E,XPLANATIONS: A PRAGMATIC BASIS FOR EARLY CHILDCOMPETENCE

Maria Si lvia Barbieri . Federica Colavita and Nora Scheuer

l . Exp lana t ions : wha t a re they?

Explanat ions occur f requent ly in every-day l i fe . In sp i te ofthis, the concept of explanation is not easy to define. Within a givencontext , u t terances or sequences of u t terances can be in tu i t ive lyrecognized as explanat ions by people par t ic ipat ing in or observ ingthe interactional si tuation. There are explanations via examples (forexamp le , 'Th is i s a squ i r re l . L i ke the one you saw las t week in thepark'), explanations via definit ions (for example, 'These are patches.Do you know what they are for? To mend holes in c lo th ' ) , orexplanat ions by means of cause-ef fect re lat ions ( for example, ' Ice

crearn melts because of the heat ') . Various and converging cues cancon t r ibu te to th i s ou tcome: l i ngu is t i c cues , such as ce r ta inconnec t i ves (because) ; pa ra l ingu is t i c cues such as in tona t ion inspoken language and punctuation in writ ten language (for example,"Mommy and daddy are shopping. They buy bread, meat and fruits:apples, apr icots , p lums, and bananas") ; text coherence ind icators(consider, for example the dif ferent causal emphasis of these twosentences: 'she had a chi ld and got married' versus 'she got marriedand had a chi ld ' ) ; and last ly , contextual ind icators which cantransform the act of showing into one of explaining (as happenswhen the speaker is more competent than the partner). In order tounderstand what al l these phenomena have in common, we apply apragmat ic approach to the not ion of explanat ion. Our approachfocuses mainly on the interpersonal aspects and dif fers from thel ine of thought which interprets explanations as a mere search forcauses. This la t ter l ine, which is best represented in developmenta lp s y c h o l o g y b y t h e w o r k o f P i a g e t ( 1 9 2 3 ; 1 9 2 4 ) , s t u d i e sexplanat ions as a problem concern ing psychology of reasoning.

L P r A P a p e r s i n P r a g m a t i c s 3 , N o . 1 ( 1 9 8 9 ) , L 2 9 - L 5 4

:

1 3 1

Our l ine of research, instead, takes into account some basicfeatures of the concept of explanation in natural interaction: l . Theinteractive dimension, i .e. the fact that someone explains somethingto someone e lse; otherwise the d is t inct ion between expla in andunderstand wi l l be missed. 2. A d i f ference of knowledge andauthority between the one who offers and the one who receives theexplanation. This lack of knowledge in the one who receives theexplanation just i f ies the act of the speaker according to Grice'sconversat ional maxims ( 1975) which request not to say too muchnor too l i t t le. 3. The aim of restructuring the partner's organizationof information.

These considerations al low us to try a definit ion of the act ofg iv ing an explanat ion on which we are focusing and to speci fythe condi t ions of i ts successfu l per formance in conversat ion. Wedefine giving an explanation as fol lows: i t is an interactional movewhich takes p lace when one par tner of fers a p iece of newinformation (explanans) which refers to an object of joint attention(explanandum). This new informat ion makes c lear what waspreviously obscure. The need for i t may be direct ly expressed bythe par tner by verbal or non-verbal means, or i t may bepresupposed by the speaker for the sake of goals related to the on-going interaction. This tentat ive definit ion al lows us to include inthe concept of explanat ion facts f requent ly tak ing p lace inin te rac t iona l con tex ts wh ich a re in tu i t i ve ly recogn ized asexplanat ions. For example, just i f icat ions, paraphrases of lex ica lmeanings, definit ions of objects and indications of functions.

Our I ine of research has been developed fo l lowing authorswork ing on ord inary explanat ions (von Wright 197 | ; Antak i &Fie ld ing l98 l ) who have d is t inguished two forms of explanat ions:descr ip t ive explanat ions, which expla in something by g iv ing i t aname that the audience wi l l understand; and causal explanat ions(o r agency exp lana t ions when they re fe r to human ac t ions ) ,de te rmin ing why someth ing has happened ( fo r an ana logousdis t inct ion in cogni t ive psychology see a lso Caste l f ranchi & Par is iI 9 8 0 ) .

I

L 3 2

Our work has both theoret ical and empir ical purposes. On thetheoret ical level , we intend to ident i fy the condi t ions that need tobe ful f i l led when giv ing an explanat ion. On the empir ical level , weh a v e c a r r i e d o u t a q u a l i t a t i v e i n v e s t i g a t i o n s t u d y i n g t h ee x p l a n a t i o n s g i v e n b y t h r e e - y e a r - o l d c h i l d r e n i n s p o n t a n e o u sinteract ion and explor ing which behavioural cues could be takeni n t o a c c o u n t t o d e c i d e w h e t h e r t h e s u b j e c t s h o n o u r t h e s econditions. Though we think that our model of explanation and thecondi t ions we propose are general ly val id, we chose three-year-olds because they are l inguist ical ly competent enough to be easi lyunderstood, while their early age allows us to investigate which arethe most basic features of explanat ion in interact ion.

2 . Cond i t ions de f in ing the ac t o f exp la in ing

Fol lowing the l ine establ ished by Sear le 's work on promises( 1 9 6 9 ) a n d i n f l u e n c e d b y t h e c o n s i d e r a t i o n s o f S b i s a ' ( 1 9 8 9 ) o neveryday explanat ions, we here propose the set of condi t ions wethink an ut terance must fu l f i l in order to count as a sat isfactoryp e r f o r m a n c e o f a n e x p l a i n i n g a c t , d i s t i n g u i s h i n g b e t w e e npreparatory, essent ia l , s incer i ty and af f i l iat ive condi t ions.

Preparatory condi t ions lay the appropr iate grounds for thespeech act. Two types were distinguished:

1. The speaker should bel ieve that the new informat ion isnecessary to the partner 's knowledge or understanding. In otherwords, the speaker should believe that the partner does not knowthe explanans or that is not able to establ ish i ts relat ion wi th theexplanandum by him/hersel f .

2 . When exp la in ing , the speaker must be sure tha t theexplanandum is at the centre of the partner's attention.

As regards preparatory condi t ion 1, the need for theexplanat ion may be expressed ei ther by a direct request f rom thepartner, or by non verbal means, or by contextual cues indicatinglack of understanding. I f th is type of preparatory condi t ion is nothonoured (nei ther the explanans nor i ts l ink wi th the explanandum

r

1 3 3

a r e n e w ) , t h e n G r i c e ' s m a x i m s f o r b i d d i n g r e d u n d a n c y i nconversa t ion i s v io la ted . In adu l t conversa t ion , overexp la in ing mayacqui re the connotat ion of an insul t , s ince i t impl ies an assumpt ionof the par tner 's d i f f icu l ty in understanding.

If preparatory condit ion 2 is not fulf i l led (partner's attentionis m iss ing ) , then commun ica t ion fa i l s . The par tne r w i l l no tunderstand the explanation, i .e. he/she wil l ei ther not notice that anexplanat ion has been g iven, or wi l l not be sure about the ent i tyrefened by the explanans.

W e h a v e a l s o d i s t i n g u i s h e d t w o t y p e s o f e s s e n t i a lc o n d i t i o n s :

l . The explanat ion must be const i tu ted of both an explanandumand an explanans.

2. The speaker must show concern for the partner's inferentialp a t t e r n s .

I t is important to d is t inguish essent ia l condi t ion I f rompreparatory condi t ion 2, that is , f rom the requi rement that theexplanandum should be an object of jo in t a t tent ion. Essent ia lcondit ion I requires, instead, that the explanandum, in order to beconsidered as such, must be l inked to an explanans. The relat ionbetween explanandum and explanans is recognized e i ther byinference, or because it is explicitly expressed with a connective.

Essential condit ion 2 indicates the speaker's concern for thepartner's way of reasoning. The speaker must build an explanationthat will make sense to the partner. This type of condition is notfulf i l led when partners do not share a common understanding ofreality. If the speaker's assumptions are not shared by the partner,the explanat ion wi l l be considered absurd, unreasonable orunaccep tab le .

Sincerity condit ions refer to the fact that the speaker mustconsider him/herself competent to give an explanation on a certainmatter. It is this subjectively felt authoritativeness which gives thepower to become the explainer. This power may have dif ferentsources. I t may derive from factors external to the interactionalsituation, as when the speaker is an expert on the topic beingexplained, or i t may develop progressively within the interactional

-

L34

si tuat ion, ss when one of the par tners turns out to be morecompetent in that f ie ld . I f these condi t ions are not honoured, thenthe explanation is deprived of i ts value and may be even refused. I tis to be noticed that this condit ion does not refer to the truth-valueo f the speaker ' s s ta tements , bu t to h i s /he r au thor i t y andcompetence as opposed to the partner's ignorance. In this situationthe partner is given the "duty" of taking into account the new pieceof information (Sbisi 1989). We may assume that this condit ion ishonoured by both par tners as far as the explanat ion is notc h a l l e n g e d .

Meet ing af f i l ia t ive condi t ions means honour ing the socia lre la t ion be tween par tne rs . These cond i t i ons re fe r to thesymmetr ica l -asymmetr ica l ax is of the re lat ionship, der ived f romsocia l ro les. This convent ional ly at t r ibuted author i ta t iveness makesit more l ikely for one partner to become the explainer. In anasymmetrical relat ionship, such as that between chi ld and adult, orbetween student and teacher, i t is easier for the one who is in a"upper" posit ion to give an explanation and for the one in an"lower" posit ion to accept i t as such. I f the one in an " lower"posit ion gives an explanation, he/she might even feel compelled toadapt h is /her language to the s i tuat ion by us ing, for instance,caut ious express ions. In a symmetr ica l re lat ionship, on the otherhand, both partners have the same chances to act as explainers.

3 . Emp i r i ca l ev idence

3.1. MethodThe subjects of our analysis were three Italian girls aged three

years (Si lvia, Carmela and Serena) attending a kindergarten nearReggio Emil ia. We observed them in dif ferent si tuations: free play,playing with puzzles and looking at a picture book. Each of thesesituations was videotaped twice. A f irst t ime, for 20 minutes, withthe teacher par t ic ipat ing act ive ly , and a second t ime, for l5minutes, wi th the teacher p lay ing a passive ro le. The v ideotapes,last ing in to ta l 105 minutes, were t ranscr ibed at tent ion having

I

1 3 5

been pa id to non-verba l behav iour as we l l . The to ta l co rpusamounted to abou t 2 ,400 u t te rances d i f fe ren t l y d rs t r ibu tedbetween adul t and chi ldren depending on whether the adul t wasactive or passive. In the adult-act ive situations, the adult producedabout 1,200 ut terances unevenly d is t r ibuted between book reading(540), free play (320) and puzzles (360) respectively, while thethree chi ldren al l together produced about 670 utterances with nod i f fe rences among the th ree s i tua t ions . In the adu l t -pass ives i t u a t i o n s c h i l d r e n p r o d u c e d a b o u t 4 1 0 u t t e r a n c e s e q u a l l yd is t r i bu ted among the th ree s i tua t ions ; the adu l t , who wasins t ruc ted to in te rvene as b r ie f l y as poss ib le and on ly whenaddressed, produced only 65 utterances.

Though we think that the concept of explanation encompasses awider range of phenomena besides explanations containing a causalpredicate, we decided to analyze in this work only this last typebecause they can be selected with clear and independent cri ter ia.In order to select explanation we applied a sl ightly adapted versionof Hood and Bloom's method (1979: 4). We adopted either formalcri teria, l ike expl ici t causal connectives, or semantic cr i ter ia, l ike animplici t reference to two events or states having a causal relat ionbetween them. Unlike them, we accepted that one part of theexplanation (usually the explanandum) could be expressed by non-verbal means, or even be a signif icant contextual event. After thisselect ion we carefu l ly examined the in teract ional s i tuat ion whichpreceded and fol lowed each explanation, looking for cues whichwould indicate whether the condit ions we set were honoured bvboth speaker and partner.

3.2. Criteria of analysis and examplesThe honour ing of these four condi t ions may be studied by

examining the in teract ional context of the explanat ion" In somecases this test may be done in an easy and direct way, as when theexis tence of an object of jo in t a t tent ion is ascer ta ined. At othert imes, the fact that cer ta in condi t ions have been respected mayonly be in ferred f rom the par tner 's behaviour . For instance, the

-

r 3 6

par tne r ' s accep tance o f the exp lana t ion shows tha t i t wasnecessary . On o ther cases , he /she may re fuse the exp lana t ion ,somet imes even ind ica t ing wh ich o f the cond i t i ons has beenviolated: a bored ' l already know' points out that the information isnot new (fai lure of preparatory condit ion I ) , whereas an annoyed'Don't talk about things you don't know about' shorvs that sinceritycondit ions have not been fulf i l led. We wil l now present the cues werel ied on in order to determine whether , and how, a l l fourcond i t i ons were observed by our sub jec ts du r ing in te rac t ion ,i l lust rat ing them wi th some examples.

3.2.1 . Preparatory condi t ionsP r e p a r a t o r y c o n d i t i o n 1 ( t h e s p e a k e r s h o u l d b e l i e v e t h a t

the partner does not know the explanans or is not able to establ ishthe relat ion between explanans and explanandum by him/herself).

The clearest possible cue is a request for an explanation. Whenthis does not occur, we can rely on indicators that the speaker mayhave no t i ced , e i the r non-verba l ( i . e . the par tne r ' s puzz ledexpression) or verbal ( repet i t ions, misunderstandings) whichindicate that the partner has not understood something.

When none of these cases occur, however, we must considerwhat the explanat ion spontaneously of fered is about . I f thespeaker 's explanat ion refers to h is /her own wishes, goals orintentions, then i t is l ikely that he/she assumes that the partnerdoes not know what he/she has in mind.

Last ly , an unexpected event which const i tu tes a cogni t iveproblem also prepares the ground for an explanation, since thespeaker may assume that it constitutes a cognitive problem for thepartner too.

Our subjects do not recognize the situation which is apparentlythe most favourable: a request for an explanation. They did notanswer any of the explicit requests made by the adult, nor thclsemade by another chi ld (except in one case which we wil l examinela te r ) .

1 ^ -L - t /

( I ) Free play in the kitchen. ' f he teacher offers rhe chi ldrensome real bread.

Serena: Oresto non si pud mangiare'You can't eat this'

Teacher: Ouesto si. Perchd?'Of course you can. Why not?'

Serena: . . . . .(does not answer)

Teacher: Ouesto si. si pub mangiare'You can eat it. '

(2) The gir ls are doing j igsaw puzzles. A piece gers sruck andSilvia is not able to take it out.

Si lvia: Perchd non si togl ie questo qui?'Why doesn't this come out?'

Serena: . . . . .(does not answer)

Instead, all three girls seem capable of offering explanationsspontaneously. The problem here is whether their explanations aresimply a verbal expression of what they are thinking, or whetherthere is some concern for the partner's needs. The opportunity foran explanation may be grasped not only from a direct r"quesi, butalso from partner's cues such as misunderstandings or repeti t ionswhich indicate she has not understood something.

(3) Free play in the kitchen.

Silvia: (to Serena) Vuoi mangiare?'Do you want to eat?'

r

l ,3 B

Serena: (to Silvia) Sono piena'I'm full. '

Carmela: (to Silvia) Dh da mangiare a Serena'Give Serena something to eat.'

Silvia: (to Carmela) Ma no/ che d piena!'Oh, no! / She's full. ' 1*;

Carmela's insistence shows that she has ignored Serena'sprev ious answer, mot ivat ing Si lv ia 's refusal ( 'Oh, no! ' ) wi th theexplanans that just i f ies i t ( 'She's ful l ' ) .

Most frequently the need for an explanation derives from thefact that the child is justifying his/her own needs, wishes, intentionsor requests. In such cases the chi ld's being expl ici t comes frompresupposing that the partner is not capable of inferr ing i t byh i m / h e r s e l f .

(4) The girls are looking at a book.

see the book) /

' l l

can't see, I can't see. I said I can't see! ' (She explains why shepushed Si lv ia) .

In the following example Serena justifies her refusal to

(*) The bar (D which appears in the examples always precedes thee x p l a n a n s .

Serena: (she pushes Silvia who is not letting herIo non vedo. io non vedo. Ho detto non vedo!

1 3 9

immediately exchange puzzles by rendering explicit her intentionto finish the on-going activity.

(5) Each girl is solving a puzzle. Silvia has finished hers andwants to exchange puzzles with Serena.

Silvia: Facciamo cambio. Sere?'I-et's switch, Sere.'

Serena: Aspetta che finisco'Wait

/ I haven't finished.'

Finally, the need for an explanation may be created by anunexpected or problematic event in the context.

(6) This situation immediately fol lows example (2).Si lvia hasreferred several times to her piece of puzzle being stuck. Sheasks the teacher for help, receiving as an answer the advice totry to do it by herself. She also asks Serena for help, who doesnot answer (see example (2)). Finally she bursts out:

Silvia: Perchb non si toglie questo qui?.../Ci vuole un martello!'Why doesn't this come out? / We need a hammer.'

The explanans "we need an instrumentarrived at by the problem resulting fromefforts.

Here we present another example of ane v e n t .

to exercise force" isseveral unsuccessful

unexpected significant

-

L40

(7) The gir ls are playing together. Carmela leaves annoyed afterbeing pushed by one of her partners.

Serena: (to Silvia)/Le hai fatto male!'/ You hurt her!'

Carmela's leaving is explained by Si lvia being blamed forhaving hurt her.

Preparatory condi t ion 2 (when expla in ing, the speaker mustmake su re tha t the exp lanandum is w i th in the f i e ld o f j o in tat tent ion) . There are two s i tuat ions in which jo in t a t tent ion isalready guaranteed: in the case of face to face interaction and in thecase where one partner verbal izes the explanandum and the otherone verbal izes the explanans. When joint attention is not alreadyguaranteed, it is the speaker who must establish it. In order to openinteraction and cal l for the partner's attention, the speaker mayeither verbal ize the explanandum, ensure attention by non verbalmeans, or combine verbal and non-verbal means.

In the fol lowing two examples, joint attention is alreadye s t a b l i s h e d .

(8) Carmela has unsuccessfullyt imes, by both verbal and nonpull the book away from Silvia,

Carmela (to Silvia) Me lo dai?'Can I have it?'

S i l v ia : ( to

asked Silvia for the book severalverbal-means: she then tr ies tosaying

Lasciami./ un po' io,*'Gimme it. / It's my turn.'

Carmela, covering the book with her arm)

I

Nooo...Not now/ I'm reading.'

L4L

In this case, since they are both handling the book, jointattention is of course guaranteed. From another point of view, theexample is in terest ing because Carmela 's explanat ion ra ises thelevel of the interaction: after Carmela's explanation, Si lvia feelscompe l led to exp la in her pos i t i on too . Dur ing the p rev iousinteraction they had opposed each other with mere (unexplained)requests and refusals.

Another case in which joint attention is already establishedoccurs when the explanandum is verbalized by one partner and theexplanans by the other, as in the following example:

(9) Serena tries to put in a piece of puzzle.

Serena: (to the teacher) Non ci vuole entrare dentro'It doesn't fit.'

Teacher: lPerchd lo sai che bisogna girarlo dal l 'al tra parte.sembra uguale..

the same...'

'/Cause' you knowr )ou have to turn it round' It looks

Serena: (she tries and the piece fits; she smiles) C'd andato'That's it. '

This is a very special example because here the partner is theadult. We did not find a similar example in child-child interaction,perhaps because of the early age of the participants.

Now we turn to a situation in which joint attention must beestabl ished. One means of achieving this is by verbal izing thee x p l a n a n d u m .

(10) Free play in the kitchen. The girls keep certain distancefrom each other.

r

I u /

Silvia: (to Serena, while putt ing on an apron)posto i l erembiu le. / s iamo le cuoche!-

an apron / We are cooks! '

Dopo ci mett iamo a

''fhen we put on

Even if a shared game had been taking place, the act of puttingon the apron would probably have not been enough to ensure thepar tne r ' s a t ten t ion , s ince she was no t l ook ing a t S i l v ia . Byverbal izing the explanandum this is achieved. From another point ofv iew, th i s case i s i n te res t ing a lso because i t shows how g iv ingexplanat ions dur ing p lay in teract ion helps the speaker to take thelead in the game. Hisiher proposals carry more weight and seemmore sensib le when just i f ied.

The chi ld may also combine verbal and non-verbal means inorder to attract the partner's attention.

( I I ) Serena t r ies to open a t in . She does not succeed andapproaches the teacher, showing her the t in.

Serena: (to the teacher) Me lo apri. che non ci riesco?'You open it I I can't.'

Or the chi ld may use non-verbal means only.

(12) Serena: ( in another moment ,A{on ci riesco ad aprire' fCan't open i t . '

showing the teacher a t in )

r 4 3

The gesture of showing the t in const i tu tes the explanandum,i .e . a non-verbal request for help. The explanans, on the other hand,i s ve rba l and jus t i f i es the reques t by re fe r r ing to her ownh e l p l e s s n e s s .

We can infer from our data that preparatory condit ions areusually fulf i l led by chi ldren at this age, except in those cases wherethe need for the explanation is expl ici t ly verbal ized by the parrnerthrough a d i rect request . This corresponds wi th the developmentsequence proposed by Hood and Bloom (1919) according to whichthe chi ld f i rs t makes causal s tatements, then answers to requestsfor explanat ions and f ina l ly makes requests for explanat ions. Hood& B loom's accoun t fo r these f i nd ings re fe rs to the fac t tha tanswer ing a request for explanat ion requi res that the chi ld be ableto encode causa l re la t ions w i th "a semant i c i n ten t ion tha t i sinf luenced by the message of the other speaker" (Hood & Bloom1979, 33 ) , o r , i n ou r op in ion , to adap t to the o ther speaker ' sv iewpoint . In the i r longi tudinal s tudy of 8 ch i ldren aged 2.6 ro 4years they found main ly causal s tatements wi th contents of denia ls ,refusals , and d i rect ions, a lmost to ta l ly re lated to ch i ldren 's ownneeds and in tent ions. Though the l inguis t ic devices for express ingcausal i ty in I ta l ian are somehow di f ferent f rom Engl ish, s ince weuse the same connec t i ve perc h d bo th fo r i n te r roga t i ve o r rddec la ra t i ve sen tences , and the syn tax fo r I ta l i an in te r roga t i vesentences is much s impler than in Engl ish, we have found s imi larresults. I t seem to us that at three years language rnainly functionsto regulate joint act ivi ty and that explanations are l ikely to occur todirect a common course of act ion f iust i f icat ions fol lowing requests,express ions o f w ishes , and in ten t ions ) o r to avo id d is rup t ions(exp lana t ions fo l l ow ing m isunders tand ings , i nadequa te ac t ions o rundes i red ac t ions wh ich cou ld a f fec t the speaker ) . There . fo re ,though most explanai ions are l inked to a speaker 's personal p lan, inthe in teract ion they turn out to be adequate and per t inent . Asregards preparatory condi t ion 2, the speaker a lways makes suretha t the exp lanandum is w i th in the f i e ld o f j o in t a t ten t ion by

r

1 4 4

var ious means , the rnos t f requen t o f wh ich i s ve rba l i z ing thee x p l a n a n d u m .

3.2.2. Essent ia l condi t ionsEssent ia l condi t ion I (presence of an explanandum and an

explanans) . Here we checked the presence of both the explanandumand the explanans. In doing this we did not l imit ourselves to verbalexpress ions, s ince somet imes par t o f the explanat ion (usual ly theexplanandum) is const i tu ted by a gesture or an act ion. We foundthat the explanandum may even be consti tuted by an event in thecontext . I f th is event ca l ls for the par tner 's at tent ion (preparatorycondi t ion 2) , i t becomes an explanandum in the case an explanansrefers to i t .

In our data, whi le the explanans is a lways verbal ized, theexplanandum is expressed in the var ious ways ment ioned above. Inmost cases i t is verbal ized, sometimes i t consists of a gesture or anaction, and in a few cases i t ei ther consists of an event or i t must beinferred from discourse. Herb we wil l only exempli fy the latterpossib i l i ty . I l lust rat ions of verbal ized explanandum and explanansmay be found in examples (3) and (5); example (4) i l lustrates anexplanandum consis t ing in a gesture or an act ion; example (7)presents an explanandum consis t ing of an event .

(13) Free play in the kitchen. Serena offers a spoon to Silvia.

Serena: Vuoi questo cucchiaio?'Want this spoon?'

Si lvia: Ma / io ce I 'ho gia' B u t / I g o t o n e . '

In this example the explanandum is a tacit refusal, whose onlyl inguist ic cue is but, a word which indicates some sort of opposit ion.

In the fol lowing example the explanandum is a completelytacit refusal which, however, can be inferred.

I

L 1 + 5

(14) Each gir l is solving a puzzle.

Si lv ia : ( to Serena)Facc iamo cambio?'Wanna swap?'

Serena: /Faccio cambio con Carmela'/ I gonna swap with Carmela.

The explanans is Serena's alternative plan, wir ich explains whyshe does not want to exchange puzzles with Si lvia.

Essent ia l condi t ion 2 ( the speaker must show concern aboutthe partner's inferential patterns). In order to explore whether thiscondit ion is observed we may use only an indirect cue, i .e. evidencein further interaction that the partner has accepted the explanation(he/she does or says something that reveals that he/she has noticedthe information which has been offered to him/her). This wouldindicate that the explanat ion seems sensib le to the par tner . Aweaker cue consists in the fact that the partner does not refuse theexplanat ion. The subjects s tudied in our research have s imi larinferential patterns since they are of the same &Be, share a similarcultural background, and have been attending the same creche for along t ime. Thus, i t was highly improbable that they would refuseeach other's explanations. We could therefore infer concern aboutthe par tner 's in ferent ia l pat terns only f rom the ind i rect cueconsist ing of the partner's behaviour.

Let us consider some cases in which the explanat ion issuccessful. In some cases the partner accepts the explanation: i .e.she acts accordingly to the speaker's request or proposal, or acceptsthe given information (nodding, repeating, etc). In other occasions,the par tner merely acknowledges the new informat ion: that is ,even i f she does not agree with the explanation (she does not dowhat has been asked to or disagrees with the given explanation),her behaviour indicates that i t has been taken into account.

r

I n examp les ( l l ) and (12) the teacher sa t i s f i es the ch i ld ' sr e q u e s t a n d o p e n s t h e t i n . I t i s i n t e r e s t i n g t o n o t i c e t h a t t h eexp lanans tha t j us t i f i es the reques t does no t have a mere ly po l i tep u r p o s e . S i n c e t h i s t e a c h e r u s u a l l y e n c o u r a g e s c h i l d r e n t o s o l v eprob lems by themse lves , Serena 's exp lanans has the purpose o favoid ing the teacher 's refusal by render ing expl ic i t in advance thatshe has a l ready t r ied by hersel f .

We now turn to the analys is of two somewhat more complexe x a m p l e s .

(15) E,ach g i r l is doing a puzzle. Serena has just f in ished hers.

S e r e n a : ( t o C a r m e l a , w h o i s s t i l l w o r k i n g o n h e r s ) F a c c i a m ocanrbio. _C.a!!C

Wannaswap, Carme' lCarmela: . . .

(does not answer)Si lvia: (to Serena) No/ faccio cambio_con te

'No / I wanna swap with you.'

Serena: (to Si lvia) No /faccio cambio con lei. Dopo. dopo facciocambio con ts

'No / I 'm gonna swap with her.. . Then you. '

Si lvia: (to Serena) Si. aspetta /che lei f inisca. /vuoi che faccia unc a s i n o ?

'O.K. Wait / she's st i l l doing hers / You wanther to make a mess of i t?'

Serena: (approaches Carmela and helps her to f in ish. When apiece fa l ls down, Serena p icks i t up) .

A remarkab le in te r lac ing o f exp lanat ions occurs dur ing th issequence. F i rs t , S i l v ia re fuses Serena 's p lan and jus t i f ies her re fusa lw i t h h e r d e s i r e t o e x c h a n g e p u z z l e s w i t h S e r e n a h e r s e l f . B u t S e r e n ai n l u r n r c l ' u s e s S i l v i a ' s p r o p o s a l a n d j u s t i f i e s t h i s r e f u s a l b y a

I

L47

rea f f i rmat ion o f he r o r ig ina l p lan . A t th i s po in t S i l v ia accep tsSerena 's p lan ( 'O .K . ' ) bu t requ i res someth ing ( 'wa i t ' ) . She a lsojusti f ies her request (Carmeia has not f inished yet) and she explainswhy i t is necessary to wait for Carmela to f inish (otherwise she wil lmake a mess of i t) . In order to decide about Si lvia's concern forSerena 's in fe ren t ia l pa t te rns we a re in te res ted in Serena 'ssuccessive behaviour . She nei ther ins is ts nor presses Carmela, butinstead helps her to f inish. In this way she shows that she has takenSilvia's explanation into account, which turns out to be successful.

(16) The gir ls are looking at a book. Carmela is sl ightly apart,while Serena complains to Si lvia, who does not al low her to seethe book because she is covering i t with her arm. Serena f irstpushes Si lv ia say ing

Serena (to Si lvia) Non__wdq (Si lvia does not move her arm).'I can't see'

Non vedo! . . .T i met to un e last ico. veh! T i met toun e last ico qua.

' I can't see... I ' l l put a rubber band on you,ah! I ' l l put a rubber band here! 'Si lvia: ( ignores this)Serena: Io t i butto fuori dal la f inestra

'I' l l throw you out the window.'Si lvia: Ma c'd freddo... / Io non ci vogl io andare. Apro la f inestra escappo dentro.

'But i t is cold, / I don't want to go there. I ' l l open thewindow and run back inside. '

Serena: (point ing at Carmela) l -e i . le i . . la but t iamo fuor i da l laf i n e s t r a

'Let 's throw her, her.. . out ihew i n d o w . '

r

I 4 B

In th is example, Si lv ia 's explanat ion ( I don' t want to go therebecause i t 's cold) is part ia l ly successful . I t b locks Serena's plan ( tothrow Si lv ia out of the window) only concerning the object to whichshe tr ies to apply i t . Af ter Si lv ia 's explanat ion, Serena tr ies to applythe same plan to a different partner, Carmela.

Another group of cases al lows us to infer that the explanat ionhas been accepted: when the partner imitates the behaviour of thespeaker (17) or repeats the explanat ion, as a whole or in part (18)

as a sign of agreement.

(17) Free play in the k i tchen.

Si lv ia: ( laughing, she cleans her hands on her apron) Adesso cioul iamo le mani /che sono cosi sporche

'Now weare cleaning our hands / they are so dirty. 'Serena: ( imitates Si lv ia, who looks at her) .

(18) Serena and Carmela are sit t ing next to each other. Theteacher places a puzzle on the table in front of Carmela.Carmela places i t between herself and Serena, as a non-verbalsuggestion to share the puzzle.

Serena: (to Carmela) No /questo b tuo'No

/ this is yours.'Carmela: Ouesto d mio. Ne vuoi un altro?

'This is mine. Do vou want another one?'

Successful explanations, such as those i l lustrated by examples(15) , (17) and (18) const i tu te the major i ty in our data.

I

L49

A third way of inferring the success of an explanation is to seewhether the same explanation is used by another partner in furtherinteraction. Some such cases appear in our data, as when Carmelasays to the teacher 'You open i t l l can ' t ' , exact ly reproducingSerena's explanation in example ( I I ) . These examples of "deferredimi tat ion" are f requent in cases of explanat ions which regulatein te rac t ion , such as 'Wa i t / I haven ' t f i n i shed ' o r 'No (don ' ttouch)/this is mine (yours) ' .

It must be said that the evidence for checking whether or notessential condit ion 2 has been honoured is the weakest both inquali tat ive and quanti tat ive terms. In about one third of the caseswe are unable to decide whether this type of condit ion has beenfu l f i l l ed , s ince these exp lana t ions a re ne i the r re fused , no rquestioned, nor accepted. I t is therefore impossible to ascertainwhether the given explanation seemed sensible to the partner.

Another means for ascertaining the explanatory success wouldbe veri fying whether the explanans refers to previously establ ishedrules and therefore to a shared conversational logic. From this pointof view it is interesting to notice that the rules used in our subjects'explanations are frequently those characteristic of the kindergarten:sharing things, taking turns, etc.

3.2.3. Sincerity condit ionsTh is cond i t i on s ta tes tha t the exp la ine r mus t cons ider

him/herself competent to give an explanation on a certain matter.We can explore how this condition is fulfil led in various ways, butmainly by analyzing what the explanation refers to. If it regards thespeaker's own feel ings, intentions, plans, capacit ies, and previousexperience, then the speaker is obviously the most authoritat iveperson on the matter. I f , instead, the explanation regards otherpeople or the physical world, then we can rely only on an indirectindicator: i.e. the fact that the partner accepts the speaker to give anexplanation about a certain matter (a counterexample would besomething l ike: 'Who do you think you are to speak about this?').A lmost a l l the explanat ions in our data refer to the speaker 'sin tent ions and p lans. Only in two examples does the explanat ion

r

1 F ^

1 ) U

refer to external ob jects : in example (6) the speaker 's author i tyderives from the fact that nobody else had wanted to take i t (Si lviahad asked the teacher and Serena for help severa l t imes, wi thoutrece iv ing any) ; i n examp le (15 ) the exp lana t ion re fe rs to thepar tner 's competence in accompl ish ing a cer ta in task, and nobodyquestions the fact that the speaker gives an explanation about thematter . In the absence of formal cr i ter ia when we analyzeexplanat ions which do not refer to the speaker 's own p lans,in tent ions, and wishes the only possib le cue is the fact that theauthority of the speaker is not questioned.

In a certain w&/, our young subjects seem concerned aboutrespect ing s incer i ty condi t ions. In the fo l lowing example Serenarefuses to explain precisely because she is not able to honour suchc o n d i t i o n s :

(19) Serena and Silvia beat the table with their hands. Carmelaimi tates them.

Serena: (to Silvia) Carmela vuole fare come noi'Carmela wants to do like us.'

Si lvia: (to Serena) Perchd?'Why?'

Serena: Io non lo so'Don't know.'

Serena is explaining the meaning of Carmela's action, but she isnot able to go further in the causal chain which would al low her tof ind a reason for this act ion. A descript ive explanation interpretsthe act ion by saying what the act ion means, but the explanat ionprocess stops at the f irst step because a reason for the action is notf o u n d .

From another point of view, the child seems to be aware in acertain way of the power derived from explaining. I t is interestingto not ice that , main ly dur ing symbol ic p lay, explanat ions acqui re

r

1 5 1

the funct ion of guid ing par tners ' behaviour . and thus of contr ibut ingto accomplish or to maintain the speaker's leadership. In fact, thechi ld who says 'we put on an apron/we are cooks' or 'now we'l lmake a pizzal let 's get the salt ' is guiding and organizing the game.

3.2.4. Aff i l iat ive condit ionsThis condit ion refers to the honouring of the social relat ion

between speaker and l is tener . Here we observed whether there lat ion between par tners was symmetr ica l (ch i ld-ch i ld) or not( c h i l d - a d u l t ) . T h e n w e e v a l u a t e d t h e r e l a t i v e a m o u n t o fexplanations the chi ld offered to each type of partner. We alsoexplored whether the topic of these explanat ions var ied accord ingto the type of partner, in order to see whether, when addressing anadul t (asymmetr ica l re lat ionship) , the chi ld expla ined th ings aboutwhich she could consider hersel f par t icu lar ly competent : s i tuat ionsin which the adul t had not been present , on-going games orreferences to the speaker's own intentions and wishes. In our data,most of the explanations were addressed to other chi ldren, whereasonly a few were addressed to the adult. When we analyzed thecontent of these latter explanations, we found that even this aspectis inf luenced by aff i l iat ive condit ions. In fact, 7 out of 8 cases werejusti f icat ions of requests for help deriving from the speaker's ownwishes or dif f icult ies. The other example was an explanation about ag a m e s i t u a t i o n i n w h i c h t h e a d u l t h a d n o t p a r t i c i p a t e d(example(17)). Therefore, we may say that not only did our subjectsgive few explanat ions to the adul t , but they a lso l imi ted suchexplanat ions to mat ters about which thei r "author i ty" was st rongerthan the adul t 's .

4 . C o n c l u s i o n s

We interpret the act of explaining as a social act occurring ina n i n t e r a c t i o n a l c o n t e x t . T h e c o n d i t i o n s w e d e s c r i b e d a b o v econst i tu te the set of ru les govern ing i ts per formance.

r

L 5 2

The qual i ta t ive analys is carr ied out , though based on a l imi tedcorpus of data, a l lows us to conclude that at three years ch i ldrenknow how to exp la in ( i . e . they mas te r the cond i t i ons and waysrequ i red to g ive an exp lana t ion ) . We cons ider neoVygo tsk ianframeworks to be the best too l for in terpret ing our resul ts as awhole. NeoVygotskian theories stress the role of social interaction inthe development of language and cogni t ion (Bruner 1983) , wi thoutdenying, however , an act ive e laborat ion of the input f rom theenvi ronment on the par t o f the subject . Severa l considerat ionssupport this theoretical choice. The f irst one is that even very youngchi ldren are able to g ive explanat ions in soc ia l in teract ion in ac o r rec t way . The second i s the fac t th a t ou r s ub j ec ts useexplanat ions main ly to just i fy speech acts of the i r own, such asrequests and refusals. As a matter of fact, according to our data,ch i ldren use explanat ions as inst ruments to in f luence and d i rectthei r par tners ' behaviour considerably more f requent ly than as ameans to inquire about the world for the satisfact ion of their own orothers' cognit ive requirements. Lastly, many of the explanations weanalyzed regard the regulat ion of behaviour according to socialrules, turns or the sharing of objects. If we accept the idea that thechi ld expla ins what she knows best , what in terests her most , andwhat serves her main goals, we may suppose that i t is only laterth a t e xp lana t ions w i l l become in s t ruments fo r bu i ld ing upknowledge about the physical world. Therefore, we hypothesize thatexplanations are - in their origins- closely l inked to human action,and to the in ten t ions and p lans o f the ind iv idua ls who a renegotiat ing within interaction. As they grow older, chi ldren wil l beable to supply explanations about a wider range of objects andsi tuat ions which wi l l be increasingly far ther away f rom thespeaker 's act ions and v iewpoints . The causal chain wi l l becomemore complex and the capacity to take into account the partner'sinferential patterns wil l sharpen. The ski l l in using verbal meanswil l also progress.

The close bond between explanation and communication canhelp us understand why the pragmatic condit ions of explaining aremastered so early. According to Hood and Bloom (1979) our subjects

I

r l q ?

were at the beginning of the i r expla in ing capaci ty . In our data, suchrequests for explanat ions were not of ten d i rected to ch i ldren, andthese requests that were made, were for the most part unanswered.There fo re , ou r in te rp re ta t ion re l i es ma in ly on the p rev iousl i terature. Never theless, the chi ldren in th is s tudy were able tog rasp the need fo r an exp lana t ion by us ing the ve ry s t rongcontextual cues which a l lowed them to g ive per t inent explanat ions.These contextual cues were not subtle signals of doubt but obviousmistakes, unpleasant act ions that affected the speaker, and requestsfor help that involved the partner direct ly.

As a f inal discussion point, we want to consider the value wemay at t r ibute to each condi t ion. Two of them, preparatory andessential condit ions, are more important than the rest in making agiven in teract ional move become an explanat ion. Preparatory andessential condit ions are those more closely l inked to the need forin fo rmat i ve exchange (p repara to ry cond i t i on I ) and to i t sef fect iveness (preparatory condi t ion 2 and essent ia l condi t ions Iand 2). The fulf i lment of sincerity and aff i l iat ive condit ions, on theother hand, is not required in order to have an explanation. Theyare not b inding condi t ions. However, they af fect the par tner 'sjudgement of the quality of the information being offered, and thepo l i teness w i th wh ich i t i s expressed . Never the less , a l l f ou rcondit ions were taken into account, in dif ferent degrees, by thethree-year old chi ldren we observed.

r

L 5 4

REFERENCES

,

Antak i , C. , F ie ld ing, G. (1981) "Research on Ordinary Explanat ions" . ;I n C h a r l e s A n t a k i ( e d . ) , T h e P s y c h o l o g y o f O r d i n a r yExolanations of Social Behaviour. London: Academic Press, 27 - i-5 5 .

Bruner, J., S. (1983) Child's talk: Learning to-use language. New York: i

W.W. Norton.

Caste l f ranchi , C. , Par is i , D. (1980) L inguaggio conoscenza e scopi .Bologna: I l Mulino.

Grice, P. (1975 ) Logic and conversation. In P. Cole and J. L. Morgan(eds.), Syntax and Semantics: Speech Acts, volume 3. New York:

Academic Press, 41-58.

Hood, L. , B loom, L. (1979) What , when, and how about why: al o n g i t u d i n a l s t u d y o f e a r l y e x p r e s s i o n s o f c a u s a l i t y .Monographs of the Society for Research in Child Development4 4 , 6 , 1 8 1 . 1 - 4 7 .

Piaget, J. (1923) Le langage et la pensee chez I 'enfant. Neuchatel:Delachaux et Niestle'.

Piaget, J. (1924) Le judgement et le raisonnement chez I 'enfant.Neuchatel: Delachaux et Niestle'.

Sbisa', M. (1989) Att i l inguist ici di spiegazione. In M.S. Barbieri (ed)Soiesazione ed interazione. Torino: Loescher.

Searle, J. (1969) Speech acts: An essay in the phi losophy of language.London: Cambridge University Press.

I

Vygotsky, L.

Von Wright,Cornell

1 5 5

S. (1962) Thougt and Language. Cambridge: MIT Press.

G. H . (1971) Exp lana t ion and unders tand ing . I rhaca :University Press.

r