A5 Pea 80 Negation

8/22/2019 A5 Pea 80 Negation

1/32

TheSocial Foundations ofand Thought

J E R E M Y M . A N G L I NT . BE RRY H RA ZE L. I O NR O G E R B R O W NR O S A L I N D C H A R N E YM I C H A E L C O L EJ A C Q U E L I N E J . G O O D N O WP A T R I C I A M. C I K E E N F I E L DP E G G R I F F I NB A R B E L I N H E L D E RK E N N E T H K A Y E

A L A S T A I R M U N D Y - C A S T L ED A V I D R . O L S O NR O Y D . PEAJ E A N PIAGETJ A N S M E D S L U N DH E N R I T A J F E LC O L W Y N T R E V A R T H E NE D W A R D T R O N I C KP E T E R C . W A S O ND A V ID J . W O O D

With a foreword by George A . Millerand an afterword by J e r o m e S. Bruner

W . W . N O R T O N & C O M P A N YN E W Y OR K L O N D ON


2/32

T h e Development of Negationin Early Child LanguageRoy D. ea

While the child's task of learning words seems relatively straightforward ithas always bedeviled language theorists. In fact, in having to construct wordmeanings from words used in conversational settings replete with actions,gestures, objects, and events, the preverbal child faces one of the most d f i -cult yet critical problems of life. Empiricist accounts of the acquisition ofwo rd mea ning, such as Quine's ( 1 9 7 4 ) , assum e the adequacy of an ostensiontheo ry, by which w ord s paired with objec ts provide th e child with the refer-ential connection required for language symbolization. Compelling argu-ments have been given against ostension as the source of word meaning(B run er, 1974-75; Harrison, 1 9 7 2 ) , but empirical considerations weighagainst it as well. The fact that children use negatives among their earliestw ords prov ides d irect counter-evidence to such empiricist doctrine. Negationhas no referent, unlike nominal terms or adjectives (such as color words),and is inherently relational in nature. It is not even logically possible for thechild to be taug ht the use of n egation by ostension. fo r althou gh there ar e afinite nu m be r of norm ally used w ords truly ascribable to a particular ob ject,there are an infinite number of words not truly ascribable to i t . In otherwords, there are far fewer things that an object is, than that it is not. Ofcourse, negation has meanings other than the truth-functional meaning ofnot-x I have been describing, but for all of the negative meanings to be con-sidere d the sam e point holds-one can not refer to negation.Fortunately there are more seductive reasons for studying the develop-ment of negation in children's language and thought than to rebut an osten-sion theory of word learning. Negation is a pervasive and essential concep-


3/32

Development o f Negation 157be without logic, science, o r explici t cor rect io n, all of which ar e rel iant o nproposi tional negation (Al tn iann . 19 67 ; Ha rr i son , 197 2; Wi lden , 1 97 2) .Compara t ive analyses o f co m m un ~ca t i onhave shown that negat ion is cen-t ral to human language. yet conspicuously absent f rom the natural com-municat ion repertoires o f o th e r an im a ls ( A l t m a n n , 1 9 6 7 ; S e be o k, 1 9 6 2 ) .Philosophers have taken speclal note that proposi t ional negat ion operateson sentences as par t of a m etalangu age a n d i s, therefore , o f a higher logicaltype than the language i t opera tes upon; they have emphas ized tha t theabi li ty to conceive of proposi tions as true o r fa lse (de pe nd en t on the op era -t ion of negation by which these valu es ar e def ined ) is on e of th e centra la spec ts o f language co rnprehe ns~on nd use (Dum met t , 19 73 ; Kan t , 1963 :M arshal l , 1 9 7 0 ) . It would be na lve to c laim that knowledge of the t ruth-con ditions fo r sentences 1s all we kno w as sp ea k er s of a natu ra l lan gu ag e,given the rich variety o f acts we acco mp lish in ou r uses of lan gu ag e (f o rexam ple, Searle , 1969. 1 9 7 6 ) . but such rules. in tegrally tied to the con-cept ion of negation, are a major par t of o u r knowledge abo ut langu age.

T he pragmat ic c o nd ~t lo ns f negation use provide yet ano ther mot ivat ionfor the study of negatlon de velo pm ent . Nega tives in n atur al langua ge serveto mark a d i screpancy f rom a posi t ive assumption that someone is pre-sum ed to bel ieve, whether oneself o r an ot he r . Th is social fact ab ou t theuses of negatlon ralses a central question in the acquisit ion of communica-t ive competence. For ~f the app ro pr iat e use of negation requires inferringthe belief states of o ther p erso ns, on e might expect the you ng chi ld to havegreat difficul t~es, or the reason that kn owledg e abo ut the physical pro per -ties of objects , so c~ a l e l a t~ on s, nd event cont ingencies in social a nd physi -cal interact ions has just begun to develop. Whether the differing worldviews of chi ldren lead t h e ~ r ses of n e g a t~ o n skew of the pragm at ic condi-tions of a dult negation 1s o ne pro ble m we m ust con fro nt.A major quest in the study of the development of negat ion and languagein general is to explain the emergence of t ruth-funct ional negat ion, used todeny proposi t ions which are not t rue, from the very early expressions ofnegation that are affective in n ature . Th e lat ter uses of neg at ion, as an inter-personal tool for con straln lng agency by reject ion, are vastly different fromthe use of negat ion to deny statements. Truth-funct ional negat ion is meta-l inguist ic and depends on a knowledge of t ruth-condit ions for predicat ions.In the terminology of early tw en t~e th- ce ntu ry iscussions of langu age devel-opment , negat ion prowdes a paradigm case of the emergence of the lan-guage of "intellect" from the language of "affect."

In thls chapter I plan to explore som e funda me ntal asp ects of chi ldren 'sacquisi tion of the semant ics of nega t ion, paylng p a rt ~ c u la r t tent ion (a s onewho has worked w ~ t h erry Rrun er wo ul d) to the t ransit ion from prel inguis-


4/32

1.58 Roy D . Pead r e n a r e t h en p r e sen t ed w h ~ c h n d i ca te th a t d i f fe r en t m ean in g s of n eg a t i o ne m e r g e In a n i n v a ri a nt w q u e n c e where,^\ ind iv idua l s vary as to w hich lex i -ca l i t ems they use to ex pres s the sL iine nega tive me an in gs a nd In t he neg a-t ive mean lngs mos t p revalen t In the l anguage they use T h e lnva r ianc e ise x p l a i n e d b y d e v e l o p n l e n ts In c o g n i ti \ e r e p r e s e n t a t ~ o nw h ich a l l o w f o r t h en e w m e a n i n g s , a n d t h e v a r i a t i o n 15 ~ ~ e w e ds resu l t ing f rom the spec t f i cexperiences o f d if fe ren t ch i l dr en in m o th e r - ch i l d d i s co u r s e A n a cco u n t oft h e f o r m a t i o n of n e g a t ~ v eword rn tdn ing \ 1s p rov ided wh lch emphas izes th einteraction be tween paren ta l u5es of nega t lon in speci f ic types of se t tm gsan d ch i ld ren 's f ir st uses of negat1Lc term5 Fo cus ing on the in ter ac t ion s Inwhlch ear ly nega t ions ge t used by b o th c h il d a n d a d u l t , r e m ov e s s o m e ofth e m y s t e r y f r o m n eg a t i o n s em an t i c s d n d f r o m th e p r o ces s b y w h lch ch l l -d r e n c o m e to k n o w th e p rag rn a ti c co n d i t i o n s of n e g a t i o n u se

Meanings of Nega t ion i n SentencesI n t h e f ew s y s te m a ti c s t ud i es d e ~ o r c d o t h e s e m a n t i c d e v e l o p m e n t of

n eg a t i o n , s ev e ra l m ean in g s of n eg a t~ o n av e p lay ed cen t r al r o l es . A n in flu-en t ia l t r ipar t i t e d iv i sion was p ropos ed by 13loom ( 1 9 7 0 , p . 1 7 3 ) :I . Not l ex i s t ence , where the retercnt was not manifest in the context, whereth er e was an expectat ion of 11s existence, arid ~t was correspondingly negated inthe linguistic expression:2 . Re jec t i on , where the referent actually ex~sted r was immlnent within thecontextual space of the speech event and via\ rejected or opposed by the child;and3 . Der z~a l ,where the negatlve utterance asserted that an a ctual (o r sup po sed )predication was not the case.

B lo o m f o u n d th a t t h e s y n t ac t i c ex p r r s s lo n o f t h e se di ff er en t n eg at i o n m ea n -in gs p r o c e e d e d in t h e o r d e r n o n e x ~ s t e n c c , e j e c t ~ o n , n d d e n i al f o r a l l t h r e eof t h e A m e r i ca n , E n g l i s h -s p eak in g ch i l d r en s h e s t u d i ed .

M cN e i l l an d M cN e i l l ( 1 9 6 8 ) a l s o rea li zed t h e i ns ig hts w h ich a co n ce n -t r at i on o n n e ga ti on s e m a n ti c s c o u ld p r o v ~ d e o r n eg a ti on d e v e l o p m e n t ,b ey o n d th e s t r u c tu r a l d e s c rtp t to n s ca r e fu l l y p r o v id ed b y Be l lu g i ( 1 9 6 7 ) i ne a rl ie r s y n t a c t i c st u d i e s . T h e i r s t u d y r e p o r t e d d a t a f o r o n e c hi ld f r o m t h eag e of tw o y ea r s t h r ee m o n th s l ea r n in g J ap an es e a s a fi rs t lan g u ag e . T h eMc Nei l l s p rop osed a sys tem of th ree b inarb fea tu re con t ras t s fo r d i f fe ren tm ean in g s of n eg a t i o n : E x i s t en ce - T r u th , E x t e r n a l - I n t e r n a l , an d E n ta i lm en t -N o n en ta i lm en t . T h es e s t u d i e s by Blo o m an d t h e M cN e i l l s a r e o f t en i n t e r -


5/32

D eve lo pr nenr o f N r g a t io n I 5 9c r epanc l e s be t u t . cn t hem t ha t have even been no t ed by M cNed l ( 1 9 7 0 , p9 6 ) A b r ~ e f e b lc u of t h e f in d ~ n g sw ~ l lm ak e t h ~ s learT h e b l cN e~ l l \ hough t t ha t t he ta s k o f t rac lng t he cou r s e of semanticdevelopnicnt tor r icgat lon would be considerably s l m p l ~ fi e d y o b s e r v ~ n gheacqul sl tl on of J c i p ,~ r l e \ e l nce t he f ou r co mm on f o r ms o f s ~ m p l e egat lon InJapanese habe tl ifTerent sernantlc functions a n d (+) r ( - ) f ea tu re marker s on each of the th ree bm ar y dimensions mentioned a b o v e :

1 N a l ( a l t r ) 15 d d e n ~ a l f a prevlous predlcat~on,which I w ~ l l all nonent a ~ l ~ n ge n ~ a l2 Na l ( a d / ) 15 d n dssertlon of nonexistence of X glven an expectatlon, by the

speaker or l~ste ner that X was prev~ouslypresent (where X is an object orevent)3 l y n I S d 5tr,11i;htforward ejection, glossed as ' I do not want" and4 I ~ y a S d d e n ~ ~ lf a previous propos~t~onl u s t h e ~ m p l ~ c a t ~ o nhat sonlethlng else IS true t h i r may be called e n t a~ l ln g e n ~ a l

T h e M c N e ~ l l s i e w e c i e ga t lo n d e v e lo p m e n t a s t h e ac q ui si ti on of d i m e n s ~ o n stha t a re marked ( e ) n a semant ic feature analys is . A cen t ra l p rob lemwith th is v iew, however , i s the oddi ty of the "Exis tence-Truth" d imens ion.wh ich is no t a d~ n i en s l on t al l, un li ke t he o t he r two . T h i s d i m e n s ~ o n ,heysugges t, concerns the "condi t ion" of neg a t ion , th e ex i s t ence o r l ack of e ~ th e rsom e thing o r the t ru th of so m e sen tence . B ut s en tences an d th ings a re o fd if fe ren t types. Only sen tences ha ve t ru th -va lues, s o the ex i s tence of th ~ n g sand the tru th of s en tences a r e no t d iamet ri ca lly oppo sed a s a r e the 2val ues o f t he o t he r t wo d ~m en s i on s . i nce t ru t h- val ues i nvo l ve a rb i t r a ry andrule-governed t r u t h- co nd ~ t i on s fo r sentences , the " t ru th" aspe ct of theM c N e i l l s ' d ~ m e n s ~ o nS of a h igher o rder and log ica l type than the "ex i s -tence" aspect . 141th the c on se qu en ce tha t an y t fea tur e analys is with thesetw o c o n d i t ~ o n s s a r t ~ f i c ~ a l .

T h e M cNei ll s conc l uded t ha t t he J apanes e ch i ld 's s eman t i c deve l op n~ en tof negat ion proc eede d In the or de r of nonexis tence, no nen tai l ing denial ,r e j ec t i on , and en t a~ l i ngden ia l . Th e di f fe rence be tween B loom's and theMcNei ll s ' f indings is that th e Ja pa ne se chi ld pu rpor te dly "had th e Idea ofl ingust ical l> regl5terIng the t ruth of s tatements before she had the idea ofl ingu is t~c al l j eglzterlng he r inn er s ta tes in re la t ion to o ut er ones" ( 1 9 6 8 , p .61 ) . whereas al l of Bloom's subjects expressed denial af ter reject ion in syn-tactic expressions.

T h e devc lopn icn ta l h ~s to r i e s f nega t ion sem ant ics o f fe red by B lo om andthe \ l c Ne ~ l l s r l c ,~ ltn ~ t h he expressio n of negat ive me an ing s in scntenceych il d ren p r od uced , they d o no t r eal ly add r e s s ques t ions ab ou t t he deve l op -


6/32

I 60 Roy D. eatence first" may be an artifact of the need for specifying a referent of thenonexis tence proclamation (Bloom , 197 0, p. 2 1 9 ) . S ince the communica-tion r equ irem ents fo r eRcctive discourse can differentially affect the com-plexity of expression for diflerent meanlngs of negation, the developmentalord er of expression f or negative meanlngs rn sentcnces tells us little aboutthe development of negative meanings per se. A second reason is that chil-dren display a rich variety of uses for negat~vewords even in the singleword utterance period. From a wider perspective, as Vygotsky's psycho-genetic m eth od (1 9 7 8 ) has show n, the study of the history of behavior thatbegins a t i ts source, and t races the d y n a m ~ c elations between i ts com po-nents, holds the promise of explanatorj analyses of processes of develop-me nt instead of descriptive accou nts of develop me ntal products.

Negat ion in the Single-Word Uttera nce PeriodBefore progressing to a description of com mun icative contexts of single-word negatives, we must confront a problem rarely acknowledged f or childsemantics in general, a nd not at all for child nega tion. Wha t exactly is "ne-

gation"? L ik e man y words, "negation" does not have any one central o rdefining essence, but a number of meanings that partake of family resem-blances to o n e anoth er (Wit tgenstein , 19 58 ) .Dictionary definitions tell us that negations express such meanings asdenials, refusals, prohibitions, an d sta tem ents of no nexistence. On e criticalthread running through these expressions is the likelihood of their beingexpressed by the wo rds "no" or "not," the prim ary negative particles inEnglish. T h e dict iona ry definitions include the m ean mg s of negation distin-guished by Bloom and the McNeills, but the problem of polysemy for anegative w ord su ch as "no" o r "gone" is a difficult on e, not unlike the lexi-cographer 's problems with polysemy in general. And unlike other areas ofsemantics, such as kinship or spatial terms, we have no standard l inguisticstudies on negation semantics to refer to. Negation is so fundamen tal a partof lexical m ean ing in general (fo r exam ple, in an tonym y) that a "semanticfield" for negation would include many of the words in a language (Milleran d Jo h n so n- La i r d, 1 9 7 6 ) .

Ho w fine a line sh ou ld be draw n between different mean ings of negation?Conce ivably, we could have either one general me aning o r a proliferation ofmeanings, literally one for each distinct occasion of use. At least this setsthe limits. T h e level of generality chosen may be m otivated by different co n-cerns. Miller (1978) indicates two potential s trategies: a lexicographicapp roa ch whic h proliferates senses of a word in i ts occupa tion with makingas many con cep tua l distinctions as possible, specifying d iscourse and selec-tional restr ictions for each specif ied sense; and a second approach which


7/32

Development o f Negation 161only introduces a new sense when combining it with another sense wouldyield a uselessly overgeneral meaning. The lexicographic approach eventu-ally has i ts limits, an d th e latter ap pro ac h is restr icted by how o ne con -strues "uselessly overg eneral." W e will no t settle th e prob lem in this briefdiscussion, but i ts consequences f o r negat ion seman tic development deserveat tent ion.O ne str iking fea tu re o f early verbal negation is the great range of con-texts in which the f irst negative words are used before they are ever com-bined with oth er w ords to form sentences. A problem for the child languageinvestigator analyzing the contexts of use for single-word negation is todelineate psychologically real semantic categories of negation for the child.It would be a great help if the child produced just two different negativewords, such a s "no" an d "gone." each o nly app earing in set si tuations, per-haps "no" only t o reject objects and "gone" only when people walk ou t ofrooms. But actual negative usage is much more complex than this. Childrenrapidly generalize negative words to new situations with the same tenacitythat they carry o ver object wo rds to new category exemplars , and the devel-op m ent al psycholinguist 's task requ ires detailed de scription s of contex ts ofnegat ion use (both discourse and si tuat ional) and the dif f icul t work oftrying to detect nuances of meaning that have later consequences in, forinstance, later structu ral d ifferentiation of negation expression. A t this time,a complete picture of early negative use is not available, but the followingobservat ions provide a sketch for one.

I would like to suggest a tentative resolution of the meaning delineationproblem by proposing that cer ta in families of negative meanings can beidentified in the wide variety of uses of single-word negation, and that sev-eral specific questio ns a bo ut th e emergence of different meanings for nega-t ion in chi ldren's speech may then be answe red.As focus for discussion, f igure 1 prov ides a n overv iew of conte xts inwhich negatives were used in the study I will describe or by children fol-lowed in pre viou s child language studies. ' F o r pu rpos es of exp osition, thechild's negative utterances are l isted separately as to whether they are adja-cent to adu l t ut terances o r init iated by the chi ld a nd , hence, no nadjacent .O n the adja cen cy side of the f igure, the types of ad ult utterances that pre-cede the child negatives a re l isted. Fuller elab oratio ns o f variations of thed i f f e r en tcon texts a re sp ecified in the r ight-hand side of the f igure; these arenot presented in the left-hand side purely for reaso ns of sp ace. T h e prolifer-ation of distinct contexts in these descriptions will lead us back to the farni-l ies-of-negative-meanings account a lready m entioned.Negation is frequen tly used to reject pare ntal p rohibitions or im peratives,and t h s negative context f requent ly recurred in long rounds of turns ( thatis. i terative loop s of ad ult command-child rejectio n). Dis app earan ce nega-

I This literature is reviewed in Pea (1 97 8) . The most descriptive published ac-co u n t s to date are Bloom (1973) and Leopold (1949).


8/32

FIGURE1

Coesfohuon

oInthf

woy

ofe


9/32

Development of Negation 163tions, such as "gone," were usually coupled with where-questions in dis-course, but occasionally with declaratives, each of which often cued a searchfor a non-present object which, after search, was not found. A variety ofdisappearance contexts had negative words expressed in them, such aschild-caused object disappearance and other-caused object disappearance,and each of these variants may vary along yet other parameters--objectin/out of possession of child at the time of disappearance, negative wordproduced before (as a negative wish) versus after the object disappears,and so on.Negatives are also often responses to yes/no questions and declarativesand may serve to deny the proposition expressed in the question form. Theyes/no questions answered by the child's negatives may be placed in one ofat least three categories, requiring different kinds of evidence for anaccounting of the truth of the proposition in question. Many questions, suchas "Do you want a cookie?", "Can you open it?" or "Can you see themouse?" depend on the child's desires, abilities, or perspective. Yet an-other important class of yes/no questions and declaratives express propo-sitions whose truth depends on facts of the world external to the child'sviewpoint, such as "Is this Daddy's pipe?" or "Did we go to the fair?" Allthree classes of yes/no questions may also vary in the dimension of time, sothat the proposition in question relates to present, immediately prior, or dis-tant past events.Other uses of early negation are for make-believe, as in disappearancecomments such as the sentence "Roy gone," used with laughter by one childas Roy stood in the child's plain view. Negatives are also sometimes used toagree with prior negative utterances (see Bellugi, 1967; Pea, 1978) .The heuristic diagram presented in figure 1 should not be considered acomplete list of early contexts and meanings with which negative words areused, but it captures the diversity of situations in which young children usenegatives. One could even proliferate further subcategorizations of thenodes of the figure, differentiating yet more situational variants. But the listof early contexts for "no" and related words used by children is not a typol-ogy of early meanings for negation, for there is no reason to assume thatthese 'types" (and subtypes, if our diagram delineated more subtle meaningdifferences) are in any sense distinct in children's conception of their ownuse of the negative words in these situations. One could take a lexicographicapproach as an initial guideline for distinguishing different negative mean-ings that may be psychologically differentiated for the child. The reason,however, why this is a bad first start is that there are certain families of neg-ative word contexts, with many features in common for any given family,that naturally cluster together. Though one still can not conclude that thesemantic categories thus obtained match up with distinct negative meaningsfor the child, the categorization is at least motivated by similarities in thechild's behaviors and situational contexts of the different category members.


10/32

164 Roy D . PeaFo r example, the child's use of "no" to reject a n object offered to her,accompanied by a pushing-away gesture, is very m uch like the child's use of"no" in rejection of a prohibition cons training her action.The semantic categories used here for meanings of child negation are aresult of such groupings of negative word uses in different contexts, andthey resemble in some respects the categories described by Bloom and theMcNeills.Rejection negatives were defined as those action-based negations withwhich the child rejects an event, person, object, or activity which is in theimmediately perceivable context of rejection or imminent in the mother'sbehavior or utterance (whose truth-value is contingent on the child's cur-rent m otivations) . Th e "imminent" disjunct of th e definition refers to usesof negatives used in contexts in which highly routinized activities providecues for the child (such as a diap er in the m other's ha nd ) of imminent to-be-rejected actions, and negatives used in response to desire questions suchas "Do you want a cookie?"Disappearance negatives a re uses of negatives with reference to the disap-pearance of something which had been present just prior to the child'sutterance but is n o longer perceivable to the child. Th e "something" may bea n object o r person , a n activity, o r some form of sensory stimulation (suchas the sound of a washer o r radio) which disappears o r ceases.Related to the negatives of disappe aran ce but distinct in important waysar e negatives of unfulfilledexpectation. Negatives expressing the meaning ofthis semantic category are used to comment on constraint on activity,absence other than immediately prior disappearanc e or cessation, or, m oregenerally speaking, some aspect of the child's continuing line of activity(such as search or play) which does not occur (for example, blockage ofmovement of bicycle, toys tha t ar e malfunctioning) o r is not found ( a blocknot in its habitual location).Truth-functional negation consists of the use of a negative in response toan utterance expressing a proposition that is true o r false given the facts ofthe situation it refers to and, hence, the truth-conditions of language. Suchtruth-functional negations express logical judgments and constitute a subsetof the negatives captu red by Bloom's "denial" category defined earlier. Thisstands in contrast to some rejection negations, which negate a propositionthe truth of which depends o n the child's own m otivations.Self-prohibition neg ation is a form of egocentric symbol use in which thechild app roaches a previously forbidden object o r begins t o d o somethingwhich has been prohibited in the past and then expresses a negative. Suchnegations are not necessarily accompanied by the child's avoidance of theforbidden act.


11/32

Development o f Negation 165stantial proportion of all the negatives children produced in the studyreported here, unlike the potential meaning categories of negative agree-ment, negative wish, make-believe negative, and non-truth-func tional denialnegation. Such negations d o occu r during the first year of language use, butonly infrequently.

The proliferation of different uses for negation by the age of two yearsdeserves special emphasis, especially because it stands in contrast to argu-ments by Steffensen (1978) tha t the word "no" during this period is "lack-ing in semantic content" and only fulfills the conversational requirement ofresponding to yes/no questions. The richness of early negation semanticsbecomes apparent when one studies children's spontaneous use of nega-tion and negation in response to parental linguistic forms other thanyes/no questions.

Co gnitive Re presenta tion and Meanings of Nega tionOne can now ask, given our characterizations of different semantic cate-gories of negation, in what order the first expressions of different negative

meanings develop. Longitudinal data from six children indicate that chil-dren first convey inner states via rejection negation, then make com-ments on the disappearance of things in the world, and only later truth-functionally negate false statements about the world. Individual differencesoccurred in the developmental ordering of unfulfilled expectation negationand self-prohibition negation, although for all children such meanings pre-ceded the expression of truth-functional negation. These findings may beexplained by analyses of the progressively higher-order representationalrequirements, or cognitive bases , of the concep tion of different meanings fo rnegation. The general "cognitive complexity" of negation has been a stand-ard observation in studies of adult cognition and child language, but thisunanalyzed notion deserves reconsideration.Negation is a semantic domain which readily lends itself to conceptualanalyses of representational complexity. The claim is that for the child toconceive the different meanings of negation that I have described, increas-ingly abstract forms of cognitive representation are required. Whether oneadvocates the developmental theory of Bruner, Piaget, or Werner, a familiardevelopmental progression is the emergence of symbolic or abstract repre-sentations from more primitive mental activity limited to concrete motor-affective sensorimotor intelligence. Th is progression places im portan t limitson negation sem antic development.Rejection negations express inne r attitudes of rejection toward behaviors,


12/32

166 Roy D. earepresentation because it is imm ediately present in the context of the rejec-tion. Darwin (1 87 2; also Jakob son, 19 72 ) has emphasized in his com pa ra-tive studies o f emotional expression, over a century ago, that the early nega-tive expressed universally with nasal sounds (Jespersen, 1917) and theheadshake, or conventional gesture of negation, are each natural signswhich directly express attitudes of aversion.Disappearance negation, on the other hand, typified by the child's com-ment of "gone" or "no more" as a ball disappears from sight, requiresabstract cognitive representation. Unlike the object of rejection negation,the ball is no longer present, and the negative comment must abstractlyden ote the vanished object or event o f disappearance.

These considerations of the cognitive demands of different meanings fornegation lead to the prediction that rejection negation, rooted to concretemotor-affective activities, will be the first meaning of negation that childrenexpress, followed by the emergence of disappearance negation, whichrequires the elaboration of more complex cognitive representation. Thoughsuch a prediction may seem intuitively obvious, recall that the account ofsemantic developm ent proferred by the McNeills' (1 9 68 ) claims that th e"natural order" of emergence of meanings for negation is "from outer toinner," from the negative meanings of no t-here and false toward ex pressionsof rejection.Truth-functional negation also requires abstract cognitive repre senta tion,of yet greater complexity and of a different logical order. Negatives withthis metalinguistic meaning of not-x consist of comments on the use ofabstract language by an oth er person a nd a re by definition of a higher logicaltype than disappearance negation (Bateson , 19 68, Wilden, 1 9 7 2 ). F ro mthe perspective of development in cognitive representation, truth-functionalnegatives should not be used by the child prior to the meanings of rejectionand disappearance.It is im portan t to observe that, theoretically, all three func tions he re maybe expressed gesturally, with the negative headshake, as well as in singlewords. For example, the child could shake her head in rejection, whenobjects disappear, or in response to yes/no questions or false stateme nts. Amuch more complex developmental landscape than this is revealed in thecourse of the children's negation development. We will return to the seman-tic categories of self-prohibition and unfulfilled expectation negation in alater section.

METHODS


13/32

Development o f Negation 167Each child was visited individually in his o r her hom e once each mon th fo rabout ninety minutes, and videotaped for a total period of thirty minuteswhile playing, feeding, bathing, and during other home activities. Motherswere also queried about their child's cu rrent uses of th e head shake a nd neg-ative words but were uninformed as to the specific questions of the s tudy .Transcriptions from the video records included all speech produced bythe children and adults present, with additional relevant nonlinguisticbehavior, for example, actions, pointing, and context (gaze direction, ob-jects present) noted in temporal relation to the vocalizations.The data for analysis were headshakes and verbal negations, consisting ofsingle o r multiword utterances made u p of at least one negative word o r aword co-occurring with a headshake. Negative words were such morphemesas "no," "not," "n't," approxim ations of "no" such as /n a / o r /na/ andwords such as "gone," "all gone," "away," and "stop." The se negativeswere transcribed from the video records and from unfilmed but tape-recorded observations which had been supplemented with context notes atthe time of the observations.

RESULTSAll six children first expressed rejection negation, and they expressedthis meaning in headshakes before they did in speech. Previous studies byCarter (1974) and Ruke-Dravina (1972) also found a gestural priority.Headshakes were first used around the age of one year to one year onemonth with an age range for the group of ten months to one year twomonths. Fo r example, R T did no t spontaneously use negation until one year

one month, when he shook his head on five different occasions, threedifferent times in response to different questions of the form "DO youwant-?" where th e referent was visible, once in rejection of a prohib i-tion from his mothe r, and once to reject bicycle clips offered to him .Four of the children first used the headshake to reject food, diaperchanging, face cleaning, and other ritual maintenance acts. The other twochildren first used the headshake in response to questions pertaining toritual functions, such as "Do you want some milk?" or "Want to go tobed?" that were redundant given the objects o r actions that were at ha nd .Considerable differences were found with respect to the time periodbetween the first uses of nonimitative uses of headshakes to reject and thelater nonimitative expression of negatives in speech. This gap ranged fromone to nine months for the children studied. Rejection was generally thefirst negative meaning encoded verbally. Tw o of the six children first used


14/32

168 R o y D . PeaTable 1 summarizes the ages of the children at first appearance of the dif-ferent functions of negation, both in gesture and in speech.

TABLE 1Age in Mon ths at First Appearance o f Negation M eanings

Child Rejection Disappearance Truth- Unfulfilled Self-Functional Expectation ProhibitionGesture W ord Gesture W ord Gesture W ord Gesture Word Gesture Wo rd

It is important to note that the cognitive representation analysis does notspecify that the communicative function of rejection will be the first mean-ing to be expressed v e rb a l l y . It may be the case that if the headshakealone is efficacious for the child in communication for expressing rejectionthen it will not necessarily be the first meaning of negation expressed ver-bally. The central issue is not the order of verbal expression of these mean-ings of negation, but the order of expression of negative meanings via con-ventional forms, whether in gesture or speech.

Four of the six children went on to express negative comments of disappearance with the word "gone," predominantly in contexts where objects orpersons had disappeared immediately beforehand. They all also expressedimportant intermediary forms of negation between the affective negations ofrejection and the more abstract negatives of disappearance. In the monthprior to the use of disappearance negation for each of these children,"gone" was used in contexts where objects fell out of their possession butremained in their clear line of sight, and were often then picked up by thechildren immediately after they had made the remark. But at this time thechildren had not made such comments when things actually disappeared. Afifth child had not used the word "gone" for disappearance negation beforethe completion of the studies, but like the other children, consistently said"ga" when towers of blocks fell over or objects fell from his hands. Thisdevelopment-from "gone" in the sensorimotor present to "gone" for non-present objects-suggests that constraints in cognitive representation delaythe expression of disappearance negation for absent referents. Such a find-ing would be expected on this cognitive approach. It also provides new sup-port for Werner and Kaplan's (1963) developmental principle of "dis-tancing," which claims developmental continuity between language fusedto action and perceptual schemas in the present, and language distant and


15/32

Development o f Negation 169Truth-funct ional negat ion was the last of these three negat ion funct ionsto appear fo r the f ou r chi ldren expressing this type of nega t ion a t a l l .Bloom's (1970) m or e gen eral category of "denial" negat ives includ es, a s

ment ioned ear lier, cases where th e chi ld d enies previous ut terances who setruth is contingent on the chi ld 's motivat ions in the si tuat ion, rather thaninherent t ruth condi t ions of th e s ta tement involved. A n e xam ple of denialnegat ion which is no t t ruth-fun ct iona l negat ion is Gia 's reply "no, no w I d oit" to the adul t quest ion "now can I d o it?" Al though the chi ld ha s negatedthe adul t proposi t ion "I can d o it," the t ruth of that proposi t ion wa sdependent on G ia' s choice. Truth-funct ional negatives, by c ontra st , a re neg-at ive comments expressing judgments of the falsi ty of statements aboutpropert ies , names, o r act ions. S everal examples i llustrating this mean ing ofnegat ion are given in table 2. Negatives with this meaning were not f i rst

T A B L E 2Exam ples of Truth-Fu nctional Negation U se

Names1. (CB, age 2; l)Experimenter: Is that a biscuit? (Epointing at CB's apple)CB: No, apple.2. (RT,age 1;lO)RT: Trains. [RT points to a train he has made with 2 railroad cars; atthis point each car is a "train" for him, multiple cars are"trains"]Experimenter: Is that a train?R T : No. Trains.

Possession3 . (CS, age 1 7)

CS: Big. [CS points t o a large pair of shoes next to her small pair; thelarge pair are N ed's]Experimenter: Is this big? [E points to CB's small shoe]CS : No, mine.Ned. Big. [CS points to one of Ned's big shoes]Presence/Absence

4. (CS, age l ;7 )Mother: Is the ball gone? [M shows CS the ball]CS: No.Misinterpretation Correction

5 . (CS, age 1 9)CS : Womble. [CS is searching fo r the Womble doll]Experimenter: Ball? [E intentionally misunderstands C S]CS: No.


16/32

Roy D. eaexpressed liy head shak es alone, bu t in speech within a month of the onsetof syntactic constructions for all four children. The experimental study ofchildren's negative corrections of false statements provides further supportthat this meaning of negation emerges around the-age of two years (Pea,1 9 7 9 ) .A typical use of truth-functional negation took place in a conversationab ou t the tem perature of a bicycle seat. R T (o ne year ten m onth s) had justhad his diaper removed and sat on a very cold metal bicycle seat. Whenasked "Is it warm?" he answered "No, cold!" This meaning of negationprovides the earliest verbal index of rudimentary logical thought, reflectingthe awareness of the propositional o peration of negation (Pea, 1 9 7 9 ).Bloom (1973), in her studies of single-word speech, also provides s u pport for the sequence of emergence of meanings of negation found here."Away" was first used by her subject to reject things as they were thrownaway and afterwards when things disappeared. Truth-functional negativeslater appeared as correct responses to yes/no questions about names andpredicates.One can also ask whether the first conventional negative symbol wasextend ed to express new negative meanings. Recall that the headshake wasfirst used to express the first negative meaning of rejection. But the firstexpressions of both disappearance and truth-functional negation were notheadshakes, fo r any of the children studied . Rath er, d isappearance n egationwas realized in words such as "gone," "stop," and "byebye," while rejec-tion, when expressed in speech, was predominan tly coded with "no," andoccasionally "don't." Truth-functional negatives were usually "no" o r"not," an d on ly two children ever accompanied the truth-functional nega-tive word with a headshake.It is intriguing that three children combined the headshake with anaffirmative word to convey a negative meaning before they used wordstogether in sentences. For instance, "pussy" with an accompanying head-shake was used when a cat disappeared from a television screen, and"mumma" with a head shake functioned as a self-prohibition not to touc hmommy's glasses. These very creative dual-channel negatives were neverthe first ap pe ara nc e of a new negative meaning, as o n e might expect giventhe genetic primacy of the headshake, but later developments of only briefuse.T h e cognitive interpretation of negation sem antic developmen t cha rtedhere has the virtue of providing systematic predictions of developm entalsequenc e fo r the emerge nce of different meanings of n egation. T he focus ison th e develop me ntal m ap sketched by elabora tions in cognitive representa-tion that allow for the emergence of new meanings for negation from theorigins of conventional negation in communication. Children first conveynegation throug h headshakes for rejection, then com m ent o n disappearan ce,and later use truth-functional negation in judgments about language use.


17/32

Development o f Negation 171Cross-language research would provide a test of the universality of thissequence of emergence.

Topics of Ne gation and th e Gene sis of Nega tion PragmaticsAnother way of relating these findings is in terms of development in therange of topics for negations children use. Since rejection negation is thefirst meaning that negatives are used for, the first topics of negation are con-cretely present in the child's immediate field of activity and transcend thehere and now only when negative comments come, several months later, tomark disappearance. In disappearance contexts, the topic of negation hasjust recently gone out of view. One might predict that subsequent develop-ments in the distancing of the topic from the child's negative utterance willgrade from these immediately prior disappearances to topics of a more gen-eral natural, to rulelike assumptions on the child's part. This is strictly truefor only two (CB, RT) of the four children who used verbal negation at allduring the course of study, and involves the relation between negations ofdisappearance and of unfulfilled expectation.Unfulfilled expectation negatives, one might expect, based on norms con-structed through experience, will emerge only after disappearance negations,which are not unlike a verbal "orienting" to a change (presence to absence)in the immediate sensory field. CB and R T met this expectation, as did thechild described by Bloom (1973), but CS and JK conveyed unfulfilledexpectation negatives before disappearance negation. Such variation couldreflect individual differences, or the developmental invalidity of the distinc-tion between negative contexts of disappearance and unfulfilled expectation,

or an overgeneral definition of unfulfilled expectation negation. Evidencepoints to the third choice. Since all of the uses of unfulfilled expectationnegation occurring prior to disappearance negation (for CS and JK) werein action contexts (either when toys did not work or when the child's move-ment was blocked), as opposed to unfulfilled expectation negatives com-menting on expected existence or location, the action uses seem to consti-tute a primitive precursor of the existence and location uses (which for anygiven child never occurred before the first expressions of disappearancenegation).The more mature uses of unfulfilled expectation negation provide animportant insight into the development of negation pragmatics. The moregeneral assumptions on which existence and location negations of unfulfilledexpectation are based are at first local norms, or habitual associations, suchas the fact that frequently encountered or acted-upon objects have standard


18/32

172 R o y D . Peathat the child learns, naturally occurring sequences, or configurations ofactivities, objects, events, and behaviors (Nelson, 1978; Schank and Abel-son, 1977). These scripts serve as normative knowledge, so that when dif-ferences from them arise, the child marks the discrepancy with negation.This description fits well with the pragmatics of ordinary language negationfor adults, with the scripts defining the child's assumed affirmative contextsfor negation. They provide the groundwork for continuity in the development of negation pragmatics, since it is also discrepancies from positiveassumptions that motivate negation use by adults. Table 3 illustrates thegrowing range in topics of negation for the four children using verbal nega-tion.

TABLE 3Dev elopm ental Change in To pics of Negation and Num ber o f Negations

Age in monthsaChild 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 g Total

aThe underscoring of the numbers of negations produced at each visit indicatesth e highest level of topic abstraction for the negations used. This does not meanthat the occurrence of simpler topics ceased, only tha t the range of topics withregard to "distancing" from the topic (see text) increased. Key: single line = topicin the here & now; double line = topic recently disappeared; triple line = topicexpected bu t not found.b Child was not visited a t this age.

e Only in response to "Where's X?questions.The character of unfulfilled expectation negation is best captured by afew examples. When CB (one year ten months) looked into my toybag, forthe previous ten months always containing a ball she favored, she said"ball, Mummy," to which her mother answered "I don't know where it is,"to which CB replied "no ball, Mummy." Nothing had disappeared; the balljust did not appear where she expected it. During the same visit, CB founda teapot lid, usually on top of a teapot that was now missing, and as sheheld up the lid, she remarked "no teapot." Similar negative comments weremade by CB when:1. only half an apple, or part of a button was shown in a picture (hence,incompletely)


19/32

Development o f Negation 173These cases show that CB's habitual norms, like the other children's, werequite idiosyncratic to her household, yet of great importance to her. Sowhen RT (one year nine months) looked at an empty chair where his catusually sat, he said, "pussy gone." O r when C S (o ne year nine m on ths) wasasked where A untie Sonya was (sh e was in the garden , where CS saw herea rlie r), she replied, "gone get dinner," probably because Sonya frequentlywent out to pick up fish and chips. Since CS was in the single-word utter-ance period at the time and was using "gone" produ ctively, this was verylikely a ritual phrase.One may expect, since negatives in the adult language are used to markdeviations from expectations or norms, that the major development fromnegative comments based on such local norms as the habitual location ofobjects would consist of moves toward a m ore gen eral level of agreement orconventionality. Such a view is ha rd t o suppo rt on close inspection.Potential examples of negations with topics of general knowledge may beheard from the mouths of two-year-olds, but a fuzzy area of indeterminacyexists between cases of habitual location and ones based on w ha t (o n firstview) seem to be general conventions. J K (one year six months) pullsdown the trousers of her G.I. Jo e doll and , upon discovering he h as on nounderpants, says, "no pants." Now is this negation based on general knowl-edge, a convention adults would share of a form such as "under trousers,one wears underwear"? O r is it one based o n the local, habitual associa-tion of co-occurrence of pants beneath trousers? The distinction, particu-larly in common cases like this one , is shaky, yet it is at the heart of on eaccount of negation pragmatic development (Volterra and Antinucci,1978) .T o carry the breakdow n of the distinction further, on e need only reflecton how the assum ptions denied in the negative speech of adu lts are alsooften based on local, idiosyncratic norms or expectations, such as "there'sno beer in the fridge" when one's spouse was supposed to buy some at thestore, o r "there's no fire left" after waking u p cold from a n ap nea r thehearth. The habitual locations of objects, such as clothes, keys, and furni-ture, are of great importance for our everyday affairs, just as they are forchildren.Many apparent differences between the topics of adults' and children'snegations are only illusory. But when the ch ild's topics of negation movebeyond the here and now, a re they always similar to those of adults? No.Volterra and Antinucci (1978) have characterized the differences thatoccur as "pragmatic misfires," cases where the presuppositions (or topics,in my terminology) of the adult and child just do not match up. Theydescribe an exam ple where a child looks at a hospital tower an d says "Lookthere is no bell." Yet adu lts kno w it is chu rch towers which hav e bells and


20/32

174 Roy D. easuppositions into line with the listener's." But how does such an alignmenttake place?In their analysis of negation development for Italian and American chil-dren, Volterra and Antinucci (1978) stress that the determining factor inthe emergence of presuppositions for negation based on "normative classes"shared by adult and child is the developing ability to represent cognitivelysuch abstract general knowledge.But developments in cognition alone cannot explain such presupposi-tional changes, which must come from the sociocultural conventions of thechild's language community. The process of aligning child-adult languageknowledge is accomplished largely through dialogue, and the older child,when such pragmatic misfires occur, is better able to repair the breakdownof understanding with the listener. The few studies of children's responsesto clarification requests, though not addressing negative speech per se, indi-cate the complexity of identifying the source of the misunderstanding(Cherry, 1979; Garvey, 1978; Stokes, 1976).In adult conversations using negation, too, our knowledge frequentlydoes not match up with our listener's expectations. If it did, as Brown(1973, personal communication), Givon (l978), Labov and Fanshel(1977), Wason ( 1965, this volume), and others have pointed out, therewould be no conversational point to negative utterances. But adults are veryadept at clarifying knowledge differences when pragmatic misfires occur andat making explicit their assumptions and background when engaging in dis-course. The ability to synchronize assumptions through clarification movesin a conversation when misfires do occur is thus a critical component in theemergence of pragmatically successful negation, once the topic becomesabstracted from the present.Volterra and Antinucci ( 1978) describe general developmental pattenisin distancing the topic of negation beyond the present similar to thosedescribed in this chapter (table 3 ) . There is, however, one other critical dif-ference between the formal linguistic analysis proposed by Volterra andAntinucci and this account of negation topic development. They formallycharacterize what I have called the "topic" of negation as a "presupposi-tion." The presupposition is then attributed to the child's listener, so thatthe listener is held to believe p, where p is the presupposition negated by thechild.One of Volterra and Antinucci's examples will make their approachclear. A sixteen-month-old child says "gone" after a flame is blown outfrom a match. Their interpretation of this utterance is that the child negatesthe presupposition "the flame is present" which he infers that the listener


21/32

Development o f Negation I75of the speech context and the knowledge state of his listener, yet inferringthe beliefs of another person is certainly not a requirement of making suchan utterance, since children use many negatives without addressing anotherperson, particularly when objects disappear. Such inferences of beliefswould also require more sophisticated social cognition than two-year-oldsseem to have available (Gearhart, 1979; Shantz, 1977 ) . Independent evi-dence from experimentation is necessary before claiming that children inthis period have the ability to infer specific beliefs of another person.

Discourse Con texts of Negation and Lexical Varia tionin Negation ExpressionThe development of meanings of negation for children from one to twoyears old has been explicitly related to the growth of representational abili-ties. Cognition alone, while allowing more abstract topics of negation, canprovide neither the meanings of negation nor the specific words children useto express them. The meanings and forms of negation are conveyed by

speakers of the language by virtue of the child's engagement in a sociocul-tural and linguistic community. This transmission becomes clear from thestriking individual differences in negation expression for the children stud-ied longitudinally. Discourse contexts varied across mother-child dyads, andvariations in the words children used to express negative meaningsoccurred.Topic initiation provides a useful focus for a look at discourse contextsfor negation. Adult-initiated negatives are those which the child producesafter an immediately preceding adult utterance. Self-initiated negations arethose which are not adjacent to adult speech but are spontaneously used.The central finding from an overall analysis of self-versus-adult-initiatednegation is somewhat surprising: 40 percent (or 367/922) of the total neg-atives produced by the four children who spoke during the period of studywere initiated by the child. This means that children were making negativecomments on their own a considerable amount of the time, when things andpeople disappeared, when toys would not work as they should, when thingswere not found in their habitual locations, and in self-prohibition.Negation topic initiation for any given mother-child dyad is not, unfor-tunately, very informative if one seeks out developmental trends. One mightsuspect that children would at first produce negatives predominantly inresponse to adult utterances and, with development, come to be able tointroduce negations spontaneously which before required maternal conver-


22/32

176 Roy D. eapression of propositions on her own (Keenan, Schieffelin, and Platt 1976;Ninio and Bruner 1978). But topic initiation for negation does not seemto follow such a pattern. The picture is one of great interindividual vari-ability and substantial up and down fluctuations across sessions in theproportion of self-initiated negations for any individual child.

Large individual differences occurred in the overall predominance of selfor adult initiated negation. For example, CS, whose mother was extremelyprohibitive and constraining in her child-rearing, produced only 20 percent(or 64/315) of her overall negatives spontaneously, since most of themwere in emotionally charged negative retorts to her mother's commandsand prohibitions. JK produced 72 percent (or, 111/155) of her negativesspontaneously; this was largely a consequence of her mother's fostering ofindependent play, often accompanied by what might be described as ego-centric speech. The other two children are near the 40 percent average(CB: 49 percent, or 75/153; PT: 39 percent, or 117/299).

One consequence of such differences is that the different children pro-duce radically divergent overall proportions of different meanings of nega-tion; 58 percent of JK's negatives were comments on disappearance,whereas only 7 percent expressed rejection. Recall that it was JK's motherwho fostered independence and solitary play. CS's language environment,however, was characterized by constraint and commands, and 37 percent ofher negations expressed rejection, while only 26 percent were comments ondisappearance. So in one important sense, the predominant types of dis-course contexts for negation provided by mothers can affect the predomi-nant meanings of negation children choose to express. It is in this light thatthe agreement in developmental ordering of negative meanings for the dif-ferent children is all the more striking.Specific negative meanings were sometimes expressed by words idiosyn-cratic to the conversations of particular mother-child dyads. JK and CS, forexample, unlike CB and RT, shook their heads as they named the possessorof prohibited objects in expressing self-prohibition negation, a reflection ofthe parents' prohibitions in statements such as "that's Mommy's" modeledwith headshaking. Specific prohibitive formats between mother and childyielded other special self-prohibition negatives, such as JK's "don't eat it"for soap bars, and CSs "mustn't bite" when playing with her sister, each ofwhich were used during the single-word utterance period.

Such lexical variation was not limited to expressions of self-prohibitionnegation. Only JK used "done" in rejection of objects she was tired of, onlyCS used "I 'nt" to reject commands (it is an Oxford working class slangcontraction for "I won't"), and only CB used "alldone" for disappearance


23/32

Development ofNegation I77Lexical variation was not the rule, however, in the words used to express

a specific negative meaning. All of the children used headshakes and "no"for rejection, "gone" and "allgone" for disappearance, and "no" for truth-functional negation. But the existence of such clear cases of lexical variationprovides proof that the forms of negation are a result of the conversationalenvironments provided by the child's language peers.

An Znteractionist Perspective on the Originsof Negation SemanticsThe emergence of negation semantics may be framed in terms of three

major phases. In the first, the child's task is to discover the meaning of neg-atives in the speech of the adult, to form an initial basis or core to themeaning of "no" and related negatives. In the second, the child first usesnegation and begins to generalize negation to novel situations. In the thirdphase, discourse contexts for negation shape the elaboration of negativemeanings conveyed by the child, a phase where the child enters a dialecticof language interaction and apprenticeship in which her negatives getresponded to and in which negatives that are heard in discourse addressedto the child are assimilated to the child's current (and accommodating) con-ceptions of negation. The processes by which conventions of adult negationget stored and adapted as the child's lexicon for negation changes in bothlexical forms and meanings are the ultimate aim of explanatory accounts ofsuch developments, but for now our analyses are confined to descriptions ofmaternal speech, the child's speech, and probable inferences as to how thedialectic of communicative exchange contributes to language development.The key to such processes is ultimately to be found in careful studies of theconversational contexts in which early word meanings are negotiated. Thefirst two of these phases are the focus of the following section, whichdescribes the origins of negation in interactional contexts and the child'searliest reactions to and uses of negation.

The Genesis of Negation ComprehensionThe predominant meaning of negation in the life of the child during the

first year of life is prohibition. Prohibitions from the parent, usuallyexpressed with "No!" and headshake, are addressed to children as they


24/32

178 Roy D. Peahandling electrical sockets, hot or sharp objects, or damage propertydeemed valuable (books, records, glasses).Spitz (1957) sees the child's uncompleted act in conjunction with theparent's negative word or gesture as a major source of the first meaning of"no" for the child. His accoun t assumes tha t the child's frustrated id drivesthereby endow the negative word and gesture with a specific affective cath -exis that ensures the child's remembrance of the negative symbols. Thechild's first use of negation, on this view, is a result of identification with theprohibiting parent, and refusal or rejection is the first meaning since thesymbol is imbued with aggressive cathexis in the unpleasurable experiencesassociated with its memory traces. Apart from the status of the psychoan-alytic com pon ents of this theory, it is certainly consistent with observationsthat "no" is on e of the first words children learn (Gopnik, 19 78 ; Leop old,19 49 ; Nelson, 1 9 7 3 ) and one of th e m ost consistently used words through-ou t the single-word utterance period (Bloom , 19 73 ), the finding that rejec-tion is the first meaning of negation children express, and the prevalence ofprohibition negation in the language of parents to children for my casestudies. Th e case for prohibition a nd constraint a s the context for children'sfirst lea rning of a meaning for negation is best pu t in a genetic descriptionof prohibition comprehension.

For child safety, the parent must establish an effective verbal means ofconstraining child exploration. Child language diarists often note the earlycomprehension of "no" in prohibition a t nine months to ten mo nths (L eo -pold, 1939, p. 112 ; Lewis, 19 63, p. 4 3 ), and the claim is made that thechild halts actions more as a result of the loudness, pitch, duration, andsuddenness of onset of "no" tha n by virtue of its conventional meaning .Such prominent attitudinal characteristics of the negative speech signalmay provide some substantive ground to Spitz's claim that negative wordand gesture become endow ed with a "specific affective cathexis" as a resultof the adult's prohibitions. But how exactly do parents get children tocomply with verbal prohibitions? A general framew ork for this accomp lish-ment m ay b e charted fro m observations m ad e during the period from eightmo nths t o eleven m onths fo r four English child ren; it also provides insightsinto the child's form ation of a con cept of negation as constraint.Pare nts a t first physically constrain ch ildren when they issue verb al pro -hibitions. Ju st as in ea rly com m and situations, where the mother says "Giveme the ball" as she gently pries it from the hand of the child and lateraccomplishes the same act with an open palm and ultimately the verbalcom m and alone, the prohibiting m other says "no" and shakes he r he ad a sshe pulls o r lifts the child away from the forbidden act. Soon the ad ultwishes prohibition compliance at a distance by language and gesture alone,and within a month of physical constraints, children manifest signs of


25/32

Development ofNegation I79before renewing progress to the forbidden act. The child also demonstratesforms of inhibition with regard to the prohibited object or act, which areaction analogues of later self-prohibitions. Intentionally disobeying is yetanother sign of understanding.Several distinctive levels of prohibition reaction reflect increasing under-standing and approach what the adult counts as prohibition negation com-prehension.It is first important to note that the child brings well-developed (if primi-tive) means for rejection to the task of prohibition comprehension. Displea-sure is displayed by physical means; the child rejects objects or actions byturning her head aside, pushing or throwing the aversive thing away, or flail-ing her arms (see Buhler and Hetzer, 1935). The means for communicat-ing rejection here are as strikingly direct as were the parent's initial prohibi-tions of forced compliance by physical means. Parents also frequently inter-pret these behaviors as expressive of negation and expand them with lexicalnegatives: "no, no, don't want it." Ryan (1974) has emphasized the impor-tance of such intent interpretations for the eventual linguistic expression ofintention.The child does not, however, initially use behaviors from the physicalrejection repertoire to respond to parental prohibitions. Cries are used atfirst, and it is a month or more before the onset of physical rejection of pro-hibition. The children's prohibition reactions quickly change form. Whenthe parent first uses verbal prohibition without physical constraint, the childignores the prohibition in nonunderstanding. Then if the prohibition ismade persistent, by repetition, louder voice, or physical constraint, it maybe effectively heeded by the child. A standard part of this last scenario isthe look from the child to the prohibiting parent, generally with bodyaligned in the direction of the prohibited object, or even still touching it.The child looks to the prohibitor but does not immediately withdraw. Prohi-bition sequences while the child continues to touch the prohibited object aremarked by numerous replays of touch-prohibition-look rounds. The child'swithdrawals, if they occur at all, are very short-term.Two of the four children at nine months presented striking inhibitorybehavior when such repeated prohibition cycles were prevalent. After aninitial string of prohibitions and withdrawals, the child begins to approachthe forbidden object and spontaneously inhibits a repetition of the prohib-ited action; the child looks at the prohibited object, arm aims out toward it,is self-impeded in midstream, then slung back and forth several times inapproach-avoidance fashion, and withdrawn (CS was prone to touch aradiator that was often hot, and SR had a favorite rubber tree leaf).For each of the children there next occurred what adults often describe(Lewis, 1963) as guilty, "permission requesting" looks to the parent, at


26/32

I 80 Roy D . Peaused in anticipation of prohibition. Looks would be made in the parent'sdirection before any prohibition had been made in the immediate context,as the child prepared to touch what the parent felt he or she "knew" shouldnot be touched. Concurrent with these anticipatory "permission requesting"looks were instances where the children reacted against prohibitions byusing primitive physical means of rejection (arm waving, fretting sounds)which were at a distance for the first time from their "objectv-the parentalprohibition. Werner and Kaplan ( 1963) note that such progressive distanc-ing between the child and the object of reference results in a shift fromego-bound things of action to ego-distance objects of contemplation. Theprimitive means of rejection once directly affecting the rejected thing (thechild pushed things away) now represents that same pushing away orrejecting at a distance.One other feature of prohibition reactions in this period is the "sneakysmile," where the child's actions superficially resemble earlier times whenthe prohibition was ignored out of nonunderstanding, except that now thechild smiles at the parent in defiance and continues with the forbiddenaction after demonstrating prohibition compliance on earlier occasions. Thechildren at this point, from ten months to eleven months, have begun to useprimitive rejection and open defiance to reject constraint and prohibition.Apart from the later development of the headshake, already described,this completes the major changes in the development of children's prohibi-tion reactions and eventual comprehension. A noteworthy feature of thisdevelopment is the progressive growth of the dyadic nature of prohibition.The child learns to react to the prohibition in two quite divergent ways:compliance, a measure of which, from the adult viewpoint, is nonrebelliouswithdrawal (passivity), and defiance, where the child comes to display anautonomy from the parent's wishes and exerts a control over his or her ownbehavior (activity). In learning the constraining nature of prohibitive actsvis-a-vis the prohibitive communication relationship, the child is learning toinhibit actions in the socially prescribed way. The child is also, however,learning how to inhibit others by constraining them via negative reactions toprohibitions. Prohibition is thus an area of early language comprehensionwhere the realization of role reversibility so central to language becomesnaturally emergent as the child attains active agency.These observations complement those of Lewis (1963) and Spitz(1957) in marking out several intermediate steps of "understanding" theadults' negative prohibitions before the child ever begins to use negativesymbols. Children first reject parental prohibitions at a distance by uncon-ventional (nonsymbolic) physical means which they had previously utilizedto physically, or directly, reject things. Spontaneous inhibitions of fre-


27/32

Development o f Negation 181tural and verbal self-prohibition, a subsequent development of great impor-tance for understanding the development of negation semantics.

TheFirst Uses of NegationWe have already seen that the first meaning of conventional negationconveyed by children is rejection. Spitz (1957) claims that prohibitionrefusal is invariably the first use of negation, but the longitudinal studies ofnegation here do not support this claim. Several children first used negationto reject food at feeding, while others first used rejection negation inresponse to actions such as diaper changing, or in response to ritual ques-tions such as "Do you want a drink?" Children's initial extensions of nega-tive symbols to new contexts were confined to these variants of rejection-of action, object, prohibition, or question (also see Carter, 1974).Soon after the establishment of rejection negation, the use of negation inself-prohibition appeared for all but one of the six children studied. Manycareful observers of child development have noted early instances of nega-tion, whether in gestural or verbal form, which occur when the child isabout to touch or is touching a "forbidden object" (Bloom, 1973;

Bruner, Caudill, and Ninio, 1976; Edwards, 1978; Escalona and Corman,1971; Greenfield and Smith, 1976; Piaget, 1962; Spitz, 1957). Such adevelopment is typically brief in duration.All of the examples of "self-prohibition" cited in this literature involveacts which the child has been forbidden to do in the past, and what appearsto occur is that the child issues a self-command to not do what he or she isabout to do or is in the course of doing. As an operational definition, "self-prohibition" is somewhat of a misnomer for this behavior, because the childoften touches the object anyway. The situation might be more accuratelydepicted by saying that the child is saying to self what the parent has saidwhen the same thing has been done in the past. But what is the import ofthis observation?Past interpretations of this phenomenon have been psychoanalytic innature. Escalona and Corman ( 1971) remark that self-prohibition demon-strates that the child has "internalized something like a conscience." Spitz's(1957) interpretation of this type of negation follows that of Anna Freud(1952) in her discussion of preliminary phases of superego development.He suggests that the child has assumed the role of prohibitor and is engagedin the make-believe game of carrying out the forbidden action playing egoagainst self. Whether one accepts the "conscience" or "self-awareness"interpretation or neither, the use of "no-no" or headshake when in themidst of a forbidden action is a significant step on the way to the develop


28/32

182 R o y D . Peasort but only the child's internal decisions of "I want this; I don't wantthis." By externally rehearsing the two roles of the prohibited, socially con-strained act, the child is in sequence playing out the roles of self-as-action-proposer and other-as-action-constrainer. The awareness of this contrast ismost striking when the child actually stops the action as if the parent hadbeen the one to say "no" rather than self, as several observers have noted(see Greenfield and Smith, 1976).Knowledge of this contrast between affirmative and negative messages isthus displayed; the child both initiates and negates the initiation. Sowhereas in the past the affirmative and negative were conveyed in the rela-tionship between parent and child, it has now become transmuted to withinthe self.Such transmutation has several important consequences. One is the inter-nalization of conventional norms for permissible-nonpermissible acts, theother is the fact that this contrast of affirmation-negation is internally repre-sented with the linguistic "no" as negative form. This may be comparedwith the cognitive interpretation of "egocentric speech" in preschoolers byVygotsky (1962), who argues that such speech serves an external planningfunction which, when it diminishes in frequency and finally disappears, hasin fact become internalized and mediates thought processes. The develop-ment of all higher mental functions, including concept and word meaningformation, is in Vygotsky's (1978) theory a result of the transmutation ofthe interpsychological, or, socially accomplished process to the intrapsycho-logical, or, internally regulated process. In this theory, we view "self-prohi-bition" as a turning "inward" of the externally accomplished interpersonalnegation of prohibition. "Self-prohibition" provides an external index, soonto disappear, of the child's conceptualization of an affirmative-negative con-trast that incorporates external social norms extrinsic to the child's immedi-ate need satisfaction. This internal binary contrast provides one of the con-ceptual bases for later negations of judgment involving truth-conditions forutterances.The child later comes to use such truth-functional negation to expressjudgments on the use of language by others to predicate properties andexistence of things. Edwards (1978) and Bateson (1968) have also pre-sented proposals relevant to the phenomenon of self-prohibition and conso-nant with this Vygotskian orientation.Edwards (1978) suggests that one important source of early wordmeanings for the child are the constraints that the adult imposes on thechild's actions by social prohibition, as well as those imposed by the naturalconstraints of the physical world. He discusses cases where the child comesto use words in contexts which seem to derive from the use of the same


29/32

Development o f Negation 183Edwards focuses on three realms of meaning: negation, possession, and

several words "which in the adult grammar would be classified as Verbsand Adjectives." He shows the close connection between prohibitions andthe child's use of rejection negation and of possession. "No," "no touch,""mummy's," and "don't," among other words and phrases, are all used byone of the children in what appear to be self-prohibitive contexts. The childproceeds to touch objects that have been prohibited by the parent in thepast, such as her parents' books, glasses, pencil, and watch. When the childdenotes "ownership" of the prohibited objects in her vocalizations, Edwardssuggests that this is due to a primitive ownership notion based upon pre-vious impositions that have conveyed the idea of "privileged access" to cer-tain objects by others. One example clearly shows that these social con-straints are basic to the child's conception of possession at this point (A isthe child, S is the mother, E is the experimenter) :

Language ContextA: No touch. [A looking at E's tape recorder]S: No touch.No.That's correct.A: Mummy ' s tape. [A pointing and looking at the taperecorder]

The tape recorder, A has been told many times before, is not to be touched.Alice's mother had never been to a recording session before and did notown a tape recorder, yet the child assimilated the forbidden object to hergeneral schema of objects her mother possesses which are not to be touched.

Similar constraints are involved in A's use of the word "leave" in self-prohibitive contexts,which seems to derive from her parents' phrase "leaveit" or "leave it alone," rather than being a Verb. In addition, words onemight view out of context as Adjectives convey quite different meanings incontext, such as "big" when two objects will not fit together (i.e., one ofthem is "too big"), "sharp" for a kitchen knife that is a forbidden object,and "stuck" when two objects will not go together as a child wishes. Mostof these words involve physical constraints imposed on the child's actions.

So from a radically different perspective than the psychoanalytic accountof Spitz (1957), Edwards (1978) draws the similar conclusion that con-straints on the child's actions provide an early source of word meaning andthat such a notion of contraint is psychologically real as a component of thechild's early word meanings for negation stemming, ultimately, from thecommunicational contexts of social constraints on the child's actions.

Edwards' data highlight one area of early word meaning where the effect


30/32

184 Roy D. easociocultural conventions for negation use. Obeying negation prohibition isa salient early affirmative context for negation.T h e second p roposal is Bateson's ( 1 9 6 8 ) , where, in tracing a speculativeevolution of arbitrary denotative "naming" and truth-functional negationfrom an earlier iconic signal code, he suggests that "It appears likely thatthe evolution of the simple negative aro se by introjec tion o r imitation of th evis-a-vis, so that 'not' was somehow derived from 'don't' " (1968, p . 62 6) .Bateson does not mention self-prohibition, b ut the "don't" negation hedescribes as accomplished in interaction patterns when one organism pro-poses a pa ttern of action that a no ther forbids with "don't" is wh at we wehave called prohibition. Bateson proposes, following the passage citedabove, that we "look for the evolutionary roots of the simple negativeamong the paradigms of interaction."Though Bateson's hypothesis refers to processes of phylogenesis, we ar eclearly dealing with an analogous problem in tracing the ontogenetic se-quence fro m the "no" that serves the "don't" function of rejecting, to th euse of "no" and "not" for expressing truth-functional judgments. I havesuggested that the constraint on actions serving as a source of early negationmeaning becomes internally represented an d that "no" as constraint pro-vides the cognitive index, or, root negative conception necessary for thelater use and understanding of truth-functional negation. The child's prog-ress in establishing the metalinguistic nature of such mature negation isreflected in the progressive severing of th e tem poral link between topic an dutterance of negation, which is accomplished through development in cog-nitive representation. Constraints on children's behavior provide the seedsfo r the semantics an d pragmatics of negation.

ReferencesAltmann, S . A. The structure of primate social communication. In S. A. Altmann(Ed.), Social Communication Among Primates. Chicago: University of ChicagoPress, 1967, 325-362.Bateson, G. Redundancy and coding. In T. A. Sebeok (Ed.) , Animal commur~icatio~Techniques of study and results of research. Bloomington, Indiana: Indiana Uni-versity Press, 1968, 614-626.Bellugi, U. The acquisition of the system of negation in children's speech. Unpub-lished doctoral dissertation, Harvard U niversity, 196 7.Bloom, L. Language development: Form and function in emerging grammars. Cam-bridge, Mass.: MIT Press, 1970.. One word at a time: The use of single-word utterances before syntax. JanuaLinguarum, Ser. M inor: No. 154. Th e Hague: M outon, 1973.


31/32

Development o f Negation 185Biihler, C. and Hetzer, H. Testing children's development from birth to school age.Translated by H. Beaumont. London: Allen & Unwin, 1935.Carter, A. L. T h e developm ent of c om m uni cat ion in the presensorimotor period.Unpublished doctoral dissertation, University of California, Berkeley, 1974.Cherry, L. The role of adults' requests for clarifications in the language development

of children. In R. 0. reedle (Ed.), Discourse processing: Multidisciplinary per-spectives vol. 2. Hillsdale, N.J.: Abex Publishing Co., 1979.Cromer, R. F. The development of language and cognition: The cognition hypoth-esis. In B. Foss (Ed.), New perspectives in child development. London: PenguinBooks, 1974, 184-252.Darwin, C. T h e expression of the emotions in m a n and animals. New York: ThePhilosophical Library, 1955, originally published 1872.De Villiers, J. G. and Flusberg, H. B. T. Some facts one simply cannot deny. lournalof Child Language, 1975, 2, 279-286.Dummett, M. Frege: Philos ophy of language. New York: Harper & Row, 1973.Edwards, D. Constraints on actions: A source of early meanings in child language.In I. Markova (Ed.), T h e social con text of language. New York: Wiley, 1978.Escalona, S. K. and Corman, H. H. The impact of mother's presence upon behavior:The first year. Hu man Deve lopment , 1971, 14, 2-15.Freud, A. A connection between the states of negativism and of emotional surrender.International lournal of Psychoanalysis, 1952, 33.Garvey, C. Contingent queries and their relations in discourse. In E. 0. eenan (Ed.),Studies in developmental pragmatics. New York: Academic Press, 1978.Gearhart, M. T h e cooperative construction of social episodes: A developmentalanalys is of role play in peer dyad s. Unpublished doctoral dissertation, Graduate

Center, City University of New York, 1979.Givon, T. Negation in language: Pragmatics, Function, Ontology. In Cole, P. (Ed.),Syntax and semantics. Pragnatics vol. 9, New York: Academic Press, 1978.Gopnik, A. The development of non-nominal expressions in 1-2 year-olds: Why thefirst words aren't about things. Paper presented at the First International Congressfor the Study o f Child Language. Tokyo, August 7-12, 1978.Greenfield, P. M. and Smith, J. H. T h e structure of com mu nica tion in early langriagedevelopment. New York: Academic Press, 1976.Harrison, B. M ean ing and structure: A n essay in the philosophy of language. NewYork: Harper & Row, 1972.Jakobson, R. Motor signs for 'yes' and 'no.' Language in Society , 1972, 1,91-96.Jespersen, 0. egation in English and other languages. Copenhagen, 1917.Kant, I. Imrnanuel K ant's critique o f pure reason. Translated by N. K. Smith. Lon-don: MacMillan, 1963, originally published 1787.Keenan, E. O., Schieffelin, B. and Platt, M. Prospositions across speakers and utter-ances. Paper and Reports on Child Language Development, 1976, 12, 127-143.Labov, W. and Fanshel, D. Therapeutic discourse: Psychotherapy as conservation.New York: Academic Press, 1977.Leopold, W. F. Speech development oj a bilingual child: A linguist's record. Vocabu-lary growth in th e first t w o years, vol. 1. Evanston, 111.: Northwestern Uni-

versity Press, 1939.- Speech development of a bilingual child: A linguist's record. Grammar andgeneral problem s in the first t w o years, vol. 3 . Evanston, Ill.; Northwestern Uni-versity Press, 1949.Lewis, M. M. Lang uage, thought and personality in infa nc y and childhood. NewYork: Basic Books, 1963.Marshall, J. C. Can humans talk? In J. Morton (Ed.), Biological and social factorsin psycholinguistics. London: Logos Press, 1970, 24-52.McNeill, D. T h e acquisit ion o f language. New York: Harper & Row, 1970.


32/32

Roy D. eaMiller, G. A. and Johnson-Laird, P. N. Language and perception. Cambridge, Mass.:Harvard University Press, 1976.Nelson, K. Structure and strategy in learning to talk. Monog. Soc. Res. Child Dev.1-2, Serial No. 149, 1973, 38.. Cognitive development and the acquisition of concepts. In R. C. Anderson,R. J. Spiro, and W. E. Montague (Eds.), Schooling and the acqrtisition o f kno w l-edge. Hillsdale, N.J.: Lawrence Erlbaum Associates, 1977, 215-239.Ninio, A. and Bruner, J. S. The achievement and antecedents of labelling. Journalo f Chi ld Language, 1978, 5, 1-15.Pea, R. D. T he deve lopm ent of negation in early child language. Unpublished doc-toral dissertation, University of Oxford, England, 1978.. Logic in early child language. A n n . N.Y. Acad. Science, forthcoming.Piaget, J. Play, drea ms , and imitation in childhood. Translated by C. Gattegno andF. M. Hodgson. New York: Norton, 1962.Quine, W. V. 0 . W or d and object: Cambridge, Mass.: MIT Press, 1960.. Th e roots o f re ference. New York: Open Court, 1974.Ruke-Dravina, V. The emergence of affirmation and negation in child language:

some universal and language-restricted characteristics. In K. Ohnesorg (Ed.),C oll oq uiu m Paedolinguisticum. The Hague: Mouton, 1972, 221-241.Ryan, J. Early language development: Towards a communicational analysis. InM. P. M. Richards (Ed.), The integration of a child into a social world. London:Cambridge University Press, 1974, 185-213.Schank, R. C. and Abelson, R. P. Scr ipts, plans, goals, and understanding. Hillsdale,N.J.: Lawrence Earlbaum Associate, 1977.Searle, J. R. Spee ch acts: A n essay in the philosophy of language. Cambridge: Cam-bridge University Press, 1969.. A classification of illocutionary acts. Language in Society, 1976, 5, 1-23.

Sebeok, T. A. Coding in the evolution of signalling behavior. Behavioral Science,1 9 6 2 , 7 , 4 3 0 - 4 4 2 .Shantz, C. V. The development of social cognition. In E. M. Hetherington (Ed.),Rev iew o f child development research vol. 5 . Chicago: University of Chicago Press,1977.Spitz, R. A. N o and yes: O n the genesis of huma n com munic atior~.New York: In-ternational Universities Press, 1957.Steffenson, M. S. Satisfying inquisitive adults: Some simple methods of answeringyes /n o questions. Journal of Ch ild Langua ges, 1978, 5, 221-236.Stokes, W. Children's replies to requests for clarification: An opportunity for hypoth-esis testing. Paper presented at the First annual Boston University confe ren ce o nlanguage development, October, 1976.Volterra, V. and Antinucci, F. Negation in child language: A pragmatic study. InE. 0. Keenan (Ed.), Studies in developmental pragmatics. New York: AcademicPress, 1978.Vygotsky, L. S. M ind in society: Th e de ve lop me nt of higher psychological processes.Edited by M . Cole, V. John-Steiner, S. Scribner, and E. Souberman. Cambridge,Mass.: Harvard University Press, 1978.. Thought and language. Translated and edited by E. Hanfmann and G . Vakar.Cambridge, Mass.: MIT Press, 1962, originally published 1934.

Wason, P. C. The contexts of plausible denial. Journal of Verbal Learning & VerbalBehavior, 1965, 4 , 7-1 1.Werner, H. and Kaplan, B. Symb ol formation: A n organismic-developmental u p-

A5 Pea 80 Negation

Documents

Transcript of A5 Pea 80 Negation