Vous êtes sur la page 1sur 14

Language and social man (Part


rrejudicc and animosity of race, culture and class. These cannot be tngineered away. One of the more dangerous of the terms that have been
t'oined in this area is 'social engineering'; dangerous not so much because it \rrggests manipulating people for evil ends - most people are alert to that rlirnger - but because it implies that the social environment can be fashioned likc the physical one, by methods of demolition and construction, if only the pllns and the machines arc big enough and complicated enough. Some of the rl li)rtunate effects of this kind of thinking have been seen ftom time to time rrr t he field of language and education. But social wellbeing is not definable, ,,r irttainable, in these terms. 'l]ducation'may sound less exciting than social engineering, but it is an ,,1(lcI concept and one that is more relevant to our needs. If the engineers ,r r(l the town planners can mould the physical environmenl, it is the teachers \r lro cxert the most influence on the social environment. They do so not by

Language and social man (Part 1)


1 Languag and the enYironment If we ever come to look back on the ideology of the 1970s,
as suggested by

the write of an imaginary 'retrospect from 1980' published, inThe Observer in the first issue ofthe decade, we are likely to see one theme clearly standing

ipulating the social structure (which would be the engineering

out, the theme of 'social man'. Not social man in opposition to individual man, but rather the individual in his social enyironment. What the write was forecasting - and he seems likely to be proved accurate - was, in effect, that while we should continue to be preoccupied with man in relation to his surroundings, as we were in the 1960s, the 1970s would show a change of cnrrhasis from the purely physical environment to the social environment. 'l his is not lr ncw concern, but it has tended up to now to take second place; rvt. hrvc hccn nrorc involved over the past twenty years with town planning irrrrl rrhirrr l'cncwul. with the flow of traffic around us and above our heads, irr(l llr,,st lcecntly with the pollution and destruction ol our material r( \our( c\. 'l his irrcvittrbly hs distracted us from thinking about the other

,r1rrloach) but by playing a major part in the process whereby a human being l,, t orrres social man. The school is the main line of defence against pollution

)l or! r'rrvilorrllrc rr t. t hitl which consists of people - not people as mere ol hrrrrrirrrily. so t|ltny lo thc squarc mile, but other individuals with u,lronr wc hirve r.lcirlings ()l t nlorc ol lcss personal kind. 'lhc 'errvilorrnrcnl' is social as wcll as physical, and a state of wellbeing,
l)rr I

rr rlrc human environment; and we should not perhaps dismiss the notion of (l( l(nce'too lightly, because defensive action is often precisely what is n, , rlctl. Preventive medicine, after al[, is defensive medicine; and what the ., lrool has failed to prevent is left to society to cure. ln I hc development of the child as a social being, language has the central r, ,l( . Llnguage is the main channel through which the patterns of living are r r ,rr\nr itted to him, through which he leans to act as a member of a 'society' rrr rrrrtl through the various social groups, the family, the neighbourhood, ,rr,l so on - and to adopt its 'culture', its modes of thought and action, its I ,, lr( l\ ir nd its values. This does not happen by instruction, at least not in the I'r, rclrrol yearsi nobody teaches him the principles on which social groups

which rlepcnds on hirrnrony with thc environment, demands harmony of both kinds. Thc raturc of this state of wellbeing is what environmental studies are about, Ten years ago we first came to hear of 'ergonomics', the study and control of the environment in which people work; many will remembe London Transport's advertising slogan 'How big is a bus driver?', announcing the design of new buses'on ergonomic principles'. This was characteristic of the conception of thc environment at that time. Today we would tind more emphasis laid on thc sooial aspects of wellbeing. No one would assert that the shapc of the bus driver's seat is unimportant; but it no Ionger seems to be thc wholc story. There are other aspects of environmental design which sccm at least as significant, and which are considcrably more difficult to adjust. Consider for example the problem of pollution, thc tlclcnsivc irspcct of environmental design. The rubhish crccr, lhe ( ( ) l r I r r i r I i ( ) I ol lrir and wltcr,cvcnthcnrosllclhitl proccssesol rlrysicirl l)()llulior irl)l.:tI to l)c nl)rc
rrr rrr

.rr,,,rgrrnized,ortheirsystemsofbeliefs,norwouldheunderstanditifthey rr r, 1. Il happens indirectly, through the accumulated experience of numer,,u,, srlrll events, insignificant in themselves, in which his behaviour is | r,l( (l irnd controlled, and in the course of which he contracts and develops , r '.r rrrirl rclationships of all kinds. All this takes place through the medium , 'l Lr n,lugc. And it is not from the language of the classroom, still less that of , ,,rrrr ol law, of moral tracts or of textbooks of sociology, that the child l, r r\ irlx)ut thc culture he was born into. The striking fact is that it is the rrr, l r r tlinlry cvcryday uses of language, with parents, brothes and sisters, "
rr, r,lrlrorr lhood children, in the home, in the street and the park, in the shops .r rr,l tl. I lirins and thc buscs, that scrve to transmit, to the child, the essential ,rr rlrlics ol socicty and thc naturc of social being.

tl.itcliblc lltrn lhe rollrrliorr irr llr,..srxrrl


I r I


I tlr.rt ir t:rrrsr.,rl


I lr rs, rr hlie l, is whirt this chaptcr is about. It is a general discussion of lhe r,l,rtrrrrr ol lilrgurAc 1() social nran, irntl in rarticular language asitimpinges ,,rrtlrr oI ol lllr. tcit(ll('r irs ir crcirl()r ol social man-oratleastasamidwife rr tlr( ( r( ir l ioI l)l()(css.'lllrl lhis (l()cs [()t rnean sintply language in school is lr, r l y 1 l ll rrrr'rrrs. r'lrlher'. llrrrgrrirge in thc t()(irl c()rrtcxt ofthc interactr,'rr lrr.t\\'rt. lur irrrlivirlrrrl irrrrl lris hrrnrrl trtvitortllcnl: lretwccn onc irdi, r ,


The sociolinguistic perspective

Language and social man (Part



vidual and others, in fact. But thc point of view to be adopted will be an educational one, emphasizing those aspects of language and social man that arc most relevant to the teacher in the classroom. It might seem that one could hardty begin to consider language at all without taking account of social man, since language is the means whereby
people interact. How else can one loo k atlanguage excepl in a social context? In the last resort, it istrue that the existence oflanguage implies the existence ofsocial man; but this does not by itselfdetermine the point of vantage from


logic and

which language is being approached. Let us think for a moment of an individual human being, considered as a single organism. Being human, it is also articulate: it can spea!and understand language, and perliaps read and write as well. Now the abiiity to speak and undetand aiises, ind makes sense, only because there are other such organisms around, and it is natural to think of it as an inter-organism phenomenon to be studied fom an inter-organism point of view. But it is also possible to investigate language ftom the standpoint of the internal make-up of that organiim: the braln structure, and the cerebral processes that are involved in its speaking and
understanding, and also in its learning to speak and to understand. So there isn intra-organism perspective on language as well as an inter_organism one.

sbsrace; phonic/


grammar and






emphasis between them, tends and fashions in scholarship which lead to concentration on one, for a time, at the expense ofthe other. in the 1960s the major emphasis was on what we are calling intra-organism studies, on the investigation of language as knowledge, of .what the speaker knows', running parallel to, and probably occasioned by, the relative neglect of man,s social enyironment. Thete has now been a move back towards a greater concern with the social aspects of language, a restoring of the balance in linguistic studies, with account once more being taken of the inter-organism factor-that oflanguage as social behaviour, or language in relation to social man. A diagrammatic representation of the nature of linguistic studies and their relation to other fields ofscholarship will serve as a point ofreference for thc subsequert discussion (figure 1). The diagram shows the domain oflanguage study - of Iinguistics, to give it its subject title - by a broken line; everyihing within that line is an aspect or a branch of linguistic studies. In the centre is a triangle, shown by a solid tine, which maks offwhat is the central area of language study, that of language as a system. One way of saying what is meant by 'central' here is that if a student is taking linguistics as a university subject he will have to cover this area as a compulsory part of his course, whatever other aspects he may choose to take up. There are then

The two standpoints are complementary; but there tend to e shifts of








,tl ogv l ".. '"

anthropo ogy



_f--human biologY



I lr.rr. outside this triangle, are the principal perspectives on language that rrs beyrnd a consideration solely of language as a system, and, in so

rlrrrrr,. 'i,rr" on other disciplines. Any study of language involves some ,rr rr r rr r( rr l() ()ther disciplines; one cannot draw a boundary round the subject r rr I rrrrr litc it fiom others. The question is whether the aims go beyond the , lrrr rrlrtiorr of language itself; and once one goes outside the central area, ,,r, r,, rr(lriring not only into language but into language iIL relation to .,,rrr, tlrrrrg c lsc. Thc diagram summarizes these wider fields unde the three lr, ,r,lrrris. 'languagc as knowledge', 'language as behaviour', 'language as

certain projections from thc triangle, representing special sub-disciplines within this central area: phonetics, historical linguistics and dialectology the last of these best thought of in broader terms, as the study of language varieties. These sometimes get excluded from thc ccntral rcion, bul probably most linguists would agrcc in placing thc nr willrin it I il onc coultl givc a thrcc-dinrcnsirnal rcl)rcscrttliotr thcy w0tr lrol l0ol lilr. r.\( t(.sccnccs.

ol tlrcsc takes us into the realm ofliterature. which is all too often il il wirs sonrcthing insulated frtm and even opposed to language: \\, rr)n(( rtr'rle rnirirrlyorr litcrature hclc -wc don't do muchon language', ,rr rl r orrr'crrlrrting ()n lile[lrrc'nradc it .rossihle to ignore the fact that l ' r,rrllrl' is rrrirtlc 0l lrrnrrirge, Sinlilirrly (hc untlcrgraduatc is invited to , lrr x rrl lrclu ccrt lirrtg. rttttl lil.'. ltt lltct lllc .listittct ion t hat is bcing irrrrlictl is rr Ir'r lr'( llv rrr(. ir rt itt,,lttl ottc lrc l wcc tt I wr t tlillt r'(. tll c tll l)lltscs ot ol'ie lll lll i, 'tts.
I lr( lrsl ,l( (l rs


Thc sociolinguistic perspectrve

rr hrrt

Language and social man (Part



one in which the centre of attcntion is the linguistic system and the othcr having a focus elsewhere; but it is wrongly named, and thcrefore, perhaps, liable to be misinterpreted. One can hardly take literature seriously without taking language scriously; but language here is being looked at from a special point of view. The other two headings derive from the distinction we have just been drawing between the intra-organism perspective, language as knowledge, and the inter-organism perspective, language as behaviour. These both lead us outward from language as a system, the former into the region of psychological studies, the lattet into sociology and related fields. So in putting language into the context of'language and social man,, we are taking up one of the options that are open foq the relating of language study to other aields of inq uiry. This. broadly, is rhe ociolinguistic option; and the new subject of sociolinguistics that has come into prominence latety is a recognition of the fact that language and society - or, as we prefer to think of it, language and social man - is a unified conception, and needs to be understood and investigated as a whole. Neither of these exists without the other: there can
be no social man without language, and no language without social man. To recognize this is no mere academic exercise; the whole theory and practice of education depends on it, and it is no exaggeration to suggest that much ofour failure in recent years - the failure ofthe schools to come to grips with social

I irlguage can be considered from cither of these points of view; the fist

wc called on the diagram 'language as behaviour', the second 'language rs knowledge'. 'Language and social man'means languagc as a function of I lr( whole man; hence language man to man (inter-organism), or language as

rrrtn behaviour.
l'lrcse are two complcmentary orientations. Thc distinction between them a difficult one to make; in itself it is rather obvious and simple. But it


it is possible to embed one perspecrr\( insidetheother: to tteat language behaviour as if it were an aspect of our L rrl)wlcdge of language (and hence to see it in terms of the capacity of the lrrrrrrirn brain), and also, though in a rather different sense, to treat the rr r,lrvidual's knowledge of language as a form of behaviour. In other words \\r' ( irn look at social facts from a biological point of view, or at biological l,r( l\ I}om a social point of view.
Irrrr lrccome complicated by the fact that I lre study oflanguage as knowledge is an attempt to find out what goeson


- can be traced to a lack of insight into the nature of the relationships between language and society: specifically of the processes, which are very largely linguistic processes, whereby a human organism turns into a social being.
2 Inter-organism
and intra-organism perspectives

The diagram in section 1 suggests

a context for language study, placing it in the environment of other fields of investigation. It also suggests where 'language and social man' fits into the total picture of language study. The discussion of the diagram will perhaps have made it clear (and this harks back to what was said at the beginning) that whe n we talk of.social man'the contrast we are making is not that of social versus individual. The contrast is rather that of social versus psychophysiological, the distinction which we have attempted to draw in terms of intet-organism and intra-organism


When we refer to social man, we mean the individual considered as a single entity, rathcr than as an assemblage of parts. The distinction we are drawing here is that bctween the behaviour of that individual, his actions and interactions with his cnvironment (especially that part of his environment which consists of other individuals), on the one hand, and on the othe hand his biological nature, and in particular the internal structure of his brain. In the first of these perspectives we are regarding the indivitlral as irn intcgral whole, and looking at him frtr the outsi(lcl in thc secorrtl wr. irlc Iircusing

our altcntion ol] thc pilrts. antl ltxrking or tltc iltsirh.. illr tlrc works.

rrr',rli lhc individual's head. The questions being asked are, what are the ,rr, r lrrrrisms of the brain that are involved in speaking and understanding, rr I rr lrlt must the structure of the brain be like in oder for the individual to l,,,rlrle 1() speak and understand language, and to be able to learn to do so? No\\ ()ne important fact about speaking and understanding language is rlr rr rt rlways takes place in a context. We do not simply 'know' our mother r',r,,r( 'rs an abstract system of vocal signals, or as if itwassomesortof a ,r,rrrrrrrr book with a dictionary attached. We know it in the sense of I rr,rs rrr lxrw to use it; we know how to communicate with other people, lr,,rr r clrotse forms oflanguage that are appropriate to the type ofsituation ',, lrrrt ourselves in, and so on. All this can be expressed as a form of lrr,,rrltrhc' we know how to behave linguistically. llri rt lorc it is possible, and is in fact quite usual in what is nowadays , ,rll, ,l sociolinguistics', to look at language behaviour as a type of knowl, rl,, . so lhat although one's attention is focused on the social aspects of I rrr,rrr,.r' on language as communication between organisms - one is still r,,l rn1 rr llrt is essentially an intra-organism kind of question: how does the rrllrr rrlrrrl know how o behave in this way? We might refer to this as , r , l, soe irlinguistics: it is the exte rnal behaviour of the organism looked at t lr,,rr rlr( l)()int of view of the internal mechanisms which control it. \\, \iri(l irbovc that the two perspectives were complementary, and it ,r,,rrlrl lrl reirsonable to conclude that they are really insepalable one from ilr,,rlr(!.ltu(ifsothcinscparabilityholdsinbothdirections. It istrue that tlr, rrr lrv itlrllrl's potcntial for linguistic interaction with others implies certain tlllllt,,, rrl)ul thc irrlcrnal makc-up of the individual himself. But the con\ L,' r\ irlso lllrc. l hc lircl that thc brain has the capacity to store language rrrrl fi.,r' il lo e llcctivc corrrnttnicittion implics that communication takes 'l r, , tlrrrt lhc irtlividutl ltits it 'hchaviour potcntial'which characterizes his trtr r.r( tirr rrillr olhel itrlivirltirls ol his srccics. ',Irr r' rr) r[)lrl)l llt'ltrtttittt Irtitttt r.v(llvcrl ill its Plcsclrt li)rm through the l'r,'r, \\ (rl ltttttlttt l|(itt1s ( orllrlllllti( illillA wilh oltc iln()thcl'. tlrc lilllcr


The sociolinguistic perspectrve

Languagc and social man (Part



perspcctive is likely to be highly significant from an evolutionary point of view. But that is not our main point of departurc here. There is a more immediate scnse in which the individuat, considered as one who can spcak and undestand and read and write, who has a.mother tongue,, needs to be seen in a social pcrspective. This conccrns the part that Ianguage has played in his own development as an individual. Let us start with the notion of the individual human organism, the human being as a biological specimen. Like the individual in many other species, he is destined to become one of a group; but unlike those of all other species, he achieves this - not wholly, but critically - through language. It is by means of language that the ,human being' becomes one of a group of'people'. But ,people,, in turn, consist of 'persons'; by virtue of his participation in a group the individual is no longer simply a biological specimen of humflnity - he is a person. Again languag is the essential element in the process, since it is largely the linguistic interchange with the group that determines the status of the individuals and shapes them as persons. The picture is as in figure 2:

Strcial oles are combinablc, and the individual, as a member of a society, )ccupies not just one role but many at a timc, always through the medium of lrnguage. Language is again a necessaly condition for this final elemcnt in

rl)c process of thc development of thc individual, from human being to as a l)r'rson to what we may call 'personality', a personality being intcrpreted of of a number is seen as the configuration rlc individual r, complex. Here the
r,,les definetl by the social relationships in which he enters; from these roles lr( synthesizes a personality. Our model now looks like figure 3:


ig. 2

US now interpret this in terms of a perspective on language. We have \()me way round in order to reach this particular angle of vision, ,,"0( r , r r in Iy ove rsimplifying the picture and perhaps seeming to exaggerate the ll r I', )r I nce of language in the total process. The justification for this is that be most releYant in an 'r, lrrvc been trying to achieve a perspective that will ,,lr(irli()nrl context. From this point of view, language is the medium


lrr,, rllh which a human being becomes a personality, in consequence of his

ln other words, instead of looking at the group as a derivation from and extension of the biologically endowed mental power of the individual, we
participation in the group. Instead of starting inside the organism and looking outwards, we can adopt a Durkheimian perspective an start from outside the organism in order to look inwards. But when we do adopt this perspective it becomes apparent that we can take the dialectic one stage lurther, and that when we do so language will still remain the crucial factor. The individual as a.person, is now a potential 'member': he has thc capacity to function within society, and once more it is through language that hc achieves this status. How does a society differ from a group, as we conceivc it hee? A group is a simple structure, a set of participants among whom thcre are no special relations, only the simple coexistencc that is implied by participation in the group. A society, on ihe other hand, does not consist of participants but of relations; and these relations define social roles. Being a member of society mcans occupying a social nrlc; and it is again hy mcans ol languagc lhlrl ir .pelson, bcconlcs p()lcnliillly lhc occrpillrl ol ir socitrl r.lllc.
explain the nature of the individual as a deriyation from and extension of his

rrrl,r'r'ship of society and his occupancy of social roles. The concept of l,,rr,s. behaviour, as a form of interaction between man and man, is "ti rrrr((l irlound, as it were, so that it throws light on the individual: the l,,r rr ir li()n of the personality is itself a social process, or a complex of social plays the key part Irr , 1 \\( s. r nd language - by virtue of its social functions rr rr I l( ncc i ust as the view of language as knowledge, which is essentially an rrrlrr rlrllrl orientation, can be used to direct attention outwards, through ,.ll, lr ( onccpts as the speech act, towards language in society, so the essenrr rll\ \o( iirl intcrpretation of language as behaviour can be used to direct ,rlh r rri )n ()nt() thc irdividual, placing him in the human environment, as we , r , r , rt rl it e a rlic r, a ntl cxplaining his linguistic potential, as speaker-hearer ,rrrrl s r itcr-rcrrtlcr, in thcsc tcrms. This does not presuppose, or preclude, ,llr\ |irlieulrr thcoly ilbotlt thc nature of the mental processes that are llr llr r'r I irr his rrtstc ry ol langttitgc, cithcr in how he speaks and understands

rrr rrr lw hc lcirrrrl l() (lo s() in thc lirst place. There are conflicting l,'r, lr,,t,,liiert lltc()ries()rt lltcsr: (ttcslirtns.;tswc shilll scc in the ncxtsection; lrrt orr l)rr\( lll l)('tsl)((liv(' is tetl(l'tl ilt lllis rcspcct. I lrl rrlrilitv lo \l)( irL rtt(l trrr(lt.l'sllll(1, illl(l llle (levcl(lPtl)clrl ol lhisilbility ir


The sociotinguistic perspective


Language and social man (Part



the child, are esscntial ingredients in the lifc of social man. To approach these from the outsidc, as inter-organism phenomena, is to take a functional view of language. The social aspect of language bccomes the reference point tor the biological aspect, rather than the other way round. In the next two sections we shall consider bricfly what this means.

torrguc consists in

rrthcr dctailcd blueprint of thc structurc of language. l,earning his mother fitting the patterns ofwhatever language hc hcas around

3 A functional

approach to language and language development

In the preceding section we outlined

ccrtain childrcn, pcrhaps because of their social background, have not acquired enough ol this commodity called language, and in order to help them we must send reliefsupplies. The implication is that there is a gap to be filled, and from this derive various compensatory practices that may be largely irrelevant to the children's needs. Now this is a false and misteading view oflanguage and ofeducational failure;and while one should not make too much of one item of terminology, we prefer to avoid the term ,language acquisition' and return to the earlier and entirely appropriate designationf
'language development'.

means whereby the various social relationships into which he enters are established, developed and maintained. This means lhat we are taking a functional view of language, in the sense that we are interested in what Ianguage can do, or rather in what thepeaker, child or adult, can do with it; and that we try to explain the nature of language, irs internal organization and patterning, in tems of the functions that it has evolved to selve. First ofall, therefore, we should look briefly into the question oflinguislic function, and say a Iittle about it in regard both to what language is and to how it is learnt by a child. Let us take the latter point first, and consider a functional approach to the question of how the child learns his mother Ionguc. This process, the learning of the mother tongue, is often referred to as'languagc acquisition', This seems rather an unfortunate term because it suggcsts that language is some kind of a commodity to be acquired, and, irllhough thc nrctaphor is innocent enough in itsell if it is taken too literally Ihc conscq ucncr:s can bc rather harmful. The use of this metaphor has led to thc helicl ir what is known as a 'deficit theory' of language learning, as a mcans ol cxplaining how children come to fail in school: the suggestion that

a general perspcctive on language and language learning in which society rather than the individual is at the centre of the picture, and the individual's language potential is interpreted as the

lrur intr thc famcwork which hc alrcady posscsses. The environmentalist \ r( w considcrs that languagc lcarning is not fundamentally distinct from ,t lrt r kinds of learning; it depends on thosc samc mental faculties that are rrrvolvcd in all aspects of the child's learning proccsscs. Rather than having I'rrll into his genetic makeup a set of concrete univcrsals of language, what tl, r'hild has is the ability to process certain highly abstract types ofcognirive
r, lrrtion which underlie (among other things) thc linguistic system; the very ,, r il ic properties of language are not innate, and therefore the child is more

on his environment - on the language he hears around him, with f he contexts in which it is uttered - for the successful learning ,,1 lrs rother tongue. In a sense, therefore, the difference of views is a r, ( r,cnce of the old controversy of nature and nurture, or heredity and

( I llc r

t, ,,,

, r\ rlr)ntlcnt, in a new guise.


,rt h

of these views can be criticized, although the criticisms that


trr,rlly made often relate to particular models of the learning process that lr.rr, o e]ss5y connection with a nativist or enyironmentalist position. 1,, , rrnrpl, it is sometimes assumed that an environmentalist interpre-

rrrr,'r inrplies some form of behaviourist theory, an essentially stimulusr,',',,rrsc view of learning; but this is totally untrue. Equally, the nativist r r, ,r rs lry no means dependent on the notion that learning proceeds by
lrttrrr, 5 into the marked slots which nature provided and running the ,,, ,, lrrr(, l() tcst whether the match is appropriate. The differences between

rrrr\r,,r irnd environmentalist are differences of emphasis, among other tlrrrrl., rrr tlrcir ideas concerning the essential character of language, which rr rrr lr.rrr two rather different traditions. Broadly speaking. the nativist ,,,,,,1, I r( llects the philosophical-logical strand in the history of thinking ,l,,,rr lllrAutge, with its sharp distinction between the ideal and the real (,\lr, lr ( homsky calls 'competence' and 'performance') and its view of lllr,r.rt[ irs rlrs - essentially rules of syntax. The environmentalist rep,,,rt,,tlr(cthnographictradition,whichrejectsthedistinctionofidealand r, rl ,l, lirts what is grammatical as, by and large, what is acceptable, and , , . Lurltrrgc i\s resource resource fot meaning, with meaning defined in
r ll r'. ,,1 lll nct oD. To this extent the two interpretations are complementary ,,,rlr, r rlrillr e()ntradictory; but they havc tcnded to become associated with ,,,rlllr, rrr,i rsychological theories and thus to be strongly counterposed. ll|l(,,tu l(.nt ()ltcn put firrward in supportof anativist approachmustbe 'lr rr .',r'r I rrs lrrllircious; this is thc theory of the unstructured input, according t,, 'rlrr lr tlrt elriltl cnrrot bc dcpcndent on what he hears around him 1,,, rrr , rrlrrl llr. hcrrs is no rl()c than bits and pieces - unfinished or rrt'r rrrn rtr( 'rl sL nlcnces, lrrll ol hcsitirtions, backtracking, unrelated fragrr, lll . ,rrl tlr(. likL. 'llris itlcir scenrs to havc ariscn because the earliest I'rln r, r'lrrl'\ ol torrrrctltrI tliseorr'se ll)rl linguists analyscd were usually r,,,rl'1 ol rrIt IIr'r'IrrrrI (ottv( rsrl()ns. wlritlr rkr lt'rd lo bc vcry scrappy, ilr,, llrr' \l irIL r\ llc lurvinl'. l() |llirr r\ llrt,y 1r rrkrrrg rrrrrl llre l)rcnliscs itrc

rcferred to as the 'nativist' and the 'environmentatist' positions. Everyone agrees, of course, that human beings are biologically endowed with the ability to learn language, and that this is a uniquely human attribute - no other specics has it, however much a chimpanzee or a dolphin may be trained to operate with words or symbols. But thc nativist vicw holds that there is a specific language-learning faculty, tlislircl li.orl otlrcr lcrrning facultics. arrd that lhis providcs thc hrrulrrr itlltrrrt rvitlr r rr,:rrtvlrrtle lrntl

of approach to the question of language devclopment. These have bcen

In the psychological sphere, there have recently been two alternative lines


The sociolinguistic perspective


Languagc and social man (Part



constantly shifting, and which are also largely insulated from the immediate situation, so that there are no contextual clues. But it is not in fact true of the ordinary everyday speech that typically surrounds thc small child, which is fluent, highty structured, and closely related to the non-verbal context of situation. Moreover it tends to have very few deviations in it; I found mysclf when observing the language spoken to, and in the presence of, a small child that almost all the sequences were well formed and whole, acccptablc cven to the sternest grammatical lawgiver. Of course, the fact that thc notion of unstructured input is unsound does not disprove the nativist theory; it merely removes one of the arguments that has been used to support it.

r spective. can throw a great deal of light on the nature of language itself. I iurguage is as it is because of what it has to do.

.r.,1ing thc question as an incidental bonus

u lrrrl is

fifth, though this is not so much a reason for for having done so. One of the young child is that of knowing rblcms language of a very in studying the l,r,
these we might add a


More important than the grammatical shape of what the child hears,
however, is the fact that it is functionally related to observable features of the situation around him. This consideration allows us to give another account of language development that is not dependent on any particular psycholinguistic theory, an account that is functional and sociological rather than structural and psychological. The two are not in competition; they are about different things. A functional theory is not a theory about the mental processes involved in the learning of the mother tongue; it is a theory about the social processes involved. As we expressed it in the first section, it is concerned with [anguage between people (inter-organism), and therefore learning to speak is interpreted as the individual's mastery of a behaviour potential. In this perspective, language is a form of inteaction, and it is learnt through interaction; this, essentially, is what makes it possible for a culture to be transmitted from one generation to the next. In a functional approach to language development the first question to be asked is, 'what are the functions that language serves in the life ofan infant?' This might seem self-contradictory, if an infant is one who does not yet speak; but the paradox is intentional - before he has mastered any recognizable form of his mother tongue the child already has a linguistic system, in the sense that he can express certain meanings through the consistent use of vocal sounds. There are, perhaps, four main reasons for putting the question in this form.
1. We can ask the same question at any stage in the tife of the individual, up to and including adulthood; there have in fact been a number of functional theories of adult and adolescent language. 2. It is much easier to answer the question in respect of a very young child; the earlier one starts, the more clearcut the functions are (whereas with an approach based on structure, the opposite is the case; it is in general harder to analyse the structure of children's speech than that of adults). 3. We can reasonably assume that the child is functionally motivated; if Ianguage is for the child a means of attaining social ends - that is, cnds which are important to him as a social being-we need lottk no litrthcr than this for the rcrsons why hc lcarns it. 4. A llnctionirl ir-rpr-olch t() llnglrirll('. il it irrllrrlrr rr rlt vcirrtttr. ttlttl

language and what is not. We can answer that, in a functional context, ing that any vocal sound (and any ge sture, if thc dcfinition is made to ' rr, lrrtlc gesture) which is intcrpretable by reference to a recognized function , ,l lr n rruage is language - provided always that the relationship of sound to rrr, rr rrirrg is regular and consistent. The production of a sound for the purpose , 'l rr;rctising that sound is a meats of learning language, but is not itself an ll r,,lir cc of language. The production of a sound for the purpose of attracting ,rll rlion Ls language, once we have reason to assert that'attracting attentr,,r ir :r meaning that fits in with the functional potential of language at this

\r ry

.t,rt\' ('l dcvelopment. I ooking at the early stages of language development from a functional i r, rr roint. we can follow the process whereby the child gradually 'learns lr' ,\r' r( ) rroan' - for this is what first-language learning is. If there is anything

'rlrrlrthcchildcanbesaidtobeacquiring,itisarangeofpotential,whichwe ,,'rlrlrcle r to as his'meaning potential'. This consists in the mastery of a small lrnlx r ol clementary functions of language, and of a range of choices in
mr .rrrrA within each one. The choices are very few at first, but they expand ,rl,r,llr rrs thc functional potential of the system is reinforced by success: the , , rr rr,l., t hirt the child makes do in fact achieve the desired results, at least on a ,r'r rlh ir rl number of occasions, and this provides the impetus for taking the t,,,,,,..\ lrrrlher. As an example, Nigel, whose language I studied in suc-

,, rr,

\ix-wcekly stages from the age of nine months onwards, started t,l',r, rrtlv with just two functions and oneortwo meanings ineach. At 10$ lr,,rrtlrs, rlhcn he first had a recognizable linguistic system, he could express ,r r,,r.rl i)l twclvc different meanings; these were derived from four clearly r,l' rrrlr,rl)le linctions (the first four in the list below) and included, among ,,rlrr', lrirl wc might translate as 'do that right now!', 'I want my toy bird rl,,,r 1 ,11 'rlicc to see you; shall we look at this picture together?' By 16{ r"rtlr., \\'l)en hc was <n thc threshold of the second phase of language ,1, \ , l, l)rr( nt. I he movc into the mother tongue, he had six functions and a ' r',r,rl ,'l lillv rncanings that he could, and regularly did, express (Halliday

l' ",r)






N igc

l's progrcss I used


the framewotk a set of seven initial


1r,,n.,. s lolltws:

I Irr,.Irrrrl!( rrIrI ( l wllrl'): sirtislying nratcrial necds ' l{r'lrrlrrtrrry ('(l(' rs I l(ll yorr'): corrtrollirrg the behaviour of others I IrrIlt,r( IrorrirI ('rc rrrrtl yorr'): B( llinA irlong with othcr pcoPlc I I','r..olll ( lrr'rt I r'orc'): irlt'rrtilyirr1 irrttl cxrressing lhc scll' ' lllrrrrrlr'('lcll tttc rvlry ): r'rrIrtirt lltc wotltl :tt'otttttl tntl itsitlc onc


The sociolinguistic perspectivc

Language and social man (Part



6 Imaginative ('lct's pretend'): creating a world of one's own


Informative ('I've got something to tell you'): communicating new information.

ln order for language to be a mcans of lcarning, it is essential for the child

r. l,c able to encode in language, through words and structures, his !'xperi, ll( L of processes of the external world and of tht) pcople and things that
l,,r licipate in them.

These headings served as a useful basis for folowing the developmental progress of an infant whosc early speech sounds, although still prelinguisric in the sense that thcy were not modelled on the English language, were used to convey these kinds of intention - to obtain goods or services that he required (instrumental), to influence rhe behaviour of those closest to him (regulatory), to maintain his emotional ties with them (interactional), and so on. The meanings that he can express at this stage - the number of different things that he can ask for, for example - are naturally very restricted; but he has internalized the fact that language serves these purposes, and it is significant that for each of them he has one generalized expression, meaning simply, 'I want that'or'do that!'etc., uhere the interpretation is given by th situation (e.g. 'I want that spoon'or'go on singing'), as well as a number of specific expessions, very few at first but soon growing, and soon becoming independent of the presence of the object or other visible sign of his intent. So by adopting a functional standpoint we can go back to the beginning of the child's language development, reaching beyond the point where he has started to master structures, beyond even his first wods, if by ,words' we mean items derived from the adult language; and taking as the foundations of language those early utterances which are not yet English or French or Swahili or Urdu but which every parent recognizes as being meaningful, quite distinct from crying and sneezing and the other nonlinguistic noises the child makes. At this stage, the child's utterances cannot readity be ,translated'into the adult language. Just as we cannot adequately represent the sounds he makes by spelling them, either in the orthography of the mother tongue or even in phonetic script, because the system which these symbols impose is too detailed and specific, so also we cannot adequately represent the meanings the child expresses in terms of adult grammar and vocabulary. The child's experience differs so widely from that of the adult that there is only a very partial correspondence between his meanings and those that the adult is predisposed to recognize. But if his utterances are interpreted in the tight of particular functions, which are recognizable to the adult as plausible ways of using language, it bccomes possible to bridge thi: gap between them - and in this way to show why the infant's linguistic system ultimately evolves and develops into that of the adult, which is otherwise the most puzzling aspect ofthe language devclopment process. By the time he reache the ag of l8 months, Nigel could use language effectively in thc instrumental, regulatory, intractional and personal functions, and was beginning to use it for pretend-play (the 'imaginative' function), and also heuristically, tbr the purposc of exploring the environment. Now for the first timc hc launched into English, making rapid strides in vocabulary itnd glrrnnrirr; rntl it was very clear from a study of his spccch thilt his prirrt irrrl rtrotivt, lirr tkrirg qr wirs lhc tse ol lirngtrirgc irs lr lclrlning rlr..vilc


l.nnguage and social structure

lr ..t r'l itln 3, we considered the process of [earning the mother tongue from a lllr(rional point of view, interpreting it as the progressive mastery of a rrrrrrlrcr of basic functions of language and the building up of a 'meaning
in respect of each. Here we are adopting a sociolinguistic perspec-or rather a perspective which in terms ofthe earlier ,lr',, rrssion would be inter-organism. Language is being regarded as the
t r(




,)n language

, rr,,rrling

of a 'behaviour potential'into a 'meaning potential'; that


of expressing what the human organism 'can do', in interaction human organisms, by turning it into what he 'can mean'. What lr, (,rr nrean (the semantic system) is, in tum, encoded into what he'can ,\ lllrc lexicogrammatical system, or grammar and vocabulary); to use ,,rrr ,,rvn lblk-linguistic terminology, meanings are expressed in wordings, \\,,rrlgq are, finally, recoded into sounds (it would be nice if we could ,r ,,,rrrrrrlings') or spellings (the phonological and orthographic systems). l, rrrrr likc meaning, wording and spelling are so familiar in everyday t( , , lr I hat we are hardly aware of them as ways of talking about language. ll,ri , \{ ry time we say, to a pupil, or to a committee chairman perhaps, I tlrrrrl. l ou'll have to altet the wording', we afe mking systematic assumpr,,ll . .rl),)ut language, bringing into play what Peter Doughty calls 'a "folk 1r,,,',r'.rr. . u "common sense" about the language we live by' (Doughty ,,,/ t,)/1. lt). I lrr', rr, r spcctive is valuable to the linguist because it affords an insight into rr /r 1 f,1,q 1c is as it is. There is no a priori reason why human language 'lr, r lr I lr;ve tr kcn just the evolutionary path that it has taken and no other; ,, rl l,r,rrrrs could have produced a symbolic system ofquite a different kind. llrr rl \\, r'r)[sidcr what language is required to do for us, there are certain lr,',, r r, n\ \\ lrich it m ust fulfil in all human cultures, regardless ol differences r tlr, rlrr,ricrrl ilnd matcrial cnvironment. These are functions of a very
, ,

,.. ,r rr(rns rr lr ( !rhcr

, rr, r,rl l.trttl.

I I ,rrt'ulll( hirs to intcrprct the whole of our experience. reducing the rrr,l, lrrr, lv vruictl phc nonrcnu of the wold around us, and also of the wold rr ,1, r,,. llr( l)r()(rsscsol ()ur own consciousness! to a managcable numbcr ,1,r.,,r'.. t)l rlretlorrrcrrr: tyl)es ()l proccsses, cvents and actions, classes of
,,1,,, t

. rr.rrrlr' ln(l ilsliluti()ns. utrl thc likc. Lrrrt,;l lrir\ lo ( )il)re\s cr'rlilir clcnrenlary logical rclations, like'and' .rl ,'r .llr(l'rl '.rrrrrtll irs lhose ( r(.irle(l by lirngtrrgc itscll such as'namcly'. '

t ,rtrl ltr'rttr' t Lrtrt,ttirll| lrrs lrr crrr'rs ()rr l)ttlril)rlior. r\ sl)(ilk('rs. itt lhr'stcclt


1'hc sociolinguistic perspective

Language and social man (Part



situation; thc roles we take on ourselves and impose on others; our wishes, lclirgs, attitudes and judgements. 4. Language has to do all these things simultaneously, in a way which
relatcs what is being said to the context in which it is being said, both to what has been said before and to the 'context ol situation'; in other words, it has to be capable of being organized as elevant discourse, not just as words and sentences in a grammar-book or dictionary. It is the demands posed by the service of these functions which have moulded the shape of language and fixed the course of its evolution. These functions are built into the scmantic system of language, and they form the basis of the grammatical organization, since the task of grammar is to encode the meanings deriving tiom these various functions into articulated structures. Not only are these functions served by all languages, at least in their adult form; they have also determined the way human language has evolved. So when we study the language development of young children, we are really investigating two questions at once. The first concerns the language they invcnt for themselves, on the basis of the set of elementary uses or functions of languagc which retlect the dcvelopmental needs, potentialities and achievements of the infant - instrumental, regulatory and so on. The second concerns their transition tc the adult language, a language which is

However, the uniqueness of the individual, in terms of his personal xperience, must be qualified by reference to the culture. Our environment rs shaped by the culture, and the conditions under which we learn language :rrc largely culturally determined. This point is significant at two levels, one ,,1 rvhich is very obvious, the othe less so. It is obviously true in the sense I lrirt a child learns the language he hears around him; if he is growing up in an lrrglish-speakingsociety,helearnsEnglish.Thisisamatterofthe/ingrlic
, llvitonment, which is itself part

of the culture, but in a special


l\lorcover he learns that dialectal variety of English which belongs to his l,irlicular socioregional subculture: working-class London, urban middle, lrrss Northern, rural Dorset and so on. (He may of course learn more than ,rrrl tlialect, or more than one language, if the culture is one in which 'rr(hlinguisticdiversityisthenorm.)Itisequallytrue,butmuchlessobvious,in .rrroIls 5s5s namely that the culture shapes our behaviour patterns, and a 'rL :r I dcal of our behaviour is mediated through language. The child learns
l,r! llother tongue in the context of behavioural settings whee the norms of rlr( ( ulture are acted out and enunciated for him, settings of parental control, r r',tr rction, personal interaction and the like;and, reciprocally, he is'sociali., ,l into the value systems and behaviour patterns of the culture through rlr, rrsc rf Ianguage at the same time as he is learning it. \!lr cln now see th(] relevance of this to linguistic theories of educational ,rlrrr'. which were referred to briefly in the last section. Thefe has been ',rr, lr rliscussion of educability lately, and various theories have been put r u ;r r tl. C)ne school of thought has concentrated on the effect of the child's //,,,///\1( cnvironment - namely, the particular form of language he has ,', , 'rr rr rlr to speak. In practice, since educational failure is usually associated ' rrlr llrc Lrrban lower working class, this means the particular socioregional 'l,Ll((t; irnd we find two versions of the'language failure' theory here, ,,,,r, linrcs known as thc'deficit thcory' and the 'diflerence theory'. Accord,

still functional in its origins but where the concept of'function'has undergone a significant change: it is no longer simply synonymous with 'use', but has become much more abstract, a kind of 'me tafunction' through which all the innumerable concrete uses of language in which the adult engages are
givcn symbolic expression in a systematic and finite form. To what extent the individual child traces the evolutionary pth in moving from one to the other is immaterial; it appears that at a certain point he abandons it, and takes a lcap directly into the adult system. Be that as it may, he has to make the transition, and in doing so he caves out for himself a route that reflects the particular circumstances of his own individual history and experience. Geoffrey Thornton expresscs this yery well when he says tha the language which eacb child learns inhcritancc. It is an inheritancc because he is endowed. as a human being, with the capacity to Iearn language mercly by growing up in an cnvtronmcnt in which language is beiig uscd around him. lt is unitue, bccause .. . no two peoplc occupy identical placcs i an environmcnt where language lcarning is taking placc, and this must mean that the language Iearnt is unique to the individual. (Doughty e, a/. 1972, 18).
is a unique

rllr, rr|cs. Now this is not mcrcly nonsense; it is dangerous nonsense. ll,l',rtrnirtelyithasrarelybeenexplicitlydenied; probably because, as the \,,r, r rrirr cducator Joan Baratz put it, 'linguists. . . consider such a view of

,,,. r() lhc deficit theory, the whole dialect is simply delective; it Iacks ,,rr, ( \sential elements-it is deficient, perhaps, in sounds, or words, or


rrrrl'r' so absurd as to make them feel that nobody could possibly believe

I t,, ,,1 rl ir lcrchcr bclicvcs that there is, and that some or all of his pupils | , .rl' orr'. lhcrr. irs Flcdcrick Williams has very convincingly shown in his
r,, ,.rr.,lir

,,r l tl)crc li)re to re fute it would be a great waste of time ' (Williams 1970a, Lt I lrr'c is nr such thing as a dcficient social dialect. But, on the other

This takes us back to the perspective outlined in section 2. Biologically we are all alike. in so far as the language-learning capacity is concerned; we have this ability, as a specics, just as we have the ability ro stand upright and walk, and it is quite independcnt oI the usual mcasures of intelligcncc' in whatcvcr lirrm. Ecologically, on thc othcr hand, cach onc ol us is urritrrr. sincc


,trt,'rtir)( ir Anrcricirr scl\xrls. hc thcrcby predisposes the children to

l:rilrlc. l'ltis

is krrrwr as

tlre'stcrcotypc hypothcsis': childen, no

tlrru rrlrrlls. will core l() l)chilvc likc thc stcrcotype to which they are

lhc cnvironmcntl pilttern is ncvcr cxtctly rcl\.iltc(1, irrrrl orrt irrlivitlrrirl's


is rlevcr

l)r siro( ls ilr)()llt(.r's.

,,,, r1'rrcrl (Willirrrrrs I ()7Orr. r. lt. I i"i). llrr', tlrlrr lcrrrls rs irrto llrt rlilli r'rcr" vclsion ol lhc lhcory. ilccordingto l, lr lll( l)rol)l( rr r\ tl()l llrirl llrt clrilrl's sPcr'th is tlelicicrt hu( lhlrt il is I ll, rr ||l rlill( rr'rl. |ll rrrr|)Ir(;rIrorr, lrrrr sorrt 'r'r'r'ivtrl sll||lrlirrtl 0t ltrtr.


The sociolinguistic perspective

Language and social man (Part



This would obviously be important if it meant that the child did not understand the language in which he was being taught (as happens with many immigrant children). But for the native English-speaking child, this is not the problem. Wherever he comes from, and whatever section of society he comes from, the speech differences are relatively slight and superficial, and in any case he has heard the teacher's language frequently on television and elsewhere, so that he never has more than very temporary difficrrlty in understanding it, and in fact is usually rather competent at imitating it - an activity, however, which he tends to consider more appropriate to the ptayground than to the classroom. So the difference theory resolves itself into a question of prejudice: if the child fails as a result of differences between his language and that of the school, it is not because there are difficulties of understanding but because the child's variety of English carries a social stigma: it is regarded by society as inferior. If'society' here includes the teacher, the child is, effectively, condemned to failure from the

distinct linguistic forms or codes and re.re cod.es essentally trunsmt the culture

n d so constran behavour. [His italics.]

ll wc accept that, as the American sociological linguist William Stewart , r I[cssed it, 'so much of human behaviour is socially conditioned rather than 1,r'rrctically determined', it is not difficult to suppose an intimate connection l,, r\\cc Ianguage on the one hand and modes of thought and behaviour on ilr( olher.
lris view is associated first and foremost with the work of the great Whorf, who wrote 'An accepted pattern of rr'.rrrq words is often prior to certain lines of thinking and modes of 1,, lrrviour.' Whorf emphasized that it is not so much in 'special uses of l.rrrllrrrge'(technical terms, political discourse etc.) as'in its constant ways of rr rrrrr[ing data and its most ordinary everyday analysis of phenomena that ,r, recd to recognize the influence [language] has on other activities, culr ir I irnd personal' ( 1956, 134-5). Bernstein (1971, 123) points out that, in ',r \\ lr,,r I 's thinking, 'the link between language, culture and habitual thought , ,,,t nrediated through the social structure', whereas his own theory
\ rrr'r'ican linguist Benjamin Lee
l,lirccs the emphasis on changes in the social structure as major factors in shaping
r r hanging a given culture through their effect on the consequences of fashions ,,1 spcaking. It shares with Whorf the controlling influence on experience r\( r ihcd to 'frames of consistency' involved in fashions of speaking. It differs I tr, r rrr l Whorf by asserting that, in the context of a commori language in the sense ,,1 ;r gcneral code, there will aise distinct linguistic forms, fashions of speaking, s lri.h induce in their speakers different ways ofrelating to objects and persons.

To that extent, then, the difference theory, unlike the deficit theory, is at least partially true'. tbere are prejudices against certain varieties of English, and they are shared by some teachers. But they are by no means shared by all is difficult to believe that this factor by itself could be explanation of the full extent of educational failure, especially sufficient since children have a great capacity for adaptation - ifone form of behaviour does not pay off they will usually switch to another, and they are quite capable of doing so where language is concerned. Moreover the prejudices are gctting less, whereas the general view is that educational failure is
teachers: and


we return to this discussion in chapter 5 below, with reference to the work of Basil Bernstein. Educational failure is really a social problem, not a linguistic one; but it has a linguistic aspect, which we can begin to
understand if we consider the cultural environment in the second of the two senses mentioned above. It is not the linguistic environment, in the sense of which language or dialect the child learns to speak, that matters so much as the cultural or subcultural environment as this is embodied in and transmitted through the language. In other words, the 'language difference'may be significant, but if so it is a difference of function rather than of form. It is this fundamental insight which lies behind Professor Bernstein's theoretical and empirical work in the field oflanguage and society; together with a further insight, namely that what determines the actual culturallinguistic configuration is, essentially, the social structure, the system of social relations, in the family and other key social groups, which is characteristic of the particular subculture. Bernstein (1971, 122) writes:

Ir,Iir,,r(i hasinvestigatedowthisconnectionismade,andsuggeststhatit

rlrr,rrrgh linguistic codes, or fashions


of speaking, which arise as a conn(c of the social structure and the types of social relation associated
As Mary Douglas put it, 'The control [ofthought] is not in the speech relations which generate thought and speech'
i I

i 'rlrrr llr', lr l in the set of human ,r,,' ll2).




lr, r I



,1, r, n(el)ctwcenonetypeofcodeandanother; it is the relative emphasis I L, , rl orr thc diffcrent functions of language, or, to put it more accurately,

[c these linguistic codes, or fashions of speaking? They relate, cssenlunctional interpretation of language. It is not the words and the s t [cture s - still le ss the pro nunciation or 'accent' - which make the

r, lLr(l\ol'rrcaningthatarctypicallyassociatedwiththem.The'fashionsof

I rl.rIllt rrc

, lr, ,*,r'irrlizrrliorr ol thc chiltl in thc lamily.

r,( nrol'c()t'lcssstrrl gly.inparticularcontexts,cspeciallythoserelating

socioscmrntic in naturc; they are patterns of meaning that

Hcncc, although each child's

A nrmber of fashions of speaking, famcs of consistcncy, arc possiblc in any

given languagc and . . . these fashi()ns oI spcakirlS. Iingtr\li( lirrtnsr crrcles, arc l() lhisvicw. thcmsclvcs lirncti()n(ri 1hc li)rrn s()ciitl r(litlirtts lirl r' ^((,)lrlirllt ,lrrrrlrrr('li('rl!.lill(\ lhcli)rrrrol llle stxiitl rclillit)l) ()1. nrr)1.,1( rr( rirllY, lllr'rr,r,'l

r,., r.rll( l(iltri,llt elrvirortell is uniruc, he also sharcs certain common r ,trrr,,,IIIroIII(r r.lriltlrt.rrrll lt silttilirrsocil blrckgrounr-l; not merely in the ,, , rlr, r;rl s( rl\(. llrlll lll(, liltr.riitl (.r]vit()nllcnls rtriry wcll bc illike in fact 'r, \ nr.r\ rrr)l l)ul in tlt( rltr'rtr strrsc lhtrt tlre Ii rrrs ol'sociirl lcllrliolland 'lt r,,l( ,\',1( lt\\uloln(lr)i lrirrr lr:rvr. llrt.i (lli.t ()n tlt(.kirrtl ol clloiL.tsilr


The sociolinguistic perspectrve


Language and social man (Part



meaning which will be highlighted and given prominence in different types of situation. Peter Doughty comments: 'the lerms elaborated arld restricted refer to characteristic ways of using language to interact with other human beings; they do not suggest that there are two kinds of "meaning potential"' (Doughty et al. 1972, 104 5). This dependence on social structure is not merely unavoidable, it is essential to the child's development; he can develop only as soclal man, and therefore his experience must be shaped in ways which make him a member of society and his particular section of it. It becomes restrictive only where the social structure orients the child's thinking away ftom the modes of experience that the school requires. To quote Bernstein again, 'the different focusing of experience . . . creates a major problem of educability only where the school produccs discontinuity b(:tween its symbolic orders and those of the child' (1971, 183-4). In other words, the processes of becoming educated require that the child's meaning potential should have developed along certain lines in certain types of context, especially in relation to the exploration of thc environment and of his own part in it. To what extent this requirement is inherent in the very concept of education, and to what extent it is merely a feature of education as it is at present organized in Britain and other highly urbanized societies, we do not know; but as things are, certain ways of organizing cxperience through language, and of participating and intcracting with people and things, are necessary to success in school. The chiltl who is not predisposed to this type of verbal exploration in this type of cxpcricntial ancl intcrpcrsonal context is not at home in the educational wrr'lrl'. as llcrnstcin puts it. Whether a child is so predisposed or not turns ()u( l()t to bc any innatc property of the child as an individual, an inhcrent linr itation on his mi:ntal powers. as used to be generally assumed; it is merely the rcsult rf a mismatch between his own symbolic orders of meaning and thosc of the school, a mismatch that results from the different patterns of socialization that characterize different sections of society, or subcultures, and which are in turn a function of the underlying social relations in the family and elsewhere. Mary Douglas says of Bernstein that he asks 'what structuring in society itself calls fbr its own appropriate structures of speech' (1972, 5); and she goes on to add 'A common speech form transmits much more than words; it transmits a hidden baggage of shared assumptions', a 'collective consciousness that constitutes the social bond'. It is all too easy to be aware of subcultural differences in speeci forms, because we are all sensitive to differences of dialect and accent. Unfortunately this is precisely where we go wrong, because differences of dialect and accent are in themselves irrelevant; in Bernstein's words, 'There is nothing, but nothing, in the dialect as such, which prevents a child fronr internalizing and learning to use universalistic meanings' ( 1971 , 199), antl dialect is a problcm only il it ismade a problem artificially by thc prejudicc and ignorance of othcrs. It is nruch hardcr to bcconrc rwrrr. r ) I l llr !1(//i( /rl diffcrcnccs. which arc rnaskcd by rliirlcclr v:rrirlion (irl(ls lrir lr lrv n(J nrciuls lwrtys cottcsrortrl lo rliirlccl rIisIitrtliotts), rtttrl u lt, lt rlo t,rt rr|rr'rr i Ilt

rbvious form of differences in vocabulary or grammatical structure. We are rtill far from being able to give a comprehensive or systematic account ofthe lrrrguistic realizations of Bernstein's codes or of the ways in which language ,)l)crates in the transmission of culture. But the perspective is that of lanlrrrgc and social man, and the functional investigation of language and lrrrrguage development provides the basis for understanding. Irr cssence, what seems to happen is this. The child first constructs a l,rr ruage in the form of a range of meanings that relate directly to certain of lrs lrrsic needs. As time goes on, the meanings become more complex, antl lr, rcplaces this by a symbolic system - a semantic system with structural r, ;rliations-basedonthe language he hearsaroundhim; thisiswhatwe call rln\'nother tongue'. Since this is learnt, and has in fact evolved, in the ., r v icc of the same basic functions, it is, essentially, a functional system; but r, llrnctionality is now built in at a very abstract level. This is what was ,,lt u'ctl to at the beginning of this scction, when I said that the adult lrri,ristic system has, in effect, the four generalized functional components, ,r 'r r'tafunctions', experiential, logical, interpersonal and textual. These l, ,r rr r I hc basis for thc organization of meaning when the child moves from lrr. ,rIiginal protolanguage into language proper. I t rr hc does not abandon the original concrete functional elements of the


nr ns he

rl: rrnd out of them evolve the social contexts and situation types that ,r.,l,, rp the patterns of use of language in daily life including those , ,,,rr1 \rs tlrat Bernstein has shown to be critical in the socialization process. ll, r.rrr lies the basis of the significant subcultural variation that we have r,, , lr lr)oking ai- fr whkh particular contexts of use will the child bring to t" .tt tlich portions of the functional resources of the system? Seen tiom a 1,,,r'rr\tie point of view, the different 'codes', as Bernstein calls them, are l llr !( nl strategies of language use. All human beings put language to , rl.un typcs of use, and all oI them learn a linguistic system which has ,,,lrrrl irr that context; but what aspects of the system are typically l, t,l,r\'((l irrrd cmphasized in one type of use or another is to a significant r, It (l( ternlincd by the culture-by thesystemsof social relationsinwhich rl,,,lrlrlAr()wstrp,includingtheroleshehimselflearnstorecognizeandto , 1,,,1 :\llchildren ltavc access to the meaning potential ofthe system; but ,1,, \ rr\ (lillc, lrccause social groups differ, in their interpretation ofwhat ,lLL rf itti()n ilcntntls.
,r ,

invented it. These still define the purposes for which language is

-, Lrrgu:tgr irrd situation

l,,l,lr('r)irow trr, rrrl I ltc ir l rrgrurgc grows up with them. By the age of two ,,,,1 r lr.rllor evr.t crrlic. tlte chiltl lr s nr aste rcd the adult langu age system; ,l' lr.rrlwo[isrrllllrt.'t.. llt.willsrcrrrlllrclcstol hischildhood-therestof

lrrl,u,tll',il\$altitvc\ltcssc(l,islt|ol(nliirl:il is wltirl I ltc sltcitkcr can tlo. \\ l, rr ,r ln t.,)ll r'trrr ,lr, lrr tlrt ltnlrrrslit scIl\r, lllltl is wlIl lre clrrt rio rs

,, lrll' r.rt.rr

rrrtrstr.titrl llrr. lrtlrrll /arrgrirr,qr'.


The sociolinguistic perspective

Language and social man (Part



speaker/hearer, is equivalent to what he 'can mean'; hence the description


language as a 'meaning potential'.

To dscribe language as a potential does not mean we are not interested in the actual, in what the speaker does. But in order to make sense of what he does, we have to know what he can do. This is true whatever our particular angle on language, whether we are looking at it as behaviour, or as knowand led--ge (Ctromity;s 'competence'), or as art: what iJ, the actual sentences

woids that constitute ur direct experience of language, derives irs significance from whal could be. But it is in the social perspective that we are best able to explain what s, because we can pay attention to situations of language use, tking account of the nonlinguistic factors which serve as the conlrolling environment. lt is at least theoretically possible to look at the 'actual' in isolation from the social context (so-called 'theories of performance'), but it has not yet been shown to be very fruitful.
When we come to examine the adult language in its contexts of use, we at once run up against the difficulty that the one thing we cannot specify is the 'use' ol any given utterance. Nor can we enumerate the total set of Possible uses for language as a whole. We cannot draw up a general list setting out the adult's uses of language, in the way that we were able to do for the developmental functions in the language of the very small child. Or ather - which amounts to the same thing - we could draw up a hundred and one such lists, anrl there would be no means of preferring one list over another. Then when we came to consider actual instances we should have to recognize that in any particular utterance the speaker was in fact using language in a number of ifTerent ways, for a variety of different purposes, all at the same time The use of language is not a simple concept.

Neverth;less it is a very helpful one, without which we cannot explain either the variation we find within a language - the different styles, levels of fbrmality and so on - or the nature of language itself. The latter is outside our scope here, although we referred in the preceding section to the inherently funttional organization of the linguistic system. But the former is fundamental to any consideration of language in an educational context The ability to control the varieties of one's language that are appropriate to different uses is one ofthe cornerstones oflinguistic success, not leastforthe school pupil. (For an excellent discussion of'differences according to use', see Doughty et al. 1972. ch. 11,'Diversity in written English'by John
Pearce.) The basic concept here is that of'context of situation', originally suggested by Malinowski (1r23) and subsequently elaborated by Firth in his 1950 paper'Personality and language in society'(see Firth 1957, ch. 14). Esseniially what this implies is that language comes to life only when functioning in some environment. We do not experience language in isolation if we did we would not recognize it as language - but always in relation to a scenario, some background f persons and actions and everts lirrt wllich thc things

whiclr itrc snicl dcrivc thcir lneaning. This is rclcrt'ctl lo ls lllr'sillllti()ll', so lrrgrItgc is siritl to lirrction il'c()nlctls ol silrIrtiot rttrl rtrt! rtttortll ol

,,,.rrrrr,.s.Ashesays,'certaingroupsofchildren,throughthefbrmsoftheir ,,, r.rlr,,irlion, lre oriented towards receiving and offering universalistic ,,1, .rrJt\ it\ (crt oitr contexts' ( 1 971, 196). This in itself is not important; but ,r I','(()les inrportrnt if there are certain rype.r of situation which play a ',,rlrirl|rlinthccltild'stotal development, since these are theoneswhere ,,, ,,"",, lo usc lirrguagc in ways that are least dependent on the here and ]l:,,i ll., ltrtls us l() lhc notion ol r situation type. Looking at how people ,, ru,rllY rrsr.. l rrgrrirge in ililily Iilc. wc fird that the apparently infinite ,,rrl!( r ()l tlillelcrrl rrssihlc slultlions rcprcscnts in reality a very much rrr,rllr'r llrllrlrer ol ,]r.rrelirl/f/r..r.tl silrrtior. which we can dcscribe in such r, rr.,:t\'l)lity(ts il\ll(lilfl lrovicc il it Ailnc',,r)l()thcr rcading bcdtime l,,rt lo r'lrilrl . '(ltsl()ll(.1 orrlcrirri rxrrl: ovcr the lclel)h()rc', '(clchcr ,,rr,lrrrl' IUIils' rlisr'rs..iol (]l it l)()(,nr' rrrrrl tlrt. likc. N()l illl tllcsc situirti()ll

l.,tltuage which lails to build in the situation as an essential ingedient is l,l.( ly to be artificial and unrewarding. ll is important to qualify the notion of'situation'by adding the word ,, l(. vrnt.' The 'context ol situation' does not refer to all the bits and pieces ,,1 tlrc ruaterial environment such as might appear if we had an audio and ' r, l, o Iccording of a speech event with all the sights and sounds surrounding I r( ultcrances. It refers to those 1'eatures which are relevant to the speech rl,.rl is taking place. Such features may be very concrete and immediate, as rlr \ lcnd to be with young children whose emarks often bear a direct I r .Itrrirtic relation to the environmen t, e.g.some more! ,l want some more of '\ lr,rt I'vc just been eating.' But they may be quite abstract and remote, as in r rr r lrrrical discussion among experts, whee the 'situation, would include ,,, l l lr ings as the particular problem they were trying to solve and their own r, rnrrg and experience, while the immediate surroundings of objects and , \ , rt\ would probably contain nothing of relevance at all. Even where the ,,,r'lr tloes elate to the immediate environment, it is likely that only , , r,ll leatures of it will be relevant; tbr example, is it the presence of a t,,r lr('ular individual that matters, or is it a certain role relationship, no ,,,rr( r who is occupying the roles in question? IfJohn saysl love you,it l,r,,'l nrbly does matter that it was said to Mary and not Jane; but if he says | 'tt \t,u l) u t up a prescription me?, what is relevan t in that situation is the ,,, ()l (lispensing chemist, Jbr and not the identity of the individual who l, rl't,( r\ to be occupying it at that particular time and place. lrr r,1 ,",-,,, the ability to use language in abstract and indirect contexts of ru.rtr()r is what distinguishes the speech of adults from that of children. I , .r n irA language consists in part in learning to free it from the corstraints ',1 rlrl irrnrediate environment. This process gi?s veryearly in lite, when 1', , lril(l lirst learns to ask for things that ae not visible and to recall objects rr, I r'r < rts which he has observed earlier. But it is a gradual process, which | ,l , . l)lircc in different ways with different children; this is one of the ,rr.rlrlts which Bernstein has found to be significant - which rypes of !ru,rt()r scrve as thc gateway to more abstract and generalized contextual


The sociolinguistic perspective

Language and social man (Part



in the types are equally interesting, and some are obviously very trivial; but on kin<i depends this of category abstract f any ilt ,..or, the iiportance of'context the notion of significance the and of it, make what we are going to oi iituation'-for ihe present discussion is that some situation types play a

il a crucial role in the chlld's move into the adult language' For example' as toy' such mother or father is playing with a child with some constructional some contain to is likely a set of building bricis,-this type of situation remarks of guid lce and explanalion, with utterances like I don't think that
one will go

',rtrtion where it simply has to be taken for granted that tbr every child, by rlr( lime he arives in school, Ianguage is a means of learning; and this is an .rrsrrmption that is basic to the educational process. Less obvious, but ', r'lraps no less fundamental, is the assumption that language is a means ol ln rsonal expression and participation: that the child is at home, linlrrirtically, in interpersonal contexts, where his early use of language to l'rrr'.ict with those emotionally important to him, and to express and ,ll r e lop his own uniqueness as an individual (the interactional and personal

.'lhe context of situation for this utterance is instruction relating to his handling of gaining is child n. in i,,t rcll the .,i."o- nn although any o-ne insiance is not by itself going to make much be highly Jiii.i.".., an oc.li.uluiion o[ exeriences of this kind may just to /his nlrt late re re marks ,lgritrcunt. ana it it regularly happe ns that the towerto but bricks, particular with these pirticular lower that iSbeing built Lrifing rn general - in other words, if the context of situation is not limited to tt i.tril physical surrounclings, but extends to more general and less " environments, at would be implied by a remark such as /ft immecliate

in there;


too wide


rrctions), has in the same way been taken up and extended into new ealms rrclrning. No doubt both these assumptions are true, as they stand: every r,,r rrirl child has mastered the use of language both for entering into per-

i,nuller ones have to 8o at the top - then language is now serving a primary have a function in this aspJct of the child's development Hence-he rvill

about sirong sense of this use of language, of language as a means of learning it and with interact to ability own environment arid about his irr.

,,rr:rl l'elationships and for exploring his environment. But the tixd of rr( irr)ings that one child expects to be associated with any particular context ,,1 \lrtion may differ widely from what is expected by another. Here ,r, :rlc back to Bernstein's codes again, which we have now approached r,nr irnother angle, seeing them as diflences in the meaning potential $ lrrr lr nray be typically associated with given situation types. As we have , , r. tl)cse diffeences have their origin in the social structure. In Ruqaiya I I r,,; r rr's words, 'the "code" is defined by reterence to its semantic propefties' ,r,l llrc semantic poperties of the codes can be predicted from the elements

f'rlyti"or control - :t1,. it. typc. of situation which seem to be most central to the child's socializiition ltrvc t.reen identified by Bernstein in the most general terms 'context' in the sense I lc rc lcrs l() tlrc tr! lls 'critical socializing contexts', using ol ir gcrrcratizccl situatio1 typc. He identifies tlre 'regulative' context' 'where ,lr" itrt.l i, ntitclc aware of the rules of the moral order, and their various trackings'; thr: i'lstl uctionIl' context, 'where the child learns. about the ni.,".ti," n"trr" nf objects and persons, and acquires skills of various kinds'; to : iirJ unagin at ive or innovating' context, 'where the child is encouraged 'interthe and terms'; own his on worlil hii re-create and experirnnt p"i.ona' context, 'where the child is made aware of affective states - his already i*n, an,l others' (Bernstein 1971, 18i, 19tt) These turn out to.be has first child the which through anticipatea in the developmental functions reguinstrumental' the his own: of starte; to build up a linguistic system iuiory ona ro on, escribid in section 3 above Fo example, those types of situaiion which involve explanation and instruction, Bernstein's 'instruciionul .ont"rt', typically pick up the developmental thread that first aooeared in the iorm ol i 'heuristic' function, the child's early use of irigu"g. to.,,ptore his environment. They are therelbre critical also in the cfria'searning ot/ anguage, since it is through using language in situations ol these types th;t he builds on and expands his meaning potential' This is where the notions of context of situation, and situation type' clrilrl slroultl t e.o,re ifnpo,tunt tbr the school. The school rcquircs thill thc olrviottsly' lltat hc lltosl ol illl Iirst ways: ccrtllit't i la,rg,ragc bc ablc to ri.c ( x l s ( l' ( ( r l l \ r olx l il l l slrrttltl hc itlllc lo tlsc lllEllilAe l()leltl'' l'ltcltilcllcl '
l l )


,,,)riitl structure which, in fact, give rise to them' (Berstein 1973, ,,li) N,, (he very young child, in his first ventures with language, keeps the r,i,r, trr)ns of language fairly clearly apart; when he speaks, he is doing only ,,r, llirrg at a time - asking for some object, responding to a greeting, ' ,t,r, rsing rterest or whatever it is. When he starts learning his mother r',,,1'u( . l)owever, the contexts of situation in which he uses it are already ,, ,rl,l,. r lrd rranysided, with a number of threads of meaning running ,,rrrrlt:rrrcously. To vary the metaphor, we could say that all speech other r l'.o r tlrC l)rotolrrguage of infancy is polyphonic: different melodies are kept , ,,,r, sirlc hy side, and each element in the sentence is like a chord which ,rrrrl)rles sornething to all of them. This is perhaps the most striking ' l r rL lcristic ol human language, and one which distinguishes it from all ,tlr, r .,t,rrrlxrlic conmunicrtion systems.

r, llrIislc'
I 1,, . Ll1l lx )in l is n rcllcction ol the contexts of situation in which language js ,, , ,l ,rLl llre wlrys in which onc type ofsituation may differ from another. t 1, , ol lrrrgrristic silr.irtion dillr Iront one another, broadly speaking, in rl,r, I r( \l)(cts; lirsl. wlriil is actually taking place; secondly, who is takinB ,,rt ,rrrrl llrirrlly. wltirl I)irrl llrc lirnguirgc is playing. These three variables,

r,lrt(,lttlrcr,(l(lcrnrirrclltcl'irrrgcwillrinwhicltnrcaningsareselectedand rl,, l,,rr\ wlriclr irt ustrl lir lltcir cxlrcssion. Ir othcr words, thcy dctcr,irr, tlrr' rr.l,isl(,r'. (S.. lirl)l( I,l). 15.) IIr, Irot|r)|tr,I r(,1rslr'r r\illortr'c vr'rysirtrrL irttrIve rylrwr'r'lirI llrclt'rsl()


The sociolinguistic perspective


Language and social man (Part



the fact that the language we speak or write varies according to the type of situation. This in its;tf ia no more than stating the obvious What the theory of register does is to attempt to uncover the general principles which govern this riation, so that we can begin to understand waf situational lactors determine rvat linguistic features. It is a fundamental Propey of all languages that they display variation according to use; but surprisinglv little is fno*n nbout th nature of the variation involved, largely because of the

!et difficulty of identifying the controlling factors An excellent exampie of register variation (and ol how to investigate and describe it) is providd by Jean Ure in a paper entitled 'Lexical density and register.liifur;ntiation' (1971). Here.fean Ure shows that, at least in Englisi, the Iexical density ol a text, which means the proportion of lexical items

(content worcls) to words as a whole, is a function first of the medium (that is, ivhether it is spken or written - written language has a higher lexical density than speech) ncl, within that, of the social function (pragmatic Ianguage, or 'language ti action', has the lowest lexical density of alt) This is probably true'of ll languages; but whether it is or not, it is a basic fact about English and a very god-illustration ol the relation between the actual and the potential thal we referred to at the treginning of this section W could say' iollowing Dell Hymes, that it is part of the speaker's 'communicative compctence' that he knows how to distribute lexical items in a text according to dilLre nt kinds of language use; but there is really no need to introdrrce here lhc artiliciat concept;f to-peten."', or'what the speaker knows', which nrcrcly udrls an cxira lcvel of psychological interpretation to what can be exrlirincrl rtrorc sitnply irr direct sociolinguistic or functional terms it is casy to bc mislecl hcre by posing tlle question the wrong way, as a ,r,,urlrcr. ,,i writcrs on thc subject have clone. They have asked, in effect, 'whxt lcaturcs ol languagc are determined by register?', and then come up with instances of neat'-synonymy whete one word diffes from another in

lrring to processes of cultivation; and this is one aspect of the relation of situation the subject matter of gardening is part of the social ( ontext. But, in fact, the probability of such tems occurring in the discoursc r\ itlso dependent on what I and my interlocutor are doing at the time. If we r re actually engaged in gardening while we are talking, there may be very lcw words of this kind. Jean Ure quotes an amusing example from some Itrrssian research on register: 'The recording was of people frying potatoes, ,rrrLl lrying potatoes was what they were talking about; but since, it seems, rr, it her frying nor potatoes we e represented lexically in the text, the recordLnr was a mystification to all who had not been in the kitchen at the time.' I lrc image of language as merely the direct reflection of subject matter is ',rrrrlistic and unsound, as Malinowski pointed out fifty years rgo; thcre is ll r r rch more to it than that, and this is what the notion of register is all about. Wlrat we need to know about a context of situation in order to predict the lI[Lristic features fhat are likely to be associated with it has been sumnr;r'izcd unde thee headings: we need to know the 'field of discourse', the r( ror of discourse' and the 'mode of discourse'. (See Halliday 1 a/. 1964, rr lrr'rc the term 'style of discourse' was used instead of'tenor'. Here I shall ,rr'li'r the term'tenor', introduced by Spencer and Gregory (Enkviste//. lr)r)t). A numbe of other, more or less related, schemata have been ,,,,xrsed; see especially Ellis 1965, 1966; Gregory 1967.) John Pearce .rrrrririzes these as follows (Doughty et al. 1912. lE5-6)l
lillrguage to
I icld refers to the institutional setting in which a piece of language occurs, and

.[]brces not only the subject-matter in hand but the whole activity of the .l)crker or participant in a setting fwc might add: 'and of the other partici-

level of forrnality, rhetric or technicality, like 'chips' and 'French-fried potatoes', or 'deciduous dentition' ancl 'milk teeth'. But these are commonplaces which lie at the fringe of register variation, and which in themselves would hardly need any linguistic or other kind of'theory'to explain

li r)or . . . refers to the relationship between participants . . . not merely var iarion Ir lonality . . . but . such questions as the permanence or otherwise of the ri lirtionship and the degrcc of emotional charge in it. . . . \l(xlc relers to the channel of communicatio adopted: not only the choice
,( trvccn spoken and written medium, but much more detailed choices lwe might :r,lrl: 'nrl othcr choices elating to the role of language in the situation']. . . . I lrcsc


them. Asked in this way, the question can lead only to trivial answers; but it is the wrong question to ask. All language functions in contexts of-situation' and is relatable to those contexts The question is not what peculiarities ol' vocabulary, or grammar or pronunciation, can be directly accounted for by refeence to thi situation. lt is whch kinds of situational factor deterrnine wlcl kinds of selection in the linguistic systcm. The notion of register is thus a lorm of prediction: givcn that we know the situation, the social context ol Ianguage use, we can predict a great deal about the lanluage that will occur' wit-h re-asonatle probability of being right. The important theoretical qtrcstion then is: wht exactly do wc neeil to know about thc s('cial contcrt irt ocler to make such Predictions? I-ct r.s lnilkc th is lrl()re c()llcl-clcll I itrtr titlliitt'' itlrolll I r t t t r I ' I tltlty lrr' ttttlt.c likcty ltl ttse tvtrtt|s lllill ill't lltc llilllt('\ ()t l l r r l . r l . l rllllt.lr rvrlrll:

rrtc thc general concepts needed for describing what is linguistically

llrli(ir]t in the context oI situation. They include the subject-matter, as an , t,, ( tolthe'liclclofdiscourse' of t he whole setting of re levant actio ns and , ,, rrs wilhin which thc languagc is functioning tbr this is where subject ,.Lrr( r l)cl()rgs. Wc do not, in [act, Iirst decide wl]at we want to say,



, r

,rr,l,r'rrrlcrrtlyollhcscttng,andfhcndressitupinagarbthatisapproprate l,rl ll r llrc c()rlc xl, ls sonrc writcrs cln language and language cvents seem to ,. ,llrr'. llrc'c()ntcnf is l)irrl ol thc total planning that takes placc. There is ",, ,lr-:rr lrrre l)clw('er llrc 'wlrrl'irnd 1hc 'how'; all language is language,,r r,( , llr ir c() lL \l {)l silurli() , rrrtl rrll ol it rclatcs to the situation. in the ,l,.tr.rr't \(ns( rr rvlri,.lr l:llr rrsirrr llre lcl1l l]crc. | .,lrrrulrl lrr'( rrtirkr' rr rrssirr, r'lr'rt rrr't lo tlilrlecls, whiclr iuc parl ()l tl]c tLr tlllf ol lrrrr'rrrrl,L rlltrlror'irrl rrrt. irlllr(llrlllr rrol rrirttrrrilv telcvrtttl itt lltc


The sociolinguistic perspective

Language and social man (Part



educational context except as the tbcus of linguistic attitudes Our language is also determined by who we are; that is the basis of dialect, and in principle a dialect is with us all our lives - it is not subject to choice. [n practice, however, this is Iess and less true, and the phenomenon of'dialect switching' is widespread. Many speakers learn two or more dialects, either in succession, dropping the first when they learn the second, or in coordination, switching them according to the context of situation. Hence the dialect comes to be an aspect of the register. If for example the standard dialect is used in formal contexts and the neighbourhood one in intbrmal contexts, then one part of the contextual determination of linguistic teatures is the determination of choice of dialect. When dialects come to have different meanings for us, the chice of dialect becomes a choice of meaning, or a choice between different areas of our meaning potential. Like the language of the child, the language of the adult is a set of socially-contextualized resources of behaviour, a 'meaning potential' that is related to situations of use. Being 'appropriate to the situation' is not some optional extra in language; it is an essential element in the ability to mean. Of course, we are all aware of occasions when we feel about something said or written that it might have been expressed in a way that was more appropriate to the task in hand; we want to'keep the meaning but change the wording.'But these are the special cases, in which we are reacting to rurcly conventional features of register variation. ln the Iast resort, it is irnpossiblc to draw a line between'what he said'and'how he said it', since th is is bascd on a conception of language in isolation from any context. The (listirclion between onc rcgister and another is a distinction ofwh is said as nruch ils ollr.w it is said, without any enforced separation between the two. ll a scvcr ycar-old insists on using slang when you think he should be using morc lirrr)irl language, this is a dispute about registers; but if he insists on talking about his football hero when you want him to talk about a picture he has been painting, then this is cqually a dispute over registers, and one which is probably much more interesting and far-reaching for both teacher and

l,rilliant paper published twenty years ago by T. F. Mitchell, called 'The lrrnguage of buying and selling in Cyrenaica' - though since the language ',lr(led was Cyrenaican Arabic and the paper was published in a learned t,,ru nal in Morocco. it was not at first widely known (MitcheU 195 7). But for r, sclrch of this kind to be relevant to a teacher who is professionally ,,rrccrned with his pupils' success in language, it has to relate to social ( ,)rtcxts that are themselves ol significance, in the sort of way that Bernt( ir's 'critical contexts' are sign i|cant fbr the socialization of the child. The

l() l)e a 'social' account of language but hardly a 'sociological' one, since the {1)ncepts on which we are drawing are not refered to any kind of general '(}cial theory. Such an account can be very illuminating, as demonstrated in a

, r rtcria would then be sociological rather than simply social - based on some tlrco ry ol social structure and social change. In this respect, the earlier terms

lrl,e llirth's'sociological Iinguistics', or'sociology of language' as used by It( rnsteir, are perhaps more pointed than the currently fshionable label ., rr'iolinguistics'.


Vareties in language

I)lrlect ( dilectal variety') variely 'according to the usea l,rloct isl ^ wlri you speak (habitually) l,,lormined by who you are (socio-region t)l origin and/or adoption), and ,,rt,rL,ssing dversity of socia structure (t)irlterns of social hierarchy)

Regsler ('diatypic variety')

variety 'according to the 0se'

A register is: what you are speakng (at the time) determned by what you are doing (nature

expressing diversity

of social activty being engaged in), and oI social process (socal division of labour)

.,, ir trrinciple dalects arel r',r,{,,nl wavs of Saying lhe same lhing ,,ilrr,Ild to differ in:
Iilr,x)tics, phonology, lexicogrammar (but

pupil concerned. Thus our functional picture of the adult linguistic systern is of a culturally specific and situationally sensitive range of meaning potential. Language is the ability to 'mean' in the situation types, or social contexts, that are generated by the culture. When we talk about'uses of language', we are concerned with the meaDing potential that is associated with particular situation types; and we ae Iikely to be especially interested in those which are of some social and cultural significance, in the light of a sociological theory of Ianguage such as Bernstein's. This last point is Perhaps worth stressing. The way that we have envisaged the study of language and social man, through the concept of 'meaning potential', might be referred to as a
kind of 'sociose mantics', in the sense that it

rr)l in semantics)

So in principle registers are: ways of sayg different lhngs and tend to differ ini semantics (and hence in lexcogrammar, and somelmes phonology. asrealization of ths) Exkeme cases:


rlrlirrrquages, molher-in-law languages

restricted IangUages, languages for

specia! purposes


,rl rlslanceS:

.,rlr ullllral varielies (standard/nonslan-


Typical instances: occupatioa varieles (lechnical. semitechnical) Principal controlling variables: field (type of socal action); tenor (role relatioships); mode (symbolic organzation)

| ,,,,, rt),rl oonkolling variables: ,,,rirl (il;rss, caste; provenance (rural/

rl)iUr). qcneration; aqe; sex

the study of meaning in a social

or sociological framewok. But there is a diffcrcnce lrctwce tr 'social' antl 'sociological' here. lf wc tlescrilc thc contcxl ol si(llllli()!l it lr.'r lts ol r1ltx' obscrvrrliorts rtbortt thc sellirlSs itt wlticlt litttgrr:tlt ir ttscrl. lltr lorrlrlle sltitl


l,i ,,{nl


lr rrllV lxrkl rlrlll(lo:r l(,witr(ls (liitkx)ls s '.yllrlx )l ol rxx:Ill rlivrlllitly

Characlerized by: major dislinclions ol spoken/writlen; languaq(! in aclion/lanquage in reflection