PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Tõugu, Pirko]On: 25 February 2010Access details: Access Details: [subscription number 919373900]Publisher Psychology PressInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

European Journal of Developmental PsychologyPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713684600

References to social norms by preschool children and their linguisticexpressionPirko Tõugu a; Tiia Tulviste a

a University of Tartu, and Centre of Behavioral and Health Sciences, Tartu, Estonia

First published on: 20 March 2009

To cite this Article Tõugu, Pirko and Tulviste, Tiia(2010) 'References to social norms by preschool children and theirlinguistic expression', European Journal of Developmental Psychology, 7: 2, 249 — 264, First published on: 20 March 2009(iFirst)To link to this Article: DOI: 10.1080/17405620802194092URL: http://dx.doi.org/10.1080/17405620802194092

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

References to social norms by preschool children

and their linguistic expression

Pirko Tougu and Tiia TulvisteUniversity of Tartu, and Centre of Behavioral and Health Sciences,

Tartu, Estonia

The study is based on the domain approach to the social norms and itinvestigated whether preschool children’s references to social norms are related totheir language expression. Children aged 3;8–7;5 years (N¼ 73; 35 boys and 38girls) participated and were video-recorded playing in triads for 20 minutes. Thenorms mentioned were coded as moral or conventional rules according toPiotrowski’s (1997) coding system. Older children and more talkative childrenmention more conventional rules (p5 .05 and p5 .001, respectively); boys tendto refer to moral rules more often than girls, yet the use of moral rules was notinfluenced by age or talkativeness of the child. This supports the idea thatchildren treat social norms of different domains differently: the use of moral rulesdoes not seem to be influenced by the characteristics of the child, while thereferences to conventional rules depend on the age and talkativeness of the child.

Keywords: Conventional rules; Language skills; Moral rules; Preschoolers.

Social norms are the customs and rules of conduct that dictate what isallowed, what is forbidden and what is expected in a certain context. Socialdevelopment is a process of creating an understanding of the social worldbased on the conclusions of and the theories on previous experience (Turiel,1983, p. 1). Language has an important part to play in socialization:Gleason (1988) has identified three ways in which language influences socialdevelopment. She points out that first of all language is used to provideexplicit instructions of what to do, feel and think. Second, children are

Correspondence should be addressed to Pirko Tougu, Institute of Psychology, University of

Tartu, Tiigi 78–336, 50410 Tartu, Estonia. E-mail: pirko@ut.ee

The study was possible due to the grant no 6511 of the Estonian Science Foundation and the

technical devices used for recording belong to the Estonian Centre of Behavioral and Health

Sciences.

We would also like to thank Enn Veldi for his help.

EUROPEAN JOURNAL OF DEVELOPMENTAL PSYCHOLOGY2010, 7 (2), 249–264

� 2008 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

http://www.psypress.com/edp DOI: 10.1080/17405620802194092

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

provided with clear instructions of what to say on certain occasions, andthird, there are subtle and indirect socializing effects resulting frominteraction (e.g., interrupting some children more than others).

What follows from Gleason, is that the analyses of the spoken languageand interpersonal communication provide a means for investigating socialnorms. Tesson, Lewko, and Bigelow (1987) have claimed that the socialnorms that people follow in their interpersonal interaction are attainable onthe level of language. In a qualitative study, Bhatia (2000) has tried to showhow parents, friends and peers use language to transmit the social norms ina particular culture to the child through commands and prohibitions.

Besides grown-ups, peers also provide an important socializing context(Nicolopoulou, 2002). Yet, as pointed out by Blum-Kulka and Snow (2004),peer interaction has not received much attention. The aim of this studywas to discover what the social norms of conduct that children themselvesmention in their peer groups are, and to study the connections betweenthe use of social norms and the linguistic measures, age, and gender ofchildren.

Social norms: Acquisition and research

Turiel (1983) has shown that children’s social knowledge is organized in twodomains: moral and conventional. Conventions or conventional rules arerelative to the social context (Turiel, 1983). Conventional rules are thedifferent norms and agreements that regulate social conduct and theinteractions among individuals that participate in a social system (Turiel,1983). They apply only in certain circumstances and they can be changed byagreement. In order to understand these rules experience is needed—children create the understanding of the social system and the applicableconventional rules due to their belonging to different social groups (family,friends, classmates).

Moral rules are universal rules that determine what is fair and good, andthey are valid regardless of whether they are explicit or not (Turiel & Nucci,1978). Moral issues are categorical and obligatory (Turiel, 1983); moral rulesare applicable not because there is general agreement or consensus, instead,they are structured according to the underlying ideas of justice, rights, andbenevolence (Nucci & Weber, 1991).

Turiel and Nucci (1978) first presented the distinction between moral andconventional rules. And, in general, the domain approach to socialknowledge later elaborated by Turiel (1983) has been widely utilized (e.g.,Siegal & McDonald Storey, 1985; Smetana, 1989; Smetana, Kochanska, &Chuang, 2000). Similarly, the present study was based on the understandingthat social norms fall into two domains and in social interactions childrenmake use of moral and conventional rules.

250 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

In order to investigate children’s understanding of social norms,interviews about real or imaginary transgressions have been used (e.g.,Smetana, Schlagman, & Walsh Adams, 1993). The ability to differentiatebetween moral and conventional rules is already apparent at the age ofthree, when children consider moral transgressions to be more severe thanconventional transgressions (Smetana & Braeges, 1990). In general,preschool children already consider breaking moral rules more severe thanbreaking conventional rules and agree that moral transgressors deserve amore severe punishment (Helwig & Turiel, 2002). In order to develop intheir understanding, children seem to need experience and contacts withtheir peers: Siegal and McDonald Storey (1985) found that children whohave been in childcare for a longer period of time, differentiate betweenmoral and conventional rules better than children who have just joined aday-care centre.

Besides interviewing children, researchers have also paid attention to thesocial norms that parents mention while talking to the child (Piotrowski,1997; Smetana et al., 2000). Piotrowski’s study (1997) showed that mothersand older peers refer to conventional rules more often than to moral ruleswhile interacting with the child. Less attention has been paid to how thesocial rules are used in peer interaction. Nobes (1999) has concentrated onthe characteristics of the rules children impose themselves, their under-standing of the universality of rules, and the role of context in imposingsocial norms. Piaget (1932/1950) has claimed that authority is important forchildren in order to define the norms; at the same time he has suggested thatfor the understanding of justice to develop, authority is not enough, and thebest environment for moral development is the company of peers (Piaget,1932/1950, p. 294). Turiel (1983) also points out that for the understandingof moral rules to develop, interaction between children is necessary. Blum-Kulka and Snow (2004) have argued that peer interaction is important forpragmatic and cognitive development due to its more equal participationstructure. Considering the importance and the impact of peers, it would benecessary to study how much children refer to social norms in their peergroups, what are the social norms children mention in peer interaction, andwhat are the different variables that influence or determine the conversationabout social norms. Unfortunately, such studies are few: Nucci and Nucci(1982a) found that young schoolchildren respond differently to theviolations of moral and conventional rules. Nucci and Turiel (1978) foundthat children in kindergarten had a tendency to respond to the transgres-sions of moral rules and not to those of conventional rules. Tulviste andKoor (2005) studied the social norms in peer interaction, and theydiscovered gender differences in the use of social norms. In their studyboys referred to moral rules more often than girls did and girls mentionedconventional rules more frequently. Similarly, Nucci and Nucci (1982b)

LANGUAGE AND SOCIAL NORMS 251

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

have discovered gender differences in the use of specific responses to moraland conventional transgressions.

Examining the use of social norms, the context of the interaction shouldalso be considered. Studying the transgressions of rules, Killen and Smetana(1999) suggested that contexts vary in their degree of structure and thereforein the necessity to follow the rules: they discovered that in day care duringfree play children break more moral rules and during organized activitiesmore conventional rules. Nicolopoulou (2002) has pointed out that in playchildren construct an ‘‘imaginary situation’’ that is structured by rulesrecognized and accepted as necessary by children themselves and notimposed by adults. In the current research children were observed duringfree play, so they were given the freedom to structure their own context.

The modern understanding of moral development of children includes theidea of shaping their judgement about wellbeing, justice and rights in peerrelations (Helwig & Turiel, 2002). Language has an instrumental role inpassing judgement. And considering the importance of language in thoughtdevelopment and the relevance of it in socialization process, it is feasible tothink that the level of language development is related to the acquisition andthe use of social norms.

Language skills and behaviour

On the one hand, language development is connected with the understandingof norms: having acquired some language skills children begin to understandtheir mothers’ explanations following the order ‘‘Stop it!’’ or why it is notallowed to have a water fight at the dinner table. On the other hand, languagedevelopment also includes improved skills of expression: as the vocabularydevelops, it becomes easier to articulate the social norms already acquired.Talkativeness may be of particular interest here given the value placed onverbal exchange for understanding social norms. Some studies suggest thatthe performance of highly talkative preschool children was significantlysuperior on all aspects of their linguistic abilities (Evans, 1996; Landon &Sommers, 1979; Van Kleek & Street, 1982). In general, research has shownthat girls outperform boys in their linguistic development: their developmentis generally faster in vocabulary (Bauer, Goldfield, & Reznick, 2002), inphonology (Dodd, Holm, Hua, & Crosbie, 2003), and pragmatics (Thompson& Moore, 2000). Girls have also been found to be slightly more talkative thanboys (see Leaper & Smith, 2004, for a review). It is also possible that betterlanguage abilities make a contribution to the extent to which childrenparticipate in discourse, for example how frequently they explicitly mention itwhen they perceive that certain norms have been violated.

Smetana and Braeges (1990) have found that children with betterlanguage skills begin to differentiate moral and conventional rules earlier.

252 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Stowe, Arnord, and Ortiz (1999) have tried to discover the connectionbetween language development and behaviour: they found that a low levelof language development is related to antisocial behaviour with 4- to 5-year-old children. This meant that children who were linguistically less skilledbroke more social norms.

At the same time the connection between expressing social norms duringnaturally occurring peer interactions and children’s language developmentlevel has not been investigated. Yet, such a study would provide additionalinformation concerning the study of the moral development of the child andpeer socialization. Also, the studies of moral development in Estonia, a fastdeveloping post-soviet country, are few. And at the time when, for example,studies in parental values in Estonia have shown changes towards a lessauthoritarian parenting style (Tulviste & Ahtonen, 2007), such aninvestigation into moral development is much needed.

This study set out to discover what the social norms that children refer toin peer group interaction are, and how the expression of norms is influencedby the linguistic features of the child and other variables such as the child’sage and gender. Children learn social norms through talk, thus one mightexpect that more talkative children and those with better linguistic abilitieswould refer more often to social norms in interaction with other children.Less talkative children and those whose linguistic abilities are not as goodmay not be able to respond as well or as quickly to the perceived violation ofsocial norms by their playmates. Following from what has been pointed outabove, the study set out with hypothesis that it is the older and linguisticallymore expressive children that mention most of the social norms in a group.Also, since girls have often been found to develop faster linguistically, theywere expected to mention more social norms than boys. Yet, as Tulviste andKoor (2005) have discovered, there could be a difference in the type of rulesmentioned by girls and boys.

METHOD

Participants

Seventy-three Estonian children in day care (Mage 5;6 years; range 3;8 to 7;5)participated in the study, 35 of them were boys (M¼ 5;7 years; range 3;8 to7;5), and 38 girls (M¼ 5;5 years; range 3;10 to 7;1). The parents of thechildren were contacted in four state-run kindergartens in Tartu, the secondbiggest town in Estonia. Children and their parents were informed of theaim of the study and their participation was voluntary. All data werecollected in Tartu, Estonia; all participants were Estonians with Estonian astheir mother tongue. All children that participated attended kindergartenfull time.

LANGUAGE AND SOCIAL NORMS 253

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Procedure

It was explained to the children that the researchers wanted to film themplaying with peers, and they were asked to choose two friends to play with.A radio transmitter was attached to the children and they were video-recorded playing for about 20 minutes (range 17.15 to 22.45 minutes;M¼ 20.39 minutes). Children mostly played in same-sex groups of three, intheir everyday childcare environment that provided toys for any activity. Allvideo recordings were transcribed using the CHAT transcription system(MacWhinney, 1991) to calculate the amount of speech of all the childrenparticipating. Originally 75 children participated, but two (1 boy and 1 girl)were excluded from further analyses since they left the group before 20minutes was up. The most preferred game of the groups of girls and girl-dominated groups was house, a few also played board games and others(ball games, hide and seek, etc.); boys mostly opted for board games andconstruction games.

Linguistic measures

Childes CLAN program (MacWhinney, 1991) was used for the linguisticanalyses of the transcripts. The following measures were used:

1. The mean length of utterance in words (MLU).2. The total number of word roots produced measuring the richness of

vocabulary.

Estonian is an agglutinative language, where the grammatical relations areexpressed by a large number of suffixes added to the word stem. In thecurrent study, the same words in different cases or with different inflectionswere counted as one root; at the same time nouns and verbs using the samestem were considered as two different roots. Different nonce words,exclamations, and sounds used by children (till-tall, ossa, mkm, noh, ohoh,ah, aa, vot, etc.) were removed from the different word count.

3. The number of words produced per minute measuring the child’stalkativeness.

Coding

Piotrowski, (1997) introduced a coding system based on the domain theoryof Turiel (1983) that has proven useful in analysing observational data.Piotrowski distinguished five subcategories of moral rules and foursubcategories of conventional rules; for the description of the subcategories

254 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

see Table 1. The social norms referred to by all the children in the transcriptswere coded by two trained coders separately. The codings were then checkedagainst the original recordings. The coders were in agreement 84% of thetime. Disagreements were resolved in discussion.

The following instances were coded as expressions of social norms: (a)claims on what is right and how things should be done (‘‘Look, this is howyou swing!’’); (b) reactions to transgressions by peers (‘‘No, you can’t!’’,‘‘Don’t!’’, etc.); (c) orders that contain a rule to be followed (‘‘Clean thisup!’’); and (d) in cases of questions that involved a rule, the answer wascoded (‘‘Is this how you do it?’’—‘‘Yes’’).

For analyses the nine different subcategories were cumulated to form thegroups of ‘‘moral rules’’ and ‘‘conventional rules’’, respectively. Also, theratio of social norms mentioned as compared to the total number ofutterances was used in order to eliminate the effect of talkativeness inanalyses where talkativeness was not the variable of interest.

As the context or the activity that the children are engaged in may have asignificant impact on the amount and type of rules mentioned in the

TABLE 1Subcategories of social norms (Piotrowski, 1997)

Moral rules

1 Rights: rules of possession or ownership. ‘‘You can’t, it’s mine!’’

2 Positive justice: turns, sharing and equity. ‘‘You don’t get any more, you already

have so many.’’

3 Harm to others: unkindness, teasing and

hitting.

‘‘Don’t fight!’’

4 Past kindness, prosocial behaviour. ‘‘You have to help us clean.’’

5 General exhortations about morality: no

special topic is mentioned.

‘‘You can’t!’’

‘‘Don’t!’’

Conventional rules

1 Rules of a game. Comments about a

structured game.

‘‘The one who was found first counts.’’

2 Destruction of property/dirtiness of the

house. Comments concerning messes,

cleanliness, includes appropriate

placement of objects.

‘‘The car does not go here.’’

3 General conventions. Includes comments

concerning silly behaviour, requests or

denial of permission.

‘‘Sit nicely!’’

‘‘This in not how you mix it.’’

4 Miscellaneous conventions. Includes gender

appropriate behaviour, assignment of

chores, appropriate language, politeness,

role-appropriate behaviour, references to

an authority figure.

‘‘Don’t be silly!’’

‘‘This Argo said that we have to play

nicely.’’

LANGUAGE AND SOCIAL NORMS 255

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

interaction, the type of play or context was also categorized. It was coded ona separate occasion and three categories were used: pretend play thatimitates social situations, construction games, and other (including differentboard games, and others such as, hide and seek).

RESULTS

Age and gender differences in linguistic measures

The mean number of roots produced during one play session was 111.66(range 20 to 249, SD¼ 50). The mean length of utterances (MLU) was 3.04words (range 1.95 to 4.19, SD¼ 0.53). The mean number of words perminute was 19.62 (range 2.88 to 58.1, SD¼ 0.54).

The general linear models (GLM) two-way MANCOVA for ChildGender (boys vs. girls)6Linguistic Measures (MLU, roots, words perminute) with the age of the child in months as a co-variable showed asignificant effect of the age of child, Wilks’ Lambda (3, 68)¼ 4.92 (p5 .01).The follow-up ANCOVA showed that the effect of age was significant forthe number of words per minute, F(1, 70)¼ 7.25, p5 .01, whereas olderchildren were more talkative. All the correlations among the linguisticmeasures were statistically significant (see Table 2).

Age and gender differences in the use of moral andconventional rules

The mean number of references children made to moral rules during playwas 4.38 (range 1 to 16, SD¼ 3.42) and the number of references toconventional rules was 12.43 (range 1 to 43, SD¼ 9.22).

The following modification was made in order to provide a better pictureof the use of conventional rules and the differences between boys and girlsand children of different ages without the confounding variable of

TABLE 2Correlations between age, linguistic measures, and the social norms used by children

(1) (2) (3) (4) (5)

(1) Age in months –

(2) Mean length of utterance (MLU) .17 –

(3) Roots .20 .65* –

(4) Words per minute .37* .58* .91* –

(5) Moral rules used .30 7.10 .19 .20 –

(6) Conventional rules used .48* .36* .63* .74* .22

Note: *p5 .01.

256 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

talkativeness. The ratio of moral rules mentioned as compared to the totalnumber of utterances by the child (moral rules per utterance) was used as ameasure for moral rules and the ratio of conventional rules mentioned ascompared to the total number of utterances (conventional rules perutterance) was used as a measure of conventional rules in the followinganalyses.

GLM ANCOVA analyses for Child Gender were performed on theMoral Rules and Conventional Rules separately with the age of children asa co-variable. The analyses yielded a significant effect of the age of child onthe use of Conventional Rules, F(1, 69)¼ 6.19, p5 .05, as older childrenmentioned conventional rules more often than younger children. There wasa tendency, but no significant effect of gender on the use of Moral Rules,F(1, 60)¼ 3.35, p¼ .07, whereas boys referred to moral rules slightlymore often than girls. A three-way ANOVA with Gender6Type ofPlay6 Social Norms showed a significant effect irrespective of gender ortype of play. There was no interaction between type of play and gender.

Influence of talkativeness on the use of social norms

In order to investigate the complex influence of the language measures,including talkativeness and other variables, on the use of social norms,multiple regression was chosen as a method. Talkativeness was included as avariable along with gender and age in months, and the analyses were carriedout on the total number of social norms mentioned. Talkativeness waschosen as the variable of interest because it revealed the strongestcorrelation with the use of conventional rules (see Table 2). Other linguisticmeasures were not included as they were highly correlated with talkative-ness. Table 3 presents the results.

TABLE 3Different variables as predictors of the use of moral and conventional rules

Variable Beta

Partial

correlation

Semipartial

correlation p-value

Use of moral rules: R2¼ .12, p¼ .06

Gender 7.25* .04

Age in months 7.07 7.11 7.10 .62

Talkativeness .262 .24 .23 .05

Use of conventional rules: R2¼ .58, p5 .001

Gender 7.10 .21

Age in months .176* .25 .167 .04

Talkativeness .695* .71 .653 .00

Note: *Variables are significant predictors of the use of moral and conventional rules.

LANGUAGE AND SOCIAL NORMS 257

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Moral rules cannot be well predicted by the variables as they account foronly 12% of the variance (p¼ .06); the best predictor is the gender ofchildren (b¼7.25), because boys mentioned them more than girls.

Significant predictors of the conventional rules include the mean numberof words per minute (b¼ .695) and the age of children (b¼ .176), which bothalso predict the use of conventional rules independent of each other. Thismodel accounts for 58% of the variance (p5 .001). Thus, older and moretalkative children frequently mention conventional rules.

DISCUSSION

The study set out to investigate the connections between the social normsreferred to during free play and the linguistic measures of the childrenhypothesizing that the older and linguistically more expressive childrenwould refer to social norms more frequently. Also, gender differences wereexpected. The results are discussed as follows.

Linguistic measures

The MLU for the children who participated in the study was about 3 words.Earlier studies found that the MLU in words for Estonian 4-year-oldchildren was 2.5 (Tulviste & Raudsepp, 1997), for 6-year-old children atmeals 2.73, and during puzzle solving 3.22 (Tulviste, 2001). All these studiesconcentrated on mother–child interaction, therefore the MLUs are not trulycompatible with those of child–child interaction. The absence of the parentcould account for the longer MLUs of the Estonian children in the presentstudy because the interaction is then not restricted by adult dominance. Inorder to be understood by the other child, the child should express himselfor herself more explicitly than in the case of being understood by adults.

The present study did not discover any gender differences in linguisticmeasures. There was no significant difference between the MLUs, thenumber of word roots used, or the number of words per minute produced byboys and girls. Similar results have been reported by Tulviste, Mizera, DeGeer, and Tryggvason (2008) for Swedish and Finnish preschool children inpeer interaction in dyads. Nevertheless, Estonian boys talked less thanEstonian girls. Triadic interaction has not been studied as much, and inEstonia there have not been any previous studies of triads. In general,studies of gender differences in linguistic abilities have revealed controversialresults. Maccoby (2002) has concluded in a review article on genderdifferences that the difference between boys and girls in linguistic skills hasvirtually disappeared.

Similarly, the present study did not reveal any apparent effect of age onthe use of word roots or the mean length of utterance. The results of the

258 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

previous studies also showed that the relationship between MLU and age isnot consistent for children older than 3 years (see Hickey, 1991). Also, theMLU and the number of word roots used do not necessarily provide acomplete picture of the linguistic skills of the child. The childrenparticipating in the study in some cases used quite complicated words andforeign words in Estonian (finis, ‘‘finish’’; kart, ‘‘go-cart’’; vastumurk,‘‘antidote’’), and English expressions (‘‘that was close’’). Yet, at the sametime they made mistakes and used grammatically incorrect forms of words(vanat, annasid, sassisime, odest, etc.). For a better understanding of thesedynamics, qualitative analyses would be necessary.

The differences between the age groups were nonetheless apparent for themeasure of talkativeness. Younger children tended to talk less than the otherchildren. At the same time, all the linguistic measures were strongly correla-ted. Thus, it is fair to conclude that these measures still express some aspect ofchild linguistic expression; whether it is the actual level of development,extraversion, or something else, is a question for further research.

Social norms

Killen and Smetana (1999), and Smetana (1989) discovered that during freeplay children have more moral transgressions than conventional transgres-sions. Also, Nucci and Nucci (1982) discovered that the response of childrendiffers depending on the domain of the transgression. In the present study,in general, children mentioned conventional rules almost three times moreoften than moral rules. Similarly, Piotrowski (1997) found that in themother–child interaction, both parties refer to conventional rules more oftenthan to moral rules. In our case, for example, children playing differentboard games frequently verbalized the rules of the game: Ja mis siis, et sarundasid, kaitsekad ei loe midagi [It doesn’t matter that you attacked, thedefence cards don’t count]; Uhe kaiguga saab mitu korda runnata, aga mitteuhe koletisega [You can attack many times during one turn, but not with thesame monster] (see also Appendix 1).

The age of the children did not have an effect on the references to moralrules; at the same time age was an important factor when the referencesto conventional rules were taken into account. Older children referred toconventional rules significantly more often than younger children. As for theconventional rules, the same idea is also expressed by the significantpredictive power of the age of children. On the one hand, such results couldbe expected because research has shown that older siblings and mothersrefer to conventional rules more often (Piotrowski, 1997). On the otherhand, such results give some reason to conclude that references to moralrules are not considerably influenced by the characteristics of the child.Because there were no age differences in the references to moral rules, it seems

LANGUAGE AND SOCIAL NORMS 259

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

that children of all ages acknowledge them and see to it that they are upheld.As for the conventional rules, it could indicate that older children have abetter knowledge of the game rules and/or authority in imposing them. Forexample, a number of children chose to play board games. They are fairlystructured games and during play the rules of a particular game (Kelle kordon? [Whose turn is it?]; Ring kaib nii pidi [The circle goes this way]) as well asgeneral conventions of how something is done (Nii ei segata [This is not howyou shuffle]; Sa ei oska keerutada—nii kaib! [You don’t know how to turnit—this is how you do it!]) could become a subject of discussion. It is notunreasonable to suggest that older children may have more experience withsuch games, and therefore they discuss the rules more often. At the same timeit could imply that the understanding of conventional rules or the ability toarticulate them in a group develops with age.

There was no well-pronounced gender difference in references to moralrules. Once talkativeness was removed from the use of social norms, therewas only a tendency of boys mentioning moral rules more often than girls,but that tendency did not reach statistical significance. Nevertheless genderwas the only significant predictor of the references to moral rules. In asimilar study, Tulviste and Koor (2005) also found that boys mention moralrules more often. Studies have shown that boys are often physically moreaggressive in play than girls (Ostrov & Keating, 2004). Their play includesdomination and competitiveness (Maccoby, 2002) and their games are alsomore active (Lewis & Phillipsen, 1998). Therefore there might have beenmore moral transgressions in boys’ play; for this reason moral rules arementioned more often (see Appendix 2). Also, if we take into account thefact that all of the variables measured explain only 12% of all the moralrules mentioned, it seems reasonable to conclude that the rest of the variancewould be explained by the actual moral transgressions in the game to whichthe children respond.

The study of the link between moral and conventional rules and linguisticmeasures showed that all the correlations between conventional rules andlinguistic measures were significant and positive. Also, one of the linguisticmeasures, namely talkativeness, predicts the use of conventional rules ratherwell. The fact that the references to moral rules are not influenced by the ageof the child or his/her linguistic measures, might indicate that moral rulesare more important in a play group and moral transgressions elicit areaction no matter how old or how talkative the preschooler is, how richher/his vocabulary is or how complex his/her sentences are.

It is important to note that talkativeness, independent of age, is aconsiderable predictor of the use of conventional rules. Therefore, it seemsthat the linguistic skills are important for imposing conventional rules and ininteraction between equal partners. More articulate children are the oneswho refer and draw attention to the rules of convention in a group of three.

260 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Or it could well be that more talkative children react to the breaking ofconventional rules or simply state the rules in a play group. Whether thedifference is a developmental phenomenon, or has more to do with thedominance of more talkative children in a group, has to be answered byfuture research using appropriate methods to make the distinction.

Previous research has found that reticence does not mean just talking less.The performance of highly talkative preschool children has been found to besignificantly superior on all measures of their linguistic abilities (Evans, 1996;Landon & Sommers, 1979; Van Kleek & Street, 1982). Though languagedevelopment is an important factor in the social development of a child, therelationship between the linguistic skills of children and the acquisition anduse of social norms has not been widely studied. The current researchproposes an interesting link between the linguistic measures of children andtheir social development—the use of social norms. The results seem toconfirm the idea that children do not treat moral and the conventional rules asequal. The present study indicates that in their play children also use thedifferent rules in a different manner and to a different extent. The number ofreferences to moral rules does not seem to depend on linguistic measures ofthe children or their age. At the same time, for the conventional rules to bereferred to linguistic measures and age of the child are important. The studyfound that older children and those children who generally talk more mentionconventional rules more than the others; moral rules seem to be referred toindependently of the influence of the age or talkativeness of the child.

As for further considerations, the scope of such a study could be enlargedby including variables that provide good measures for comparison, such asinterviews with the same children about their actual understanding of socialnorms. It would also be important to look behind the numbers and examinethe social norms that children mention in their interaction on a moredetailed level. Such data might provide a good basis for interestingcomparisons, for example, with minority children in Estonia.

Manuscript received 19 September 2007

Revised manuscript accepted 7 May 2008

First published online 23 March 2009

REFERENCES

Bauer, D. J., Goldfield, B. A., & Reznick, J. S. (2002). Alternative approaches to analyzing

individual differences in the rate of early vocabulary development. Applied Psycholinguistics,

23, 313–335.

Bhatia, S. (2000). Language socialization and the construction of socio-moral meaning. Journal

of Moral Education, 29, 149–156.

Blum-Kulka, S., & Snow, C. (2004). Introduction: The potential of peer talk. Discourse Studies,

6, 291–306.

LANGUAGE AND SOCIAL NORMS 261

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Dodd, B., Holm, A., Hua, Z., & Crosbie, S. (2003). Phonological development. A normative

study of British English-speaking children. Clinical Linguistics & Phonetics, 17, 617–643.

Evans, M. A. (1996). Reticent primary grade children and their more talkative peers: Verbal,

nonverbal, and self-concept characteristics. Journal of Educational Psychology, 88, 739–749.

Gleason, J. B. (1988). Language and development. In F. S. Kessel (Ed.), The development of

language and language researchers (pp. 269–280). Hillsdale, NJ: Lawrence Erlbaum

Associates, Inc.

Helwig, C. C., & Turiel, E. (2002). Children’s social and moral reasoning. In P. K. Smith &

C. H. Hart (Eds.), Childhood social development (pp. 475–490). Oxford, UK: Blackwell

Publishing.

Hickey, T. (1991). Mean length of utterance and the acquisition of Irish. Journal of Child

Language, 18, 553–569.

Killen, M., & Smetana, J. G. (1999). Social interactions in preschool classrooms and the

development of young children’s conception of the personal. Child Development, 70, 486–501.

Landon, S. J., & Sommers, R. K. (1979). Talkativeness and children’s linguistic abilities.

Language and Speech, 22, 269–275.

Leaper, C., & Smith, T. E. (2004). A meta-analytic review of gender variations in children’s

language use: Talkativeness, affiliative speech, and assertive speech. Developmental

Psychology, 40, 993–1027.

Lewis, T. E., & Phillipsen, L. C. (1998). Interactions on an elementary school playground:

Variations by age, gender, race, group size, and playground area. Child Study Journal, 28,

309–331.

Maccoby, E. E. (2002). Gender and relationships: A developmental account. In W. Craig (Ed.),

Childhood social development. The essential readings (pp. 201–219). Oxford, UK: Blackwell

Publishers.

MacWhinney, B. (1991). The CHILDES project: Tools for analyzing talk. Hillsdale, NJ:

Lawrence Erlbaum Associates, Inc.

Nicolopoulou, A. (2002). Peer-group culture and narrative development. In S. Blum-Kulka &

C. E. Snow (Eds.), Talking to adults. The contribution of multiparty discourse to language

acquisition (pp. 117–155). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

Nobes, G. (1999). Children’s understanding of rules they invent themselves. Journal of Moral

Education, 28, 215–234.

Nucci, L. P., & Nucci, M. S. (1982a). Children’s social interactions in the context of moral and

conventional transgressions. Child Development, 53, 403–412.

Nucci, L. P., & Nucci, M. S. (1982b). Children’s responses to moral and social conventional

transgressions in free-play settings. Child Development, 53, 1337–1342.

Nucci, L. P., & Turiel, E. (1978). Social interaction and the development of social concepts in

preschool children. Child Development, 49, 400–407.

Nucci, L., & Weber, L. (1991). The domain approach to values education: From theory to

practice. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and

development (pp. 251–266). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Ostrov, J. M., & Keating, C. F. (2004). Gender differences in preschool aggression during free

play and structured interaction: An observational study. Social Development, 13, 255–278.

Piaget, J. (1950). Cooperation and the moral development of the idea of justice. In The moral

judgment of the child (M. Gabain, Trans.; pp. 195–326). London: Routledge & Kegan Paul.

(Original work published 1932)

Piotrowski, C. (1997). Rules of everyday family life: The development of social rules in

mother–child and sibling relationships. International Journal of Behavioral Development, 21,

571–598.

Siegal, M., & McDonald Storey, R. (1985). Day care and children’s conceptions of moral and

social rules. Child Development, 56, 1001–1008.

262 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

Smetana, J. G. (1989). Toddlers’ social interactions in the context of moral and conventional

transgressions in the home. Developmental Psychology, 25, 499–508.

Smetana, J. G., & Braeges, J. L. (1990). The development of toddlers’ moral and conventional

judgment. Merrill-Palmer Quarterly, 36, 329–346.

Smetana, J. G., Kochanska, G., & Chuang, S. (2000). Mothers’ conception of everyday rules for

young toddlers: A longitudinal investigation. Merrill-Palmer Quarterly, 46, 391–416.

Smetana, J. G., Schlagman, N., & Walsh Adams, P. (1993). Preschool children’s judgments

about hypothetical and actual transgressions. Child Development, 64, 202–214.

Stowe, R. M., Arnord, D. H., & Ortiz, C. (1999). Gender differences in the relationship of

language development to disruptive behavior and peer relations in preschoolers. Journal of

Applied Developmental Psychology, 20, 521–536.

Tesson, G., Lewko, J. H., & Bigelow, B. J. (1987). The social rules that children use in their

interpersonal relations. In J. A. Meachum (Ed.), Social relations: Contributions to human

development (Vol. 8, pp. 36–57). Basel, Switzerland: S. Karger.

Thompson, R. B., & Moore, K. (2000). Collaborative speech in dyadic problem solving:

Evidence for preschool gender differences in early pragmatic development. Journal of

Language and Social Psychology, 19, 248–255.

Tulviste, T. (2001). Can differences in mother–child interaction be explained by context and

collectivistic attitudes expressed by mothers? Applied Psycholinguistics, 22, 541–554.

Tulviste, T., & Ahtonen, M. (2007). Child-rearing values of Estonian and Finnish mothers and

fathers. Journal of Cross-Cultural Psychology, 38, 137–155.

Tulviste, T., & Koor, M. (2005). ‘‘Hands off the car, it’s mine!’’ and ‘‘The teacher will be angry

if we don’t play nicely’’: Gender-related preferences in the use of moral rules and social

conventions in preschoolers’ dyadic play. Sex Roles, 53, 57–66.

Tulviste, T., Mizera, L., De Geer, B., & Tryggvason, M.-T. (2008). Cultural, contextual, and

gender differences in peer talk. Manuscript submitted for publication.

Tulviste, T., & Raudsepp, M. (1997). The conversational style of Estonian mothers. First

Language, 17, 151–163.

Turiel, E. (1983). The development of social knowledge: Morality and convention. Cambridge,

UK: Cambridge University Press.

Turiel, E., & Nucci, L. P. (1978). Social interactions and the development of social concepts in

preschool children. Child Development, 49, 400–407.

Van Kleek, A., & Street, R. (1982). Does reticence mean just talking less? Qualitative differences

in the language of talkative and reticent preschoolers. Journal of Psycholinguistic Research,

11, 609–629.

APPENDIX 1

GAME RULES USED IN THE GROUP OF THREE

Bridget (BRG), Eda (EDT) and Michelle (MCH) playing hide-and-seek.

EDT: Sina oled mangust vaba. [You are out.]BRG: Kui mina mangust vaba olen, siis see tahendab, et mina peidan. [If

I’m out, then this means that I’m hiding.]EDT: Tema on mangust vaba. [She is out.]BRG: Michelle loeb. [Michelle is counting.]

LANGUAGE AND SOCIAL NORMS 263

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010

BRG: Ei, Edith loeb. [No, Edith is counting.]BRG: Meie peidame. [We will hide.]EDT: Mitmeni? [How long?]BRG: Sina loed, kui me kutsume. [You will count until we call you.]MCH: Seeparast, kes on vaba, see ei *. [Because, the one who is out, does

not *.]BRG: Kui me kutsume, siis void tulla. [You can come when we call.]EDT: Michelle leitud! [Found Michelle!]BRG: Michelle loeb. [Michelle will count.]BRG: Mina * keda viimasena leitakse * keda enne leitakse see loeb.

[I * the one who is found last * who is found first will count.]MCH: Okei. [OK.]

APPENDIX 2

EXAMPLES OF MORAL RULES REFERRED TO

Mike, Mart (MRT) and Rob (ROB) playing cards.

ROB: Jagasid valesti. [You dealt wrong.]ROB: Martin, sa jagasid valesti. [Martin, you dealt wrong.]MRT: Sa vaatasid! [You looked!]MRT: See ei ole aus mang! [It’s not fair play!]ROB: On kull. [Yes, it is!]MRT: Niimoodi pole aus mang. [It’s not fair play like that.]ROB: Ma ei anna neid kaarte! [I won’t give these cards!]MRT: Ara anna. [Don’t.]ROB: Mina tahan jagada. [I want to deal.]ROB: Sina jagasid juba. [You already dealt.]%act: Puuab Mardilt kaarte ara votta. [Tries to take cards away from

Mart.]MRT: Ei vota! [You can’t take them!]

264 TOUGU AND TULVISTE

Downloaded By: [Tõugu, Pirko] At: 13:55 25 February 2010