Getting the full picture:loadout
a reflection on the work of M.A.K.Halliday
Diana Kilpert*
创始人英文
Department of English Language and Linguistics,Rhodes University,PO Box 94,Grahamstown 6140,拖车费英文
South Africa
Abstract
Halliday and Matthiesn (Halliday,M.A.K.,&Matthiesn,C.M.I.M.,1999.Construing Experience through Meaning:A Language-Bad Approach to Cognition.Casll,London &New York)is a significant work on language and linguistic theory.It deals with the construal of human experience as a mantic system,from the perspective of systemic-functional lin-guistics (sfl),the theory that has grown out of Halliday’s work over the past four decades.In this review article I look at the variety of linguistic interests and applications that the book address.The review is followed by a general discussion of the way the theory expands and enhances our view of language and how it reconnects language and society in the grammar.I examine some rervations about the terminology,and give some specific detail of sfl analy-sis.The discussion is t in the context of a wider debate,initiated by Roy Harris,about the scope of linguistic inquiry,its methods and sphere of influence.I outline some critical con-cerns voiced by Harris and consider some points of congruence between what Harris demands of linguistics and what Halliday has achieved,with the aim of eing how sfl might satisfy s
ome of Harris’s criteria for integrational linguistics.My overall concern is to draw attention to aspects of Halliday’s linguistic theorizing that enable us to achieve a fuller picture of lan-guage than has hitherto been possible.#2002Elvier Science Ltd.All rights rerved.Keywords:Linguistic theory;Systemic functional linguistics;Integrational linguistics;Semantics;Cogni-tive science
In talking of the power of language,I do not mean only its power as exploited in political contexts,but what it achieves at every institutional and personal *Corresponding author.
E-mail address:d.kilpert@ru.ac.za (D.Kilpert).
160 D.Kilpert/Language Sciences25(2003)159–209
level in human lives.What Ifind surprising,in this light,is the discrepancy between the potency of language and the trivial picture that is so often pre-nted of it-not least by some of the folk who most strongly caution against its powerful effects.(Halliday,1997:25)
emit
1.Introduction
Michael Halliday’s linguistics is notable for its extravagance.It has enlarged our current picture of language by foregrounding motifs of opening up,expanding,and eing things from multiple perspecti
ves and,by thus bringing in elements that a parsimonious orthodoxy sidelines,it has broadened the scope of linguistic inquiry and enabled the discipline to extend its sphere of influence and to speak to the needs of the consumer.
Before discussing some of the ways Halliday has achieved this,I will explain the angle this review article takes.Progress in linguistics has been regarded during the past four decades as very much a process of trouncing your opponents and tting yourlf up in their place;the assumption being that you cannot prove yourlf right withoutfirst proving someone el wrong.This adversarial approach can be counter-productive.As Halliday and Fawcett point out,‘it leaves little room for the cond scholar to take a positive line—to suggest,perhaps,that s/hefinds much(or at least some!)common ground with thefirst approach,and that s/he might not have thought of the ideas being prented without the stimulus of the first scholar’s work’(1987:2).If the orthodoxy favours dividing rather than consolidating,then consolidators are liable tofind their work sidelined as ‘‘unscientific’’,or even‘‘anti-intellectual’’.1I want to tip the balance a little towards celebrating rather than carping,and focus on what we can u,rather than on what we should reject.This article,then,is a reflection on,not a disction of, M.A.K.Halliday’s work and the theory that has grown out of it,systemic-functional linguistics(sfl).
Halliday himlf makes remarkably little overt criticism of mainstream linguistics. He rejects‘trial by intellectual combat’,suggesting that‘the better cour is surely to make a straightforward statement of one’s ca—which may often ufully include an honest attempt to summarize the pros and cons of other possible approaches, and perhaps a discussion of the new approach’s applicability for some specific pur-po—and then simply to let other scholars consider the alternatives offered,and make up their own minds as to which is more helpful’(Halliday and Fawcett, 1987:2).This includes making an effort tofind something to build on even in the formal model which is the antithesis of everything he believes in:
1In the words of Noam Chomsky,in personal communication to R.A.Harris:‘I was told that my work would arou much less antagonism if I didn’t always couple my prentation of transformational grammar with a sweeping attack on empiricists and behaviorists and on other linguists.A lot of kind older people who were well dispod towards me told me I should stick to my own work and leave other people alone.But that struck me as an anti-intellectual counl’(R.A.Harris,1993:51).
D.Kilpert/Language Sciences25(2003)159–209161
By generative[Chomsky]meant explicit:written in a way which did not depend on the unconscious asstudent是什么意思
sumptions of the reader but could be operated as a formal system.His tremendous achievement was to show that this is in fact possible with a human language,as distinct from an artificial‘‘logical’’language.But you have to pay a price:the language has to be so idealized that it bears little relation to what people actually write—and still less to what they actually say. (Halliday,1994a:xxviii)
I will briefly explain my understanding of what kind of science it is that Halliday does.Butler(1989)2looks at the variety of theoretical perspectives within SFL and traces some of the differences of opinion to two readily recognizable ways of doing science,which can be crudely summed up as,on the one hand,the desire to keep your options open by accommodating multiple perspectives,and,on the other,the desire tofind the one‘correct’way and to prove its correctness beyond reasonable doubt.While admitting the value of the former,Butler evinces a hankering after the latter.He points out that‘the Popperian style is regarded by Halliday as just one possible valid way of‘‘doing linguistics’’,his[sc.Halli-day’s]own preference is for the creative development and refinement of insights, which may often be rather vague atfirst’(1989:8).Butler’s opinion is that‘although the imaginative,creative style of working which Halliday himlf strongly favours is vitally important at the stage of hypothesis formation,true progress can never be made unl
realize过去式
ess the hypothes are truly testable,and unless we go on to test them publicly and explicitly’(1989:29).He express doubts,for example,about the ‘functional components hypothesis’,saying that,although it is‘a uful pedagogical and practical convenience’,it‘rests on very dubious foundations’(1989:29).He suggests that Halliday and Fawcett are covertly admitting to the dubiousness of this hypothesis when they say that‘The notion of st valuable in practice as an expository aid—both for onelf,in teaching students,and in intro-ducing students to systemic theory’(Halliday and Fawcett,1987:7,original empha-sis)’(1989:29).
Here I want to disagree with Butler.The reference to their expository value was not an attempt to make excus for the metafunctions.Halliday does not posit the as a discovery about some entities that actually exist in language,but ts them up as a way of talking about language.I cannot emphasize this point too strongly;in my view, this is the esntial strength of Halliday’s linguistics.Butler also says he thinks that‘the categories offield,tenor and mode need much clearer definition before we can jus-tifiably u them in tting up and testing hypothes’(1989:29).I would suggest that this opinion needs revising now that,a decade later,the concepts have proved their worth many times over as aids to discour analysis and language teaching. (‘The test of a theory of language,in relation to any particular purpo,is:does it go?’Halliday,1985:6.)The metafunctional con
cept is a‘frame of reference’2I am aware of some unfairness in arguing against opinions expresd more than a decade ago,which may well have since been modified.I hope I will be forgiven for taking Butler’s1989position as ufully reprentative of some current debates about Halliday’s work.
162 D.Kilpert/Language Sciences25(2003)159–209
(Halliday,1997:4).It is‘something to think with’(Halliday,2000:222);we can evaluate it and e how far it helps in solving problems,‘but we cannot test it for being right or wrong’(Halliday,1996:19).
Halliday’s idea of‘creative development and refinement of insights’is not to t up concepts as hypothes to be rigorously tested and expod to falsification,and then—how many decades down the line?—finally to allow the consumer to u them.That idea rests on misguided emulation of the hard sciences and a failure to distinguish between a theory and a hypothesis(Halliday,1996:19,referring to Hjelmlev’s distinction).Butler worries unnecessarily,I think,that‘the consumers of systemic linguistics will assume that the claims which have been made have been thoroughly tested’(1989:31).But a linguistic theory is not quite the same thing as, say,an inadequately tested drug.A linguistic theory such as Halliday’s grows through being ud—and the test of its worth is how well it enables us to talk about language,in all the applications for which lingui
美元汇率破7意味什么
stics is needed.This is not just the simple idea that we judge our work by how uful it has been to others (Butler,1989:24).The point is that the theory cannot develop other than by being put to u.
Butler complains makes is that Halliday nowhere explicitly‘reject[s]an earlier claim in favour of a new and different one,or giv[es]reasons for the replacement of one claim by another;’(1989:7).This again is to misunderstand Halliday’s way of working.In linguistics there is no‘one true path’:a new concept may be not so much a replacement as an alternative,and an older insight may be as valid as a new one for a particular purpo.For example,it is interesting to note how many systemicists still u thefirst edition of the Introduction(1985),as a companion to the cond (1994a).In similar fashion,Halliday’s three-part article in the Journal of Linguistics (1967/1968)has never lost its relevance and ufulness.What would we want him to reject?Butler points out that Halliday has‘explicitly dissociated himlf from the ‘‘God’s truth’’view of linguistics which maintains that there can be only one correct grammar’(1989:6),preferring to allow that a number of different views might all be ‘right’at different times and for different purpos.3
A fundamental problem with the Popperian style as applied to linguistics is that it holds out little hope of ever giving us the full picture of language.Butler describes this style as‘the practice of developing falsifiable hypothes relating to specific and often small areas of language’(1989:8).The many enter
pris of this kind that have occupied many linguists over many years do not em to be getting us very much clor to an understanding of the whole.I would argue that Halliday’s strategy is more likely to take us there.I have wondered whether in his description of Darwin’s 3In a new work,Fawcett develops elements of Halliday’s initial model,Scale and Category Grammar.He says that‘later developments in Halliday’s thinking have left most of the concepts prented in‘‘Categories’’[sc.Halliday,1961]with a curiously And the concepts that have superded them in Halliday’s current model em to hover—insightfully or unsatisfactorily, depending on your viewpoint—somewhere between the levels of meaning and form’(Fawcett,2000: Introduction).
D.Kilpert/Language Sciences25(2003)159–209163 method of working he might to some extent be describing his own:‘Darwin’s strat-egy is that of accumulating mass of evidence rather than moving forward logically one step at a time’(1990:103).But underlying the‘mass of evidence’there must be one unifying creative insight that proves to have been esntially right all along; otherwi all the painstaking investigation of‘small areas’will have been for noth-ing,and we arrive at the frightening conclusion:‘Tout est a`refaire’(Roy Harris, 1997:308).
Until it is realid that much metalinguistic description is bad on a myth—‘the nineteenth-and twen
every moment of my lifetieth-century reification of‘‘grammar’’as a linguistic compo-nent which has a real existence independently of grammarians’communicational purpos’(Roy Harris,1997:234)—the fruitless arguments about the‘rightness’of a particular picture will continue.Modifications and re-arrangements,[by, e.g. Fawcett,Martin,and Gregory(Gregory,1987)],should rather be en as variations on a theme,not as competitors for a prize.Or,as Martin puts it in a recent dis-cussion of variation within the theory,‘what is required is a clear exposition of alternatives,and some n of which kinds of consumers mightfind which alternatives attractive’(1998:52).
All the foregoing is to explain why I have not explored in this article the‘rights and wrongs’of some recent debates within and around SFL,interesting and valuable though the are.4The most important point of Butler’s article,for my purpos,is his statement that
托辞Although there is a multiplicity of views within systemic linguistics,there can be no doubt that the work of Michael Halliday forms the central core:it was Halliday’s work whichfirst developed the programmatic ideas of Firth in the late50s,and it is he who has been the undisputed leader of the systemic move-ment since then.Tho systemicists who views diverge to some extent from Halliday’s have reached their positions largely by reacting to the rich am of innovative ideas which he has mined over the past three decades.(1989:2)
It is this‘central core’that I am concerned with in this article.I believe that Hal-liday’s bigger picture of language will enable others to work on the details without losing sight of the whole.To emphasize what it is that he has achieved,I have counterpointed my discussion of his work with various attacks on linguistics that have been made by that most irreverent critic,Roy Harris,5and by others who have made similar criticisms,such as Robinson,Singh,and R.A.Harris.This may em at odds with my suggestion earlier that the adversarial approach in linguistics might 4For the reader interested in the debates,one example is Huddleston(1988),responded to by Mat-thiesn&Martin(1991),followed by a further respon from Huddleston(1991);and an entertaining recent example is de Beaugrande’s(2001)respon to Widdowson(2000).
5For example,Roy Harris(1980,1981,1987,1990,1997).(Not to be confud with R.A.Harris,to whom I also refer in this article.)
164 D.Kilpert/Language Sciences25(2003)159–209
profitably be toned down,but Harris’s critique is of a different kind:it is not a coup attempt aimed at installing a particular model,but rather a plea to linguists to acknowledge their limitations,and to reali the extent to which they are led astray by unquestioned orthodoxies,in particular by the‘langu
age myth’mentioned earlier.I believe that Halliday succeeds in avoiding such pitfalls.I have therefore considered his achievements to some extent in the light of Harris’s critique,to e where there might be some points of congruence between what Harris demands of linguistics and what Halliday has made of it.I claim that Halliday’s work gives us some reason to believe that the problems Harris identifies may not be insuperable, not least becau he has begun to give us the fuller picture of language which is in esnce what Harris calls for.
In the review and discussion that follow I have assumed that the reader is aware of some milestone works,for example,Halliday and Hasan’s Cohesion in English (1976),and Halliday’s Language as Social Semiotic(1978)and An Introduction to Functional Grammar(1985,1994a),but is not deeply into the theory;so I must apologize in advance for explaining matters that may em obvious to some.For tho who are new to the theory,some uful sources are the following.Halliday’s Introduction(1994a)is esntial.Some good beginner-level introductions are Thompson(1996),Martin et al.(1997),and Butt et al.(2000).See Martin(1997:438) for a further list,and Matthiesn and Halliday(1997)for an online introduction. Two excellent short overviews are Halliday and Martin(1993:22–50)and Halliday and Matthiesn(1999:507–564),and two of the best comprehensive accounts are Thibault(1987)and Halliday(1996).Readers needing background in the
functions aspect of sfl are referred to Halliday(1994a),and tho wanting to go more deeply into the systems to Martin(1992)and Matthiesn(1995).
2.Review of Halliday and Matthiesn(1999)
Some of the fuller picture Halliday is aiming for can be found in this large and inclusive new book.It is the result,as the authors explain,of notes made from dis-cussions between the two of them over some years.It‘was it at a summer school in the midwestern United States;on a tour during a visit and ’(ix).They apologi for not having disguid the ‘sporadic’nature of the book’s genesis,which does give it something of a peripatetic quality,but this is due to the variety of topics covered rather than to lack of polish: much revision and rewording has gone into each ction,as will be evident to any-one familiar with sfl literature.6Language is a subject that is notoriously difficult to 6Readers of Halliday’s work over the years will be familiar with the technique of restatement-with-increments with which he has built up his theory.Reading his work chronologically,one is aware of a powerful process of growth:a subtle combination of established ideas with new.As Hodge remarks,‘the ries of restatements form a polyphony,not a mere repetition’(1988:156).
あいりみく
