7/30/2019 Arguing and Reasoning in Understanding Historical Topics http://slidepdf.com/reader/full/arguing-and-reasoning-in-understanding-historical-topics 1/32 Arguing and Reasoning in Understanding Historical Topics Author(s): Clotilde Pontecorvo and Hilda Girardet Source: Cognition and Instruction, Vol. 11, No. 3/4, Discourse and Shared Reasoning (1993), pp. 365-395 Published by: Taylor & Francis, Ltd. Stable URL: http://www.jstor.org/stable/3233745 . Accessed: 27/04/2013 02:03 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Cognition and Instruction. http://www.jstor.org This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AM All use subject to JSTOR Terms and Conditions
32
Embed
Arguing and Reasoning in Understanding Historical Topics
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/30/2019 Arguing and Reasoning in Understanding Historical Topics
Arguing and Reasoning in Understanding Historical TopicsAuthor(s): Clotilde Pontecorvo and Hilda GirardetSource: Cognition and Instruction, Vol. 11, No. 3/4, Discourse and Shared Reasoning (1993), pp.365-395Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/3233745 .
Accessed: 27/04/2013 02:03
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp
.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
.
Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Cognition and
Instruction.
http://www.jstor.org
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
The positive effect of social interactionon children'sreasoninghas frequentlybeen recognized.Social interactionprovidesa social supportsystem,particularlyfor the acquisition of procedural knowledge (Brown & Palincsar, 1989;
Pontecorvo, 1990). If learning s mainlya social process,an "entry nto a culture
via inductionby more skilled members"(Bruner,1986), what is learnedin a
social context is a process of behaving(Clancey, 1990).
Althoughsubstantial hangesarebeginning o be introducedn school contexts
and to appearn new
syllabusesand
curriculan
Italyand in othercountries,thesocial and constructive eaturesof learningarestill underestimatedn mostschool
teaching. For this reason, aiming to study the social processes of knowledge
acquisitionin a naturalsetting,we implementedsocial interactions n school by
designing innovative learning settings, which we call discussions, that make
possible a cognitive apprenticeship Collins, Brown, & Newman, 1989) in the
scientific ways of arguingand reasoning.Discussions are collective situations aimedat solving a problemin groupsof
differentsizes;theycan be led by a teacherormanagedautonomouslyby children
(Pontecorvo, 1986). School discussion can be a powerfulcontext for practicingand learning new reasoningbehavior if certainconditions for their realization
arecompliedwith, for example,starting romcommonexperience,having a very
problematicobject of discussion, and changing the rules of school discourse
when the teacher s leadingthe group(Pontecorvo,1990).As collectivediscourse,a discussion is based on more general conversationalskills that childrenhave
masteredbeforeenteringschool and even preschool:They have to learnto direct
those skills toward the goal of the instructional alk (Orsolini & Pontecorvo,
1992).
Requestsforreprints hould be sent to ClotildePontecorvo,Dipartimento i Psicologiadei Processi
di Sviluppoe Socializzazione,Universithdi Roma"La Sapienza,"Via dei Marsi78, 00185 Roma,
Italy.
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
(Cole, 1990), first of all by language (Vygotsky, 1934/1990), which assumes
particulardiscursive forms in school, as Olson (1986) has shown. School
discourse has been the object of many sociolinguistic studies (Cazden, 1986;
Mehan, 1979; Sinclair & Coulthard, 1975). These studies have been more
interestedn identifying he conversational eaturesof this form of discoursethanin looking for the role of collective speech in knowledge acquisition.
Ourapproachs to look atschooldiscourse,as it is implementedndiscussions,as instructional alk throughwhich teaching-learninggoals can be carried out.
Speech is the main mediationaltool for the transmissionand constructionof
knowledge at school and also for the shift from the interpsychological o the
intrapsychological evel (Vygotsky, 1960/1974, 1978; Wertsch & Sammarco,
1985). In our research,however, speech is not considereda tool thatguides or
indirectly points to some "material"actions. Rather,"what has to be done"
togetheris a discursiveaction,and the actionthat has to be carriedout throughdiscourse is a social knowledge constructionthat is the object of ongoing
negotiationbetweenparticipants.The presenceof a goal-and-tool-mediatedctionin every typeof instructional
talkencouragedus to use the developmentsof the Vygotskianapproachproposed
by Leont'ev's theorization1959/1976, 1975/1977; Wertsch,1981) to frame and
explain the teaching activityand its specific actions andoperations.With these
threeconstructs, t is possible to identifydifferent evels of analysisof the overall
teaching-learningsetting-as we have alreadydone elsewhere (Pontecorvo&
Orsolini, 1992)-taking into account both its molar and molecularaspects.The
activity-theoryperspectivecan be useful for explainingthe level of actionwhere
there is a cultural and interpersonalmediation between teacher and child or
between childrenworking togetheron the same problem.Actions, which can
also be divided into subactions,arecharacterized y theirbeing drivenby a goal(and/orsubgoals)about which the interlocutors an share awareness.
The activityconstruct n Leont'ev's meaningis referred o in the most globallevel of analysis. It is a molar category that can explain the sociocultural
interpretation imposedon the context by the participant(s)"Wertsch,1985b,
p. 203). It is "the non-additive,molar unity of life for the material,corporeal
subject.... It is the unitof life that is mediatedby mental reflection" Leont'ev,1975/1977, pp. 67-68). The social definitionof the instructionalactivity used
by the participantss almost never fully defined at an explicit level. It can be
reconstructed nly at the level of operations-that is, when theagentsareactually
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
operatingwithin the social and interactivecontext in specific and distinctive
ways thatwe can reconstruct s being theexpressionof their"activity"definition.
In the case of class discussions, however, children's (and teachers')operationsare conversational nes. They arecarriedout throughverbal nteractionandshow
typicalconversational eaturesas studiedby sociolinguistics.Speech is used byall interactantsas an instrument or meaning constructionand negotiationand
for sharing topics and perspectives.
DOMAIN ROCEDURESANDEPISTEMIC CTIONS
Actions andoperationsareprimarilydiscursiveones, thatis, more or less actedout through discourse. They have to be assumed in their peculiar features,
however, which aredeterminedby the specific domain or object of knowledge.Because what has to be transmittedn school is a culturalobject of knowledge,which is characterizedby particularepistemological operations(e.g., types of
explanation, ways of reasoning,conceptualframeworks), t is essential to take
into account hepeculiar eaturesof eachknowledgedomain.Inpreviousresearch,we have studiedhow childrencarryout the peculiarways of knowing required
by naturalscience topics in classroom discussions (Castiglia, Pontecorvo,&
andreasoningon scientifictopics. One of us (Girardet,1991) has worked on the
particular pistemic procedures hat are actuallyused by 9- and 10-year-olds n
teacher-ledschool discussions devoted to the explanationand interpretation f
historical events. In bothcases, the main focus of the teaching-learningactivitywas on procedural knowledge, although substantial pieces of declarative
knowledgewere always involved in thelearning asks.But,as Glaser and Bassok
(1989) emphasized,the importantdifferencebetweenexpertand novice ways of
Empire;only one group discussion was carriedout without the teacher before
the one analyzedhere.
The absence of the teacher and of any other adult guidance gives criticalimportanceto the way in which we presentedthe task to the children. The
following written text was given to the groupsof children:
Ammiano Marcellino is a Roman writer of the 4th century. In his
descriptionhe says that the Huns had habits similarto beasts.What
do you think he meant? Was he rightor wrong?Discuss it with your classmates, and write down the reasons that could
cause him to think in this way and whetheryou agree with him or
not.
The childrenwere asked to reach a consensuson: (a) what the historianmeant
by "habits imilar o beasts," b) his reasonsforsayingthis, (c) whetheror notthis
judgmentwas well grounded, d)thereasons or theHuns'"strange" ehavior,and
(e) the children'sreasons for agreeingor disagreeingwiththe historian.
The researchers'expectation(based on previous data: Girardet,1991) was
that the children would be engaged in the double distancingattitudesrequired
by the task, both toward the author and the Huns, by rejecting the evaluativeposition of the Roman historian and by contextualizingthe particularhabits of
the Hunswithintheir culturaland social organization. n bothcases, it is possibleto considerchildren'sinterpretative ctivity as aimed at negotiatinga contextof
plausibility, as Perelmanand Olbrechts-Tyteca 1958) defined it. Within this
context, the author'sjudgmentcan be acceptedor rejected,and, if rejected,the
Huns' behaviors can be accepted under certain conditions. Some groups have
graduallydeveloped more articulatedperspectives by discovering intermediate
positions that could take into account differentpoints of view.
Corpus f Data
The corpusis composed of six tape-recordeddiscussions for a total of 4 hr, 20
min of talk. Discussions were fully transcribedusing the following conventions:
xxx = word or sentencesthat were not identified
= shortpause
...= long pause
{ } = overlappings
The punctuation ries to representutterance ntonation.Each turn is numbered
progressivelywithin the discussion sequence and is marked by the subject'sname. Because the turn was not a significantunit in our data (as opposedto the
collective discussions in which interventions are shorter, because they are
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
propelledmainly by teachers'questions), we divided the turnwhen necessaryinto idea-units that correspondto linguistic clauses, as other researchershave
This studyhas the descriptiveaim of showing if andhow the peculiarepistemic
procedurescharacteristicof historical reasoning can be practicedby children
when they are in a social situationthat supportstheir individuallinguisticand
cognitive activity. We assume the Vygotskian construct that children can
internalize and appropriate(Newman, Griffin, & Cole, 1989) the culturalreasoning proceduresof a scientific domain when, being supportedby the task
requestandby the social context(Wertsch,Minick,& Arns, 1984), they areputinthe conditionof applying heirgeneralargumentativekills to a specificdomain
problem(Pontecorvo,1990).
LevelsofAnalysis
Three levels of analysis were used. The first one is molar and identifies the
generalmood of thediscussion,what we have called in previouswork theframe
of discourse.Framehas been defined(Pontecorvo& Orsolini, 1989; Pontecorvo
& Zucchermaglio,1989) as a partor a phaseof a discussionthatis characterized
by a discursiveactivity and by a relatedcognitive function. Such functionsare
usually pursued by the teacher, who proposes her or his general goals to the
children'sgroupand oftenreproposes recycles) themin the course of discussion
(Pontecorvo & Orsolini, 1992). Because the functional and goal-directed
leadershipprovided by the teacheris not operating n the discussion here, the
object of discourseis framedby the children,who negotiatetheirinterpretationof the task thathas to be accomplished.In our case, the frame,as the dominant
cognitive activity that results from the negotiationbetween speakersabout a
historicaltask,comes closer to the methodological ssues of the historian's ools.
The second level, which is embedded in the first level, consists of smaller,well-identifiedreasoning sequences in which particularepistemic actions (or
subactions)arepursued.Sequencescorrespondo the level of actioninLeont'ev's
theory,and, in our case, they are similarto the explanationstrategiestypicalof
the historicaldomain.
Within these reasoning sequences, the third level of analysis looks at the
molecularoperationscarriedout throughthe idea units: the smallest units in
which the discourse is analyzed. Each idea unit is submitted to a double
categorization, looking at the specific argumentativeoperations (i.e., the
sociocognitivemotor of the discussion) and at the epistemic operationsused bychildren.At this level, we have also marked what was unexpressedby speakers
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
andleft at an implicit level. In most cases, it is shared and does not requireany
explicit formulation.
RESULTS
FramesofDiscourse
Froma general analysisof the six discussions,two main framesemerge as waysin which the task was interpretedby the children'sgroups.Both are very close
to those used by thehistorians;but,althoughframeA corresponds o the explicitaim of the curricular ctivity,frameB is ratherunexpected, given the children's
ageand
grade.
A. Acceptabilityof the judgment of AmmianoMarcellino. This frame is
What does it mean? What were the motives for their behavior?
B. Authenticityand reliability of the source. This frameis subdividedinto
the following questions:Is what AmmianoMarcellino said true?Was he
well documented?Is the source an authenticone?
In both cases, the participantsdiscussed whetherAmmiano Marcellino wasrightor wrongand whetherthey could or could not agreewith his interpretation,and they did it by reconstructingdifferent "contexts of plausibility." Two
groupdiscussions (COR and AOR)' are for the most partwithin frame A and
interpret he task accordingto its explicit aim. One groupdiscussion (ALU) is
completely within frame B, whereas threegroupdiscussions (DOR, BOR, and
BLU) present an alternationof the sequences in frame A and frame B. The
interpretations f the task and their relatedframes are summarized n Table 1,wherethe numberscorrespondto the turns of talk. Gaps between turnnumbers
indicate out-of-tasksequences.
Discussion equencesandEpistemicActions
The second level of analysis permits us to identify homogeneous reasoning
as being more or less typical of the explanationproceduresof the knowledgedomain. Sequencesare identified within the frames as argumentativephases in
which a dominantcollective goal-mediatedaction is pursued.
We have identifiedthe following relevant actions:
'The first letterof the acronymsof the groupdiscussions refers to the orderin which the groupworked on the task (A, B, C, D); the last two letters refer to the two classroom teachers(LU and
OR).
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
SummaryContentof the Six Protocolsby Frames and Topics
1. COR:Is the judgmentof A. M. an acceptableone? (A)(1-20) Whatdoes "to be like a beast"mean?
(44-65) Is it possible to be different withoutbeing inferior?
(73-84) Do we have somethingin common with the Huns?
2. AOR: What is the meaningof being bestial or civilized? (A)
(1-30) Whatdoes "to be like a beast"mean?
(31-43) Whatdoes "to be civilized" mean?
(44-62) We have to locate the Huns in their time and context.
3. ALU: How reliable is A. M.? Is he an authenticsource?(B)
(1-42) Did A. M. really see them?Did he live at this time? Have we got enough proofto
decide?
(48-75) What did A. M. gain from saying/writingthese things?How could he know?
4. BOR
(1-9) What do we have to discuss?
(16-60) Is A. M. a reliableauthor? B)
(61-80) Were the Huns really like beasts? (A)
(81-92) How could we decide? (B)
(93-125) We have to locate the Huns in their time and context. (A)5. BLU
(1-23) Why are the Huns like beasts? (A)
(24-25) Does A. M. tell the truth? B)
(26-58) Whyare the Huns like beasts?
(A)6. DOR
(1-19) What do we have to do?
(20-32) Which sentencebest describes the Huns?
(33-37) Does A. M. tell the truth? B)
(38-57) Which sentence best describes the Huns?
(58-72) Does A. M. tell the truth? B)
Note. Numbers in parenthesesto the left of topics indicate turns of talk spanningthe topic.Frame(A or B) is given following appropriate opics. A. M. = AmmianoMarcellino.
1. Terminologicalandconceptualdefinitions.
2. Categorizationof social actors and of sociohistoricalphenomena.3. Locatingevents and phenomenain time and space.4. Interpreting ctions,plans, and intentionsof social actors.
5. Locatingactors and actions in theirhistoricalcontext.
These actions havebeendevelopedby the children n bothframeA and frame
B. We provide examplesof some of them in what follows.
OperationsndCategorizationystem
The categorization ystemhas been developedto take into account the relations
betweendiscourse and reasoningat the molecularlevel of the idea units where
the operationsoccur. A distinction was made betweentwo types of operations:
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
(a) the argumentative perations hatgive an accountof the collective discursive
activity and (b) the epistemic operationsthroughwhich the specific knowledge
domain is analyzed. As is specified further,an internallink, which should beverified in furtherresearch,could be assumedbetweenepistemic and argumen-tativeoperations,correspondingo the structure f collectivereasoningembedded
in the argumentativeactivity.
Argumentative perations. Theseareusedby thespeakers s a meansof
constructingand supporting heirreasoning(Toulmin, 1958). They areused bythe children as a means of constructing and supporting their claims and,
consequently, their reasoning and thinking. They can appear as any of the
following:
Claim: Any clause thatstates a position (thatcan be claimed).
Justification:Any clause that furnishes adequategroundsor warrants or
a claim.
Concession:Any clausethatconcedes something o anaddressee,admittinga point claimed in the dispute.
Opposition:Any claimthat denies what has beenclaimedby another,with
or withoutgiving reasons.Counteropposition:Any claim that opposes another'sopposition, which
can be more or less justified.
The categorizationof argumentativeoperations permits us to focus on the
developmentof the discourse-reasoningconstructedeither by a single speakeror by more than one. The link that is necessarilyestablishedbetween claims and
justifications s basic andcrucial,even for reasoning hatdevelopsmainlywithin
a single participant,whereasthe chainingbetweenopposition, counteropposition,and sometimes concession underlines the collective dimension of the shared
reasoning.Only at the argumentativeevel have we formulatedand categorizedwhat stays implicit in the collective discourse.In most cases, they are"implicitclaims,"which the speakersdo not need to make explicit; in a few cases, there
are also "implicit justifications"or both. Implicit idea-unitsare markedwith a
letterfollowing the numberof the clause to which they are linked (e.g., 1.4a in
Excerpt1, which appears n a later section).
Epistemic operations.Theseare
groundedn the
explanation roceduresand arespecified in termsof the particularhistoricalcontent to whichthey refer.
Epistemic operationscorrespond o the explanationprocedures hatareused for
describingand interpretinghistorical events. When used on the real sequencesof discourse, heyarein most cases accompaniedby a reference o thepeculiaritiesof the historicaltopic with which the speakersaredealing.
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
Given the descriptive characterof this study and the particular ormatof the
conversationaldata in which we are interested, n what follows we describeand
commenton some protocolexcerpts.Reasoningsequencesexemplifyingdifferent
epistemicactions,withinthe two activities definedas frame A andframeB, are
categorized atthe level of the molecular
operations.We
provide examples,developedin some of the children'sreasoning sequences,that arerepresentativeof the mainepistemicactions.Accordingto our initialassumptions,such actions
are attemptsat appropriatinghe explanatoryproceduresof historicalreasoning
by practicingwith some of the historian'sreasoningtools.
The main title of each protocol corresponds o the main epistemic function
carriedout by the sequence;the subtitlespecifies the content of the frame and
in some way thetopic of the collective discourse.(Inexcerpts,A. M. = Ammiano
Marcellino.)2
Terminological efinition. In Excerpt 1 of frameA, the group aced the
problem of if and how the Huns' habits could be defined as normal. The
mainly constitutedby justificationsof personalclaims advanced both by the
opposerand the opposee.
Conceptualdefinition. InExcerpt of frameA, thetopic ookssimilar othe previousone, but herethe focus is on the conceptual efinition f beingcivilized. Aftera long silenceuntil thispoint,Andrearecycled(44) his opposition
Excerpt2
(2) COR (Turns 44 to 56)
Conceptual definition: Were the Huns inferior or different? (frame A)
TALK
SEQUENCE
Andrea44.1 no,no,no
44.2 I say that he is not right
Alessandra45.1 but why do you thinkthey are
civilized?!
45.1a you wouldbewrongif you said that
they were civilized
Filippo46.1 are they civilized?!
46.1a beingcivilized means to be like us
46. 2 are they like us?
Andrea47.1 they are not really civilized
47.1a there are differentdegreesofcivilization
47.2 they have one.... some laws to follow
Filippo48.1 but it was written on the text
48.2 that they didn't have laws
Andrea49.1 that'strue,but this was their custom
49.2 we cannot say that they werebarbarians
49.3 if that was their custom
Alessandra50.1 Andreabut
ARGUMENTATIVEOPERATIONS
Opposition
Claim
Counteropposition
Implicit claim
Counteropposition
Implicit justification
Counteropposition
Concession
Implicit claim
Justification
Justification
Opposition
Concession
Claim
Justification
Opposition
EPISTEMICOPERATIONS
Evaluationof the author'sjudgment
Categorizationof people
Categorizationof people
Appeal to an analogy
Recategorizationof people
Appeal to an existingcondition
Appeal to the source
Appeal to a lacking condition
Recategorizationof people's behavior
Appeal to the consequences
Recategorization
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
Filippo54.1 wait a moment, we can say that they Claim Definition
didn't use to eat like us
54.2 we don'tgo down the road,pick roots Justification Appeal to an analogyand eat them, we don't do that
54.3 we eat cooked food.., raw meat, we Justification Appeal to an analogydon't eat raw meat
54.3a they are different rom us Implicit claim
54.4 we are more civilized thanthey
Claim Recategorizationof a people
Andrea55.1 it's true,but they were more Concession Appeal to time
backward
55.1a whoeverprecedesus in time is Implicit justification
justified in being less civilized
Filippo56 Yes, that'strue Concession Predication on 55.1
to the historian'sjudgment, which he had already presented in I and 3 (see
Excerpt 1). After havingreceived strong oppositionthat included ironicand/orrhetoricalquestions (45 and 46), Andrea insisted that it was possible to have
differentways of behavingand different aws (inthe sense of rules)to be followed
(47 and 49). These differences could be justified because of one's being"backward"55). This concession was recognizedas acceptableby others.
The epistemic action carried out here related to the representationof the
concept of being civilized. The sequence was more clearly focused than the
previousone on the possibility, brought up by Andrea,of a graduatedposition;that is, the concept that there are different degrees of civilization and that
civilizationmust be evaluatedwithrespectto its time. Theargumentativetrategyused by the opposers was mainly "extremization" f the opposee's arguments,whereas the opposee produceddifferenttypes of graduation,as concessions, a
kind of argumentative caling (Ducrot, 1980), and therebysucceededin havinghis more balancedposition accepted.
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
Themain pistemicctionwas o understandnd xplain,nthedoubleenseoffered yVonWright1971),hemotives nd ituationf thispopulationyrelatinghem otheir istoricalnd ulturalontext. rom nargumentativeointofview, hesequenceascharacterizedyalackofstraightppositionetweenthe nteractants.herewasaprevalencefconcessionsndpartialegationsfstatementshat ouldbe toostrong,orexample:Theyouldn't avea good
19.5 because therewere not many pens Justification Appeal to materialand paper conditions
19.6 or, I mean, Idon't think A.M. is right Claim Evaluationof author's claim
19.6a he cannot have written this text Implicit claim
19.7 these are my impressions Claim Predicationon own claim
19.8 and even if A.M. had written this Claim Predicationdocument in ancienttimes Counterfactual strategy on the source authenticity
initiating
19.9 but it must be seen how he succeeded Claim Appeal toin looking at them a necessary condition
(eyewitness)
19.10 since they had very bestial habits Justification Appeal to datafrom the source
19.11 so they could even kill him Justification Appealto a consequenceof source's data
19.11a thus either it is false that they Implicit claimwere bestial or it is false that he had Counterfactual strategylived at that time and had seen them initiating
Nicola20.1 Right! what Paolo said is right Claim Predication on 19.1
20.2 because he could not have lived at Claim Appeal to anecessarythat time condition
203 also because I think that if he had Claim Predicationlived in those times, in the Middle Counterfactual strategy on timecontemporaneityAges initiating
20.4 not everyone could have, let's say, in Justification Appeal to socioculturalthe Romantimes and so not everyone contextcould write (scarcediffusion of writing)
20.5 and they could not producea Justification Appeal to sociocultural
description of people with such contextanomalous laws
20.6 I think that no one could have done Claim Predicationthis description then Counterfactual strategy on cultural impossibility
ending
Filippo
21.1 instead, I think something that Claim Predication on own claimperhaps does not fit in
21.2 that is, I mean I partly agree and Claim Predication on own claim
partly I don't agree
21.3 because when A.M.,well, he could Justification Appealto a consequenceeasily have been killed for the author of data from
the source
21.4 or he could have had some problems Concession Appeal to a consequencein seeing, in getting in touch with the for the author of data fromAlans or Huns the source
21.5 because either he was an Alan or a Claim CategorizationHun of the authoras member of
the people of the source
21.6 or what he has written is somewhat Claim Predicationfalse on the source'suntruth
385
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
Paolo22.1 1think at that time they could not Justification Appeal to a necessary
read and write condition (lackofinstrumental abilities)
22.2 thus it was very hard for A.M. to Claim Predicationhave written that document on the improbability of the
attribution of the sourceNicola
23.1 because, if he had writtenon these Claim Predicationsheets of paper,on sheets of paper Counterfactual strategy on the possibility of the
initiating material production
23.2 I think that, at this time, the sheets Justification Appeal to timewould alreadyhave turned to dust as reason for the source's
material deterioration23.3 in short, as time goes by, the sheets Justification Appeal to time
turntodust as reason for the source'smaterial deterioration
23.4 Two or threethousand years have Justification Appeal to
passed, I believe the amount of time
23.4a source s not authentic Implicit claim
23.5 so the sheets wouldn'thave been Claim Appeal to the consequencesfound any more, they would have Counterfactualstrategy of timepassingon thesourcebeenturnedto dust ending
Filippo24.1 it could easily have been written on Claim Predication
a stone, for instance on alternative procedures(other materials on which
the source was written)24.2 or remnants of huts
might have been Claim Predicationfound on alternative procedures(othersources)
24.3 and they would prove naturally... in Justification Appeal to material aspectsthe building, perhaps in the way it of the socio-cultural contextwas used, how they used it
Nicola25.1 1 think that if he had written it on a Opposition Predication
stone Counterfactualstrategy on the possibility of the
initiating condition (writing material)
25.2 they couldn't have written that is Justification Appeal to datawritten there everything, everything from the source(amountof
information)
25.3 simply becausewritingon stone is not Justification Appeal to a general principlethe same as writingon paper
25.4 thus I think that all this news Claim Predicationwould not have been understox)d;n Counterfactualstrategy on thequalityof source'sshort,do you understand? ending information
25.4a but since they are understandable Implicit justification
25.4b thus the source is false Implicit claim
conclusion that the source was false (21.6) because of: the lack of paperand
pencil (19.5), the inevitabledeteriorationof the paperif it existed (23), a lackof writingand descriptiveskills duringthat time (20), and the impossibilityof
writing such a text on a stone (25). Counterfactual easoning (19.8 to 19.11)directly challenged the content of Ammiano Marcellino'sjudgment:What he
says is not true,because if he had really seen this populationwith his own eyes
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
(19.9), the Huns would have killed him, being cruel and"like beasts."So, either
he was one of them,and thus the statementthat he was a Romanwould be false,
or what he wrote about the Huns(21.6) was false, andtheywerenot "like beasts."Thedominantpistemic ctionhereconcernedhereliability ndauthenticityof the historical source, which was radically challenged by the children's
analyzingboth culturaland materialimpossibility, as shown in the preceding.
Excerpt 5 was characterizedby the presence of three complex counterfactual
strategies (starting at 20.3, 23.1, and 25.1) by which children used single
argumentativeperationso reacha morecomplexargumentativeoal. Giventhe main topic of the sequence,argumentativeoperationswere characterizedbythe demonstrationof the impossibility-based on conditionsand consequences
and being both material and cultural-of the authenticityof the source.
DiscourseOrganizationnd Interaction ode
The linearorganizationof the written text does not do full justice to either the
complex articulationof the individualreasoningor the differentmodes throughwhich the shared discourse and reasoningwere carriedout. At this point, we
were pushed to find a better representationof the links existing among the
differentpartsof the discourse-reasoning, rying to representgraphicallytheirinternalandsometimes hierarchical rder,as is doneby Resnicket al. (this issue).
Also usingthe turnas a unit of analysis,it was possibleto discoverthe internal
structure of the discourse with its different levels of articulation.The main
position expressedby the speaker,whose claim could be more or less justified,was supportedn most cases by anorganizedset of justificationsand claims that
had the subordinate unction ofjustifyingthe mainposition.Thesemorecomplex
argumentscould be structuredaccordingto differentmodes, and some of them
could be identifiedmoreeasilyon the basis of theirepistemiccontent; orinstance,
they maderepeateduse of analogiesor of counteropposed xamples.The example reportedin Figure 4 representsTurn4 of Excerpt 1 in which
Alessandraexpressedherposition in five clauses. In her first clause, Alessandra
statedherposition throughan evaluation "According o me AmmianoMarcellino
is quite right"), after which followed an argumentativestructure aimed at
justifyingthepositionandorganizedon two levels: first,a claimandajustificationwith an appealto the source ("becausefrom this document t is clear/ that these
people were not normal"); hen, a second claim (anotherevaluation)groundedon a justificationthat used an appealto an example ("makingcuts from birthto
stop the beardgrowing/ I think it's really a bestial habit").The speakerstended to use the same structure n differentturns of the same
conversation as if they were following an argumentativestyle: Alessandra
repeateda similar structuren Turn6 (see Figure5) and in Turn 53 (see Figure6) of the same discussion.
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
This type of graphicrepresentationnhancesunderstanding f the interaction
modes used by the speakers duringthis discursiveactivity, which can be much
more intensive than is indicatedby a simple count of frequencies.In Figure 5, the discussion reported n Excerpt 1 is representedgraphically.Eachspeaker'snameappears n a column:The vertical ine enables us to examine
the argumentdeveloped by a single speaker,whereas the horizontal line makes
it possible to focus on the exchangesamong speakers.When we consider the same speaker(Andrea n Turns1 and 3, or Alessandra
in Turns 4 and 6 of Figure 5), it is possible to look at the structureof the
individual'sdiscourse and at the ways in which they coconstructreasoning in
agreementor oppositionwith the other speakers.When the exchangesbetween
speakersare focused, it is possible to distinguishbetween different interactionmodes. For instance,oppositioncan concern the refusalof one single claim, as
in the case of Filippo in Turn 2 opposing Andrea's Turn 1. But it can also
concern an entire argumentative tructure,as in the case of Alessandrain Turn4 (opposingAndrea'sTurn3):This opposition s carriedout through he complex
argumentative tructureanalyzed in Figure 4.
Such graphicalrepresentationsmake possible an overview of the dynamicsof the interactionalexchanges that produce differenttypes of patterns.In the
reasoningsequencepresentednFigure5, eachspeakerexposes his or herthinking
in a rather structuredway, whereas in othermoments of the same discussion,which arereportedn Excerpt2 andpartiallyrepresentedn Figure6, thedifferent
arefaster,often supportedby singlejustificationsor by simplified argumentativestructures.
DISCUSSIONANDCONCLUSION
Our differentdataanalysesshow that autonomous nteractionalactivitiescan be
extremelyrich situations n terms of theproductionof high-level reasoning,even
in young children. Such group discussions could be defined as situations of
"cognitive apprenticeship"Collins et al., 1989) in which reasoning is both a
situated and a distributedaction existing in the social realityof negotiationwith
othersand with the task.As Bruner 1990) alsoemphasized,a person's knowledgeis not just in one's own head but also in the notes, in the underlinedbooks, in
the informationsources, in the friendsone can call on for help, in what results
from social interactionwith others.
The complexity of children's sharedreasoningmade it necessaryto developa pluralityof analyticalinstruments o account for the different dimensions of
the reasoning,on both the argumentative nd epistemic levels.
The identification of general frames of discourse and of more specific
epistemic actions was particularly ffective in emphasizingthe characterof the
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
"social negotiation" of children's cognitive activity. The absence of the
teacher-usually the personwho aims to bringthe children to an understandingof the
topicandto
acquirecorrecthistorical
knowledge--compelledthe children
to engage in much negotiationactivity in interpretinghe task assignedto them
(Edwards& Mercer,1987). The children understood he features of the specifichistorical activity, although they carried it out within different frames and
consequentlythroughdifferentepistemic actions, which are very close to the
operationswere productivein focusing the structureof the shared reasoning,both within the individual and the group, in ways that were relevant to the
historical topic. On this basis, it is possible to confirm our hypothesis thatelementary chool childrencanpracticeandgraduallymastercognitiveskills and
tools of a specific knowledge domain when they are supported by a learningenvironment hat offers a meaningfulproblematic ramework hat can be shared
in a group.The identification of more complex argumentativestructures (as other
researchershave also done: see Resnick et al., this issue) reveals different
interactionmodes and argumentativestyles in subjects and in groups. With
longitudinaldata,it may be possible to verify the evolution and/orthe "passage"
of a reasoningstructure romone child to anotheras an effect of the interactionwith peers in appropriate earningenvironments. ndeed,the exchangeconcerns
not only the appropriation f informational lements but also the acquisitionof
reasoning strategies,the core of which is given by the structures f justification.From a proceduralpoint of view, children could succeed in practicingand
mastering he methodologicalandexplanatory ools of the historian,as we have
shown throughthe protocol analysis.From a declarativepoint of view, the lack
of adult guidance sometimes leads children onto divergent paths: Adequateteacherinterventionwould have been extremely useful in such cases, offering
Comparing hese discussions with those led by teachers(Pontecorvo, 1993),
however, we can say that children'sautonomouscollective discourseis often on
a higher cognitive level thanthatguideddirectly by theteacher(Girardet,1991).The aim of masteringthe tools of arguingand thinkingwithin a subjectfield is
probably better reached through this type of unguided discussion dispute.
Nevertheless, such discourse occurs within the school context and provides a
kind of assessment of what has alreadybeen done in class in the knowledge
domain of history. Problem-solving group settings can also be regarded as"contextualizedmeasures"of how much childrenhave masteredthe actionsand
goals proposedto them by their teachersthroughthe curriculum.At the same
time, thechildren,who engagedthemselvesautonomously n discussingwhether
the judgmentof the Romanhistorian was well grounded,showed that they not
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions
only understoodhe specificfeatures f the historicalask but also weremore
generally ocialized o a representationf the learning ctivityas an "active"
counterpartf theteaching.notherwords, hesechildrenmplicitly hared heideathat n schoolone has to engage n understanding,xplaining, easoning,andarguing. heengagementf thesechildrenn a difficultask, hownbytheir
eagerness o find a well-groundednswer o the questionsproposed o them,demonstrateshat heyshare conceptionf learning-inLeont'ev's1975/1977)words, they share a particular efinitionof the learning"activity"-as a
self-directedrocessmediated y culturalools andbygroup-distributedction.
ACKNOWLEDGMENTS
A previousItalianversionof this articleappearedn 1993 in the bookLa
Condivisione Della Conoscenza [Shared Knowledge]. (C. Pontecorvo, Ed.).Florence:LaNuovaItalia.
Thepreparationf thisarticlewassupported ya grant f the ItalianMinistryof Universityand of Scientific and TechnologicalResearch(60%, Rome
University) or a researchprojecton "Mediationalmeans and interactivemodalitiesn competence cquisition."
REFERENCES
Bloch, M. (1949). Apologie pour l'histoire ou mitier d'historien [Apology for history or the trade
of a historian].Paris:Colin.
Brown, A. L., & Palincsar,A. S. (1989). Guided, cooperative learningand individualknowledge
acquisition.In L. B. Resnick(Ed.),Knowing, earning,and instruction:Essays in honor of Robert
Glaser (pp. 393-451). Hillsdale, NJ: LawrenceErlbaumAssociates, Inc.
Bruner,J. S. (1986). Actualminds,possible worlds. Cambridge,MA: HarvardUniversityPress.Bruner,J. S. (1990). Acts of meaning:Four lectureson mind and culture.Cambridge,MA: Harvard
UniversityPress.
Castiglia,D., Pontecorvo, C., & Zucchermaglio,C. (1983). Discorso e ragionamentoscientifico in
classe [Discourseand scientific reasoningin class]. Scuola e Cittc, 17, 371-387.
Cazden, C. B. (1986). Classroom discourse. In M. C. Wittrock(Ed.), Handbookof research on
teaching(3rd ed., pp. 4-91). New York:Macmillan.
Clancey,W. (1990, October).How shouldexpert systemtechnologybe applied or education?Paper
presentedat the NATO Workshopon Intelligent TutoringSystems, Sintra,Portugal.
Cole, M. (1990). Culturalpsychology:A once andfuturediscipline?In J. J. Berman Ed.),Nebraska
Symposiumon Motivation 1989: Cross-culturalperspectives (pp. 279-335). Lincoln: University
of NebraskaPress.Collins, A., Brown, J. S., & Newman, S. E. (1989). Cognitive apprenticeship.Teaching the crafts
of reading, writingand mathematics. n L. B. Resnick (Ed.), Knowing, earning, and instruction:
Essays in honor of Robert Glaser (pp. 453-494). Hillsdale, NJ: Lawrence ErlbaumAssociates,
Inc.
Ducrot,O. (1980).Les echellesargumentativesArgumentativecales].Paris:Les Editions du Minuit.
This content downloaded from 152.74.16.35 on Sat, 27 Apr 2013 02:03:16 AMAll use subject to JSTOR Terms and Conditions