "Comments in Academic Articles" in Grunwell, P. (ed): Applied Linguistics in Society. London: GILT, 98-108.

58

II. Hedging Strategies in Academic Discourse

Papers from the BMJ (Note inconsistent numbering) (In the Appendix) Harrinson, L. 1853 "Peculiar case of accidental uterine haemorrhage", British Medical Journal 1, 1, Jan 7, 1-12. Buchanan, George 1883 "Penetrating gunshot wound of the abdomen: excision of the bullet from the rector spinae muscle: recovery", British Medical Journal 1, Mar 31, 609-611. Tedder, R.S. et al. 1991 "Hepatitis C virus: evidence for sexual transmission", British Medical Journal302, 6788, Jun 1, 1299-1302. (In the text) Lowther George 1853 "Case of stricture of the oesophagus", British Medical Journal 1, 7, Feb 18th, 149-150. Pinnock, D Denham, Wood, Paul 1943 "Blast injury to the abdomen by depth charge", British Medical Journal 1, May 1, 537-539.

J. Skelton, How to tell the Truth in The British Medical Journal

59

Appendix Harrinson 1853 PECULIAR CASE OF ACCIDENTAL UTERINE HAEMORRHAGE By I. HARRINSON, Esq., Surgeon (Read before the Reading Pathological Society, Oct 27, 1852.) On Sunday, July llth, 1852, at half-past 11 o'clock, am, I was called to Mrs. G , Queen's-road, aged thirty-six years. She was a well-developed woman, having tolerable health, and was about seven and a half months advanced in her third pregnancy. She had suffered less during the present than her former pregnancies. From her distressed condition, I could only gain the following particulars as to the method 5 of her seizure. She was sitting in her kitchen, between 9 and 10 o'clock, a.m., preparing some culinary compound, when she suddenly felt feint; her bowels were twice moved in quick succession, and severe pain in the abdomen immediately succeeded. At half-past 10 o'clock I found her lying on a sofa, complaining of iaintness and of extreme pain in her abdomen. She fancied that labour was coming on. There was no discharge. Some brandy was given, and she rallied. I requested 10 that she should be got to bed, and promised to see her again in an hour. On my return, she was more distressed, more faint, and in more pain; in a few words, she was in a state of collapse. On examination externally, the abdomen was very tense, and the seat of pain was principally at the fundus of the uterus, which reached and occupied the epigastric region. The pain was that of 15 distension, of bursting, and indescribably distressing. On examination per vaginam, I discovered the cervix uteri not obliterated, the os uteri just admitting the finger, the membranes unnurtured, and the head presenting. Amid the enveloping agony, she was now and then slightly conscious of the pain of labour. I determined at once to rupture the membranes, which was readily accomplished. The waters 20 flowed profusely and clear, there was not a trace of blood. About this time she had a convulsion, during which life appeared to be all but extinguished.. Dr Cowan now saw her, soon after 1 o'clock, p.m. After a careful examination, we agreed that a grave accident had happened, but could not determine its precise nature, nor its seat; nor whether it was perforation or rapture of some abdominal or pelvic viscus. 25 From, however, the condition of pregnancy, and the absence of any other recognized disease, we felt justified in thinking that the uterus must be the organ involved, and that rupture, partial or complete, ofthat organ had probably occurred The indications appeared to be, to relieve the pain, and to keep up the failing powers as well as possible, which were attempted to be fulfilled by opium and by brandy. One drachm of tincture 30 of opium was given immediately, and brandy and water at frequent intervals. There was an almost incessant cry for cold water, which was ungrudgingly supplied. Three o'clock, p.m. Much the same; no relief. The same dose of tincture of opium was repeated. The os uteri was progressively dilating.

60

II. Hedging Strategies in Academic Discourse

At 5 o'clock Dr Cowan again visited her. I propsed that we should spare her little remaining 35 strength, by now completing labour with the perforator and the forceps; premising, of course, that that could be done readily and with safety. Dr Cowan agreed. The head was perforated and extracted with little difficulty. In a short time, one hand being kept firmly applied to the abdomen, the body readily escaped. Without waiting to separate the child, the hand was still kept on the abdomen, and I was surprised to find that it had scarcely diminished in size or in tension. 40 Finn pressure was kept up, and I was about to observe to the nurse that we had one or two more children to come, when one of the most fearful gushes took place I ever witnessed. Placenta, clots and fluid blood, to the amount of some quarts, were expelled at the same instant The uterus was at once carried down into the pelvis, and there fixed with the hand. There was no further escape of blood. The uterus showed no disposition to relax, nor did it apparently in the slightest degree. 45 She was immediately relieved ofthat dreadful feeling of distension, and lay easy for a time. Anticipating what was to follow, while I was guarding the uterus, the nurse was employed in giving as much brandy and water as she could get down. Another drachm of tincture of opium was given. It was now about half-past six o'clock. After a short time the restlessness increased; she tossed about incessantly, and the extremities became cold. At eight o'clock she apparently fell into a quiet 50 sleep. At half-past eight o'clock, Dr Cowan again saw her. It was agreed to let her sleep a little longer. She slept, but it was the sleep of death. At ten o'clock her breathing was observed to change, and in a few minutes she died, twelve hours from the time she was first seized. Autopsy performed next morning. Dr Cowan was present The uterus was found well contracted: no blood was discovered within it No appreciable disease in any organ. 55 I may here remark, that the cord was twice round the neck of the child, and was very short I much regret that I did not measure it There was no disease of the placenta. What occurred, then, to produce so lamentable an accident, undoubtedly was, a sudden separation, probably nearly entire, of the placenta from the uterus; and the only discoverable cause, an unusual shortness of the umbilical cord, and that twisted round the neck of the infant 60 In the retrospect of such a case, a few points present themselves for consideration. Could a more accurate diagnosis have been arrived at? and if so, could the treatment have been materially modified? To describe her condition in a few words, it may be said that she was in a state of collapse, with extreme pain. These symptoms, however, would equally belong to perforation or rupture of any of the abdominal or pelvic viscera, or even to the rupture of an aneurism.

65

The only characteristic I could name, would be the state of the abdomen. It was distended to the utmost; was dull to percussion; and had an unusual, peculiar, doughy feel. Her strength was supported after the most approved methods and opium, that sustainer of the vital energies, was carefully administered. Labour was accomplished as soon as the state of the 70 uterus allowed. When the uterus was emptied, no further discharge was permitted. No example at all similar has occurred to me before. No writer that I can find has described one exactly corresponding and some authors do not even name the possibility of such an occurrence. Dr Robert Lee, in his "Clinical Midwifery", D Copeman in "Crosse's Midwifery", Drs Churchill, Ramsbotham and Blundell, etc., give good cases and descriptions of "accidental uterine 75 haemorrhage11, as ordinarily happening; but I can find no author giving an account on the diagnosis of such cases as the one related, where no discharge appears externally. Readm& October 1852

J. Skelton, How to tell the Truth in The British Medical Journal

61

Buchanan 1883 PENETRATING GUNSHOT-WOUND OF THE ABDOMEN: EXCISION OF THE BULLET FROM THE ERECTOR SPINAE MUSCLE: RECOVERY By GEORGE BUCHANAN, Professor of Clinical Surgery in the University of Glasgow. The following is one of those rare cases, of which examples have occasionally been recorded, of a man's body having been penetrated by a bullet, with subsequent recovery. On the afternoon of June 11th, 1882, Mr. M, aged 25, a healthy vigorous young gentleman, volunteer, athlete, cricketer, etc., was visiting a steamship in the harbour, in which he had been engaged in assisting to fit up the engines, that being the business for which he was preparing. 5 The captain of the ship showed his visitor and a companion a revolver, and, while the companion was holding the pistol, it accidentally went off, and the contents of the barrel were lodged in Mr. M/s body. The companion was standing in front and a little to the right of Mr. M., and the barrels of the pistol were directed towards him. Mr. M. fell down immediately when struck, and at once became faint He was placed on a sofa, and Dr. Renton was sent for, being a friend of Mr. M. Finding 10 that no haemorrhage was going on, and that Mr. M. could be moved without causing him much pain, Dr. Renton sent a message for the ambulance wagon, and had the patient removed to a private room in the Western Infirmary under my care, and I was sent for. I arrived at the hospital about 7 p.m., and found Mr. M in bed in the following condition. His countenance was pale, almost livid, and anxious; his skin cold and clammy^ pulse almost 15 imperceptible, so small that it could not be counted. The patient was inclined to vomit, and had vomited several times the small quantities of brandy and hot water which had been given. He was quite conscious, and could give an intelligent account of what had happened, specially insisting that the accident was most certainly fortuitous, and that no blame was attachable to the young gentleman who held the pistol; but he was not allowed to speak much. 20 On examining the abdomen, a small ragged wound was seen, three and a half inches above the umbilicus, and an inch to the left of the middle line. It was not so big as a threepenny-piece; its edges were dry and inverted, those of two sides resting against each other, so that it was practically closed by the contraction. On putting my finger-tip into the wound, I found its entrance was resisted; and, on gently insinuating a probe into it in the direction in which the bullet must have entered, 25 it went a little way outwards and down-wards, and backwards, and then it was resisted, showing that the opening had become valvular and practically closed. There was no haemorrhage, and little or no pain in the site of the wound, but a good deal of pain in the back, opposite to it, and in the left thigh in front The most careful manipulation failed to detect any evidence of the presence of the bullet, or any 30 bits of clothing, in the abdominal wall around the wound; and an equally careful examination of the parietes failed to give any indication that the bullet might have, by some curious torsion or oblique movement, travelled round among the layers of the abdominal walls out of reach at the back. I, therefore, believed that the bullet had passed through the abdominal cavity, and was lodged somewhere in the back, although I could not find any evidence of its presence there; and that, 35 probably, some internal viscus was wounded, thus giving rise to the collapse which is so frequent a sign of perforation of the bowel. I formed a most unfavourable opinion of the patient's state, and asked one of his friends to telegraph to his father and brother, who resided near Dublin, to come to Glasgow next day, but stated my fear that a fatal result would take place before they arrived.

62

II. Hedging Strategies in Academic Discourse

Tedder et al 1991

5

10

15

20

25

30

35

40

45

50

55

60

Introduction A number of bloodborce systemic viral infections— 65 homosexual men. There was no association with the for example, human immunodeficiency virus (HIV), presence or absence of a sexually transmitted disease hepatitis B virus, and Ebola virus—may be transmitted diagnosed at that clinic visit. between partners during sexual intercourse. There Of the serum samples in the reactive category (table is epidemiological evidence that sexual contact is a III), a higher proportion were from homosexual men risk factor for acute ηοπ-A, non-B hepatitis,' which 70 (24/32) and were positive for antibody to HIV (21/32) suggests that the agent responsible may be sexually and antibody to hepatitis B core antigen (19/32) than transmissible. The recent cloning and expression of an was found in the non-reactive category (232/987, RNA viral genome from the plasma of chimpanzees 547987, 164/976 respectively; paKTOpbi BapnaTHBHOCTH peneebix BbicKasbiBaHHft". in OCHOBM peiesoft flejiTejibHocTH. MocKBa: PyccKHft 3 .

reopnH

Lyons, J. 1977 Semantics. 2 vols. Cambridge: Cambridge University Press. Markkanen, R. and Schröder, H. 1987 Hedges and its Linguistic Realization in German, English and Finnish Philosophical Texts. VAKKI-seminaari VI, Vaasa, 47-57.

V. Namsaraev, Hedging in Russian Academic Writing in Sociological Texts

79

Markkanen, H., Steffensen, M.S. and Crismore, A. 1990 Quantitative Contrastive Study of Metadiscourse: Problems in Design and Analysis of Data. Paper presented at the XXV International Conference of Contrastive Linguistics and Cross-Cultural Studies. December 6-9, 1990. Rydzyka, Poland. Meyer, B.J.F. 1975 The Organization of Prose and its Effects on Memory. Amsterdam: NorthHolland. Namsaraev, V. V. 1993 XeflXHHr B pyccKHX HayHHbix cratb x ΠΟΟΟΙΙ,ΗΟΠΟΓΗΗ. (Ha Marepnane nyoJiHKauHft B xypnajie "CouHOJiorimecKHe HccneflOBaHHJi" B 1978-1993 r.r.). Licentiate thesis. Vaasa: University of Vaasa. Opitz, K. 1981 Formalcharacter als Indiz f r Fachsprachlichkeit: ein deflatorischer Ansatz, in K hlwein, W. and Raasch, H. (eds.): Sprache: Lehren - Lernen, Vol. I, T bingen: Narr. Palmer, F. R. 1979 Modality and the English Modab. London: Longman. Palmer, F. R. 1986 Mood and Modality. Cambridge: Cambridge University Press. Perkins. M. 1983 Modal Expressions in English. London: Frances Pinter. Salager-Meyer, F. 1994 "Hedges and Textual Communicative Function in Medical English Written Discourse", English for Specific Purposes vol. 13, N 2, 149-170. Simpson, P. 1990 "Modality in Literary-Critical Discourse", in Nash.W. (ed.): The Writing Scholar: Studies in Academic Discourse. London: Sage Publications, 63-94. Skelton.J. 1988 "The Care and Maintenance of Hedges", ELI'Journal 42 (1), January 1988,37-43. Vande Kopple, W. 1985 "Some exploratory discourse on metadiscourse", College Composition and Communication 36, 82-93. Vande Kopple, W J. 1988 "Metadiscourse and the Recall of Modality Markers", Visible Language, Vol. XXII, 2, 3,233-272. Weinreich, U. 1966 "On the semantic structure in English", in Greenberg, J.H. (ed.): Universals of Language, Second Edition, 142-217. Williams, J.M. 1981 a "Literary Style: The Personal Voice", in Shapen.T. and Williams,]. M. (eds.): Style and variables in English, Cambridge, MA: Winthrop, 116-216. Williams, J.M. 1981b Style: Ten Lessons of Clarity and Grace. Glenview, IL: Scott, Foresman.

III. Interactive Aspects of Hedging

AVON CRISMORE and WILLIAM J. VANDE KOPPLE Hedges and Readers: Effects on Attitudes and Learning

1. Introduction

In this paper we discuss some findings of studies that we have been conducting of hedges, which are linguistic elements such as perhaps, might, to a certain degree, and it is possible that. We have been using two general questions to guide our study. First, we have been asking what kinds of effects it is possible for hedges to have on readers. With this general question we have associated more specific questions such as the following: Is it possible that hedges can move readers to change their attitudes toward the author or the subject matter of a text? And can hedges lead people to alter their overall ideologies or the systems of belief that sustain their lives? Can hedges in a written text affect what people learn by reading that text or even how they configure in long-term memory what they have learned? Can hedges lead readers to alter their views about how knowledge is or should be generated and communicated within an academic field or between academic fields? Second, as we found evidence that hedges do have effects on readers, we began asking what aspects of the overall context of reading those effects are dependent on or interact with. With this general question, too, we have associated more specific questions. Some of these have to do with the nature of the text being read. For instance, what difference does the genre or the conventionalized kind of a text make on how hedges in that text affect readers? In other words, would hedges in a scientific research report have different kinds of effects on readers than hedges in a scientist's autobiography would? Similarly, what difference does the nature of the subject matter of a text make on how hedges in that text affect readers? To take one specific example, do hedges associated with very controversial material have different kinds of effects on readers than do hedges associated with material that most readers find not nearly so controversial? Other of the more specific questions have to do with the nature of readers. For example, what differences do different purposes for reading a text make on how hedges affect readers of that text? Specifically, would hedges have effects on those reading a text in order to pick up the gist of the information it conveys that are different from the kinds of effects they would have on those

84

. Interactive Aspects of Hedging

reviewing in a professional journal that same text for their colleagues? Or what differences do readers' beliefs about knowledge make on how hedges affect them as they read a text? To take one specific example, if readers believed that knowledge is primarily a social construction, would they be likely to be affected differently by hedges in a text than they would if they believed that knowledge is primarily the result of discovering the nature of things in the world and the relationships between those things? Still other of these more specific questions have to do with the potential relationships between readers and writers. For instance, if all or most readers are of a different gender from the writer, if all or most readers have less actual or perceived social influence or political power than the writer, if all or most readers are of different ages than the writer, or if all or most readers have different basic beliefs about life from the writer, how prominent a role would such possible differences play in determining how hedges that the writer uses would affect those readers? This paper will address some of these research questions. 1.1. Our -working definition of hedges

What we call hedges are sometimes also called weakeners (Brown/Levinson 1978), downtoners (Holmes 1982; Quirk/Greenbaum/Leech/Svartvik 1985), detensifiers (Hübler 1983), and indicators of degrees of reliability (Chafe 1986). The classification system for hedges that we have found most helpful for our work has been developed by Prince and her associates through examinations of the discourse among medical doctors as they discussed cases in a pediatric intensive care unit (Prince 1976; Prince/Frader/Bosk 1982). In our work, we have focused primarily on the kind of hedging expression that Prince and her co-workers call a plausibility shield. The shields in general produce "fuzziness in the relationship between the prepositional content and the speaker" (Prince/Frader/Bosk 1982, 85); they do not affect the truth values of the propositions associated with them. When people add the specific kind of shield called a plausibility shield to a proposition, they indicate that they are not certain about the truth value of the proposition. They use plausibility shields such as I think or perhaps to signal a tentative or cautious assessment of the truth of propositions. In so doing, they reduce the "degree of liability" or responsibility that they might face in expressing the proposition (Hübler 1983, 18). It is interesting to note in this connection that Prince and her associates found plausibility shields throughout their corpus but discovered that they were particularly characteristic of the physicians' talk when those physicians diagnosed illnesses and planned courses of treatment, two extremely sensitive domains of discourse within the corpus. As Prince, Frader and Bosk point out, it is clear that different plausibility shields are associated with different levels of uncertainty, but it is not altogether clear how many levels of uncertainty would be needed in an adequate model of natural language production and processing. Chafe (1986) makes a contribution by showing how different levels of uncertainty are expressed by various adverbs and modal verbs. And Halliday (1985, 86) begins classifying such levels by distinguishing a level of probability from a level of possibility. On each

A. Crismore and W. Kopple, Effects on Attitudes and Learning

85

of these levels he goes on to make finer distinctions, but such fine distinctions lie beyond our main purposes here. 1.2. Disagreement about the effects of hedges We were initially drawn to study the effects of written hedges because the research on hedges has left many interesting areas largely or entirely unexplored. For instance, most of the research has focused on spoken hedges, not written ones. And some of the research has not been sufficiently attentive to aspects of the context in which the subjects in the research did the experimental reading. Specifically, not all of the research has been attentive to such potentially important aspects of context as how well those encountering or using hedges know one another,how much power those using or encountering hedges have relative to one another, and how those using or encountering hedges feel about the subject matter the hedges are associated with. Furthermore, we were drawn to study the effects of hedges because the work that has been done on how hedges generally affect people contains many apparent disagreements. For example, R. Lakoff (1975) classes hedges with the markers of powerless speech. And as Bell, Zahn, and Hopper (1984, 28) point out, researchers have argued that people who use hedges in making claims and assertions "will be perceived by others as less powerful and less credible" . In addition, Hosman and Wright (1987) have asked subjects to judge how authoritative and attractive transcripts of courtroom testimony are. Generally the transcripts with hedges led subjects to view them as less authoritative and attractive. At the same time, "a high level of hedges resulted in greater attributions of guilt (...) than did an absence of hedges (...)" (Hosman/Wright 1987, 180). Not all researchers fully accept the results of such studies, however. For instance, Hosman and Wright (1987, 183), showing the kind of contextual sensitivity that seems essential in working on the effects of hedges, point out that "hedges may be more negatively evaluated in a simulated legal context than in other contexts" because the legal context leads people to value certainty so highly. Beyond this, Bell, Zahn, and Hopper (1984) report on studies that suggest that some researchers might have found negative effects for hedges because they included an unnaturally high number of them in their materials. Finally, some researchers have discovered positive effects of hedges. Newcombe and Amhoff (1979), for example, found that hedged speech was sometimes perceived as warmer than nonhedged speech. In the light of both the gaps in the research on hedges and the apparent conflicts within that research, it is not surprising that there is a wide array of advice to writers about hedges in general and plausibility shields in particular in works on rhetoric, composition, and style. Those who give advice about hedges seem to fall into three different camps. In the first camp are those who urge writers to eliminate all hedges in their prose. Some of these advisors see hedges as adding nothing to prepositional information. Therefore, they call hedges such things as "deadwood" (Mahaney 1985, 364) "throat-clearing" (Lindgren 1982, 177), "clutter words" (Lannon

86

ΙΠ. Interactive Aspects of Hedging

1986, 135) and "empty qualifiers" (Millward 1980, 205). Others argue for the elimination of hedges by claiming that they rob prose of its certainty and power (Payne 1975, 71; Zinsser 1976, 96), that they reveal a writer avoiding commitments and difficult critical judgments (Rawlins 1980). Such advisors often call hedges "weak qualifiers" (Bamet/ Stubbs 1983, 335) or "the leeches that infest the pond of prose, sucking the blood of words" (Strunk /White 1959, 59). Advisors in the second camp are more positive about hedges. They take care to note that hedges attached to statements of fact are unnecessary, but also that many statements do not express facts. Rather, they express opinions, judgments, hypotheses, guesses, predictions, and the like. Thus these advisors direct writers to examine the nature of their material carefully. If that material is not factual, then writers might be well advised to hedge it (cf. de Beaugrande 1985, 17; Dougherty 1985, 186; Hariston 1981, 54; Lannon 1983, 135; Ruggiero 1981, 227; and Woodman/Adler 1985, 491). Those who give this kind of advice point out that good academic writing is dotted with hedges (Dillon 1981, 91), that much student writing "stands in need of a delicate qualifier here or there rather than the removal of them" (Dillon 1981, 91), and that hedges can "keep prose responsible" (Packer/Timpane 1986, 220). Finally, a third camp of advisors has an even more favorable view of hedges. These advisors stress the benefits of hedges and call for writers to be alert for chances to use them. For example, Singer (1986) claims that hedges help make a lext "friendly". Laib (1985, 590) proposes that "Statesmen and teachers may prove more effective with a degree of effacement, since it leaves room for discussion, differences of opinion, and compromise". And Kress and Hodge even suggest that hedges and related elements might convey the "major content of an utterance" (cited in Ruthrof 1981, 196). We designed our study in order to begin to respond to some of the gaps and the conflicts within both the research findings on hedges and the advice to writers about hedges. We hoped to explore and stimulate discussion on what effects hedges have on readers. And we began our work with the assumption that we would never be able to understand the effects, of written hedges without examining aspects of the context of reading carefully and asking how those aspects affect the manner in which readers respond to hedges. In so doing, we hoped to offer insights to those studying processes of reading as well as to those examining the relevant intricacies of contexts of reading. 1.3, The background for our study

As indicated earlier, previous research on hedges has usually explored spoken language and has tended to focus on the producers of hedges, the speakers (e.g., Holmes 1982; 1984; Preisler 1986), although a few studies have looked at the effects of hedges on listeners (e.g., Hosman/Wright 1987; Newcombe/Arnhoff 1979). But hedges in written texts and their effects on readers have remained almost unexplored. Bruce (1981) argued that when devices like hedges are added to a written text, readers become more involved and active as they read. This involvement then results in increased understanding of both authors and content and in intensified or changed attitudes. He suggested that devices such as

A. Crismore and W. Kopple, Effects on Attitudes and Learning

87

hedges might have at least four specific effects: (1) delight by readers in technique - a pleasure that helps contribute to a reader's engagement in the communication process; (2) the feeling of an exotic experience; (3) access to knowledge otherwise inaccessible; and (4) ethos - more credible authors. Markkanen and Schröder (1989, 2) emphasized that hedging is not an inherent characteristic of a text but a product of writer-reader communication: "(...) the linguistic devices used for hedging get their meaning only through the response they produce in the readers." Studies of hedging in written texts, therefore, focus on the social or interpersonal function of language (Halliday 1983; 1985) and expand on the previous hedge research in spoken texts. When we looked for research studies of hedging in written texts and the effects on readers, the only one we could find was a study by Crismore (1985). Crismore added both hedges and emphatics (elements stressing certainty such as of course, and assuredly) to some social studies passages and investigated whether readers recalled the subject matter better when it was written with or without hedges and emphatics. She found that hedges and emphatics facilitated the recall of students who were not anxious about reading social studies material, but detrimentally affected the learning and attitudes of students who were highly anxious. Moreover, on some subtests, hedges and emphatics increased the learning of high-ability students but not that of low-ability students. She also investigated whether hedges and emphatics made a difference in readers' attitudes toward the subject matter. She found that readers' attitudes improved significantly when passages contained hedges and emphatics but not when the same passages did not. But the manipulated passages contained both hedges and emphatics; thus the roles of these elements were not separated, and it was impossible to know which one had the positive effects on learning and attitudes. Therefore, to begin gathering information about how readers respond to hedges and what might help writers decide how to better use them, we conducted an experiment to explore how hedges affect readers' attitudes and learning. Van Dijk (1982) defines opinions and attitudes as cognitive information organized into sets of evaluative beliefs. Attitudes are schema-like networks or systems of general opinions that organize more specific opinions and beliefs, while ideologies are complex systems that organize attitudes. He argues that language users assign evaluations to authors and to their textual material and that the very processes of understanding may be affected by these attitudes. Because written texts often move readers to change their beliefs, opinions, and attitudes, we wanted to investigate attitude changes that readers might experience. We operationally defined attitude change as the difference between scores on pre- and post-reading attitude tests on the subject matter. Specifically, we wanted to pursue research questions such as the following: Would students' attitudes toward the subject matter change or stay the same? If they did change, would they become more positive or more negative? Finally, after reading the selected passages, what attitudes would readers have concerning the author, the text, and themselves as readers? We also investigated how hedges affect readers' learning from prose. We operationally defined learning as the difference between scores on pre- and

88

ΙΠ. Interactive Aspects of Hedging

post-reading retention tests for propositional information - the subject matter - expressed in the passages. Some of the specific research questions we pursued were as follows: Would it make any difference to readers' learning from a passage where authors put hedges and how many they used in that passage? Would hedges presented personally, in forms such as I suppose that, affect readers differently from those presented impersonally, in forms such as it is possible that? Such questions guided us as we devised versions of a passage from a science textbook and versions of a passage from a social studies textbook. Our goal was to investigate the effects of hedges presented in different conditions on readers' attitudes toward and learning of the subject matter of the passages. 2. The context and methods of our study

2.1. Subjects Our subjects were ninth-grade students from three middle-class junior high schools in a fairly large midwestern city in the United States. They were average students as revealed by their composite scores on the Iowa Test of Basic Skills (median = 295.50). In the spring of the school year, students were randomly assigned to either experimental (N = 74) or control conditions (N = 35). In each condition females and males were equally distributed. All data were gathered across a two-week period in regular classrooms.

2.2. Text materials We selected a passage about 1,000 words long from a science textbook and a social studies textbook, both of which textbooks were commonly used in junior high and high schools. The topic of the science passage was one that has been controversial in our society at least since the time of Charles Darwin: organic evolution and the development of primates. The social studies passage covered two controversial topics: change as progress in the first half of the passage and women's liberation in the second half of the passage. Because such controversial topics naturally lend themselves to hedging by authors, and because these passages were hedged to some extent in the first place, they are especially appropriate for studies like this. To adapt these passages for students in the control condition, we removed all hedges, producing a "factual" text. To adapt it for students in experimental conditions, we added hedging clauses to the beginning of targeted sentences. As often as possible, we targeted sentences that had been hedged in the original passages. For example, students in the science control condition read this sentence: Scientists can explain the biological change that took place in primates on the basis of the evidence. Students in some of the science experimental conditions read the example sentence with a hedging clause added: It seems to me that scientists can explain the biological change that took place in primates on the basis of the evidence. Hedging clauses appeared in several different ways: (a) in either personal voice or impersonal voice; (b) in only the first half of the passage, only the sec-

A. Crismore and W. Kopple, Effects on Attitudes and Learning

89

ond half of the passage, or throughout both halves of the passage; and (c) with low or high intensity. Hedges presented in personal voice contained personal pronouns (I, to me, for me), and those presented in impersonal voice contained the third-person pronoun it. Table 1 presents the hedges in personal and impersonal voice. Among the eight hedges in personal voice, to me occurs twice (clause internal), for me occurs twice (clause initial), and / occurs four times (clause initial). Among the eight hedges in impersonal voice, it (clause initial) occurs each time. Also, four of these impersonal constructions include passive transitive verbs, and the other four include intransitive verbs. Table 1: The Hedge Constructions Used 1. 2. 3. 4. 5. 6. 7. 8.

Personal Voice It seems to me that For me it is conceivable that / presume that It appears to me that / theorize that For me it is possible that / suppose that / hypothesize that

Impersonal Voice It seems that It is conceivable that It is presumed that It appears that It is theorized that It is possible that It is supposed that It is hypothesized that

When hedges were presented in the low-intensity condition for one half of a passage, five of them appeared. Thus, the passage for one experimental condition would have hedges in low intensity in its first half, and the passage for another experimental condition would have hedges in low intensity in its second half. The five hedged sentences were fairly evenly distributed throughout one or the other half of the passage, with three of the five hedged clauses introducing the first sentence of a paragraph, and two introducing sentences internal to paragraphs. When hedges were presented in the low intensity condition for the entire passage, one set of five hedges mentioned above in either personal or impersonal voice was used twice (once for each half of the passage), producing a total often hedges in the passage. On the other hand, when hedges were presented in the high-intensity condition for one half of a passage, three hedging clauses in addition to the five mentioned above were added. They were added to consecutive sentences near the middle of one half of a passage. Thus, one passage would have eight hedges in its first half and another would have eight in its second half. When hedges were presented in the high-intensity condition for the entire passage, one set of eight hedges would be expressed twice, producing a total of sixteen hedges in the passage. Further, when hedges were presented in the high-intensity condition for one half of a passage, four of the eight hedges were attached to sentences that adult expert readers had judged as expressing ideas essential for understanding the passage; these sentences were labelled "central idea sentences." The other four hedges in one half of a passage were attached to sentences that the same readers read and judged as expressing ideas somewhat peripheral to the gist of the passage; these sentences were labelled "peripheral-idea sentences." Examples of sentences expressing central ideas and of sentences expressing peripheral ideas appear in Table 2.

90

ΠΙ. Interactive Aspects of Hedging

Table 2: Examples of Manipulated Central and Peripheral Idea Sentences Domain

Central Ideas

Science

It seems that if a ground-ape could stand on his hind legs, he would have an advantage. It appears to me that the key to their survival was the ability to change eating habits.

Social Studies

I theorize that equality will have to be supported and tested by the courts, by all levels of government, and by men as well as women. It seems to me that with the beginning of the Industrial Revolution, it became common to regard all change as change for the better or progress.

Domain

Peripheral Ideas

Science

I theorize that when they came to the other patch of forest, they might have found apes in possession of that territory. It is conceivable that somewhere in the rocks there exist fossils of the tree-living apes of that period of increasing dryness.

Social Studies

I hypothesize that it is not progress to go even faster. It is possible that the air in the upper atmosphere is disturbed with severe and uncertain consequences.

2.3. Pretests andposttests

After preparing the different versions of the passages, we devised several tests; one measured student beliefs about knowledge, certainty, authority figures, and textbooks. Another measured how positively or negatively students felt about the topic that they would see addressed in a passage. Still another measured their attitudes toward the author of the text, some features of the text, and themselves as readers. The final test measured how much they had learned aboutthe subject. Personal beliefs and opinions questionnaire (PBOO). Research has de-monstrated how important students' values and beliefs about knowledge, certainty, and authority can be for comprehension (e.g., Perry 1970; Schommer 1990). Therefore, we developed a 25-item questionnaire with a four-point scale (1 = very likely; 4 = very unlikely) to measure our students' viewpoints about knowledge, certainty, and authority. For example, we asked: How likely are you to examine an issue that is important to you from several different angles before making a decision on it? admire people who are not always so sure and positive about what they say? challenge or raise questions about what the author of your text book writes?

Attitude test I: content attitude. Five-item questionnaires, the Science Content Attitude Test (SCA) and the Social Studies Content Attitude Test (SSCA), were developed to measure students' attitudes toward the content of each of the passages. These tests were deliberately text dependent. Two questions were based on content expressed in the first half of the passage, and three questions were based on content expressed in the second half of the passage.

A. Crismore and W. Kopple, Effects on Attitudes and Learning

91

These tests required students to indicate on a four-point scale how positive or negative they felt about ideas about which they had probably developed prior attitudes and which are discussed in the test passages. The SCA and SSCA tests were given to students before and after they read the passages, in an effort to measure how their attitudes toward the content of the passages were affected by the processes of reading and by hedges. The following five questions were used on the science passage: How would you describe your attitude toward the idea that once there were new kinds of monkeys with a particular development of the bones and muscles of the arms, shoulders, and chest that allowed diem to swing by the arms from branches? scientists can explain biological changes in primates on the basis of the different kinds of evidence they have gathered so far? early man developed from the apes that were unfortunate enough to be forced to the ground when dry periods came and forests disappeared? there is a missing link between early primates and early man? there were different species of early men and those with better brains survived?

The five questions for the SSCA social studies passage are shown below: How would you describe your attitude toward the idea that: progress means limiting quantity, size, and speed, in order to work, progress means not having the most or the biggest or the fastest? in our country there are good reasons for keeping technical developments in check? women should free themselves from the dominance of men? women should question the traditional roles diey have be expected to play? women should be allowed equal opportunity in the worlds of business, industry, and government, as well as in the arts and sciences?

Attitude test II. To discover students' attitudes toward characteristics of the author, passage, and themselves as readers, we developed a three-part, fivepoint scalar test with eight items in each part. The students' responses could range from 1 (very positive) to 5 (very negative). Sample traits for the author that this test focused on included interestingness, warmth, confidence, credibility, and morality. Sample aspects of the passage that this test focused on included level of formality, connectedness, readability, persuasiveness, and bias. Finally, sample aspects of the students as readers included motivation to continue reading, self-confidence as a reader, and the degree of evaluation they brought to texts. Students took the test after they read the science passage and also after they read the social studies passage. Retention tests. To assess learning from reading, two cognitive measures were developed: the Science Retention Test (SRT) for the science passage and the Social Studies Retention Test (SSRT) for the social studies passage. We used multiple-choice tests with four alternatives to measure students' retention of ideas previously identified as either central or peripheral. Each retention test consisted of sixteen questions, eight assessing recall of central ideas, and eight assessing recall of peripheral ideas. For students in the experimental conditions,

92

III. Interactive Aspects of Hedging

eight questions focused on ideas in sentences with hedges attached, and eight focused on ideas in sentences without hedges. For students in the control condition, all sixteen questions assessed recall of ideas in sentences without hedges. Examples of questions on central-idea sentences and on peripheral-idea sentences are shown in Table 3. Table 3: Examples of Questions Based on Manipulated Central-Idea and Peripheral-Idea Sentences Domain

Central Ideas

Science (a.) (b.) (c.) (d.)

The apes comprising the "missing link" are those that lived in trees during the time of less and less moisture that never learned to use sticks as tools, without hemoglobin in the blood, that never lived in trees.

(a.) (b.) (c.) (d.)

Considering all change as change for the better became common with the beginning of the Agricultural Revolution, greater yields per acre of croplands, flights from New York to Paris, the Industrial Revolution.

Social Studies

Domain

Peripheral Ideas

Science (a.) (b.) (c.) (d.)

When wandering apes came to a new patch of forest, they probably found others of their kind living there, many healthy trees, adequate water supplies, dying apes.

(a.) (b.) (c.) (d.)

We probably cannot consider it progress when we produce bigger products than anyone else resist the burden of grinding toil, feed more of the hungry than anyone else, spend more money on building more advanced missiles.

Social Studies

In sum, each retention test consisted of one eight-item subtest for central ideas and one eight-item subtest for peripheral ideas. These central-idea and peripheral-idea subtests were then further subdivided. The eight items on each subtest were divided equally into items assessing statements with and without hedges. The six subtests, then, included (1) Central Ideas, (2) Central Ideas With Hedges, (3) Central Ideas Without Hedges (4) Peripheral Ideas, (5) Peripheral Ideas With Hedges, and (6) Peripheral Ideas Without Hedges.

2.4. Design and procedure The design for the experimental group was a 2 (Personal, Impersonal Voice) X 2 (Low, High Intensity) X 3 (First Half, Second Half, Both-Halves Location) factorial design. Each student was randomly assigned to a particular experimental condition or to a control group. Students who read science and social studies passages with hedges read them in the same condition but in counterbalanced order.

A. Crismore and W. Kopple, Effects on Attitudes and Learning

93

After gathering students' scores on the Iowa Test of Basic Skills from the schools, we administered the following pre-tests to students on the first day of testing: 1. 2. 3. 4. 5.

Personal Beliefs and Opinions Questionnaire (PBOQ) Wide Range Achievement Test (WRAT) (Jastak Associates, 1962) Attitude I Content Tests (SCA; SSCA) Science and Social Studies Attitude II Tests Science and Social Studies Retention Tests (SRT; SSRT)

On another day, approximately one week later, students read either a science passage or a social studies passage, those in the experimental conditions reading passages with hedges, and those in the control group reading passages without hedges. After reading a passage, students were given the appropriate posttests for attitudes and retention.

3. The findings from our study 3.1. Iowa test of basic skills/wide range achievement test (WRAT) findings

Based on ANOVA statistical findings, the scores on these two subject characteristics tests were higher for the combined hedge experimental groups than for the control group but were not significantly higher. 3.2. Personal beliefs and opinions questionnaire (PBOQ)findings The findings from the Personal Beliefs/Opinions measure help us better understand specific aspects of our student readers: their beliefs about knowledge, authority, certainty, controversies, multiple perspectives, critical reading, author credibility, goals of education, and the like. Table 4 shows the ranked scores of all students combined as to their beliefs and opinions reported as most likely, somewhat likely, and least likely about the aspects mentioned above. Some specific findings are highlighted in the following. Students most strongly believed that science and social studies teachers and textbooks were authorities who were helpful to them and would not intentionally mislead them when giving opinions or making claims. Students reported they themselves examined an issue that was important to them from several viewpoints before taking a position on it; they were also somewhat likely to look for evidence to support claims and pay attention to the philosophy (ideology) of their science and social studies textbooks. Interestingly, they reported that they were more likely to admire people when those people were certain rather than uncertain about what they said. In other words, they were rather negative about people who hedged their statements. They were also somewhat likely to accept and respect people with beliefs and views that were in conflict with theirs and to empathize with people. In addition, they were somewhat likely to believe that knowledge is subjective and that the goal of education is

94

III. Interactive Aspects of Hedging

to think critically and evaluate controversial issues. Yet they indicated they were least likely to challenge science and social studies textbooks and pay attention to the background experiences or personalities of authors of textbooks. They were also least likely to consider themselves as experts in aspects of science or social studies. Table 4: Ranked PBQJScores for die Hedge and Control Groups Combined *(1 = Very Likely; 4 = Very Unlikely) Item

Average Scores

Most Likely Considering science/social studies teachers as authorities Admiring people who aren't always certain about what they say Seeing people in authority over you as helpful Considering science/social studies textbooks as authorities Examining an issue important to you from several angles before making a decision on it

1.7 1.8 1.9 1.9 1.9

Somewhat Likely Accepting and respecting people with beliefs and views different from yours Epadiasizing with (putting yourself in the shoes of) others Paying attention to whedier science textbook authors give evidence for claims they make Believing knowledge is a set of subjective personal opinions with some opinions better than others Defining the goal of education as helping students evaluate and take a stand on controversies Paying attention to the philosophy and basic beliefs (ideology) of science/social studies textbook authors Admiring people who are not always certain about what they say

2.1 2.1 2.1 2.3 2.5 2.5 2.6

Least Likely Challenging and raising questions about what science/ social studies textbook authors write Believing classmates consider you a possible authority in some aspect of social studies Trying to get a clear picture of science/social studies textbook authors' personality Paying attention to the background experiences and education of science/social studies textbook authors Believing that people in authority over you would give opinions in order to mislead you intentionally (on purpose)

2.7 2.7 2.8 3.0 3.4

The differences between the scores of the hedge and control groups were not large except for the scores on a few items. In general, students in the hedge

A. Crismore and W. Kopple, Effects on Attitudes and Learning

95

group were a little more positive than the control group in the following beliefs: that knowledge is a set of subjective beliefs with some opinions better than others; that the goal of education is to help students learn to evaluate and take a stand on controversial issues; that people who are not always sure or positive about things are admirable; that social studies teachers are authorities; that people with views conflicting with theirs should be accepted and respected; and that classmates might consider them experts in some aspects of science and social studies. Thus the hedge population seems to be somewhat more confident, open, and flexible; and to a greater extent than the control group does, it seems to include people who are critical thinkers and who see value in uncertainty. The scores of the control group indicate that they were a little more likely than the hedge group to empathize with people, to care about the personalities of their textbook authors, and to look for evidence when authors of social studies textbooks made claims. On the Personal Beliefs and Opinions Questionnaire, we found several statistically significant differences between genders. The females' scores were more positive than the males' in regard to empathizing with other people, accepting and respecting those whose beliefs and views differed from their own, paying attention to the background experiences and education of authors of the science and social studies textbooks, and seeing people in authority over them as helpful. Females were also more anxious than males while reading a science text and when seeing authors using hedges in a science text. The males' scores were more positive than the females' in believing that classmates would consider them experts in some area of science or social studies. Further, males, more so than the females, felt a sense of power and control over their life in school, felt competent in science and social studies in and out of the school context, and felt comfortable when taking a test in science or social studies. 3.3. Attitude test I: content attitude findings To derive measures of our readers' attitudes toward the subject matters before they read about them, we assigned a number to each possible response on the content attitude tests. We assigned the "very positive" response a 1, the "somewhat positive" response a 2, the "somewhat negative" response a 3, and the "very negative" response a 4. The total scores on the tests could range from 5 (very positive) to 20 (very negative). Then we calculated the mean response for all of our subjects for each of the five items on the Content Attitude Tests for science and social studies. Content attitude I: science (SCA) findings. The means for science revealed that our readers had quite negative attitudes toward the content of the passage. The sum of the means for all five questions on the Content Attitude Science Test was 14.41 (SD = 2.63). This mean is very close to that which we would have obtained had all subjects chosen the "somewhat negative" response for all five questions. That readers were especially negative about the implications or explicit claims about primate evolution made in the passage is shown by an examina-

96

III. Interactive Aspects of Hedging

tion of the mean scores for Questions 3, 4, and 5, the questions that focused on ideas about evolution. The mean for Question 3 was 3.54 (SD = 0.65), that for Question 4 was 2.83 (SD = 0.99), and that for Question 5 was 3.18 (SD = 0.91). As is evident in these means, the students' negative attitudes snowed themselves most clearly in response to questions 3 and 5. Question 3 reads as follows: "How would you describe your attitude toward the idea that early man developed from the apes that were unfortunate enough to be forced to the ground when dry periods came and forests disappeared?" And question 5 reads as follows: "How would you describe your attitude toward the idea that there were different species of early men and that those with better brains survived?" When our subjects answered these questions before reading the passage, they indicated that they had a stance between somewhat and very negative toward the ideas that were raised. Perhaps our subjects' attitudes toward many of the ideas in the science passage were negative because those ideas conflicted with the religious beliefs of our subjects, all of whom lived in the relatively conservative midwestern United States. Content attitude I: social studies (SSCA) findings. The sum of the means for all five questions on the Attitude I Social Studies Test was 10.22 (SD = 2.51), revealing that students were much more positive about the ideas in the social studies passage than those in the science passage. The means for the five questions are as follows: Question 1 (progress means not having the most, biggest, or fastest) = 2.34; Question 2 (there are good reasons for keeping technical developments in check) = 1.82; Question 3 (women should free themselves from the dominance of men) = 2.18; Question 4 (women should question their traditional roles) = 2.33; and Question 5 (women should have equal opportunities in business, industry, government, arts and sciences) = 1.55. Students' positive attitudes are especially evident in Question 1 and Question 5; their stance on the issues addressed was between very and somewhat positive. Perhaps our subjects' attitudes toward the ideas expressed in the social studies passage were more positive than their attitudes toward many of the ideas in the science passage because the ideas in the social studies passage did not conflict with or maybe even were consonant with our subjects' religious beliefs. Content attitude I: science (SCA) difference scores findings. Students responded to the content attitude questions both before and after they read the passages. The differences between these scores showed whether their attitudes stayed the same or became more positive or negative after the reading. Our statistical analysis of the difference scores on the Science Content Attitude Test showed that hedges did in fact affect students' attitudes: their attitudes became more positive. All the science hedge treatment groups had attitude difference scores more positive than those for the control group. As we see in Table 5, the positive changes were almost three times greater for the total population of hedge groups than for the control group (statistically significant). The prescores for the total content attitude test show that the hedge group was initially much more negative than the control group (14.41 vs. 13.80). Questions 1 and 3 showed the most dramatic differences between the two groups. On Question 1, about the evolution of monkey bone and muscle development, the total population of students in hedge groups became more

A. Crismore and W. Kopple, Effects on Attitudes and Learning

97

positive (-.29) while the control group became more negative (+.26). On question 3, about early man's evolution from apes, those in hedge groups became more positive (-.27), while the controls had almost no change in attitudes (-.05). Table 5 also shows that the control group had more positive attitudes on the pre-test than did the hedge groups. We found that although the mean difference score for the total population was -.76, there were some gender differences: the mean difference score for the females (N = 38) was -1.03, while for the males (N = 36) it was only -0.47. The positive attitude changes for the females were more than twice as great as for the males. Finally, Table 5 also reveals that the hedge variables most instrumental to positive attitude changes were both-halves location, high intensity, and impersonal voice. Table 5: Difference Score Results for the Science Attitude I Tests )ifferenc( ; Scores Ranked for Attitude Changes I. Science: Attitude 1 1 Variables/Groups

Total Test N Difference Scores

A. Experimental Groups Both Halves High Intensity Impersonal Voice Total Population Second Half Personal Voice

22 37 37 74 26 37

-1.28 (3.30) -1.05 (2.69) -.94 (3.03) -.76 (2.73) -.58 (1.56) -.54 (1.56)

14.09 (3.09) 14.49 (2.42) 1424 (2.56) 14.41 (2.63) 14.69 (2.67) 14.54 (2.73)

12.82 (3.47) 13.43 (2.93) 13.30 (3.32) 13.65 (3.06) 14.12 (2.69) 14.00 (0.54)

B. Control Group

35

-.20 (2.27)

13.80 (2.80)

13.60 (2.27)

Total Test Pre Scores

Total Test Post Scores

II. Science Attitude I Difference Scores for Combined Hedge Groups* and Control Groups Question Pre

1 2 3 4 5 TOTAL

2.60 2.60 3.40 3.00 320

Hedg e Group Post (0.89) (0.89) (0.89) (1.00) (0.89)

2.00 2.20 3.00 2.80 2.71

(0.71) (0.45) (1.00) (0.89) (2.40)

Differences Pre -.60 (0.89) -.40 (0.89) -.40 (0.55) -.20 (0.45) -.49 (0.89)

14.80 (3.56) 12.71 (1.???) -2.09 (2.40)

2.00 2.46 3.54 2.83 2.97

COD trol Group Post

Differences

226 2.43 3.49 2.71 2.71

+26 (1.07) -.03 (1.04) -.06 (0.84) -.12 (1.02) -.26 (0.92)

(0.89) (0.85) (0.82) (0.99) (0.95)

13.80 (2.80)

(0.85) (0.85) (0.78) (1.10) (0.89)

13.60 (3.08)

-.20 (2.27)

"Variables = Personal Voice, High Intensity, and Both Halves Location

However, for some treatment groups, such as Treatment 6, personal voice rather than impersonal voice was more effective in producing positive attitude changes. Part I of Table 6 shows that Treatment 6 (hedges in the personal voice, high intensity and both halves of the passage) produced a difference score of2.09, a score many times larger than the contro 1 group's difference score of -.20.

98

ΙΠ. Interactive Aspects of Hedging

Content attitude I: social studies (SSCA ) difference scores findings. Our findings for the Social Studies Attitude I test were quite different from those for the difference scores on the Science Content Attitude I Test. Table 6, Part I, shows that the social studies mean difference scores for the total population of students was -1.22, much higher than the -.76 score for the science test. The difference score for the males (-1.24) was about the same as for the females (-1.20); this finding stands in contrast to what happened on the science content attitude test, for which the males' difference score was only -.47 and the females' difference score was -1.03. The results for the hedge variables were also interesting. The mean difference scores for the intensity variable were about the same for low intensity (-1.23) as for high intensity (1.21). These scores are in contrast to the mean difference scores for the intensity variable in the science test, for which we found that high intensity (-1.05) was much more effective in changing attitudes than low intensity (-.46). Analysis of the voice variable in the social studies test revealed that personal voice produced somewhat more positive attitude changes (-1.27) than did the impersonal voice (-1.16). On the science test, however, the mean difference score for personal voice was only -.54, while for impersonal voice it was -.95. The mean difference scores for the location variable on the social studies test were much higher for the second-half location, which focused on women's liberation (-1.60) than for the first-half location (-1.35) or for bothhalves location (-.67). For the science hedge groups, the location variable produced quite different results: the score for both-halves location was much higher (-1.27) than the scores for the first half (-.50) or second half (-.58). When we combined all the hedge experimental groups together, we found that their attitudes became more positive (-1.22) from the pretest score (10.28) to the post-test score (9.04). But when we looked at the controls, we found that their attitudes became even more positive (-1.53) from the pretest score (10.50) to the post-test score (8.97). Looking more closely at the hedge experimental groups, we found that four groups had difference scores higher than the controls and that the rest had lower scores. Thus the averaged scores for the hedge groups apparently canceled out the positive effects of some of the experimental groups. For example, the experimental group that saw hedges in personal voice, low intensity, and in both locations (Treatment 4) is illustrative of the strong attitude changes experienced by the four hedge groups mentioned above. The results that appear in Part II of Table 6 show that even though the control group and the Treatment 4 group began with the same prescore of 10.50, the Treatment 4 hedge group had a difference score of -2.50, while the control group had a difference score of only -1.53. The scores for the other three hedge groups were also impressive, with difference scores of-2.83, -2.43, and -1.86. All the other hedge groups had difference scores ranging from -1.33 to -0.29. We found little disparity between the mean difference scores for the males (-1.24) and those of the females (-1.20). But the voice variable did interact with gender: the males had higher difference scores with personal voice (-2.54) than the females did (-.58). The females had higher difference scores with impersonal voice (-2.06) than the males did (-.40). These results were statistically significant (p = .002).

A. Crismore and W. Kopple, Effects on Attitudes and Learning

99

Table 6: Difference Score for the Social Studies Attitude I Test I. Social Studies vs. Science Attitude I Difference Scores for the Hedge Variables Social Studies (N = 74) Combined Hedge Groups Difference Score = -122

Science (N = 74) Combined Hedge Groups Difference Score = -.76 Gender

Gender Males -1.24 (N=33)

Females -120 (N=41)

Males Females -.47 (N=36) -1.03( N=38)

Low -123 (N=35)

High -121 (N=39)

Personal -1.27 (N=37)

Impersonal -1.16 (N=37)

Personal -.54 (N=37)

Second -1.60 (N=25)

Both -.67 (N=24)

Location Second Both First -.58 (N=26) -127(N=22) -.50 (N=26)

Intensity

Intensity

Low High -.46 (N=37) -1.05 (N=37) Voice

Voice Location First -1.35 (N=25)

Impersonal -.97 (N=37)

II. Social Studies Attitude I Difference Scores for Treatment 4 Hedge and Control Groups Treatment 4 Hedge Group Post Pre Question 2.17 (0.75) 1.50 (0.55) 1 2 2.17 (0.98) 1.50 (0.84) 2.50 (123) 1.83 (1.17) 3 4 2.17 (0.98) 1.50 (0.84) 1.50 (0.55) 1.67 (0.82) 5 TOTAL

10.50 (2.81) 8.00 (2.83)

Differences -.67 -.67 -.67 -.67 + .17

Pre 2.33 (0.93) 1.75 (0.73) 2.31 (0.89) 2.61 (0.93) 1.50 (0.81)

Control Group Post Differences 2.08 (0.90) -25 1.42 (0.65) -.33 1.92 (1.08) -.39 2.11 (0.98) -.50 1.44 (1.86) -.06

-2.50 (1.38)

10.50 (2.35)

8.97 (2.47)

-1.53 (2.50)

The most important findings from the Content Attitude Tests were that readers' attitudes and the context in which the hedges were presented (personal vs. impersonal voice, low vs. high intensity, first-half vs. second-half vs. bothhalves location) made a difference for the effects on readers: some combinations of the hedge variables resulted in positive attitude changes, and some resulted in negative attitude changes. 3.4. Attitude IIfindings After reading the science passage and the social studies passage, students used the Attitude II Science and the Attitude II Social Studies Tests lo rate their attitudes toward the author, text passages, and themselves as readers and learners. We compared the attitudes of the hedge groups and the control group for both the science and social studies passages. We combined the attitude scores for the science and social studies passages and averaged them to get an overall sense of the students' attitudes when they read with and without hedges. We found that scores for 4 of the 24 items had a difference of .15 or above between the

100

III. Interactive Aspects of Hedging

The control group had more positive attitudes than the hedge groups toward the authors; specifically, the control group found the authors more confident (not surprisingly) as well as more believable and moral than the hedge groups did. For the hedge groups, the text seemed more informal, more connected, more biased, and had more of the author's personality present than for the control group. When asked to indicate their attitudes toward themselves as readers and learners, students in the hedge groups' had more positive scores than those in the control group for being engaged with the passages, being motivated to continue reading the passages, and being active readers. Thus, when we combined the Science and Social Studies Attitude II scores, we found mixed results for the effects of hedges. When we examined the scores of the hedge group and control group for the science and social studies passages separately, we found more negative scores for the science passage than for the social studies passage. Students in the hedge groups for the science passage did not find the author as warm as the control group did, but those in the hedge groups found the text easier to read than the controls did. Students in the hedge groups also saw themselves as easier to persuade, more inclined to be evaluators of texts, but less self-confident as learners than the control group. Students in the hedge groups for the social studies passage saw the author as being more an interpreter of texts, more respectful of readers, and more warm than the students in the control group did. Students in the hedge groups also indicated that they preferred the unfamiliar more than the controls did. Findings for the Attitude II Tests appear in Table 7. We did not find any significant differences between the males' and females' responses to these tests. Table 7: Hedge and Control Groups Difference Scores for Attitude II* Science and Social Studies Item Hedge Group

Science Control Group

Difference Scores

Hedge Group

Social Studies Control Difference Group Scores

Description I. Author Characteristics Interpreter of Facts 3.20 Respects Reader 2.78 Considerate of Readers 2.72 Interesting 2.97 Warm 3.42 Confident 2.32 Believable 3.41 Moral 3.26

3.21 2.71 2.65 3.09 3.09 1.94 3.12 3.09

.01 .07 .07 .12 .33 .38 .19 .17

3.49 2.34 2.45 2.62 2.95 1.88 1.93 2.45

3.83 2.54 2.50 2.66 3.14 1.71 1.71 2.20

.34 .20 .06 .04 .19 .17 .22 .25

II. Text Characteristics Natural Language Formal Connected Easy to Read Easy to Identify With Au^orPasonalityFiesent Persuasive

2.50 2.56 2.88 2.56 3.24 3.56 3.27

.07 .24 .27 .30 .04 .33 .12

2.18 2.73 2.40 1.99 2.40 2.78 2.55

2.20 2.46 2.60 1.87 2.51 2.87 2.46

.02 .27 .20 .12 .11 .09 .09

2.57 2.80 2.61 2.26 3 .20 3.23 3.39

A. Crismore and W. Kopple, Effects on Attitudes and Learning Item

ΙΠ. Reader Characteristics Motivated to Continue Active Reader Easy to Persuade Self-Confident as Learner Prefer the Unfamiliar Evaluator of Texts/Authors Close to Author Respect for Author

101

Science Social Studies Hedge Control Difference Hedge Control Difference Group Group Scores Group Group Scores 3.33 2.97 3.69 2.59 3.30 2.75 3.66 2.77

3.76 3.30 4.03

224 3.33 2.94 3.70 2.70

.43 .67 .34 .35 .03 .19 .04 .07

2.78 2.57 3.15

220 3.03 3.03 3.08 2.35

2.65 2.71 3.29 2.29 3.21 3.00 3.12 2.30

.13 .14 .14 .09 .18 .03 .04 .05

*A 5-point Bipolar Lickert Rating Scale: 1 = positive; 5 = negative

As the Attitude I Tests did, the Attitude II Tests show that readers' attitudes can change dramatically when they encounter hedges. Hedges affected readers' attitudes toward the subject matter, the author, the text, and themselves as readers and learners. Usually the readers' attitudes became more positive when they read hedges, but this effect no doubt depended on the attitudinal objects, namely, what their attitudes were about. 3.5. Retention tests (SRT and SSRT) findings

We descriptively and statistically analyzed the students' performance (i.e., learning gains based on pre- and post-test difference scores) on the Science and Social Studies Retention Tests and found a remarkably consistent pattern of results. Interestingly, the students who learned the most were not those who read the science and social studies passages without hedges. Rather, they were those who read the passages with hedges specifically, with hedges in personal voice, the second half, and low intensity. (These same variables produced the largest gains in positive attitudes for the Social Studies Attitude I Test, but not for the Science Attitude I Test.) Thestudents who had made the greatest learning gains were followed by those who had read a version with hedges in the second half, then by those who had read a version with hedges in low intensity, then by those in the control condition, and then by those who had read a version with hedges in the impersonal voice. These groups were followed by several others. The learning scores of those at the bottom of our ranked list contrast sharply to those at the top. The students who made the smallest learning gains were those who encountered hedges in impersonal rather than personal voice, both-halves rather than second-half location, and high rather than low intensity. The ranking of the means for various conditions for the science passage appear in Table 8 (Page 105). Based on our analyses of the difference scores on the science and the social studies retention tests, we can note that hedges affected females differently than they did males. For the science retention test, the females' mean difference score was 4.63 and the males' was 3.53. However, for the social studies retention test, we found the opposite result: The males' mean difference score was 3.48 while the females was 3.12. The females had much higher difference

102

III. Interactive Aspects of Hedging

scores for science than for social studies, but the males' difference scores were about the same for both tests. The scores indicate that hedges helped the females learn more from the science passage than from the social studies passage and that hedges helped the males learn equally well from both the science and social studies passages. Table 8: Ranked Means and Standard Deviations for the Total Retention Test Difference Scores Groups/Variable

N

Science

N

I. Experimental Groups Combined Variables Personal Voice, Second Half Location, Low Intensity Second Half Location

8 12

6.63 (3.42)

6

4.17 (2.48)

26

25

Low Intensity

37

Personal Voice

37

Total Population

74

First Half Location

26

Impersonal Voice

37

High Intensity

37

Both Halves Location

22

5.54 (3.22) 4.81 (3.13) 4.70 (3.00) 4.09 (3.32) 3.69 (3.08) 3.49 (3.31) 3.37 (3.15) 2.86 (2.75)

4.12 (3.31) 3.77 (3.68) 3.49 (3.49) 3.28 (3.37) 3.12 (3.06) 3.08 (3.74) 2.85 (3.05) 2.58 (3.68)

6

1.33 (3.39)

6

1.33 (4.13)

35

4.77

36

3.97

Combined Variables: Impersonal Voice, Both Halves Location, High Intensity II. Control Group

(2.65)

37 37 74 25 37 37 24

Social Studies

(3.29)

3.6. A summary of the effects of gender differences

Although our primary focus in this chapter is on the effects of hedges for people in general, we, as noted above, found effects of gender differences, which are worth a brief summary here. Statistical analysis of the PBQ^questionnaire revealed that of the total population of students, males were significantly more negative than females when it came to admiring people who were unsure and uncertain about what they said (for example, people who hedged); paying attention to whether science and social studies textbook authors supported their claims; and thinking that classmates considered them as experts in science. Females, on the other hand, were more negative than males when asked whether they considered science and social studies teachers as authorities.

A. Crismore and W. Kopple, Effects on Attitudes and Learning

103

3.7. The major attitude findings: A discussion

The results for the Attitude I Tests (SCA and SSCA) showed that, in general, the hedge variables of high intensity, impersonal voice, and both-halves location were more effective in producing positive attitude changes than other variables for both science and social studies. For science, all hedge conditions produced greater positive attitude changes than the control condition did, and the attitude changes for females were twice as strong as for the males. For social studies the control group had greater positive attitude changes than the hedge groups considered together, and there were no differences in positive attitude changes between females and males. However, students in four treatment groups (for example, the treatment group with the hedge variables personal voice, low intensity, and second-half location), had greater positive attitude changes than students in the control group. It appears that hedges within controversial written texts act as a powerful inducement for readers to change their attitudes positively toward the subject matter, perhaps to some extent regardless of relatively fine distinctions in where the hedges are presented, in what voice they are presented, and how often they are presented. In our tests, greater positive attitude changes usually took place when hedges were present in the materials to be read than when they were not, but the specific attitudinal objects and aspects of the overall context of the experiment surely played a role in how hedges influenced attitude changes. Because of the many salient aspects of our experimental context, and because the specific variables that most afl'ected readers' attitudes were not always the same as those that most affected readers' learning, it is a challenge to explain our results fully. Raija Markkanen and Hartmut Schroder (1989, 12) have emphasized that "the linguistic devices used for hedging get their meaning only through the response they produce in the readers. Therefore, the analysis of hedging in a particular text should begin with an analysis of the communicative situation." We must now ask how these effects on readers' attitudes in our study are related to the rhetorical context created in the study, for as Watzlawick, Beavin, and Jackson (1967, 21) point out, "Failure to realize the intricacies of the relationship of an event and the matrix in which it takes place (...) either confronts the observer with something mysterious or induces him to attribute to his object of study certain properties which the object may not possess." The context of our study included the school classroom, original and modified science passages on primate development, social studies textbook passages on change and progress and on women's liberation, and the students and their attitudes toward the topics. The schools and classrooms selected for our study are typical in that they are part of an educational system whose goal, in part, is to dispense factual knowledge to students by means of textbooks and teachers and then to test students on their memory for that information. One of the first to examine the status of textbooks in such a situation, David Olson (1989, 234) claims that: "They [textbooks] are taken as the authorized version of a society's valid knowledge. The students' responsibility is primarily to master this knowledge." Olson adds that students typically do not feel that they have the right to disagree with the authorized texts but must master them and be prepared to defend them because of

104

III. Interactive Aspects of Hedging

the difference in status between writer and student and between teacher and student. Most students regard what they read and hear in their classroom as true; because of their schooling, they view textbooks (and their teachers as extensions of textbooks) as truthgivers. They have learned to expect typical textbooks and teacher talk to be presented objectively, impersonally, and without stance or voice. For example, authors of science textbooks strive for clarity of meaning and for explicit and unambiguous presentation of factual data; and classroom teachers, as Felman and Weitsch found in their study, rarely use expressions of uncertainty; hedges and expressions of attitudes (such as may, might, could. I think, believe, hope, and feel) are usually lacking in the talk of teachers to students (cited in Olson 1989). Presented with one flat assertion after another in much of their schooling, many students learn to value certainty rather than contingency. And, in relation to both their textbook and their teachers, they generally assume the docile, acquiescent, non-authoritative stance described by Luke, de Castell, and Luke (1989). Such findings and claims are confirmed to an extent by the responses to our Personal Beliefs and Opinions Questionnaire (PBOQ). The students, ninthgrade adolescents, were novice readers who had a set of expectations about what written instructional materials should be like. The students reported that they considered science and social studies textbooks and teachers as helpful authorities and were not very likely to care much about their credibility or ideologies - or raise questions about what such authorities claimed. And they did not believe that others considered them as experts in science or social studies. As the Science Content Attitude Test (SCA) showed, our subjects initially had quite negative attitudes toward the topic of the science text. These attitudes also showed themselves outside of the testing context; as many students left the testing room, we noticed that they scoffed about the possibility of humans' having evolved from apes. The science materials might have threatened some of their deeply held beliefs about the history of man, the nature of the world, and the character of the forces controlling the world. The Social Studies Content Attitude Test (SSCA) showed that our subjects were less negative about the topics of change as progress and women's liberation, perhaps because these secular topics did not threaten any deeply held beliefs. All such factors involving how readers see themselves, the reading situation, the kinds of texts, and the specific topics of those texts would have to be considered in a full explanation of why all of the hedge groups reading the science passage had greater positive attitude changes than did the control group and of why several of the hedge groups reading the social studies passage had greater positive attitude changes than did the control group. Attitudes are difficult to change, especially when people have strong commitments to them because of some personal stake in them or because of their upbringing. Research has demonstrated the tenacity with which people adhere to their theories and opinions and attitudes (e.g., Nesbett & Ross, 1980), and rhetoricians for centuries have discussed the challenge involved in changing the minds and behaviors of hostile audiences (e.g., Horner, 1988). Tyler and Voss (1982) claim that whether attitude change occurs or not depends on the nature

A. Crismore and W. Kopple, Effects on Attitudes and Learning

105

of the communication - its relevance to the person's value system and life goals. For both the science and the social studies passages, hedges may have led to positive attitude changes because they, sounding their notes of uncertainty, alerted readers that these passages were written by a real person or by real people and that the information was not some socially authorized set of information which they simply had to memorize but was information allowing them some room to make up their minds about. This allowing them some room to make up their minds about information is in itself a means of granting students more power than they probably often have in school situations. Further, since the students probably were able to see the information as controversial and not as firmly established as fact, they might even have appreciated the sense of honesty that the hedges brought to the material. They might even have seen an author who dares to hedge as possessing some humility, a trait that they might well have found very attractive in a person writing materials for them to read at school. We can offer some additional speculative comments about the students in the hedge groups for the science passage, all of whom had more positive attitude changes than did the students in the control group. The students in the hedge groups may have become more positive in attitudes toward the subject matter than those in the control group did because the students in the hedge groups were better, more critical and more reflective readers. Evidence from the Iowa Tests, the Wide Range Achievement Test, and our Personal Beliefs and Opinions Questionnaire suggests that this was the case. The hedges, in their explicit signaling of the author's stance toward the subject matter, may have both activated the students' attitude schemata and intensified their use of reflective, critical-reading strategies. Several studies have shown the importance of readers' organized networks of attitudes and knowledge (schemata) and the interactive influences of affect and comprehension (Fiske/Linville 1980; Martins 1982; and van Dijk 1982b). When a text like our science text is threatening and readers are hostile, hedges may help the more able readers process the text deeply and eventually change their initially negative attitudes. But the degree of attitude change for readers seeing hedges in controversial texts may depend on whether the hostile readers are male or female. The females in the hedge groups changed their attitude much more noticeably than the males, perhaps because they were even better readers than the males, because (as some of the data on the Personal Beliefs and Opinions Questionnaire indicate) they had more positive attitudes toward people with viewpoints different from theirs, or, even more likely, because they read differently. The research on gender differences in reading and speaking suggests that females read with different goals than males do. Females seem to be more interested in harmony, congruence, and interpersonal relationships than males are (Coates/Cameron 1989; Flynn/Schweichart 1986; Holmes 1982), so perhaps they are easier to persuade than males are, especially when an author uses signs such as hedges that can be seen as conciliatory and perhaps as contributing to inclusiveness.

106

ΠΙ. Interactive Aspects of Hedging

For the students in the hedge groups reading the science texts, hedges presented in impersonal voice worked better to change attitudes than did those presented in personal voice. But gender made a difference for this variable. The females' attitude difference scores for the science attitude test were greater when they saw hedges in impersonal voice. The males' difference scores were higher for personal voice. This finding is probably related to something shown on the Personal Beliefs and Opinions Questionnaire. The females had more positive attitudes toward authorities than did the males. Females probably saw the hedges in impersonal voice as closely related to authorities and associated their positive feelings about authorities with the hedges, through this process becoming more positive about the subject matter of the passage. Or perhaps the females read the impersonal hedges as being written by more than a single author and therefore associated them with greater authority. The high intensity conditions were more effective than the low intensity conditions in changing students' attitudes positively toward the science material. We think this may be because when students' beliefs and attitudes are in conflict with, and perhaps threatened by, those the author expresses, they probably need a good measure of hedges in order for those hedges to break through the threat and get students to feel more positive about the subject matter. How many hedges are necessary is still unclear, but the eight per half in the highintensity conditions were more effective than were the five per half in the lowintensity conditions. In addition, in the high intensity conditions students saw hedges associated with both central ideas and with peripheral ideas. Students probably needed to see hedges on both of these levels of the text structure in order to counteract their negativity toward the ideas in the passage. The second location was a more powerful location than either first- or bothhalf locations for changing attitudes, no doubt because it was at that point that students really understood the authors' point about humans' evolution from apes and the implications for their beliefs about science and religion. The first half was more background information. The second half, where the students probably saw a threat to their beliefs, was most important. 3.8. The major retention findings

The females in the hedge groups had a higher mean difference score for the science retention test (4.63) than the males did (3.53); this finding is similar to that for the science attitude test. For the social studies retention test, the males had a slightly higher mean difference score (3.48) than did the females (3.12); this finding is very much like that for the social studies attitude test. In general, those students who learned the most from reading the science and the social studies passages were those who read the passages with hedges included - specifically, hedges in personal voice, low intensity, and the second half of the passage. (This pattern quite consistently matches the pattern producing the greatest positive changes on the social studies attitude test, and elements of this pattern played a role in producing positive changes for several treatment groups on the science attitude test.) This finding was most apparent for the science passages, and it is for the results of the science retention test that

A. Crismore and W. Kopple, Effects on Attitudes and Learning

107

we can at this time offer the fullest explanation. But many aspects of this explanation will no doubt apply to the results of the social studies retention test and perhaps to aspects of the attitude tests as well. First, we offer some comments about why hedges in general might have aided the learning of our subjects. As noted repeatedly above, our ninth-grade subjects initially had quite negative attitudes toward the subject matter of the science passage. These subjects could have responded in several ways to the controversial science material that they had negative feelings about. They could have disregarded it altogether - the "I don't want to continue reading this" response. However, these students did not stop reading, probably because they were cooperative in what they viewed as a situation in which students were subordinate to the researchers and textbook authorities. They could have read the science material dismissively, processing it superficially, assuming it was not true and not worth remembering. Such reading results when students have failed to resolve conflicts between their existing knowledge and beliefs and the new information and opinions presented in texts, and such reading naturally leads to little learning (Lipson 1982). Finally, students could have read the material carefully, considered it carefully, and evaluated it thoroughly. They could take pains to analyze ideas they felt negative about in order to determine whether their attitude was justified, and if so, to build a good case against the ideas. In fact, based on theoretical and empirical investigations such as those by Hübler (1983), it appears that controversial situations often lead to intense attempts to clarify one's own views, to understand better the views of others in order to rebut and perhaps refute them, and even occasionally to develop a synthesis from the various ideas, all of which processes can positively influence learning. And Berlyne (1960) assumes that conceptual conflict leads to epistemic curiosity, which in rum leads to an active search for additional information, a more thorough cognitive analysis, and a more precise understanding of the differing perspectives of others. Reading and responding to the science passage in this active, analytical manner, therefore, would probably lead to significant learning. We believe that the non-hedged (control) passage encouraged students to read dismissively and that certain hedged passages led to noticeable learning gains because they encouraged students to read evaluatively. These students read evaluatively because hedges clearly mark the controversial passage as an expression of opinions or hypotheses. Many students probably considered the material as hypothetical and conjectural and might have appreciated the author's honesty in marking such material with hedges. In their signaling of truth assessments, hedges call readers' attention to the fact that those assessments were made by others and should be evaluated before being accepted. In the evaluation process, the students could decide for themselves about any controversial ideas; the hedged texts were not above criticism. Olson (1989, 239) stresses that "children, too, have the right to hold their opinions, to express them, and to have them taken seriously if they are honest, informed, considered, or significant." We suggest that hedges were a welcome addition to the text for our subjects because in the course of evaluation the hedges proba-

108

III. Interactive Aspects of Hedging

bly encouraged them to put their own opinions up against the authors' and to hold their opinions as valid too. Bruce (1981) argues that when devices like hedges are used in texts, readers have the chance to react to ambivalent, contradictory feelings and thoughts. Bruce also argues that these effects lead to active involvement of readers, which in turn results in increased understanding of both authors and content and in intensified or changed attitudes. Dillon (1985) proposes five dimensions of the footing or alignment between authors and readers. He suggests that the continuum running between the point where authors and readers will meet and are equal to the point where they do not and are nonequal will be determined in part by whether or not the authors use hedges, which tone down the intensity with which a proposition is conveyed. The hedges in our science passage, then, may have encouraged students to see themselves not in a subordinate but in a more nearly mutual relationship with the author; in seeing themselves in this way, they could have become more likely to engage their critical thinking skills and to evaluate their opinions and attitudes about the subject matter. We believe that there are several reasons why on the science retention test hedges in personal voice had an advantage over those in impersonal voice. First, the personal voice could have aided learning by bringing a concrete (even imaginable) presence into the text. In Olson's terms, hedges could have helped connect language to the one who produced it. And first person pronouns, according to Dillon (1985), can increase the solidarity between authors and readers, thus fostering a dialogical, participatory relationship to the text. An author's presence in a text is a manifestation of the interpersonal function of language. Halliday (1985) explains that modality is an inherently personal part of language, since with modality markers like hedges, speakers or writers intrude into the speech act by taking a position on the truth value of the subject matter. It may be that using personal voice to express hedges draws attention to them, thus making hedged material easier for readers to recognize and process. But what is more importa ('It seems to me that'). Markkanen (1989, 145) has noted that the first person singular pronouns I and me can also be hedges, particularly when they occur together with verbs like think, suppose and seem, "because then the writer can be interpreted as meaning something like 'this is only my view, other people may think differently'". The use of personal voice probably lessens the force of the material for those who feel negative about it even more than impersonal hedges would. As Olson (1989, 240) notes, when a text originates with an identifiable author, readers "know that it is just his interests, and that he may or may not have satisfied the preparatory conditions for making an assertion. Hence his utterances are open to criticism." It is probably easier for a ninth grader to set his or her own opinions over against those of one other person than over against those of several others or of institutions such as school systems and publishing corporations. If these speculations are correct, they also help explain the finding that hedges in the second half of the science passage resulted in the most learning for readers. Although the passage is controversial throughout, it is in the second half that the material with the greatest potential to give rise to negative attitudes appears. The first half of the science passage presents a theory of early pri-

A. Crismore and W. Kopple, Effects on Attitudes and Learning

109

mate development. It theorizes about biological changes for early primates, details the development of the first apes, and hypothesizes about the gradual change from plant-eating to meat-eating apes. But it does not explicitly state that the meat-eating apes with the ability to stand erect are often considered the direct precursors of humans. The passage goes on to speculate about when this change took place and to explain a process for determining the time required for differentiation of species that are closely related. It appears, therefore, that the imporlant factor was not the second half of our controversial passage per se but rather the location of the most controversial material. In future research, we will have to distinguish more clearly location in text from location of the most controversial material. Related to this, Markkanen and Schröder (1989) argue that researchers should try to see which particular parts of a text contain a particularly high amount of hedging and then, based on an analysis of the rhetorical situation, try to explain why. Hedges in low intensity, not high intensity, helped lead to the greatest learning gains on the science retention test. An important fact to consider here is that each hedged sentence is longer than its non-hedged counterpart; the passages with hedges in high intensity were longer than those in low intensity. Moreover, each hedged sentence is more syntactically complex and delays the introduction of its topic (what it is about) longer than its non-hedged counterpart. Therefore, with these additional words, syntactic complexity, and delays of sentence topics, the extra hedges in the high intensity passages could have impeded efficient processing. Apparently, then, if full-clause hedges are to be used to get readers to grapple with and evaluate a controversial text, they should be used lightly. Perhaps the positive results would have been produced with even fewer than the five hedges used in the low intensity condition, especially if hedges accompanied the most controversial material. It is important, as noted above, to remember that the effects of the hedge variables depended to some extent on gender. Another factor to bear in mind about our findings is readers' attitudes toward the particular discipline or content area of the reading. Much in our explanation of the results on the science retention tests rests on the fact that our subjects had quite negative attitudes toward the ideas about evolution. Toward the ideas in the social studies passage our readers were only moderately or weakly negative. That fact would have to be woven into an explanation for why hedges in particular social studies conditions aided students' learning more than the unhedged social studies passage did. And we have yet to explore the effects of hedges on readers' learning of material that they initially have moderately or very positive attitudes toward. Depending on what readers' initial attitudes are toward various kinds of content, the effects of hedges on those readers' attitude changes and learning gains might vary considerably. 4. Some implications and extensions of our findings

Evidence from this study and other hedge studies shows that we can offer some cautious generalizations about the effects of hedges on hearers and readers. In

110

III. Interactive Aspects of Hedging

a study of the effects of hedges in courtrooms, Hosman and Wright (1987) found that witnesses' use of hedges led jurors to see the witnesses as somewhat powerless. The jurors tended to believe witnesses who appeared certain and assertive more than they believed witnesses who appeared unsure of themselves and who hedged their statements; yet Ulmschneider (personal communication) cautions that it all depends on who is using hedges in the courtrooms since, in her observations, hedges create an impression of power when used by judges. In our study, hedges were somewhat effective for ninth graders' positive attitude changes toward and learning of controversial social studies material; and hedges were very effective for these same ninth graders' positive attitude changes toward and learning of controversial science material. On the basis of studies other than this one, we can generalize somewhat, too, about how writers use hedges to make rhetorical moves in their texts. In a study of Charles Darwin's use of metadiscourse in The Origin of Species, Crismore and Farnsworth (1989) found that Darwin locates hedges quite differently in his introductory first chapter than he does in the crucial fourth chapter, in which he sets forth his controversial theory of natural selection. In Chapter One, most of the hedges are in the last half of the chapter while in Chapter Four most are in the first half. Darwin's hedges are situation-specific. In his line of argument, Chapter One presents a different rhetorical situation than does Chapter Four since the first chapter introduces the problem and gives supporting and opposing viewpoints on the origin of species, while the fourth chapter presents the climax of the argument Darwin makes for natural selection. Thus where Darwin probably felt most susceptible to attack, he used the most hedges. Furthermore, studies of texts written by professional social science writers suggest that they use fewer hedges than do professional science writers (Crismore 1989) and that scientists writing for general audiences use fewer hedges than they do for professional audiences (Crismore/Farnsworth 1990). Cultural differences can also be lound in the use of hedges. Markkanen and Schröder (1987), for example, found that authors of German and Finnish philosophical texts used more hedges than authors of such texts in English. According to Clyne (1991), it is conventional for Germans to use hedges extensively in scholarly texts even double or triple hedges. Markkanen and Schröder (1989) suggest that this could be due to their intellectual style of building "big theories" in argumentation. The structure of a language may also have an influence on the use of hedges. Finnish, for example, offers more possibilities for impersonal expressions and thus more possibilities for not accepting responsibility for one's statements than does English (Markkanen/Schröder 1989). Clearly, we cannot understand hedges in texts and their effects without first considering the particular effects on particular readers in particular situations and without considering the interactions of subject matter, readers, readers' beliefs and attitudes, writers, conveyed beliefs and attitudes, and the text itself. Just as there is no general model of reading or writing, there is no general model of hedging. It is dangerous for teachers to teach the notion that hedges weaken student writing. Surely one implication of our hedge study is that giving ninth graders opportunities to see hedges in textbook passages promoted critical thinking and learning.

A. Crismore and W. Kopple, Effects on Attitudes and Learning

111

It is also dangerous for researchers to assume that writers always follow the same rules for hedging and that the results of experimental studies can be interpreted without an examination of how the rhetorical context interacts with the study variables. Using hedges depends on who is doing what with particular subject matters to whom - in other words, the effects of hedges depend on the constraints of the rhetorical situation.

References Barnet, Sylvan and Stubbs, Marcia 1983 Praaical Guide to Writing (4th ed.). Boston: Little Brown. Bell, Robert A., Zahn, Christopher,)., and Hopper, Robert 1984 "Disclaiming: A Test of Two Competing Views", Communication Quarterly, 32, 28-36. Berlyne, Daniel E. 1960 Conflict, Arousal, and Curiosity. New York: McGraw. Brown, Penelope and Levinson, Stephan 1978 "Universals in Language Usage: Politeness Phenomena", in Goody, E. N. (ed.): Questions and Politeness, Strategies in Social Interaction. NY: Cambridge, 56-289. Bruce, Bertram C. 1981 "A Social Interaction Model of Reading", Discourse Processes, 4, 273-311. Chafe, Wallace 1986 "Evidentiality in English Conversation and Academic Writing", in Chafe, W. and Nichols, J. (eds.): Evidentiality: The Linguistic Coding of Epistemology, 261-272. Norwood, NJ: Ablex. Coates, Jennifer and Cameron, Deborah 1989 Women in Their Speech Communities. New York: Longman. Clyne, Michael 1991 "The Sociological Dimension: The Dilemma of the German-speaking Scholar", in Schröder, H. (ed.): Subject-Oriented Texts: Languagesfor Special Purposes and Text Theory, Berlin and New York: de Gruyter, 49-67. Crismore, Avon and Famsworth, Rodney 1990 "Metadiscourse in Popular and Professional Science Discourse," in Nash, W. (ed.): The Writing Scholar: Studies in the Language and Conventions of Academic Discourse, Newberry Park CA: Sage, 118-136. Crismore, Avon and Farnsworth, Rodney 1989 "Mr. Darwin and His Readers: Exploring Interpersonal Metadiscourse and Dimensions of Ethos", Rhetoric Review 8, 91-111. Crismore, Avon 1989 Talking to Readers: Metadiscourse as Rhetorical Act. New York NY: Peter Lang. Darwin, Charles 1967 On the Origin of Species: A Facsimile of the First Edition, with an Introduction by Ernst Mayr. New York: Atheneum. de Beaugrande, Robert A. 1985 Writing Step by Step: Easy Strategies for Writing and Revising. Chicago: Harcourt Brace Jovanovich. Dillon, George L. 1981 Constructing Texts, Elements of a Theory of Composition and Style. Bloomington: Indiana University Press. Dougherty, Barbey N. 1985 Composing Choices for Writers: A Cross-Disciplinary Rhetoric. New York: W. W. Norton.

112

III. Interactive Aspects of Hedging

Fiske, Susau T. and Linville, Patricia W. 1980 "What Does the Schema Concept Buy Us?", Personality and Social Psychology Bulletin 6 (4), 543-557. Flynn, Elizabeth A. and Schweickart, Patrocinio (eds.) 1986 Gender and Readings: Essays on Readers, Texts, and Contexts. The John Hopkins University Press. Halliday, Michael A.K. 1985 An Introduction to Functional Grammar. London: Edward Arnold. Halliday, Michael A.K. 1973 Explorations in the Functions of Language. New York: Elsevier-North Holland. Hariston, Maxine 1981 Successful Writing: A Rhetoricfor Advanced Composition. New York: W. W. Norton. Holmes, Janet 1982 "Expressing Doubt and Uncertainty in English", Regional English Language Center (RELC), 13, 9-28. Holmes, Janet 1984 "Modifying Illocutionary Forces", Journal of Pragmatics 8, 345-365. Homer, Winifred B. 1988 Rhetoric in the Classical Tradition. New York: St. Martin's. Hosman, Lawrence A. and Wright, John W., II 1987 "The Effects of Hedges and Hesitations on Impression Formation in a Simulated Courtroom Context", Western Journal of Speech Communication, 51, 173-188. Hübler, Axel 1983 Understatements and Hedges in English. Amsterdam: John Benjamins. Laib, Nevin K. 1985 "Territoriality in Rhetoric", College English, 47, 579-593. Lakoff, Robin 1975 Language and Women's Place. New York: Harper & Row. Lannon, John M. 1983 The Writing Process, A Concise Rhetoric. (2nd ed.). Boston: Little Brown. Lindgren, James 1982 "Style Matters; A Review Essay on Legal Writing", The Yale Law Journal, 92, 161-187. Lipson, Marjorie X. 1982 "Learning New Information from Text: The Role of Prior Knowledge and Reading Ability" Journal of Reading Behavior, 14,243-261. Luke, Carmen, de Castell, Suzanne and Luke, Allen 1989 "Beyond Criticism: The Authority of the School Textbook", in de Castell, S., Luke, A. and Luke, C. (eds.): Language, Authority, and Criticism, New York: Palmer, 245-260. Mahaney, William E. 1985 Workbook of Current English (3rd ed.). Glenview IL: Scott, Foresman. Markkanen, Raija 1989 "Hedging and Modality", in Jäntti, A. (Hg.): Probleme der Modalität in der Sprachenforschung, Studia Philogica Jyväskyläensia 23. Jyväskylä: University of Jyväskylä, 137-148. Markkanen, Raija and Schröder, Hartmut 1989 "Hedging as a Translation Problem in Scientific Texts", in Laur6n, Ch. and Nordman, M. (eds.): Special Languages: From Humans Thinking to Thinking Machines, Clevedon and Philidelphia: Multilingual Matters, 171-175. Markkanen, Raija and Schröder, Hartmut 1987 "Hedging and its Linguistic Realizations in German, English, and Finnish Philosophical Texts: A Case Study", in Erikoiskielet ja käännösteoria, VAKKISeminar VII, Vaasa, Finland. Martins, Daniel 1882 "Influence of Affect on Comprehension of a Text", Text 2 (1-3), 141-154.

A. Crismore and W. Kopple: Effects on Attitudes an Learning

113

Milward, Celia 1980 Handbook for Writers. New York: Holt, Rinehart, and Winston. Nesbett, R, E. & Ross, L. 1980 Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs NJ: Prentice-Hall. Newcombe, Nora, and Arnkoff, Diane B. 1979 "Effects of Speech Style and Sex of Speaker on Person Perception", Journal of Personality and Social Psychology, 37, 1293-1303. Olson, David R. 1989 "On die Language and Authority of Textbooks", in de Castell, S. Luke, A., and Luke, C. (eds.): Language, Authority and Criticism: Readings on the School Textbook, New York: Palmer, 233-244. Packer, Nancy H. and Timpane, John 1986 Writing Worth Reading, a Practical Guide. New York: St. Martin's Press. Payne, Lucille V. 1975 The Lively Art of Writing (3rd ed.). Chicago: Follett. Perry, William G., Jr. 1970

Forms of Intellectual and Ethical Development. New York: Holt, Rinehart, and

Winston. Preisler, Bent 1986 Linguistic Sex Roles in Conversations: Social Variation in the Expression ofTentativeness in English. Berlin: de Gruyter. Prince, Ellen R 1976 The Syntax and Semantics of Neg-raising, with Evidence from French", Language, 52, 404-426. Prince, Ellen F., Frader, Joel, and Bosk, Charles 1982 "On Hedging in Physician-Physician Discourse", in Di Pietro, R. J. (ed.): Linguistics and the Professions, Norwood NJ: Ablex, 83-97. Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey, and Svartvik, Jan 1985 A Comprehensive Grammar of the English Language. New York: Longman. Rawlins.John P. 1980 "What's So Wrong with 'In My Opinion'?", College English, 41, 670-674. Ruggiero, Vincent R. 1981 The Art of Writing. Sherman Oaks CA: Alfred Publishing. Ruthrof, Hans 1981 The Reader's Construction of Narrative. London: Routledge and Kegan, Paul. Schommer, Marlene 1990 "Effects of Beliefs about the Nature of Knowledge on Comprehension", Journal of Educational Psychology, 82(3), 498-504. Singer, Harry 1986 "Friendly Texts: Description and Criteria,", in Dishner, E.K., Bean, T.W. Readence, J.E. and Moore, D.W. (eds.): Reading in the Content Area: Improving Classroom Instruction, 2nd ed., Dubuque LA: Kendall/Hunt, 112-128. Strunk, William and White, E.B. 1959 The Elements of Style. New York: Macmillan. Tyler, Sherman and Voss, James F. 1982 "Attitude and Knowledge Effects in Prose Processing", Journal of Verbal Learning and Verbal Behavior, 21, 424-538. van Dijk, Teun 1982a "Introduction", Text 2 (1-3), 1-5. van Dijk, Teun 1982b "Opinions and Attitudes in Discourse Comprehension," in Le Ney, J.F. and Kintsch, W. (eds.): Language and Comprehension, New York: North-Holland, 35-51. Watzlawick, Paul, Beavin, Janet H., and Jackson, Dawn D. 1967 The Pragmatics of Human Communication: A Study of International PATTERNS, Pathologies, and Paradoxes. New York: Norton.

114

III. Interactive Aspects of Hedging

Woodman, Lenora, and Adler, Thomas P. 1985 The Writer's Choices. Glenview IL: Scott, Foresman. Young, Benjamin 1990 "Post Critical Fables and Post-Modem Criticism (A Review Essay)", Discourse, 10, 133-139. Zinnser, William 1976 On Writing Well. New York: Harper and Row.

ANNA MAURANEN

Hedging in Language Revisers' Hands

/. Hedges as an interculturalproblem in academic texts

Academic discourse is a world where observations suggest that something might be the case, where states of affairs appear ίο hold, where it seems reasonable, to suggest, and where we might infer; in other words, it is a world of uncertainties, indirectness, and non-finality - in brief, a world where it is natural to cultivate hedges. Hedging is generally taken to mean those expressions in language which make messages indeterminate, that is, they convey inexactitude, or in one way or another mitigate or reduce the strength of the assertions that speakers or writers make. Typical expressions are, like those above, such that make assertions less categorical and direct, and are thereby regarded as less certain, as well as more polite than would be their unmodified, direct counterparts. On this view, saying (la) below is more polite and less certain than saying (Ib): (la) (Ib)

The conclusion appears to be that - within the present sample - there is no unitary "smoking culture" among the girls... The conclusion is that there is no unitary "smoking culture" among the girls...

Since LakofFs (1972) seminal paper, the literature seems to focus on two broad main areas of hedge use: first, issues concerning the certainty of propositions, and secondly, issues to do with the interactive, or interpersonal aspects of language, notably the question of politeness. Because the notions of certainty and uncertainty occupy a central position in the characterisations of hedging, modal expressions have received considerable attention in the discussion. Hedging and modality are interrelated concepts, with either one or the other subsuming the other in different models. For example Simpson (1990) presents hedging as one function that modal expressions can have, whereas Markkanen/Schr der (1989) take hedging to be the wider concept. In this study, hedging and modality are treated as overlapping domains of language use, with neither covering the other completely. LakofFs original notion of hedging as increasing or decreasing 'fuzziness' in the use of concepts is also close to vagueness. Vagueness has been analysed in depth by Channell (1990, 1994), and it also occupies a prominent place in Zuck and Zuck's (1985) study of hedges in newspaper texts. Vagueness includes among other things expressions denoting impreciseness (about, sort of), which

116

. Interactive Aspects of Hedging

relate to certainty in that they leave uncertain the exact quantity, quality, identity etc. of the object of discussion. Altogether, there is a large and varied number of linguistic expression categories associated with the notion of hedging. Nevertheless, exhaustive and clear-cut lists of hedging expressions would probably be impossible to devise, since there is no simple correlation between different functions of hedging and those of linguistic categories as defined in terms of the linguistic system. Hedging is characteristically context-sensitive: certain kinds of expressions seem to possess a potential, as it were, for taking on a hedging function in some contextual environments. Although hedging is an integral part of academic discourse, there is no general consensus among scholars of academic discourse concerning its use and functions. The issue has been investigated in for instance the field of language for specific purposes, but it has been much less popular than such topics as text structures, terminology, or thematic choices. The currently dominant way of conceptualising hedging in academic discourse appears to be in terms of interpersonal strategies, above all those to do with politeness (see for example Myers 1989, Markkanen/Schröder 1989, Simpson 1990, Luukka 1992, Valle 1993). These analyses draw on Brown and Levinson's (1978) notions of politeness and face, and some of them focus largely on expressions of modality (e.g. Myers 1989, Simpson 1990). Brown/Levinson's theory of politeness rests essentially on the notion of face, which includes an individual's postulated need to keep up both a positive and a negative face, that is, to be approved interpersonally and to be unimpeded personally. Extended to academic discourse, actors' face-saving strategies have been viewed from a two-sided perspective: saving the opponent's face and saving one's own. The latter aspect, the need to save one's own face is seen as the primary motive for scientific hedging by Markkanen/Schröder (1989), while for instance Myers (1989) does not differentiate these in terms of importance. Brown/Levinson's notion of politeness, together with other similar early developments in pragmatic politeness theory, has recently been criticised both for postulating a super-rational actor making strategically appropriate choices (see for example Janney/Arndt 1993), and also for not taking cultural variation sufficiently into account (for example Ide 1988, 1989). In connection with academic writing, the former criticism is particularly relevant, and it seems that there are also other grounds for being more cautious about subsuming all motivation for hedging under the principle of politeness. This is not to say that politeness is unimportant in academic discourse. It merely means that even in a small sample of academic texts hedging appears in roles where politeness need not be invoked. Hedging has not only presented a theoretical problem for modelling and describing language use, but also a practical one in for example translation. The question of translating hedges has been the particular concern of Markkanen/Schröder (1989), who point out that since hedges are realised by different means in different languages, and since writers also seem to have individual stylistic preferences with respect to hedges, their translation is not a simple matter. Markkanen/Schröder (1989, 174) emphasise that it is imperative for

A. Mauranen, Hedging in Language Revisers' Hands

117

the translator to understand writer intentions in order to translate correctly. A correct translation in their view means that the effect of the translated text is equivalent to that of the original. Given this goal, cultural variation in hedging and related phenomena such as modality and expressions of politeness in general seems to demand a special sensitivity to the communicative values of expressions from a translator. Markkanen/Schröder found in their study that bilingual or trilingual writers who translated their own texts made cultural adjustments to their hedges. They then raise the question whether such adjustments are possible in a "normal" translation situation, when the translator cannot of course know the intentions of the original writer. In a similar vein, Tabakowska (1989) discusses the difficulty of translating epistemic modals. She approaches the problem from the angle of attitudinal equivalence in translation, and emphasises the need to take into account such factors as the speaker's reliance on or appeal to other speakers, and the speaker's degree of involvement, in addition to the more commonly discussed degree of confidence in propositions. Translation problems are in many ways similar to those encountered in the practice known as language revision, where native speakers of a language revise texts by non-native writers. This is now becoming an increasingly common way of utilising native speakers' expertise in improving the linguistic quality of nonnative texts. For instance in Finnish universities language revision is quickly replacing the translation of academic papers, particularly in languages that are generally well known in the country, such as English. An earlier study on non-native academic writers' texts and the linguistic revision of their texts by native-English revisers (Ventola/Mauranen 1990) indicated that revisers made very few corrections in modality expressions, even though the non-native writers did not appear to use these expressions in a native-like manner. For one thing, the writers employed fewer modal expressions than comparable native speakers, and in addition, they used little variety in these expressions. The results thus suggest that the non-native writers might have needed help with their modal expressions, but that the revisers did not appear to meet this need. However, the study made no attempt to go into the reasons for revisers' seeming unhelpfulness, since it was a textlinguistic analysis. Nor did it include a variety of hedges, but described modal verbs and adverbs only. An interesting question therefore remains to be answered with respect to language revisers' treatment of hedges: does correction avoidance apply to hedges at large, and if so, what reasons do revisers have for doing this? This paper purports to give a voice to language revisers, and to explore the notion of hedging from their point of view. It reports a small study, which looks at three academic research reports and the way in which language revisers treat hedging in them. The texts were selected from different disciplines, and are written in English by non-native academics. A survey of hedging expressions and their revisions was followed by an interview with the revisers, asking them to discuss their solutions with respect to correcting some hedges, and to explain their attitudes towards the revision of hedging in scientific writing: what problems are involved, if any, and what would be a good end result?

118

III. Interactive Aspects of Hedging

The next section describes the texts and the interviews a little more closely. In section 3, the main hedge types found in the texts will be presented briefly. Section 4 presents revisers' comments about the processes and problems of revising hedging in non-native academics' texts. Section 5 discusses some examples of actual revisions, and Section 6 is devoted to summarising and discussing the revisers' observations about the nature of hedging. 2. Investigating hedge revisions Three papers were selected from among around forty texts which had been sent in for language revision to the Language Centre of Helsinki University, and whose reviser comments were available for investigation. They represented approximately the average English skill level of the research writer group, as judged by the number of corrections made by the revisers in all the texts. The texts were selected from fields which usually produce a good number of papers for revision, and which at the same time were different from each other. Because these discipline areas tended to be prolific, it was possible to find revisers who had considerable previous experience with texts in the same fields. The choice then fell on three papers, one natural science and two social science texts, which used different research approaches: one was empirical, one theoretical, and one speculative-prescriptive. The natural science text (HL) was a biochemical paper reporting basic empirical research. One of the social science papers (ML) was a theoretically oriented text assessing an economic model, and the other (PK) an educational text, which was written to be spoken at a conference, and which was neither empirical nor theoretical, but speculative and somewhat normative. The revisers were all native speakers of English, as well as qualified and experienced teachers at the Language Centre. In addition, they all had a long experience as revisers, and were well acquainted with revision in these particular fields. The revisers were contacted about a week before the interview, and given a copy of the paper to be discussed. They had revised the papers about two years earlier, and it was therefore felt that they should be given a chance to go over the paper at their leisure. The copies included their original revisions and comments. When the individual interview appointments were made, the revisers were also informed about the topic and nature of the interview, in order to give them a chance to prepare for the interview if they so wished. The actual interview was semi-structured and relatively informal. Each reviser was asked general questions about how they felt about revising hedging, whether they found it difficult, whether they would like to improve the revision system or the revision process as regards hedging. They were also asked to comment specifically on the hedging habits of the writer in question, and their own corrections or the lack of it in given instances. In addition to the questions in common, the interview was a recorded dialogue between the revisers and the interviewer. The revisers were thus able to initiate topics they felt were relevant or important, like for example their own experiences as writers, even if the interviewer had not asked questions about them.

A. Mauranen, Hedging in Language Revisers' Hands

119

3. Hedging in the texts

To provide a background to the revisers' comments, this section will briefly look at the kinds of hedges which appeared in the papers that were discussed. Hedges were taken to be expressions which in their context reduced the certainty or precision of propositions. Thus, by employing hedges, a writer refrained from making straightforward assertions concerning states of affairs, but instead qualified these by either indicating that the propositions in the text are uncertain (e.g. may, seem, tend, perhaps), vague (e.g. around, about, some kind of), or of limited applicability (in some respects, at least], or that the writer is not committed to the proposition (// may be true thai, One could say thai). In the identification of hedges, the textual context plays a crucial role. As pointed out earlier, expressions which are typically used as hedges, have also other uses, and their potential for acting as hedges is only realised in some contexts, in interaction with other linguistic features. For instance modal verbs have other readings, and so do expressions such as in principle, in general, etc. Thus, the expression in general in example 2 may or may not be a hedge: (2)

The academic qualifications are in general high. (PK)

If the context is as in 3, there is no hedging effect, but the statement is read as a generalisation which acts as a basis for the conclusion in the next sentence. (3)

The academic qualifications are in general high. [...] thus the marks in die matriculation examination and the GPA must be high. (PK)

If, however, the expression is read in the context of (4), it imparts a hedging effect, indicating that the generalisation is of limited validity: (4)

The academic qualifications are in general high. [...] There is, however, a clear difference between male and female applicants. It has been much easier for male applicants to get into the university. (PK)

The context of each potential hedge has thus been utilised in the identification of hedges in the present study as well, although a systematic enquiry into the relations of contextual features and the hedging expressions is well beyond the scope of the study. The epistemic and interpersonal uses of hedges are here distinguished from one another in principle, even though some linguists prefer to look for unity rather than division between the epistemic and the interpersonal. For example Halliday (1985) treats modality as part of the interpersonal function of language. Similarly, Simpson (1990; 1993) subsumes modality under the interpersonal function of language in his studies of academic discourse. He defines modality as the speaker's attitude towards the truth of a proposition expressed by a sentence and the situation or event described in that sentence. This definition is clearly epistemic, and this epistemic use is then connected to the politeness function: "In addition to the speaker's uncertainty, modalized utterances are motivated pragmatically by tentativeness, tact and distance - in short, considerations of politeness" (Simpson 1990, 73). However, even though the epistemic and politeness uses of hedges may be related, there are also grounds for distinguishing them. Particularly in the aca-

120

III. Interactive Aspects of Hedging

demic context, it seems that there are several reasons for conveying epistemic uncertainty. There simply are many things that a scientist or scholar does not or even cannot know; for instance, even if we are able to isolate some plausible causal relationships, most people particularly in the humanities or social sciences would be reluctant to say that these exhaust the possible influencing factors. This would go against currently accepted epistemological beliefs in these fields. Thus, the hedging verb seems in the following example (5) from the economics text, is interpretable as a comment on the epistemic status of the inference. It would seem far-fetched to seek face-saving or self-protective motives for it, since the statements around this sentence are quite unhedged and straightforward, and it is not in fact the hedged sentence that is the writer's main point, but rather the last sentence. (5)

The arguments for the condition ews < 0 are questionable. The case ewsX) seems the more plausible alternative. However, the main analysis showed that, in the time period [O,T], both ewsX) and ews = 0 are present, although the latter case is expected to take place close to the terminal time. (ML)

Similarly, the writers of the science text in the sample were exploring little known territory, and used hedges to convey the speculative nature of their findings and explanations (6). The hedges indicate which statements, or which parts of the knowledge the writers themselves regard as uncertain: (6)

This indicates that the excimer monitored at 470 nm is kinetically distinguishable from the monomeric species monitored at 377 nm. More probably the precursor of excimer E2, centered at 470 nm is around 420nm emitting intermediate state M2, which is formed relatively quickly [...]. This state, which is also observed in timeresolved spectra at 235 K and 114 K,but with lower intensities [...], may be a monomeric conformer, which by a single hindered diffusional rotation can lead to excimer. (HL)

Although it is not unreasonable to maintain that part of the motive for hedging in such contexts is to shield the writers against counterevidence and more successful explanations, the epistemic concern is nevertheless much more prominent than in cases like 7 below. Here the hedge perhaps does not reflect the status of the writer's knowledge of the matter, but is a part of his argumentative strategy. The writer is arguing for the reverse position to the one referred to in the extract; he is advocating the view that a teacher's professional thinking and decision-making processes are important. At this stage he is making a concession to the counterposition. The concession is hedged, which takes off some of its force, and probably also renders it a lower status in terms of importance and interest. (7)

In that position it perhaps is not important how the teacher because he does not have much to decide. (PK)

thinks and decides

The writer's hedge in this case thus seems to be part of a text strategy which anticipates readers' reactions, shows awareness of other opinions, and adapts the message accordingly - that is, above all an interpersonal strategy. It seems, then, that hedges can be used for more epistemic and more interpersonal reasons, and that it is therefore useful to maintain this distinction in their analysis. In practice not all instances can be comfortably assigned to one category without strong overtones from the other, but the fact that some

A. Mauranen, Hedging in Language Revisers' Hands

121

expressions can be multifunctional does not invalidate the postulation of different functions if there are other grounds for them. The distinction between primarily epistemic and primarily interpersonal uses was thus applied in this study, and will be maintained in the brief discussion of hedges in the sample texts below (sections 3.1. and 3.2). In addition, the discussion contains a further section illustrating the clustering of hedges in text (section 3.3). 3.1. Epistemic use of hedges In the texts at hand, expressions of epistemic uncertainty were quite common. Typical instances were modal expressions by which the writer indicated some reserve concerning the certainty of the proposition, as in example 8: (8)

This might indicate that the X*-like transient is a precursor of an excimer. (HL)

Sometimes the hedge also appeared to lend an air of unimportance to the statement. In 9, for instance, seems appears in a sentence which contributes a minor supportive element to the main argument, which continues in the next sentence. The hedge in such a context appears to enhance the impression that the second sentence is a passing remark, whose precise truth value is not only uncertain but also of little consequence. The hedge perhaps in the final sentence again serves to indicate the speculative status of the writer's conclusions. (9)

Based on this fact, one can reach certain conclusions. The situation also seems to be the same in other countries. Perhaps it is related to what Lortie describes as conservatism [...] (PK)

On the borderlines of hedging were modal expressions referring to possibilities which obtain in the text-external actual world. In these, the uncertainty is projected away from the speaker, onto the states of affairs themselves, as in example 10: (10)

In these model alternatives, if the penalty parameters g and p are big enough, switching to a non-supervision control regime may take place very early, or may not take place at all. (ML)

These expressions do not always play a very strongly hedging role, but seem rather to be very close to non-hedging statements, like example 3, concerning what is possible in the actual world: (11)

LB films can be prepared as mono- or multimolecular layers containing chromophoric molecules. (HL)

A number of expressions of vagueness were also found in the texts, such as example 12. They can be subsumed under a broad notion of epistemic hedging since they convey a meaning which is uncertain with respect to exact facts or figures. The latter hedge in 12 is vague about the identity or type of the process described. (12)

The E2 emission around SOOnm can be regarded as an emission of an excimer type transient formed from molecules by some kind of ground-state interaction [...] (HL)

Here the first expression of impreciseness gives rise to the interpretation that a more exact figure is unnecessary or unimportant, while the second hedge (some

122

ΠΙ. Interactive Aspects of Hedging

kind) is indefinite rather than inexact. It refers to an unknown mechanism, and conveys an attitude of speculation. Even if from the writer's point of view such vagueness may be a matter of deliberate choice, for example in order to background some information (cf. Channell 1990, Myers 1994), for the reader it is always a matter of uncertainty in that he or she is left uncertain about precise quantities or qualities. What the uses of hedges in epistemic roles seem to have in common is a kind of openness. They convey a sense of suggesting one way of interpreting a state of affairs or one way of looking at it, or one possible truth, while at the same time leaving space for other possibilities. These other possibilities, like the statements on the chosen alternatives, may concern states of the actual world, or states in terms of theoretical models, or in the interpreter's mind. The point is that the world that is looked at is rendered as potentially containing other possibilities. 3.2. Interpersonal use of hedges With some hedges, interpersonal considerations seem to override any epistemic motivations that hedges may also have. In these cases, it seems natural to speak of hedges as strategic devices for personal protection. In many cases the primary object of protection seems to be the writer (or the writer's face, in the currently most widely used terminology, as was already pointed out in the introduction). Example 13 is an instance of interpersonal hedging. (13)

One could say that this kind of skill - if it is a skill any more - is a second order concept which directs the teacher's actions and which is totally dependent his way of thinking.(PK)

Here the writer's certainty about the propositional content of what he is asserting is not really at issue. The writer is making a suggestion (that a particular skill should be understood as a second-order concept) in a tentative manner, possibly to protect himself from refutations or counterclaims. Whether this tentativeness is there to protect the writer from an attack or to show consideration to somebody else who treats the question very differently, it is in any case clearly a matter of interpersonal relations and tact in a way which is quite different from the epistemic examples above. It is in cases like this that the hedge analyses based on politeness models seem most relevant. The second hedge in example 13 (if it is a skill any more) indicates the writer's detachment from the concept that lie is using. This seems also primarily an interpersonally motivated hedge, since it indicates the writer's stand, not his epistemic belief, with regard to this concept. 3.3. Clustering of hedges Hedges play a number of interesting roles in text, and the co-text is in turn relevant for the interpretation of hedges. This interplay offers an interesting area for research in itself, but there is space in this section only to take up one aspect of hedges in context. This is the tendency of hedges to cluster in certain places

A. Mauranen, Hedging in Language Revisers' Hands

123

in the text more than in others. Two examples are taken up here for illustration. The first example (14) illustrates the clustering of hedges within the scope of one sentence. (14)

/ think that Finland is a special case in many respects when we talk about teacher education, although there naturally are many common features and details in teacher education in general. (PK)

Here the sentence begins by hedging the neustic (see e.g. Lyons 1977), or a reporting clause of a subjective explicit type (Halliday 1985, 333). The vagueness of in many respects both further reduces the strength of the assertion of Finland's status as a special case and also backgrounds the exact respects in which this is or is not the case. The concessive clause in turn hedges the entire proposition again. Longer passages also show clustering of hedges, for example in sequences which are speculative or deal with uncertain or sensitive issues. A passage illustrative of a sensitive question is presented below (example 15), where the writer is trying to say something that his audience both wants and does not want to hear: it is flattering to them personally, but unflattering professionally. The text appears to have been written so as to be spoken, and the writer is addressing an audience which consists of educationists, many of whom are former schoolteachers. He discusses issues of teacher education, and at this stage he is addressing the problem that a certain number of student teachers never take up a teaching career after graduation. The explanation suggested is that it is the brightest students who do not take jobs in schools. Since both the speaker and most of the audience belong to this group, this is flattering to them, yet not entirely appropriate to articulate in public. It also points to a potentially serious mismatch between teacher education and the profession it is preparing students for: successful programmes would naturally have the best students excel within the profession, not outside it. The writer begins by attributing the problematic facts to another author: "Kerr [...] says quite frankly: [...] of those who do take teaching positions, the best leave the classroom [...] We can reasonably expect only the numb and the dull to linger in teaching careers [...]" After presenting these inconveniently frank words as quotations, the writer goes on as follows: (15)

This is a very serious statement and / hope that we are not yet at that stage. But we must admit that there is some truth in it. I suppose that there are many former teachers among the participants of this Congress, including myself. Why have we not remained in the schools? Certainly there are many reasons and / dare not claim that we in particular would be the best and the most intelligent, as Ken perhaps would say. Most oflen we can find two main reasons for leaving: [...] (PK)

The writer hedges both the claim (I hope that we are notyet at that stage, Certainly there are many reasons, and I dare not claim) and the counterclaim (But we must admit that there is some truth in it), and attributes the difficult viewpoint once more to the other author, although hedgingly (as Ken perhaps would say). He then goes on to suggest reasons for leaving the profession which are not directly to do with higher ability or intelligence, and after the initial vagueness (most often, main) no more hedging follows. The linguistic devices used to make this

124

III. Interactive Aspects of Hedging

viewpoint palatable are of various kinds, and illustrate the point made earlier that there are no hard-and-fast correlations between form and function among hedges. What is particularly clear here is that the hedging is not limited to single words, particles or phrases which modify individual utterances or elements within them. The expressions which can be regarded as hedging in the three papers ranged from individual words modifying other individual words to whole clauses or even longer text sequences which modified clauses or sequences of clauses. There was no obvious connection between the size or syntactic type of the expression and its expressive function in the text, although some tendencies were observable. Thus words like about, around, and some tended to modify individual sentence elements, while whole clauses or sentences, typically concessive clauses, tended to hedge other clauses or longer passages. 4. The practice of revising hedging 4.1. The revision process

Each of the revisers emphasised that hedging is above all a matter of the writer's choice. It is up to the writer to decide the level of certainty he or she believes is appropriate, and the language reviser should only interfere with this decision if the outcome is linguistically unnatural or awkward. On the one hand the judgement of the right level of certainty and commitment was seen as being the writer's right, so that interference from an outsider would be quite inappropriate and unwelcome; on the other hand it was seen as the writer's responsibility, which no outside agent should attempt or need to take on. As one of the revisers put it: "most of the time we have to assume that the people who are writing these papers in English are capable of putting forward what they want to say". Thus the revisers' answer to the question of whether the translator or reviser 'correctly' represents the original writer's intentions is that what they see in the text is what they assume was intended. However, there was one exception to this general line. One of the revisers remarked at one point that in his experience those who have the best command of English use the most hedges. If this is so, it means that there is indeed a problem of language proficiency, and this would constitute a counterclaim to the basic assumption repeatedly made by each of the revisers that the writers knew what they were doing. However, since no systematic evidence is available on the relationship of proficiency and hedging at this skill level, the matter remains unresolved without further research. Two of the three revisers thought hedging is one of the most difficult areas in revision, whereas one said it is not particularly difficult. Despite this difference of opinion, all three were highly unanimous as regards the requisite skills in revising hedging. Above all, the native speaker's intuition about what is natural in the language seemed to be the chief guide in language revision. It is clear, however, that for these three revisers, the intuition had also greatly benefited from their long experience with academic texts. Another thing they all emphasised was familiarity with the field, which was seen to constitute an

A. Mauranen, Hedging in Language Revisers' Hands

125

important prerequisite for successful revision. The necessary familiarity had been achieved by these revisers either by getting well acquainted with an individual writer's work, or by knowledge in the field more generally. In addition to these major points of agreement, the revisers also had different individual interpretations of the issues involved. One reviser suggested that a major source of difficulty was the enormous text-external knowledge of the field that was necessary. He thought that any given paper was related to several others that had preceded it in the field, and felt that in order to make a judgement of the appropriate level of hedging, he would have to read a large number if not all of those papers. This line of reasoning is clearly reminiscent of the idea of intertextuality. Academic papers can indeed be conceptualised as if they were elements in a long chain or a large network of papers, in the context of other similar texts. They can be positioned within layers of contexts in fact, from the broad context of disciplinary tradition to the narrow context of research and writing around a specific question. In some sense the choice of an appropriate level of hedging is a function of all these other texts. However, this is more true of an idealised theoretical world of texts than the actual world of writing and revision. It is unlikely that writers themselves consider all their hedge choices in the light of all potentially relevant other texts in the field. Another source of difficulty that was implicated by one reviser was that of knowing the writer's intentions. Since it is obviously impossible for practical reasons to stop and ask at every hedge what the writer wanted to say (let alone in the cases where he or she is not hedging, but might have been), the match of the end product with the intentions regarding the text cannot be achieved. This view relates to the translation problem brought up by Markkanen/ Schröder (1989); if the original writer's intentions are inaccessible, can a 'correct' translation be achieved, or a 'correct' revision? The question of being faithful to original intentions obviously provides a very different angle to the authenticity of a revision or a translation from the intertextuality problem. Both throw light on some of the factors that have to be considered in translating or revising hedging. Yet somehow they both seem to imply a view of hedge choices as if it were a question of putting together a puzzle where all the pieces fit together. It is unlikely that such an exact solution is feasible or even desirable; even if it was possible to ask the writer what his or her intentions were, or to read a large number of the papers relevant to a given text, it would hardly be possible to shed all indeterminacy from the choices. Writers are unlikely to have precise and clearly formulated intentions, and there is no logical end to the texts that may be relevant to a given text. What matters in practice is how appropriate the final text version is for its purpose. To achieve an optimal solution in this respect, the writer and the translator should join forces and negotiate the hedge choices by bringing in their respective kinds of expertise to the process. The revisers made some suggestions that might improve current revision practices, such as creating more favourable conditions for writer-reviser interaction: revisers and writers should have more time for individual consultation, and in general more opportunities for collaboration on individual texts. They also spoke warmly about their experiences of working with writers over a longer

126

III. Interactive Aspects of Hedging

period, which made it possible to get to know the writers' work and intentions better, and to cooperate in developing the texts stage by stage. For the most part, though, the revisers were not very keen to suggest radical changes to the ways in which revision is carried out. This is of course natural in view of the fact that they placed hedging firmly within the writer's domain. If the choices rest with the writer, it is primarily the writers who should be made more sensitive to the effects of hedging and the available possibilities of expression. 4.2. Revisers as writers

Two of the revisers probed their own experiences as writers of academic texts. Both had relatively recently been writing an MA thesis, and both reported receiving comments to the effect that they should be more definite, that is, to hedge less. Their reactions to such comments were different. One implicated lack of knowledge: he explained that he had hedged because there was no way he could have known enough about the issues in the field to warrant full certainty. This in his opinion did not mean that he did not know where he was going, but rather that he felt there was so much written and going on in his field that he could not keep up with it all, and therefore he wanted to make hedged statements. The other reviser, in contrast, had felt pleased about the comments encouraging her to hedge less, because she felt she was not a very self-confident person, and the advice to be more certain about what she was saying encouraged more confidence in her own work. These two different reactions again seem to reflect the two sides of uncertainty that have here been associated with hedging: on the one hand, epistemic uncertainty, that is, uncertainty of knowledge, and on the other hand interpersonal, or, in fact, personal, uncertainty, or insecurity. The first reviser-as-writer related his hedging to his relative lack of knowledge, while the second related this to her uncertainty as a person. What they seemed to have in common, though, was the role of a novice in the field. It would seem natural that novices are less certain of the field and themselves than experts, and the third language reviser in fact said that she had observed in her work that the novice-expert distinction is relevant in accounting for the amount of hedging in a paper.

5. Hedges revised

Few of the hedges that the writers had used had been touched by the revisers. It is therefore not surprising that when they were asked whether they felt that the writer was using hedges appropriately, they tended to give affirmative answers, and explain or justify the writer's hedging rather than question or criticise it. Example 16.0 is from the conclusions section of a text, and one of the most hedged sentences in the whole text (the Ο stands for the original version in revision examples, and R for the revised version):

A. Mauranen, Hedging in Language Revisers' Hands (16.O)

127

The formation of the dimeric species might be a preceding step before the formation of an immediate precursor of the excimer. (HL)

The reviser had corrected the syntax a little (the ing-form into a relative clause), but not the hedge: (16.R)

The formation of the dimeric species might be a step which precedes the formation of an excimer.

On being asked to comment on this hedge, the reviser said that it is there because "we're not too sure how the dimeric species is formed, there's a lot of uncertainty of how we get to dimeric species". The reviser thus accepted the writer's choice and attributed a motivation to it. In a similar fashion, in cases where writers had made straightforward claims that were neither hedged nor corrected, the revisers tended to find good reasons for the writers to write as they did. This can be seen in remarks like the following, where the reviser is commenting a point in the text which reads 'the increase of absorbances is evident', and says "we have a curve and you can see it actually, it's there, you can say it because it is increasing, because it is rising". On the whole, then, the revisers seemed to accept the writers' hedge choices. Those corrections that the revisers had made by for example adding modal verbs to the text were mostly explained by referring to such reasons as 'style', 'naturalness', or 'idiomaticity', rather than to the reviser's feeling that the writer had been either too direct or too hedgy. On some occasions, though, both stylistic reasons and the surmised writer intentions were implicated for the same change, as in the following: (17.O) (17R)

I think that maybe he is not wrong. (PK) I think that perhaps he is not wrong.

The reviser commented the change like this: "partially it's style. I think that he was more sure that he was not wrong and the maybe makes it less sure than perhaps*. For the very few corrections that the revisers did make, the clearest motive appeared to be that the reviser had felt the writer was trying to say something that he was not able to convey through the linguistic means that he had chosen, as in the above example. Occasionally the reviser was not very sure of having interpreted the intended meaning correctly, as here: (18.O)

(18.R)

When the supervision dimension is added to an efficiency wage model, // is rational to firm to use the supervision control only when the supervision control is effective one, when die effort gains from die wage controls are uncertain and die firm's wage policies are constrained .(ML) When die supervision dimension is added to an efficiency wage model, it is reasonable to assume that die firm uses supervision control only when this is effective, when the effort gains from the wage controls are uncertain and when corporate wage policies are constrained.

Commenting on her correction, the reviser said: "this was because of the linguistic problem that I wasn't sure what he meant really so I was only hoping that I'd sort it out from his language, what he was actually trying to say and, the use of the word rational, I couldn't think where he got rational from anyway". The fact that the original version attributed rationality to the model firm

128

ΠΙ. Interactive Aspects of Hedging

whose behaviour was being discussed, but the revision moved the rationality or reasonableness to the writer, was acknowledged as a problem: "it's the assumption that is reasonable not the firm in what I'm saying ". However, the problem was resolved in practical terms because the original writer had not questioned the revision: "I'm saying it's the assumption. But he didn't come back on it. They very rarely do". In several cases the revisers found it clearly difficult to explain why exactly they had made the correction. This probably relates to their comments that they do not want to make changes if the text "reads well". The revisers said that they did not consciously look for hedges but reacted only if something felt linguistically awkward or not quite natural in style.

6. How revisers see hedging In addition to discussing revision work and the processes and problems involved, the revisers also probed their understanding of the nature of hedging in academic papers, and writers' reasons for using such expressions. The interviews brought to the fore two major areas in the need to hedge: the nature and uncertainty of knowledge, and the interpersonal side of shielding oneself or creating a certain impression in the audience. Questions of knowledge were associated both with the nature of knowledge itself, and the changing nature of the objects of study and discussion. Because the first kind of motivation follows from the way knowledge is, it is seen as necessary, and the writer is taken to have no choice but to hedge, because 'nothing is certain', 'nothing is black and white'. This understanding is obviously basically epistemic. The uncertainty of knowledge was also associated with constant change, such as the fact that new results may undermine earlier conclusions and theories, and in fact are almost certain to do so sooner or later. Awareness of this fact of a scientist's life was implicated in the need to hedge conclusions, for example. Another aspect of change was the changing nature of the object of study, which was implicated in educational research: the field of education itself changes relatively fast, and the analysis of today's situation may be dated in a few year's time. Therefore it is wise to hedge your assertions, and in this way give them a longer life. The other type of motivation was more a matter of using deliberate strategies. The writer was assumed to hedge in order to impart a certain effect on his or her audience, as is evident from comments like "he is using modesty" and "he is using hedging to get the sympathy of the audience", or "he's suggesting that even though it's suggested-we don't believe it". Even though the strategic, interpersonal motivation for hedging seems to contain a much stronger element of choice than the epistemic motivation, the writers were still seen as having very good reasons for hedging: "[...] he's creating this level, instead of doing it directly; the danger is that if he doesn't hedge it may seem as if I know it all and I'm going to tell you what it all is, what it's

A. Mauranen, Hedging in Language Revisers' Hands

129

all about". The strategic motive also appears to carry an implication of insincerity; the writer operates on two levels at the same time. One level is what he or she 'really' knows or wants to achieve, the other is a deliberate use of certain linguistic devices for reaching this goal. There are thus two meanings involved, one of which is the writer's own, and the other that which he or she wants the audience to read into the discourse. The following interview extracts illustrate how the reviser reads out extracts of the text and comments on them: 7 suppose there are many former teachers among the partiapants of this Congress' "he knows damn well that there are" 7 dare not claim that we in particular would be' - "it really means I think we are"

Despite reading such double-speak into the writer's meanings, the revisers apparently had no difficulty interpreting both of the levels simultaneously. One of the revisers described a similar attitude with respect to his own hedging as a writer: "I hedge a lot but I know where I'm going to, if you look between the hedges, you find I've got a very strong point of view, if the reader cares to look carefully he finds that I think this way [...]". If readers and writers know of these two levels and are to read them appropriately, the question arises of why is such hedging necessary? If the interpersonal motivation of hedging lies in creating an effect of politeness and modesty, but at the same time the strategy is transparent and the opposite or the more direct and less polite intention is also readable into the text, it seems that hedging is either a waste of effort, or it is part of the language game that we play and accept. One possibility is that the use of hedging in these contexts has more of a ritual function. On this view, the politeness or modesty interpretations would be basically subordinate to the more clearly social function of hedging in academic contexts, that is, indicating social affiliation and in-group status. However, there appear to be at least two kinds of social motivations for fulfilling appropriacy expectations in academic papers: one is the micro-level tactfulness that Myers (1989) perceptively describes, and which is necessary for protecting personal investment and individuals' faces. The other motivation appears more at the level of groups and communities: certain established, ritual ways of writing are incorporated into the discourses of the community, and playing by those rules of expression shows group affiliation and insider status, while it also serves in a gatekeeping function, i.e. distinguishes between insiders and outsiders. These more macro level motivations are closer to the observations concerning the use of modal expressions by Fairclough (1992). All revisers took up the question of interdisciplinary differences, which they believed to be considerable. Each compared two disciplines, but different ones: hedging was estimated to be relatively more common in education than law, more common in psychology than in economics, and more common in experimental sciences than in mathematics. The revisers related the assumed relative frequency of hedging to the nature of knowledge in these fields: the more the field was perceived as being in a constant state of change, the more hedging it was supposed to need. In contrast, the lack of hedging was associated with fields which were assumed to change little, such as law and mathematics, as well as economics which rests on formulae and proofs.

130

III. Interactive Aspects of Hedging

An interesting difference of opinion related to disciplinary differences came up in connection with two different hedging strategies used by the writers. One of the writers (PK) employed plenty of hedging all through his paper until it was time to draw conclusions. These were not hedged, and therefore stood out from the rest of the text. Another writer (HL) used very clearly the reverse strategy, where quite a few hedges accompanied the conclusions, even though the experimental methods and results were reported in a straightforward manner. When the revisers of these two papers were asked to comment on the writer's strategy in each case, they made comments which in effect justified the writer strategy that was used in the text they had revised. The writer who used hedges relatively frequently throughout the paper except for the conclusions was an educationist. The reviser felt that the writer was doing the right thing, that it was natural not to hedge conclusions, because that would "leave us hanging in the air". She felt that at this stage, that is, towards the end of the paper, there is no more reason to show consideration or insecurity, and that if he had been hedging at that stage she would have expected more text to come. She thought that the conclusions were pulling together what the writer had been going through, and presenting what he stands for. She also likened this text to another text, which she felt was using the same strategy very successfully. This other text was from the field of law, and as described by the reviser, was hedging all the way through until the very end of the text, where the writer made a clear, unmodified statement as a conclusion. The reverse writer strategy was used in the biochemistry text. The reviser of this text felt that this is the only reasonable way of doing it: in his view, results can be discussed in a straightforward manner, without hedging, because they are clear and there for anyone to see. Results are something that you can show as a writer and see as a reader, but theory is more precarious, as are conclusions: "when you're talking about results, you'll say the results were this, well they were [... you can ] speak solid, but when it comes to getting the theory out, or explaining the results, then there's always a lot of hedging around; you can't be so certain" The third example text (ML) was on economics, and it hedged relatively little all the way through. The strategy was nevertheless closer to the biochemical paper than the educational one, that is, the hedges clustered towards conclusions rather than premises. The reviser's comments were in line with this, although she made them when she was speaking about hedging in general. She felt that if it was a question of an experiment which can be quantified and the results are very tangible there was less need to hedge than on occasions where opinions were discussed. These different interpretations of the appropriate place for hedging can be related to Simpson's (1990) findings on modalisation in a literary-critical text. He found that it was not the most important claims that were modalised, but those that were relatively trivial for the field. Substantial claims, and those that were likely to be controversial were presented as categorical assertions, and as self-evidently true, while less "risky" information was presented as if dubious or likely to cause affront (Simpson 1990, 89, 91). Such results would seem to be in line with the strategy employed in the educational text here.

A. Mauranen, Hedging in Language Revisers' Hands

131

It seems, then, that there are at least two possible ways of dealing with the writer's claims: either to hedge them modestly, in contrast to the empirical evidence that is presented as speaking for itself, or alternatively by highlighting the claims in contrast to hedged and modified speculations over matters which have more of a background status. These two strategies may relate to disciplinary traditions, or to the research approach adopted in the reported studies. Without wishing to speculate here on the possible reasons behind this, it is worth pointing out that these strategies have one textually relevant feature in common: they both highlight the conclusions by treating them differently from the premises. The change of manner in hedging foregrounds conclusions, whichever way the change may take place. 7. Conclusions The two main functions that were assigned to hedges in the text analysis presented here were recognised by the language revisers interviewed for this study: the epistemic and the interpersonal function. These appeared to form the basis of understanding hedging for revisers. The textual roles that hedges can also play, such as organising the text, and foregrounding or backgrounding elements or passages received much less attention here, as they have also done in the literature (but see for example Simpson 1990, Channell 1990, and Myers 1994). However, the two strategies of hedging either the premises or the conclusions pointed towards interesting new possibilities for analysis: not only is the text-organising potential of such strategies worth further investigation, but their possible explanations beyond the ones offered by the revisers. For example, the two strategies may reflect choices open to any individual in the academic world, or they may reflect disciplinary differences, or perhaps differences of academic status, which was implicated now and then by the revisers in explaining differences in the propensity to hedge. The matter is not very easy to settle one way or the other on the basis of available evidence, since it has been typical of studies in this rather narrowly specialised field to deal with a very small number of papers, often only one, or within one discipline only. But there is certainly space for more research here. Comparisons of disciplines in terms of hedging strategies would be most welcome, since we know little about the differences of hedging in different disciplines, although it seems that people like language revisers who regularly work with academic texts seem convinced that such differences exist. The language revisers did not make many corrections to hedges, which is in broad agreement with Ventola/Mauranen's (1990) earlier finding that modality was not corrected very much. What is more interesting is that the revisers presented this as a deliberate and motivated choice. They felt that hedging was the writer's domain, and therefore preferred not to interfere with it. The revisers seemed willing to maintain the author's voice as far as possible, while they would revise anything they felt was linguistically incorrect. The dividing line between the normative sphere of language use and the sphere of personal freedom of choice nevertheless remains unclear. So does the role of language pro-

132

III. Interactive Aspects of Hedging

ficiency and hedging. Do highly proficient non-native speakers actually hedge more than lower-level users, as one of the revisers suggested, or do they just do it more appropriately, and with more subtle nuances? Markkanen/Schröder's (1989) original question of whether it is possible to achieve a correct translation which is faithful to the original writer's intentions and the effect of the original text, given that the original writer's intentions are inaccessible, and the means for expressing hedges are culturally and linguistically variable, seems to get a negative answer. Language revisers are not in a much better position than translators for penetrating and correctly conveying the past intentions of writers. However, the bright side of the matter is that this may not be crucial to a good end product of the translation or revision process: the translation may serve its purpose in its target environment without being strictly faithful to its originator. Translators can be assessed from various angles, and using various sets of priorities, as for example Zabalbeascoa (1994) has recently pointed out in a model of the contexts of translations. Only one of these sets of priorities is the degree to which the translation is an accurate reflection of the source text. If we switch our priorities from the past of the text to its future, the appropriateness of its hedges are to be judged from the perspective of its target context. To achieve appropriacy in its future context of use, an academic text needs the joint efforts and combined expertise of the writer as a member of the academic community, and the reviser as a native speaker of the language.

References Brown, Penelope; Levinson, Stephen 1978 "Universals in Language Usage: Politeness Phenomena", in Goody, N. (ed.): Questions and Politeness. Cambridge: Cambridge University Press, 56-324. Channell, Joanna 1990 "Precise and Vague Quantities in Writing on Economics", in Nash, W. (ed.) The Writing Scholar: Studies in Academic Discourse. Newbury Park: Sage, 63-94. 1994 Vague Language. Oxford: Oxford University Press. Crismore, Avon; Markkanen, Raija and Steffensen, Margaret 1993 "Metadiscourse in Persuasive Writing. A study of Texts Written by American and Finnish University Students", Written Communication, 10/1, 39-71. Fairclough, Norman 1992 Discourse and Social Change. Cambridge: Polity Press. Halliday Michael A. K. 1985 An Introduction to Functional Grammar. London: Edward Arnold. Ide, Sachiko 1988 "Introduction", Multilingua 7, 371 -374. 1989 "Formal Forms and Discernment: Two Neglected Aspects of Universals of Linguistic Politeness", Multilingua 8,223-248. Janney, Richard W./ Arndt, Horst 1993 "Universality and Relativity in Cross-Cultural Politeness Research: A Historical Perspective", Multilingua 12, 13-50. Luukka, Minna-Riitta 1992 Akateemista metadiskurssia. Tieteellisten tekstien tekstuaalisia. Interpersonaalisia ja konlekstuaalisia piirteitä (Korkeakoulujen kielikeskuksen julkaisuja 46). Jyväskylä: Korkeakoulujen kiek'keskus.

A. Mauranen, Hedging in Language Revisers' Hands

133

Lyons, John 1977 Semantics. Cambridge: Cambridge University Press. Markkanen, Raija; Schröder, Hartmut 1989 "Hedging as a Translation Problem in Scientific Texts", in Lauren, C. and Nordman, M. (eds.) Special Language: From Humans Thinking to Thinking Machines. Clevedon: Multilingual Matters, 171-179. Myers, Greg 1989 "The Pragmatics of Politeness in Scientific Articles", Applied Linguistics, 10/1,1-35. 1994 "Strategic Vagueness in Academic Writing", paper presented at the symposium 'Academic Writing - Research and Applications'. University of Helsinki, May 19-23, 1994. Palmer, Frank R. 1986 Mood and Modality. Cambridge: Cambridge University Press. Simpson, Paul 1990 "Modality in Literary-Critical Discourse", in Nash, W. (ed.) The Writing Scholar: Studies in Academic Discourse. Newbury Park: Sage, 63-94. 1993 Language, Ideology and Point of View. London: Routledge. Tabakowska, Elzbieta 1989 Lexical markers in English and Polish, Multilingua 8, 21-36. Valle, Ellen 1993 The Talkative Community. Rhetorical, textual and pragmatic features in Royal Society texts, 1711-1870. Unpublished licenciate thesis: University of Turku, Department of English. Ventola, Eija; Mauranen, Anna 1990 Tutkijat ja englanniksi kirjoittaminen. Helsinki: Helsinki University Press. Zabalbeascoa, Patrick 1994 "Awareness in Translation", paper presented at the Second International Congress of Language Awareness, Plymouth, United Kingdom, April 11-14, 1994. Zuck, Joyce Gilmour/ Zuck, Louis V. 1985 "Hedging in Newswriting" in Comu, A-M., Vanparjis, J., Delahaye, M, and Baten, L. (eds.) Beads or Bracelet? How do we approach LSP? Selected Papers from the 5th European Symposium on LSP. Oxford: Oxford University Press, 172-180.

WOLFRAM WILSS

Hedges in Expert-Language Reviews

1. Introduction

The following discussion is based on the premise that the meaning of the term "review" and the purpose of reviews can be taken for granted. Anyone who is active in the field of research, either receptively or productively, has an intuitive notion of what reviews are and what they are aimed at, and regards them as a cultural technique which is invaluably contributory to coping economically with our scientific universe of discourse (Schweickhard 1992). In view of their relatedness to the past, the present, and the future of research, reviews are an essential factor of our research activities, combining and integrating mainly two functions, reporting and evaluating research findings in a more or less condensed and expert manner (Dallmann 1979; Gläser 1979; Jehle 1990; Jokubeit 1980; Pätzold 1986; Ripfel 1989). Reviews are an illustrative evidence of the well-known hypothesis that any communication contains a referential and a relational aspect. The sender not only informs the recipient about something scientific in the form of concentrated propositions, but he also suggests the way in which he thinks the latter should see the communicated subject-matter. In other words, the reviewer does not confine himself to factual determinants; he also engages in some sort of self-exposure, in so far as his text contains more or less clearly identifiable and rhetorically competent assessments of a scholar's work. "Evaluations are controversial, because they are guided by individual interests and motivations, but this does not mean that they are scientifically useless" (Nußbaumer 1991, 3; translated by W.W.). The impulse for investigating the above-mentioned double perspective is by no means incidental. It is an overdue reaction to the one-sidedness of a research paradigm which has become known as "generative theory". Meanwhile it has turned out that linguistic research which is both theory- and practice-oriented must not reduce its object, i.e. natural language, to a quasialgebraic system; but that it must deal with authentic data and concrete observations in the field of linguistic usage, thereby basing its activities on the underlying assumption that linguistics is a means to an end rather than an end in itself in the sense of an autonomous branch of research. Undeniably, generative theory has enhanced the scientific status of linguistic research, but at the

W. Wilss, Hedges in Expert-Language Reviews

135

same time it has led into a theoretical dead-end street, because it has ignored the fact that a good deal of language usage is a matter of degree. It has pursued its target at the expense of linguistic spontaneity, openness, and naturalness, taking full account of the richness, subtlety, and complexity of linguistic usage; on the other hand, linguistic usage is - for the theoretical and the practical linguist - an intricate and sometimes elusive phenomenon with a high degree of contextual dependency. In view of the fact that contextuality is inherently and invariably situation-relative, context often appears vague and unstable over time, communicative specifications, and socio-cultural settings. "Language is like a game we are often told; but if so, it is a game with soft rules: not like chess, played on board of abstract geometry, but rather like golf, to be played on this actual course or that" (Vender 1980, 209). Linguistic research which looks at concrete textual manifestations such as reviews must be aware of its principal a posteriori character. Linguistic usage is always far ahead of linguistic theory, i.e. linguistic theory follows linguistic practice, trying, as best it can, to register and to explain what goes on in the language user's mind while communicating. Seen from this angle, linguistics is a reconstructive rather than a constructive field of research. We want to understand, under text-strategical aspect, how someone proceeds when he formulates or reads a review without being fenced in by premature theoretical prescriptivism. A simply propositional argumentative frame of reference on the basis of syntactic and selection-restrictive considerations does not suffice for the description and explanation of reviews. We must enlarge our research design through the incorporation of external factors and standards with the help of which we can open a vista for the comprehension of reviews as a phenomenon of a specific type of communicative interaction. The development of linguistic theory in the last two decades has shown that theoretical standpoints such as generative theory are not irreversible. A change of theoretical assumptions seems in order for two reasons. First, generative theory shows symptoms of fatigue; as a result, one might talk of generativity as a degenerating research paradigm. Second, generative theory has failed to successfully accommodate growing factual knowledge of language and treat it naturally and insightfully as an integrated whole. What we need is not an over-ambitious super-theory with the inevitable consequence of an increasing divergence between practice and theory but a program which takes account of language as a means of communication with a rich variety of rhetorical strategies and techniques which, for the benefit of the reader, we have to present in a simple, natural, and intuitively satisfying manner, without giving up the principles of scientific work and taking refuge in ad hoc descriptive devices, however pervasive they may look. After the generative phase, we again understand language, or, for that matter, language usage, as an anthropologically, psychologically and socially embedded mixture of meanings, relations and structures serving as an instrument for efficiently coping with our everyday communicative needs and intentions. Lexical, idiomatic (phraseological), and syntactic phenomena are seen in a functional rather than an ontological framework, embracing four different

136

III. Interactive Aspects of Hedging

kinds of competences, contextual competence, meaning-ascribing competence, coherence competence, and paraphrasing competence. This approach entails four consequences: 1. The language user has the status of an important "intervening variable" in linguistic discovery procedures. 2. The subject-matter to be analyzed can be presented only through the extensive elucidation of its basic perspectives and the introduction of a considerable array of coherent concepts and descriptive devices. 3. The criticism of rhetoric as a technique of deception, as indicated first by Plato, has become immaterial. 4. The logo-centred view of language, basically as erroneous as the relativity concept, has been pushed into the background in favor of a frame of reference in which social factors with their "fuzzy-logic"-properties have more weight than "naturalistic" laws with their reductionalist impact. Linguistic activity is conceived of as a manifestation of processing linguistic data combining intentional knowledge (knowing what for), declarative knowledge (knowing what), and procedural knowledge (knowing how). The language user musters up the various knowledge bases situationally with the help of a mental grammar (behavioral grammar) integrating perception, attention, memory resources, evaluative capacities, ratiocination, psychomotoric components, text and world knowledge in such a manner as to create the preconditions for efficient sender/recipient interplay. As a research target, the investigation of this interplay is not new: About 60 years ago this conception was discussed by Bühler (1934) who thus prepared the way for dynamic meaning constitution and the recognition of "interactive" text strategies in the production and the reception of textual material. What we are interested in is not the philosophical construction of a methodically "unified" scientific conception, but an experientially plausible frame of reference which can be made operative by listing all textually relevant parameters. In doing this, we take it as self-evident that the trend must be towards greater emphasis on surface form and less reliance on the derivation of surface structures from abstract underlying representations.

2. Textual Aspects Constitutive of such a frame of reference are four text dimensions, text content, text function(s), text pragmatics, and rhetorical strategies. These four dimensions are easily recognizable in reviews. A person, as a rule an expert, expresses her/his opinion on a scientific work with a view to bringing about a (tacit) feedback between herself/himself and the respective author and to familiarizing a more or less expert readership with the achievements and failures of the work under review. The reviewer produces, on the basis of his subjective text assessment, a metatext directly related to a primary text. This is to say that,

W. Wilss, Hedges in Expert-Language Reviews

137

normally, a review is a mixture of referential and commenting text passages, with no established norm for organizing these two text functions. Clearly less marked is the appellative function, i.e. the suggestion to buy this particular work (or not to buy it, as drastically expressed in an American review with the following closing statement: "If you have money to burn, burn it; don't buy this book!"). Reviews drive home in an exemplary fashion the well-known fact that, in dealing with our environment, we tend to combine situation description with situation evaluation: Human beings react to the events around them not as neutral observers who confine themselves to the registration of the behavior of things in such and such a manner. Human beings are not neutral, they cannot "keep out", because their observations and expressions are always related to action and are embedded in actions (Bayer 1982, 16; translated by W.W.).

Like other functional and pragmatic parameters, evaluations have a surface representation; in reviews they largely determine the thematic and formal constitution of a review text. Cassirer's famous question, whether we deduct the function from the gestalt or the gestalt from the function, that is to say, whether we prefer the functional perspective over the formal perspective or vice versa is irrelevant, considering the fact that reviews represent bundles of features whose investigation reveals a cline of sophistication, subtlety, and intricacy. It will serve as a convenient reference point for our considerations if we remember that all we can reasonably expect of a textsurface-dominated approach is that we are as clear and precise as possible in describing our understanding of the subject-matter we are intent on dealing with. This is particularly important for the investigation of a phenomenon which is typical of reviews and which is known in modern linguistics under the name of "hedges", a term which linguistics has probably adopted from the language of business administration. It is meanwhile common knowledge that the interest in hedges goes back to an article which Lakoff, without making reference to reviews, published in 1972 under the title Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts. According to Lakoff, "natural language concepts have vague boundaries and fuzzy edges". As a result, "natural language sentences will often be neither true nor false, nor nonsensical, but rather true to a certain extent, true in certain respects and false in other respects" (1972, 183). Linguistic manifestations of hedges in day-to-day communication are words and phrases such as "sort of, more or less, roughly, somewhat, rather, technically, strictly speaking, for the most part, as it were", and many others (for a lexical approach to hedges see Kolde 1986). Taking up Lakoff and elaborating his concepts, Hübler (1983), has made a distinction between "understatements" and "hedges". However, his argumentation is not cogent, because both categories serve the expression of what he calls "indetermination", a notion which closely resembles LakofPs concept of fuzziness (for a detailed discussion of Hubler's approach see Markkanen 1989 who deals with hedges in the context of epistemic modality).

138

III. Interactive Aspects of Hedging

3. Communicative Functions of Hedges

The reason why people tend to use hedges is self-explanatory. Hedges make utterances more acceptable to recipients. They are face-saving strategies which enable the sender to "soften and tentativize"; they are "downtoners, as some writers choose to call them, not those that strengthen, i.e. boosters, particularly since only the former are in keeping with the everyday meaning of hedging" (Markkanen 1989, 143). Without hedges, writes Clemen (1990, 9) quoting Skelton, the world would be a rigid gestalt of an either-or mentality, lifeless and dreary. Hedges make our language more flexible and the world more subtle; it is hedges, which provide our language with spontaneity, impact and appeal.

Clemen (1990, 9) adds: Hedges push dogmatism into the background, and, anyway, who knows with absolute certainty diat something is that way or another. We need by no means name things in a direct manner, and we need not lay open all our cards in all situations. Whatever follows the colon is either uninteresting or it is not known. Conservative statements help us to protect ourselves and the addressee. Moderate expressions support polite manners in dealing with others, and sometimes it is urgent to camouflage things in order not to be regarded as rude. It is easy for a writer to criticize if he modifies what he has got to say (translated by W.W.).

Hence, the build-up of a sufficiently differentiated amount of hedging strategies is essential for the control of a large number of delicate communicative situations e.g. orders, requests, claims, questions etc. It is, therefore, almost trivial to state that Grice's co-operative principles (1975) contain a politeness component ("manner"). Leech's suggestion to expand the co-operative principle by introducing a "politeness principle" (1983) can therefore be dispensed with. As we shall see later on, hedges are highly standardized mini-programs which preclude social embarrassment in many discoursal situations; they serve as an impact-attenuating device with the help of which we can avoid mutual hostility or at least play down unavoidable confrontations to a mutually tolerable degree of discomfort (Brown/Levinson 1987; Markkanen/Schröder 1989). Hedges are a constitutive element of linguistic performance with an almost ubiquitous range of application; the trademark of hedges is inoffensiveness; in the stream of perpetually changing states of consciousness hedges have a defusing, "defocalizing" effect preventing the language user from manoeuvring himself into critical situations and thus making life for himself and for others more difficult than necessary. It is obvious that hedges, which Gofrman has referred to as "softening mechanism" (1967, 45) and which Gläser has called "höfliche Kritik" (polite criticism) (1979, 123), are of particular significance in review texts where they appear in a multitude of surface forms (Brown/Levinson 1987, 146). They are useful when a reviewer wants to be non-committal to some extent, last but not least for fear a negative assessment (justified as it is) might turn out to be a boomerang, or when he wants to resort to formulations which give the reviewed person the feeling that the reviewer's comments are not meant as a personal offence.

W. Wilss, Hedges in Expert-Language Reviews

139

This is especially crucial in the final section of a review telescoping what has been said so far into statements which the review reader might preserve in his memory more readily than other, more or less long-drawn-out review passages dealing with details of the reviewed work. The following text specimen may serve as an illustration (since I am a native speaker of German, the editor has agreed to a German documentation; all examples are, of course, original, all names have been replaced by X, no sources are given for obvious reasons): Insgesamt ein ideenreiches, mit übersetzungsmethodischen Perspektiven vollgepacktes Buch, das dem Leser - damit meine ich auch und vor allem den Übersetzungsstudenten - eine Fülle von Anregungen vermittelt und das auch durch selbstironische Bemerkungen ("Those who can, write; those who cannot, translate; those who cannot translate, write about translation") sympathisch ist. Daß das Buch auch manche argumentativ schwächeren Passagen enthält, daß es gut gewesen wäre, wenn man entstehungsbedingte Wiederholungen gestrichen hätte, sei nicht verschwiegen; auch die Bibliographie ... und die beiden Register lassen manches zu wünschen übrig. Dies schmälert jedoch den insgesamt positiven Eindruck des Buches, das aus den didaktischen Erfahrungen des Verfassers hervorgegangen ist, nur unwesentlich. Die in diesem Buch etwas vernachlässigten übersetzungstheoretischen Fragen will X in einem "text book of translation theory and practice"... diskutieren; es bleibt abzuwarten, ob er die provokatorische Behauptung, daß Begriffe wie Übersetzungseinheit, Übersetzungsäquivalenz, Übersetzungsinvarianz, übersetzungswissenschaftlich gesprochen, "dead ducks"... sind, aufrechterhalten und verteidigen kann. 9

Anyone who is to some extent familiar - either as the author or the reader of reviews - with the cognitive style of reviews is immediately aware that here we experience a blatant difference between what has been said and what has been meant and that the reviewed work, above all concerning its economical organization leaves a lot to be desired. Hedging strategies turn out to be a specific communicative game in which the production and the reception of hedges is related to the scheme "I know that you know that I know". Hence, what we need for the investigation of hedges (which are also typical of reference letters) is a descriptive framework which accommodates both the reviewer and the reviewee and which operates on a common knowledge structure. It presupposes that the reviewer has the ability to formulate hedges and that the reviewee (or any other reader) has a corresponding competence which enables him to reconstruct these hedges. The interrelatedness of these two proficiencies are based on a complicity effect or "in-group"-behavior - without the need to engage in back-and-forth negotiations with the aim of clarifying communicative intentions as a result of hermeneutic or analytical text-processing strategies. The characteristic property of reviews is their two-level structure which supports the assumption that the positing of hedges in a review requires a shared appreciation resting on an associative rather than a modular network of cognitive operations. Assigning a two-level structure to reviews means that they contain two meaning dimensions, a foreground prepositional meaning level and, underneath, a somewhat veiled (or implicit) meaning dimension representing the real opinion of the reviewer. Experienced review-writing displays a gradation from central to peripheral depending on how far and in what ways the reviewed book deviates from the expectations and the expertise of the reviewer.

140

III. Interactive Aspects of Hedging

4. Knowledge Structures in Hedges As a rule, the reviewer operates on implicit partner hypothesis because normally there is no feedback between reviewer and reviewee. A compensation for the lack of such a feedback is the feeling of "we-awareness" which is imperative in the hedging context, because hedges intentionally violate the normality conditions of linguistic communication. A text which contains hedges is not a piece of straightforward communication in a yes/no fashion. Hedges, in view of their entrenchment in our intellectual life, show in an exemplary fashion that text production comes before text reception and that the sender, in order not to risk a communicative breakdown, depends on a co-operative recipient. Sigmund Freud has pointed out that nobody makes a joke for himself. Analogically, we can say that nobody produces hedges just for himself. Thus, there must be a fit between the associative disponibility of the reviewer and the reviewee. The complementarity between reviewer and reviewee roles presupposes "active" reading of the contexts in which hedges occur and the ability to translate hedging strategies into one's own communicative experience. The significance of tacit knowledge between reviewer and reviewee is particularly noticeable in cases where the reviewer utilizes patterns which are characterized by hedges and, as such, can hardly be misunderstood. The following selection of hedging segments is not an a priori matter of preference or scientific rigor, but a factual one pertaining to at first sight non-prototypical instances that can be assimilated to classes or categories of hedges to the extent that they can be construed as matching or approximating prototypes. They form a continuous spectrum of possibilities whose segregation into distinct configurations or phraseologically petrified entities together form a coherent integrated structure: -

...trotz all seiner Mängel und Fragwürdigkeiten... ...trotz der vielen Vorzüge...hat X die Erwartungen nicht erfüllt... ...wir heißen es (das Buch; W.W.) als Zwischenlösung willkommen... .. .trotz einiger kritischer Hinweise... ...doch wiegen solche Einwände nicht schwer... .. .wenn man nicht mehr verlangt, kommt man auf seine Kosten... .. .wie anerkennenswert das Erreichte auch sein mag... ...dieser begrüßenswerte und wohl auch richtige Schluß... ...aufgrund dieser Erkenntnisse hätte es allerdings nahegelegen... .. .dieses Ziel hat der Autor allemal erreicht... ...die (Informationen; W.W.) müssen wir uns umständlich...aus dem Register mit doppelspaltig bedruckten Seiten klauben... - ...daran ist vielleicht der besondere Ehrgeiz des Autors nicht ganz unschuldig... - ...wenngleich einzelne Lehrer abhängig von ihren persönlichen Schwerpunkten sicher einiges vermissen werden...

5. Frame Structures in Hedges When employing hedges, we generally manipulate the resulting configurations holistically, as a kind of gestalt, and we know how to gauge them psychically,

W. Wilss, Hedges in Expert-Language Reviews

141

situationally, and socially. We have internalized a complex network of expectations based on previous experience and we fall back upon them when interpreting linguistic strings that we hear or see. The fact that reviewer and reviewee possess a comparable experiential background, ascertains a high level of semantic recoverability. We realise that straightforward plus/minus values are alien to hedges. Hedges are, as indicated, a matter of degree, and it is not always easy to mark offhedges from non-hedges. A presentation of hedges in the form of a list is perfectly legitimate, but does not in itself fully reconstruct the systemic nature of the actual phenomena and accommodating them in a complete range of possible hedging strategies. The subtlety of hedged formulations can be shown by the two syntagmatic strings "Trotz einiger kritischer Hinweise..." (despite some critical remarks...) and "Trotz der vielen Vorzüge..." (despite the many positive aspects...). Whereas in the first case the overall judgement is, in the last analysis, favorable, because this type of strategy making use of 'trotz' anticipates a generally positive appraisal, the opposite is true of the second case. The reviewer does mention some appealing points, but on the whole the reviewed book remains (far) behind the expectations of the reader. Besides innovative (creative), subject-centred formulations, hedges consist of a good deal of repository-type standard configurations requiring a relatively low level of cognitive input (communicative dynamism) typical of collectivestyle phraseology. It is in line with the mental nature of mankind that there are both, reflexive and routinized forms of productive and receptive linguistic behavior in communicative management. Linguistic routine can be called knowledge- and experience-based communicative skills. Our everyday observations point to the fact that the language user is inclined to employ this combination of knowledge and skills whenever he can (and sometimes even outside the range of what is situationally permissible), the explanation being that, in dealing with his environment, the subject wants to proceed as economically as possible; he moves, so to say, over from the sphere of "creatio" into the sphere of "creatum". This applies not only to manual, but also to cognitive activities reflecting a desire for relief ("Entlastung", as Gehlen called this phenomenon as early as 1957). This is to say that in our behavior, linguistic or otherwise, we look for "minimax strategies" with an optimal input/output relation. In a metaphorical way, Schreckling (1985,69) has stated that in the routine handling of situations the individual switches on his "autopilot", whereas in problemsolving activities - and innovative expressions are a specific kind of problemsolving - he uses mentally controlled strategies (Handsteuerung). In the review context, economy strategies are best manifested in "programmed" behavior expressing itself in the repetitive use of internalized modular components (Textbausteine) whose characteristic feature is their almost unlimited availability. To make the rich variety of hedged techniques and approaches generally accessible, four conditions must be fulfilled: 1. Our hedging competence must have become an integrated part of our general linguistic competence. It consists of our having accumulated a store of highly normative configurations (or style markers) which show how reviewers express themselves in specific evaluative situations. Their function is to give a review its peculiar rhetorical flavor.

142

. Interactive Aspects of Hedging

2. There must be, as stated before, a convergence (in the sense of Wittgenstein's familiarity ratings) between what the reviewer and the reviewee have at their disposal - productively or receptively - in the way of hedging-specific patterns. 3. The review writer must be able to utilize the accumulated hedging inventory independently from context in a spirit that unites a variety of approaches and outlooks into a loose-knit collective enterprise accommodating prevailing trends in review-writing. 4. The routinized handling of hedging techniques presupposes the adoption of "topical inventories" (topische Arsenale; Hess-Lüttich 1984, 285) with the help of which a reviewer can respond critically to a scientific work with ease, impact, and rapidity.

6. Examples of hedging devices in reviews Like any human being, many reviewers have a predilection for habitualization, because uniformity of processes, repetition of the same or similar features in similar or same situation is one of our primary modes of behavior. The salience of routinized hedging techniques can be made abundantly clear by the following, highly selective typology of examples (mostly from linguistic periodicals; some examples are taken from the scientific supplement of the German daily or weekly press): 6.1. Concessive relations (1) (2) (3)

(4) (5) (6) (7)

(8)

Der Xsche Paradigma-Begriff ist, wie er (the reviewee; W.W.) selbst im Postskriptum von 1969 einräumt, mehrdeutig. Präzisierungen...sind zwar hilfreich, aber auf nichtformalisierte Ansätze nicht unproblematisch zu applizieren. Der durch den exzessiven Gebrauch von Fremdwörtern geprägte Stil wird zwar ausdrücklich mit dem Dienst an "sachorientierter Systematik and Methodik" begründet. Aber es bleibt der Eindruck, der Verfasser fürchte den Vorwurf mangelnder Professionalität und tue daher zuviel des Guten. Der Leser wird zwar knapp und kompetent mit X's Werken und Wirken vertraut gemacht, die heikleren Fragen aber werden eher geschickt umgangen und apologetisch bereinigt als einer offenen und kritischen Diskussion unterzogen. Selbst wenn man Einzelheiten anders sieht, kann man sich daher der Stringenz der Xschen Argumentation und Erklärung im ganzen nur schwer entziehen. Und dieses Konzept läßt sich hören, auch wenn es nicht jedermann Konzept sein muß. Allerdings liefert X "nur" theoretische Überlegungen für sein Konzept, kein ein ziges Beispiel für empirische Forschung nach seinem Modell. Aus diesen Elementen baut X sein eigenes Modell von Bedeutung im kommunikativen Handeln auf...; in vielen Punkten kann man ihm hier zustimmen, nur scheint mir seine - berechtigte - Reaktion gegen eine frühere Überschätzung des Sprachzeichens und seiner Rolle in der Interaktion gelegentlich etwas weit zu gehen. Auf diese Weise wird bei X das Arbeitsgebiet der historischen Semantik sehr ausgeweitet, und das erscheint mir im Prinzip richtig. Allerdings soll nicht verschwiegen werden, daß der Autor selbst wiederholt an der empirischen Ausführbarkeit seines Modells zweifelt...

W. Wilss, Hedges in Expert-Language Reviews (9) (10)

143

Das Lehrbuch hat die Anmerkungen je Kapitel beibehalten; das macht die Nutzung ein wenig schwerfälliger. Gleichwohl wird an den Anmerkungen die Breite das überblickten Zeit- und Wissensraums deudich. Die Interpretation von X's Rhetorikkritik...ist zwar originell, wird aber dem Verhältnis von Rhetorik und Ästhetik bei X nicht gerecht.

6.2. Use of subjunctive (11)

(12) (13) (14)

Beim späten X, und zwar genau in jener Schrift, die...entfaltet, hätte der Verfasser lernen können, daß der Phänomenologe sich der "naiven Sprechweise des Lebens" bedienen muß, selbst wenn er sich damit von der Zunft der Wissenschaftler den Vorwurf des Dilettantismus einhandelt. Ein eigenes Kapitel befaßt sich mit "Rundfunk und Wochenschau". Hier hätte man sich doch ein bißchen mehr gewünscht... ...deshalb wäre vielleicht eine andere Formulierung des Titels angezeigt gewesen. ...die Texte...böten eine Fülle von empirischem Material.

6.3. Parentheses (bracketed additions) (15) (16)

(17)

Die Bedeutung der Interjektionen für die Gewährleistung formaler Kooperativität führt X dazu, X's Zwei-Felder-Konzept (Symbolfeld, Zeigfeld) durch ein Lenkfeld (oder terminologisch nicht ganz glücklich durch ein "expeditives Feld") zu ergänzen. Natürlich sind X's Spekulationen über das Unbewußte, die Symbolik und die Sprache "interessant" (ein Lieblingswort von X - vor allem zur Kennzeichnung von Gedanken, die er nicht genauer ausführt); es sind aber auch so viele kritische Einwände möglich, daß sich am Rande dieser Abhandlung - die übrigens nicht unbedingt in den Band gehört - meine Fragezeichen häufen. Mit solchen Interjektionen signalisiert der Sprecher - so könnte man X ergänzen eine (kurzfristige) Veränderung (Um-"Lenkung") seiner Ansprechbarkeit, eine Verschiebung seiner Aufmerksamkeit, Erwartungen, Präferenzen, möglicherweise eine Beeinträchtigung seiner Bereitschaft zur Kooperation.

6.4, Phraseological expressions (18) (19) (20) (21) (22) (23) (24) (25)

Es versteht sich, daß bei einer so fundamental angelegten, gleichwohl relativ kurz gehaltenen Untersuchung Fragen offen bleiben. Bei einer Gruppenbildung besteht überdies die Gefahr, daß der Herausgeber vorher festgelegte Ordnungsgesichtspunkte umstößt. Ein sehr eigenwilliger Gestusbegriff ist es, den X entfaltet und facettenreich belegt. Da er sowohl makro- als auch mikrostrukturelle Verwendung findet, ist der Leser streckenweise in begrifflichen Nöten. Flüchtigkeitsfehler wie nicht nachgewiesene Literatur...fallen aber ebensowenig ins Gewicht wie inzwischen überholte Aussagen, etwa zum "sehr rudimentären Stadium" der deutschen Satzintonationsforschung. Und es ist überhaupt auffällig, daß in dem sonst so überaus gründlichen Buch der eigentümliche Philosophiebegriff X's nicht einmal am Rand erwähnt wird. Das bestimmt den liebevoll-einfühlsamen Umgang mit ihnen (den in dem betr. Werk diskutierten Autoren; W.W.), freilich auch eine gewisse unkritische Identifikation mit ihren Lesern... Wenn X anschließend jedoch äußert,...fragt sich der Leser doch... Abgesehen von ziemlich vielen überflüssigen Wiederholungen und gelegendichen begrifflichen Vagheiten..., ein gutes, perspektivenreiches, informatives Buch...

144

III. Interactive Aspects of Hedging

6.5. Expanded participk constructions (26) (27)

Flüchtigkeitsfehler wie nicht nachgewiesene Literatur...fallen ebensowenig ins Gewicht wie inzwischen überholte Aussagen, etwa zum "sehr rudimentären Stadium" der deutschen Satzintonationsforschung. Dennoch scheint mir die bewußtseinsphänomenologische Psychologie mitsamt der von X entworfenen, aber nicht eigentlich ausgeführten "bewußtseinseigenen Sprachwissenschaft" heute nicht überzeugender als zur Zeit ihrer ersten Veröffentlichung.

6.6. Negation (28) (29) (30) (31) (32)

Wonach sich diese Anschlußfahigkeit bemißt, teilt X nicht mit. ...eine Problematisierung der Termini begegnet nicht... ...abgesehen davon, daß seine Quelle nicht angegeben wird... Interrogative Funktionen sind durchaus nicht immer oder auch nicht überwiegend an die interrogative Kadenz gebunden; auch gibt die empirische Erfahrung keinen Anhalt, zwischen...und...zu unterscheiden. Die Melodieverläufe sind kaum so zu verallgemeinern und lassen sich in empirischen Untersuchungen...so nicht bestätigen.

7. Conclusions

The list of examples, which could be extended ad libitum, is helpful in demonstrating what this paper seeks to accomplish, the data it considers, the questions it asks, what it accepts as persuasive evidence, and so on. The documentation permits the following conclusions: 1. Hedges are a basic interactive strategy in reviews; they occur in many types and contexts; they are a constitutive - and, as such, highly predictive - element of reviews. Hedges belong to the principal categories of rhetoric; they can, as it were, be regarded as sort of "sugar-coating" in reviews; they represent an important field of a functionally oriented language-usage grammar. 2. A considerable proportion of hedges can be subsumed under the category of rhetorical patterns or behavioreme-type "hedging mechanisms" for which Lausberg (1963), one of the pioneers in rhetorical research has coined the term "Wiedergebrauchstexte" (texts for repetitive use). The schematic character of hedges is particularly obvious in tokens which are popular because of their easy learnability, contextual ability to convince, and adaptability, thereby implanting themselves on a wide range of review activities. 3. The investigation of hedges has two consequences. First, it extends far beyond the domain of persuasive speech as represented by classical rhetoric in the sense that it includes common-language and expert-language communication. Second, it marks hedges off from other forms of speech such as rigidly scientific discourse. 4. The development of objective criteria for ways of speech that are characterized by hedges (e.g. within the framework of a taxonomy of hedges ranging

W. Wilss, Hedges in Expert-Language Reviews

145

from (rather) unobtrusive to (rather) powerful specimens with a clear delineation between covert and overt criticism is difficult, all the more since the quality of a review depends - both in its positive and its negative evaluative remarks - on the rhetorical knowledge and skills of the reviewer. The interpretation of a specific formulation in a review as hedge or non-hedge rests significantly - at least in cases where the reviewer formulates innovative and not routinely - on the subjective assessment capacity of the reader. 5. Whether a reviewer prefers a more aggressive or a more subdued, mitigating form of criticism, e.g. by "side-tracking" or "decentering" information with the help of parentheses or embedded participial constructions, depends on circumstances which are largely (or totally) unknown to the outsider, be it the reviewee himself or other review-readers. Motives for aggressiveness or polemical statements may be envy, revengefulness or simply showing off. An important factor in adopting hedging strategies are the strong forces of collegiate politeness (Brown/Levinson 1987, 2); these forces may rebound on the reviewer in cases of role-switching, i.e. if he is the one who is reviewed. 6. In view of the fact that hedges are a cross-linguistically observable phenomenon, it would be an interesting task to compare hedging devices in individual language-pairs such as German and English. Presumably a confrontative approach would reveal more similarities than differences, as is, e.g., shown by the fact that hedges are relatively easy to translate (Reinert 1990, 87). More evidence for German-English convergences in review-organizing are yes/but constructions, with the "but"-argument always having more impact than the "yes"construction, as the following English example shows which with the same substance could also be found in a German review: (33) (34)

X's paper is, as we might expect, full of ideas, brilliant, and provocative, but it is also in many respects wrong (original). X's paper is in many respects wrong, but it is, as we might expect, also full of ideas, brilliant, and provocative (my reformulation).

While in (33) the negative overall impression is predominant, the converse is true of (34): The author is wrong in many ways, but in view of his many provocative, brilliantly formulated ideas such weaknesses do not essentially detract from the qualities of that particular publication. Nevertheless, we must keep in mind, as Markkanen has suggested, "that some cultures are more "hedgy" than others, and academic writing is also influenced by the conventions of the writer's cultural background" (1989, 144). However, at the moment it seems, in my view, premature to make sweeping statements without prior empirical testing of interlingual corpora and to postulate, as Galtung (1983) has done, the existence of a Teutonic, a Gallic, a Nipponic and a Sachsonian style in the discussion of scientific subject-matter. In my opinion, it is obvious that in assessing the positive and the negative aspects of a scientific piece of research, we must reckon with a transitional area, making paradigmatic statements virtually impossible. Despite these limitations, hedges are a fruitful field of research allowing insights into the text-strategic intricacies which are almost invariably apparent in communicative situations characterized by a more-or-less rather than by an all-or-nothing attitude. But to

146

III. Interactive Aspects of Hedging

posit a continuum is not tantamount to the abolishment of the principle of methodically rigorous investigation. What we must and can do is to describe the individual hedging strategies and techniques so as to make transparent the degree of reticence noticeable in the writings of the reviewer. It is a moot question whether the description of classes of hedges should have priority over the analysis of individual occurrences, but we should remember that hedges are, at least to a considerable extent, systemic, semi-institutionalized in nature. Their properties are best discernible within an integrated approach and are adequately describable only in the context of their own functional system. It is this system which determines when hedges may occur, how they will be recognized, what they may be combined with textually, which patterns they follow, and generally, how they fit into the organizational structure of a review.

References Bayer, Klaus 1982 "Mit Sprache bewerten", Praxis Deutsch 53,15-25. Brown, Penelope and Levinson, Stephen C. 1987 Politeness. Some Universals in Language Use. Cambridge: Cambridge University Press. Bühler, Karl 1934 Sprachtheorie. Die Darstellungsfunktion der Sprache. Jena: Gustav Fischer. Clemen, Gudrun 1990 Hecken in englischen Wirtschaßstexten. MA Thesis. Universität/Gesamthochschule Siegen. Dallmann, Sabine 1979 "Die Rezension. Zur Charakterisierung von Texttyp, Darstellungsart und Stil", in Fleischer, W. (ed.): Sprachnormen, Stil und Sprachkultur. Berlin: Akademie der Wissenschaften der DDR, 58-90. Galtung, Johan 1983 "Struktur, Kultur und intellektueller Stil. Ein vergleichender Essay über sachsonische, teutonische, gallische und nipponische Wissenschaft", Leviathan 2, 303-338. Gehlen, Arnold 1957 Die Seele im technischen Zeitalter. Sozialpsychologische Probleme in der industriellen Gesellschaß. Hamburg: Rowohlt. Gläser, Rosemarie 1979 Fachstik des Englischen. Leipzig: VEB Verlag Enzyklopädie. Goffman, Ervin 1967 Interaction Ritual. Garden City/New Jersey: Doubleday. Grice, H. Paul 1975 "Logic and Conversation", in Cole, P. and Morgan, J. L. (eds.): Syntax and Semantics. Vol. 3: Speech Acts. New York: Academic Press, 41-58. Hess-Lüttich, Ernest W. B. 1984 Kommunikation als ästhetisches Problem. Vorlesungen zur angewandten Textwissenschafi. Tübingen: Narr. Hübler, Axel 1983 Understatements and Hedges in English. Amsterdam: Benjamins. Jehle, Günther 1990 Das englische und französische Lernerwörterbuch in der Rezension. Theorie und Praxis der Wörterbuchkritik. Tübingen: Niemeyer.

W. Wilss, Hedges in Expert-Language Reviews

147

Jokubeit, Werner 1980 Das Erörtern in der Rezension. PhD-Thesis. Dresden. Kolde, Gottfried 1986 "Zur Lexikographie sogenannter Heckenausdrücke", in Weiss, G. et al. (eds.): Akten des VII. Internationalen Germanisten-Kongresses Göttingen 1985, Vol. 3, Tübingen: Niemeyer, 170-176. Lakoff, George 1972 "Hedges. A Study in Meaning Criteria and the Logic of Fuzzy Concepts", in Papers from the eigth Regional Meeting Chicago Linguistic Society, April 14-16, 1972. Chicago: Chicago Linguistic Society, 183-228. Lausberg, Heinrich 1963 Elemente der literarischen Rhetorik. 3. Auflage, München: Hueber. Leech, Geoffrey N. 1983 Principles of Pragmatics. London: Longman. Markkanen, Raija 1989 "Hedging and Modality", Jäntti, A. (ed.): Probleme der Modalität in der Sprachforschung. Jyväskylä: Universität Jyväskylä, 137-148. Markkanen, Raija and Schröder, Hartmut 1989 "Hedging as a Translation Problem in Scientific Texts", in Lauren, C. and Nordman, M. (eds.): Special Language. Clevedon/Philadelphia: Multilingual Matters, 171-179. Nussbaumer, Markus 1991 Was Texte sind und wie sie sein sollen. Ansätze zu einer sprachwissenschaftlichen Begründung eines Kriterienrasters zur Beurteilung von schriftlichen Schülertexten. Tübingen: Niemeyer. Pätzold, Jörg 1986 Beschreibung und Erwerb von Handlungsmustern. Beispiel: Rezensionen wissenschaftlicher Publikationen. Berlin: Akademie der Wissenschaften der DDR/Zentralinstitut für Sprachwissenschaft. Reinert, Gudrun M. 1990 Kontrastive Untersuchung von sprachübersetzungswissenschaßlichen Rezensionen EnglischDeutsch. MA Thesis. Saarbrücken, Universität des Saarlandes. Ripfel, Martha 1989 Wörterbuchkritik. Eine empirische Analyse von Wörterbuchrezensionen. Tübingen: Niemeyer. Schreckling, Jürgen 1985 Routine und Problembewältigung beim Untenichten. München: Profil. Schweickard, Wolfgang 1992 "Die sprachwissenschaftliche Rezension als Forschungsgegenstand (am Beispiel rumänizistischer Rezensionen des 19. Jahrhunderts)" in Ernst, G. et al. (eds.): Akten des Kolloquiums 'Die rumänische Sprache im 19. Jahrhundert', Regensburg26.-2S. Aprä 1990. Tübingen: Niemeyer, 209-211. Vendler, Zeno 1980 "Review of Peter Cole (ed.) (1978): Syntax and Semantics. Vol. 9: Semantics. New York: Academic Press", Language 56,209-211.

IV. Cross-cultural Aspects of Hedging

WILLIAM GRABE and ROBERT B. KAPLAN

On the Writing of Science and the Science of Writing: Hedging in Science Text and Elsewhere ...Simple thoughts whose full import is easily grasped are indeed often best transmitted as spoken word....Scientific observations and deductions rarely fit in this category, however. In the first place, the truth of a scientific assertion is usually a function of thepresicion with which it is stated. Subtle details and equally subtle implications are what makes scientific statements powerful. The spoken word is altogether too spontaneous, often too hastily formulated, and certainly too fleeting to be acceptable as the sole carrier of ideas of such precision and subtlety....There is a second important reason why the spoken word is of limited value for scientific and technical communication....If all scientific communication were verbal the vast majority of results would inevitably be ignored, forgotten, or, at best, housed in inaccessible corners of the mind [because of the huge volume of information] ....Finally, there is a third factor to consider. For results to be regarded as truly scientific they must be made readily available for close, repeated inspection...over time and throughout the world....It goes without saying that scientists need to be skillful readers....The often ignored corollary of this assertion, however, is that scientists are obliged to be skillful writers. Only the researcher who is competent in the art of written communication can play an active and effective role in contributing to science... (Ebel et al. 1987, 4-5).

1. Introduction

It is our understanding that the book from which the above quotation is taken is perceived as a good, standard textbook for helping scientists write scientific articles. It is, one may say, influential. There is a certain irony in its broad acceptance. The claims made in the introductory section, quoted in part above, are interesting and bear some examination. The claim is made, right from the start, that spoken language is incapable of being "the...carrier of ideas of...precision and subtlety" but rather is useful largely as a carrier of "simple thoughts whose full import is easily grasped." Few linguists would be inclined to accept such a judgment. While most linguists recognize an important distinction between spoken and written language, the difference is probably perceived as occurring along another dimension (or along other dimensions). There are studies of relatively "unplanned" spoken dis-

152

IV. Cross-cultural Aspects of Hedging

course as opposed to relatively more "planned" written discourse (e.g., Ochs 1979). But to claim that spoken language can only convey simple ideas ignores most of the real uses of spoken discourse in the contemporary world. Surely Walter Cronkite and Dan Rather would deny that the nightly news consists of "...simple thoughts whose full import is easily grasped..."; surely most university professors, who lecture on everything from anthropology to zoology, or who advise students orally on the students' theses and dissertations, would find it difficult to have their oral lectures and advice so characterized. But aside from the exaggerated claim, there is another aspect of written communication which, we will argue, tends to set it apart from spoken communication; in the quotation (above) note the use of "hedges" (marked in added bold and italics) even in this relatively simple text. There is little disagreement over the importance of writing to science. Were it not for the capability of written language to permit retrieval of information in precisely the same form over time and space, modern science (with its objective of developing noetic control over nature) would not be possible. But it is a gross over simplification to claim that writing is important to science because only writing can be "...the...carrier of ideas of precision and subtlety..." It is hardly worth the effort to demonstrate that all written statements in science are not necessarily profound, precise, or subtle; as a corollary, it is hardly worth the trouble to demonstrate that scientists sometimes write (and sometimes even speak) impenetrable drivel. It is the case that all spoken language is not "simple" and "easily grasped." There is an astonishing arrogance in the claim, in the implication that the scientist is some sort of intellectual giant different from ordinary men. (See, e.g., the novels of C. P. Snow.) But it is not arrogance that is the issue here; rather, it is the perception of the role of language. Several researchers have demonstrated that science writing is not at all like the claims that are made for it. (For a summary, see Kaplan and Grabe 1991.) The writer(s) of the quoted passage claim that "...if all scientific communication were verbal the vast majority of results would inevitably be ignored, forgotten, or, at best, housed in inaccessible corners of the mind [because of the huge volume of information]." On the face of it, the claim is easily dismissed. Scientists do talk together, and the literature suggests that the real science - warts and all, with false starts, mistakes, misdirections, failures takes place orally between cooperating scientists in the laboratory. What appears in print is, at best, a distillation, carefully selected, of what actually happened in "the experimental environment." The claim of the writer(s) of the quoted passage that "...for results to be regarded as truly scientific they must be made readily available for close, repeated inspection...over time and throughout the world..." is equally silly. The evidence suggests that a great deal of scientific text is never looked at again after it appears. Indeed, it is the few statements which gain acceptability as incontrovertible "facts" that count, not because they are repeated, but because they become subsumed as basic presuppositions in the writing of subsequent scientists. More importantly, there is an implication in the claim that experiments are easily replicated. The reality is that many experiments are not replicable at all because the text in which they are reported does not provide enough data to permit replication.

W. Grabe and R. Kaplan, On the Writing of Science

153

The evidence suggests that scientific writing, no doubt constrained by a number of important conventions, really consists of text which is intended to "sell" an idea (especially in the kinds of writing characterized in grant proposals). The scientific article positions the author(s) in relation to the dominant paradigm, elicits support (or the semblance of support) from recognized figures in the context of that paradigm, and tries to establish the claim that the writer is exploring an important idea — one that has not previously been adequately explored — that the writer's results contribute significantly to the dominant paradigm, that the results are interesting and relevant, and (in the case of the grant proposal) that they are worthy of financial support.

2. Hedging: Historical Development and Cunent Uses

Before undertaking a detailed discussion of hedging as a phenomenon of science writing, the probable origin of hedges in scientific writing deserves some passing comment. In the setting of early Restoration England there was no one solution to the problem of knowledge which commanded universal assent. The technology of producing knowledge had to be built, exemplified and defended against attack. The categories of knowledge and their generation which seem to us self-evident and unproblematic were neither self-evident nor unproblematic in the 1660s. The foundations of knowledge were not matters merely for philosophers' reflections; they had to be constructed and the propriety of their foundational status had to be argued. The difficulties that many historians evidently have in recognizing this work of construction arise from the very success of that work: to a very large extent we live in the conventional world of knowledge production that Boyle and his colleagues amongst the experimental philosophers laboured to make safe, self-evident and solid (Shapin 1984, 482).

Shapin's ( 1984) contribution to an understanding of the evolution of scientific prose lies in the explication of the means by which the original British empiricists, led by Robert Boyle, attempted to assure the credibility of both their philosophy of science and of particular empirical results. Shapin points out that Boyle integrated three technologies of knowledge-production: the material, the literary, and the social. Through this integration, Boyle created an epistemological basis for "natural" experimental philosophy by explicating experimental "matters of fact." To achieve his objectives in terms of the literary strategy, Boyle advocated: i) ii) iii)

deliberate prolixity, the use of modesty and caution, and clarity

In the Proemial Essay (which serves as a general preface to the body of his experimental work, 1660) Boyle writes: ...in almost every one of the following essays I...speak so doubtingly, and use so often, perhaps, it seems, it's not improbabk, and such other expressions, as argue a diffidence of the truth of opinions I incline to, and that I should be so shy of laying down principles, and sometimes of so much as venturing explications... (cited in Shapin 1984, 495).

154

IV. Cross-cultural Aspects of Hedging

Thus, in laying down the foundation for the contemporary scientific article, Boyle urged a modesty and caution which assured the use of the hedge as an integral part of the writing of scientists - a phenomenon which has persisted into the end of the 20th century. Hedges, at least in American English, are taken to be markers of uncertainty, though they may also serve the needs of politeness (see, e.g., Prince et al. 1982). Uncertainty may be a fundamental characteristic of scientific inquiry; surely it is common enough in scientific texts. The motivation for expressing weak form negation with stronger pragmatic understanding essentially constitutes the use of hedging or of a politeness strategy (e.g., as true hedging: Some of the apples are ripe, and possibly all of them are. - that is, Some of the apples are ripe and/but possibly none of them are; as politeness strategy: I don't think I can come. - that is, [I think] I can't come). Harnish (1976, 360) claims that "the speaker wants to communicate a certain belief...without saying it, as if he might want to leave the Ί didn't say that' bridge unburned." Seventeenth and eighteenth century scientific writing, both preceding the important influence of Boyle and in the transitional period during which Boyle's ideas took hold widely, was largely epistolary and was entirely dominated (indeed, it was so dominated right into the nineteenth century) by the code of the gentleman. Virtually all early British scientists belonged to the upper classes (or were supported by those classes who had a profound influence on the style of reporting scientific observations), and wrote according to the literary traditions ofthat class. These early scientific "articles" were replete with polite forms, which not only served to establish solidarity with the scientific audience but which also served to preclude direct attacks on intellectual adversaries, requiring a special politeness-marked form of hedging. Thus, for example, the following texts [(i) by J. Ray, "A Letter written to the Publisher by the Learned Mr. Ray..." (1675) as well as the following anonymous text (ii) from 1675] illustrate the style: (i)

(ii)

Sir, I was much pleased and satisfied with the ingenious conjecture I found in your Transactions of May last, pag. 310. concerning the swimming Bladders of Fishes; and perswade my se[l]f that the Author thereof hath hit upon their true use, viz. to sustain or keep them up in any depth of water. For 1. it hath been observed by some, and I find it in Mr. Willughbyes general notes of Fishes... (cited in Atkinson 1993, 141). Honoured Sir, Having long understood the freedom of address, you have allowed to any candid and unprejudic'd persons, whose general good wishes to the Commonwealth of Learning made 'em ambitious to impart something, whereby they may promote the Empire of Man above other Creatures; I flatter'd my self, it would be no unpardonable presumption, to communicate some of my thoughts unto you, on such a subject, as that the usefulness of the matter may keep me from seeming altogether impertinent, and the smalness of my performance, argue the greatness of my desire to serve you. This I resolved to do by way of Animadversions on the two last Philosophical Transactions; and though I performed nothing more, I hop'd at least, my Observations may do them the kindness, as shades intermixt with light, to render them more conspicuous. In hopes of this, I shall draw up my thoughts under the general Tide of Some Philosophical Observations on some passages in the two last Months Transactions, viz. for Decemb. and January... (Cited in Atkinson 1993, 152).

Another relevant area, restricted to spoken discourse, lies in the use of silence. It is mentioned here in passing because in certain cultures it constitutes a man-

W. Grabe and R. Kaplan, On the Writing of Science

155

ifestation of hedging. When in doubt, it is sometimes best not to say anything. The duration of such pauses varies significantly from culture to culture. English speakers have difficulty dealing with silence; thus, they try to fill gaps in conversation with some sort of noise. Sometimes, they use simple verbalized pauses, but most often they fill space with restatement. Pauses in English spoken discourse rarely exceed one second (in part because for English speakers a pause of more than one second signals the willingness of the speaker to give up the floor). Japanese speakers, and Finnish speakers, on the other hand, tolerate extended pauses without difficulty. There is documented evidence of pauses running to ten minutes and longer in Japanese negotiation discourses (without giving up the floor). Obviously, however, pause length is not a concern in written discourse (except perhaps in written fictional dialogue). Hedging in written text is commonly defined in terms of a limited number of semantic and syntactic devices. But the matter turns out to be more complex. In our initial analyses of the texts used in this study, it became evident, as well, that any discussion of hedging - as qualification of evidence or evaluation, or as indecision or purposeful ambiguity - could not be separated completely from issues of emphasis, strong assertion, and epistemic commentary. Aside from the fact that qualifications and critical evaluations appear across the full range of lexico-syntactic forms, they also often shade into qualified emphasis and assertion, and on to assertion without qualification. For this reason, we have included in our analysis not only linguistic signals of hedging but also those of emphasis and assertion, recognizing that the two concepts together form a semantic cline; in cases in which both concepts are referred to, we sometimes use the term evidential*, drawing on Chafe and Nichols (1986). To limit the analysis to hedging alone appeared to be somewhat artificial in that deciding on which side of the line a linguistic feature would fall (as hedge or as emphatic) was a fairly subjective decision in a number of cases. For example, the word, just, can appear in sentences which indicate a minimal activity or effort was expended without any specific commitment to the actual amount; conversely, the word can be interpreted as an emphatic in the same sentence when combined with a negative, indicating that the plan was unacceptable. iii) iv)

The plan provided just enough resources. The plan just did not provide enough resources.

3. Current Research on Hedging

Hedging (and related emphatics) as analyzed in the present study builds upon frameworks proposed by Vande Kopple (1985) and Crismore (Crismore and Farnsworth 1990, Crismore and Vande Kopple 1988; see also Barton 1993, Biber and Finegan 1988, Butler 1990, Myer 1989; 1992). These frameworks provided sufficient flexibility to consider a range of linguistic marking of hedging, qualifying, commenting, and evaluating. Earlier discussions of hedging, such as found in Quirk, et al. (1985) have provided important suggestions for actual lexical forms which signal hedging and emphasis; the exposition, however,

156

IV. Cross-cultural Aspects of Hedging

does not as readily lend itself to a classificatory system for text analysis as do the above frameworks. In the various analyses of adverbials alone in Quirk, et al. (1985), consistent textual candidates for hedging and emphasis are examined across a range of approximately 100 pages (1985, pp. 543-640) and are treated under separate categories of subjuncts, disjuncts, and conjuncts. The framework proposed by Crismore and Farnsworth (1990), intended to examine metadiscourse categories in texts, covers: (1) (2) (3) (4)

code glosses - definitions and explanations for terms, modality markers - hedges, emphatics, and attributives, attitude/evaluative markers - ( e.g., it is regrettable that...), and commentaries - constraints on the discourse situation or comments on the work.

While much work on hedging limits itself to modality markers, we found that texts often present qualifications and hedges in some form in all of the four above categories. On occasion, a definition or explanation will be carefully understated so as not to over-commit the author; similarly, commentaries are often introduced with reporting verbs that, themselves, indicate a degree of qualification (see also Myers 1989,1992 for discussions pertaining to categories 3 and 4). Specific examples of our adoptions of this framework are discussed below. In related research on hedging and evidentials, a number of variations have been proposed, many of which offer additional suggestions for the study of hedging. Barton (1993) examined evidentials in 100 "Points of View" editorial essays from the Chronicle of higher education (the weekly newspaper of The American Council on Education) with 100 undergraduate-student essays written for a university writing-proficiency requirement. Because the corpus comprised academic essays, she employed both linguistic and rhetorical indicators of evidentiality. Linguistic features analyzed included categories of modals, sentenceinitial conjuncts (e.g., but, however), reporting predications (e.g., we suggest that...), prepositional phrases (e.g., of course), and degree-of-reliability evidentials (e.g., probably, generally, virtually). Rhetorical features included the following concepts: problematization (stating a problem to be discussed), persona (use of interactive pronoun reference), citation (appeal to authority), argument (claims, supports, addressing counter-arguments), and epistological stance (the assumed definition of what counts as knowledge). Her qualitative analysis emphasized the inabilities of many students to use rhetorical features of evidentiality to shape an effective essay. Butler (1990) examined a corpus of scientific texts drawn from physics, botany, and animal physiology (approximately 32,000 words) in a careful analysis of modal verb use. Taking modal verbs as markers of the weighting of evidence and the drawing of conclusions, he compared results of his corpus with the analyses of Coates and Leech (1980), Hanania and Akhbar (1985), and Huddleston (1971). He also examined the co-occurrence of modals with particular types of main verbs and the distribution of modals across sections of the scientific article format, and he commented on difference in modal usage across different scientific disciplines-pointing out possible systematic variation within the general category of scientific writing. His findings on modal verb usage, and relevant comparative information from the additional research sources are presented in Table 1.

W. Grabe and R. Kaplan, On the Writing of Science

157

Table 1: Frequencie i of Modais p>er 1000 words Modal

Butler

Huddleston*

Coates & Leech

May

3.33

3.80

1.30

.53

.49

.67

1.93

3.43

Might

Can Could Will Would Shall Should Must Need Ought TOTALS: Notes: b

.47

.81

1.31

.65 .37 .21 .68 .06 .00

1.38 1.12 .28 .62 .86 .04 .01

2.19 1.77 2.70 3.15 .27 .91 1.01 .04 .07

9.96

11.59

14.07

Huddleston (1971) is a 135,000 word corpus of scientific texts Coates and Leech (1980) is the 1,000,000 word Brown corpus

Hanania and Akhbar (1985), working with a 95,600 word corpus of 20 Nonnative-English student MA Theses in Biology, Physics, and Chemistry, reported a much lower overall modal-verb use level of 6.99 modals per 1,000 words. In addition, Butler supports a finding which appeared in both Huddleston (1971) and Hanania and Akhbar (1985) with respect to variation in modal uses across scientific disciplines. All three studies note a higher use of modals in Physics writing than in Biology writing. Moreover, Physics writing makes much greater use of the modals: can, could, would, and should; Biology, on the other hand, makes greater use of may and might. Why this consistent pattern should appear remains an open question. It does, however, point out an interesting difference in hedging and evaluative signals between writing in two scientific disciplines. Butler also examined his corpus for differences in modal uses across sections of the text. Not unexpectedly, he found that modals were used much more frequently in introduction sections (9.74 modals per 1,000 words) and discussion sections (12.51 modals per 1,000 words) than in results sections (3.14 modals per 1,000 words) and methods sections (.65 modals per 1,000 words). Finally, Bulter's comparison of modal use with the results obtained by Coates and Leech suggests that scientific prose does not make as extensive use of modal verbs as does a much wider range of other text types. Crismore and Farnsworth (1990) examine specifically the extent to which academic science writing and more popular science writing reveal differences in the types of hedging and emphasis, as noted above. They compared two articles written by Steven Jay Gould (the professional one co-authored with David Woodruff). The two texts totaled approximately 19,000 words. In their qualitative analysis, Crismore and Farnsworth demonstrated, through a series of examples, how the two articles differed in their signalling of metadiscourse, and in categories of hedges, emphatics, attitude markers, modality markers, and author commentary markers. Of specific concern to the present study, they reported quantitative results for the following categories (reported in Table 2, below):

158

IV. Cross-cultural Aspects of Hedging

Table 2: Variation across two genres in hedging and emphatics (All numbers are per 1,000 words)

Note:

a

Professional article

Popular article

Modals Hedges Emphatics Evaluatives

10.7 7.8 2.9 2.3

9.4

TOTALS

23.7

21.9

7.3 2.1 .1

Evaluatives are exemplified as follows: it is regrettabk that, fortunately, most importantly, notabk, with littk justification, etc.

The above results conform generally with Butler's results for overall modal use. For example, Table 2 shows professional science to have 10.70 modals per 1,000 words. This matches well with Butler's total of 9.96 modals and Huddleston's total of 11.59 modals. In addition, their findings suggest that, for the categories discussed in Table 2, there is little difference in the extent of overall hedging and emphatic marking between professional and popular science writing. Crismore and Famsworth offer an extensive argument to the effect that professional science writing, as opposed to popular science writing, contains much more overall metadiscourse as compared with other categories of meta-commentary which are included in Table 2. It will be the purpose of the remainder of this study to analyze a specific corpus of texts and examine the textual features which typically mark hedging, emphasis, and other signals that indicate the degree to which an author asserts or qualifies the "truth" of the ideational content of the text. A primary concern of this paper is to situate and examine scientific writing within a somewhat larger frame of other text types; to do this, the current analysis will compare science writing, in both professional and popular forms, to three other functional written genres: news editorials, annual business reports, and fictional narratives. The three additional text types provide a ground against which to explore the extent of evidential marking (Chafe and Nichols 1986) in science texts. It seems that the presentation of results only from various categories of scientific text would be relatively meaningless without a comparison against similar phenomena in other varieties of text. The finding of the present analysis of texts will then be compared with similar analyses by Barton (1993), Butler (1990) and Grismore and Farnsworth (1990).

4. Method 4.1. Corpus

The corpus of 50 texts for this study1 was drawn from a somewhat larger corpus of texts used by Grabe (1984, 1987) to examine patterns of linguistic variaThe full corpus may be had at cost from William Grabe, Department of English, Northern Arizona University, Flagstaff; AZ 86011 USA.

W. Grabe and R. Kaplan, On the Writing of Science

159

tion across a range of expository text types. Five text types, composed of ten texts each, were chosen through a set of random selection techniques outlined in Grabe (1984). Each text is approximately 470 words in length and represents a coherent text segment. The five text types chosen include: professional natural science, popular natural science, newspaper editorials, annual business reports, and fiction narratives.

The professional and popular science categories represent an extension of the analysis presented in Crismore and Farnsworth (1990). The other three categories represent a wider context by which to examine the frequencies of hedging and emphatics in science texts. Both the news editorials and the business reports are text types which would seem to invoke extensive use of hedging and emphatics, though for different reasons: editorials seek to evaluate events in a certain light and to persuade readers; annual business reports promote a positive image of the corporation and emphasize bright future prospects. Narrative was include as a text type both because it is far removed from the functional purposes of the other text types, all of which have needs for employing hedging and emphatics, and because it is not generally included in the larger category expository text. As an example of the sampling procedures, the professional science texts (n=4,905 words/20.83%) were chosen by referring to the University of Southern California Serial Journal Index in its Doheny Library. Journals beginning with the words "Journalof... "were chosen for ten different natural science disciplines. Natural Science disciplines were chosen from the set of disciplines listed on the Reader Service card that can typically be found in the Journal Science. After locating ten representative journals, text segment selection was randomly decided upon by turning to page 40 of the current volume. If no coherent segment of at least 400 words was available, the next target page was page 70 of the volume (then page 110, 150, 190, etc.). This procedure continued until a coherent segment was found. Further details of the selection process may be found in Grabe (1984). The texts used in the category included texts from the following ten journals: Journal Journal Journal Journal Journal Journal Journal Journal Journal Journal

of Anatomy, of Materials Science, of Applied Physics, of Physical Oceanography, of Ecology, of Physics: Mathematics and..., of Endocrinology, of Spacecraft, of General Microbiology, of Statistical Physics.

The other four categories are described here more briefly, though similar careful types of sampling procedures were used. Popular Science texts (n=4,669 words/19.83%) were chosen by selecting ten journals and magazines listed in

160

W. Cross-cultural Aspects of Hedging

the Reader's Guide to Penodical Literature, and which were available in the University of Southern California library. News editorial texts (n=4,709 word/20.01%) were selected from two newspaper - The Los Angeles Times, and The Christian Science Monitor. Business Reports (n= 4,616 words/19.61%) were chosen from among corporations listed in the Dow Jones list of leading industrial, transportation, and utility companies, and which were available in the University of Southern California School of Business Library. Fictional Narratives (n=4,638 words/19.71%) were chosen by taking half from the Brown Corpus and half from then current best sellers based on an informal poll of salespersons in two Los Angeles bookstores. The total corpus for this study consists of 23,537 words. 4.2. Decision procedures Determinations of what counted as a hedge or an emphatic were based on a number of general considerations. The research reported above, in combination with initial readings of the present text corpus, revealed six ways in which hedging marking is used in this corpus: (1) (2) (3) (4) (5) (6)

hedging qualifies the exactness of measure or commitment to a measure taken; hedging identifies the degree of commitment to the truth of some statement; hedging qualifies the commitment to a future outcome or plan; hedging allows the writer to remain uncommitted to specific details and specifications that may not be necessary; hedging conditions the likelihood or prediction of an outcome, and hedging, in more literary texts, takes place through metaphoric extensions and analogies which do not require detailed explanation or specification, or are not readily explainable.

Emphatics also were identified according to a number of the above uses, such as close exactness of measurement, high degree of commitment, etc. In addition, emphatics were noted for forms which actually emphasized the information presented through unusual and extreme descriptors, through unusual commentary, and through forms which evaluated while permitting no qualification. As discussed above, hedging and emphatics are here treated as points on a semantic cline of evidentiality, and thus, both types of linguistic markers need to be taken into consideration. Below are some example sentences which indicate the types of decisions made in developing our counts. In the first example, both "...only realistic..." and "...believe..." function as hedges; they do not state in fact that the formulation is in error, and they allow for possible objections to the writer'(s) position. The phrase "...very seriously..." suggests, at the same time, an emphasis on the lightness of the critical position taken in the sentence. • It is only realistic to believe that the formulation is very seriously in error (ProSci #2).

In the second example, there are a set of four hedges (in bold) and one modal (in italics). There are no emphatics. We did not count common clausal transition markers (e.g., however, but, if#) as a category of hedging or emphatics. This

W. Grabe and R. Kaplan, On the Writing of Science

161

sentence provides an example of a transition marker (However..) that could have been counted as a hedging marker. • However, even for the atmospheric boundary layer in which convection is believed to dominate usually, it is being recognized (Prost, et al., 1982: Maniuns, 1982) that the contribution [of an interfacial velocity jump] may often be ignored (ProSci #2).

The third example is taken from popular science and indicates both a number of hedges and emphatic markers. Emphatics include "...major criticism..." and "...unusually low..."; hedges include "...attributed this..." and "...acknowledged other factors...." There is also a modal (may). • A major criticism of the Vietnamese data was its unusually low rate of miscarriages and birth defects in unexposed populations. The Vietnamese attributed this to their underdeveloped status with few industrial pollutants and synthetic compounds in their environment but acknowledged other factors, including genotypic variation, may be involved (PopSci #5).

The fourth example, from news editorials, indicates the pervasiveness of qualification, emphasis, and evaluation in this text type. There are two emphatics ("...indeed..." and "...no question..."), one hedge ("...be assumed...") and five modal verbs (could, is to, will, need not, would). • Indeed, it [peacekeeping force] could prove a help in the long run. There is no question that U.S. troops are involved in a hostile situation and, if the president is to manage this latest Mideast crisis effectively, he will need congressional cooperation. It need not be assumed that such cooperation would be widiheld (edtl #9).

Two final examples, taken from annual business reports, reveal the very different signalling dynamics in reports intended to sell the company's present record and to promote a positive future. In some cases the extensive use of emphatics is straightforward, as companies highlight their most successful products. In other cases, reports attempt to place weak performances in the best light possible. • In large businesses serving markets whose growth is slowing - autos and steel, for example - sustaining a leading market position wiU no longer take the heavy capital investments of the past. Instead, funding at reduced levels can go mainly to sustain the low-cost leadership of these businesses (BusRept #2). • To offset die heavy yield dilution caused by such low fares, TWA created a clearly differentiated service product, Ambassador Class. It features superior comfort at a modest premium fare over Coach, yet well-below first-class fares. This new service first demonstrated its strong appeal, especially to business travelers, on TWA's transatlantic flights (BusRept #6).

In the first segment above, the report says it will dump unprofitable businesses, lay off workers, and invest less money in those parts of the corporation in the future. The segment has one emphatic ("...no longer..."), three hedges, and one modal as part of the means to soften the intent of the passage. In the second segment, TWA is explaining how it intends to turn around its financial

162

IV. Cross-cultural Aspects of Hedging

losses, hoping to achieve its intent by using many positive emphatic markers! In addition, two hedges ("...heavy yield dilution..." and "...modest...") obscure the problems - the airline is losing money, and it will charge more for its tickets. 4.3. Counting procedures All texts were read, and four categories were coded. Each text was first examined for modal verb (and periphrastic counterparts). Texts were then marked for verbal hedging (hedging involving main verbs), non-verbal hedging (all other markers of hedging), and emphatics (all lexico-syntactic markers). The separation of counts into these four categories allowed comparisons for modal use, hedging and emphatics. The separate category for verbal hedging also reflects the variety of verbs of reporting (i.e., stated, claimed, indicated, etc.), mental verbs (i.e., believed, thought, considered, etc.), and evaluative verbs (i.e., is dismissed, is confined to, presumed, made progress toward, etc.). The various types of texts also highlighted distinct lexical items as hedges and emphatics because of the differing functional purposes for qualifying or emphasizing information. For example, in business reports, verbs such as plans to, began to, enables, point to future corporate activities which are intended to persuade investors to keep their money in the corporate stocks. These same verbs would not typically be considered verbs of hedging in the contexts of the other text types. Typical linguistic markers of verbal hedging included the following verbs: claim think of appear

assume assert seem

say believe argue

report finds agree to

suggest are convinced slip back from

Typical linguistic markers of non-verbal hedging included the following: sometimes most of the likely that possibility is capable of approximately with positive assessment

typically evident that nearly no more than usually the hope that consideration of

modestly in large part the so-called if that is what it is

Typical linguistic markers of emphatic expression included the following: of course virtually all raw(power) insist that unrelenting clearly sizzling unrelenting not willing to accept

abysmal quality surely the best Malty at rock bottom

fmatty quite simply incompatible excessively in fact indeed no question that very mysteriously

There is no question that some of the choices made in creating these categories reflect subjective decisions. It is virtually impossible to cross a wide range of text types and not be forced to make decisions about linguistic markers of hedging and emphasis which will vary somewhat from text type to text type. Moreover, lexical items not normally treated as qualifying markers become strong markers of hedging in specific textual contexts, particularly in newspaper editorials; e.g.:

W. Grabe and R. Kaplan, On the Writing of Science

on theface of it become inaeasingly blurred aside from if history is a guide nourished the ground for continuing conflict

163

to have meaning be premature dubious bkssing

4.4. Analysts The four categories of hedging and emphatics discussed above were counted, and counts for each category in each text were normed to be equivalent to texts of 1,000 words in length. All quantitative results below are presented as normed data. Each text was given a frequency count for modal verbs, verbal hedging, non-verbal hedging, and emphatics. In addition, modal verbs were counted separately in order to provide a comparison with Table 1 above, summarizing the work of Butler (1990). No inferential statistics were used. The corpus is not sufficiently large for generalizing claims on its own (without comparison with other results); furthermore, the specific counts for hedges and emphatics were not checked for inter-rater reliability and the data should, accordingly, be treated with some caution.

5. Results

Modals: Table 3 below presents the results of the modal-verb uses by text type. The modals are listed in the same order as in Butler (1990), as presented in Table 1 above. Table 3: Results of modal counts across five text types Modal

Prof Sei

Pop Sei

Bus. Kept

May Could Will Would Shall Should Must Need Ought Won't Have to Be able to Be to

01.75 00.41 01.98 00.56 01.34 00.46 00.61 00.76 00.38 00.00 00.00 00.00 00.37 00.00 00.19

01.99 00.00 02.76 00.64 01.44 0128 00.00 0022 00.45 00.00 00.00 00.00 00.63 00.60 00.00

00.00 0023 01.30 0024 03.96 0023 00.00 00.42 00.00 00.00 00.00 00.00 00.00 0020 0020

TOTALS

08.82

10.00

06.78

Might

Can

Editls

Narr

TOTALS

01.71 00.00 02.47 01.87 05.44 04.55 00.00 01.73 02.68 01.05 00.00 00.43 00.39 00.64 01.47

00.00 00.00 01.06 03.19 00.63 03.81 00.00 00.00 0022 00.00 00.00 00.00 01.07 0021 0022

05.45 00.64 09.57 06.50 12.81 10.33 00.61 03.13 03.73 01.05 00.00 00.43 02.46 01.65 02.08

24.42

10.42

XXXXX

Notes: , Counts based on a norm of modals per 1,000 words The totals in the right-hand column are not averaged comparisons but sumtotal comparisons. They are meant only for rough comparison across modals; they are not indicators of average modal use across text types. Averages per 1,000 words could be derived by dividing by five.

164

IV. Cross-cultural Aspects of Hedging

Total counts of verbal hedges, non-verbal hedges, and emphatics, in combination with totals from the modal counts are presented in Table 4. All totals are normed to number of occurrences per 1,000 words. Table 4: Frequency totals for modals, hedges, and emphatics across five text types ProfSci PopSci BusRept Editls Narr 08.82 Modals 10.00 06.78 24.42 10.42 Verbal Hedges 05.91 08.78 09.13 10.75 02.75 Non-verbal 10.94 Hedges 22.36 14.20 23.03 07.40 TOTAL 25.67 41.14 HEDGES 30.11 58.29 20.77 07.67 06.51 19.54 Emphatics 22.81 14.23

A further result of Table 4 is the ratio of hedges and evidential to total words in the text for each text type. For Professional Science [ProfSci], the ratio is approximately 25 hedges per 1,000 words which translates to one hedge every forty words of text; • for Popular Science [PopSci], the ratio is approximately one hedge every 25 words; • for Business Reports [BusRept], the ratio is approximately one hedge every 33 words; • for Editorials [Editls], the ratio is approximately one hedge every 17 words, and • for Narratives [Narr], the ratio is approximately one hedge every 50 words.

For evidentials (combined hedges and emphatics) the ratios are higher: the the the the the

category Professional Science has one evidential every 30 words; category Popular Science has one evidential every 21 words; category Business Reports has one evidential every 19 words; category Editorials has one evidential every 13 words, and category Narratives has one evidential every 29 words.

6. Discussion

The variations in hedging and emphatics revealed in this corpus of texts lead to a number of relatively safe general conclusions. First, it is readily apparent that hedging and emphatic signalling occur in quite different configurations and different frequencies in different types of texts. Also of interest is the way that lexical forms and phrases can take on a hedging function because of the specific functional intent of a given text; this is consistently apparent in the annual business reports and in the newspaper editorials. Such a conclusion, however, points out a potential problem with automated counts of such phenomena as hedging and emphatics in that automated counts presume a hedging role for certain linguistic forms and disregard other linguistic forms which take on hedging and emphatic roles for specific contexts.

W. Grabe and R. Kaplan, On the Writing of Science

165

Second, the comparisons of modals, hedges, and emphatics across text types points out that all text types make fairly extensive use of these devices - no text is neutral or straightforwardly objective. At the same time, the results support the intuitively held notion that editorial writing is strongly persuasive and evaluative, as this set of texts made the greatest use of modals, hedging, and emphatics in general. Third, the specific results of the modal counts are fairly consistent with the results of Butler (1990) and Crismore and Farnsworth (1990) on a general level. Butler notes that his corpus shows 9.96 modals per 1,000 words of text; Crismore and Farnsworth noted approximately 10 modals per 1,000 words of text; the present results place modals in the combined science categories (professional and popular) at approximately 9.4 modals per 1,000 words. Our results also support the implication, noted in the commentary of Table 1, that modal usage is somewhat less frequent in science text than in certain other text types. On more specific levels, there are notable differences across the three studies. The frequency distributions across the different modals in science texts do not match well with Butler's results as shown in Table 1, though they are more in line with results reported by Coates and Leech (1980) for a more general corpus. These results for science texts suggest some qualification to the assertion in Butler that, ...there are certain uses of may and can that are characteristic of scientific and possibly other academic texts, while some of the meanings of will, would, could, and should are less appropriate in scientific writing than in some other kinds of written English (1990, 144).

Our results do not support this claim to the extent that the distribution of modal verbs in our corpus does not match the above assertion. To explore the science text data a bit further, we examined the section of the articles that each segment was drawn from. Four were drawn from introductions, four from methods sections, and two from discussion/conclusion sections. This moderately even distribution would suggest that the text sample constitutes a reasonable representation of science writing overall, though we certainly recognize that a sample of ten texts does not provide an adequately representative sample. The different results that we have achieved, however, do place into question the generalizability of some of the specific modal useresults reported in Butler (1990). Fourth, our results, on a general level, support the study by Crismore and Farnsworth (1990), demonstrating that professional science writing and popular science writing are not very different in terms of their uses of modals, hedges, emphatics and attitudinal markers. In detail there are, however, some differences. In their study, Crismore and Farnsworth found slightly more frequent uses of modals, hedges, and emphatics in professional science texts; we have found the reverse to be the case. However, the differences across the two studies are quite small for most hedging and emphatic uses. The only serious difference was found with non-verbal hedging. This difference may be the result of the much more inclusive criteria that we adopted for this category. Overall, we confirm their finding that modals, hedges, and emphatics do not vary greatly between professional and popular science writing. In this one area, at least,

166

IV. Cross-cultural Aspects of Hedging

one could suggest that students reading and writing from popular science resources will not be misled when they move, later on in their development, to more professional academic texts. Fifth, this study reveals the potential for exploring hedging and emphatic markers across a wider range of texts than science texts or academic texts. At this point, it might be fair to say that the myth of the objective science articles has been dispelled through text discourse research. Perhaps charting the variability of such features in science and academic texts together with their occurrences in a wider range of text types will provide new insights and pose new questions. In turn, these insights and questions may suggest improvements in teaching advanced writing and in our own writing habits. The research explicated here brings into serious question the validity of the kind of textbook for science writers quoted at the start of this paper. It seems to us that the sort of advice offered in that textbook is more likely to mislead science writers than it is to help them; certainly, it misrepresents the realities of science discourse, and it misrepresents the realities of the language of science texts.

References Atkinson, D. 1993 A historical discourse analysis of scientific research writingfrom 1675 to 1975: The case of the Philosophical Transactions of the Royal Society of London. Los Angeles: University of Southern California. Ph. D. Diss. Barton, E. 1993 "Evidentials, argumentation, and epistemological stance." College English 55, 745-769. Biber, D. and E. Finegan 1988 "Adverbial stance types in English." Discourse Processes 11, 1-34 Butler, C. 1990 "Qualifications in science: Modal meanings in scientific texts" in W. Nash (ed.) The writing scholar: Studies in academic discourse. Newbury Park, CA, Sage, 137-170. Chafe, W. and J. Nichols (eds.) 1986 Evidentially :The linguistic coding ofepistemology. Norwood, NJ. Ablex. Coates, J. and G. Leech 1980 "The meanings of the modals in modern British and American English." York Papers in Linguistics, 8, 23-34. Crismore, A. and R. Farnsworth 1990 "Metadiscourse in popular and professional science discourse" in W. Nash (ed.) The writing scholar: Studies in academic discourse. Newbury Park, CA: Sage, 118-136. Crismore, A. and W. Vande Kopple 1988 Readers' learning from prose: The effects of hedges. Written Communication 5, 184-202. Ebel, H. F. et al. 1987 The art of scientific writing: From student reports to professional publications in chemistry and related fields. Weinheim: VCH. Grabe, W. 1984 Towards defining expository prose within a theory of text construction. Los Angeles, CA: University of Southern California. Ph. D. Diss.

W. Grabe and R. Kaplan, On the Writing of Science Grabe, W. 1987

167

"Contrastive rhetoric and text type research" in U. Connor and R. B. Kaplan (eds.): Writing across languages: Analysis ofL2 text. Reading, MA: Addison-Wesley, 115-137. Hanania, E. and K. Akhtar 1985 "Verb form and rhetorical function in science writing: A study of MS theses in Biology, Chemistry, and Physics." ESP Journal 4, 49-58. Harnish, R. M. 1976 "Logical form and implicature" in T. G. Bever, et al. (eds.): An integrated theory of linguistic ability. New York: Crowell, 313-391. Huddleston, R. 1971 The sentence in written English: A syntactic study based on an analysis of scientific texts. New York: Cambridge University Press. Kaplan, R. B. and W. Grabe 1991 "The fiction in science writing" in H. Schröder (ed.): Subject-oriented text: Languages for special purposes and texts theory. Berlin, Nes York: Walter de Gruyter, 199-217. Myers, G. 1989 The pragmatics of politeness in scientific articles. Applied linguistics 10, 1-35. Myers, G. 1992 "In this paper we report...': Speech acts and scientific facts." Journal of Pragmatics 17, 295-313. Ochs, E. 1979 "Planned and unplanned discourse" in T. Givon (ed.): Discourse ana syntax New York: Academic Press, 51-80. Prince, E. F. et al. 1982 "On hedging in physician-physician discourse" in R. J. DiPietro (ed.): Linguistics and the professions. Norwood, NJ: Ablex, 83-97. Quirk, R. et al. 1985 A comprehensive grammar of English. New York: Longman. Shapin, S. 1984 "Pump and circumstance: Robert Boyle's literary technology." Social Studies of Science 14, 481-520. Vande Kopple, W. 1985 "Some exploratory discourse on metadiscourse." College Composition and Communication 36, 82-93.

MINNA-RIITTA LUUKKA and RAIJA MARKKANEN Impersonalization as a Form of Hedging

1. Introduction All language use is interpersonal in the sense that the message expressed is meant for other people to read or hear, and its meaning is always a result of negotiation between the participants. This leads the speaker or writer to choose different communicative strategies1 according to who the audience is. Our assumption is that one of these strategies is that of hedging, through which the speakers or writers can avoid taking full responsibility for or committing themselves fully to the content of the message expressed. The idea of a communicative strategy of hedging makes it possible to include in it much more than expressions that make words "fuzzy" (in LakofPs 1972 sense of the term) or expressions that soften or mitigate utterances. We claim that it also allows the inclusion of the use of impersonalization, i.e. the avoidance of explicit reference to persons or the use of "constructions lacking a specified human agent" (Siewierska 1984, 273). It can be claimed that, for example, by using a passive construction instead of the first person pronoun (It is claimed that the Earth is flat instead of / claim that the Earth is flat) a speaker can avoid taking full responsibility of what s/he is saying. This can be likened to the use of may or perhaps in the same context (The Earth may be/is perhaps flat), which are normally acknowledged as hedges. In both cases the speaker leaves it open whether s/he believes in the flatness of the Earth. We therefore suggest that there is a general strategy of hedging, which consists of different sub-strategies, impersonalization being one of them and the use of expressions like perhaps another. Thus, we are suggesting a kind of hierarchy of strategies with hedging as a superstrategy and impersonalization as a sub-strategy. As a communicative strategy impersonalization comes close to detachment, which is the term Chafe uses for the choice of devices that "suppress the direct involvement of an agent in action" (Chafe 1982, Chafe/Danielewicz 1987). Chafe considers detachment to be particularly a feature of written language, 1

In this context we use the concept 'communicative strategy' to mean a conscious or uncon scious plan which a language user has in order to make his/her message acceptable to the receiver.

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

169

whereas its opposite, involvement, is a feature of spoken language. He also distinguishes three types of involvement: involvement with oneself, involvement with audience, and involvement with the concrete reality of what is talked about. Involvement is manifested, among other things, in frequent reference to oneself and direct reference to the audience. Detachment as a particular feature of written language leads to the avoidance of these direct references and the use of impersonal expressions. Like other forms of hedging, impersonalization can be motivated by two opposing desires. On the one hand it can be seen as motivated by the desire to protect one's own face. By not taking full responsibility of the content of a message a speaker or writer avoids possible criticism on the part of the audience. Therefore, the choice of impersonalization can depend on such factors as the speaker's or writer's relationship to the addressee(s), his/her position vis-avis the addressee(s), his/her personality, etc. Impersonalization, along with other types of hedging, can also be seen as a strategy to save the addressee's face. To quote Brown/Levinson (1987, 190), "one way of indicating that S does not want to impinge on H is to phrase the FTA as if the agent were other than S, or at least possibly not S or not S alone..." This form of negative politeness can also lead to impersonalizing the hearer, i.e. not impinging on him/her with direct address ("you"). Thus, our strategy of impersonalization includes the avoidance of direct reference to the addressee, for example when a speaker giving a talk says As can be seen on the handout instead of As you can see on the handout. Furthermore, we suggest that avoiding direct reference to a third person could be motivated by the desire of not impinging on that person, for example when ones says or writes vaguely // is often suggested that instead of X suggests that. Another explanation for the avoidance of direct reference to third persons could be the speaker/writer's desire to be generally vague, i.e. the same reasons that explain the use of hedging expressions in general. On the whole, then, impersonalization is a way of making one's message more acceptable to the audience and thus to increase its chances of ratification, which is the prime motivation Hübler (1983) gives for the use of hedges. The motivation for impersonalization described above can be applied to both spoken and written discourse. It is often said that scientific writing is prone to impersonalization, that it is in fact the norm in that type of writing, to the extent that a writer actually has no choice but to use it. Impersonalization is said to contribute to the objectivity of scientific discourse. What matters is not the persons involved but the content itself, i.e. the point argued about or the experiment reported on. However, as suggested above, impersonalization, like other forms of hedging, can be seen as resulting from the writer's desire to save his/her own face, to protect him/herself from potential criticism on the part of the readers. Therefore the amount of hedging in a text can depend on such factors as the position of the writer within the field, the potential readership, and even the writer's personality (cf. Markkanen/ Schröder 1992). As a matter of fact, Korhonen/Kusch (1989), who analyzed devices used for reference to self in philosophical texts, found out that the strongest influence on the writer's choice comes from the writer's age and posi-

170

IV. Cross-Cultural Aspects of He dging

tion within the scientific community. Experienced writers who have an established position of authority use direct reference to themselves more than do younger writers. In line with this, Myers (1989) sees the role of hedging in scientific discourse as a way of presenting oneself as "a humble servant of the discipline, showing deference to the scientific community". This explanation seems particularly applicable in the case of young writers.

2. Procedure and data Our basic assumption, then, is that knpersonalization is a strategic choice the speaker or writer makes. S/he has the choice of referring to persons (him/herself, the addressee, other persons) either explicitly or implicitly. When choosing to use explicit reference the speaker/writer is more direct and more open, takes responsibility of his/her opinions or refers directly to the source of information. When choosing implicit reference, the writer/speaker is indirect, avoids taking responsibility and avoids direct reference to the source of information. However, it is important to note that both choices are "personal" in the sense that they contain a reference to person(s), either explicitly or implicitly. Thus, both I daim that χ equalsy and In this report it is claimed that χ equalsy refer to person(s), the first one explicitly and the second one implicitly. In other words, both differ from χ equalsy, which contains no personal reference. Thus, we first draw a distinction between completely impersonal and personal utterances. Secondly, within the personal type we distinguish between explicitly and implicitly personal utterances. And it is the use of the last mentioned utterances, i.e. the implicitly personal ones that we call 'impersonalization'. Our starting point in this article, then, is that the avoidance of explicit personal reference, which we call impersonalization, is a sub-type of the strategy of hedging. The motivation behind its choice is the desire to save face, either one's own, the addressee's or some third person's face. At the same time this strategy can serve as a way of detachment from the topic. We also assume that the use of this strategy is more frequent in written than in spoken language, and that academic writing is an excellent example of the contexts in which it is used. We will therefore look at its occurrence in academic discourse. To be able to prove that its use is more common in writing than in speaking, we will also compare its occurrence in written texts with its use in spoken academic discourse. Another aspect we will focus on is the possible differences in its linguistic realization in languages and in the cultural norms of its use. For this purpose we will compare English and Finnish academic texts. The data used for the above purposes consist of spoken and written texts2 by two authors. The spoken texts are papers delivered at conferences, the written texts are the versions of the same texts prepared for publication. One of the authors is a native speaker of English, henceworth called NSE, the other one For the sake of convenience we use the term 'text' to refer to both spoken and written language material.

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

171

is a native speaker of Finnish speaking and writing in Finnish (NSF). These two will be analysed separately and then compared. The same NSF also has a paper delivered and written in English, which we will use in order be able to compare her behaviour in the two languages. Our data, then, consists of the following types of texts: SPOKEN 1 NSE 1 NSF 1 NNS (E)

WRITTEN 1 NSE 1 NSF l NNS(E)

Although our data is very small and our analysis basically descriptive, we have also analysed the data quantitatively mainly for the purposes of comparison between texts and different types of impersonalization. The quantitative analysis was done by counting the occurrences of each type of personal reference and relating them to the total number of utterances in the texts. The results are normally given as percentages, i.e. a particular type occurring in such and such percentage of the total number or utterances. Finding a unit suitable for the analysis of both written and spoken texts is often difficult. We are using as our unit of analysis "utterance", by which we mean independent clauses containing a predicate and corresponding roughly to T-unit. The following example from one of the spoken texts illustrates our way of dividing texts into "utterances" : Utterance 1. Utterance 2. Utterance 3. Utterance 4.

okay I'll start off by saying that I am a psychologist and like all psychologists I do these highly artificial studies with single words and laboratory setting with no context and very little language involved I'll start with giving you some background on this study that I'm going to talk about today can you all hear me properly

3. Comparison of the systems of personal reference in English and Finnish Before launching on a discussion of the use of the strategy of impersonalization in the data, it is necessary to describe the devices available for it in the two languages involved. In this description the first step is to compare the systems of explicit personal reference in English and Finnish and then proceed to the possibilities that the languages offer for its avoidance. 3.1. Explicit personal reference In both languages a person can be explicitly referred to, of course, by using the person's name or a noun/noun phrase that singles out that person, but more often by using a personal pronoun. Table 1 shows the systems of personal pronouns in English and Finnish (Table 1 page 172). English has a system of five persons, Finnish one of six. The difference is due to the fact that English does not make a distinction between singular and plural in the second person, whereas Finnish does. However, English makes a distinction in the third person singular between masculine and feminine, which

172

IV. Cross-Cultural Aspects of Hedging

Table 1. Systems of personal pronouns English person 1st 2nd 3rd

singular I you he/she (it)

Finnish plural we you they

singular minä sinäTe hän (se)

plural me te he (ne)

is not done in Finnish. In English the pronoun it is often included in the personal pronouns, as a neutral third person singular, although it is not used in reference to persons except in rare cases (the baby - it.). Its Finnish counterpart se (plural ne), which is considered a demonstrative pronoun by grammarians, is frequently used in colloquial speech to refer to a person. However, in formal language se is not used in this way. Another difference between the two languages is that in Finnish the second person plural te is used also when addressing one person deferentially. A more notable difference between the two languages is that in Finnish a personal suffix is attached to the verb in all six persons, not only in the third person singular as in English. This means that in the first and second persons the pronoun can be omitted as the suffix indicates the person (minä istun = "I sit"), which makes the person less emphatic than in English (kirjoitan = I write). The majority of Finnish verbs are inflected in all persons, but there are exceptions like the modal verbs, of which only the third person singular is used and the noun or pronoun in the subject position takes the genitive form (Minun täytyy läbteä I must go", Sinun täytyylähteä "You must go"). There are also cases in which, what in English would be subjects, get "lowered" to the position of an adverbial or an object in Finnish. Thus, in Minua väsyttää "Iam tired" minua is an object. In Minultajäikirja kotiin "ließ the book at home", minulta is an adverbial. The verb in these cases takes the third person singular form. This is possible when the semantic role of the person is that of an experiencer rather than an active agent. The English personal pronouns have object forms and corresponding possessive and reflexive pronouns. The Finnish pronouns can be inflected in all the same cases as nouns. The possessor role of a person is indicated by the genitive form and a possessive suffix attached to the following noun. Also in this case the pronoun is optional in the first and second persons (minun kirja«/ = kirja»z "my book"). In colloquial speech the possessive suffixes are generally left out and abridged forms of the pronoun used (mun kirja = minun kirjawz "my book").

3.2. Avoidance of personal reference The personal pronouns are part of deixis, i.e. they belong to those linguistic items which "relate utterances to the spatio-temporal coordinates of the act of utterance" (Lyons 1977, 636). The first and second persons typically refer to the participants in the communication situation, whereas the third person pronouns may have referents outside the situation. However, even though typically deictic and referential, some of the personal pronouns can also be used

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

173

impersonally, to refer to an unspecified human agent. They do not then identify individuals but indicate "anyone" or a vague group of people. Even the first person singular pronoun can be used impersonally but only when talking hypothetically, as pointed out by Kitagawa/Lehrer (1990, 742), who give the following example. We form a frame or script for this kind of situation... Thus, in order to be able to take the subway in New York I simply need a "taking a subway" script or frame, if I have one... But at the same time, I may - even if I take the subway daily - be reminded of yesterday's trip when I met this strange man...

The impersonal uses of you and we are, however, much more common. They could both replace / in the above example and would actually be more normal in that type of context. However, Kitagawa/Lehrer (1990, 752) maintain that, although the impersonally used /, we, and you are often interchangable, they still reflect their "personal" use and are thus "distinct from each other with respect to rhetorical force and pragmatic implications". Siewierska (1984, 244) is on the same lines when she points out that the impersonal you explicitly involves the addressee, who "is represented as a potential protagonist and his participation is not offset by the overt concomitant participation of the speaker as with we". The use of these pronouns is made even more complex by the fact that bothyou and we can also be used in reference to a vague group of people, i.e. they can refer to specific but not identified individuals, as docs you in the following example (in Kitagawa/Lehrer 1990): You're - I don't mean you personally - you're going todestroy us all in a nuclear war. (said to an American a European)

In addition to the impersonally used personal pronouns English also has the non-deictic pronoun one for impersonalization. According to Quirk et al. (1972, 222), one means "people in general" but with a particular reference to the speaker. They also say that its use is chiefly formal and is often replaced by the more informal you, i.e. the difference is a stylistic one. However, Siewierska (1984, 243) maintains that one is slightly more vague thanjyo«. Also in Finnish the first person plural pronoun me can be used like the English we to refer either to "anyone" or to a vague group. The second person singular pronoun sinä is not used for generic or vague reference in writing although it seems to be sneaking into colloquial Finnish, perhaps as a result of influence from languages like English. The first person singular minä is possible in the hypothetical example about travelling in the subway given above. But a more natural alternative in Finnish would be the so called generic verb form, which means the third person singular form of the verb used without a subject, as seen in the following translation of the above English example: Niinpä voidakseen matkustaa maanalaisella New Yorkissa tarvitsee yksinkertaisesti "maanalaisella matkustamisen" käsikirjoituksen, Jos omaa sellaisen... Mutta samalla saalfaa - vaikka matkustaakin maanlaisella joka päivä - muistaa edellisen päivän matkan, jolloin tapasi...

According to Hakulinen (1987), avoidance of explicit personal reference is deeply ingrained in the structure of the Finnish language. This is clearly seen in the frequent use of the generic verb form. By using the generic form speakers

174

IV. Cross-Cultural Aspects of Hedging

can not only avoid reference to themselves but also make their opinions sound like general truths, as in Uutta kirjauisuutta lukee mieküään (literally "New literature reads with pleasure", i.e. "One reads new literature with pleasure"). A further possibility for avoiding explicit personal reference in both languages is the use of a passive construction. In English this means the use of the agentless passive, which is more "impersonal" than the other alternatives described above, because it does not have any implications like one and you. However, even in cases in which the passive is used the agent can often be retrieved from the context. The English passive has also another central motivation for its use: topicalization of patient. It may therefore be difficult to decide in individual cases what has been the motivation for its choice. Still, even when the prime motivation is topicalization, the agentless passive offers the possibility of leaving the agent ambiguous. The so called passive form in Finnish is different from the passives of IndoEuropean languages. It indicates that the agent is human but unspecified. Morphologically the passive is formed by adding a passive marker (-tta/-tta, ta/-ta) to the verb stem and after that a special person suffix - Vn. Thus the passive form of the verb istua "to sit" is tstutaan (istu+ta+an). Furthermore, the passive is inflected in different tenses and moods. The Finnish passive is closely connected with the personal inflection of verbs, to the extent that it is considered a generic fourth person (Hakulinen/Karlsson 1979). Because it always indicates a human but indefinite agent, Shore (1986) suggests that Finnish has no real passive at all but an indefinite verb form, which resembles the passive of Indo-European languages, the shared characteristic being that both make it possible to hide the agent. From the fact that the unspecified agent is always human in connection with the Finnish passive it follows that for example the sentence Mies tapettiin ("The man was killed") implies murder and cannot refer to a traffic accident, in which case one would have to say Mies kuoli ("The man died"). Another difference when compared with languages like English is that an agent cannot normally occur with the Finnish passive. Thus, it cannot be used for topicalization, which in Finnish is simply done by changing the word order of an active sentence: Jussi söi omenan 'John ate the apple" vs. Omenan soiJussi"The apple ate John, i.e. The apple was eaten by John". However, in written - often bureaucratic - language the agent is sometimes expressed through a circumlocution (Viranomaisten taholta ilmoitetiun...)"'it was announced on the part of the offcials ..."). An advantage that the Finnish passive has when compared with the English passive is that it can be used with both transitive and intransitive verbs. Thus it is possible to say Stella laulettiinja tanssittiin (literally "There was sung and danced"). Both languages also offer the possibility to avoid personal reference through nominalization. In Finnish the verb used with the nominalized phrase is often tapahtua ("happen"). Thus, instead of saying Minä pisteytin tulokset seuraavasti ("I counted the results in the following way") the speaker or writer can say Tulosten pistetyttaminen tapahtui seuraavasti ("The counting of the results happened in the following way"). Nominalization is the most impersonal of all impersonalization devices.

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

175

Table 2 summarizes the possibilities for avoidance of personal reference in English and Finnish. Table 2. Devices used for avoiding direct personal reference ENGLISH agentless passive nominalization pronoun one pronouns I, you, we

FINNISH passive generic verb form nominalization pronouns me (sinä)

3.3. Problems of interpretation

There are problems of interpretation in the analysis of some cases of impersonalization: it is often difficult to decide which person is hidden behind the impersonal expression. Because the same devices are used for avoiding reference to different persons, the interpretation depends entirely on the context. Thus, for example the passive can be interpreted in different ways in both English and Finnish. It can be used to avoid explicit reference to the writer him/herself as happens in the following examples, in which the context makes it clear who is meant. This argument is developed more fully in C (to appear) (C is the writer himself) Tässä kirjoituksessa pyritään ennen kaikkea tarkastelemaan vieraiden kielten opetusta oppimisen kannalta. ("In this article an attempt will be made to look at forcing language teaching from the point of view of learning.")

The passive can in both languages also refer to "anyone", "people in general", as in the following: How is knowledge of language put to use? Baddleyn tyomuistimallin mukaan lukemisen yhteydessä muodostetaan artikulaation avulla fonologinen työmuistitiedosto ("according to Baddley's model of working memory a phonological working memory register is formed with die help of articulation")

A third possibility of interpreting the passive is to see it as hiding a vague group of people like "researchers in the field" or "language teachers". This interpretation is possible in the following two examples. Kognitiivisessa psykologiassa oppiminen nähdään aktiivisena ajattelutoimintana. (In cognitive psychology learning is seen as an active thinking process") second language learners are being treated as if diey're pale and inefficient copies of marvellous monolinguals

In this last mentioned type it is sometimes difficult to decide whether the speaker or writer wants to be included in the vague group or not. If s/he does, these cases could also be seen as avoidance of first person reference. In the spoken versions there is occasionally vacillation between explicit and implicit reference, as in the following Finnish example, in which the speaker starts by referring to herself with the first person pronoun, then switches to the passive, then immediately back to the first person pronoun. In spite of this

176

W. Cross-Cultural Aspects of Hedging

switching she seems to be referring to herself all the time, although it is also possible to interpret the passive as referring to a vague group, perhaps a research team. tota, ensimmainen tehtävä mitä mä tein oli sellanen jossa tota, nämä lapset kuulivat...ja sitten nämä nauhoitettün taas toiselle nauhurille. ja sitten katsottiin erilaisia asioita, mä muun muassa mittailin reaktioaikoja (well, the first task I did was one in which these children heard "..."and dien these were recorded onto another tape-recorder, and dien different things were looked at, I among other things measured reaction times)

Also the Finnish generic verb form has a double function: it can be used for avoiding explicit reference to either the speaker/writer or to the audience. Again it is the context that clarifies the interpretation, as in the following examples, in the first of which reference is to the speaker, in the second to the audience: Aiemmin tuli todettua se tuttu tosiasia, että suomen kielen konsonantismi on sangen yksinkertainen ("Earlier happened to mention (= I happened to mention) that the Finnish system of consonants is quite simple") Taulukosta 6 voi lukea, missä järjestyksessä sijapäätteet ilmaantuvat suomalaislasten puheeseen. ("In table 1 can read (= you can read) in what order the case endings appear in the speech of Finnish children")

The same kind of interpretation problem is met in the case of the first person plural pronouns we and me in both languages. In addition to explicit reference to speaker/writer and audience, both pronouns are used in references to people in general or to a vague group of people. The former one is the more likely interpretation in the following extracts: it's saying how do we use die language knowledge mat we have in our minds for sociological and psychological ends eli jokainen meistä omaksuu kielen ja meidän hermojärjestelmä on sen kaltainen että kielen omaksuminen on sille hyvin tyypillistä ("in odier words everyone of us acquires language and our nerveous system is such that language acquisition is very typical for it")

However, in the next example a vague group of researchers is indicated by wey "we who work in the field of language acquisition": if we bring in the target L£ too we are in die danger of warping our analysis of die learner's own grammar

Problems of interpretation are also caused by we and me when they are used in speaking. They can be seen as rhetorical devices through which the speaker tries to draw in the audience or to create an atmosphere of togetherness. The alternative interpretation is that the pronouns are used in order to avoid explicit reference to oneself. Both interpretations are possible in the following example. mä näytän että miltä se näyttäs jos me otettas nämä eri järjestyksessä elikkä alotettas siitä että... ("I will show what it would look like if we took diese in a different order in odier words if we started from...")

MR. Luukka and R. Markkanen, Impersonalization as a Fonn of Hedging

177

Ambiguous are sometimes also the Englishjw» and one. In the following examples they can be interpreted either as referring to the speaker or to any representative of a vague group of linguists working in the field of second language learning. both have their advocates in linguistics but you can't find anything other than references to them in second language acquisition as soon as one goes beyond statement of regular patterns in this kind of data

4. Companion of spoken and written versions 4.1. Spoken and written texts by the NSE

The native speaker of English uses an equal amount of personal references (explicit or implicit) per utterance in the spoken and written versions: the percentage for the spoken version is 47,5% and 48% for the written version. Out of these personal references 64% are explicit ones in the spoken version and 36% implicit ones. In the written version the relationship is reverse: 64% implicit and 36% explicit references. Thus, on the whole, the written version is much more impersonal than the spoken version. The most noticeable difference between the two versions is that there is a high frequency of direct 1st person references in the spoken version: in 24% of the utterances the speaker refers to himself directly by using / whereas there is not a single / in the written version. The / in the spoken text is used in three different functions: in references to what the speaker is doing or going to do ("speaker as doer"): I am going to talk about today... I'll end up with my set of views... I'm trying out some of the arguments on you...

in references to the speaker as "experiencer" or "knower": I don't know of a major linguist since Roman Jacobson...

in so-called parenthetical clauses, i.e. independent clauses that interrupt the ongoing speech, most of which function as hedges: the main answer sadly I suppose is that linguists...

Contrary to the spoken, the written text does not contain any parenthetical clauses, which is to be expected as these clauses are a feature of spoken language rather than written. Thus there are no hedges of the type I suppose in the written text. In implicit references to himself as 'doer' or 'knower' the writer uses the passive. Clearest cases are those in which the writer uses the passive and then gives the source of information, which is the writer himself, as in the following: This argument is developed more fully in X (to appear) pC is the writer's name)

There are other cases which are not that clear but could be interpreted as implicitly referring to the writer:

178

IV. Cross-Cultural Aspects of Hedging The knowledge of the second language has to be looked at as an independent grammar in its own right The learner's interlanguage may be compared with the target L£ as spoken by monolinguals

In the last two examples the alternative interpretation is that the reference is to a vague group of people like "linguists" or "people doing second language research". Another way of implicit reference to oneself used by the NSE in the spoken version is one (it only occurs once in the written version).: a third area one can mention I suppose is...

In the next example, however, the speaker may be referring to himself or to a vague group, in which he himself is included: as soon as owe goes beyond statement of regular patterns in this kind of data...

A similar consideration applies to the NSE's use of the pronoun you, which can be interpreted as referring to a vague group which also includes the speaker himself, as in the following: you can't look at language knowledge without looking at how it is actually acquired

In the next extract, however,you clearly refers to "anyone", i.e. it is truly generic, but of course it then includes also the speaker or writer: before jo« can begin to relate caterpillars to butterflies you have to describe a caterpillar as a caterpillar

There are proportionally more of both generic references and references to vague groups in the written version than in the spoken one. Another clear difference between the two versions is in references to third persons. In both texts these are names of the persons whose works have been used as sources of information. There are clearly more of these in the written text (16%) than in the spoken one (2%). In the written text the sources are naturally given formally, with years of publication within the text, which is not always done in the spoken version. There is an interesting problem for interpretation in the NSE texts related to references to third persons. Both texts give an overview of linguistic theories as related to second language learning. Therefore the topic is the views of various linguists. This leads to frequent mentioning of names, which are not references to persons as sources of information but persons as topics of discussion. We have attempted to keep these two types separate, which is not always easy, and have only included the sources of information into our account of third person reference. Of the following two examples the first one contains a reference to a person as topic, the second one to persons as sources of information: Weinreich's nineteen fifty three book languages in contact seems to me one of die truly brilliant books in our area As Sridhar/Sridhar (1986) point out...

Table 3 summarizes the differences between the two versions and shows clearly how much more explicitly personal the spoken version is when compared to

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

179

the written one. The devices used for each person are given in their order of frequency of occurrence. Table 3. Devices used for personal reference (NSE) PERSON

I

SPOKEN TEXT pronoun I passive

WRITTEN TEXT passive

one

one you he/she

we generic or vague

(you) pronoun you name passive + name pronoun we pronoun we pronoun you passive

name passive + name

passive pronoun you pronoun we

we

As table 3 shows, there are no explicit references to the first or second persons in the written text. Even in the vague and generic references the passive is preferred to pronouns in the written version. The passive seems to make these vague references even more vague. 4.2, Spoken and written texts by NSF The Finnish spoken and written texts show clear differences both as regards to the frequency of personal references and the devices used for them. In the spoken text the NSF's style is much more personal than in the written one. She frequently refers to other researchers, to herself, to the audience or to herself and the audience together. The majority of her utterances (70%) contain explicit or implicit references to persons. In the written text the frequency of occurrence of utterances without any personal reference is clearly higher (55%). There is also a difference between the texts in the explicitness of these references: in the written version eight out of every ten references are implicit, whereas in the spoken version the corresponding frequency is only five out of every ten utterances. The written text is thus clearly more impersonal than the spoken text. The clearest difference is in the references to self. In the spoken version they are clearly explicit: the speaker refers to herself by using the first person singular pronoun mind. She uses implicit reference to herself only when reporting on an experiment she has carried out. In the written version, on the other hand, most of the references to self (80%) are made by using the passive, i.e. implicitly. Of the following two examples the first one comes from the spoken version, the second one represents the written version: minätein tätä sillä tavalla että mäannoin kaksi saraketta, lauseita ja niistä... (I did this so that I gave two columns, sentences in which...) Oppilaille esitettiin kaksi viiden lauseen joukkoa ja heidän oli löydettävä... (The pupils were given two groups of five sentences and they had to find...)

Even when the writer refers to herself explicitly she often uses only the verb with the first person suffix and leaves out the pronoun. In the spoken version

180

W. Cross-Cultural Aspects of Hedging

almost all explicit references to self contain the pronoun. Thus, also in this sense the spoken version is more explicitly personal. The following examples illustrate this difference between the two texts: Spoken: taustateoriana mä oon käyttäny tässä, joo mä oon käyttäny tämmöstä bädlin ja hitsin työmuistiteoriaa (as a background theory I have used here, yes I have used this theory of working memory) Written: Muistipuskurin toiminnan malliksi valifsin Baddleyn and Hitchin (Baddley ja Hitch 1974) tyomuistimallin... (As the model of the functioning of the memory buffer chose+I Baddley's and Hitch's...)

In the spoken version the majority of explicit personal references are to the speaker herself, but in the written version the majority are to third persons, i.e. other researchers, whose names are mentioned. In the written version there are no direct references to the readers (te/"you") or to the writer and readers together (me/"we"), whereas the spoken version contains a few of these interpersonal references, although fewer than references to the speaker herself. The following is an example of the speaker addressing the audience directly: tota mää aion puhua lähinnä ensin käsitteestä työmuisti vain sen verran että le tiedätte mitä minä sillä tarkoitan (well I am going to talk first about the concept of working memory enough so that you will know what I mean by it)

In most cases, however, the speaker refers to the audience implicitly by using the generic verbform: ja Jos nyt katsoo vaikka tätä ettei huku numeroihin katsoo tätä oikeanpuoleista saraketta (and if now looks (one looks) for example at this column on the right so that does (one does) not get lost in numbers)

The spoken version contains slightly more of vague and generic references than the written version. These are mostly expressed through the passive. Some of them could be interpreted as having a general reference or as referring to a vague group of unidentified persons. The context helps in deciding between these two interpretations. In the next example the speaker refers to a vague group of people, researchers who are using a particular test: sünä carrolin ja steltonin testissä vastaava tehtävä on semmonen jossa anmtaan kaksi lausetta josta toisesta...on alleviivattu jokin sana (in carrol's and stelton's test the corresponding task is one in which two sentences are given in one of which a word is underlined)

In the following example, however, the speaker uses the passive for generic reference: virheet mitkä siellä syntyy hyvin paljon heijastaa yksinkertaistusprosesseja joista saadaan pseudosana semmoseen muotoon että sitä pystytään siellä muistissa pitämään niin kauan että se voidaan... (the errors that are made greatly reflect the simplification processes of which a pseudoword can be formed in a form which can be kept in memory so long that it can be...)

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

181

In the written text there is not a single first person plural pronoun used in referring to the writer and readers together. Instead, it is used for generic reference. The generic verb form is used more for generic or vague reference in the spoken text than in the written one. The passive is used in both texts for several purposes: to refer to the speaker/writer, to the listeners/readers either alone or together with the speaker/writer, in generic references to anyone, and to a vague group of people. However, there is a difference between the texts: in the spoken version it is used most frequently for vague reference, whereas in the written text its main function seems to be in references to the writer herself. Table 4 summerizes the devices used for personal reference, both explicit and implicit, by the NSF in her spoken and written texts. The divices are given in the order of frequency of occurrence. Table 4. Devices used for personal reference (NS1:) PERSON

I

you he/she

we generic

or vague

SPOKEN TEXT pronoun minä (I) passive possessive suffix person suffix generic verb form pronoun sinä (you) name passive pronoun me (we) passive generic verb form pronoun me (we)

WRITTEN TEXT 'passive person suffix possessive suffix

name

passive pronoun me (we) generic verb form

5. Influence of language on personal reference A more reliable picture of the influence of language on the use of personal reference or its avoidance can be achieved by comparing texts produced by the same person in the two languages. In this part of our research we used the spoken and written versions of an English and a Finnish text produced by one and the same person. The topic and the genre of the English and Finnish texts greatly resemble each other: they are both reports on a research experiment. At a general level the proportions of explicit personal references and avoidance cases are similar in the English and Finnish texts: there are more explicit references in the spoken versions than in the written ones. However, in the spoken English versions of the NNS(E) the proportion of explicit references is clearly higher (69%) than in her Finnish spoken version (45%). Her English and Finnish written texts are more alike in this respect. There is also suprising similarity between the English and Finnish texts in references to self: the frequencies of explicit reference are almost equal in both the spoken and written versions of both her English and Finnish texts. A slight difference detected between the spoken versions is that the English text is somewhat more personal than the Finnish one. For example, direct ref-

182

IV. Cross-Cultural Aspects of Hedging

erences to the audience are used more often than in the Finnish text, in which direct address is avoided much more frequently. However, referring vaguely to other persons seems to be easier in the mother tongue: this was done much less frequently in the English texts. As the devices for explicit and implicit reference are by and large similar in the two languages, it is difficult to say whether the English texts of the NNS(E) resemble more her own performance in the mother tongue or the texts by the native speaker of English. The only clear differences are found in generic or vague references to persons. In this the Finn's spoken English text resembles more the text of the native speaker of English than her own Finnish text. She refers to vague or generic persons through the pronouns we andyou more than the passive. In her Finnish spoken version she uses the passive most frequently in this function. However, in her written English version her generic or vague references follow the Finnish style: she uses the passive throughout, except for two cases of the pronoun we. Another difference is that the native speaker of English does not use one single explicit reference to himself in his written version, whereas the Finn refers explicitly to herself by using the first person prounoun equally frequently in both her spoken and written English texts. Tables 5 and 6 compare the ways explicit and implicit personal reference is used by the native speaker of Finnish, in her mother tongue and in English, to the ways the native speaker of English uses them. The different devices used are given in the order of frequency of their occurrence. Table 5. Spoken text PERSON

I

you he/she generic/ vague

English (NSE) pronoun I passive one you pronoun you name pronoun we pronoun you passive one

Finnish(NSF) pronoun I passive possesive suf. verb + person suf. generic verb pronoun you name passive generic verb pronoun we

English (NNE) pronoun I we passive pronoun you name pronoun we pronoun you passive

Table 6. Writen text PERSON

I

English passive one

Finnish (NSF) passive verb + person suf. possessive suf generic verb

English (NSE) passive pronoun I we

you

-

-

-

he/she

name passive name passive pronoun you pronoun we

name

name passive + name passive pronoun we

generic/ vague

passive pronoun we generic verb

Although the devices used for personal reference - whether explicit or implicit - are similar in English and Finnish at a general level, there are minor differ-

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

183

ences, as shown in chapter 2 above. But these differences do not seem to be decisive enough to cause any noticeable mother tongue transfer when the Finn uses English. However, it can be speculated that Finnish might have some influence on the behaviour of Finns in English in this area. For example, the less conspicuous way that Finnish offers for explicit reference to self and to the audience through the use of only the personal suffix without the pronouns makes reference to self easier - it seems to be half-way to impersonalization in being less conspicuous than the use of the pronouns. This may make the explicit reference to oneself easier also in English, as is seen in the written English version of the Finnish writer in this study. Another example of mother tongue influence could be the use of the passive. As pointed out above, the Finnish passive resembles more an indefinite person than the passive of IndoEuropean languages and can be used in more functions. This may encourage its use by Finns also in English, which again is seen in the fact that the Finn in this study uses it for generic/vague reference almost exclusively in her written English text. More important, however, than these differences in the actual linguistic devices between the languages, are the norms for the choice between explicit and implicit reference. And these seem to be very similar in Finnish and English. Our results indicate that the general tendency for the avoidance of explicit reference is similar in scientific texts regardless of language or actual text type. The main tendency is for spoken texts to be more explicitly personal than written ones. This tendency seems to be clear both in texts that report on an experiment and those which discuss a more general theoretical issue. Moreover, the difference in the explicitness of personal references between the spoken and written versions is very similar in both Finnish and English. However, when we compare the English and Finnish spoken versions we notice that in the English texts the references are more frequently explicit than in the Finnish texts. Interesting is also that the native speaker of Finnish behaves in this respect like the native speaker of English: both her spoken and written English versions are more like those of the NSE than her own Finnish texts. Figure 1 illustrates this point (Figure 1 page 184). When we compare the variety of personal references we notice that it is greater in the spoken texts than in the written ones. Both the Finnish and English spoken versions contain references to the speaker him/herself, to the audience, to the audience and the speaker together, other researchers and vague groups of people. The frequency of references to self, explicit and implicit, is slightly higher in the Finnish than the English spoken versions. Other differences are to be found in the total number of personal references and in implicit references to second persons. Once again the NNS(E) text resembles the NSE text more than her own Finnish text, as can be seen in Figure 2 (Figure 2 page 185). In the written texts the variety of persons referred to is smaller: they contain only implicit references to self, references to other scholars and vague groups. In the written Finnish text the proportion of references to self, both explicit and implicit is clearly greater than in the English texts. Differently from the spoken versions, the English text written by the Finn is in this respect more like

184

IV. Cross-Cultural Aspects of Hedging

her own Finnish text than that of the native speaker of English. This result is illustrated by Figure 3 (Figure 3 page 185).

English text (NSE)

Finnish text (NSF) 100

100

implicit references explicit references

spoken text

written text

36 64

64 36

implicit references explicit references

spoken text

written text

55 45

77 23

English text (NNE)

implicit references explicit references

spoken text

written text

31

65

I explicit references E3 implicit references

35

Figure 1

M.R. Luukka and R. Markkanen, Impersonalization as a Fonn of Hedging

185

Personal references in spoken texts

vague person explicit 3rd person implicit we explicit we implicit you explicit you implicit 1 explicit 1

English

Finnish

NNEnglish

16 2 0 2

22 3 2 2 8 2 6 23

10 1 0 4 0 4 5 24

0 2 1 24

L] vague person £3 explicit 3rd person Π implicit we D explicit we Q implicit you E3 explicit you Η implicit I Η explicit I

V

Figure 2

Personal references in written texts

60 50 40 30 20 10 0 vague person explicit 3rd person implicit we explicit we implicit you explicit you implicit 1 explicit 1

LJvague person 0 explicit 3rd person 0 implicit we Ο explicit we

'////////Z,

English

Finnish

NNEnglish

27 16 0 2

16 6 0 0 1

9 14 0 0 0 0 23 4

0 0 3 0

0 17 4

Figure 3

Q implicit you E3 explicit you IZD implicit I Η explicit I

186

IV. Cross-Cultural Aspects of Hedging

6. Concluding remarks

This paper has taken the view that avoidance of explicit personal reference is a form that the strategy of hedging can take. The results therefore indicate that scientific writing requires more hedging than scientific speaking (cf. also Luukka 1992 and Luukka 1994). The explanation for the above phenomenon has to be sought for in the general requirements and norms of academic discourse. The requirements seem to be somewhat contradictory: on the one hand, what is required of scientific texts is precision, i.e. it has to be made clear who is responsible for the information. On the other hand a certain modesty is required: the topic is more important than people. In particular, emphasis on one's own person must be avoided. These requirements seem to be stronger for writing than for speaking. It is easier to say "I" than write it. It also seems easier to say "I" when the speaker is referring to what s/he is doing (e.g. "Next I will take up the issue of...") than using "I" when putting forward arguments or opinions. In the spoken data of this study most of the direct references to self are to the "speaker as doer", not to the "speaker as knower". In the written version, when the writer gives his/her own views or earlier research results, this is done by using the same technique that is used when referring to other sources of information, for example by saying "As is shown by X (1993)" instead of "As I have shown...". Similarly, it is easier to refer explicitly to the listeners with "you" in speaking because they are present. The speaker knows who s/he is speaking to, whereas in writing a scientific text the writer may only have a vague notion of what kind of people will probably read it. Explicit reference to audience would be out of place in scientific writing; hence it is not used at all in the written data of this study. References to third persons seem to be the opposite from references to first or second persons. In the present data there are much fewer explicit references to third persons in the spoken versions than in the written ones. Thus, writing seems to require more precision, i.e. explicitness in this respect than speaking. Because of the case study nature of our research it is impossible to draw general conclusions about differences between the Finnish and English cultural norms of scientific writing and speaking. Our results only suggest possible differences. However, it can be said with fair certainty that the difference between the spoken and written scientific register seems to be the same for both languages. It may be that English texts are slightly more explicit in their personal references than Finnish texts, and this seems to be more likely in the spoken versions. Interesting is also the fact that our Finnish subject adopts the English style in her spoken English version but not to the same extent in her written version. Our main aim in this study has not been to find cultural differences between Finnish and English texts but to widen the scope of the concept of hedging. By connecting implicit personal reference with it we hope to have been able to emphasize its interpersonal aspect, i.e. that it should not only be looked at from the speaker/writer's perspective. We also hope that our discussion widens the rather narrow view of personal references given by grammarians. When looked at from a functional, pragmatic perspective they appear to be much more complex and difficult to interpret, but at the same time more interesting.

M.R. Luukka and R. Markkanen, Impersonalization as a Form of Hedging

187

References Brown, Penelope and Levinson, Stephen 1987 Politeness: Some universal language usage. Cambridge: Cambridge University Press. Chafe, Wallace 1982 "Integration and involvement in speaking, writing and oral literature", in Tannen, D. (ed.): Spoken and Written language: Exploring orality and literacy. (— Advances in Discourse processes, vol IV). Norwood, New Jersey: Ablex, 35-54. Chafe Wallace and Danielewicz, Jane 1987 "Properties of spoken and written language", in Horowitz, R_ and Samuels, SJ. (eds.): Comprehending Oral and Written Language. San Diego: Academic Press, 83-113. Hakulinen, Auli 1987 "Avoiding personal reference", in Verschueren, J. and Bertucelli-Papi, M. (eds.): The Pragmatic Perspective: Selected Papers from the 1985 Pragmatics Conference (Pragmatics and Beyond Series). Amsterdam: John Benjamins, 141-153. Hakulinen, Auli and Karlsson, Fred 1979 Suomen kielen lauseoppia. Helsinki: Suomalaisen kirjallisuuden seura. Hübler, Axel 1983 Understatements and Hedges in English (Pragmatics and Beyond IV:6). Amsterdam: John Benjamins. Kitagawa, Chisato and Lehrer, Adrienne 1990 "Impersonal uses of personal pronouns", Journal of Pragmatics 14, 739-759. Korhonen, Riitta and Kusch, Martin 1989 "The rhetorical function of die first person in philosophical texts - the influence of intellectual style, paradigm and language", in Kusch, M. and Schröder, H. (eds.): Text, Interpretation, Argumentation. Hamburg: Helmut Buske, 61-77. Lakoflf, George 1972 "Hedges: a study of meaning criteria and the logic of fuzzy concepts", in Papers from the Eighth Regional Metting of Chicago Linguistic Society. Chicago: Chicago University Press, 183-228. Luukka, Minna-Riitta 1992 "Varmuuden kahdet kasvot tieteellisessä tekstissä", Virittajavol 2, no 4, 361-379. Luukka, Minna-Riitta 1994 "Metadiscourse in academic texts", in Gunnarsson, B-L., Linell, P. and Nordberg, B. (eds.): Text and Talk in Professional Contexts. (Selected Papers from the International Conference Discourse and the Professions). Uppsala: ASLA, 77-88. Lyons, John 1977 Semantics. Cambridge: Cambridge University Press. Markkanen, Raija and Schröder, Hartmut 1992 "Hedging and its realizations in German, English and Finnish philosphical texts: A case study", in Nordman, M. (ed.): Fachsprachliche Miniaturen (Scandinavian University Studies in Humanities and Social Sciences). Frankfurt am Main: Peter Lang Verlag, 121-130. Myers, Greg 1989 "The pragmatics of politeness in scientific articles"; Applied Linguistics 10, 1-35. Quirk, Randolph; Greenbaum, Sidney; Leech, Geoffrey and Svartvik, Jan 1972 A Grammar of Contemporary English. London: Longman. Shore, Susanna 1986 Onko suomessapassiivia. Helsinki: Suomalaisen kirjallisuuden seura. Siewierska, Anna 1984 The Passive: A Comparative Linguistic Analysis. London: Croom Helm.

TARJA NIKULA

Interlanguage View on Hedging 1. Introduction

Hedging can be described as a strategy by which speakers mitigate and soften the force of their utterances. It is a phenomenon probably shared by most languages, if not by all. It is true that different languages may have quite different linguistic realizations for hedging, ranging from morphological devices to syntactic constructions. Yet interpersonal concerns, such as politeness and face saving, constrain language use across cultures so that it is often beneficial for speakers to express themselves in a hedged form rather than directly and unequivocally. This paper looks at hedging from an interlanguage perspective (Selinker 1972, 1992). More particularly, as appropriate hedging constitutes an aspect of pragmatic competence, the viewpoint is that of 'interlanguage pragmatics' in that the main interest will be on hedging in a foreign language.1 In Rasper's words (1992, 1), interlanguage pragmatics is "a branch of second language research which studies how non-native speakers (NNS) understand and carry out linguistic action in a target language, and how they acquire L2 pragmatic knowledge". The reason for focusing on interlanguage rather than native language is that appropriate hedging can be a difficult thing to master for foreign language speakers (see e.g. Holmes 1982; Blum-Kulka and Levenston 1987). For native speakers it rarely causes problems; having internalized their language and culture they know where and how to use hedges without having to pay much conscious attention to it. As Amdt/Janney (1991, 522) put it, "people growing up in a culture learn techniques not only for expressing logical relations between concepts in speech, but also for expressing feelings and attitudes toward things, events, partners, etc.". Non-native speakers, on their part, may be unaware of the role of hedging, and they may therefore leave their speech unhedged, concentrating only on the transmission of referential information. Given that hedging serves important interpersonal functions this can at its worst result in pragmatic failure which easily reflects unfavourably on speakers The term interlanguage pragmatics has usually been used in connection with non-native speakers. Kasper/Blum-Kulka (1993, 1-2), however, suggest that tying interlanguage pragmatics to non-native speakers only may narrow its scope too much as there appears evidence of interlanguage styles even among native speakers, brought about by language contact.

T. Nikula, Interlanguage View on Hedging

189

so that they may be regarded as intentionally impolite or offending rather than having insufficient pragmatic skills (see Thomas 1983; Riley 1989). This is especially true when speakers are otherwise fluent because people easily expect that someone who speaks their language well on the surface level also masters the underlying social conventions. This paper is based on a recent study which focused on the use of modifying expressions in conversational settings (Nikula 1992, 1993). The main focus was on conversations by Finnish speakers of English (NNS) whose use of both hedging and emphasizing modifiers was compared with native speakers of English (NS(E)), and with the same Finns' conversations in Finnish (NS(F)). The Finnish participants were university students of English; they were advanced speakers, and their English was fluent as far as vocabulary and grammar were concerned. The native speakers of English were British university students of linguistics, of about the same age as the Finnish students. The data thus contained three types of conversations - NNS, NS(E), and NS(F) - and there were four conversations of each type, each lasting about 30 minutes.2 The data can best be described as consisting of informal, non goal-directed conversations between equals. Brown and Yule (1983) distinguish between transactional and interactional discourse type on the basis of whether the focus is on optimally efficient transmission of information or on maintenance of interpersonal relationships. Although it is difficult to draw any clear distinction between the two types, the conversations in this study can be regarded as mainly interactional because the main task of the speakers was to collaboratively keep the conversations going. The findings cannot thus be generalized to all kinds of contexts but, broadly speaking, they give an insight of the use of modifying expressions in informal interactional situations. Although both hedging and emphasizing expressions were analyzed in the study, this paper deals with hedging only3, with the following research questions in focus: (i) How much does the use of hedges by non-native speakers of English differ from native speaker performance? Hedging has attracted increasing attention in linguistics recently but much more research is still needed of speakers with different native languages and different target languages before an adequate account of hedging in a foreign language can be given, (ii) Are differences so great that they can affect non-native speakers' successful commuAll groups were given a general topic at the beginning of the session (the rather long duration of university studies in Finland (NNS), die question of whether students should pay the poll tax (NS(E)), and the role of the English language in Finland (NS(F)). The purpose of the topic was only to get the conversation going; the participants were free to drift from the topic or change it entirely whenever they would feel like it. Note that the non-native speakers discussed in English among themselves rather than widi native speakers of English. This may have had an impact on dieir behaviour but given that die use of modifying expressions is largely an unconscious phenomenon (see e.g. Faerch and Kasper 1989, 243), it is probable diat they would behave in much the same way also in authentic encounters with native speakers. This is largely due to the fact that modifiers with an emphasizing function were much rarer than hedging expressions in all three sets of data. This seems to suggest that in diese conversations, mitigating was interpersonally more important than emphasizing, which is an interesting finding as such.

190

IV. Cross-cultural Aspects of Hedging

nication in English? All differences need not be detrimental to communication but some of them can play a crucial role in how non-native speakers are perceived. It is therefore important to investigate what the biggest problems for non-native speakers are in the use of hedges so that these can be taken into consideration in foreign language teaching, (iii) What is the role of learners' Lj in the way they use hedges in English? As Blum-Kulka et al. (1989, 26) point out, it is still an unanswered question in interlanguage pragmatics to what extent characteristics in learner language are typical of interlanguage as such, and to what extent they are influenced by learners' Lj.

2. Hedges 2.1. The concept of hedge The term hedge is in itself vague, and over the years different writers have defined it differently. In this paper, hedging is seen as a strategy which renders speakers' messages more tentative and vague, and thus reduces the force of what they are saying. The strategy of hedging can be realized in various verbal and non-verbal ways; as Brown/Levins on (1978, 151) put it, "the semantic operation of hedging can be achieved in indefinite surface forms". Instead of trying to account for hedging strategies in general, however, this paper focuses only on the linguistic level and on such lexical expressions as I suppose, probably, sort of or a bit, for example. This is because focusing on lexical devices serves as a good starting point for analysing hedging behaviour of non-native speakers. Note that the term lexical is here used in opposition to, for example, prosodic or syntactic hedging strategies such as passivization, instead of referring to single lexemes only. Thus expressions such as I don't know or or something like that are also regarded as lexical hedges. It is difficult to give any kind of concise list of hedges because they basically form an open-ended category. Depending on the formality of the situation, for example, speakers may use a completely different set of expressions to add to the vagueness of their messages. Compare, for example, or stuff like that and as it were, which are likely to be used in informal and formal settings, respectively. A further complication is that what functions as a hedge depends on the context so that the same expression may sometimes be a part of the proposition proper and sometimes modify a proposition. For example, the adverb just mitigates the speaker's opinion in I just have my doubts about him being a good psychologist, but it is part of the proposition in if you wanted you could apply it [the loan]just for oneyear. It is therefore in principle necessary to look at each expression in its context of occurrence before it can be labelled a hedge. It is, however, possible to give a general description of different types of hedges and in this paper, a division is made between two major categories. Firstly, there are expressions such as perhaps, might, I suppose, which are closely connected to the concept of epistemic modality in that they signal speakers' assessment of the truth of the proposition as a whole. More particularly, they indicate that speak-

T. Nikula, Interlanguage View on Hedging

191

ers are not fully committed to the validity of the proposition they are conveying.4 Following Prince et al. (1982), such expressions will be called 'shields'. The other category of hedges contains expressions such as sort of, kind of, somewhat, a bit (see e.g. Lakoff 1972; Skelton 1988). Rather than indicating the exact degree to which speakers are committed to the truth of the proposition as a whole, these expressions focus on a word or an expression, making its meaning more fuzzy and imprecise (e.g. sort of funny, a bit old-fashioned). These expressions will be called 'approximators'.5 Prince et al. (1982, 85) formulate the division by saying that while approximators introduce fuzziness within a proposition, shields introduce fuzziness between the speaker and the proposition. It can also be argued that shields bring speakers and their personal assessments quite explicitly to the open whereas speakers can avoid giving emphasis to their role in interaction by using approximators. Shields could thus be characterized as markers of uncertainty and approximators as markers of denotational vagueness. Yet they both help reduce the force of what speakers are saying and in that sense both types function as hedging devices.6 Although this paper mainly focuses on shields and approximators, attention will also be paid to expressions such as / mean, you know, well and like, which Östman (1981) calls pragmatic particles and Schiffrin (1987) discourse markers. Some writers include these expressions in hedges (e.g. Brown/Levinson 1978). They are kept separate from the category of hedges proper in this paper, however. One reason is that their scope of influence is often hard to define and they cannot therefore be divided into shields and approximators in the same way as hedges; they are often ambivalent in that they seem to have characteristics of both. Another problem is that they do not always function as modifiers. Speakers often use them to organize discourse. These expressions can, for example, function as turn-taking or turn-yielding devices. They can also occur as pause fillers or hesitation markers, and speakers can use them out of habit rather than in an interpersonally salient function. In fact, they have often been Lyons (1977, 797) defines epistemic modality as follows: "Any utterance in which the speak er explicitly qualifies his commitment to the truth of the proposition expressed by the sentence he utters, whether this qualification is made explicit in the verbal component or in the prosodic or paralinguistic component is an epistemically modal or modalized utterance". Lyons thus makes it clear that epistemic modality is not the property of the verbal expression alone. This paper, however, concentrates only on the verbal component. It is possible to make a similar division between emphasizing expressions. Thus, for example, certainly and obviously indicate speaker's full commitment to the truth of the proposition, whereas very and extremely typically focus on a word or an expression, intensifying its meaning The following approximators and shields were used by the native and non-native speakers of English in the present study. Note that this is not meant to be an exhaustive list of hedges. It rather illustrates hedging expressions that are likely to occur in informal face-to-face con versations. In different contexts speakers might use a different set of expressions. Approximators: kind of, sort of, a bit, a little, fairly, not really, not very, pretty, about, just, almost, quite, actually, basically, or something like that, some(thing), and everything, or whatever, and that sort ofthing, and stuff like that. Shields: I think, I suppose, I guess, I feel, I find, I figure, I don't know, I'm not sure, maybe, probably, perhaps, supposedly, possibly, may, could, would, should, might, sounds, seems, tend, tag questions.

192

W. Cross-cultural Aspects of Hedging

negatively evaluated as mere rumbles. Yet it is true that they frequently occur in mitigating functions as well. Their meaning is difficult to define without context, however, and they can remain ambiguous even then so that it is eventually up to the hearer to interpret them. This is why they will be called 'implicit modifiers' in contrast to shields and approximators which are more explicit in that their downtoning meaning is easier to interpret even when they occur out of context. 2.2. The interpersonal significance of hedging It was pointed out above that the function of hedging is to tentativize utterances by reducing their force. This does not reveal the whole significance of hedges, however, because one also has to account for the interpersonal reasons for speakers' willingness to remain vague. It is nowadays widely acknowledged that politeness is a chief motivation behind hedging. In their well-known theory of politeness, Brown/Levinson (1978, 1987) make a division between negative and positive politeness on the grounds of whether speakers take into account others' want to remain unimposed or their wish to be accepted and approved by others. That is, negative politeness is associated with respect and social distance and positive politeness with solidarity. Hedging is often seen as a negative politeness strategy because it helps reduce the force of speech acts that might threaten addressees' negative face, such as requests or questions, for example. The function of hedges on such occasions is, as Hübler (1983,23) puts it, "to make sentences more acceptable and thus to increase their chance of ratification by the hearer". It is not, however, always easy to keep negative and positive politeness apart and hedges can also be used to protect the hearer's positive face. It often depends on the type of speech act modified whether hedges are seen as negative or positive politeness markers. In Holmes' (1984, 350) words, "to reduce the force of an 'unwelcome' speech act is to express positive feelings towards the hearer which should increase the solidarity of the relationship". Positive and negative politeness functions of hedges are thus closely intertwined. Politeness is frequently equated with protecting the face of others, and saving hearer's face is also often regarded as the main purpose of hedging. However, speakers quite often use hedges in order to protect their own face wants as well. For example, hedgy messages can protect speakers from potential unfavourable judgements by others in that they cannot be made fully responsible for something they expressed tentatively in the first place. It is thus equally difficult as with separating between negative and positive politeness functions of hedges to say whether they are directed more towards the hearer's or the speaker's own face. As Goffman (1967, 6) puts it, one's own face and the face of others are "constructs of the same order". But even if it is difficult to pin down the exact politeness function of hedges it is nowadays generally accepted that their use does contribute to perceptions of politeness. It would therefore be important for non-native speakers to understand the role of hedges and the motivations behind speakers' hedgy language, and to make use of this knowledge in their interlanguage.

T. Nikula, Interlanguage View on Hedging

193

Negative and positive politeness are usually associated with face-threatening acts that speakers wish to redress. But even in the absence of obvious facethreatening acts speakers are responsible for maintaining good atmosphere and good relations with others. The concept of 'involvement' can be used to refer to such personal and interpersonal investment to the situation as a whole. Involvement is a concept that is often hard to define. Tompson (1991, 169) describes it as emotional commitment, and according to Villaume (1988, 362) it is "the degree to which a person focuses on the flow of conversation and is sensitive to the interplay of self and other in patterns of emergent meaning". In discussing the properties of spoken and written language Chafe (1985) maintains that while written language is by its nature detached, spoken language shows a variety of manifestations of involvement. He distinguishes between three kinds of involvement: involvement of the hearer with him/herself, involvement with the hearer, and involvement with the subject matter. There are various involvement signals, prosodic and kinesic cues being of great importance (e.g. gaze, proximity, speech rate; see Cappella 1983), but it is interesting from the viewpoint of this paper that Chafe (1985) also discusses certain expressions that are here regarded as hedges and implicit modifiers in terms of their ability to function as involvement signals. For example, expressions with first person pronoun and reference to a speaker's mental processes (e.g. shields I suppose, I figure, I guess) signal ego-involvement, and modifying expressions which appeal to the addressee can signal involvement with the hearer (e.g. you know, tag questions). It is thus possible to look at hedges not only as mitigating expressions but also as expressions which contribute to the personal and interpersonal atmosphere of the situation. That hedges can signal different meanings does not diminish their interpersonal significance, however. On the contrary, it is exactly this potential multifunctionality of hedges that is often beneficial for successful communication as it leaves room for meaning negotiations and reformulations for both speakers and hearers. Finally, it is important to point out that as the notions of politeness and successful communication can vary across cultures, also the interpersonal functions of hedges may be different in different cultures. It must therefore be borne in mind that as this paper deals with speakers of English and Finnish, the perspective is that of western cultures where individualism is an important concept in interaction and where politeness can be understood on the basis of mutual face work. Recent research has indicated, however, that in many nonwestern cultures group membership rather than individualism is often considered more important (Janney/Arndt 1993, 18), and it is possible that in such cultures also the role of hedges would be different.

3. Characteristics ofinterlanguage hedging 3.1. Quantitative differences

One of the main assumptions behind the study described in this paper was that hedging, as other phenomena of the pragmatic domain, can be very difficult

194

IV. Cross-cultural Aspects of Hedging

for foreign language speakers. It came thus as no surprise that the Finnish speakers of English used hedges considerably less than the native speakers of English. There were 301 hedges in the four NNS conversations and 705 in the four NS(E) conversations. In general, the learners had a tendency to use only a few hedges, which typically occurred at the beginning of their turns. Quite often, however, they left long stretches of talk completely unmodified. In the NS(E) conversations speakers usually used hedges more evenly throughout their turns. Examples 1 and 2 illustrate the difference between the two groups.7 (Both hedges and implicit modifiers are italicized in the examples.) Example 1:NNS 51 52 51 53 54 52 51 53 54 52

what about this four years at the university plus continuing education while at work? / think that's for for most women it's im- it's fairly difficult because if you want to have children and do everything you're supposed to do it's not that easy yeah and anyway now they aim at continuing education /mean why does it have to be four years and this [it can't be more and still] we to go to, yes [mm and- yeah] and, we aren't all going to be teachers so we aren- maybe we take a job that has nothing to do with the language and then we have to take continuing education, [in] linguistics and all [mm] that's right so you won't be a, teacher? no but I hope I have a job which has something to do with the language ((laughs)) me too

Example 2: NS(E) 51 52 51 53 52 51 53 52 53 SI 54 SI

like who agrees? I suppose we have to be vocal here we can't just put up our [hands here mm yeah] weü I don't think we should no why not? because well I can't afford to for one thing and / don't think people who're considerably less well off than others should have to pay you still use the services though yeah well maybe we should pay but, wett I suppose we could be then asked to pay proportionally less I think, much less yeah [and there's (-) what like five pounds or] (laughs) but you still have to pay things like going on buses and everything don't you?

The difference between the native and the non-native speakers is even bigger if also implicit modifiers are taken into account, as can be seen also in the exam7

For reasons of readability, the examples have the following marking conventions only: [ ] for overlapping speech; (—) for unintelligible speech, and, for short pauses.

T. Nikula, Interlanguage View on Hedging

195

pies above. There were 127 implicit modifiers in the NNS conversations and 587 in the NS(E) conversations.8 Implicit modifiers were thus very frequent in the NS(E) data. Moreover, the native speakers often used them in connection with potentially face-threatening contexts, such as disagreements or expressions of personal opinions, for example. It is therefore reasonable to assume that they do not only serve a gap-filling function but also have an interpersonal role in interaction. James (1983, 201) suggests that they are often responsible for creating "a certain informality of style and intimacy of relationship". That is, they seem to have the capacity of signalling that speakers feel at ease with each other and see the situation as relaxed. In this sense, implicit modifiers can function as involvement markers because they show speakers' emotional commitment to the situation. The Finnish speakers of English used implicit modifiers very rarely, failing to communicate the sense of involvement associated with the use of these modifiers. The lack of implicit modifiers is thus partly responsible for the impression that the learners sounded, on the whole, more detached and more formal that the native speakers in these corresponding face-to-face situations.9 Besides using shields, approximators and implicit modifiers more frequently than the non-native speakers, the native speakers also very often used all these types in complex combinations. This made their speech quite heavily modalized at times. Coates (1987, 129), similarly, draws attention to native speakers' skill to use a range of modifiers in complex and subtle combinations. Example 3 illustrates this kind of multiple use of modifiers. Example 3:NS(E) SI

/ think it's just, I hate filling in forms because you know that none of the boxes are gonna fit your particular lifestyle [...] I don't know sort of I mean sometimes I actually enjoy you know sort of scribbling things all over it

Such complex combinations of different hedges could not be found in the NNS conversations at all. This can partly be due to the fact that learners are not aware of the role of modifiers and they thus cannot make appropriate use of them. Another reason may be that foreign language speakers probably quite easily think that fluent speech is something that contains as little hesitations, pauses, or extraneous words and phrases as possible. They may therefore be unwilling to use a bunch of modifiers in an utterance when speaking a foreign language, regarding it as a sign of disfluency. As Sajavaara/Lehtonen (1978, 49) point out, foreign language speakers are never taught how to be appropriately 'disfluent'. This is because foreign language teaching is often largely based on 8

9

When considering hedges and implicit modifiers together, and the number of words they constitute, they represented 7.3% of all the words in the NNS data whereas in the NS(E) data the corresponding figure was 11.2%. It is exactly the implicit nature of these modifiers that most probably makes it quite difficult for non-native speakers to interpret their meaning. They therefore seem to find it easier to resort to hedges whose meaning they can be fairly sure of, such as maybe or I think, for example (see also Kärkkäinen 1990, 74).

196

IV. Cross-cultural Aspects of Hedging

the use of literary standards which contain no 'disfluencies'. Yet outside classrooms, where there are interpersonal constraints on language use, expressing bare facts with simple unmodified declaratives is often inadequate. Given that hedges play an important role in linguistic politeness by making speakers appear more indirect and vague, there is a real possibility that nonnative speakers can be perceived too direct if they use hedges scarcely. Such interpretations, obviously, always depend on the context. It seems on the basis of the present native speaker data, however, that the use of hedges is a rule rather than an exception in informal face-to-face encounters; this applies to native speakers of both English and Finnish. Especially expressions of personal views and opinions were almost always modified with hedges. The multifunctional nature of hedges is particularly clear in such cases because expressing opinions tentatively can simultaneously protect the hearer's face from imposition and shield the speaker's own face from unfavourable judgements of others. The non-native speakers' tendency to offer unhedged opinions can thus give a rather unyielding and categorical impression. Compare examples 4 and 5 which illustrate the difference: the native speakers hedge their views much more frequently, thus signalling their attitudes towards what they are saying. In comparison, the non-native speakers report their views without modifications, going straight to the point instead of resorting to mitigating devices. Example 4: NS(E) 51 52 SI 53

weU I think the idea of the old rates you know the idea of actual property tax was okay because generally speaking the people who own property tend to be the richest ones anyway that's right yeah I said it is a good [idea I mean] students [don't own property and they're but / mean I hate] to be very sort of capitalist and and whatever but I do believe up to a certain point that if / mean I don't know it's not something I feel strongly about so whatever I say is pretty, ((laughs)) but til mean there is the point that you know if if people are earning you know a lot of money say and they've obviously trained hard for it and worked hard for it, then why should they be punished by a- by paying more?

Example 5: NNS 51 52 51 52 51 52 51 52 53 51 S3 52

wish I could fight against becoming a teacher as long as I can yeah many people fight against that ((laughs)) mm so everybody says they are not going to be teachers but hi the end everybody, almost everybody will because it's it's the easiest way I'm studying politics too, now politics? politics, in valtio-oppi aha if I ever wanted to go to the foreign ministry ((laughs)) it- it's handy to have some studies in politics too mm I'm such that I hate politics and history and political history (laughs)10 well there's nothing wrong in being a teacher but yes there is one thing and that's the wages the wag- money mm

T. Nikula, Interlanguage View on Hedging

197

Examples 4 and 5 further illustrate that the use of hedges can also affect the level of formality. The use of hedges is characteristic of informal speech situations in particular, which is why their rare occurrence in the learners' speech made them sound more formal than the native speakers despite the fact that the setting, the topic, and the relationship between the participants were about the same in all conversations. This sense of formality was brought about by the fact that, with the infrequent use of modifying expressions, the non-native speakers' turns were very densely packed with information as almost every word carried referential rather than interpersonal information. As a result, they seemed to be more preoccupied with the content of their messages than with their interpersonal effect. This apparent concentration on the content of the message can easily result in the impression of formality. 3.2. Qualitative differences

Apart from the smaller number of hedges in the NNS conversations than in the NS(E) conversations in general, there were also certain qualitative differences in the way the two groups used hedges. The non-native speakers, for example, resorted to a much smaller variety of hedges. This was reflected, for example, by the finding that there were only 14 shields and approximate« that they used more than five times in the four conversations whereas in the NS(E) conversations the corresponding figure was 26. The smaller variety of hedges is not a very interesting finding as such; it is quite natural for learners to have a smaller set of expressions at their disposal simply because they speak a foreign language. What is interesting, however, is the almost complete lack of certain hedges in the learners' speech, such as sort of, kind of, and tag questions, for example. These hedges were very frequent in the native speaker conversations; their combined number was 128 in the NS(E) data and only 11 in the NNS data. The non-native speakers also used implicit modifiers / mean, you know, well and like much less than the native speakers, as was indicated above. These particular hedges and implicit modifiers are certainly familiar for the students who have been in contact with the English language for years, yet they do not employ them in their interlanguage. One reason for this may be that these expressions are all very much part of informal, face-to-face register which has not been the focus of formal language teaching at schools and which the speakers, consequently, have had little access to. The fact that the two non-native speakers who had spent a longer period in an English speaking country used the majority of these expressions in the NNS data suggests that experience of authentic informal use of language can improve speakers' pragmatic skills.11 10 Although this paper focuses on lexical hedges only, it is worthwhile to point out that in this particular NNS example laughter may well serve mitigating functions. That is, instead of making use of verbal hedging strategies, die non-native speakers indicate by laughing that what they say need not be taken at its face value. 11 Thomas (1983,110), however, points out diat a stay in the target culture does not automatically guarantee greater success pragmatically; pragmatic skUls are not simply 'absorbed' when living in the target environment.

198

IV. Cross-cultural Aspects of Hedging

Besides using a narrower range of hedges the non-native speakers favoured one shield in particular: I think was clearly the most frequent hedge in their speech with its 65 occurrences. The native speakers used it frequently as well (112 occurrences), but the learners used it proportionally more often as I think represented 22% of all their hedges; in the native speech its proportion was 16%. It was an even more interesting finding that the two groups used this shield differently. The native speakers typically used it in combination with other hedges so that there could be little speculation about it having a mitigating function as well because multiple hedges tend to reinforce each other. They also quite often used it in the form I don't think which can be seen as a more tentative form than I think (see e.g. Bublitz 1992, 571). For the non-native speakers I think was usually the only hedge in an utterance, at least not immediately accompanied by other hedges. They also typically began their turns with this shield. Examples 6 and 7 illustrate the difference between the two groups. Example 6: NNS 51

52 53 S2 SI

/ think that people who study languages, the students are quite dif ferent from the other- other students here cos they, they travel a lot they don't have boyfriends and girlfriends, they don't get married, they don't have children, cos they want to be free and, travel and and, like that they want to live their live in full, there was too (—) who don't go abroad (-) professionals I think I don't fit that pattern ((laughs)) (-) independence too like it's so, I like travelling abroad and 11 like studying languages, practising it but, there are buts

Example 7: NS(E) 51 52 52 53 S2 54 S2

just like he probably doesn't really know what he's talking about, (—) the architecture yeah yeah that would be a no-no I think offer your head ((laughter)) probably, I don't know he just I mean he seems very different to previous, previous monarchs yeah / thinkhe's got some green ideas hasn't he? yeah / mean he is sort of he is concerned about, er architect- / mean I think his views are good

That the non-native speakers favoured this particular hedge may cause some problems because besides having a mitigating function, I think can also emphasize the force of an utterance. This is true especially when it is utterance-initial where it is easily spoken with emphasis (see Holmes 1985, 33). It is thus possible that the scarce use of other supporting hedges together with the non-native speakers' tendency to begin their turns with / think helped contribute to an impression that the speakers wanted to bring their views into focus rather than mitigate them. This is an effect which is probably quite the contrary to the speakers' intentions and which might lead to misunderstandings in actual encounters with native speakers.

T. Nikula, Interlanguage View on Hedging

199

3.3. Lack of involvement in interlanguage As was indicated above (section 2.2), certain hedges and implicit modifiers can create a sense of involvement, either by indicating that speakers are interested in their hearers and their views or by showing speakers' interest and commitment to the flow of conversation in general. Tag questions and the implicit modifier you know, for example, appeal to the addressee and can thus signal other-involvement. These modifiers occurred frequently in the NS(E) conversations, and they are, in general, particularly frequent in informal situations. The non-native speakers used these modifiers very rarely and they may therefore appear as showing little concern for their coparticipants. As Chafe (1985, 116) points out, the function of other-involvement markers is that they are signs of "concern for the dynamics of interaction with another person", and the failure to indicate such concern can make speakers appear quite indifferent to their interlocutors. It was also pointed out that hedges and implicit modifiers that contain I-pronoun together with a reference to speaker's mental processes signal ego-involvement (Chafe 1985). Even though the non-native speakers used / think fairly often, they used hardly at all other expressions that signal speaker's personal involvement such as I mean, I suppose, I gather, I find, and I figure which were very frequent in the NS(E) conversations. Thus, despite the learners' favouring / think, their non-use of other speaker-oriented forms made them seem, on the whole, more personally detached from what they were saying than the native speakers. The native speakers were more likely to discuss also rather general topics in terms of their own views which they indicated by using modifiers with first person pronoun whereas the non-native speakers' use of language appeared more impersonal in comparison, largely due to the small number of hedges with personal reference. Compare, for instance, examples 8 and 9 which illustrate how the general tone of the NNS and the NS(E) conversations tended to differ as far as the use of modifiers was concerned. In example 8, the non-native speakers discuss a topic in very impersonal terms, distancing themselves by the frequent use of you as an indefinite pronoun whereas in example 9, the native speakers more explicitly make reference to their personal views and experiences, signalling thus more ego-involvement with the message. Example 8: NNS 52 53 52 53 S2

you just have to change the whole contents of studies or the there should be much more possibilities to study during the summer but who wants to sacrifice three months' vacation yeah [especially if you don't and in many cases] it isn't a vacation you go abroad or you work [and mm] most people work it's just you have to work if you want to finance your winter anyway

Example 9: NS(E) SI

...and my youngest sister and me who were pretty broke anyway / mean we're even more broke now, so

200

IV. Cross-cultural Aspects of Hedging S3 SI

yeah / suppose it's sort of elderly people who live on their own they've gained so I suppose it has done somebody some good mm /still 1think that the only fair thing would be to tax people on what comes in that's the only seems to be the only fair way of establishing what goes out

The discussion above indicates that hedges have the capacity of functioning as cues of speakers' personal attitudes and feelings and speakers who fail to provide such cues can sound quite detached and uninvolved. It is thus useful to look at hedges also from the viewpoint of involvement because it gives a broader picture of their interpersonal significance than looking at their use as markers of politeness alone. 3.4. LI influence on interlanguage

In addition to comparing conversations by non-native and native speakers, it is also useful to have a look at learners' Lj performance to tackle the question of pragmatic transfer. It is quite likely that native language plays a role in interlanguage but this can happen in complicated and subtle ways so that it is unreasonable to make any simplistic equations between LI and IL. It is also necessary to bear in mind Rasper's (1992, 21) notion that it is important to dissociate the result of pragmatic transfer from communicative effect. Therefore, it is oversimplification to conclude that difference from L2 brought about by LI influence necessarily equals miscommunication. But even though transfer need not always have an unfavourable effect on communication, the question of LI influence on interlanguage hedging is nevertheless theoretically interesting. Besides this theoretical interest, there was also another reason for investigating the learners' conversations both in English and in Finnish. This is the fact that concentrating on learners' foreign language performance alone can easily lead to assumptions that the way they speak a foreign language is a direct reflection of their native culture and native language behaviour. Thus, although it is probably quite true that when Finns use foreign languages in international encounters, they are often perceived abrupt or too straightforward (cf. Hiukka 1991), this is certainly as much an indication of their insufficient pragmatic skills in a foreign language as an exact mirror image of the way they behave when speaking Finnish. Kärkkäinen (1990, 74) has also investigated Finns' performance in English and she suggests that the fact that Finns use modal expressions rarely in English might reflect that they are "more likely to go about their business in a very straightforward and 'honest' way" when using their native language. But as her study contains no data in Finnish the suggestion remains unsupported by evidence. It was pointed out earlier that also in the present study, the non-native speakers' performance was characterized by the scarce use of hedges, which corroborates Kärkkäinen's findings about the small amount of modifying material in Finnish speakers' English. It was, however, difficult to attribute the small number of hedges to native language influence because when the same students spoke Finnish, their utterances contained an abundance of hedges as is illustrated by example 10. In addition to using approxi-

T. Nikula, Interlanguage View on Hedging

201

mators and shields, the speakers in the NS(F) data also made frequent use of implicit modifiers niinku(n) and sitteen (the rough meaning equivalents in English are 'like* and 'such a/in such way', respectively). Example 10: NS(F) 51 52

53 51 52

vaikka tuolla kadulla keskenämme ruvettais englantia puhumaan nun tuntus väbän ouvolle ma laas oon sitä mieltä että mun mielestä than mä tykkään puhuu niinku suomalaisten kans niinku vaikka tämmösissä ülanteissa ni tykkään puhua englantii ja ruotsii et musta se on uv&jolenki keskustella tälleen koska siinä on just se että kaikki ollaan niinku tavaüaan samassa jamassa siinä joo niin on eikä sitä silken kuuntele eikä niinku sillä tavalla välitä niinku on joienkin vapaampi ku on saman ikäsiä ja ei niinku kiinnitä sillä tavalla huomiota puhumiseensa mut heti jos tulee joku opettaja ni mä ainakin muu tan heti niinku jotenkin ehkä väbän puhettani sillä tavalla niinku että valitsen sanani ei kirosanoja^ näin edespäin (...if we started to speak English with each other on the streets for example, it would seem a bit strange in my opinion I think I quite enjoy speaking you know widi odier Finns like in these kinds of situations I enjoy speaking English or Swedish I find it nice somehow to discuss like this because there's die fact that we are all like in a way in the same situation yeah mat's right and you don't sort ofpyy attention and like in a way you don't care you know you feel somehow more free when there are people of me same age like you don't son of pay atten tion to what you're saying but if mere's a teacher present at least I change you know somehow maybe a little me way I talk like choose my words no swear words and so on)

There were 501 hedges altogether in the NS(F) conversations and 320 implicit modifiers; the proportion of modifying expressions of all words varying between 8% and 15% in the four conversations. It is important to point out that comparing the proportional amount of modifying expressions in the Finnish and English conversations is difficult because the two languages are structurally very different. Finnish is an agglutinative language which means that one Finnish word can express a meaning that needs to be expressed by several words in English (e.g. talo+i+ssa+mme+kin: noun + plural marker + case ending + possessive suffix + clitic 'also in our houses'). Due to such differences in word formation, simple type-token relationships do not give a conclusive answer to the question of which group uses hedges more, proportionally. Yet the number of hedges in the NS(F) conversations is illustrative enough for showing that hedging is by no means a rare phenomenon in casual Finnish conversations, either. There is, of course, the question whether hedging serves similar politeness functions in Finnish and in English, that is, what is 'the same' across cultures. Janney and Arndt (1993, 20) point out that there may well be as many different notions of politeness and face as there are cultures and languages. Similarly, hedging may not serve the same interpersonal functions in

202

IV. Cross-cultural Aspects of Hedging

Finnish and in English. As both groups represent western cultures, however, and as the speakers of both NS(E) and NS(F) conversations seemed to use hedges in much the same ways, it can fairly safely be assumed that hedges served to a large extent similar functions in the two sets of native speaker data: those of mitigating the message either to make it more acceptable to the hearer or to guard speaker's own face from unfavourable judgements by others. Thus the finding that Finnish speakers used hedges abundantly when speaking their mother tongue but expressed their views clearly more abruptly when speaking English supports the assumption that hedging is indeed a difficult thing to master in a foreign language and it does not develop automatically along general language proficiency. Although the small number of hedges in the interlanguage data cannot be assigned to L^ influence as such, there were certain characteristics in the learners' use of hedges that seemed to suggest native language influence. For example, the non-native speakers tended to favour hedges that have rather close semantic equivalents in Finnish. Thus expressions or something like that, I think and maybe, which were among the most frequent hedges in the NNS conversations, can be compared with Finnish hedges tai jotain sellaista Or something like that', musta/mun mielestä think' and ehkä 'maybe'. These all occurred frequently in the NS(F) conversations in hedging functions. These hedges were used frequently by the native speakers of English as well, which means that learners' favouring them need not have any unfavourable effect on communication. Perhaps the use of a narrower range of modifiers only serves to indicate that speakers are non-native. Thus, as was already indicated above, Lt influence need not equal miscommunication. Native language influence may also explain why the learners used rarely such approximators and implicit modifiers as sort of, kind of, I mean, andyouknow, which often reflect that speakers see the situation as informal and relaxed. These expressions do not have as close semantic equivalents in Finnish as the ones mentioned above which may be a reason for the non-native speakers using them so rarely. The speakers did express similar meanings when they spoke Finnish, but they made use of different linguistic means. For example, the Finns quite frequently attached morphological endings, so called clitics, to their words in order to modify the force of the message, and the use of clitics served rather similar functions to the use of implicit modifiers in English. Especially the clitic -hAn often seemed to have parallel functions with implicit modifiers in English.12 In the following example it signals shared knowledge between the speaker and the hearer, simultaneously diminishing the imposition of the utterance. It thus corresponds roughly to the function of you know in English: Eihan me voida muuttaa mitään (No+clitic we can change anything 'We can't change anything you know'). Similarly, the approximator semmonen/tämmönen (literally 'such a') was used frequently by the Finnish speakers and it was used in much the same ways as 12

Clitics resemble implicit modifiers also in being multifunctional. For example, Markkanen (1991, 192) shows that the clitic -hAn can have various functions depending on the context: It can be a marker or shared knowledge, it can be used as an emphasizer, or it can function as a softening device.

T. Nikula, Interlanguage View on Hedging

203

approximators sort of ana kind of by the native speakers of English. It seems, however, that such functional correspondence is more difficult for learners to perceive than semantic similarity. To sum up, the findings indicate that the way Finnish speakers of English use hedges differs both quantitatively and qualitatively from native speaker performance. It could thus be argued that they use hedges erroneously, but applying the concept of error is always problematic when talking about pragmatic phenomena. What Arndt/Janney (1991, 233) say about emotive contrasts in speech in general applies to hedges as well: "they are not true or false, or correct or incorrect according to rules or formal requirements of any kind, they are more appropriate or less in a given situation, or more successful or less in producing a particular impression or reaching a particular goal with a partner in a given instance". In this vein, the biggest problem for the non-native speakers was that they used hedges rarely, and that they often failed to use them appropriately in interpersonally salient positions. As a result, they can easily be judged overly formal, detached, or indifferent to their interlocutors because their speech lacks the interpersonal flavour that the use of hedges conveys. Thus, even though hedging is certainly not the only aspect of pragmatic competence, knowing how to use hedges appropriately would nonetheless be a big step towards more successful communication in a foreign language.

4. Implications for language teaching The fact that appropriate hedging does not seem to develop automatically along with foreign language skills in general sets an obvious challenge for foreign language teaching. Teaching aspects of language that belong to the pragmatic level can, however, be very problematic. Littlewood (1983, 184) maintains that providing learners with an abundance of pragmatic rules to remember can increase communicative anxiety rather than facilitate communication. Alexander (1988, 71) is along the same lines in pointing out that "just because something may be analyzed and hence made potentially teachable does not make it automatically desirable to be taught". Thomas (1983, 99) also discusses the delicate nature of teaching pragmatic principles. She points out that pragmatic preferences are often closely connected to speakers' values and beliefs, to how they see the world, and people may be very reluctant to change themselves in this respect when speaking a foreign language. Blum-Kulka (1991, 255) draws attention to the fact that as interlanguage speakers already master one language system they have general pragmatic knowledge at their disposal. That is, they have an intuitive knowledge of what communication is. It would seem reasonable, therefore, to take this general knowledge base into account in language teaching and start approaching pragmatic phenomena through learners' native language first. This ties in with considerations about the ways in which to improve non-native speakers' pragmatic skills. It is often suggested that as far as pragmatic phenomena are concerned, awareness raising is probably a better approach than explicit teaching. Thus,

204

IV. Cross-cultural Aspects of Hedging

given the existence of the general pragmatic knowledge base discussed above, it would seem natural to start increasing learners' awareness of pragmatic aspects of language from their native language rather than from the foreign language. According to Bardovi-Harlig et al. (1991), speakers can quite easily be made aware of pragmatic rules in their native language and what it means to enact these rules appropriately and inappropriately. Such awareness of Lj is needed in order to help non-native speakers appreciate that similar phenomena can also be found in foreign languages. As far as hedging is concerned, learners ought to be made conscious of the fact that it is not only acceptable but sometimes quite necessary for successful communication that speakers express themselves vaguely rather than directly, and that this applies both to their native language and to foreign languages. Problems of teaching pragmatic phenomena also lead to the question whether it is, generally speaking, desirable for non-native speakers to completely converge to native speaker norms. According to Janicki (1986, 177), maximal convergence is not necessarily evaluated positively by native speakers. Furthermore, Blum-Kulka (1991, 269) maintains that for many non-native speakers the desirable goal might be to diverge from the native norm because being different from native speakers can function as an assertion of cultural identity. Non-native speakers may indeed choose to diverge from the target language norms but this does not preclude the need to raise speakers' awareness of pragmatic phenomena in the first place. For, as Thomas (1983, 96) points out, the target of language teaching should be to equip foreign language to express themselves exactly the way they choose, whether rudely, tactfully, or in an elaborately polite way, the goal of consciousness raising being to prevent them from being "unintentionally rude or subservient" (emphasis original). It thus seems reasonable to suggest that in order to be able to diverge from the target language socio-cultural norms to mark cultural identity, foreign language speakers first need to become aware of those norms.

5. Concluding remarks

This paper has investigated one aspect of interlanguage pragmatics, the way non-native speakers use hedges in conversational settings. Although the findings clearly indicate that non-native speakers' use of hedges differs both from the target language and the native language norms, this paper has only given a preliminary insight to interlanguage hedging and there is plenty of room for future research. Firstly, contrasting separate sets of native and non-native data made it difficult to judge in any exact way the kind of interpersonal consequences that non-native speakers' use or non-use of hedges might have in genuine encounters with native speakers. It also outruled the possibility of investigating how much native speakers' use of hedges might affect non-native performance, for example in terms of accommodation. It is therefore an obvious future requirement to investigate also real NS-NNS encounters to give a more reliable account of how non-native speakers' way of using hedges affects the

T. Nikula, Interlanguage View on Hedging

205

overall success of communication. Secondly, focusing attention on lexical hedges only does not give a complete picture of hedging strategies. It would therefore be worthwhile to adopt a wider perspective on hedging, to investigate what linguistic and non-linguistic means are available for interlanguage speakers to mitigate the force of their messages. Finally, it is important to bear in mind that appropriate use of hedges is only one area of the interpersonal use of language. Much more research is therefore needed before we can fully understand what it means to be a pragmatically competent foreign language speaker.

References Alexander, Richard J. 1988 "An applied linguistic look at some aspects of the pragmatics of natural discourse", English Language Research Journal, Vol. 2, 61-77. Arndt, Horst and Janney, Richard W. 1991 "Verbal, prosodic, and kinesic emotive contrasts in speech", Journal of Pragmatics, Vol. 15, No. 6, 521-549. Bardovi-Harlig, Kathleen; Hartford, Beverly A.S.; Mahan-Taylor, Rebecca; Morgan, Mary J. and Reynolds, Dudley W. 1990 "Developing pragmatic awareness: Closing the conversation", ELTJournal, Vol. 45/1, 4-15. Blum-Kulka, Shoshana 1991 "Interlanguage pragmatics: The case of requests", in Phillipson, R., Kellerman, E., Selinker, L., Sharwood Smith, M., and Swain, M. (eds.) Foreign/Second Language Pedagogy Research, Clevedon: Multilingual Matters, 255-272. Blum-Kulka, Shoshana; House, Juliane and Kasper, Gabriele (eds.) 1989 Cross-Cultural Pragmatics: Requests and Apologies. Norwood, N.J.: Ablex. Blum-Kulka, Shoshana; House, Juliane and Kasper, Gabriele 1989 "Investigating cross-cultural pragmatics: An introductory overview", in Blum-Kulka, S., House, J. and Kasper, G. (eds.) 1989, 1-34. Blum-Kulka, Shoshana and Levenston, Edward A. 1987 "Lexical-grammatical pragmatic indicators", Studies in Second Language Acquisition, Vol. 9, 155-170. Brown, Gillian and Yule, George 1983 Discourse Analysis. Cambridge: Cambridge University Press. Brown, Penelope and Levinson, Stephen 1978 "Universals in language usage: Politeness phenomena", in Goody, E. (ed.) Questions and Politeness. Cambridge: Cambridge University Press. Reprinted, with a new preface, in 1987 as Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Bublitz, Wolfram 1992 "Transferred negation and modality", Journal of Pragmatics, Vol. 18, No. 6, 551-577. Cappella, Joseph N. 1983 "Conversational involvement: Approaching and avoiding others", in Wiemann, J.M. and Harrison, R.P. (eds.) Nonverbal Interaction, Beverly Hills: Sage, 113-148. Chafe, William 1985 "Linguistic differences produced by differences between speaking and writing", in Olson, D.R., Torrance, N. and Hildgard, A. (eds.) Literacy, Language and Learning. The Nature and Consequences of Reading and Writing. Cambridge: Cambridge University Press, 105-123.

206

IV. Cross-cultural Aspects of Hedging

Coates, Jennifer 1987 "Epistemic modality and spoken discourse", Transactions of the Philological Society, 1987,110-131. Faerch, Glaus and Kasper, Gabriele 1989 "Internal and external modification in interlanguage request realization", in Blum-Kulka, S., House,}, and Kasper, G. (eds.) 1989, 221-247. Gofrman, Erving 1967 Interactional Ritual. Essays on Face-to-Face Behaviour. London: Allen Lane the Penguin Press. Hiukka, Kristiina 1991 "Strengths, Weaknesses, Threats and Opportunities of Communication in Intercultural Business Negotiations. How Do Finns Succeed as Negotiators?" A paper presented at 1991 Biennial Convention of World Communication Association, July 1991, University of Jyväskylä, Finland. Holmes, Janet 1982 "Expressing doubt and certainty in English", RELCJournal, Vol. 13, No. 2, 9-28. 1984 "Modifying illocutionary force", Journal of Pragmatics, Vol. 8, No.3, 345-365. 1985 "Sex differences and mis-communication: Some data from New Zealand", in Pride, J.B. (ed.) Cross-Cultural Encounters: Communication and Mis-Communication, Melbourne: River Seine, 24-43. Hübler, Axel 1983 Understatements and Hedges in English. Amsterdam: John Benjamins. James, Allan R. 1983 "Compromisers in English: A cross-disciplinary approach to their interpersonal significance", Journal of Pragmatics, Vol. 7, No. 2, 191-206. Janicki, Karol 1986 "Accommodation in native speaker-foreigner interaction", in House, J. and Blum-Kulka, S. (eds.) Interlingual and Intercultural Communication. Discourse and Cognition in Translation and Second Language Acquisition Studies, Tübingen: Gunter Narr Verlag, 169-178. Janney, Richard W. and Arndt, Horst 1993 "Universality and relativity in cross-cultural politeness research: A historical perspective", Multilingua, Vol. 12-1, 13-50. Kärkkäinen, Elise 1990 "Face saving and epistemic modality", in Nyyssönen, H., Kuure, L., Kärkkäinen, E. and Raudaskoski, P. (eds.) Proceedings from the Second Finnish Seminar on Discourse Analysis. Oulu: University of Oulu, 64-76. Kasper, Gabriele 1992 "Pragmatic transfer", in Bley-Vroman, R. and Sato, C. (eds.) University ofHawai'i Working Papers in ESL,Vo\. 11, No. 1,1-34. Kasper, Gabriele and Blum-Kulka, Shoshana 1993 "Interlanguage Pragmatics: An Introduction", in Kasper, G. and Blum- Kulka, S. (eds.) Interlanguage Pragmatics, New York: Oxford University Press, 1-32. Lakoff, George 1972 "Hedges: A study in meaning criteria and the logic of fuzzy concepts", Papers from the Eighth Regional Meeting of Chicago Linguistic Society, 183-228. Littlewood, William T. 1983 "Contrastive pragmatics and the foreign language learner's personality", in Sajavaara, K. (ed.) Cross-Language Analysis and Second Language Acquisition I, Jyväskylä: Department of English, 179-188. Lyons, John 1977 Semantics. Cambridge: Cambridge University Press. Markkanen, Raija 1991 "Metadiscourse in intercultural communication", Proceedings of the VAKKISeminar XI, Erikoiskiekt ja käännösteoria, Vaasa, Finland: University of Vaasa, 184-194.

T. Nikula, Interlanguage View on Hedging

207

Nikula, Tarja 1992 Lexical certainty modifiers in non-native and native discourse. An unpublished Licentiate thesis. University of Jyväskylä, Department of English. 1993 "The use of lexical certainty modifiers by non-native (Finnish) and native speakers of English", in Bouton, L. R and Kachru, Y. (eds.) Pragmatics and Language Learning, Vol. 4, Urbana, IL: Division of English as an International Language, University of Illinois at Urbana-Champaign, 126-142. Östman, Jan-Ola 1981 You know: A Discourse-Functional Approach. Amsterdam: John Benjamins. Prince, Ellen R, Frader, Joel and Bosk, Charles 1982 On hedging in physician-physician discourse", in Di Pietro, R. J. (ed.) Linguistics and the Professions, Norwood, N. J.: Ablex, 83-97. Riley, Philip 1989 '"Well don't blame me.' On the interpretation of pragmatic errors", in Oleksy, W. (ed.) Contraslive Pragmatics, Amsterdam: John Benjamins, 231-249. Sajavaara, Kari and Lehtonen, Jaakko 1978 "Spoken language and the concept of fluency", in Lautamatti, L. and Lindqvist, P. (eds.) Focus on Second Language. A Special Issue of Language Centre News, Jyväskylä: Language Centre for Finnish Universities, University of Jyväskylä, 23-55. Selinker, Larry 1972 "Interlanguage", IRAL, 10,209-230. 1992 Rediscovering Interlanguage. London: Longman. Schiffrin, Deborah 1987 Discourse Markers. Cambridge: Cambridge University Press. Skel ton, John 1988 "The care and maintenance of hedges", ELTJournal, Vol. 42, No. 1, 37-43. Thomas, Jenny 1983 "Cross-cultural pragmatic failure", Applied Linguistics, Vol. 4, 91-112. Tompson, Sandra 1991 Toward a characterization of high-involvement style with reference to mail openings", in Bouton, L.F. and Kachru, Y. (eds.) Pragmatics and Language Learning, Vol. 2, Urbana, IL: Division of English as an International Language, University of Illinois at Urbana-Champaign, 167-185. Vdlaume, William A. 1988 Interaction involvement and the use of referential and formal anaphora in conversation", Language and Speech, Vol. 31, No. 4, 357-374.

HEINZ KREUTZ

Some Observations on Hedging Phenomena and Modifying Devices as Regional Markers in the Speech of Young East Germans

1. Introduction and General Background

The events of 1989 and since have brought massive political and social change to central and eastern Europe. In the case of Germany political unification also meant a unification of two speech communities with common cultural traditions which had been subjected to opposing socio-economic systems for almost half a century.1 The question of divergent developments of the German language in East and West Germany2, respectively, has been a long-standing issue, dating back as far as the mid-fifties. A multitude of books and articles on the issue have been published (cf. esp. Hellmann 1976, and v. Polenz 1993) by language experts and journalists alike on either side of the border, and the topic of'EastSpeak' vs. 'West-Speak' has repeatedly received much public attention. The current study was given impetus by the divisive stance between Ossis' and 'W«iw'3, a situation that has emerged only since 1989. Of interest is the fact that the two Germanics were allied to opposing power blocs (capitalist vs. socialist), and that the societies were structured in fundamentally different ways at all possible levels. For example, in the public domain communication in the various - respective FRG and GDR - institutions varied markedly. It has been

Härtung (1990); Fleischer (1987), and v. Polenz (1988) employ the term "Kommunikationsgemeinschaft'' (communication community) as opposed to "Sprachgemeinschaft" (speech community). They argue that whilst the overall German "Sprachgemeinschaft" remained intact, two heterogeneous "Kommunikationsgemeinschaften" did develop. More recently, this difference was emphasised by the east German politician W. Thierse (1992, 36), who states that east Germans did not have a different German language, but a different relationship to that language. In the following I shall employ the terms East/West German/y and FRG/GDR when referring to the political entities of pre-1989, and east/west German/y as merely pertaining to geographical areas of the new, unified Germany. The expressions OssiVWessi" are colloquialisms denoting Germans in east and west, respectively. The terms did not exist before unification, they are emotionally loaded in terms of inferiority (east) and (self-perceived) superiority (west).

H. Kreutz, Some Observations on Hedging Phenomena

209

argued that there existed in the GDR three distinct 'Kommunikationsbereiche' (communicational domains) - public (all government institution, including schools and the mass media), semi-public (e.g. church and opposition groups) and private (family, friends, neighbours and the like) - with varying degrees of ritualised discourse (Fraas/Steyer 1992, 175). Official communication patterns were generally in accordance with official party directives. Little is known about communication in the private sphere (but see Hellmann 1989; and especially Debus/Hellmann/Schlosser 1986 for alternative attempts at analysing spoken everyday language on the basis of GDR films). However, it has been argued that the prescriptive nature of the public register also had an effect on the private domain, and it has been assumed that the average GDR-speaker had to have command of 'innere Zweisprachigkeit (internal bilingualism), and was forced to constantly switch between registers, public to private and vice versa (Hellmann, esp. 1991; Schlosser 1990b). The need to switch was also required in situations where East Germans communicated directly with West Germans: The communicative effort that was required for smooth interaction has always been on speakers from the east, in terms of clarifying, repeating, elaborating4. 1.1. Recent discussions More recent observations (Böhm 1992; Ylönen 1992; Blei 1991) seem to point in the direction that differences in relation to how east or west Germans communicate do indeed exist. They do not support the 'Sprachspaltungsargument regarding linguistic differences between East and West Germany, but they are in favour of the notion that divergent tendencies took place, and that the effects are still evident today. The general understanding is that it is quite easy to identify whether someone is from the east or the west from their general behaviour, and that east and west Germans are quite aware of this fact (Schnibben 1993). However, when asked to specify actual differences people are often lost for detailed description, and normally some vague ideas are expressed about general behaviour, the way they dress, behave or act in particular situations6. It has been assumed that the similar, opposing patterns are also a feature of the language used by east and west Germans. There are supposed to exist apart from the obvious dialectal differences - certain linguistic indices that

Manfred Hellmann (1989, 36) (eminent German linguist and) expert on linguistic variation between east and west, sums this phenomenon up as "Leistung sprachlicher und kommunikativer Reparaturarbeit" (need for linguistic and communicative repair work), which existed before 1989, and continues in some areas even today. 'Sprachspaltung' (language division) is a term applied in the past by the more extreme proponents of the existence of linguistic differences between FRG and GDR as an inevitable consequence of opposing ideological allegiances. (Cf. esp. older publications by Weiskopf 1955; Moser 1964; and for a general overview Hellmann 1976). Perhaps the most commonly encountered perceptions of the differences in recent times have been that whilst west Germans are self-assured and assertive, east Germans are supposed to be quite the opposite: deferential and insecure.

210

IV. Cross-cultural Aspects of Hedging

mark the speech of an east German, but again, when trying to ascertain what these markers are, the matter becomes more difficult to assess.7 The emerging picture is thus that research prior to 1989 on the difference between East German German and West German German was rather heterogeneous. The reader is referred to Hellmann (1987) for a comprehensive summary of earlier research publications and von Polenz (1993) for work done since 1989. Opinions have ranged from "Spaltungstheorie" (theory of the total division and drifting apart of the two varieties of German, e.g. Moser 1961, 1964; V. Schmidt 1978; Klemperer 1952 representing the extreme position; whilst Hellmann 1973, 1984; Schlosser 1990a, b; Fleischer 1987 have taken a more moderate stance) to a 'no-differences-at-all' position (esp. Oschlies 1989; Drosdowski 1991; Domaschnew 1991).8 Depending on the linguistic level under scrutiny, either viewpoint may be justified, but overall, the truth lies somewhere between the two extremes.9

2. Semantico-pragmatic explanations The difficulty, however, is that many words are coloured by previous usage (Lerchner 1991; Fix 1993a and b), in that they were used in the past with different connotations, sometimes even denotations.10 It has been argued that a certain reluctance on the speakers' part to use large parts of his lexicon will affect his overall speech production and contribute to what may be termed 'Kommunikative Unsicherheit (communicative uncertainty). Assuming that east

' Two interesting examples from my own data may be drawn on to illustrate this point: during a discussion between an east German and several west Germans the use of the word "Hoffnung" (hope) is the cause of some misunderstanding, until the speaker from east Germany tries to clarify: "kann es denn nicht sein, daß dieses Wort Hoffnung in diesem Teil Deutschlands etwas anderes bedeutet als bei uns?" (isn't it possible that the word hope means something different in this part of Germany than in ours?). Likewise, a speaker from east Berlin reflects: "...weil, wenn wir über bestimmte Dinge reden, wir immer erklären müssen, wie wir das meinen..." (because, whenever we talk about certain things, we always have to explain what we mean). ° While extreme divisionist positions were clearly ideologically motivated, concentrating too much on political and societal differences and exaggerating the effect party controlled language might have on the people of East Germany, the opposite view of a homogeneous German language - except for some minor lexical differences - seems too superficial. 9 It has been repeatedly claimed that the main differences exist at the semantic, lexical and pragmatic level, while syntax and phonology have hardly been affected at all. Estimations regarding the total number of GDR-specific words and expressions before the "Wende" ranged from 800 - 2500 (Hellmann 1990,268), and in certain areas of the lexicon (e.g. collective pronouns) differences amounted to more that 74% (Frein-Plischke 1987, 311). 10 Examples of terms requiring re-labelling, expressions that denoted "something" in the old, no longer existing, system, and that were often narrowly defined in a Marxist-Socialist sense, are numerous; they are still semantically affected (Schlosser 1990a), and speakers may be reluctant to re-apply them in a new context

H. Kreutz, Some Observations on Hedging Phenomena

211

Germans have to re-learn considerable parts of their lexicon11, and bearing in mind that evidence is almost exclusively from written language data, it seems logical to assume that: a) re-learning would also affect the spoken language, and b) the domain of usage is not any more exclusively the public sphere.

Saying that the communicative situation (^ Kommunikationssituation} in Germany today is not the same as the one before those momentous events in November 1989, or even in the months following the 'Wende', would be stating the obvious. The east German society has undergone dramatic change, experiencing drastic assimilation into the west German system, which has led to changes throughout and, not unimportantly, in terms of language. This is hardly surprising: the reciprocal relationship between language and society has been documented on many occasions (e.g. Fishman 1971; Hymes 1974; or Gumperz 1971) and, interestingly enough, especially in the East-TWest- German context: prior to 1989 the impact of social organisation and ideological structuring on the German language was one of the most productive areas of linguistic investigation. It seems that since 1989 little has changed: having had limited access to the East German speech community, researchers now engaged even more intensively in the examination and analysis of what may be broadly termed the German communication situation ('Deutsch-Deutsche Komunikationssituation'}. In the last five years more than 250 publications were produced, a certain indication that interest in the field has not waned. There is a complete re-structuring of an entire society in progress, and this makes for linguistic change par excellence, linguistic change that can be observed. Hellmann (1993, 188) in an analysis of written discourse (Leipziger Volkszeitung) lists a range of what he calls "Neubildungen und neue Gebrauchsweisen schon vorhandener Wörter, (...) Wiederbelebung früher gebrauchter Wörter" (new formations and new application of already existing words, (...) revitalisation of words formerly used). In this context it seems natural to raise the question to what extent communication in general, not just media communication, was affected by the transition from an old GDR-Weltbild to a new western one. Clearly there existed remnants of the old communication culture, there came into existence new taboo expressions and prevailing discourse patterns which criticised the old and new social system. With the advent of democracy virtually all East German lexemes that had to some degree become specific to the political or economic system acquired new or partially new meaning (Schlosser (1993, 219) calls it "neuer Wert" new value, connotations) in a much changed society. Perhaps a more modern/different linguistic usage became superimposed, and with it old ways of thinking ('Denkweisen/'-muster*) and linguistic habits ('Spracbgewohnheiten') 11

Estimates on the number of new items (single lexemes or whole expressions, including entirely new concepts) that need to be acquired by east Germans range from 2000 to 4000 (Fix/Schlosser, personal communication), permeating all aspects of life in a pluralist society, be it applying for a job or unemployment benefits, enrolling at university or attending parent-teacher interview: the social and administrative framework, and the terminology that accompanies it, have changed.

212

IV. Cross-cultural Aspects of Hedging

became replaced by new ones. Important terms from the public and official domain had always been taken up by organs of public discourse, and these became, virtually overnight, liberated ('befreit*). The ideological predetermination of the German language in the east gave way to a liberation of public discourse.12 Schlosser has noted on several occasions that linguistic exchange from top (public/official) to bottom (private/every day) was virtually impossible. The organs of public discourse claimed in terms of semantic occupation ('Bedeutungsbesetzung') much of the ideological, political, and economic vocabulary. In the GDR every day language (Alltagssprache') had entirely different role and status than in the FRG, since interference in public discourse and communication from the private sector was impossible. The formulaic language of the SED party apparatus became taboo once its restrictions on the population had been lifted after November 1989. Some would claim, however, that this process had already taken place in the private domain well before the Wende. It is perhaps interesting to note that the assumption that political programs and economic information would be easily understood by east German lacked foundation: linguistic differences, especially at the pragmatic level, between East and West Germans had been well documented, but were simply ignored. The exposure to different forms of public discourse forms, to name but one area, predicted a lack of understanding and comprehension. This communicative inexperience with western discourse forms, naturally, did not simply affect reception, but, and perhaps even more so, production. It is here where most of the linguistic work along an alternative line of investigation still remains to be done. It is a fact that the reality of "eine Sprach- aber unterschiedliche Kommunikationsgemeinschaften" (Fleischer 1983, 258, one speech- but different communication communities) were simply ignored. How can a people who publically proclaimed that "in unserem Lande ist die Kommunikation zwischen Staat und Gesellschaft offensichtlich gestört" (Fraas 1993, 260, in our country the communication between the state and society is obviously distorted) be expected to immediately establish a rapport between itself and a new state it has only limited knowledge of?13 The term 'Entnennung' (to my knowledge first introduced into the discussion by Schlosser (1990a) has been repeatedly mentioned in discussion on the new communication culture of the former GDR: a de-labelling and liberation of one's own language from the ideological burden (Ballasf) of the regime. Schlosser (1993, 223) describes this as follows "Gratwanderung zwischen Realsozialismus und Kapitalismus existierte auch sprachlich. Es gab eine gewisse Scheu (emphasis my own, and it wiU be argued

It should perhaps be noted at this point that the typically East German division between public and private domain, and relevant linguistic usage also disappeared almost immediately. The emergence of alternative/different discourse patterns ('Diskursmuster') especially in the public domain may well be a reflection of previously existing private 'habits'. Note the separation of "Staat" and "Gesellschaft": thus was the alienation of the government and its people that over time the two lexemes which in the GDR context would be at least partly synonymous have developed separate lexical fields with diverging connotations: the "Staat" negative, the "Gesellschaft" not so.

H. Kreutz, Some Observations on Hedging Phenomena

213

that Scheu and Hedging are intrinsically linked), Begriffe wie Kapitalismus und Sozialismus zu verwenden. Es handelt sich dabei um keinen Zufall: es ist eine Folge der Einsicht, daß diese und ähnliche Begriffe infolge einseitiger und ideologischer Besetzung durch die SED mehr oder weniger unbrauchbar geworden waren (...) die DDR-Opposition (...) mußte auf ganze Begriffsfelder verzichten, oder sie durch Formulierungen ersetzen, mit denen man noch keine allgemein klare Vorstellungen verbinden konnte."^

Again, it would seem that this combination of uncertainty in terms of communicative situation, context, and culture might be conducive to the production of more careful modes of expression and hedgings. I shall return to the notion of'kommunikative Unsicherheit1 below. First, however, I would like to turn to the language data analysed for the purpose of the present study.

3. Description of the data

The data consists of approximately four hours of video-taped TV-discussions, round-table-talks and talkshows. It was chosen for a number of reasons which have been discussed in more detail elsewhere (Kreutz 1993). The decision was made to opt for video recordings of TV discussions which were held during the six months between October 1989 and March 1990, the historical period which has since been referred to as 'Wende''. Initially interviews with 'Ubersiedkr* (former GDR-citizens resettling in West Germany) had been planned, with a view to how they functioned communicatively in their new environment (one area of interest was the switching between registers (^innere Zweisprachigkeit internal bilingualism), which a number of researchers had claimed was a definite feature of former GDR-German (Clyne 1984; Schlosser 1990a; Hellmann 1987, 1989), but which had not as yet been subjected to systematic research). It quickly became evident that the interview situation proved too artificial, informants were not displaying their normal speech behaviour in the presence of an outsider, and they also showed a strong tendency to integrate themselves as quickly as possible into the West German society. The discussions involve large numbers of (predominantly young) East German speakers debating a range of topics pertinent to the social and political changes that were taking place at the time. There was a definite advantage in terms of quantity: by gaining access to the video recordings at the IDS in Mannheim15 a much more comprehensive corpus could be obtained than would have been possible if the material had been collected by means of personal interviews. One main drawback of the use of The tight rope walk between real socialism and capitalism also existed in linguistically. There existed a certain shyness to use terms like capitalism and socialism. This is no accident: it is the result of a realisation that these and related terms had more or less become unusable, due to ideological predetermination by the SED party (...) the GDR opposition (...) had to do without entire lexical fields, or replace them with expressions which did not fully lend themselves to the production of clear representations IDS = Institut für deutsche Sprache, the National German Language Research Institute.

214

IV. Cross-cultural Aspects of Hedging

TV-language material is the difficulty to classify the language as naturally occurring speech. However, if one applies the criteria 'spontaneous' and 'emotional', certain claims at representativeness are possible1·6. Härtung, prominent East German linguist, claims that17"solange davon ausgegangen werden darf, daß eine Äußerung in der vorliegenden Form auch in anderen Kontexten und Situationen möglich ist, darf vorausgesetzt werden, daß sie allgemeine Gültigkeit hat" (as long as it can be assumed that an utterance in its existing form is also possible in different situations and contexts, it may be presupposed that they have general validity (Härtung 1990, 448)). Similar approaches (i.e. 'interpretative Analyse') in conjunction with Mediensprache' (language in the media) have been used successfully elsewhere (e.g. Linke 1985; Mühlen 1985; Hoffmann 1982; Holly/Kühn/Püschel 1986 on talkshows, and 'politische Fernsehinterviews' (television interviews with the participation of politicians). As regards the calibration and categorisation of the data, it should be mentioned that observations regarding modifying expressions such as hedgings are not random, but viewed in a wider thematic context and according to individual speaker profiles. Information relevant for the exemplification of the analytical method applied precedes the transcription extract in the Appendix.

4. Theoretical background and some results The methodology applied is essentially a discourse analytical one. Attention has been paid to the wider context, the overall topic, and the type of interaction between individual speakers, for it has been assumed that these factors have a bearing on the occurrence/frequency of 'items'. Initially, these items were categorised according to the various types of 'Spezifika', whereas at a later stage of the analysis they were subjected to closer examination in relation to instances of communicative uncertainty. 'Kommunikative Unsicherheit (communicative uncertainty) may be assessed by various means, and in the present context the analysis shall be confined to the one area of modifying devices such as hedgings and downtoning particles. (See Kreutz 1993 for a discussion of speech pauses and hesitation phenomena, i.e. the occurrence of hesitation in the environment of particular - marked items). The process of linguistic liberation may be well advanced, however, remnants and terminological, lexical, and semantic gaps are likely to persist for ^ R. Schmitt (1993, IDS Mannheim, personal communication) uses the term "Situationsemergenz", (dynamics of the situation) which basically means that in most conversational settings, especially when topics of personal or emotional relevance are being discussed, the speakers) will at some point become oblivious to the formality of the surroundings and produce natural and spontaneous speech. *·' This claim of being able to draw general conclusions from linguistic data as long as one may assume that an utterance is likely to occur in the same form in any other context may be somewhat sweeping. The point he is making, however, seems clear: It allows us to analyse a much wider range of spoken language material (esp. institutionalised discourse) within existing discourse analytical frameworks.

H. Kreutz, Some Observations on Hedging Phenomena

215

some time, and the necessary linguistic inventory at the disposal of east German speakers at the time of the Wende and thereafter has essentially been a West German one, and in the context of the current 'discourse of discontent' and Ostalgie' ('eastalgia', an obvious pun on 'Nostalgie' nostalgia) a general reluctance to accept lexemes and expressions that have prior connotations may be presupposed. After all, it is a feature of many discourse forms and genres to resort to more careful means of expression and argumentation if there is doubt or uncertainty involved. The linguistic means resorted to are often what has become known as hedging (Coates 1986; Holmes 1984; Lakoff 1975; Simpson 1993; Clyne 1991).

5. Hedging As will be shown in examples of analyses below, the use of terms, expressions and lexemes is not properly assessable without relating to situation and context. Schlosser (1993, 229) describes the linguistic habits of the new east Germans as characterised by "(...) Abgrenzung sowohl gegen den Zwangskollektivismus der SED, als auch gegen die Überbetonung des Individualismus in der westlichen Gesellschaft (...)" (dissociation from the forced collectivisation of the SED, and from the over-emphasis of individualism in western societ ^· This is communicative no-man's land: caught between "Zwangskollektivismus der SED" and "Überbetonung des Individualismus" leaves the speaker in a situation where language use becomes severely affected by the state of flux it is in. This is exactly the situation of east Germans during and after the Wende, and to some degree even still today. The struggle for new linguistic and expressive means, the awareness, self-consciousness of one's own linguistic, communicative, pragmatic inappropriateness is bound to generate communicative caution and uncertainty, and creates an excellent environment for modalisations and hedgings. Particularly in oftlcial/semi-public contexts and discourse forms the inferiority of the every-day language ('Alllagssprache') soon became evident: "Der bis dahin weitgehend auf Nichtoffizielles beschränkten Alltags spräche fehlte der Halt, den das SED-Offizialidiom nicht geboten hatte." (every-day language, hitherto limited to non-official contexts, lacked the stability which had not been provided through the official SED-register (Schlosser 1993, 230). The present paper will try to explore markedness of East German speakers via an alternative line of linguistic investigation from a more pragmalinguistic perspective. Some suggestions to that effect were proposed by Good (1993, 249), who is critical of the term 'Vereinigungsdiskurs' (discourse of unification): many observations are purely anecdotal, virtually everybody feels entitled to comment on the language situation between East and West Germany, without proper empirical backup. Not only individual lexemes are affected by potential divergent development of the German language in East and West, but also other linguistic levels (Sprachebenen), levels of communicative behaviour. Naturally, a description of those requires special analytic tools. Good (1993,

216

IV. Cross-cultural Aspects of Hedging

250) recommends a pragmatic approach, which may capture more completely the reasons behind the communicative difficulties between east and west Germans. In general, in the situation of 1989 and thereafter a high degree of communicative uncertainty may be expected from speakers from east Germany trying to cope with the new. People who share the same mothertongue feel that they cannot communicate properly. Lexical differences clearly existed, however, the paradigm developed mainly by Hellmann (Hellmann 1973; Hellmann 1976; Hellmann 1984), and Schlosser18, as useful as it has been, could perhaps not be fully sustained, since, as became quickly evident, lexical differences were not the main impediment to effective communication. Fraas (1993, 260) justifiably asks "Woher kommen die zahlreichen Kommunikationsprobleme?" (where do all the communicative difficulties arise from). It seems obvious to assume that, in view of the vast amount of research undertaken in the area of East-West German language divergence, and the limited insight the studies afford, that the problems east and west German speakers are facing when trying to communicate are not entirely at the linguistic surface level. Different experiences, ways of life, patterns of behaviour (^Erfahrungswelten, Lebensweisen, Verhaltensmuster'} have developed during four decades of separation. There is the well known and well documented phenomenon that east Germans perceive the meanings (denotation and connotation) of certain words, their implications and presuppositions differently from their west German counterparts, and vice versa (e.g. von Polenz 1993). Speakers often are not aware of these differences, they communicate, but that which is being communicated ('Kotnmunikaf) is not necessarily the same, misunderstanding may be based on semantic and pragmatic differences, there is potential for conflict. The different experiences of social and political developments for over four decades would have facilitated the formation of different standards and expectations, which are reflected in disparate communicative behaviour of East and West Germans. There exists psycholinguistic evidence that difficulties in communication may be caused by unfamiliar text segments (and I would argue that text here is to be understood in the broadest possible sense), such as lexemes or complicated syntagms. Subsequent gaps have to be bridged by drawing on existing knowledge. (This phenomenon of sometimes only vaguely understanding the gist of a text/discussion is a well known phenomenon to all language users - in areas that require specialised language or jargon). Fraas (1993, 261) quotes Strohner: "bei der Aktivierung eines Wissensbereichs werden neue Wissensbereiche hinzuaktiviert, wenn es bei der Herstellung von Sinn notwendig ist" (when activiating one area of knowledge, if necessary, new areas become activated in the construction of meaning) and she continues emphasising the "Selbstorganisation des Textverarbeitungssystems zur Erschließung neuer Wissensbereiche" (self-organisation of the text-processing system in order to It has to be stressed, of course, that especially Hellmann's categories were never intended to be applied in a pragmatic analysis.

H. Rreutz, Some Observations on Hedging Phenomena

217

access new areas of knowledge). In the case of east-west German communication problems the effect of disparagingly used lexemes and their connotations (kxematische Bedeutung$em>eiterungen/-dwergenzen') are, naturally, of much more importance than comprehension difficulties as a result of syntagmatic complicatedness (which are, it seems, predominantly intralingual ({innersprachlich'), across the two Germanics. Still, an analysis beyond the word level is surely required. The one crucial factor that needs to be taken into account in the context of linguistic unification is the lack of information and incompatible knowledge bases { Wissensbestände') which are often not recognised, simply because a mutual basis for understanding each other is simply assumed, irrespective of whether it really exists or not. The resultant conflicts are often noticed retrospectively. Here one may want to look for the key to conflicts in communication between speakers of the same native language, conflict that is not uncommon in cross linguistic settings between native speakers of different languages. In those non-native settings, however, speakers expect to encounter difficulties and are prepared to compensate for them and accommodate accordingly (incidentally, accommodation theory (esp. Giles and Smith 1979), may also provide some interesting insights). Speakers adjust a priori to differences in mentality, cultural and social background. If one wants to believe the frequent accounts of difficulties between east and west, then, there seems to be much less common ground for unhindered communication than originally expected. It may be argued that the most important pre-requisite for efficient, unhindered communication is, apart of course from sharing one and the same language, shared experience, world knowledge, and generally accepted rules of communication. This line of research may indeed provide insightful results, as it requires researchers to take into account different experiences and societal background ('Erfahren*- und Erlebniswelten der Kommunikationspartner') when attempting to explain communicative discrepancies between east and west. It has been suggested that comparison of an analysis of proto-and stereotypical perception of items provides a useful concept (esp. Fraas 1993), since differences may serve to actually detect the different background experiences. However, care needs to be exercised, and one may wonder to what extent it can be claimed that "unterschiedliche semantische Auffüllung gleicher Wortkörper mit den zugrundeliegenden unterschiedlichen Stereotypen und prototypischen Effekten (Fraas 1993, 261) (differing semantic 'mapping' of identical linguistic forms according to their underlying basic stereotypes and prototypical effects) can actually be fully traced. Although the present paper will not draw so much on neighbouring disciplines, its direction in terms of linguistic orientation is broad enough in that it draws on a pragmatic framework. Any investigation into hedging phenomena, however, is still primarily linguistic: the focus is primarily on linguistic structure, while the analysis of its communicative and pragmatic force, function and purpose comes second. In the context of the, in my view, still widespread communicative difficulties between speakers of what once were the FRG and GDR,

218

IV. Cross-cultural Aspects of Hedging

the use of hedging may serve as a starting point: if we accept Simpson's (1993, 47) view that modality (and one might argue that modality cannot really be separated from hedging) "refers broadly to a speaker's attitude towards, or opinion about, the truth of a proposition expressed" and "extends to their attitude towards the situation or event described by a sentence". Clearly, modality, and therefore hedging, are part of the interpersonal (social/affective) function of language. There cannot be an argument any more about the fact that much attention needs to be paid to the context in which an utterance is performed, and it is generally accepted that contextual information is also required for the processing and communication of meaning. This, of course is adding "a pragmatic dimension to our semantic base" (Simpson 1993, 127). Hedges have an important function in terms of softening the impact of the message on the addressee. The pragmatic significance of hedging constructions has been well documented (e.g. as markers of politeness and mitigation), and are highly relevant to the way a message is shaped and delivered by the speaker. It may be argued that, at the more unconscious level, a further function of hedging and modality is that of expressing a degree of communicative uncertainty. In the light of what has been discussed in above in relation to 'unterschiedliche Erlebnis- und Erfahrungswelten' (different background experience), it should be clear what is meant by this: a speaker who suddenly finds her/himself out of her/his communicative depth, as it were, might generate pragmatic presuppositions (Simpson 1993, 127) which may contain meanings that do no "attach conventionally to particular items or constructions which derive from their normal contexts of use", simply because a "normal context of use" for all speakers of German east and west does not exist. Pragmatic presuppositions are part of the shared conventions of language use (Simpson 1993, 128), and not in the formal patterns of its logical structure. This raises the question why a speaker should use a linguistic strategy (modalised and/or hedged) which demands greater processing effort from the interlocutor. Certainly directness, in purely linguistic terms, would lead to greater clarity in interaction, nonetheless, it is a strategy which normally is perceived as being impolite and tactless. The use of indirectness is one of the chief cross-cultural realisations of politeness, and it is a strategy by which speakers can to some extent mitigate the effect of their utterance, an essential strategy in situation where one feels less secure. In the case of East German (and as far as the present data is concerned, young) speakers, hedging and modality are an important communicative strategy, and their analysis may serve to shed some light on the linguistic differences and communicative difficulties between East and West German mainly at the pragmatic level. The main objective of the present paper then, apart from a broad discussion of the east-west German communicative situation, is to provide an impression of to what extent the speech of young east Germans is linguistically and/or pragmatically 'marked'. The notion of (especially social) markers in speech is not new and has been the objective of many publications. As a rough category of analysis the concept of'communicative uncertainty' has been introduced, which may be observed from "den Spuren eines komplizierten und problema-

H. Kreutz, Some Observations on Hedging Phenomena

219

tischen Formulierungsprozeß" (traces of a complicated and problematic formulating process, Kallmeyer/Schmitt 1992, 1). 5.1. Modifying devices: Hedging and downtoning particles: some findings Apart from hesitation phenomena/speech dysfluencies related to lexical encoding, the occurrence and application of hedgings is an interesting area to look at: the formulation of information and opinions, and the way in which they are presented may offer some valuable insights into the differences in terms of linguistic behaviour between east and west German speakers. As indicated above, numerous publications in recent years dealing with the issue of linguistic differences between east and west Germans have made mention of'kommunikativ relevante Mentalitätsunterschiede' (von Polenz 1993; Schlosser 1993) (differences in mentality which are relevant for the communication between members from the formerly two speech communities), citing examples from observations by Resendiz, Ylönen, Teichmann and Lerchner: west Germans are supposed to be more self-assured, more controversial, dominating, rhetorically well versed, speak faster and sound more colloquial. East Germans, on the other hand, are said to be more inhibited, insecure, sceptical, indirect, matterof-factly, monotonous, compliant, seeking common ground, more prone to be silent in interaction with west Germans. Discounting the fact that these observations have by no means been statistically verified, they nonetheless seem to suggest that characteristics such as indirect or seeking common ground, insecure, inhibited may manifest themselves in the way of marked usage of hedgings. There have also been concerns expressed about "inkompatible Wissensbestände" (incompatible information/knowledge background) the inevitable "Rückgriff auf die alte kommunikative Kompetenz" (reverting to the communicative competence of former times) (Lerchner 1992, 323). Taking these observations and assumptions into account, it may be hypothesised that east Germans would, as a result of being out of their "communicative depth", as it were, resort to different usage of hedgings and modality. Hedging in the present context include the occurrence and combination (e.g. "double and triple hedgings", cf. Clyne 1991, 63) of modality markers, downtoning particles and, in general, means of indirect linguistic behaviour (especially the use of hedged performatives). If one revises the literature on hedging phenomena, be it in relation to scientific or every-day discourse, it becomes clear that one important function of hedgings pertains to careful presentation of the speaker's opinion, and the downtoning of one's expressive force. (Milderung des eigenen Ausdrucks'}. Clyne (1991, 57) states that "the function of the phenomena under consideration are to reduce the weight or certainty of the propositions and to relieve authors of some of the responsibility for statements they are making." If, for present purposes, "author" is substituted with "east German speaker", it becomes possible to look at how they may try to "relieve themselves of some of the responsibility" for their utterance. Similarly Sandig (1979, 84) on particles: "coloring (= downtoning) particles are not indicators for the illocutionary

220

IV. Cross-cultural Aspects of Hedging

forces of the speech acts but modify the pure illocutionary forces for the varying purposes of interaction." Some commonly observed constructions in the present data used by East German speakers to modify their speech (i.e. hedgings and modality) were:19 ich muß sagen, dazu kann ich sagen, was jetzt weniger konkret nennbar wäre, natürlich muß man abwägen, da muß ich jetzt speziell sprechen, ich wollte eigentlich folgendes sagen, natürlich muß man das abwägen, man darf sich nicht dazu verleiten lassen, die Orientierung an der Presse kann nicht ausreichen, es entsteht die Vorstellung, das muß man so sehen but also typically GDR: das kann ich jetzt nicht einschätzen ich will nicht sagen, daß die DDR nichts einzubringen hat

The use of hedgings and downtoning particles as modifying devices are important at the discourse level.20 The pragmatic force of an utterance is altered to an extent where the interlocutors may feel compelled to take charge of the conversation/discussion. This, it has been argued, is a common feature in east-west contact situations today, and the examples above on divergence speech behaviour seem to illustrate this point. FRG and GDR societies were fundamentally different in their organisation. Expressions denoting GDR-specific speech acts (e.g. 'EINSCHÄTZEN; DURCHSTELLEN to assess; to pass on to a higher authority) may prove particularly helpful to certify more overt instances of GDR markedness. One striking feature of the data so far is the fact that certain speech acts which feature prominently in discussions - KONSTATIEREN (to make a point), NENNEN (to quote, to mention), ABWÄGEN (to reflect), DIFFERENZIEREN (to differentiate), WARNEN (to warn), AUFFORDERN to ask, to challenge) - are frequently realised by using illocutionary verbs in the form of hedged performatives, creating an effect of reduced speaker involvement, or responsibility, and Objectivisation' of utterance contents. Of additional interest may be the fact that such indirect and modified expressions frequently occur at the beginning of turns, which possibly serves as the speaker's method of expressing a certain degree of reservation about the contents of the utterance rather early, perhaps even before the internal process of finalising the articulatory plan of the utterance has not even been finalised. ° Translation of these examples would be of little value, as they function in a very different way in English (e.g. as evidentiality markers), sometimes even underpinning one's authority, quite in contrast to the modifying function they serve in German). 20 An example of hedging in conjunction with the particle "schon" (in this context used to downtone the statement and translatable as "I suppose") is the following utterance from my own data: "...und ich bin Tochter einer muß ich schon sagen kommunistischen Familie." ( (...) and I am the daughter of a I must say I suppose a communist family). By using "muß" and "schon" the speaker almost apologises for being a member of a communist family.

H. Kreutz, Some Observations on Hedging Phenomena

221

As for occurrence and frequency of downtoning particles, only limited generalisations are possible at the present stage of the analysis. The particle 'eigentlich' (actually) is frequently used in assertive sentences, marking some sort of mild contradiction. A preliminary analysis of hedging and modality markers is illustrated in Table 1 below. While it is understood that the listed items/occurrence are not necessarily exclusively East German, their relatively high incidence seems to point in the direction of divergent German usage when comparing east and West: may be this is what Weinrich called 'Nuancenkompetenz' - it is the small things that can make a big difference. The exploratory nature of this paper has been mentioned before, and since it constitutes a fragment of a much larger study, some remarks on analytical procedures, especially in relation to the information contained in Table I,21 are perhaps in order.22 If we then take a look at Table 1, it quickly becomes evident that the tabulation of such a complex field of communication - the occurrence and function of hedging - is not easily accomplished. The reader is referred to note 22 for an approximate translation of the German terminology used. While this is essentially of limited value, it may nevertheless be of some service to the reader with only a basic knowledge of German. As stated before, however, it has to be remembered that the full meaning and communicative content ('kommunikativer Inhalt*) of items such as downtoning particles and modal verbs depends to a large degree on extralinguistic criteria like context, situation and interlocutor. To fully capture all aspects of the phenomenon in question presents a challenge, and the calibration and presentation is but a first attempt at bringing some order into the chaos. The field of modality, and modal particles in the context of hedging constructions in particular, is wide open, and one is constantly forced to find some kind of footing from which to work. Initially, at the beginning, things often look reasonably neat, there exists a broad understanding and some agreement as to what constitutes hedging, modality, downtoning and so on. Actually trying to pinpoint individual occurrences and single items, however, and working from the individual utterance, is less straightforward. There is also the question of quantification vs. qualification. Hedging and downtoning elements are subject to native speaker interpretation (Burkhardt 1994), and in my view one cannot help but resort to such 'native speaker qualification', as it were. The method chosen for describing the communicative activities of east German speakers, under the conditions investigated, is a correlation of communicative patterns and authentic utterances. Some of these patterns are text-constitutive, some are context-establishing, and others are speech-organising (Holly et al. 1984). When aiming at a description of one area of the communicative competence of an east German speaker, the inductive approach reflected in Table 1 * Some of the examples in the table, in conjunction with the text sample in the appendix are, to some extent, self-explanatory. Three full examples are given in the appendix (in italics and bold print - 14 U, 48 U and 118 P4), and they should suffice to illustrate some of the points being made, although there are numerous examples of this kind which lend themselves to asimilar analysis in the data.

222

IV. Cross-cultural Aspects of Hedging

seems the more rewarding option. By scrutinising the material and classifying the patterns, and by comparing them (albeit intuitively) with the communicative competence of other (west German) speakers the researcher is likely to yield some valuable insights. Table 1 presents an attempt at systematising and analysing the communicative activities of the speaker, with a view to the overarching aspect of 'Kommunikative Unsicherheit' (communicative uncertainty), i.e. hedging and downtoning. The method of investigation followed in principle consists in commenting on text passages (see Appendix), in exemplary form, as typical realisations of specific communicative patterns. There are a great many of such patterns, and by capitalising the verb which represents the utterance act itself23(Table 1 - column 2), the impression may arise that there exists a welldetermined and all-comprising system of linguistic action patterns ('sprachliche Handlungsmuster'). There is, in fact, no complete system to be found anywhere in the literature. Recording the patterns mainly serves as a method of interpretive illustration. These patterns encompass only some of the communicative patterns which are being realised by the speakers when completing their communicative tasks in the discussions, namely such patterns which the speakers saw fit (consciously or unconsciously) to 'hedge'. The present article does not leave scope for an extensive treatment of the concept of'action' ('Handlung*), suffice it to say that action has been analysed as being sense-related ('sinngebunden'), aim-directed (zielgerichtet*), controllable, rule-governed, complex ('Handlung als komplexes Zeichen'), and interpretationdependent ('interpretationsabhängig'). Action pertains to practical as well as linguistic action ('praktische und sprachliche Handlungsmuster^. The question to what extent an all-encompassing concept of action can account for linguistic behaviour, down to linguistic actions in different contexts (hedging and downtoning 22 The following are approximate translations of the German terms and expressions employed in Table 1. 1) (column 2 - 'Utterance Act'): BESTÄTIGUNG affirmation; ASSERTION asserion; EINSTELLUNGSÄUSSERUNG expression of opinion; ZURÜCKWEISUNG dismissal (of argument, point of view); BERICHTEN report; ERGÄNZEN/ERKLÄREN explain/ explicate; ERZÄHLEN/MITTEILEN inform; 2) (column 3 - 'Performative Verb'): bestätigen - confirm; sagen - say; wehren - defend; denken - think; anfangen - begin; ausdrücken - express; abwägen - consider/contemplate; annehmen - assume; 3) (column 4 - "Modality Marker1): möchten - like, want to; müssen - must, have to; überhaupt - at all; so - well; schon - well; eigentlich - actually; irgendwie - somehow; mal well, then; bißchen - little bit; können - can, be able to; wollen - want to; 4) (column 5 - 'Communicative Function'): Betonung - emphasis; Dissens - dissent; Verallgemeinerung - generalisation; Bestätigung - affirmation; Selbstdarstellung - selfportrayal; Hinweis auf (eigenen) Äußerungsbeitrag - reference to (speaker's own) utterance; Äußerungseinleitung - utterance introduction; Einschränkung - reservation/qualification; Begründung - reason/substantiation; Einleitung des Redebeitrags - introduction of (speaker's own) turn; 5) (column 6 - 'Socio-pragmatic Function'): Distanzierung/Solidarisierung - disassocition/ showing of solidarity; Selbsdarstellung - self-portrayal; Rechtfertigung - justification; Kritik - criticism. 23 A widespread convention, see for example Holly 1990.

H. Kreutz, Some Observations on Hedging Phenomena

223

are linguistic actions reflecting communicative uncertainty), remains open to discussion. Some examples of speech act patterns ('Sprechtandlungsmuster') reflecting communicative abilities needed in a wide range of social interaction are Table 1 Location in Text

Utterance Act

Performative Verb

Modality Marker

Communicative Function

Sociopragmatic Function

DN 1/2 BT1

BESTÄTIGUNG

bestätigen

MV2 möchten

Betonung

Distanzierung/ Solidarisierung

DN 1/12 BT

ASSERTION

sagen

MV müssen

Dissens

Solidarisierung

MP überhaupt Madv4 so

Verallgemeinerung

Distanzierung

DN 1/26 BT

3

EINSTELLUNGSÄUSSERUNG

DN 1/32 BT ASSERTION

sagen

MV müssen MP schon

Bestätigung

Solidarisierung

DN 1/36 BT ZURÜCK-

wehren

MP eigentlich

Dissens

Distanzierung

denken

Madv irgendwie

Selbstdarstellung

Distanzierung

DN 2/12 ANT BERICHTEN

anfangen

MPmal

Hinweis auf (eigenen) Äußerungsbeitrag

Selbstdarstellung

DN 2/14 U

ERGÄNZEN/ ERKLÄREN

sagen

Madv vielleicht MV müssen

Äußerungseinleitung

Selbstdarstellung

DN 2/40 U

MITTEILEN

ausdrücken

MPmal Madv bißchen

Einschränkung

Solidarisierung

DN 2/44 U

ERZÄHLEN/ MITTEILEN

sagen

MP also, eigentlich MV können

Einschränkung

Distanzierung

DN 2/48 U

ERGÄNZEN/ ERKLÄREN

abwägen

MV müssen

Begründung

Rechtfertigung

DN 2/116 U

EINSTELLUNGSÄUSSERUNG

annehmen

MP eigentlich MV müssen

Begründung

Kritik

DN 2/118 P4

EINSTELLUNGSÄUSSERUNG

sagen

MP eigentlich MV wollen

Einleitung des Redebeitrag

Rechtfertigung

WEISUNG

DN 1/43 BT EINSTELLUNGSÄUSSERUNG

l DN ^MV ^MP 4 Madv

= Discourse number, 2 = speaker turn, BT = speaker initials = modal verb = modal particle = modal adverb

224

IV. Cross-cultural Aspects of Hedging

DISTANZ VERRINGERN diminishing distance, BEZIEHUNG ANKNÜPFEN establishing relationships, FÜRSPRACHE intercession, BITTEN requesting, RATSUCHEN seeking advice (Hundsnurscher 1993, 379). Returning to Table 1, the categories chosen should now be a little clearer: the utterance act is the actual linguistic act ('sprachliche Handlung') which is being performed, the performative verb and the modality marker constitute the linguistic form which is necessary to perform the act (Handlungsvollzug'). The communicative function then becomes a function of the actual linguistic form, whereas the socio-pragmatic function constitutes itself via the utterance. An example (DN 1/2BT): "Ich möchte bestätigen, daß P." (I want/would to like to affirm that P.). The utterance act is BESTÄTIGEN (affirmation), which is linguistically realised by möchten (modal verb) and bestätigen (performative verb). While the communicative function is one of emphasis, the socio-pragmatic function is one of Distanzierung (disassociation, from the new). The remaining examples listed in Table one can be captured in a similar fashion, and in the case of different communicative competencies between speakers from east and west Germany, respectively, I would argue that the sociopragmatic function of an utterance act serves as the driving force behind linguistic choice.

6. Concluding remarks Whilst the political unification of Germany was completed with great speed and efficiency, the convergence of the two formerly separated communication communities of East- and West Germany is still in progress. Linguistic and pragmatic manifestations of divergent political, social and economic developments may still be detected today. If one looks at the speech behaviour of speakers from east Germany, at how they need to function communicatively in a new all-German context - which is essentially based on former West German norms - their linguistic output seems marked, i.e. they display some kind of communicative uncertainty which may identify them as east Germans. Hedged and modified speech may be interpreted as evidence for the difficulties east Germans experience and which are part of a wider process of social integration. Fraas (1993, 262) sums up: Tür Linguisten ist es ein hochinteressantes Problem, daß Sprecher einer gemeinsamen Muttersprache massive Kommunikationsprobleme miteinander haben. Die Erklärung dieses Problems muß mit benachbarten Disziplinen in Angriff genommen werden. Die Verständigungsprobleme der Deutschen wachsen zu einem immer drängender werdenden gesellschaftlichen Problem aus. Wenn die deutsche Einheit nicht nur ein abstraktes Staatsgebilde werden soll, müssen die Deutschen lernen, sich zu verstehen." (it is extremely interesting for linguists to observe that speakers of a common native language are experiencing massive communication difficulties. An explanation of this problem will have to be tackled in conjunction with neighbouring disciplines. The communication problem of the Germans are developing more and more into a pressing social problem. If the German unity is to become more than an abstract notion of a state, the Germans will have to learn to communicate with each other).

H. Kreutz, Some Observations on Hedging Phenomena

225

If we consider the former East German situation in the light of the new - linguistically not quite - unified Germany we will find areas prone to communicative friction. Hedgings and modalised/downtoned utterances prevail in such situations, not simply because of a need to be polite, but also, and perhaps more so, to save face in a social context that by definition renders 'eastGermanness' as an inferior attribute. This article essentially consists of two main parts, background (preconditions) and current investigation. The relationship has been explored between the specifically GDR communicative background and one aspect of the language which has emerged from this environment - namely hedging, language, therefore, is seen as 'social action' ('gesellschaftliches Handeln*). An exploratory analysis has been presented which may serve to shed some light on the often perceived differences in communicative behaviours of east and west Germans. For this purpose, it has been necessary to look at communicators in a semiinstitutionalised setting and examine their speech in a range of communicative contexts. There remain certain questions which could not be addressed, due to the customary limitation of scope and length: 1) 'patterns of communicative action' ('Kommunikative Handlungsmuster*) - a description of such patterns can take place under a formal aspect or under an illocutionary aspect. For example, BESTÄTIGUNG affirmation may have a range of illocutionary functions, such as emphasis, reinforcement, ascertaining comprehension ('Verständnissicherung'), etc. The dichotomy of function and form, as expressed in Table 1 captures this problem to a limited extent only. 2) 'communicative activities' (Kommunikative Handlungen') are of a rather complex nature, and a simple classification of such activities merely under speech act categories would not suffice. One needs to resort to discourse categories if one wants to capture more completely various forms of behaviours and, especially in the context east-west German, the type of communicative activities which are relatively new, as a consequence of social, political and institutional restructuring. To describe the communicative competence of a person in its totality may not be possible at all, even if one was to develop a complete inventory of speech act types or dialogue types, drawing on situational contexts for their realisation. It is, however, possible, to describe segments of a person's communicative competence, by looking at communicative behaviours in a limited range of settings. One area which has emerged as perhaps more conspicuous than other in the context of Ost-West-Kommunikation may be that of hedging, as realised through communicative uncertainty by modalisation and downtoning.

226

W. Cross-cultural Aspects of Hedging

APPENDIX In the following I will present an example of how certain segments have been interpreted. I shall concentrate on two lexemes: "progressiv" (progressive) and "Reisekader" (cadre, group of specialists with certain privileges who were politically reliable, here: to travel abroad to nonEastern Bloc countries). Whilst "Reisekader" was a GDR-specific lexical item denoting "something" that had no West German equivalent, the lexeme "progressiv" is very much a specific occurrence of a semantic nature: it is/was, of course, known and used by both speech communities, but, as it were, with different shades of meaning, or connotations. I will return to this in the interpretation below, first, however, it might be useful to reproduce the transcribed segments in which they occur. The segments are taken from the show "Doppelpunkt", a talk show for younger (up to the age of about 35) viewers. The show was broadcast on 13. Nov. 1989, as a live transmission. There is before a participating studio audience an inner round of four speakers (two male, two female) who three weeks earlier had fled the GDR via Hungary. The main theme for discussion is the faltering East German state, and reasons for choosing to live in West Germany. Transcriptional remarks: 13, 14, etc. MS U

-

P4

-

/./ /../ /.../ etc.

-

number of turn Michael Steinbrecher, compere of the show Uwe, 27-year old East German, member of the inner round of guests (P = "Publikum" = audience) young (age unknown) member from the audience who is participating in the discussion silent pause, 500ms silent pause, > 1 Sec. omitted segments emphasised segments (my emphasis, for illustration purposes only)

13

MS:

wie war das bei euch im Beruf so?

14

U:

da muß ich vielleicht noch etwas dazu sagen /./ bei mir war das so also /./ entstehen Eindrücke in der BRD daß viele aus der DDR kommen weil /../ materielle Gründe der Anlaß sind /./ ich denke aber die materiellen Gründe stehen hinter den ideelllen Gründen weit zurück /../ es ist so /./ nach meinen Erfahrungen zumindest..././ aber viel mehr und viel wichtiger finde ich die Tatsache daß ideele Dinge /./ ideele Werte eine Rolle spielen /./ daß man also gezwungen ist /.../ in der DDR staatskonform zu leben /../ ohne /../ mit dem Gesetz anecken zu müssen /./ und mal so auf alles das alles reagieren muß

47

MS:

haste denn mal versucht auszubrechen?

48

U:

ja gedanklich schon /./ und auch in Handlung aber /./ natürlich muß man abwägen 1.1 ganz klar abwägen zwischen dem /./ was einem da nützen könnte /./ und dem /./ was einem unheimlich schaden könnte /./ und ich habs gerade gesagt wenn man ausbricht dann ist man weg dann ist der Zug abgefahren dann ist es unheimlich schwer wieder aufzuspringen /./ wenn man einmal den Kontakt verloren hat und einmal in diesen Kreis /../ derjenigen gezählt wird die nicht mehr systemkonform sein können /../ beziehungsweise mit denen politisch /../ schwierig zu diskutieren ist /./ oder die eben ganz neue Ideen bringen /.../ auch progressive Ideen /../ konformistische Ideen /.../ dann ist es schon sehr schwierig da etwas zu machen

H. Kreutz, Some Observations on Hedging Phenomena

227

117

MS:

noch etwas aus dem Publikum?

118

P4:

ich wollte eigentlich auch noch was dazu sagen /../ zum Beispiel /./ bei mir war der Grund daß ich abgehauen bin /./ aufgrund dessen /..../ weil also man ist kein Reisekader gewesen verstehen sie /.../ als

There is atypical hesitation after "konformistisch" (conformist), which seems to point to some sort of terminological confusion: whilst above (14, underlined) the negative connotation of staatskonform is evident, the speaker (U) later attempts to label "konformistisch" (48, underlined) as "progressiv" (in the sense of "ganz neue Ideen", also note the enthusiastic undertone). The speaker anticipates a brighter, non-socialist future for East Germany, but is somewhat inconsistent in his attempt to verbalise this. Here it is useful to recall the narrowly defined, predetermined meaning of "progressive' and "progressiver Fortschritt" in Marxist philosophy. Historical Materialism understands "progressiv" and related derivation as part of a predetermined societal process. The speaker's predicament is apparent from his intention to convey meaning by a form that is colored by prior (marxist) usage, and which thus may only be partly suited to express the intended message adequately. He is also at risk to trigger unwanted associations and old (socialist) concepts among his peers, instead of new, emerging ones. Later in the discussion he solves this problem: "reformistische, progressive Gedanken" become "westliche Gedanken" (reformist, progressive ideas - western ideas). The "Reisekader" example is of a somewhat different (purely lexical) nature. Since there is no equivalent west German concept, a simple substitution of form becomes impossible. The speaker seems to realise this after beginning the verbalisation process, which results in two unusually long silent pauses and ultimately in communication breakdown. The compere interferes and appropriates the next turn.

References Ahrends, Martin (ed.) 1989 AEseitiggefestigt. München: dtv. Andersson, Sven G. 1983 "Deutsche Standardsprache - drei oder vier Varianten?", Muttersprache 93, 259-83. Bauer, Dirk 1990 "Zwei deutsche Staaten! - Zwei deutsche Sprachen?", Deutsche Sprache 18, 228-240. Blei, Dagmar 1992 "Neue Offenheit - neue Herzlichkeit. Beobachtungen zum Sprachgebrauch der Ostdeutschen", Deutsch ah Fremdsprache 29, 49-51. Boehm, Christian 1992 "Der Broiler lebt. Die deutsche Sprache im Wandel zwischen DDR und BRD", Zeitsdmflßr Germanistik 2, 320-340. Burkhardt, Amin 1994 "Abtönungspartikeln im Deutschen: Bedeutung und Genese." Zeitschrift für Germanistische Linguistik 2, 129-151. Clyne, Michael G. 1992 "German as a pluricentic language. In Clyne, M. G. (ed.): Pluricentric Languages. Berlin, New York: Walter de Gruyter. Clyne, Michael G. 1991 "The sociocultural dimension: The dilemma of the German speaking scholar", in Schröder, H. (ed.): Subject-oriented Texts. Berlin, New York: Walter de Gruyter. Clyne, Michael G. 1984 Language and Society in the German Spreaking Countries. Cambridge: Cambridge University Press.

228

IV. Cross-cultural Aspects of Hedging

Coates, Jennifer 1986 Women, Men and Language. London: Longman. Debus, Friedhelm; Hellmann, Manfred and Schlosser, Horst-Dieter (eds.) 1986 Sprachliche Normen und Normierungsfolgen in der DDR. Hildesheim: Olms. Dieckmann, Walther 1989 „Die Untersuchung der deutsch-deutschen Sprachentwicklung als linguistisches Problem". Zeitschrißfür germanistische Linguistik 17, 162-81. Domaschnew, Anatoli 1991 "Ade, DDR-Deutsch! Zum Abschluß einer sprachlichen Entwicklung", Muttersprache 101, 1-12. Drosdowski, Günther 1991 "Deutsche Sprache in einem geteilten Land. Beobachtungen zum Sprachgebrauch Ost und West in der Zeit von 1945-1990", in Stötzel, G. (ed.): Die deutsche Frage. Sprachwissenschafiliche Skizzen. Sprache und Literatur in Wissenschaft und Unterricht 22,21-35. Fishman, Joshua 1971 Sociolinguistics. Rowley, Mass.: Newbury House. Fix, Ulla 1993a "Noch breiter entfalten und noch wirksamer untermauern. Die Beschreibung von Wörtern aus dem offiziellen Sprachverkehr der DDR nach den Bedingungen des Gebrauchs", in Grosse, R.; Lerchner, G. and Schröder, M. (eds.): Beiträge zur Phraseologie, Wortbildung, Lexikologie. Frankfurt/M. et al. Peter Lang. Fix, Ulla 1993b "Rituelle Kommunikation im öffentlichen Sprachgebrauch der DDR und ihre Begleitumstände", in Lerchner, G. (ed.): Sprachgebrauch im Wandel. Frankfurt/M. et al. Peter Lang. Fleischer, Wolfgang 1983 "Die deutsche Sprache in der DDR. Grundsätzliche Überlegungen zur Sprachsituation'', Linguistische Studien (Leipzig) 111, 258-275. Fleischer, Wolfgang 1987 Wortschatz der Deutschen Sprache in der DDR. Leipzig: Bibliographisches Institut. Fraas, Claudia 1993 "Verständnisschwierigkeiten der Deutschen", Muttersprache 3, 260-263. Fraas, Claudia and Steyer, Kathrin 1992 "Sprache der Wende - Wende der Sprache? Beharrungsvermögen und Dynamik von Strukturen im öffentlichen Sprachgebrauch", Deutsche Sprache 20, 172-84. Frein-Plischke, Marie-Luise 1987

Wortschatz Bundesrepublik

- DDR.

Semantische Untersuchungen anhand von

Personalkollektiva. Düsseldorf: Schwann. Giles, Howard and Smith, Philip 1979 "Accomodation theory: optimal levels of convergence", in Giles, H. and St. Clair, R. (eds): Language and Social Psychology. Oxford: Blackwell. Good, Colin 1993 "Über die "Kultur des Mißverständnisses" im vereinten Deutschland", Muttersprache 103. 3,249-259. Gumperz, John 1971 Language in Social Groups. Stanford: Stanford University Press. Härtung, Wolf-Dietrich 1990 "Einheitlichkeit und Differenziertheit der deutschen Sprache", Zeitschrift für Germanistik 4, 447-465. Hellmann, Manfred W. 1973 Zum öffentlichen Sprachgebrauch in der BRD und der DDR. Düsseldorf: Schwann. Hellmann, Manfred W. 1976 Bibliografte zum Öffentlichen Sprachgebrauch in der BRD und DDR. Düsseldorf: Schwann.

H. Kreutz, Some Observations on Hedging Phenomena

229

Hellmann, Manfred W. 1984 Ost-West Wortschatzvergleiche. Tübingen: Narr. Hellmann, Manfred W. 1989 "Berichte zur Alltagskommunikation in der DDR - ein Thema für Linguisten", Muttersprache 89, 79-82. Hellmann, Manfred W. 1990 "DDR-Sprachgebrauch nach der Wende - eine erste Bestandsaufnahme", Muttersprache 100, 266-286. Hellmann, Manfred W. 1991 "Ich suche eine Wohnung. Zur vergleichenden Untersuchung alltagssprachlichen Handelns in beiden deutschen Staaten", in Schlosser, H.D. (ed.): Kommunikationsbedingungen und Alltagssprache in der ehemaligen DDR. Ergebnisse einer interdisziplinären Tagung. Hamburg: Buske. Hellmann, Manfred W. 1992 "Babylon oder: Die Leipziger Volkszeitung in der Wende", Sprachreport 2-3, 12-13. Hellmann, Manfred W. 1993 "Die Leipziger Volkszeitung vom 27.10.1989 - eine Zeitung im Umbruch", Muttersprache 3, 186-218. Hellmann, Manfred W. 1987 "Die doppelte Wende - Zur Verbindung von Sprache, Sprachwissenschaft und zeitgebundener politischer Bewertung am Beispiel Deutsch - Deutscher Sprachdifferenzierung", in Klein, J. (ed.): Politische Semantik. Düsseldorf: Wissenschaftlicher Verlag. Hoffmann, Rolf R. 1982 Politische Fernsehinterviews. Eine empirische Analyse sprachlichen Handelns. Tübingen: Niemeyer. Holly, Werner 1990 Politikersprache. Berlin: Walter de Gruyter. Holly, Werner; Kühn, Peter and Püschel, Ulrich 1986 Politische Fernsehdiskussionen. Tübingen: Niemeyer. Holly, Werner; Kühn, Peter and Püschel Ulrich 1984 "Für einen 'sinnvollen' Handlungsbegriff in der linguistischen Pragmatik", Zeitschrift ßrgermanistische Linguistik 12, 275-312. Hohnes, Janet 1984 "Hedging your bets and sitting on the fence", Te Reo 27, 47-62. Hundsnurscher, Franz 1993 "Politikersprache" (book review), Journal of Pragmatics, 19-4, 376-390. Hymes, Dell 1974 Foundations in Sociolinguistics. Philadelphia, Pa.: University of Philadelphia Press. Kalhneyer, Werner and Schmitt, Reinhold 1992 "Zur Formulierungsdynamik von Äußerungen mit Fokusopposition", (Unpublished Working Paper) Mannheim: Institut für deutsche Sprache. Klemperer, Viktor 1952 "Unsere Sprache - einigendes Band der Nation", Die neue Schule 8, 4-5. Kreutz, Heinz J. 1993 "Pragmatic and linguistic implications of the convergence between east and west Germany", in Hajek, J. (ed.) Working Papers in Linguistics 13, 63-78. Melbourne: University of Melbourne. Lakoff, Robin 1975 Language and Woman's Place. New York: Harper Colophon. Lerchner, Gotthard 1992 "Broiler, Plast(e) und Datsche machen noch nicht den Unterschied. Fremdheit und Toleranz in einer plurizentrischen deutschen Kommunikaüonskultur", in Lerchner, G. (ed.): Sprachgebrauch im Wandel Anmerkungen zur Kommunikationskultur in der DDR vor und nach der Wende. Frankfurt/M. et al. Lang.

230

IV. Cross-cultural Aspects of Hedging

Levelt, Willem 1989 Speaking. Prom Intention to Articulation. Cambridge, Mass.: MIT Press. Linke, Angelika 1985 Fernsehgespräche im Fernsehen. Bern et al.: Peter Lang. Moser, Hugo 1961 "Die Sprache im geteilten Deutschland", Wirkendes Wort 11/1,1-21. Moser, Hugo 1964 Das Aueler Protokoll - Deutsche Sprache im Spannungsfeld zwischen Ost und West. Düsseldorf: Schwann. Mühlen, Ulrike 1985 Talk ah Show. Frankfurt/M. et al.: Peter Lang. Oschlies, Wolfgang 1989 Würgende und wirkende Wörter - Deutschsprechen in der DDR. Berlin: Holzapfel. Pätzoldjörg 1992 "Zwischen Indirektheit und Sprachlosigkeit", in Welke, K.; Sauer, W. and Glück, H. (eds.): Die deutsche Sprache nach der Wende. Hildesheim: Olms. Panther, Klaus-Uwe. 1981 "Einige typische indirekte sprachliche Handlungen im wissenschaftlichen Diskurs", in Bungarten, T.: Wissenschafissprache. München: Wilhelm Fink, von Polenz, Peter 1988 "Binnendeutsch oder plurizentrische Sprachkultur?", Ztitschrifi für germanistische Linguistik 16, 198-218. von Polenz, Peter 1993 "Die Sprachrevolte in der DDR im Herbst 1990", Zeitschrifi für germanistische Linguistik2, 127-149. Quasthoff, Ulla 1979 Erzähkn in Gesprächen. Tübingen: Niemeyer. Rosendiz, Julia-Liebe 1992 "Woran erkennen sich Ost- und Westdeutsche? Eine Spracheinstellungsstudie am Beispiel von Rundfunksendungen", in Welke, K.; Sauer, W. and Glück, H. (eds.): Die deutsche Sprache nach der Wende. Hildesheim: Olms. Sandig, Barabara 1979 "Beschreibung des Gebrauchs von Abtönungspartikeln im Dialog", in Weydt, H. (ed.): Die Partikeln der deutschen Sprache. Berlin: Gruyter de Walter Scherer, Klaus and Giles, Howard 1979 Social Markers in Speech. Cambridge: Cambridge University Press. Schlosser, Horst-Dieter 1986 "Zur Rezeption von Normen der öffentlichen Sprache in der "Alltagssprache" von DDR-Filmen", In Debus, F.; Hellmann, M. and Schlosser, H.-D. (eds.): Sprachliche Normen und Normierungsfolgen in der DDR. Hildesheim: Olms. Schlosser, Horst-Dieter 1990a Die deutsche Sprache in der DDR. Zwischen Stalinismus und Demokratie. Köln: Verlag Wissenschaft und Politik. Schlosser, Horst-Dieter 1990b "Das Ende der Zweisprachigkeit", in Schlosser, H.-D. (ed.): Wiederbegegnung, Herausforderung an die politische Bildung. Frankfurt a. M.: Pädagogische Arbeitsstelle Deutscher Volkshochschulverband. Schlosser, Horst-Dieter 1992 "Mentale und sprachliche Interferenzen beim Übergang der DDR von der Zentralplanwirtschaft zur Marktwirtschaft", in Welke, K.; Sauer, W. and Glück, H. (eds.): Die deutsche Sprache nach der Wende. Hildesheim: Olms. Schlosser, Horst-Dieter 1993 "Die ins Leere befreite Sprache. Wende-Texte zwischen Euphorie und bundesdeutscher Wirklichkeit", Muttersprache 103, 219-230. Schmidt, Gerhard D. 1983 "Die deutschen Varianten des Deutschen", Muttersprache 93, 284-89.

H. Kreutz, Some Observations on Hedging Phenomena

231

Schmidt, Veronika 1978 "Klassenbedingte Differenzierung des Wortschatzes", Zeitschrifi ßr Phonetik, SprachwissenschaftundKommunikationsforschung'Sl, 3-14. Schrubben, Cordt 1993 "Der neue kalte Krieg", Spiegel 20, 154-72. Simpson, Paul 1993 Language, Ideology and Point of View. London: Routledge. Teichmann, Christine 1991 "Von der 'langue de bois' zur 'Sprache der Wende'", Muttersprache 101, 252-67. Tliierse, Wolfgang 1992 "Sprich, damit ich dich sehe. Ein anderes Deutsch, doch keine andere Sprache in der Vergangenheit der DDR", Frankfurter Allgemeine Zeitung 1. April 92, 36. Weiskopf, Carl 1955 "'Ostdeutsch' und 'Westdeutsch' oder über die Gefahr der Sprachentfremdung", Neue Deutsche Literatur 3/7, 79-88. Weydt, Harald (ed.) 1979 Die Partikeln der deutschen Sprache. Berlin: Walter de Gruyter. Weydt, Harald (ed.) 1983 Partikeln und Interaktion. Tübingen: Niemeyer. Ylönen, Sabine 1992 "Probleme deutsch-deutscher Kommunikation. Unterschiede im kommunikativen Verhalten zwischen Alt- und Neubürgern", Sprachreport 2/3, 17-20.

V. Hedging: The Concept, its Origins and a Bibliographical Guide

GUDRUN CLEMEN

The Concept of Hedging: Origins, Approaches and Definitions

1. Introduction Linguists studying the concept of hedging realize that it employs a host of devices (called hedges) which are difficult to define even after mature consideration. Reference works are not very helpful with regard to the word Hedge and its implications. It is mentioned superficially or not at all. This is not surprising considering that this linguistic phenomenon has been regarded patchily so far as a subcategory of a larger class which, in practice, tends to be treated simplistically, and for which (as in the case of discourse analysis and other linguistic domains) there is no agreed-upon definition. The various approaches to the analysis of hedging have not always made it clear that hedging is achieved primarily by setting utterance in context rather than by straightforward statement, or that discourse analysis must precede semantic grasp. In this paper, I shall try to show the origin and further development of the notion of hedging by focusing on problems concerning definitions and classifications, as well as discussing questions which have not yet been dealt with satisfactorily. A detailed analysis of hedges in English economics texts will shortly be published by the present writer. 1.1. The origins Although U. Weinreich1 dealt with the phenomenon of hedging as early as 1966, labelling it "metalinguistic operators", George Lakoff is recognized as the linguist who introduced the term "hedgeV'hedging" to describe lexical units as expressions "whose job it is to make things fuzzier or less fuzzy" (1972, 195).

Weinreich, who aims at universals of language in the framework of consistent and comprehensive theory, argues "that for every language 'metalinguistic operators' such as English true, real, so-catted, strictly speaking, German eigentlich, and the most powerful extrapolator of all - like - function as instructions for the loose or strict interpretation of designata" (1966, 163).

236

V. Hedging: The Concept, its Origins and a Bibliographical Guide

1.2. Definitions and kxico-semantic aspects Concise information is, of course, to be expected in all reference works. Those seeking precise information concerning the linguistic hedge will be disappointed. A more detailed definition can be found in A Dictionary ofStylistics (Wales 1989, 215/216), which refers to "Hedge'VHedging" both in the light of discourse analysis and in the narrower sense (based on Lakoff) as semantic modifiers. In the most recent English monolingual reference works, the following entries appear: - to hedge (v.i.) = to avoid making a decisive statement (The CasseU Encyclopaedia Dictionary 1990, 679) - to hedge... = die use of evasive or deliberately vague language (Roget's II. The New Thesaurus, Expanded Edition 1988, 488) - to hedge = If you hedge you avoid answering a question or committing yourself to something (BBC English Dictionary 1992, 539)

In the realm of recent lexical research, mainly represented by Wörterbücher Dictionaries Dictionnaires (1989, 855-862), the authors focus on the problem of classifying hedging expressions and their treatment in reference works. This entry is based on Lakoff's "list of hedges" and deals with some essential aspects without covering the subject overall. Taking into account that hedging typically depends on context and situation and is not determined by individual lexical units or phrases, it does not seem possible to establish "lists of hedges". This may be the reason why an adequate and complete description of lexical units serving as hedging devices in dictionaries and reference works is lacking. The International Handbook of Contemporary Research: Semantik/Semantics (1991, 261) contains an article that treats vagueness and related phenomena. Drawing on Lakoff's fuzzy-logical treatment of hedges, the author refers to the limits of Lakoffs concept and to such hedges as "eigentlich", "echt" and "typisch", which do not exemplify semantic vagueness.

2. Approaches 2.1. The Lakoff model as a starting point LakofFs concept of hedging arises from Zadeh's "fuzzy-set theory", and the experimental results obtained by Rosch-Heider. In his fuzzy-set theory,2 Zadeh (1965) noted that such classes as "animal", which are initially conceived as having fixed membership criteria, had a "continuum of classification grades". In a fuzzy set, the instance does not merely fit or not, but may fit to a certain extent. Zadeh's observations transformed analysis of category membership 2

In German Theorie unscharfer Mengen" according to W. Wblski (1980); "Gefranste Mengen" according to R. Klockow (1980).

G. Clemen, The Concept of Hedging

237

from being hard and fast ("a member belongs to a category or not") to the modelling of verdicts on degrees of membership ("every member belongs to a category only in a certain degree"). Rosch's investigations (1973, 1978) were based on Zadeh's theory. She developed what has since come to be called "theory of prototypes and basic-level categories", or "prototype theory". She challenged the classical theory which claims that the properties defining a category are shared by all members who have equal status in that classification. Rosch demonstrated the existence of prototype effects. She found that some members are better examples of a category than others and called those members the "prototypes" of the given category. For example, subjects will consistently rate some birds better examples of the category birds than others. Robins, for instance, are judged to be more representative of the category bird than penguins.3 The most representative members of a category are often called "prototypes" or "prototypical members". Rosch's conclusion was that a robin is not just "sort of "a. bird, but rather a "bird par excellence", whereas one might describe a penguin or a hen as a "sort of" bird.4 Prototype theory is seen as a holistic approach to describe word meaning - in contrast to the analytical method. Lakoff defines hedges as follows: "For me, some of the most interesting questions are raised by the study of words whose meaning implicitly involves fuzziness - words whose job is to make things fuzzier or less fuzzy. I will refer to such words as 'hedges'" (1973a, 471). According to Lakoff, "hedges" such as "sort of" typically modify predicates (in the sense of Formal Logic) with regard to their being assigned to a category. His primary interest is not the qualitative aspect according to truth but grading. The speaker's reality of judgement is based on critical attributes of category membership. It is via "vagueness" and "fuzziness" that Lakoff arrived at the concept of hedges. With his often cited example (a) (b)

Esther Williams is a fish Esther Williams is a regular fish

Lakoff demonstrated that hedges do not merely reveal distinctions of degree of category membership, but can also reveal a great deal more about meaning. Literally, a) is false, since Esther Williams is a human being and not a fish; b) with the "hedge" regular would seem to be true. It means that Esther Williams swims well and is at home in water. Thus, it refers to her skills and does not assert that she has the other characteristics of a fish (1973a, 473). In Lakoff (1973a, 472), a summary of expressions and phrases which he regards as hedges is provided. His selection is open to question (cf. House/Kasper 1981). The value of a "list of hedges" appears to be highly quesOf course, this applies to areas where a robin is indeed representative of the genre, bird: there are other areas where another type of bird is regarded as representing the genre (e.g., in New Zealand, the kiwi). It later transpired that Rosch's experimental results were widely interpreted wrongly. Rosch extended and further elaborated her approach (1978), making it clear that prototypes, though often thought to show something direct about the nature of human classification, do not compose an overall theory of representation for categories: "to speak of a prototype at all is simply a convenient grammatical fiction; what is really referred to are judgements of degree of prototypicality" (quoted in Lakoff 1987, 136).

238

V. Hedging: The Concept, its Origins and a Bibliographical Guide

tionable because of the universal attributes of many lexical items and expressions. In neglecting the various functions which lexical items can perform, Lakoff's model does not account for lexical use. Primarily focusing on the cognitive aspect of his studies, Lakoff revised his initial concept from 1972/1973 in the 1980s (cf. Lakoff 1982). Based on the assumption that we organize our knowledge by means of archtetypes (i.e., idealized cognitive models = ICMs)5 and that category structures and prototype effects are by-products of that organization (cf. Lakoff 1987, 68), Lakoff, following Fillmore's frame semantics and studies related to the "theory of natural categorization", now puts more emphasis on cognitive theories suggesting that the philosophical commitment to traditional realism of the sort of model-theory semantics proposed by Barwise/Perry/Montague; i.e., truth-conditional semantics should be discarded and replaced by cognitive semantics, and experimental realism should be reconsidered (cf. 1986, 157). Lakoff adheres to Kay's experimental results within the ICM theory, which claims that categories used in ICMs are not objective, but based on experimental aspects of human psychology. Kay presents a re-analysis of the data on linguistic hedges as they were first studied by Lakoff in 1972/1973. His observations have "important implications for any theory of meaning, especially the theory of ICMs" (Lakoff 1982, 63) 6 With his new approach, Lakoff attempted to account for the new view in the classical theory of concepts and categories and makes us re-examine our ideas about rational thought. Apart from including the Lakoff-type hedges in his considerations in the framework of ICMs, there seems to be no further extension of the hedging spectrum. Lakoff remains rooted in semantic theory. Verbal interaction in context is hardly considered. The Lakoff repertoire of hedges within the scope of semantics is to be regarded as modifiers in a narrower sense. Even in his most recent volume, Women, Fire, and Dangerous Things (1987), his "classical" repertoire ranges from "loosely speaking" to "regular" (1987, 122-123). Lakoff apparently realizes the inadequacy of his past approach since his hedging devices are now seen in the light of experimental results obtained by Kay (Lakoff 1987, 122). 5

6

The following example of what ICMs are, and how they work, is based on Fillmore's frame concept: Our model of a week, for instance, is idealized. Seven-day weeks do not exist objectively in nature, but are created by human beings. Tuesday, like any other day of die week, can be defined only relative to an idealized model that includes the natural cycle defined by the movement of the sun, the standard means of characterizing the end of one day and the beginning of the next, and a larger seven-day calendric cycle - the week. In an idealized model, the week is a whole, with seven parts organized in a linear sequence; each part is called a day, and the third is Tuesday (cf. Lakoff 1987, 68, 69). Kay demonstrates in the case of loosefy speaking, strictly speaking, regular and technically, which were first analysed by Lakoff (1972/73) (inadequately, as Lakoff says), that ordinary speakers of English have two conflicting dieories of how to use words to refer to tilings in the world. Kay identifies "loosely speaking" and "stricdy speaking" as "pragmatic hedges" which take for granted the ICM; i.e., they assume mat words can fit the world by virtue of their inherent meanings, either strictly or loosely. If words can fit the world, they can fit it either strictly or loosely, and the hedges "stricdy speaking" and "loosely speaking" indicate how narrow or broadly one should construe the fit (Lakoff 1987, 122f).

G. Clemen, The Concept of Hedging

239

Despite new impulses in hedging research over the past two decades (also influenced by discourse analysis), Lakoff remains rooted in the realm of cognitive science without extending his pioneering model of hedges to account for the considerably broadened spectrum of hedging research. His contribution in having initiated the notion of hedging, and thus inducing many linguists to develop his model further, nevertheless remains uncontested.

2.2. More recent approaches LakofFs model has been adopted and used as a point of departure in subsequent investigations into hedging phenomena. Divergent views can be found in the literature as to which lexical and/or syntactic hedging devices should be assigned to individual pragmatic strategies (politeness, indirectness, mitigation, vagueness, understatement)7 which focus mainly on tension-free social interaction. In any case, varying degrees of understanding the concept of hedging become evident from different domains of knowledge. Consequently, finer distinctions in examining the hedging phenomenon have emerged based on the pragmatic aspect of communication; an aspect that was hardly considered in LakofFs model. Various approaches pointed to a greater variety of motives in applying hedging devices. For instance, mitigation and the modification of utterances to save the face of participants. Face-saving strategies intended to obtain hearer acceptance, avoidance of commitment, and intentional vagueness are all given special documentation. The most interesting research activities emerged from works in the field of politeness strategies, and, closely related to it, indirectness and mitigation. It was not until the late 1970s that the concept of politeness became a major concern in pragmatics. The model proposed by Brown/Levinson (1978/1987) has become a constant source of reference and has furnished new ideas for investigating the role of hedging in the framework of politeness strategies.8 Speakers/writers use lexical and/or syntactic devices such as modal auxiliaries (e.g., can/could; may/might; shall/should; will/would; must/ought to; etc.) hedged performatives (such as: have to admit, wish to invite, can promise, etc.), impersonal structures (one suggests), conditionals (if this were.../one would) to mitigate the strength of claims, statements and utterances while tending to face-save to achieve broader acceptance from the recipient as well as to evade possible criticism. These linguistic devices which modify utterances that may otherwise seem too forceful (and which, in most cases, indicate politeness) may have a hedging function, depending on the context. In this sense, hedging should be considered to be a face-saving strategy. "Hedged Performatives" as a mode of politeness are an important aspect of hedging; thus drawing attention to hedging devices out of the "store of Cf. e.g., Fräser 1975, 1980; Brown/Levinson 1978, 1987; House/Kasper 1981; Panther 1981; Prince et al. 1982; Rounds 1982; Leech 1983; Werlen 1983; Hübler 1983; James 1983; BlumKulka/Olshtain 1984; Pindi/Bloor 1987; Markkanen/Schröder 1987; Schröder 1988; Myers 1989; Bloor/Pindi 1990; Flowerdew 1991. Cf. e.g., Fräser 1975; Panther 1981; House/Kasper 1981; Leech 1983; Werlen 1983; Blum-Kulka/Olshtain 1984; Myers 1989; Flowerdew 1991; Maier 1992.

240

V. Hedging: The Concept, its Origins and a Bibliographical Guide

modals".9 Unlike Lakoff's view, mitigation per se is not equated with hedging (c£ Fräser 1980). In the 1980s, more attention was paid to the level of empirical analyses, and the focus of research was often based on contrastive and cross cultural approaches. This resulted in further refining the "content" of hedges. Another indication of refinement can be observed in approaches that attempt to subdivide hedges. Hedges, within the group of "politeness markers", for instance, are classified as a subclass of "downgrade«", thus implying a narrower definition in contrast to Lakoff (cf. House/Kasper 1981). Cross-cultural analyses reveal that there are different frequencies in the use of hedges between English and German speakers (cf. Clyne 1991). There are theses according to which "downtoners", "understates", "minimizers" and "negations" have to be distinguished from each other and cannot be defined as hedges (Blum-Kulka/Olshtain 1984). These investigations lead to various attempts to redefine the concept of hedges, although these have not generally been accepted. It thus seems that these attempts suggest an obvious general discomfort among linguists as to what the term "hedge" designates. Presently, it appears to cover almost "everything" and requires a more systematic approach. One of the approaches put forward is based on a report on the occurrence of these phenomena in professional discussions among physicians (Prince et al. 1982). This study suggests a bipartite division of hedges into "Shields" and "Approximators"10 In addition to these categories, a third group, i.e. "Diffusers", has also been proposed (Rounds 1982).11 The literature on vagueness^- does not always include the hedging phenomenon despite the fact that the lexical and/or syntactic features of vague utterances are closely related to hedging. Vagueness is one of the many hedging strategies. As a rule, the speaker/writer will make vague statements if exact data is missing or if precise information is irrelevant in preliminary results, or with no need 9

Cf. e.g., Fräser 1975; Panther 1981; Werlen 1983; Leech 1983. "Approximators" and "shields", according to Prince et al., have little in common. "Approximators" contribute to semantics by indicating some markedness; i.e., atypicality with respect to class membership; while "shields" affect pragmatic aspects by including implicatures conveying markedness with respect to speaker-commitment, a) Hisfeet were blue conveys the proposition that his feet were blue; it implicates that the speaker knows that his feet were blue, b) His feet were SORT OF blue (or His feet were NON-PROTOTYPICALLY blue) conveys a different proposition: Speaker of b) would be as fully committed to the truth of the proposition that is conveyed as would the speaker of a), sort of is the type of hedge that affects the prepositional content but not speaker-commitment. The authors call such hedges "approximators"; c) I THINK his feet were blue. In sharp contrast, c) con veys the same prepositional content. It implicates that die speaker is less dian fully committed, or commited in some less marked way, to the truth of the proposition. These hedges are called "shields" (cf. 1982, 85 f). 11 Rounds, who argues that "diffusers" (1. jargon, 2. avoidance, 3. footnote) be included, refers to a type of hedge which does not appear in Prince et al.'s class-system. These hedges, according to Rounds, consist of rhetorical strategies which she refers to as "diffusers". 12 E.g., Wolski 1980; Pinkal 1980/81, 1981; Rieger 1989; Busse 1991; Jahr 1992. 10

G. Clemen, The Concept of Hedging

241

for categorical assertions, or when the speaker/writer is not absolutely certain. Hedging protects him from making possible false statements, provides him with a graceful way out and increases the credibility of his utterance ("It's about a hundred..."). Predications that cannot be evaluated exactly demand vagueness of expression, and hedging is typically found in such communicative situations. The function of hedging with respect to vagueness was more closely studied in the 1980s. According to Pinkal (1981), "so-called linguistic hedges" (translated by G.G.) serve to specify "vague expressions". Pinkal distinguishes "präzisierende", "depräzisierende" and "quantifizierende Hecken" (= "precise", "imprecise" and "quantifying" hedges) (translation G.G.). The role of hedging as an indicator of vagueness and imprecision is discussed in the framework of LSP (language for specific purposes) texts. These studies consider, for example, the frequency and type of lexical devices (cf. Dubois 1987) used to achieve the hedging effect, as well as the motives for the use of hedging (cf. Channell 1990). Understatement, achieved by certain lexical and/or syntactical devices in appropriate contexts, is another strategy that may have a hedging function. Hubler's distinction between "understatements" and "hedges"13 has found little support among researchers because the former is often considered to be a subcategory of the latter. There have been many contributions to the research on modal particles.^ These lexical items are rarely considered to be examples of hedges though their hedging capacity as illocutionary modifiers has been recognized. They are means to convey emotions and attitudes of the speaker/writer. According to Rudolph, the three most important functions of particles, which have been overlooked so far, are downtoning, modifying and grading. Hence, she distinguishes downtoners, modal and grade particles (cf. Rudolph 1986, 76). The demarcation of modal particles has proved difficult because the area they cover is vast. Also, they have different functions, and vary according to context. Most researchers cannot agree on the elements which constitute a parHubler's hedges constitute a very small category of those items which are conventionally regarded as hedges. He defines "understatements" and "hedges" as "two types of understate ment", distinguishable according to the different scope of linguistic devices used to achieve a desired rhetorical effect. The prepositional content itself is called "phrastic", referring to understatements: a) It's snowing in the mountains. "Neustic", on the other hand, b) It's snowing in the mountains, I suppose "represents that part of the illocution which expresses the attitude of the speaker to the hearer regarding the proposition", and "tropic" is meant to characterize "the attitude of the speaker to the prepositional content".Thus the two sentences "have the assertorical mood (tropic) in common. As far as the neustic is concerned, however, they differ in that (...) a) "represents straightforward assertion", in b) "the assertion is weakened and becomes a supposition". (Hübler 1983, 11 f) The term "Modal Particles" has been used to refer to the grammatical class of words labled "Modalpartikeln" in German although it "is virtually a non-topic in the English-language linguistic tradition, perhaps because English has no grammatical category akin to the German modal particle" (Abraham 1991, Vol. II, 331). Contributions in research into "modal particles" cf. e.g., Altmann 1978; Weydt 1979, 1981, 1983; Helbig/Kötz 1981; König et al. 1990.

242

V. Hedging: The Concept, its Origins and a Bibliographical Guide

tide, which may be the reason why hedging has not received much attention in the area of modal particles. The classifications suggested by Quirk et al. (1972, 452f.) illustrate the problem of designation: "compromisers" along with "diminishes" and "minimizers" as well as "approximate«" are classified under "downtoners"; expressions like "sort of", "kind of", etc. belong to the category "compromisers". "Downtoners" (corresponding to "Abtönungspartikeln" in German) like "compromisers" have received attention within the framework of hedging analysis (cf. Weydt 1979). Both can appear in their double function as "voluntary markers of imprecision" and as "modifiers of illocutionary force" (cf. James 1983). While research on hedging has concentrated primarily on the spoken language, from the end of the 1980s onwards, attention has shifted more to the written medium15 and to specific language areas such as LSP. Preference is given to LSP (for example, Academic Writing and Writing on Economics), often on a cross-cultural and contrastive basis, aligned with didactic purposes. One of the latest contributions (cf. Holmes 1993), an analysis of politeness strategies in verbal interaction, refers to the hedging behaviour in the language of women as opposed to men in New Zealand. The findings on women's linguistic behaviour in discourse are contrary to those of Robin Lakoff (1973) from 20 years before.

3. Methodological problems

3.1, Demarcation Researchers cannot agree on which lexical items, phrases or syntactic structures should be classed as hedges, and which attributes a word or phrase should contain to function as a hedge in a given context. There are no clear criteria. German linguists have found no satisfactory term for the pragmatic strategy involved. They use such expressions as "Heckenausdrücke", "sogenannte Heckenausdrücke", "linguistische Hecken", and "sprachliche Hecken" as well as the English term "hedges". It remains to be seen whether the German "Hecke" and similarly "hedge" are the most accurate terms for the variety of functions which the term is assumed to encompass. A systematic analysis of the various "types" of hedge (cf. Pinkal 1985, 48) which offered some classification or subdivision would serve to distinguish more exactly the phenomenon of hedging. Eliminating the term "hedge" and substituting some other designation for it would hardly gain acceptance. In this context, Skelton's proposal (this volume) is interesting. Skelton, who has for some time found the word "hedging" „an unfortunate choice", argues that the term "hedge* should be used "very narrowly" and "only for mitigations of responsibility and/or certainty to the truth of a proposition". His hierarchical breakdown is as follows: Comment16 on Truth-judgement - Comment on Value-judgement - Hedge. 15

Cf. e.g., Markkanen/Schroder 1987, 1989; Pindi/Bloor 1987; Schröder 1988; Myers 1989; Bloor/Pindi 1990; Channell 1990; Clyne 1991; Namsaraev 1991; Maier 1992. 16 "Comment", however, may clash with the same term within the theme/rheme theory.

G. Clemen, The Concept of Hedging

243

Distinctions are not always made in literature between strategies to be applied and modifying devices used to achieve hedging. We may sometimes read, "This is a hedge", independent of whether it refers to the strategy (i.e. indirectness), or whether the modifier is referred to (e.g.) an impersonal structure such as "It appears that...")

3.2. Neglected word-classes and research areas Shortcomings can still be found in analysing both Modal Particles and Adjectives in the framework of hedging. There is a need for further research here. As far as functional words are concerned, Modal Particles are one of the newest research areas. They have received little attention in relation to hedging because of the problems of classification arising from their wide range of use. The role of the Adjective as a modifier expressing subjective evaluation by the speaker/writer has still to be clarified. There have been no adequate analyses of which types of adjective act as hedges in specific contexts. Similarly, contrastive analyses taking into account the questions of cross-cultural communication can be seen as a challenge for further research (e.g. Clyne 1991). 3.3. Pragmatic strategies and modifying devices The following is an attempt to provide a list of the most important Strategies along with their relevant Modifying Devices to achieve hedging. Levels of strategies

Politeness Indirectness Mitigation Vagueness Understatement

Modifying devices Hedged Performatives Epistemic Qualifiers Modal Verbs Modal Particles/Downtowners, etc. Adj ectives/Adverbs Certain Personal Pronouns Passive/Passive Infinitive Impersonal/Indirect Constructions Parenthetic Constructions Subjunctive/Conditional Concessive Conjunctions Negation

Some modifying devices can be attributed to more than one level. (Prosodic and non-verbal means, which may be included, have not been considered). A single item cannot function as a hedge. Rather, hedges are determined by context (type of discourse), the colloquial situation and the speaker's/writer's intention, plus the background knowledge of the interlocutors. Hedging can be deduced only from the combination of the individual clausal elements plus the relevant illocution. The pragmatic function of hedging is implicit at the level of utterance and not explicit in any lexical unit.

244

V. Hedging: The Concept, its Origins and a Bibliographical Guide

4. Tentative characteristics of hedges

Hedges are used to express the speaker's/writer's subjective comment or attitude to his/her utterance or statement: "speaker's tentativeness over the truth value of the matrix clause" (cf. Quirk et al. 1985, 1114). Hedges function in a particular context. They are not inherent in the phrase. Taking into account one (or more) of the following elements may guarantee that the speaker's/writer's statement or utterance is received by the hypothetical addressee: -

mitigation putting things into a relative aspect avoiding unnecessary risks taking into account the addressee's possible objection being vague, reserved or restrictive, as the case may be, but still observing the maxims of politeness, and so modifying the speaker's/writer's responsibility for the truth-value and/or precision of a proposition as well as expressing the speaker's/writer's attitude to the proposition made.

Despite or, perhaps, because of their mitigating effect, hedges can increase the credibility of a statement (e.g. in academic texts). It is in this sense that hedging is better seen in the scope of discourse analysis as opposed to its narrower semantic aspect as a prototypical modifier.

5. Concluding Remarks

The purpose of this paper was to survey the origin of the concept of hedges and its development during the last two decades. This topic has received increasing interest, especially in the field of discourse analysis. The linguistic hedge is not a clearly defined research object, and studies have often treated it as a by-product of a larger entity. Nevertheless, a core sample has been amassed from various observations and findings which show that the concept of the hedge has been greatly extended. Definition, classification and assignment pose problems which still require clarification. Further exploration of the use of hedges and kindred elements (particularly from a cross-cultural and contrastive perspective) would offer new valuable insights into a linguistic device which seems to be growing more important in communication. Focusing on an attempt to disentangle the disagreement surrounding the various theoretical approaches to hedges, Skelton's most recent concept is worth considering. His proposal may show how certain adjectives/adverbs may be accommodated by assigning them judiciously to the speaker's/writer's Comments on Value-judgement/Truth-judgement, as the case may be, and thereby solving some theoretical problems. This will be taken into account by the present writer in a forthcoming study intended to show how hedging and related phenomena occur in English economics texts written by journalists.

G. Clemen, The Concept of Hedging

245

References Abraham, Werner 1991 "The Grammaticization of the German Modal Particles", in Traugott, E.G. and Heine, B. (eds.): Approaches to Grammaticalization (Vol. II). Amsterdam/ Philadelphia: John Benjamins, 331-380. Altmann, Hans 1978 Gradpartikelprobleme. Tübingen: Narr. Bloor, Thomas and Pindi, Makaya 1990 "Schematic Structure in Economics Forecasts" in Dudley-Evans, Tony, and Henderson, W. (eds.): The Language of Ecomonics: The Analysis of Economics Discourse. Modern English Publications and the British Council (ELT Documents 134), 55-66. Blum-Kulka, Shoshana and Olshtain, Elite 1984 "Requests and Apologies: A Cross-Cultural Study of Speech Act Realization Patterns". (CCSARP). AppliedLinguisics, Vol. 5, no. 3. 196-213. Brown, Penelope and Levinson, Stephen C. 1978 "Universals in Language Usage: Politeness Phenomena" in Goody, E.N. (ed.): Questions and Politeness. Cambridge et al.: Cambridge University Press, 56-310. 1987 Politeness. Some universals in language usage. Cambridge: Cambridge University Press. Busse, Dietrich 1991 Diachrone Semantik und Pragmatik. Untersuchungen zur Erklärung und Beschreibung des Sprachwandels. Tübingen: Niemeyer. Channell, Joanna 1990 "Precise and Vague quantities in Writing on Ecomonics" in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse. London: Sage, 95-117. Clyne, Michael 1991 "The Sociocultural Dimension: The Dilemma of the German-speaking Scholar", in Schröder, H. (ed.): Subject-oriented Texts. Berlin, New York: de Gruyter, 49-67. 1993 "Pragmatik, Textstruktur und kulturelle Werte. Eine interkulturelle Perspektive" in Schröder, H. (ed.): Fachtextpragmatik, Tübingen: Narr, 3-18. Dubois, Betty Lou 1987 "'Something on the order of forty to forty-four' - Imprecise Numerical Expressions in biomedical Slide Talks", Language in Society 16:4, 527-541. Fillmore, Charles 1982 "Frame Semantics", in Linguistic Society of Korea (ed.): Linguistics in the Morning Calm, Seoul: Hanshin, 111-138. Flowerdew, John 1991 "Pragmatic modifications on the 'representative' speech act of defining", Journal of Pragmatics 15, 253-264. Fräser, Bruce 1975 "Hedged Performatives" in Cole, P. and Morgan, J.L. (eds.): Syntax and Semantics, Vol. 3, 1975, 187-210. 1980 "Conversational Mitigation", Journal of Pragmatics 4, 341-350. Helbig, Gerhard and Kotz, Werner 1981 Die Partikeln. Leipzig: Enzyklopädie. Holmes, Janet 1993 "New Zealand women are good to talk to: An analysis of politeness strategies in interaction", Journal of Pragmatics 20, 91-116. House, Juliane and Kasper, Gabriele 1981 "Politeness Markers in English and German" in Coulmas, F. (ed.): Conversational Routine. The Hague: Mouton, 157-185. Hübler, Axel 1983 Understatement and Hedges in English (Pragmatics and Beyond IV:6), Amsterdam: John Benjamins.

246

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Jahr, Silke 1992

"Zur Vagheitsproblematik und Prototypentheorie", Deutsch als Fremdsprache 29:3, 167-172. James, Allen R. 1983 "Compromisers in English: A cross-disciplinary approach to their interpersonal significance", Journal of Pragmatics 7:2,191-206.

Kay, Paul 1979

The Role of Cognitive Schemata in Word Meaning: Hedges Revisited. Department of Linguistics, Berkeley: University of California. 1984 "The Kind of/Sort of Construction", Proceedings of the Tenth Annual Meeting of the Berkeley Linguistics Society, 157-171. Klockow, Reinhard 1980 Linguistik der Gänsefüßchen. Untersuchungen zum Gebrauch derAnführungszeichen im gegenwärtigen Deutsch. Frankfurt/Main: Haag & Herchen. König, Ekkehard; Stark, Detlef; Requardt, Susanne 1990 Adverbien und Partikeln. Heidelberg: Groos Lakoff, George 1972a „Linguistics and Natural Logic" in Donald Davidson and Gilbert Harman (eds.): Semantics of Natural Language. 2nd edition. Dordrecht/Holland, Boston/USA: D. Reidel, 545-665. 1972b "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts", Papers from the eighth Regional Meeting Chicago Linguistic Society, April 14-16, 1972, 183-228. 1973a "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts", Journal of Philosophical Logic 2, 458-508. 1973b "Lexicography and Generative Grammar. Hedges and Meaning Criteria" in Zgusta, L. (ed.): Probkme des Wörterbuchs. Darmstadt: Wissenschaftliche Buchgesellschaft, 150-161. 1982 Categories and Cognitive Models (L.A.U.T., Series A. Paper No. 96), Trier: LA.U.T. 1986 "Frame Semantic Control of the Coordinate Structure Constraint", Papers from the Parasession on Pragmatics and Grammatical Theory at the Twenty-Second Regional Meeting, CLS 22, Part 2, Chicago: Linguistic Society, 152-167. 1987 Women, Fire, and Dangerous Things. Chicago and London: The University of Chicago Press. Lakoff, Robin 1973 "The Logic of Politeness, or Minding your p's and q's", Papers from the ninth Regional Meeting of the Chicago Linguistic Society, Chicago: Linguistic Society, 292-305. Leech, Geoffrey N. 1971 Meaning and the English Verb. London: Longman. 1983 Principles of Pragmatics, London and New York: Longman. Maier, Paula 1992 "Politeness Strategies in Business Letters by Native and Non-Native English Speakers", Englishfor Specific Purposes, Vol. 2, No. 3,. 189-205 Markkanen, Raija and Schröder, Hartmut 1987 "Hedging and its linguistic realization in German, English and Finish philosophical texts. A case study", Erikoiskiekt ja käännösteoria, Vakki-seminaari VII, 47-57. 1989 "Hedging as a Translation Problem in Scientific Texts", in Lau^n, C. and Nordman, M. (eds.): Special Language, Clevedon, Philadelphia: Multilingual matters, 171-179. Myers, Greg 1989 "The Pragmatics of Politeness in Scientific Articles", Applied Linguistics 10:1, 2-35. Namsaraev, Vasili 1991 "Linguistc realisation of Hedging in Russian Scientific Text", Erikoiskielet ja käännösteoria, Vakki-seminaari XL, 195-207.

G. Clemen, The Concept of Hedging

247

Palmer, Frank Robert 1986 Mood and Modality. Cambridge: Cambridge University Press. Panther, Klaus-Uwe 1981 "Einige typische indirekte sprachliche Handlungen im wissenschaftlichen Diskurs" in Bungarten, T. (ed.): Wissenschafissprache, München: Fink, 231-260. Pindi, Makaya and Bloor, Thomas 1987 "Playing Safe with Predictions: Hedging, Attribution and conditions in Economic Forecasting" in Bloor T. and Norrish, J. (eds.): Written Language (GILT 55), 55-69 Pinkal, Manfred 1980/81 "Semantische Vagheit: Phänome und Theorien", Linguistische Berichte 70, 1-26 and 72, 1-26. 1985 "Kontextabhängigkeit, Vagheit, Mehrdeutigkeit", in Schwarze, C. and Wunderlich, D. (eds.): Handbuch der Lexikologie, Königstein/Ts.: Athenäum 27-63. 1991 „Vagheit und Ambiguität" (Metrische Vagheitsformen), in v. Stechow, A. and Wunderlich, D. (eds.): Semanlik/Semantics, Berlin: de Gruyter, 257-269. Prince, Ellen F.; Frader, Joel; Bosk, Charles 1982 "On Hedging in Physician-Physician Discourse", in Di Pietro , R.J. (ed.): Linguistics and the Professions, Norwood, NJ, 83-97. Rieger, Burghard 1989 Unscharfe Semantik. Frankfurt/Main, Bern, New York, Paris: Lang. Rosch, Eleanor 1973 "Natural Categories", Cognitive Psychology 4. 1978 "Principles of Categorization", in Rosch, E. and Uyod, B.: Cognition and Categorization, Hillsdale, NJ: Erlbaum Ass, 27-48. Rounds, Patricia L. 1982 Hedging in Written Academic Discourse: Precision and Rexibility. Ann Arbor: The University of Michigan. Rudolph, Elisabeth 1986 "Partikeln und Text-Konnexität im Deutschen", in Heydrich, W. and Petöfi, J.S. (eds.): Aspekte der Konnexität und Kohärenz von Texten. Hamburg: Buske, 73-90. Schröder, Hartmut 1988 "Gesellschaftswissenschaftliche Fachtexte und interkulturelle Fachkommunikation: Probleme für Fremdsprachenlemer und Übersetzer", Neuphilologische Mitteilungen, Vol. 89:4, 559-566. Skelton, John 1988a "The care and maintenance of hedges", ELT-Journal, Vol. 42:1, 37-43. 1988b "Comments in Academic Articles", Applied Linguistics in Society (GILT Publications), 9-108. 1997 „How to Tell the Truth in The British Medical Journal: Patterns of Judgement in the 19th and 20th Centuries" (this volume). Weinreich, Uriel 1966 "On the semantic structure of English" in Greenberg, J.H. (ed.): Universah of Language, second edition, Cambridge/Mass.: Press, 142-217. Werlen, Iwar 1983 "Vermeidungsritual und Höflichkeit. Zu einigen Formen konventionalisierter indirekter Sprechakte im Deutschen", Deutsche Sprache 11, 193-218. Weydt, Harald 1969 Abtönungspartikel. Bad Homburg: Gehlen. 1979 Die Partikeln der deutschen Sprache. Berlin: de Gruyter. 1981 Partikeln und Deutschunterricht. Abtönungspartikeln für Lerner der deutschen Sprache, Heidelberg: Groos. 1983 Partikeln und Interaktion. Tübingen: Niemeyer. Wolski, Werner 1980 Schlechtbestimmtbeit und Vagheit. Tendenzen und Perspektiven. Tübingen: Niemeyer.

248

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Zadeh, Lotfi 1965 'Fuzzy Sets', Information and Control 8,338-353.

Reference Books BBC English Dictionary 1992 London: Harper Collins, 539. Hausmann, Franz Josef; Reichmann, Oskar; Wiegand, Herbert Ernst; Zgusta, Ladislav (eds.): 1989 Wörterbücher Dictionaries Dictionnaires. Ein internationales Handbuch zur Lexikographie. (= Handbücher zur Sprach- und Kommunikationswissenschaft 1) Berlin, New York: de Gruyter, 855-862. Quirk, Randolph; Greenbaum, Sidney; Leech, Geoffrey; Svartvik, Jan 1972 A Grammar of Contemporary English. London: Longman, 452 f. 1985 A Comprehensive Grammar of the English Language. London: Longman, 1114 Roget's II. The New Thesaurus. 1988 Expanded Edition. Boston: Houghton Mifrlin Co, 488. v. Stechow, Arnim and Wunderlich, Dieter (eds.) 1991 Semantik/Semantics - Ein internationales Handbuch der zeitgenössischen Forschung. An International Handbook of Contemporary Research. (= Handbücher zur Sprach- und Kommunikationswissenschaft 6). Berlin, New York: de Gruyter, 261. The Cassell Encyclopaedia Dictionary 1990. London: Cassell, 679. Wales, Katie (Ed.) 1989 A Dictionary of Stylistic*. New York: Longman, 215/16.

HARTMUT SCHRÖDER and DAGMAR ZIMMER

Hedging Research in Pragmatics: A Bibliographical Research Guide to Hedging

L Introductory remarks By the term 'hedging research', we mean a complex research area within the fields of pragmatics, linguistics, semantics, logics and philosophy. In each of these research areas, the term 'hedge'/'hedging' is referred to in a different way. In the present research guide, we mainly view 'hedgesY'hedging' as a concept of pragmatics. In pragmatics, the concept of 'hedgeY'hedging' is linked to politeness phenomena, mitigation, vagueness and modality. A hedge is either defined as one or more lexico-syntactical elements that are used to modify a proposition, or else, as a strategy that modifies a proposition. The term 'hedging1 is used to refer to the textual strategies of using linguistic means as hedges in a certain context for specific communicative purposes, such as politeness, vagueness, mitigation, etc. The aim of this bibliographical research guide is to facilitate access to bibliographical information about 'hedging'/'hedges'. We do not only intend to present theoretical studies, but also those that explore possible applications of 'hedging' research (translation, intercultural communication, language learning, etc.). The structure of this research guide is as follows. Firstly, we deal with the development of the concepts 'hedges' and 'hedging' (chapter 2). In the main section of this guide (chapter 3), we give bibliographical information on discourse-oriented hedging research in pragmatics (chapter 3.1.), on concepts that are related to the concept of hedging (chapter 3.2.) and, furthermore, we present research of 'hedging'/'hedges' in areas other than in pragmatics (chapter 3.3). Finally, we present studies that focus on possible applications of results from hedging research (chapter 4). Our overview ends with the bibliography itself. The limits of this bibliographical research guide are the following. Our intention is not a critical evaluation of the research literature. Moreover, we wish to limit ourselves to a mere presentation of bibliographical data and hope, thus, to present a helpful reference work. We did not only choose literature that explicitly deals with 'hedges'/'hedging', but also literature that refers to concepts linked to 'hedgesV'hedging', such as politeness, mitigation, vagueness

250

V. Hedging: The Concept, its Origins and a Bibliographical Guide

and modality. We considered monographies, dissertations, articles published in journals and joint publications and entries in dictionaries. Of course, we do not presume to give a comprehensive account of all publications in hedging research. However, we decided to limit ourselves mainly to English and German research literature. This bibliographical research guide was compiled with an electronic database that was created with the program File MakerPro. The database contains 270 sets (including the articles of the present volume). The database allows the retrieval of research literature in various ways. The ranking can be implemented according to the alphabetical order of the authors' names, to the search terms, and to the year of publication. Thus, the database permitted us retrieval of information from different angles. Despite our original intention to make this database available to a larger audience, we decided to refrain from doing so. We realized during the process of creating the database that an electronic presentation of the data would not supply the user with more information and retrieval options than are possible in a traditional print medium. One of the most important advantages of an electronic database would be the elaboration of a citation index that allows exploration of a large number of research works as a cohesive network. The creation of such an index is among our goals for future research.

2. The development of the concepts hedge and hedging There are only a few entries in linguistic dictionaries dealing with the concepts 'hedge' and 'hedging'. We found entries with 'Hecken' and 'Heckenbildung' (the German equivalents for hedge and hedging) only in two recent German linguistic dictionaries (Bußmann 1990 and Gippert 1993) and, furthermore, in one English dictionary of stylistics (Wales 1989). The German dictionary entries present a semantic definition of the concepts on the basis of the early contribution of Lakoff. However, the Dictionary of Stylistics gives a twofold definition of hedging from the point of view of semantics and pragmatics (including discourse analysis and speech act theory). The following contributions seem fundamental to the early development of the concept of hedging (in chronological order). First, we have to mention Zadeh (1965) and Weinreich (1966) as the most important predecessors of hedging research. Both, Zadeh and Weinreich, dealt with the concept without using the term 'hedge' and 'hedging'. Zadeh (1965) deals in his article "Fuzzy Sets" with aspects of the so-called fuzzy set theory. He notes that classes, such as "animals", which have been conceived as having fixed membership criteria, have a "continuum of classification grades". According to fuzzy-set theory, an item always only fits to a certain extent into a class (continuum of classification grades). Weinreich (1966) deals in his article "On the Semantic Structure of English" with the phenomenon of hedging, labelling it "metalinguistic operators". He argues that "for every language 'metalinguistic operators' such as (in) English true, real, so-catted, strictly speaking, and

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

251

(in) German eigentlich, and the most powerful extrapolator of all - like - function as instructions for the loose or strict interpretation of designata" (1966, 163). The designation 'hedge'/'hedging' itself was introduced first by G. Lakoff (1972) in his article "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts". In his synchronic, non-contrastive study of the oral and written standard English, Lakoff defines 'hedges' (from the point of view of language philosophy) as words whose function is to make meanings fuzzier (eg. sort of) or less fuzzy. Lakoff argues that the logic of hedges requires serious semantic analysis for all predicates. He (1972, 195) defines hedges as follows: "For me, some of the most interesting questions are raised by the study of words whose meaning implicitly involves fuzziness — words whose job it is to make things fuzzier or less fuzzy. I will refer to such words as 'hedges'". In his article "Fuzzy-Set - Theoretic Interpretation of Linguistic Hedges", Zadeh (1972) followed Lakoff in using the new designation 'hedge' and analyzed English hedges (such as simple ones like very, much, more or less, essentially, and slightly and more complex ones like technically and practically) from the point of view of semantics and logics. The author assumes that hedges are operators that act on the fuzzy set representing the meaning of their operands. Hedges vary in their dependency on context. In a later publication, Zadeh (1975) studied the written standard English from the point of view of psycholinguistics. Furthermore, fundamental contributions were made again by G. Lakoff (1973) with the focus on lexicography, and by Rosch (1973) from the point of view of cognitive psychology. Rosch (1978, 39) deals with semantic prototypes from the point of view of cognition and argues "that natural languages themselves possess linguistic mechanisms for coding and coping its gradients of category membership". In Rosch's opinion hedges belong to those mechanisms. In his article "Hedged Performatives", Fräser (1975) analyzed modal verbs from the point of view of pragmatics. In more recent publications, Fräser deals with the hedging phenomenon from the point of view of mitigation and politeness research (Fräser 1980) and from the point of view of discourse markers (Fräser 1990). Brown/Levins on (1978, 1987), dealing with politeness in verbal interaction from the point of view of pragmatics, viewed hedges as a device to avoid disagreement. They describe hedges as a strategy or an expression of negative politeness (see chapters 5.3.1. and 5.4. in Brown/Levinson 1978). Prince/Bosk/Frader (1982) conducted an empirical study of hedging in discourse among physicians working in an intensive care unit. The authors distinguish between two types of hedges, those that affect the truth-conditions of propositions ('approximators') and those not affecting the truth conditions but showing the speaker's commitment to the truth-value of the whole proposition ('shields'). Rounds (1982) introduced the notion of 'diffusere' meaning that "they tend to disperse or cut off a source of disagreement or argument". The first monograph about hedges was published by Hübler (1983). In his book "Understatements and Hedges in English", Hübler makes a distinction between understatements and hedges, although he also uses understatement as a covering term for both. Understatement proper means for him that "the emo-

252

V. Hedging: The Concept, its Origins and a Bibliographical Guide

tional negatability (of sentences) is restricted through the indertermination of the phrastic", i.e. concerns the prepositional content of the sentence (...), whereas hedging means that it "is restricted through the indertermination of the neuistic", i.e. concerns the speaker's attitude to the hearer regarding the proposition. Other important publications on the very concept hedge and hedging are Markkanen/Schröder (1988, 1992), Darian 1995, Salager-Meyer (1995) and, furthermore, the articles by Holmes (1982a, 1982b, 1984), who deals with hedges from the point of view of teaching and learning English as a second language. Holmes defines hedges as devices for attenuating the strength of utterance (1982a) and as a part of epistemic modality (1982b, 1984). The research in the field of 'hedges'/'hedging' has mainly been undertaken by scholars from English-speaking countries, and 79 % of the publications in our bibliography are written in English. Considering the growing importance of English as an international language of research, it is interesting to note that, nevertheless, more than 19 % of the publications we gathered were written in German. Only 2 % of the works in our bibliography were written in other languages than English and German, namely in French, Finnish, Spanish, and Italian. Not surprisingly, the vast majority of our publications treat 'hedgesV'hedging' as phenomenons of the English and the German language. However, especially since the second half of the 1980's, a growing number of researchers expanded their interest beyond these two languages. The following languages/dialects (in alphabetical order) have been subjects of research: Amele: Roberts (1990) Chinese: Chen (1993), Gu (1990) Daco-Romanian: lonescu-Ruxandoiu (1989) Danish: Abraham (1990) Dutch: Abraham (1990) English: Aijmer (1984, 1987), Ayto/Walford (1986), Blum-Kulka (1985), Brainbridge (1994), Brown/Levinson(1978, 1987), Busch-Lauer (1994), Butler (1990), Clemen (1990), Clyne (1991), Crismore/Markkanen/Steffensen (1993), Crismore/Vande Kopple (1988, 1990), Damerau (1977), Dubois (1987), Edwards (1984), Flowerdew (1991), Fräser (1975), Grabe/Kaplan (in the present volume), Grundy (1989), Heinämäki (1972), Henkel (1983), Herrmann (1975), Holmes (1986, 1988, 1990), Hosman/Wright (1987), House/Kasper (1981), Hübler (1983), James (1983),Johnson/Vinson 1990), Lachowicz (1981), Lakoff, G. (1972, 1987), Lehrer (1989), Loewenberg (1982), Luukka/Markkanen (in the present volume), Markkanen/Schroder (1988, 1992), Meyer (in the present volume), Nikula (in the present volume), O'Connor (1993), Östman (1981), Pindi/Bloor (1987), Powell (1985), Preisler (1986), Rasmussen/Moely (1986), Salager-Meyer (1990), Salager-Meyer/Salas (1991), Semiloff (1977), Underbill (1988), Vinson/Johnson (1989), Weinreich (1966), Wickboldt (1985), Wierzbicka (1986, 1991), Wilss (in the present volume), Zadeh (1972, 1975) Frisian: Abraham (1990) French: Haarmann (1992), Prince (1976), Rastier (1983), Vicher/SankofF(1989), Wiggli (1982) German: Abraham (1990, 1991), Busch-Lauer (1994), Buscha/Heinrich/Zoch

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

253

(1985), Clyne (1991), Clyne/Hoeks/Kreutz (1988), Dittmar/Terborg (1990), Erben (1994), Günther (1992), Haarmann (1992), von Hahn (1983), Hinrichs (1984), House/Kasper (1981), House/Vollmer (1988), Koch (1985), Köhler (1981), Kretzenbacher (1991), Kreutz (in the present volume), Liebert (1994), Lovik (1990), Markkanen/Schröder (1988, 1992), Marui/Reinelt (1985), Öhlschläger (1989), Sanders (1992), Schröder (1989), Weinreich (1966), Weinrich (1993), Werlen (1983), Weydt/Harden/Hentschel/Rösler (1983), Wimmer (1987), Zierer (1987). Hebrew: Blum-Kulka (1982, 1985), Maschler (1994) Icelandic: Abraham (1990) Italian: Kinder (1985, 1987, 1988), Venier (1986) Japanese: Inoue (1983), Ohta (1991), Matsumoto (1989), Marui/Reinelt (1985) Polish: Duszak (1994), Wierzbicka (1991) Rumanian: Popescu, Radu (1985) Russian: Namsaraev (1991, in the present volume) Spanish: Berk-Seligson (1987). Research in the field of 'hedgingY'hedges' has increased rapidly after its inauguration by Zadeh (1965). Less than 1 % of the works in our database were published in the 1960's. The number of publications from the 1970's is about 10 %. A little more than 53 % of all collected publications were published in the 1980's. Publications of the 1990's already make up 35 % of our entries. This research area thus seems to be a flourishing field. 1965-1966 1972-1979 1980-1989 1990-1995 total:

2 30 142 96 270

0,7 % 11,1 % 52,6 % 35,6 °/o 100 %

The immense increase in research activities since the 1980's may be attributed to a broadening of the concept of hedges. In the 1970's, hedges were almost exclusively viewed from the semantic point of view. In the 1980's however, the concept of hedges was broadened because of the growing influence of pragmatic research. In pragmatics, hedges are seen as realizations of interactional/communicative strategies in contexts of mitigation, politeness, indirectness, etc.

3. Hedging research in pragmatics and other areas 3.1. Discourse-oriented hedging research Biber (1988, 240) gives a discourse-oriented definition of hedges: Hedges are informal, less specific markers of probability or uncertainty. Downtoners give some indication of the degree of uncertainty; hedges simply mark a proposition as uncertain. (...) Biber (1986a) finds hedges co-occurring with interactive features (e.g., first and second person pronouns and questions) and with other features marking reduced or generalized lexical content (e.g., general emphatics, pronoun /'/, contractions).

254

V. Hedging: The Concept, its Origins and a Bibliographical Guide

For more information on discourse-oriented hedging research, see also Markkanen/Schröder (1988, 1992, in the present volume), Salager-Meyer/Salas (1991), Salager-Meyer (1995). From the point of view of discourse analysis Prince et al. (1982) distinguish between two types of hedges, those that affect the truth conditions of propositions (approximators) and those that do not affect the truth conditions, but show the speaker's commitment to the truth value of the whole proposition (shields). The authors discovered hedges mainly in the discourse that was related to the physicians' uncertainty in the medical-technical domain. For more information on discourse-oriented hedging research, see also Markkanen/ Schröder (in the present volume), Nyyssönen (1988), Rounds (1982). Biber (1988), Ghoul (1982), Fräser (1990), Maschler (1994), Ohta (1991), Redeker (1990), Schiffrin (1991), Stubbs (1986), Vicher/Sankoff 1989 described hedges as discourse markers. House/Kasper (1981) viewed hedges as downgraders among modality markers; Abraham (1991), James (1983), Keseling (1989), Östman (1981), Rudolph (1986), Weydt et al. (1983) dealt with the hedging phenomenon under the concept of discourse particles. Weinreich (1966) dealt with the hedging phenomenon under the term metalinguistic operators. Following this Crismore/Markkanen (1993), Hinrichs (1984), Markkanen (1991) and Markkanen/Steffensen/Crismore (1993) viewed hedges as a part of metadiscourse. Blum-Kulka (1982, 1985), Blum-Kulka/Olshtain (1984), Chen (1993), Falkenberg (1981), Flowerdew (1991), Givon (1983), Gu (1990), Holmes (1984), House/Vollmer (1988), Koch (1985), Lavandera (1988), Lüger (1992), Markkanen (1985), Meyer (in the present volume), Nicoloff (1989), Panther (1981), Rolf (1984), Sökeland (1981), Werlen (1983) dealt with hedges from the point of view of speech act theory. For example, Panther (1981) distinguished in his article about indirect speech acts in scientific communication between three kinds of indirect speech acts, abstraction of the writer and reader, the use of hedged performatives and the formulation of requests. According to the author, these indirect speech acts are not necessarily an expression of politeness towards the reader. Moreover, they are used to construct the image of the author as an Objective scientist'. Flowerdew (1991) validates in his empirical study the claim that a representative speech act, namely the act of defining, can be subject to modification, just as a directive, a commissive, or an expressive can. The author's examples show that when the speech act of defining is performed, a lot more is going on the interpersonal dimension than the conveyance of truth or falsehood. The author separates indirectness and politeness from hedges. He defines hedges as a sub-category of mitigation. Last, but not least, Channell (1985), Henkel (1983), Lee (1987), Matsumoto (1989), Nikula (in the present volume), and Nyyssönen (1988) analyzed hedges from the point of view of conversation analysis. 3.2. Theoretical foundations of hedging research in pragmatics and related concepts Mitigation: Erben (1994), Flowerdew (1991), Labov (1973), Lüger (1992), Pomerantz (1984).

Fräser (1980), Henkel (1983),

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

255

Politeness: Blum-Kulka (1982), Blum-Kulka/Olshtain (1984), Bradac/Street (1989/90), Brown/Levinson (1978, 1987), Carrell/Konnecker (1981), Chen (1993), Flowerdew (1991), Fräser (1980), Giles/Bradac (1994), Gu (1990), Haarmann (1992), Hagge/Kostelnick (1989), Herbert (1990), Holmes (1993), House/Kasper (1981), lonescu-Ruxandoiu (1989), Janney/Amdt (1993), Johnson (1992), Kreutz (in the present volume), Labov (1973), Lakoff, R.T. (1974), Lavandera (1988), Leech (1983), Lüger (1992), Maier (1992), Matsumoto (1989), Morley (1987), Mulkay (1985), Myers (1989), Namsaraev (in the present volume), Ohta (1991), Östman (1981), Panther (1981), Weinrich (1993), Werlen (1983), Weydt/Harden/Hentschel/Rösler (1983). Vagueness: Blum-Kulka (1982), Busch-Lauer (1994), Caffi/Janney (1994), Channell (1980), Channell (1985), Channell (1990), Channell (1994), Erben (1994), Gruber (1993), Günther (1992), von Hahn (1983), Henkel (1983), Jahr (1992), Lakoff, G. (1992), Lüger (1992), Morley (1987), Müller (1980), Pinkai (1980/81,1985a, 1985b), Powell (1985), Rieger (1989), Rudolph (1986), Schneider (1988), Scriven (1976), Sörensen (1990), Spies (1993), Werlich (1983),Wierzbicka (1986), Wolski (1980), Zuck/Zuck (1987). 3.3. Hedging research in areas other than pragmatics Hedging research from the point of view of logics: Drösser (1994), Gippert (1993), Kiefer (1994), Kusunose (1976), Lakoff, G. (1972, 1973), McCawley (1981), McNeill/Freiberger (1994), Spieß (1993), Wolski (1980), Zadeh (1965, 1972, 1975). Hedging research from the point of view of semantics: Abraham (1984), Bußmann (1990), Channell (1980), Coates (1983), Damerau (1977), Gippert (1993), Glass/Holyak (1974), Günther (1992), Hinrichs (1984), Kay (1987), Kayser (1978), Kusunose (1976), Lakoff, G. (1982, 1986, 1987), Lee (1987), Müller (1980), Neubert (1985), Öhlschläger (1989), Perkins (1982), Pinkai (1985a, 1985b), Popescu (1985), Prince (1976), Rieger (1989), Taylor (1994), Wachtel (1980), Weinreich (1966), Werlich (1983), Wierzbicka (1986), Zadeh (1972). Hedging research from the point of view of rhetorics and stylistics: BuschLauer (1994), Crismore/Markkanen/Steffensen (1993), Crismore/Vande Kopple (1990) Crismore/Vande Kopple (1990b), Danet (1976), Duszak (1994), Fiedler (1985), Kretzenbacher (1991), Lachowicz (1981), Lakoff, R. T. (1977), Lauer (1985), Lindeberg (1994), Loewenberg (1982), McLain (1977), Meyer (in the present volume), Perttunen (1990), von Polenz (1981), Sanders (1992), Staley (1982), Stubbs (1986), Wimmer (1987).

4. Possible applications of hedging research The most important possible applications of hedging research are in the field of applied linguistics. Holmes (1982b) conducted a study of the difficulties ESL-learners (ESL = English as a Special Language) have as far as expressions

256

V. Hedging: The Concept, its Origins and a Bibliographical Guide

of epistemic modality are concerned. She pointed out "three sources of potential difficulty, the problem of establishing the precise degree of certainty expressed by particular linguistic forms; the range of linguistic forms available for signalling this aspect of meaning; and the interaction of different types of meaning in different contexts." Holmes made some suggestions which may assist second language learners wishing to develop competence in expressing and interpreting degrees of certainty and convictions in English. For more information on questions concerning ESL, see also James/Garrett (1991), R.T. Lakoff (1975), Pindi/Bloor (1987), Simpson (1990) and Holmes (1988). Adams (1986), Bamberg/Damrad-Frye (1991), Bateman (1988), Dittmar/ Terborg (1990), Green (1984) conducted research from the point of view of language acquisition studies. Markkanen (1991), Markkanen/Schröder (1988), Neubert (1985), Perttunen (1990) and Schroder (1989) viewed hedging as a translation problem and Kolde (1989) dealt with hedges and hedging from the point of view of lexicography. Communication research: Aijmer (1987), Bradac/Hemphill/Tardy (1981), Brooke/Ng (1986), Cameron (1992), Coates (1989), Crismore/Vande Kopple (1988), Di Pietro (1982), Erickson/Lind/ Johnson/O'Barr (1978), Hagge/Kostelnick (1989), Herbert (1990), Holmes (1986), Holmes (1990), Hosman (1989), Hosman/Wright (1987), Kramarae (1992), R.T. Lakoff (1977), Loewenberg (1982), McMullen/Pasloski (1992), Preisler (1986), Rasmussen/Moely (1986), Rieger (1989), Semiloff (1977), Vinson/Johnson (1989), Wright/Hosman (1983). Academic communication research: Bell/Zahn/Hopper (1984), Busch-Lauer (1994), Butler (1990), Channel (1990), Clyne (1991), Dubois (1987), Duszak (1994), Fiedler (1985), Gilbert/Mulkay (1984), Grabe/Kaplan (in the present volume), von Hahn (1983), Hyland (1994), Janney/Arndt (1993), Köhler (1981), Kretzenbacher (1991), Lachowicz (1981), Liebert (1994), Lindeberg (1994), Luukka/Markkanen (in the present volume), Markkanen/Schröder (1992), Master (1991), Mauranen (in the present volume), Meyer (1989), Meyer (1990), Meyer (in the present volume), Myers (1989), Namsaraev (in the present volume), Panther (1981), Perttunen (1990), von Polenz (1981), Rounds (1982), Salager-Meyer (1992), Salager-Meyer/Salas (1991), Sanders (1992), Schröder (1988), Simpson (1990), Skelton (in the present volume), Sökeland (1981), Ventola (1992), Wilss (in the present volume), Wimmer (1987). Intercultural communication studies: Blum-Kulka (1984), Brown/Levins on (1978), Clyne (1991), Crismore/Markkanen/Steffensen (1993), Duszak (1994), Haarmann (1992), Holmes (1982a, 1982b, 1988), House/Kasper (1981), Janney/Arndt (1993), Kinder (1988), Labov 1973), Lüger (1992), Markkanen (1991), Markkanen/Schröder (1988, 1992), Marui/Reinelt (1985), Nyyssönen (1988), Schröder (1989),Ventola (1992), Wierzbicka (1991).

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

257

References Abraham, W. 1984 "Domain, Domain Adverbs and the Theory of Metaphor", Papiere zur Linguistik 30,3-21. Abraham, W 1990 "Modalverben in der Germania", in Iwasaki, E. (ed.): Akten des VIII Internationakn Germanisten Kongresses. München: iudicium, 109-118. Abraham, W. (ed.) 1991 Discourse Particles. Descriptive and Theoretical Investigations on the Logical, Syntactic, and Pragmatic Properties of Discourse Particles in German. Amsterdam: John Benjamins. Adams, A. K. 1986 "Sentence Frame Effects on Children's Category Judgements", Papers and Reports on Child Language Development 25, 1-8. Aijmer, K. 1984 "'Sort of and 'kind of in English Conversation", StudiaLinguistica 38:2,118-128. Aijmer, K. 1987 "Discourse Variation and Hedging", Coslerus 57, 1-18. Ayto, Jo. and Walford, G. 1986 "Of or Pertaining to Dictionaries", English Today 6, 39-40. Bateman, J. 1988 "*We had a hedgehog in our Garden on Sunday'—a Strategy for Maximizing Discourse", Modern English Teacher 16:2,22-23. Bell, RA.; Zahn, C.J. and Hopper, R. 1984 "Disclaiming: A Test of Two Competing Views", Communication Quarterly 32:1, 28-36. Berk-Seligson, S. 1987 "The Intersection of Testimony Styles hi Interpreted Judicial Proceedings: Pragmatic Alterations in Spanish Testimony", Linguistics 25:6, 1087-1125. Biber, D. 1988 Variation across Speech and Writing. Cambridge: Cambridge University Press. Bloor, T. and Pindi, M. 1990 "Schematic Structure in Economic Forecasts", in Dudley-Evans, T./Henderson, W. (eds.): The Analysis of Economics Discourse. Modern English Publications and the British Council (ELT Documents: 134), 55-66. Blum-Kulka, S. 1982 "Learning to Say what You Mean in a Second Language", Applied Linguistics 3/1, 29-59. Blum-Kulka, S. 1985 "Modifiers as Indicating Devices: the Case of Requests", Theoretical Linguistics 12:2/3, 213-229. Blum-Kulka, S. and Olshtain, E. 1984 "Requests and Apologies. A Cross-Cultural Study of Speech Act Realization Patterns", Applied Linguistics 5/3, 196-213. Bradac, J.J.; Hemphill, M. R. and Tardy, C. H. 1981 "Language Style on Trial: Effects of 'Powerful' and 'Powerless* Speech upon Judgements of Victims and Villains", Western Journal of Speech Communication 45: 4, 327-341. Bradac, J. and Street, R. 1989 "Powerful and Powerless Styles of Talk—A Theoretical Analysis of Language and Impression Formation", Research on Language and Social Interaction 23, 195-241. Brainbridge, O. 1994 "Sex Differences", in Asher, R.E. (ed.): The Encyclopedia of Language and Linguistics. Vol. 7. Oxford: Pergamon Press, 3864-3868.

258

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Brooke, M.E. and Ng, S.H. 1986 "Language and Social Influence in Small Conversational Groups", Journal of Language and Social Psychology 5: 3, 201-210. Brown, P. and Levinson, S. 1978 "Universals in Language Usage: Politeness Phenomena", in Goody, E, N. (ed.): Questions and Politeness. Cambridge: Cambridge University Press, 56-310. Brown, P. and Levinson, S. 1987 Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press. Busch-Lauer, I. 1994 "Kontrastive Fachtextanalyse am Beispiel englischer und deutscher Zeitschriftenartikel der Medizin", in Spillner, B. (ed.): Fachkommunikation. (Kongreßbeiträge zur 24. Jahrestagung der Gesellschaft für Angewandte Linguistik GAL e.V.). Frankfurt/Main: Peter Lang, 71-72. Buscha, J.; Heinrich, G. and Zoch, I. 1985 Modalverben. Leipzig: VEB Verlag Enzyklopädie. Bußmann, H. 1990 "Heckenausdruck", in Bußmann, H.: L·xikon der Sprachwissenschaß. Stuttgart: Kröner, 304-305. Buüer, C.S. 1990 "Qualifications in Science: Modal Meanings in Scientific Texts", in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse. Newbury Park: Sage, 137-170. Caffi, C. and Janney, R. W. 1994 "Toward a Pragmatics of Emotive Communication", Journal of Pragmatics 22, 325-373. Cahill,RJ. and Foley, J. J. 1973 Evaluation and Evaluative Research in an Urban Bilingual Program. Debham/Mass.: Heuristics. Cameron, D. 1992 "'Not Gender Difference But the Difference Gender Makes'—Explanation in Research on Sex and Language", International Journal of the Sociology of Language 94, 13-26. Carrell, P. L. and Konnecker, B. H. 1981 "Politeness: Comparing Native and Non-native Judgments", Language Learning 31/1, 17-30. Chafe, W. and Danielwicz, J. 1987 Properties of Spoken and Written Language. California University, Berkeley: Center for the Study of Writing (Technical Report 5). Chambers, F. and McDonough, J. (eds.) 1980 Aspects of English for Special Purposes. Occasional Papers, No. 22. (Papers written by students in the M.A. in Applied Linguistics program, Essex University, Colchester/England). ChannellJ.M. 1980 "More on Approximations", Journal of Pragmatics 4, 461-476. ChannellJ.M. 1985 "Vagueness as a Conversational Strategy", Nottingham Linguistic Circular 14,3-24. ChannellJ.M. 1990 "Precise and Vague Quantities in Writing on Economics", in Nash, W. (ed.): The Writing Scholar: Studies in Academic Discourse. Newbury Park: Sage, 95-117. ChannellJ.M. 1994 Vague Language. Oxford: Oxford University Press. Chen, R. 1993 "Responding to Compliments: A Contrastive Study of Politeness Strategies between American English and Chinese Speakers", Journal of Pragmatics 20, 49-75.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

259

Cho, C.-H. 1987 "On 'Dramatic' Particles in English", Language Research 23:3, 379-393. Ghoul, J.-C1. 1982 "Les Operateurs simiotiques en discours", Recberches simiotiques 2:2, 150-165. Clemen, G. 1990 Hecken in englischen Wirtschaftstexten. Unpublished MA-Thesis. Universität Gesamthochschule Siegen. Clemen, G. "The Concept of Hedging: Origins, approaches and definitions", in the present volume. Clyne, M. 1991 "The Socio-Cultural Dimension: the Dilemma of the German-Speaking Scholar", in Schröder, H. (ed.): Subject-orientied Texts. Berlin: de Gruyter, 49-67. Clyne, M.; Hoeks, J. and Kreutz, H.-J. 1988 "Cross-Cultural Responses to Academic Discourse Patterns", Folia Linguistica 22: 3^, 457-475. Coates.J. 1983 The Semantics of the Modal Auxiliaries. London: Groom Helm. Coates, J. 1989 "Women's Speech, Women's Strength?", York Papers in Linguistics 3, 65-76. Crismore, A.; Markkanen, R. and Steffensen, M. S. 1993 "Metadiscourse in Persuasive Writing: A Study of Texts Written by American and Finnish University Students". Written Communication, 10:1, 39-71. Crismore, A. and Vande Kopple, W. J. 1990 "Rhetorical Contexts and Hedges", Rhetoric Society Quarterly 20:1, 49-59. Crismore, A. and Vande Kopple, W.J. 1988 "Readers' Learning from Prose: The Effects of Hedges", Written Communication 5:2, 184-202. Crismore, A. and Vande Kopple, WJ. "Hedges and Readers: Effects on Attitudes and Learning", in the present volume. Crosby, R and Nyquist, L. 1977 "The Female Register: An Empirical Study of LakofPs Hypothesis", Language in Society 6:3, 313-322. Damerau, F. J. 1977 "On 'Fuzzy' Adjectives'", Linguistics 196:3, 57-64. Danet, B. 1976 "Speaking of Watergate: Language and Moral Accountability", Centrum 4:2, 105-138. Danet, B. and Bogoch, B. 1992 "From Oral Ceremony to Written Document: The Transitional Language of Anglo-Saxon Wills", Language and Communication 12:2, 95-122. Darian, Steven 1995 "Hypotheses in introductory science texts", IRAL, Vol. /2, 83-108. DiPietro,R.J. (ed.) 1982 Linguistics and the Professions. New Jersey: Ablex Publishing Corp. Dirven, R. 1989 A User's Grammar of English: Word, Sentence, Text, Interaction. Frankfurt/Main: Peter Lang. Dittmar, N. and Terborg, H. 1990 "Modality and Second Language Learning", Materialien Deutsch als Fremdsprache 29, 251-286. Drösser, C. 1994 Fuzzy logic—Methodische Einführung in krauses Denken. Reinbek: Rowohlt. Dubois, B.-L. 1987 "'Something on the Order of around Forty to Forty-Four'-Imprecise Numerical Expressions in Biomedical Slide Talks", Language in Society 16:4, 527-541.

260

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Duszak, A. 1994 "Academic Discourse and Intellectual Styles", Journal of Pragmatics 21, 291-313. Edwards, C. 1984 "'Stop me if You've Heard this one' — Narrative Disclaimers as Breakthrough into Performance", Fabula 25:3/4, 214-228. Erben, J. 1994 Sprachliche Signale zur Markierung der Unsicherheit oder Unscharfe von Aussagen im Neuhochdeutschen. Berlin: Akademie Verlag. (Sitzungsberichte der Sächsischen Akademie der Wissenschaften zu Leipzig. Philologisch-historische Klasse 134/3). Erickson, B.; Lind, EA.; Johnson, B.C. and O'Barr, W.M. 1978 "Speech Style and Impression Formation in a Court Setting: The Effects of 'Powerful' and 'Powerless' Speech", Journalof Experimental Social Psychology 14: 3, 266-279. Falkenberg, G. 1981 "Epistemische Operatoren und illokutionäre Kraft", in Hindelang, G. and Zillig, W. (eds.): Sprache. Verstehen und Handeln. Bd.2. Tübingen: Niemeyer, 193-203. Fiedler, S. 1985 Text- und Stilmerkmale englischer Zeitschriftenartikel der Pädagogik", in Graustein, G. et al. (ed.): Beiträge zur englischen Textlinguistik. (= Linguistische Studien. Reihe A, Arbeitsberichte 135, Akademie der Wissenschaften der DDR), 124-133. Flowerdew, J. 1991 "Pragmatic Modifications on the 'Representative' Speech Act of Defining", Journal of Pragmatics 15,253-264. Fräser, B. / 1975 "HedgedPerformatives", Cole, P. and Morgan,). L. (eds.): Syntax and Semantics. Vol. 3. New York: Academic Press, 187-210. Fräser, B. 1980 "Conversational Mitigation", Journal of Pragmatics 4:4,341-350. Fräser, B. 1990 "An Approach to Discourse Markers", Journal of Pragmatics. 14:3. 383-395. Gilbert, N. and Mulkay, M. 1984 Opening Pandora's Box. A Sociological Analysis of Scientists' Discourse. Cambridge: Cambridge University Press. Giles, H. and Bradac.J.J. 1994 "Speech Styles: Attitudes and Inferences", in Asher, R.E. (ed.): The Encyclopedia of Language ana Linguistics. Vol.8. Oxford: Pergamon Press, 4260-4263. Gippert,J. 1993 "Heckenausdruck", in Glück, H. (ed.): Melder Lexikon Sprache. Stuttgart: Metzler, 243. Givon, T. 1993 English Grammar: A Function-Based Introduction. Amsterdam: Benjamins. Glass, A. L. and Holyoak, K. J. 1974 "Alternative Conceptions of Semantic Theory", Cognition 3/4, 313-339 Gottschalk, K.-D. 1979 "Unverbürgte und unverbindliche Äußerungen im Englischen", Grazer Linguistische Studien 9, 34-47. Grabe, W. and Kaplan, R. "On the Writing of Science and the Science of Writing: Hedging in Science Text and Elsewhere", in the present volume. Green, M. 1984 "Cognitive Stage Differences in Types of Speaker Uncertainty Markers", Language and Speech 27:4, 323-331. Gruber, H. 1993 "Political Language and Textual Vagueness", Pragmatics 3:1, 1-28.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

261

Grundy, P. 1989 "Obviously, Natural Language Maxim Hedges, x-bar Syntax, ok?", Linguislica Silesiana 10, 49-60. Gu.Y. 1990 "Politeness Phenomena in Modern Chinese", Journal of Pragmatics 14, 237-257. Günther, U. 1992 "und aso das isch gar need es Tabu bi üs, nei, überhaupt need". Sprachliche Strategien bei Phone-in-Sendungen am Radio zu tabuisierten Themen. Bern et al.: Peter Lang. Haarmann, H. 1992 "Deference, Politeness and the Status of Language - a Study in the Variability and Changeability of a Crucial Relationship", in Ammon, U. and Hellinger, M. (eds.): Status Change of Languages. Berlin, New York: Walter de Gruyter, 521-540. Hagge, J. and Kostelnick, C. 1989 "Linguistic Politeness in Professional Prose: A Discourse Analysis of Auditors' Suggestion Letters, with Implications for Business Communication Pedagogy", Written Communication 6: 3, 312-339. von Hahn, W. 1983 Fachkommunikation: Entwicklung, linguistische Konzepte, betriebliche Konzepte. Berlin, New York: Walter de Gruyter. Heinämäki, O. 1972 "Almost", Papers from the Second Scandinavian Conference on Linguistics 1972, 64-71. HenkelJ. 1983 "Mitigation, Indirection, and Women's Speech", Papers from the Mid-America Linguistics Conference 1982, 76-90. Herbert, R. K 1990 "Sex-Based Differences in Compliment Behavior", Language in Society 19:2, 201-224. Herrmann, L. 1975 "On 'in that'", Proceedings of the First Berkeley Annual Meeting of the Berkeley Linguistics Society 1, 189-195. Hinrichs, U. 1984 "Studentensprache, Spontisprache", Mutterspracht. 94:5/6, 404-416. Holmes, J. 1982a "The Functions of Tag Questions", English Language Research Journal 3, 40-65. Holmes, J. 1982b "Expressing Doubt and Certainty in English", RELCJournal 13:2, 9-29. Hohnes, J. 1984 "Hedging your Bets and Sitting on the Fence: Some Evidence for Hedges as Support Structures", Te Reo 27, 47-62. Hohnes, J. 1986 "Functions of *You Know* in Women's and Men's Speech", Language in Society 15:1, 1-21. Hohnes, J. 1988 "Doubt and Certainty in ESL Textbooks", Applied Linguistics 9:1, 21-44. Holmes, J. 1990 "Hedges and Boosters in Women's and Men's Speech", Language and Communication 10:3, 185-205. Hohnes, J. 1993 "New Zealand Women are Good to Talk to: An Analysis of Politeness Strategies in Interaction", Journal of Pragmatics 20, 91-116. Hosman, L. A. 1989 "The Evaluative Consequences of Hedges, Hesitations, and Intensifiers: Powerful and Powerless Speech Styles", Human Communication Research 15:3, 383-406.

262

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Hosman, L. A. and Wright,]. W. 1987 "The Effects of Hedges and Hesitations on Impression Formation in a Simulated Courtroom Context", Western Journal of Speech Communication 51:2, 173-188. House, J. and Kasper, G. 1981 "Politeness Markers in English and in German", in Coulmas, F. (ed.): Conversational Routines. The Hague: Mouton de Gruyter, 157-185. House, J. and Vbllmer, H. J. 1988 "Speech Act Performance in German: On the Realization of the Speech Actions 'Request' and 'Apology"", Linguistische Berichte 114, 114-133. Hübler, A. 1983 Understatements and Hedges in English. Amsterdam: John Benjamins. Hyland, K. 1994 "Hedging in Academic Writing and Textbooks", English for Specific Purposes 13/3,239-256. Inoue, K. 1983 "An Analysis of a Cleft Conditional in Japanese - Where Grammar Meets Rhetoric", Journal of Pragmatics 7:3, 251-262. lonescu-Ruxandoiu, L. 1989 "Remarks on Politeness Strategies in the Dialects Spoken in Muntenia", Revue RoumainedeLinguistiqueW·. 4, 340-355. Jahr, S. 1992 "Zur Vagheitsproblematik und Prototypentheorie", Deutsch als Fremdsprache 29:3, 167-172. James, A. R. 1983 "Comprisers in English: A Cross-Disciplinary Approach to their Interpersonal Significance", Journal of Pragmatics 7:2, 191-206. James, C. and Garrett, P. (eds. 1991 Language Awareness in the Classroom. London: Longman Group. Janney, R. and Arndt, H. 1993 "Universality and Relativity in Cross-Cultural Politeness Research", Multilingua 12, 13-50. Johnson, C. and Vinson, L. 1990 "Placement and Frequency of Powerless Talk and Impression Formation", Communication Quarterly. 38:4,325-333. Johnson, D. M. 1992 "Compliments and Politeness in Peer-Review Texts", Applied Linguistics 13, 51-71. Kärkkäinen, E. 1990 "Face Saving and Epistemic Modality", in Nyyssönen, H.; Kuure, L.; Kärkkäinen, E. and Raudaskoski, P. (eds.): Proceedings from the Second Finnish Seminar on Discourse Analysis. Oulu: University of Oulu, 64-76. Kärkkäinen, E. 1988 "Episteemisen modaalisuuden funktioista englanninkielisessä keskustelussa", Acta Universitatis Ouloensis, B 14,147-160. Kay, P. 1987 "Linguistic Competence and Folk Theories of Language - two English Hedges", in Quinn, D. and Quinn, N. (eds.): Cultural Modals in Language and Thought. Cambridge: Cambridge UP, 67-77. Kay, P. 1984 "The "Kind of/Sort oP Construction", Proceedings of the Annual Meeting of the Berkeley Linguistics Society 10, 157-171. Kayser, H. 1978 "Performativmodale, weltkonstitutive und deskriptive Einheiten als Grundelemente der 'atomaren Texte'", in Biasci, C. and Pritsche, J. (eds.): Texttheone und Textrepräsentation. Hamburg: Buske, 145-176.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

263

Keseling, G. 1989 "Die Partikel 'äh' — ein paraverbales Element im Sprachsystem?", in Weydt, H. (ed.): Sprechen mit Partikeln. Berlin, New York: Walter de Gruyter, 575-591. Kiefer, F. 1994 "Modality", in Asher, R. E./Simpson, J. M. Y. (eds.): The Encyclopedia of Language and Linguistics. Vol. 5. Oxford: Pergamon Press, 2515-2520. Kinder, J.J. 1985 "Verbal Strategies for Marking Interference in New Zealand Italian", Rivista Italiana di Dialeftologia: Scuola Societa Territorio 9: 1/9, 103-128. Kinder, J.J. 1987 "Code Switching and Social Integration in Bilingual Conversation", Australian Review of Applied Linguistics 4,37-51. Kinder, J.J. 1988 "Transference Markers in New Zealand Italian: Constructing Context and Negotiating Identity", Australian Review of Applied Linguistics 11:1, 9-21. Klockow, R. 1976 "Gänsefüßchen-Semantik: eine Ergänzung zu LakofFs 'Hedges'", in Weber, H. and Weydt, H. (eds.): Sprachtheorie und Pragmatik. Tübingen: Niemeyer, 235-245. Koch, W. 1985 "Das Modalverb als Handlungsevaluator", in Akten des VII. Internationalen Gemanisten-Kongresses Göttingen 1985. Tübingen: Niemeyer, 381-392. Köhler, C. 1981 "Zum Gebrauch von Modalverben und Passivfügungen in der deutschenFachsprache der Technik", in von Hahn, W. (ed.): Fachsprachen. Darmstadt: Wissenschaftliche Buchgesellschaft, 239-261. König, E.; Stark, D. and Requardt, S. 1990 Adverbien und Partikeln. Heidelberg: Groos. Kolde, G. 1989 "Probleme der Beschreibung von sogenannten Heckenausdrücken im allgemeinen einsprachigen Wörterbuch", in Hausmann, F. J. et al. (eds.): Wörterbücher: ein internationales Handbuch zur Lexikographie/Dictionaries: An International Encyclopedia of Lexicography. Berlin, New York: Walter de Gruyter, 855-862. Kramarae, C. 1992 "Gender: How She speaks", in Ryan, E.B./Giles, H. (eds.): Attitudes toward Language Variation. London, 84-98. Kretzenbacher, H. 1991 "Syntax des wissenschaftlichen Fachtextes", Fachsprache. 3-4, 118-137. Kreutz, H. "Some Observations on Hedging Phenomena and Modifying Devices as Regional Markers in the Speech of Young East Germans", hi the present volume. Kusunose, J. 1976 "Recent Trends in Generative Semantic", Sophia Linguistica 2,23-31. Labov, W. A. 1973 "Modes of Mitigation and Politeness", in De Stefano, J.F. (ed.): Language, Society and Education. Worthington/Ohio: C. A. Jones. Lachowicz, D. 1981 "On the Use of the Passive Voice for Objectivity, Author Responsability and Hedging in EST", Science of Science 2:6, 105-115. Lakoff, G. 1972 "Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts", Papers from the Eighth Regional Meeting of the Chicago Linguistic Society 1972,183-228. Reprinted in: Journal of Philosophical Logic. 1973, 2: 4, 458-508. Lakoff, G. 1973 "Lexicography and Generative Grammar. Hedges and Meaning Criteria", in Zgusta, L. (ed.): Probleme des Wörterbuchs. Darmstadt: Wissenschaftliche Buchgesellschaft, 150-161.

264

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Lakoff, G. 1982 Lakoff, G. 1986 Lakoff, G. 1987 LakofT, G. 1992

Categories and Cognitive Models (Series A. Paper No. 96). Printed by G. Lakoffand Reproduced by L. A. U. T. Trier. "Frame Semantic Control of the Coordiante Structure Constraint", Papers from the Parasession on Pragmatics and Grammatical Theory at the Twenty-Second Regional Meeting of the Chicago Linguistic Society 22:2, 152-167. Women, Fire and Dangerous Things. Chicago: The University of Chicago Press.

"Metaphor and War: The Metaphor System Used to Justify War in the Gulf, in Pütz, M. (ed.): Thirty Years of Linguistic Evolution. Studies in Honour ofRmi Dirnen on the Occasion of his Sixtieth Birthday. Philadelphia/Amsterdam: Benjamins, 463-481. LakofT, G. and Johnson, M. 1980 Metaphores We Live by. Chicago: University of Chicago Press. LakofT, R. T. 1974 "The Logic of Politeness: or Minding your p's and q's", Papers from the Regional Meeting of the Chicago Linguistic Society 9,292-305. LakofT, R. T. 1975 "Linguistic Theory and the Real World", LanguageL·arning25:2, 309-338. LakofT, R. T 1977 "Women's Language", Language and Style 10:4, 222-247. Lauer, I. A. 1985 "Text- und Stilmerkmale englischer Rezensionen in der Psychologie", in Graustein,G. et al. (ed.): Beiträge zur engischen Textlinguistik. (Linguistische Studien, Reihe A, Arbeitsberichte 135), 140-149. Lavandera, B. R. 1988 "The Social Pragmatics of Politeness Forms", in Ammon, U.; Dittmar, N. and Mattheier, K. (eds.): Sociolinguistics. Soziolinguistik. Berlin, New York: Walter de Gruyter, 1196-1205. Lee, D. 1987 "The Semantics of 'jusi'" Journal of Pragmatics 11:3, 377-398. Leech, G.N. 1983 Principles of Pragmatics. London: Longman. Lehrer, A. 1989 "Remembering and Representing Prose: Quoted Speech as a Data Source", Discourse Processes 12:1, 105-125. Liebert, W.-A. 1994 "Fremdagens, Eindringlinge, Piraten und unsichtbare Killer. Metaphern-Modelle als Übergangsphänomene zwischen Theorietexten, fachlichen PR-Texten und populärwissenschaftlichen Texten zum Thema AIDS und HIV-Forschung", in Spillner, B. (ed.): Fachkommunikation (Kongreßbeiträge zur 24. Jahrestagung der Gesellschaft für Angewandte Linguistik GAL e.V.). Frankfurt/Main et al.: Peter Lang, 85-87. Lindeberg, A.-C. 1994 "Rhetorical Conventions in Scholarly Articles in Economics and Business Sciences: A Study of Introductions with a Special Reference to Knowledge Claims", in Cmejrkova, S. et al. (eds.): Writing vs. Speaking: Language, Text, Discourse, Communication. Proceedings of die Conference Held at the Czech Language institute of the Academy of Sciences of the Czech Republic, Prague, October 14-16, 1992. Tübinger Beiträge zur Linguistik 392). Tübingen: Narr, 317-324. Loewenberg, I. 1982 "Labels and Hedges: The Metalinguistic Turn", Language and Style 15:3, 193-207.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

265

Lovik, T. A. 1990 "Geben Sie mir so'n Streuselapfel: Hedging in Authentic Speech", Die Unterrichtspraxis/Ieaching German 23:2, 121-28. Lüger, H.-H. 1992 Sprachliche Routinen und Rituale (Werkstattreihe Deutsch als Fremdsprache 36). Frankfurt am Main et al.: Peter Lang. Luukka, M.-R. and Markkanen, R. "Impersonalizations as a Form of Hedging", in the present volume. Maier, P. 1992 "Politeness Strategies in Business Letters by Native and Non-Native English Speakers", English for Specific Purposes 2/3, 189-205. Markkanen, R. 1985 Cross Language Studies in Pragmatics (Jyväskylä Cross Language Studies 11). Jyväskylä: University of Jyväskylä. Markkanen, R. 1989 "Hedging and Modality". Jäntti, A. (ed.): Probleme der Modalität in der Sprachforschung. Jyväskylä: University of Jyväskylä, 137-148. Markkanen, R. 1991 "Metadiscourse in Intercultural Communication", Eriskoiskielet ja käännösteoria. VAKKI-seminaari XI. Vöyri. 09-10.02.1992. Vaasan yliopisto: Kielten laitos, 186-194. Markkanen, R. and Schröder, H. 1988 "Hedging as a Translation Problem in Scientific Texts", in Lauren, C. and Nordman, M. (eds.): Special Languages: From Humans Thinking to Thinking Machines. Clevedon/Philadelphia: Multilingual Matters, 171-175. Markkanen, R. and Schröder, H. 1992 "Hedging and its Linguistic Realization in German, English, and Finnish Philosophical Texts: A Case Study", in Nordmann, M. (ed.): Fachsprachliche Miniaturen. Festschrift für Christer Laurin. Frankfurt am Main et al.: Peter Lang, 121-130. Markkanen, R. and Schröder, H. "Hedging: A Challenge for Pragmatics and Discourse Analysis", in the present volume. Markkanen, R.; Steffensen, M. and Crismore, A. 1993 "Quantitative contrastive of Metadiscourse. Problems in Design and Analysis of Data", Papers and Studies in Contrastive Linguistics 28, 137-151. Marui, I. and Reinelt, R. 1985 "ich will dem Mensch werde...- Modalisierungen in deutschen Aufsätzen japanischer Studenten", in Rehbein, J. (ed.): Interkulturelle Kommuikation (Kommunikation und Institution 12). Tübingen: Narr, 190-221. Maschler, Y. 1994 "Metalanguaging and Discourse Markers in Bilingual Communication", Language in Society 23, 325-366. Master, P. 1991 "Active Verbs with Inanimate Subjects in Scientific Prose", English for Specific Purposes 10:1, 15-33. Matsumoto, Y. 1989 "Politeness and Conversational Universals—Observations in Japanese", Multilingua 8,2/3,207-221. Mauranen, A. "Hedging in Language Reviser's Hands", in the present volume. McCawley.J. D. 1981 "Fuzzy Logic and Restricted Quantifiers", Philosophy and Grammar, 101-118. McLain, R. 1977 "The Problem of 'Style': Another Case in Fuzzy Grammar", Language and Style 10:1, 52-65.

266

V. Hedging: The Concept, its Origins and a Bibliographical Guide

McMullen, L. M. and Pasloski, D. D. 1992 "Effects of Communication Apprehension, Familiarity of Partner, and Topic on Selected 'Women's Language' Features", Journal of Psycholinguistic Research 21:1, 17-30. McNeill, D. and Freiberger, P. 1994 Fuzzy logic. Die unscharfe Logik erobert die Technik. München: Knaur. Meier-Fohrbeck, T. 1984 "Reflections on the Illocutionary Structure of Texts-With Some Suggestions for Foreign-Language Teaching", Linguistische Berichte 89, 64-81. Meyer, H.-J. 1989 "Modality in Science Texts", Special Language/Fachsprache 3/4, 127-135. Meyer, H.-J. 1990 "Functions of Modal Verbs in Physics Texts", Fachsprache. International Journal of LSP 12/1-2, 58-73. Meyer, P. G. "Hedging Strategies in Written Academic Discourse: Strengthening the Argument by Weakening the Claim", in the present volume. Morley.J. D. 1987 Grammatik des Lächelns. Japanische Innenansichten. Reinbek/Hamburg: Wunderlich. Translated into German by Weber-Schäfer, P. Original tide: Pictures from the Water Trade. An Englishman in Japan. London: Andro Deutsch. Müller, K. 1980 "Interaktionssemantik", Deutsche Sprache 8:4,289-305. Mulkay, M. 1985 "Agreement and Disagreement in Conversation and Letters", Text 5, 201-227. Myers, G. 1989 "The Pragmatics of Politeness in Scientific Articles", Applied Linguistics 10:1,1-35. Namsaraev, V. 1991 "Linguistic Realisation of Hedging in Russian Scientific Text", Erikoiskielet ja käänösteoria. Vakki seminaari. XL 195-207. Namsaraev, V. "Hedging in Russian Scientifc Writing within the Field of Sociology", in the present volume. Neubert, A. 1985 "Textlinguistik des Übersetzens", in Graustein, G. et al. (ed.): Beiträge zur englischen Textlinguistik. (Linguistische Studien, Reihe A, Arbeitsberichte 135, Akademie der Wissenschaften der DDR), 15-24. Newkirk, T. 1985 "The Hedgehog or the Fox: The Dilemma of Writing Development", Language Arts 62:6, 593-603. Nicoloff, F. 1989 "Threats and Illocutions", Journal of Pragmatics 13, 501-522. Nikula, T. 1993 "The Use of Lexical Certainty Modifiers by Non-Native Speakers of English", in Bouton, L. F. and Kachru, Y. (eds.): Pragmatics and Language Learning Urbana/Illinois: Devision of English as an International Language/University of Illinois at Urbana, 126-142. Nikula, T. "Interlanguage View of Hedging", in the present volume. Nyyssönen, H. 1988 "Modality, Interactivity and Discourse Strategies", Acta Universitatis Ouluensis. B 14, 227-238. O'Connor, M. 1993 Writing Successfully in Science. London: Chapman & Hall. Öhlschläger, G. 1989 Zur Syntax und Semantik der Modalverben des Deutschen. Tübingen: Niemeyer.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

267

Ohta, A. S. 1991 "Evidentiality and Politeness in Japanese", Issues in AppliedLinguistics 2:2,211-238. Östman, J. 1981 You know: A Discourse-Functional Approach. Amsterdam: John Benjamins. Palmer, F. R. 1987 Mood and Modality. Cambridge: Cambridge University Press. Panther, K.-U. 1981 "Einige typische indirekte sprachliche Handlungen im wissenschaftlichen Diskurs", in Bungarten, T. (ed.): Wissenschafissprache. München: Fink, 231-260. Perkins, M. 1982 "The Core Meaning of the English Modals", Journal of Linguistics 18:2, 245-273. Perttunen, J. M. 1990 The Words between. A Handbook for Scientists needing English, with Examples mainly from Biology and Medicine. Helsinki: Kustannus Oy Duodecim. Pindi, M. and Bloor, T. 1987 "Playing Safe with Predictions: Hedging, Attribution and conditions in Economic Forecasting", in Bloor, T. and Norrish, J. (eds.): Written Language (GILT 55), 555-569. Pinkal, M. (1980/1981): "Semantische Vagheit: Phänomene und Theorien", Linguistische Berichte 70, 1-26; 72, 1-26. Pinkal, M. 1985a "Kontextabhängigkeit, Vagheit, Mehrdeutigkeit", in Schwarze, C. and Wunderlich, D. (eds.): Handbuch der Lexikologie. Königstein/Ts: Athenäum, 27-63. Pinkal, M. 1985b "Vagheit und Ambiguität" (Metrische Vagheitsformen), in von Stechow, A. and Wunderlich, D. (eds.): Semantik/Semantics. Berlin, New York: Walter de Gruyter, 257-269. von Polenz, P. 1981 "Über die Jargonisierung von Wissenschaftssprache und wider die Deagentivierung", in Bungarten, Theo (ed.): Wissenschafissprache. München: Fink, 85-110. Pomerantz, A. 1984 "Giving a Source or Basis: the Practice in Conversation of Telling 'how I know'", Journal of Pragmatics 8:5/6,607-625. Popescu, R. S. 1985 "On the Meanings of the Word tarina", Limba Romana 34:3, 232-238. Powell, M. J. 1985 "Purposive Vagueness: An Evaluative Dimension of Vague Quantifying Expressions", Journal of Linguistics 21:1, 31-50. Preisler, B. 1986 Linguistic Sex Roks in Conversation: Social Variation in the Expression ofTentativeness in English. Berlin: Mouton de Gruyter. Prince, E. F. 1976 "The Syntax and Semantics of Neg-Raising, with Evidence from French", Language 52:2, 404-426. Prince, E.; Bosk, C. and Frader, J. 1982 "On Hedging in Physician-Physician Discourse", in di Pietro, J. (ed.): Linguistics and the Professions. Norwood/New Jersey: Ablex, 83-97. Rasmussen, J. L. and Moely, B. E. 1986 "Impression Formation as a Function of the Sex Role. Appropriateness of Linguistic Behavior", Sex Roks 14:3^t, 149-161. Rastier, F. 1983 "The Cohesion of Foreign Utterances", Semantikos. 7:2, 23-61. Redeker, G. 1990 "Ideational and Pragmatic Markers of Discourse Structure", Journal of Pragmatics 14:3, 367-381.

268

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Reis, M. 1984

"Pour une grammaire a trois valeurs: Zones d'inde'cidabiliti", DRLAV. Revue de LinguistiqueSl, 1-17.

Rieger, B.

1989

Unscharfe Semantik. Die empirische Analyse, quantitative Beschreibung, formale Repräsentation und prozedurak Modellierung vager Wortbedeutungen in Texten. Frankfurt am Main et al.: Peter Lang. Roberts,). R. 1990 "Modality in Amele and other Papuan Languages", Journal of Linguistics 26, 363-401. Rolf, E. 1984 "Indikatoren zur Quantifizierung von Äußerungen, mit denen sprachliche Informationshandlungen realisiert werden können", Krenn, H.; Niemeyer, J. and Eberhardt, U. (eds.): Sprache und Gesellschaß. Tübingen: Niemeyer, 53-62. Rolf, E. 1994

Rösch, E. 1973 Rösch, E. 1978

Sagen und Meinen. Paul Grices Theorie der Konversations-Implikaturen. Opladen: Westdeutscher Verlag.

"Natural Categories", Cognitive Psychology 4, 328-350.

"Principles of Categorization", in Rosch, E. and Lloyd, B. (eds.): Cognition and Categorization. Hillsday/New Jersey: Erlbaum Ass, 27-48. Rounds, P. L. 1982

Hedging in Written Academic Discourse: Precision and Flexibility. Ann Arbor:

University of Michigan. Rudolph, E. 1986 "Partikeln und Text-Konnexität im Deutschen", in Heydrich, W. and Petöfi, J. S. (eds.): Aspekte der Konnexität und Kohärenz von Texten. Hamburg: Buske, 73-90. Salager-Meyer, F. 1990 "Discoursal Movements in Medical English Abstracts and Their Linguistic Exponents: A Genre Analysis Study", Interface 4:2, 107-124. Salager-Meyer, F. 1992 "A Text-Type and Move Analysis Study of Verb Tense and Modality Distribution in Medical English Abstracts", English for Specific Purposes 11:2, 93-113. Salager-Meyer, F. 1995 "I think that Perhaps You Should: A Study of Hedges in Written Scientific Discourse", in Miller, T. (ed.): Functional Approaches to Written Texts: Classroom Applications. Volume I (The Journal of TESOL France, Volume 2, No. 2, 1995), 127-143. Salager-Meyer, F. and Salas, J. C. 1991 "A Genre-Based and Text-Type Analysis of Hedging in Written Medical English Discourse (1980-1990)", Interface 6:1, 33-54. Sanders, W. 1992 Sprachkritikastereien und was der "Fachler" dazu sagt. Darmstadt: Wissenschaftliche Buchgesellschaft. Schiffrin, D. 1991 Discourse Markers. Cambridge: Cambridge University Press. Schneider, E.W. 1988

Variabilität, Niemeyer.

Polysemie

und

Unscharfe

in

derWortbedeutung.

Tübingen:

Schröder, H. 1989 "Gesellschaftswissenschaftliche Fachtexte und Interkulturelle Fachkommunikation: Probleme für den Fremdsprachenlerner und Übersetzer", Neuphilologische Mitteilungen 89:4, 559-566.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

269

Striven, M. 1976 Reasoning. New York: McGraw Hill. Semiloff, H. 1977 "Language and Woman's Place", Language Sciences 44, 34-38. Sherzer.J. 1974 "On Linguistics and Other Disciplines: A Perspective from Anthropology", Georgetown University Round Table on Languages and Linguistics, 131-142. Simpson, P. 1990 "Modality in Literary-Critical Discourse", in Nash, W. (ed.): The Writing ScholarStudies in Academic Discourse. Newbury Park: Sage, 63-94. SkeltonJ. 1988 "The Care and Maintenance of Hedges", English Language Teaching Journal 42:1, 37^3. SkeltonJ. "How to Tell the Truth in The British Medical Journal. Patterns of Judgement in the 19th and 20th Centuries", in the present volume. Sökeland, W. 1981 "Erklärungen und Argumentationen in wissenschaftlicher Kommunikation", in Bungarten, T. (ed.): Wissenschaßssprache. München: Fink, 261-293. Sörensen, R. A. 1990 "Process Vagueness", Linguistics and Philosophy 13, 589-618. Spies, M. 1993 Unsicheres Wissen. Heidelberg: Spektrum. Staley, C. M. 1982 "Sex-Related Differences in the Style of Children's Language", Journal of Psycholinguistic Research 11:2, 141-158. Stubbs, M. 1986 " Matter of Prolonged Field Work': Notes toward a Modal Grammar of English", Applied Linguistics 7:1, 1-25. Takeuchi, H. 1991 "Measuring the Meaning of Linguistic Hedges by Fuzzy Rating Scale Method ", Mathematical Linguistics 17:8, 365-376. Taylor, J. R. 1994 "The Two-Level Approach to Meaning", Linguistische Berichte 149, 3-26. Troemel-Ploetz, S. 1994 "'Let me Put it this Way, John': Conversational Strategies of Women in Leadership positions", Journal of 'Pragmatics 22,199-209. Tsohatzidis, S. L. 1987 "An Episode from the Emergence of the Gricean Paradigm", Word$S:2, 91-97. Underbill, R. 1988 "'Like' is, like, focus", American Speed 63:3,234-246. Vande Kopple, W. and Crismore, A. 1990 "Readers reaction to Hedges in a Science Textbook", Linguistics and Education 2:4, 303-322. Venier, F. 1986 "Gli awerbi modali", Lingua e Stile 21,459-483. Ventola, E. 1992 Journal of Applied Linguistics 2:2,191-220. Vicher, A. and Sankoff, D. 1989 "The Emergent Syntax of Pre-Sentential turn Openings", Journal of Pragmatics 13:1, 81-97. Vinson, L. and Johnson, G. 1989 "The Use of Written Transcripts in Powerful and Powerless Language Research", Communication Reports 2:1, 16-21. Wachtel, T. 1980 "Pragmatic Approximations", Journal of Pragmatics 4:3,201-211.

270

V. Hedging: The Concept, its Origins and a Bibliographical Guide

Wales, K. 1989 Webber, P. 1994

"Hedging; hedge", in Wales, L. (1989): A Dictionary of Stylislics. New York: Longman, 515-516.

The Functions of Questions in Different Medical Journal Genres", English for Specific Purposes 13:3,257-268. Weinreich, U. 1966 "On the Semantic Structure of English", in Greenberg, J.H. (ed.): Universals of Language. Second Ed. Cambridge/Mass.: MIT Press, 142-217. Weinrich, H. 1993 Textgrammatik der deutschen Sprache. Mannheim: Bibliographisches Institut/ Brockhaus. Werlen, I. 1983 "Vermeidungsritual und Höflichkeit: Zu einigen Formen konventionalisierter indirekter Sprechakte im Deutschen", Deutsche Sprache 11, 193-218. Werlich, E. 1983 A Text Grammar of English. Heidelberg: Quelle und Meyer. Westney, P. 1986 "How to be More or Less Certain in English: Scalarity in epistemic Modality", International Review of Applied Linguistics 24:4,311-320. Weydt, H.; Harden, T; Hentschel, E. and Rosier, D. 1983 Kleine deutsche Partikellehre. Ein Lehr-und Übungsbuch ßir Deutsch als Fremdsprache. München: Klett. WickboldtJ. 1985 "An Examination of a Corpus of ate'+Verbal Constructions and specially the 'Hate to'-Hedge", Linguistics in Belgium 7,111-131. Wierzbicka, A. 1986 "Precision in Vagueness: The Semantics of English 'Approximatives'", Journal of Pragmatics 10, 597-614. Wierzbicka, A. 1991 Cross Cultural Pragmatics. The Semantics of Human Interaction. Berlin: Mouton de Gruyter. Wiggli, H. 1982 "Second-Language learning through Communication Characteristics of a Didactic Concept", Bulletin CILA 35, 42-59. Wilss, W. "Hedges in Expert-Language Reviews", in the present volume. Wimmer, R. 1987 "Der fremde Stil. Zur kulturellen Vielfalt wissenschaftlicher Textproduktion als Problem interkultureller Germanistik", in Wierlacher, A. (ed.): Perspektiven und Verfahren interkultureller Germanistik. München: iudicium, 81-98. Wolski, W. 1980 Schlechtbestimmtheit und Vagheit. Tendenzen und Perspektiven. Tübingen: Niemeyer. Wright, John W. II and Hosman, L. A. 1983 "Language Style and Sex Bias in the Courtroom: The Effects of Male and Female Use of Hedges and Intensifies on Impression Information", The Southern Speech Communication Journal 48:2, 137-152. Zadeh, L. 1965 "Fuzzy Sets", Information and Control8,338-353. Zadeh, L. A. 1972 "Fuzzy-Set - Theoretic Interpretation of Linguistic Hedges", Journal of Cybernetics 2:3, 4-34. Zadeh, L. A. 1975 "The Concept of a Linguistic Variable and its Application to Approximate Reasoning- , Information Sciences 8:3, 199-249.

H. Schröder and D. Zimmer, Hedging Research in Pragmatics

271

Zierer, E. 1987 "La palabra alemana 'gewisz' as a hedge", Lenguaje y ciencias 27:3-4, 63-72. ZuckJ. G. and Zuck, L V. 1987 "Hedging in Newswriting", in Comu, A.; Vanpaijs, ].; Delahaye, M. and Baten, L. (eds.): Selected Papers from the Fifth European Symposium on LSP. Leuven: OUP, 172-180.

Name Index Abraham, W. 252-255 Adams, A. 256 Aijmer, K. 252, 256 Akhbar, K. 156,157 Alexander, R. 203 Amdt, H./Janney, R. 188, 193,201,203 Atkinson, D. 47, 154 Ayto,J./Walford, G. 252 Bardovi-Harlig, K. 204 Barton, E. 155-156, 158 Barnet, S./Stubbs, M. 86 Barwise, J./Peny, J./Montague, W. 238 Bateman,J. 256 Bazerman, C. 47 Bell, R. 256 Bell, R./ Zahn, C./Hopper, R. 85 Berk-Seligson, S. 253 Berlyne, D. 107 Biber, D. 156, 253-254 Blum-Kulka, S.. 188, 190, 203-204, 240, 253-255 Boehm, C. 209 Boyle, R 153-154 BradacJ. 255-256 Brainbridge, O. 252 Brooke, M./Ng, S. 256 Brown, P. 239 Brown, P./Levinson, S. 4, 7-8, 24-25, 67, 73, 84, 116, 169, 190-192, 251-252, 255-256 Brown, G./Yule, G. 189 Bruce, B. 86, 106 Bublitz, W. 198 Burkhardt, A. 221 Busch-Lauer, I. 252, 255-256 Buscha, J./Heinrich, G./Zoch, I. 252 Bußmann, H. 3,250,255 Butler, C. 10, 155, 157-158, 163, 165, 252, 256 Caffi, C./Janney, R. 255 Cappella,]. 193 Chafe, W. 4, 7, 84, 155, 158, 168, 193, 199 Cameron, D. 256 Channell, J. 115, 122, 131,241

Carrell, P. 255 Channell,]. 255-256 Chen, R. 252, 254-255 ChoulJ. 254 Clyne, M. 72, 110, 213, 215, 219, 240, 243, 252-253, 266 Coates, J./Cameron, D. 105, Coates, J. 156-157, 165, 195,215, 255-256 Conan Doyle, Sir Arthur 49 Crismore, A. 67, 87, 110, 155-159, 165 Crismore, A./Farnsworth, R. 110 Crismore, A./Markkanen, R./Steffensen, M. 9,252, 254-256 Crismore, A./Vande Kopple, W. 252, 255, 256 Cronkite, W. 152 Damerau, R 252,255 Danet, B. 255 Darian, S. 6, 10,252 Darwin, C. 88, 110 De Beaugrande, R. 86 Debus, E 209 Dillon, G. 86,108 Di Pietro, R. 256 Dittmar, N./Terborg, H. 253, 256 Domaschnew, A. 210 Dougherty, B. 86 Drösser, C. 255 Drosdowski, G. 210 Dubois, B. 10, 47, 241, 252, 256 Duszak, A. 253, 256 Ebel, H. 151 Edwards, C. 252 ErbenJ. 253-255 Erickson, B./Lind, E./Johnson, B./CXBarr, W. 256 Fairclough, N. 129 Falkenberg, G. 254 Famsworth, R. 155-159, 165 Fiedler, S. 255-256 Fillmore, C. 238 Finegan, E. 155 Fishman,;. 211

274

Name Index

Fiske, S./Linville, P. 105 Fix, U. 210 Fleischer, W. 210,212 Flowerdew, J. 252,254-255 Flynn, E./Schweickart, P. 105 Fraas, C. 209,212,216-217,224 Fräser, B. 4,24-25,31,35, 67,240,251-252, 254-256 Geist, U. 11 Gilbert, G. 49 Giles, H. 217 Giles, H./Bradac,J. 255 Gippert,J.250,255 Givon, T. 254 Glass, A./Holyoak, K. 255 Goffinan, E. 67,192 Good, C. 215 Gould, S. 157 Grabe, W. 151-152,158-159 Grabe, W./Kaplan, R. 252,256 Green, M. 256 Grice, P. 26 Gruber, H. 255 Grundy, P. 252 Gu, Y. 252 Günther, U. 253,255 Gumperz,J. 211 Haarmann, H. 252-253,255-256 Hagge,J/Kostelnick,C. 255-256 Hariston, M. 86 Hakulinen, A. 173-174 Halliday, M. 67,84, 108, 119,123 Hanania, E. 156-157 Hamish, R. 154 Heinämäki, O. 252 Held, G. 67 Hellmann, M. 208-211,213,216 HenkelJ. 252,254-255 Herbert, R. 255-256 Herrmann, L. 252 Hinrichs, U. 253-255 Hiukka, K. 200 Hoffmann, R. 214 Holly, W. 214,221 Holmes, J. 9,84,86,105,188,192,198,215, 242, 252,254-256 Horner, W. 105 Hosman, L. 256 Hosman, L./Wright, J. 85-86,110,252,256 House, J./Kasper, G. 4, 7-8, 237, 240, 252-256 House,J./Vollnier,H. 253-254 Hübler, A. 5-6, 8, 21, 23, 64, 84, 107, 169, 192, 241, 251-252

Huddleston, R. 156-158 Hundsnurscher, F. 224 Hyland, K. 256 Hymes, D. 211

Ide, S. 116 Inoue, K. 253 lonescu-Ruxandoiu, L. 252,255 Jahr, S. 255 James, A. 195,242,252,254, 256 Janney, R/Arndt, H. 255-256 Janicki, K 204 Jespersen, O. 65 Johnson, C./Vmson, L. 252 Johnson, D. 255 Kärkkäinen, E. 13,200 Kalimeyer, W. 219 Kaplan, R. 152 Kasper, G. 13, 188, 200 Kay, P. 238, 255 Kayser, H. 255 Keseling, G. 254 Kiefer, F. 255 Kinder, J. 253,256 Kitagawa, C./Lehrer, A. 173 Koch, W. 253-254 Köhler, C. 253, 256 Kolde, G. 256 Kong, A. 47 Korhonen, R./Kusch, M. 169 Kramarae, C. 256 Kress/Hodge 86 Kretzenbacher, H. 253,255-256 Kreutz, H. 213-214,253,255 Kühn 214 Kusunose.J. 244 Labov, W. 254-256 Lachowicz, D. 252,255-256 Laerman, K. 11 Laib, N. 86 Lakoff, G. 4-7, 9, 43, 50, 64, 85, 191, 235, 240,251-252, 255 Lakoff, R. 8,215,242,255-256 Lannon.J. 85-86 Lauer, I. 255 Lausberg, Heinrich Lautamatti, L. 67 Lavandera, B. 254-255 Lee, D. 254-255 Leech, G. 156-157, 165,255 Lehrer, A. 252 Lehtonen.J./Sajavaara, K. 195 Leontyev, A. 72

Name Index Lerchner, G. 210,219 Levinson, S. 239 Liebert, W. 253, 256 Lindeberg, A. 255-256 Lindgren, J. 85 Linke, A. 213-214 Lipson, M. 107 Littlewood, W. 203 Loewenberg, I. 252,255-256 Lovik, T. 253 Lüger, H. 254-256 Luukka, M./Markkanen, R. 252,256 Luke, C./de Castell, S./Luke, A. 104 Lyons, J. 6,65,123 Mahaney, W. 85 Maier, P. 255 Markkanen, R. 254,256 Markkanen, R./Schröder, H. 5-7,10,21,23, 36, 87, 103, 109-110, 115-117, 125, 132, 252-254,256 Markkanen, H./Steffensen, M./Crismore, A. 67,254 MartinJ. 15, 105 Marui, I/Reinelt, R. 253,256 Maschler, Y. 253-254 Master, P. 256 Matsumoto, Y. 253-255 Mauranen, A. 256 McCawleyJ. 255 McLain, R. 255 McMullen, L./Pasloski, D. 256 McNeill, D/Freiberger, P. 255 Meyer, H. 37, 67, 252,254, 256 Milward, C. 86 Morley.J. 255 Moser, H. 210 Mühlen, U. 214 Müller, K. 255 Mulkay, M. 49,255 Myers, G. 10, 47, 116, 122, 129, 131, 156, 170,255-256 Namsaraev, V. 253 Nesbett, R./Ross, L. 104 Neubert, A. 255-256 Newcombe, N./Arnhoff, D. 85-86 Nichols, J. 155, 158 Nicoloff, F. 254 Nikula, T. 189, 252,254 Nyyssönen, H. 254, 256 Ochs,E. 152 O'Connor, M. 252 Öhlschläger, G. 253, 255 ÖstmanJ. 191,252,254-255 Ohta, A. 253

275

Olshtain, E. 240 Olson, D. 103-104, 107-108 Opitz, K. 64 OschÜes, W. 210 Packer, N./TimpaneJ. 86 Palmer, F. 6, 66 Panther, K. 254-256 Payne, L. 86 Perkins, M. 65, 255 Perttunen.J. 255-256 Perry, W. 90 Pindi, M./Bloor, T. 252, 256 Pinkai, M. 241-242, 255 Pomerantz, A. 254 Popescu, R. 253 Powell, M. 252,255 Preisler, B. 7-9, 86, 252, 256 Prince, E. 47, 84, 154, 191, 240,252,255 Prince, E./Frader, J./Bosk, C. 5- 6, 11, 84, 251,254 Prost 161 Püschel 214 Purves, A. 15 Quirk, RVGreenbaum, S./Leech,G./ SvartvikJ. 84 Quirk, R. 155-156,242 Rasmussen,J./Moely, B. 252,256 Rastier, F. 252 Rather, D. 151-152 Rawlins, J. 86 Redeker, G. 254 Rieger, B. 255-256 Riley, P. 189 Roberts,]. 252 Rolf, E. 254 Rösch, E. 236-237,251 Rounds, P. 240, 251, 254,256 Rudolph, E. 241,254-255 Ruggiero, V. 86 Ruthrof, H. 86 Salager-Meyer, F. 47, 51, 72,252,254,256 Sanders, W. 253,255-256 Sandig, B. 219 Schiffrin, D. 191, 254 Schmidt, V. 210 Schmitt, R. 219 Schneider, E. 255 Schneider, H. 253, 256 Schrubben, C. 209 Schommer, M. 90 Scriven, M. 255

SearleJ.

24-26,30,34

276

Name Index

Selinker, L. 188 Semiloff, H. 252,256 Shore, S. 174 Siewierska, A. 168, 173 Simpson, P. 11, 66, 115-116, 119, 130-131, 215,218,256 Singer, H. 86 SkeltonJ. 47,191,242,256 Sökeland, W. 254 Sörensen, R. 255 Spies, M. 255 Spillner, B. 9, 10 Staley, C. 255 Steyer 209 Stolze, R. 9 Strunk, W./White, E. 86 Stubbs, M. 6,254-255 SwalesJ. 49 Tabakowska, E. 117 Taylor, J. 255 Teichmann, C. 219 Thomas, J. 13, 189,203-204 Tompson, S. 193 Tyler, S./Vbss,J. 105 Underbill, R.

252

Valle, E. 116 Vande Kopple, W. 4, 67, 155 Van Dijk, T. 87, 105 Venier, F. 253 Ventola, E. 256 Ventola, E./Mauranen, A. 13, 117, 131

Vicher, A./Sankoff, D. 252, 254 Villaume, W. 193 Vinson, L./Johnson, G. 252,256 Von Hahn, W. 7,253,255-256 Von Polenz, P. 210,216,219, 255-256 Wachtel, T. 255 Wales, K. 236,250 Watzlawick, P./Beavin, J./Jackson, D. Weinreich, U. 235,250,253-255 Weinrich, H. 253,255 Werlen, I. 253-255 Werlich, E. 255 Weydt, H. 242,253-255 WickboldtJ. 252 Wierzbicka, A. 252-253,255-256 Wiggli, H. 252 Williams,}. 67 Wilss, W. 252, 256 Wimmer, R. 253, 255-256 Wolski, W. 255 Woodruff, D. 157 Woodman, L./Adler, T. 86 Wright, H./Hosman, A. 256 Ylönen, S. 209, 219 Yule, G. 189 Zabalbeascoa, P. 132 Zadeh, L. 236,237250-253,255 Zierer, E. 253 Zinsser, W. 86 Zuck,;./Zuck, L. 7,115,255

103

Subject Index

ability 25,33-38 "Abtönungspartikeln" 242 academic communication 256 academic texts 159,244 academic writing 64 action 222 active involvement of readers 86, 108 adjective(s) 243-244 admissions 32 adverb(s) 71,243-244 anticipating 35 approximators 5-6,240,242, 251 argumentation 10 assertiveness 8,13 assignment 244 attitude 90-91,241,244 attitude change 87, 105, 109 attitudes 83, 85-86, 104 attitudinal markers 165 avoidance 239 awareness raising 203-204 behavioral grammar 136 behavioreme 144 "big theories" 110 bipartite division 240 booster 9 British medical Journal 42 categories) 236-238 - basic-level 237 - membership 236-237 - structures 238 causal theory of perception 30 certainty 36-37, 43^t5, 50-53, 85, 90, 93 chronological/logical ordering 49-51, 54-55 classification 235,237,242-244 classifying 236 clustering of hedges 121-123 cognitive 236,239 - aspect 236 - model 236 - semantics 236 - science 239 - theories 236

comments) 242,244 - on truth-judgement 242 - on value-judgement 242 - subjective 244 commentative potential 45 coming-to-know 37-38 commitment 25, 27-29, 34, 38, 43, 45, 53-55, 57, 238-239 communicative activities 222, 225 communicative content 221 communicative context 6 communicative interaction 135 communicative needs 135 communicative patterns 222,225 communicative situation 211,213 communicative skills 141 communicative strategy 3, 5 communicative strategy of hedging 168 communicative uncertainty 210, 214, 216, 218,218, 222, 225 compromisers 242 conclusion 32 conditional^) 238-239,243 connotations 217 context 85, 116, 119-122, 125, 129, 132, 235-236,238-239,241-244 contrastive 240,242 - analyses 242 - approaches 240 conventions 39-40 convention analysis 254 creative ambiguity 53-54, 57 cross-cultural 240,243 - analyses 240 - communication 243 culture 145 deductive/inductive methodology 50, 52, 57 deference 8 definition(s) 235-236,240,244 degree of reliability 84 demarcation 241-242 depersonalisation 68, 71, 73, 75-77 designation 242 desire 24-25,31

278

Subject Index

detachment 168-170 detensifier 4,84 dictionary (ies) 236 didactic 242 diffusere 240,251 diminishers 242 discourse analysis 235-23 6, 23 9,244,254 discourse markers 254 discourse of unification 215 doubt 43-46, 49, 51 downgrade« 240 downtoners 84,240-242 downtoning 219,221,241

gender differences 102, 105 generative theory 134-135 generic and vague reference 173-174, 181-183 generic verb form 173, 175-176, 180-182 genre 83, 14 German communication situation 211 German language 252 Germany political unification 208 gestalt 137-138,140 goals of science 39-40 grade particles 241 grading 237,241

economics texts 235 - writing on 235 "eigentlich" 236 emphasis 32, 157 emphatics 87, 155,158, 160-165 empirical analyses 240 English language 252 English text 177-179, 181-183,186 epistemic 117, 119-122, 128, 131 epistemic modality 6-7, 13, 137, 190 epistemic uncertainty 120-121, 126 ethos 9 evaluatives 158 evidentials 164 evidential judgement 53-55, 57 evidential marking 158 evidentiality 7 excuses 23 explanations 22, 25,28-29, 31, 34-35, 40 explicit performative 24-25, 32 explicit personal reference 170-171, 173-174, 180, 186 explicitly 170,173, 178-180, 182-183, 186

"Hecken" 3,241-242 - "depräzisierende" 241 - "linguistische" 242 - "präzisierende" 241 - "quantifizierende" 241 - "sprachliche" 242 "Heckenausdruck" 3,242 hedge(s) 43-46, 51, 53, 55, 83, 85,105,134, 137, 141, 144-146, 188-194, 196-200, 202, 204-205, 235-236, 238, 240-241 - concept 64-67, 154,240,253,240 - definition 64,251 - discourse-oriented definition 253 - double or triple 72, 75 - function 65 - linguistic 238 - "list of" 236 - provocative 69 - so-called 241 - subdivide 240 hedged performatives 4, 24-26, 28, 239, 243 hedges in courtrooms 110 hedging 138-142, 144-146, 235-236, 239, 240-241 - application of 255-256 - concept of 235-236 - degree of 73,75,77 - explicit 71,74 - function 239,241 - logics of 255 - non-verbal 162 - provocative 70 - research 155-158,249-256 - rhetorics of 255 - strategies 68, 130,240 hypothetical speech act 27

face 8, 65-67, 70, 72, 75-76, 169-170, 169-170, 192-193, 195, 201-202 - threatening acts/(FTA) 67 face-saving 23, 39-40,239 face-threat 25,31 face-work 32,34 falsifiability 39 Finnish 189,194-196,200-203 Finnish text 181-184, 186 formality 190, 195, 197 frame semantics 238 fuzziness 237 fuzzy 235,236 - set theory 236 fuzzy concepts 137 fuzzy logic 136 gender

8-9,84

ICMs (idealized cognitive models) illocution 243 illocutionary act/force 4, 6, 8 illocutionary modifiers 241

238

Subject Index implicitness (vs. explicitness) 192,196, 199, 201-202 impersonal construction 239,243 impersonal expressions 6 impersonal voice 106 impersonalization 168-169,173-175, 183 implicit personal reference 170, 182, 186 imposition on the hearer/reader 23-24, 29-30, 32-33, 35, 38 impreciseness 7, 10 imprecision 241-242 inclination 24-25,38 indetermination 5-6, 68, 71, 73, 76-77 indirect construction 243 indirectness 239,243 instruments 30 intensifier 4 interactional strategy 12 intercultural communication 256 interdisciplinary differences 129-131 interlanguage 13 interlanguage pragmatics 190,192,204 interpersonal 168, 180, 186 - metadiscourse 67 - functions of language 87 involvement 193, 195,199-200 knowing what for/what/why 136 knowledge and skills 141, 145 language 241-242 - for specific purposes (LSP) 241 - spoken 242 - women/men 242 lexical 235-236,238,240, 242-243 - devices 240 - items 238,240,242 - research 236 - units 235-236 - use 238 lexico-semantic aspects 236 like 242 linguistic devices 239, 244 linguistic theory 135 linguistic usage 134,135 "loosely speaking" 238 major/minor truth judgements 45-46, 48-52, 55 maxim of quality 26, 38 metadiscourse 4, 9, 64, 67, 110,156 metalinguistic 235 - device 235 - operators 235 methods 30 methodological problems 242

279

minimizers 240,242 mitigation 4, 15,239,242,254 - of certainty 242 - of responsibility 242 modal 157-158, 165, 239,241-243 - auxiliaries 239 - particles 241-243 - verbs 156,243 modality 6-7, 13, 108 - communicative 67 - deontic 65 - epistic 65,252 - marker 108, 156 - marker 108 modalization 11 model-theory semantics 238 modification 64,68 modification of utterances 239 modifier(s) 236,242,244 - of illocutionary force 242 - prototypical 244 - semantic 236 modifying devices 219, 243 native language influence 200,202 nature of knowledge 128 necessity 31-33,38 negatability 35 negation 240,242 neglected wordclasses 243 neustic 5-6 nominalization 174-175 non-native English text 181-183 non-verbal means 243 norms of academic discourse 186 note 26,29, 31 Occam's razor 32 offering 27,29,34 origin(s) 235-236,244 paradigmatic statements 145 parenthetic constructions 243 passive(s) 46, 174-176, 179-183, 186 passive/passive infinitive 243 performatives 24-28,34-35 peripheral-idea sentence 89 personal pronouns 243 personal voice 88, 106, 108 pessimism 25 plausibility shields 84 politeness 4, 7-10, 15, 64-65, 67, 75, 116-119, 122, 129, 169, 188, 192-193, 200-201, 218,239-240,242-243,251, 255 - markers 240 - strategies 154, 239,242-243

280

Subjct Index

possibility 24-25, 34, 36-37 postmodern science 11,12 powerless language 9 powerless speech 85 pragma 9 pragmatic aspect of communication 239, 243 pragmatic competence 13 pragmatic presuppositions 218 pragmatics 239 preparatory condition 33 probability 25, 33, 36-37 proposition(s) 43,45-46,53,57, 84,242 prosodic means 243 prototype(s) 237,251 - effects 237 - theory 237 prototypical members 237 reference works 235-236 "regular" 238 referring 170, 174, 176-178, 180-182, 186 repetitions 32 research 22,28, 30, 35-36, 38-39,236,242 retention 91-92 review/review writing 134, 136-140, 142, 145 reviewer/reviewee 134, 136, 138-142, 145, 146 rhetoric 134, 144 rhetorical context 111 rhetorical effect 103 rhetorical norms 15 rhetorical patterns 144 Russian language 69-77 schemata 105 scientific article 153-154 scientific text 169, 183, 186 scientific writing 6, 9-11, 15, 152-153, 157 self-criticism 25 shields 5-6, 84,240,251 should 21-23,29, 33, 36-37 significance 46, 49, 53, 55 social ritual 129 sociological texts 69-77 solidarity 9 so-called 241 "sort of" 237-238, 242 speaker's commitment 4-5 speaker/writer 169-170, 176, 181, 186 speaking facts 27 speculative judgement 53-57 speech act 223, 254 spoken language 151-152 standard configuration 140-141.

strategic hedging 128-129 strategies 239-240,242-243 strength 21, 23,25-26,28, 31, 36, 38-39 strengthening and weakening 23,39 subcategory 235,241 subdivision 242 subjective attitude 243-244 - comment 244 - evaluation 243 subjectivisation 68, 71, 73-75, 77 subjectivity of decision 46 subjunctive 243 suggest 22, 24,28-29,38 suspect 22,26-28, 38 syntactic devices 239-242 - features 240 - structures 242 systematic approach 240 tacit knowledge 140 tentativeness 7-8 text content/function(s)/pragmatics 134, 135-138, 141 textual context 119 textualisation 9, 11, 15 theory 134-135, 139 transition marker 161 translating hedges 116 translation 12, 14, 116-117, 125, 131-132, 139 true 237 truth 237,238, 242, 244 - conditional semantics 238 - judgement 46-47,50-55,242,244 - value 4-5,7-8,43,45-46,244 "typisch" 236 uncertainty 84, 104, 115, 119-122, 126-128 understatements) 5,239,241,243,245,251 understaters 240 unspecified human agent 173, 175-176 vague 236,244 vagueness 7, 44-45, 115, 121-123, 236-237, 239-241, 255 - semantic 236 value judgement 46-47, 49-52, 54 volition 25 voluntary marker of imprecision 242 weakeners 84 weak performatives 26, 28 we-awareness (complicity) 140 writer intentions 117, 127 writing 151, 152

RESEARCH IN TEXT THEORY UNTERSUCHUNGEN ZUR TEXTTHEORIE

Intertextuality Edited by Heinrich F. Plett 1991. VIII, 268 pages. Bound. ISBN 3 11 011637 5 (Volume 15)

Subject-oriented Texts Languages for Special Purposes and Text Theory Edited by Hartmut Schröder 1991. VIII, 322 pages. Bound. ISBN 3 11 012568 4 (Volume 16) ANTONIO GARCIA-BERRIO

A Theory of the Literary Text 1992. X, 544 pages. With 2 figures. Bound. ISBN 3 11 012809 8 (Volume 17) ROGER G. VAN DE VELDE

Text and Thinking On Some Roles of Thinking in Text Interpretation 1992. XVI, 328 pages. With 9 figures. Bound. ISBN 3 11 013250 8 (Volume 18)

Semiotics and Linguistics in Alice's Worlds Edited by Rachel Fordyce and Carla Marello 1994. VIII, 277 pages. With 34 figures. Bound. ISBN 3 11 013894 8 (Volume 19)

W G

Walter de Gtuyter · Berlin · New York

RESEARCH IN TEXT THEORY UNTERSUCHUNGEN ZUR TEXTTHEORIE

Approaches to Poetry Some Aspects of Textuality, Intertextuality and Intermediality Edited by Janos S. Petöß and Terry Olivi 1994. VIII, 300 pages. With 6 coloured plates. Bound. ISBN 3 11 013893 X (Volume 20)

Aspects of Oral Communication Edited by Ufa M, Quasthoff 1995. VI, 493 pages. With 30 illustrations. Bound. ISBN 3 11 014465 4 (Volume 21)

Focus and Coherence in Discourse Processing Edited by Gert Rickheit and Christopher Habel 1995. XII, 300 pages. With 49 figures and 10 tables. Bound. ISBN 3 11 014466 2 (Volume 22)

ELISABETH RUDOLPH

Contrast Adversative and Concessive Relations and Their Expressions in English, German, Spanish, Portuguese on Sentence and Text Level 1996. XX, 544 pages. Bound. ISBN 3 11 014955 9 (Volume 23)

W DE G

Walter de Gruyter · Berlin · New York