- Social Sciences
- African Studies
- American Studies
- Asian Studies
- Communication Sciences
- Ethnic Studies
- European Studies
- Gender Studies
- Physical Sciences
- Life Sciences
- Animal Communications
- Cell Biology
- Evolutionary Biology
- Food Science and Technology
- Human Anatomy
Tolerance and Intolerance: Children's and Adolescents' Judgments of Dissenting Beliefs, Speech, Persons, and Conduct
by Cecilia Wainryb, Leigh A. Shaw, Camelia Maianu
Tolerance and Intolerance: Children's and Adolescents' Judgments of Dissenting Beliefs, Speech, Persons, and Conduct
Cecilia Wainryb, Leigh A. Shaw, Camelia Maianu
Updated: February 25th, 2013
Child Development, December 1998, Volume 69, Number 6, Pages 1541-1555
Tolerance and Intolerance: Children's and Adolescents' Judgments of Dissenting Beliefs, Speech, Persons, and Conduct
Cecilia Wainryb, Leigh A. Shaw, and Camelia Maianu
Tolerant and intolerant judgments of different types of dissent were examined. One hundred sixty participants (aged 7,3,10,4,13,6, and 20,l) made judgments about dissenting beliefs, speech, practices, and people engaged in those practices. Across all ages, participants were more tolerant (1) of the holding of dissenting beliefs than of their expression, (2) of the expression of beliefs than of the persons engaged in acts based on those beliefs, and (3) of the persons than of the acts. Tolerance of dissenting beliefs and speech increased with age. Although, at all ages, participants were intolerant of the practices, they were relatively more tolerant of practices grounded in cultural contexts. Participants were more tolerant of people espousing dissenting information than dissenting values. We concluded that tolerance and intolerance coexist at all ages and hinge on what individuals are asked to tolerate and on the sense in which they are asked to tolerate it.
Tolerance for dissenting beliefs and conduct is con- sidered to be essential to the sustenance of demo- cratic societies. In On Liberty, John Stuart Mill (1956) argued that tolerance for conflicting opinions encour- ages the exchange of ideas and provides an opportu- nity for identifying inaccurate propositions and pur- suing the truth, and for developing efficient solutions to problems. In addition, tolerance is thought to be essential for the protection of individual autonomy and rights. Tolerance is thus valued on both prag- matic and moral grounds. Nevertheless, a number of social scientists and philosophers have argued that, however desirable on balance, tolerance is not an ab- solute value, and that not every decision to limit cer- tain types of ideas, behaviors, or institutions is an in- stance of narrow-minded intolerance (Bishop, 1989; Hatch, 1983; McClosky & Brill, 1983; Williams, 1972). Instead, societies and individuals must wrestle with questions about the kinds of dissent that are to be tolerated and the grounds on which they are to be tolerated or suppressed. This is particularly true when the dissent refers to different moral beliefs and practices which, although rooted in cultural tradi- tions, may also be seen as oppressive or harmful, such as the subordination of women in some Islamic countries (Goodwin, 1994), bride burning and dowry deaths in India (Bumiller, 1990), or nosebleeding rites among male teenagers in New Guinea (Herdt, 1990). Some have argued that a call for blanket tolerance of diversity (i.e., the injunction that all diversity ought to be tolerated) may result in indiscriminate approval
Or unjust and practices and foster an unwarranted and conservative commit- ment to an undesirable status quo (Baumrind, 1998; Hatch, 1983).
Upholding a tolerant attitude, therefore, requires establishing grounds for distinguishing acceptable from unacceptable dissent. Although philosophers and social scientists have wrestled with the vexing question of the boundaries of tolerance (without, we must add, arriving at an incontrovertible resolution), questions regarding the distinctions that children and adolescents draw to make such decisions, the grounds to which they appeal in so doing, and the age-related changes in patterns of tolerance and in- tolerance remain largely unexplored. Indeed, in spite of the current concern in the social sciences, among educators, and in the popular press for tolerance of diversity, psychological research on these questions has been surprisingly limited.
Most of the existing research on tolerance consists of large-scale public opinion studies conducted over several decades by sociologists and political scien- tists, surveying levels of political tolerance in the adult population (Davis, 1975; Lawrence, 1976; McClosky, 1964; Nunn, Crockett, & Williams, 1978; Prothro & Grigg, 1960; Stouffer, 1955). These large- scale surveys have consistently shown that American adults indeed display both tolerant and intolerant at- titudes. On the one hand, Americans have been shown to endorse freedoms and civil liberties when stated in abstract terms; as an example, data on free- dom of speech (McClosky & Brill, 1983) show that when respondents are asked whether they endorse
01998 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920/98/6906-0006$01.00
freedoms in general terms ("I believe in freedom of speech for all no matter what their view might be"), very high levels (81%-89%) of acceptance of free- doms and rights are obtained. On the other hand, the majority of the same respondents also do not support the application of those freedoms and liberties to cer- tain groups in certain contexts; as an example, over two-thirds of the same respondents said they would deny members of the Nazi Party or of the Ku Klux Klan the right to free speech. Researchers of political socialization have extended these findings to chil- dren (Zellman & Sears, 1971) and adolescents (Avery, 1988; Dennis, Lindberg, McCrone, & Stiefbold, 1969; Jones, 1980; Owen & Dennis, 1987; Patterson, 1979) and demonstrated that, not unlike adults, children and adolescents support democratic rights in the ab- stract but often do not endorse the same rights in spe- cific contexts. Overall, the picture emerging from these surveys is one where individuals across a broad age range are both tolerant and intolerant of diversity in belief and behavior; because of their particular na- ture, however, these survey data cannot answer questions such as why individuals may be tolerant of some issues and not of others, or how those toler- ant and intolerant attitudes develop. Recent research by Helwig (1995,1997) demonstrated that developing concepts of rights and their varying coordination with other social considerations underlie the mixture of endorsement and suppression of civil liberties and may explain, in part, what appears to be a puzzling combination of tolerant and intolerant attitudes.
In comparison to the extensive treatment of toler- ance by sociologists and political scientists, the dearth of developmental research is somewhat per- plexing, particularly given the general interest in fostering tolerance among young children and adolescents. The little developmental research and theorizing that exists has focused on a single dimen- sion of tolerance, namely, judgments about the per- sonal worth of dissenting individuals. In a series of studies (Enright & Lapsley, 1981; Enright, Lapsley, Franklin, & Steuck, 1984), children and adolescents were asked to give their opinion about certain issues and were subsequently asked to judge a hypothetical person who allegedly took the opposite stand. The issues about which dissenting opinions were sam- pled in those studies were diverse, and included deci- sions regarding conflicts between different kinds of considerations or goals (e.g., keeping a prize or re- turning it to the person who has a prior claim to it; obeying a teacher or helping a friend; allowing the American Nazi party to hold a march in a predomi- nantly Jewish community or protecting the commu- nity from potential harm and offense).
Adopting a social-cognitive developmental per- spective, Enright and his colleagues examined mean age differences in modal responses across different dilemmas. Based on their findings, they proposed that tolerance entails a developmental progression from a generalized intolerant attitude during the childhood years, through a tolerant evaluation of dis- agreeing individuals during adolescence, and ending with a level "above open-mindedness" (Enright & Lapsley, 1981, p. 1056) reflected in the tendency of college students to suspend judgment about dis- senting others. Furthermore, Enright and Lapsley (1981) reported that significant relations exist be- tween the level of belief discrepancy and the levels of logical and moral reasoning, and concluded that a cognitive structural constraint operates on the devel- opment of tolerance.
Interestingly, this type of research did not make distinctions among different types of dissenting be- liefs in terms of their content. As noted above, the studies on belief discrepancy included conflicts be- tween different types of social considerations. As a result, although in all cases children were asked to pass judgment on a proponent of a position with which they themselves disagreed, at times the dis- senting position supported injustice or harm to oth- ers, other times the dissenting position was over a conflict between two rights or freedoms about which there is not a clearly moral or immoral resolution, and yet other times the dissenting position did not involve moral considerations at all but, rather, in- volved matters of personal preference or convention. In the belief-discrepancy studies, all items were im- plicitly considered to be conceptually equivalent (in- asmuch as they always involved a position with which the respondent disagreed) and were combined into a single score. It is possible, however, that indi- viduals make distinctions in judging different types of dissent. For example, it may be that individuals are tolerant of the proponents of ideas with which they themselves disagree if those ideas have to do with matters of personal preference or convention more than if they have to do with unjust proposi- tions. Indeed, data from a more recent study which also elicited judgments about the hypothetical propo- nents of a number of disagreeable ideas (Sigelman & Toebben, 1992) indicated that, across a broad age range, the evaluations of the proponents of disagree- able ideas significantly varied as a function of their evaluations of the ideas themselves (disagreeable ideas were rated on a 5 point scale from "good" to "terrible"). Furthermore, the age trends in that study were not consistent; adolescents were more tolerant than young children with regard to the proponents of some ideas, young children were shown to be more tolerant than adolescents with respect to the propo- nents of other ideas, and no age differences in toler- ance were found with respect to the proponents of yet other ideas. These data suggest that children and adolescents take into account the content of the dis- senting belief in judging whether or not it should be tolerated; it is, moreover, possible that children and adolescents differ in their interpretations of at least certain types of dissent.
Another important characteristic of the belief- discrepancy tradition has been its exclusive focus on the dimension of tolerance for the individuals who propose disagreeable or dissenting ideas. Undoubt- edly, refraining from negatively judging the propo- nents of ideas with which one disagrees is an essen- tial component of tolerance. However, there may be other senses or ways in which one may be tolerant. People, for example, may tolerate not just persons with whom they disagree, but also dissenting ideas, public discussion of dissenting ideas, or actual prac- tices based on such ideas. Furthermore, people may actually make distinctions between how tolerant they are of (or whether or not they tolerate), for example, ideas, speech, persons, and conduct, and may appeal to different grounds for making those distinctions. Moreover, it is possible that the various dimensions of tolerance develop differently. Indeed, there is some evidence in the Sigelman and Toebben (1992) study that tolerance for the public expression of dis- senting beliefs increases with age, but willingness to interact socially with the proponents of dissenting be- liefs remains low across age groups. It may, therefore, be important to examine simultaneously various di- mensions of tolerance, to assess what distinctions, if any, children and adolescents make among the differ- ent ways of being tolerant and whether age-related changes in the various dimensions follow similar or different paths.
Overall, the findings of both survey and develop- mental studies suggest that individuals draw distinc- tions in their judgments of tolerance. The lack of con- sistent age differences in tolerance (and, indeed, the consistent pattern of survey findings indicating that both tolerance and intolerance are part of children's and adults' judgments) suggests that what individuals are asked to tolerate and the sense in which they are asked to be tolerant play a central role in their judgments. In the present study, we set out to exam- ine the distinctions that children and adolescents draw between what they tolerate and what they do not tolerate in reference to both the content of what they are asked to tolerate and the sense in which they are asked to be tolerant.
Wainryb, Shaw, and Maianu 1543
Although diversity of belief cuts across many do- mains (as examples, people hold, among others, dif- ferent concepts of human nature, different religious beliefs, different beliefs about mores and etiquette), in this study we focused on diversity of moral values and practices (specifically, those that entail potential injustice or harm to others). Not only do diversity in this domain and tolerance for diversity in this do- main constitute important foci of controversy among social scientists and philosophers (Baumrind, 1998; Bishop, 1989; Hatch, 1983; Williams, 1972) and inter- esting questions in their own right, but they also seem an appropriate arena for examining questions regarding the boundaries and distinctions that chil- dren and adolescents draw between tolerance and in- tolerance.
Over 40 years ago, Asch (1952) cogently argued that, when elucidating the meaning of diverse socio- moral practices, a distinction must be drawn between moral beliefs, which are value judgments about what is right and wrong, and informational beliefs, which are beliefs about matters of fact. Consider, as an ex- ample, a father who forcibly thrusts stiff cane grasses into his teenage son's nose until blood flows. It is likely that most people in the United States would judge such an act to be wrong and would consider it unacceptable for a parent to do it. In making such a judgment, Americans would probably rely on a deeply held moral belief that parents ought to care for their children's well-being and that it is wrong for parents to inflict harm on their children. Note, how- ever, that holding the moral belief that it is wrong to hurt one's children is not, in and of itself, sufficient to make the judgment that nosebleeding one's child is morally wrong. Although this may not be readily apparent, to reach such a decision one's reasoning must also include the belief that nosebleeding a child causes the child harm and produces no benefit; this, however, is a matter of factual beliefs, not values. To make this point more intelligible, consider the possi- bility of someone having a different factual belief. Imagine, for example, that someone believed it to be true and considered it a matter of fact that bloodlet- ting is the only way to rid boys of female pollutants that otherwise block their growth and prevent them from attaining reproductive competence. Given such a belief (which is, incidentally, firmly held in some New Guinea cultures; see Herdt, 1990), the act of bloodletting would no longer be seen as entailing harm. Indeed, under such an understanding of the "facts" of the case, bloodletting would actually be seen as promoting boys' development and, when coupled with the moral belief that parents ought to care for their children's well-being, would be judged a parental duty (not unlike the way Americans view certain painful practices such as vaccinations, teeth extractions, or surgery). The aforementioned sug- gests, then, that to understand and make judgments about socio-moral practices (including practices dif- ferent from one's own) it is necessary to account for both the underlying moral beliefs and the underlying informational beliefs.
Not only is the distinction between moral and in- formational beliefs a central epistemological distinc- tion,' but it also appears to tap a distinction that indi- viduals actually make in their judgments. A growing body of psychological research has demonstrated that, across a broad age range (starting as early as 5year-olds; Wainryb & Ford, 1998), individuals dis- tinguish between moral and informational beliefs and take into account both types of beliefs in their judgments of socio-moral practices. Research has, furthermore, shown that opposite moral evaluations of social practices often hinge on differing informa- tional beliefs (Turiel, Hildebrandt, & Wainryb, 1991; Turiel, Killen, & Helwig, 1987; Wainryb, 1991, 1993).
The reason the distinction between moral and in- formational beliefs is important to the present study of tolerance and intolerance is that it may be relevant to the distinctions that children and adolescents make between what should and what should not be tolerated, and to the grounds on which they make such distinctions. Consider, again, the example of bloodletting. It is likely that one would judge people who engage in such a practice because they believe that bloodletting is necessary for enabling normal de- velopment (presumably an informational belief with which most Americans would disagree) as acting out of ignorance or misinformation. By contrast, one might judge people who engage in such a practice because they believe that it is all right for parents to hurt their children (presumably, a moral belief with which most Americans disagree) as acting out of some sort of badness or immorality. In drawing boundaries between the tolerable and the intolerable, therefore, participants might be more tolerant of in-
1. In proposing the distinction between moral and informa- tional beliefs, we do no mean to imply that the two types of be- liefs can always be unambiguously disentangled; indeed, what one believes to be true may, at times, be "contaminated" by one's moral beliefs, and vice versa. Nevertheless, the existence of mixtures and overlaps between the two types of beliefs does not, in and of itself, preclude the validity of the distinction. In- deed, it has been proposed (Wainryb, 1991) that this distinction is not only of theoretical and methodological importance for ex- amining moral diversity in cases where the two types of beliefs are easily disentangled, but can also inform analyses of judg- ments in cases in which the two types of beliefs cannot be distin- guished in a straightforward way.
formational dissent than of moral dissent, as they use the type of underlying belief to infer intentions.
Because, in general, diversity of belief might be part of a cultural belief system or, alternatively, be idiosyncratic to an individual and inconsistent with predominant cultural views, in this study we also contrasted tolerance for dissenting moral and infor- mational beliefs which were commonly held within a culture with tolerance for the same dissenting beliefs when they were not grounded in a cultural belief sys- tem. We thought it likely that participants would be more tolerant of dissent grounded in a cultural belief system than of comparable dissent lacking such a cul- tural grounding.
In addition to examining the effects of varying the content and the cultural status of the dissent, this study explored the distinctions that children and ad- olescents make among the different senses in which individuals may be asked to be tolerant (hereafter re- ferred to as the "dimension" of tolerance), as well as the grounds on which they make those distinctions. To pursue these questions, we examined partici- pants' tolerance for the holding of dissenting beliefs, for the public expression of such beliefs, for the actual practices based on such beliefs, and for the persons engaged in those practices. In general, we expected that participants would view the holding of dissenting beliefs and the public expression of such be- liefs as not causing harm to other people and thus would be relatively tolerant of both. By contrast, we thought it likely that when judging actual practices based on dissenting beliefs, their focus would be on the harm or injustice ensuing from those practices; less tolerance, therefore, was expected on this dimen- sion. When making judgments about a person, partic- ipants might focus on the person's intentions and be relatively more tolerant of persons espousing differ- ent informational beliefs than of persons espousing different moral beliefs; alternatively, participants might be guided by the harm ensuing from the per- son's acts and be relatively intolerant across the board.
Finally, as opposed to a linear progression from intolerance to tolerance, we expected that both toler- ant and intolerant judgments would be found at all ages. Age-related patterns, as they interact with the dimensions of tolerance, the types of dissenting be- lief, and the context in which beliefs are embedded, were examined.
The sample included 160 participants, 20 males and 20 females in each of four grade levels: first grade (mean age 7,3), fourth grade (mean age 10,4), seventh grade (mean age 13,6), and college undergraduates (mean age 20,l). Participants were middle class and primarily (8O0/0) European American (the sample also included 5% Asian, 4% African American, and 11% Hispanic), and attended public schools and a public university in a mid-size Western city.
Procedures and Assessments
The overall purpose of this study was to examine participants' tolerance for moral beliefs and practices with which they themselves disagree. The stimuli in this study, therefore, were short stories describing a person who engages in a harmful or unfair practice because he or she has either a moral belief or an infor- mational belief with which participants themselves disagree.
The type of dissenting belief in the stimuli was varied in a between-subjects manipulation. Half the partici- pants were presented with a character who engages in a potentially harmful or unfair act because he or she has a moral belief different from the participants' belief (Moral); the other half of the participants were told about a character who engages in the same act because he or she has an informational belief differ- ent from that of the participants (Informational). To allow for generalizability across content areas and to reduce the effects of a monomethod bias, two compa- rable versions of the stimulus were designed, one dealing with psychological harm and the other with inj~stice.~
One version described a teacher who puts her students down and calls them names when they make mistakes. In the Moral condition, the teacher believes that "it is okay to be mean to children and hurt their feelings"; in the Informational condition, the teacher believes that "the way to teach children is to put them down when they make mistakes." The other version described a father who allows his sons to play outside and visit with their friends but makes his daughters stay home and does not let them visit their friends. In the Moral condition, the father be- lieves that "it is okay to be nicer to boys than to girls and to give boys more privileges than girls"; in the Informational condition, the father believes that "boys are smarter than girls and know how to stay out of trouble, and girls are not as smart and always get in trouble."
The context of the dissenting belief was varied in a within-subjects manipulation. Participants were pre- sented with the same event twice. In one case, the
2. The entire set of stimuli and examples of fully scored pro- tocols from each age group can be obtained from the main au- thor.
Wainryb, Shaw, and Maianu 1545
dissenting (moral or informational) belief was pre- sented as being held by a person in a country where everyone traditionally shares that belief (Cultural Context). In the other case, the dissenting (moral or informational) belief was presented as being held by a person in a country where people do not share such a belief (No Cultural Context). The presentation order of the two contexts in the within-subjects manipula- tion was counterbalanced within each grade level.
The dimension of tolerance was tapped by assessing participants' judgments about the holding of the dis- senting belief (Belief ), the public expression of that belief (Speech), the act based on the belief (Act), and the person engaged in the act (Person), using the fol- lowing assessments:
Belief: This [father/teacher] I'm telling you about believes that [dissenting moral or informational be- lief]. Do you think it is alright for [him/her] to believe that, or do you think that [he/she] should not believe that? Why?
Speech: This [father/teacher] wants to go to [the next PTA meeting/a big teachers' conference]; [he/ she] wants to give a speech and tell all the [parents/ teachers] that [dissenting moral or informational be- lief]. Do you think that people should stop this [fa- ther/teacher] from saying that, or do you think that they should let [him/her] say what [he/she] be- lieves? Why?
Act: This [father lets his sons but does not let his daughters play outside and visit with friends/ teacher puts her students down and calls them names] because [he/she] really believes that [dis- senting moral or informational belief]. Do you think that [he/she] should be allowed to do that, or do you think that [he/she] should not be allowed to do that? Why?
Person: This [father lets his sons but not his daugh- ters play outside and visit with friends/teacher puts her students down and calls them names] because [he/she] really believes that [dissenting moral or in- formational belief]. Do you think that [he/she] is a pretty good person or not such a good person? Why?
To establish that the underlying beliefs were such that participants disagreed with them, we followed a two-step procedure. First, a number of different moral and informational beliefs relevant to practices involving harm or injustice were pilot tested with children in different age groups; care was taken to ascertain that participants disagreed with beliefs hy- pothesized to be dissenting moral beliefs on the basis of value judgments (e.g., "it's unfair to give boys more privileges than girls"; "it's wrong to hurt chil- dren's feelings like that") and with beliefs hypothe- sized to be dissenting informational beliefs on the ba- sis of factual judgments (e.g., "it's not true that girls are not as smart as boys"; "children don't really learn when you put them down, they learn when you ex- plain things to them and let them make their own mistakes"). The next step was to ensure that no par- ticipants agreeing with the dissenting beliefs were in- cluded in the sample. To that end, a baseline assess- ment was included prior to the main assessments of the study; participants were told about the dissenting belief and were asked their opinion about it (e.g., "There is a person who believes that when children make mistakes, the best way to teach them is to put them down and to call them names. Do you think that's how children learn not to make mistakes, or do you think that's not how children learn? Why?"). Participants who agreed with the belief were replaced. Participants were individually interviewed; interviews were tape-recorded and later transcribed for analysis.
Scoring and Reliability
Scoring categories were formulated based on scor- ing systems used in previous related studies (David- son, Turiel, & Black, 1983; Wainryb, 1991, 1993) and were further elaborated by scoring 20% of the inter- view protocols. Participants' judgments were scored as tolerant (and assigned a score of 1)if they stated that it was all right for the character to hold the dis- senting belief, that he/she should be allowed to pub- licly speak about it, that he/she should be allowed to engage in an act based on that belief, or that he/ she was a good person. Judgments were scored as nontolerant (and assigned a score of 3) if participants stated that the character should not hold the dis- senting belief, that he/she should not be allowed to publicly speak about it, that he/she should not be allowed to engage in the act, or that he/she was not a good person. Judgments were scored as mixed (and assigned an intermediate score of 2) if participants gave both tolerant and nontolerant responses to a question.
In addition to scoring the judgments as tolerant, mixed, or nontolerant, the justifications given for those responses were also coded using the following categories (multiple justifications were allowed):
Harmlessness: reference to the lack of harm ensu- ing from an act (e.g., "it's okay for him to think what- ever he wants, there isn't any harm in just believing something").
Tradition: reference to the belief's or act's cultural grounding (e.g., "it's okay for her to do that because that's what everyone else in her country does, that's the way they do things there").
Intellectual exchange: reference to the positive consequences believed to ensue from expressing and
discussing ideas with others (e.g., "they should let her give that speech even if what she says is not true because other people can also speak about their opin- ions and they can discuss it and maybe she'll figure out that she was wrong").
Positive intent: reference to the character's positive intention (e.g., "he's a good person because he really thinks that he's doing what's best for his daughters, he really thinks that he's protecting them and he's trying to do right by them").
Ignorance: reference to the character's ignorance or socialization (e.g., "she's not a bad person because she doesn't know any other way, she was brought up to think that way").
Harmful/Unfair consequences: reference to the act's harmful or unfair consequences (e.g., "he shouldn't do that because it's really unfair to let only the boys have fun, he should give all the kids the same kinds of privileges, he should treat them the same, that's just not fair what he's doing").
Leads to action: reference to the assumption that there is a connection between believing something and acting on that belief (e.g., "he shouldn't think that because then he'll treat his daughters that way").
Falseness: reference to the false or inaccurate na- ture of the content of the belief or speech (e.g., "they shouldn't let him say that because girls are just as smart as boys, he shouldn't go around saying things that aren't true").
Scoring reliability was assessed through recoding of 20% of the protocols. Interjudge agreement was 97% for the scoring of judgments (Cohen's kappa = .94), and 89% for the scoring of justifications (Cohen's kappa = .87).
All participants disagreed with the dissenting be- liefs; no participants were replaced. The large major- ity of participants justified their disagreement with dissenting moral beliefs with references to value judgments (89%) and their disagreement with dis- senting informational beliefs with references to fac- tual judgments (98%).
Preliminary analyses of participants' judgments and justifications by sex and version yielded no sig- nificant effects or interactions. Those variables were dropped from subsequent analyses. A repeated-measures ANOVA by dimension, belief context, be- lief type, and age, with dimension and belief context as repeated measures, was performed on partici- pants' judgments. Subsequently, post hoc compari- sons using Duncan multiple-range tests and Bonfer- roni t tests were performed to test for significant between-subjects and within-subjects effects, respec- tively. The repeated-measures ANOVA yielded sig- nificant effects for dimension, F(3,435) = 105.03, p < .001, belief context, F(l, 145) = 24.53, p < .001, belief type, F(1, 145) = 10.70, p < .001, and age, F(3, 145) = 17.34, p < .001. Overall, participants were more tolerant of holding dissenting beliefs than of the ex- pression of such beliefs, were more tolerant of the expression of dissenting beliefs than of the persons engaged in the acts, and were more tolerant of the persons than of the acts themselves, ps < .01. Across all dimensions, participants made more tolerant judgments when the underlying dissenting beliefs were informational (as opposed to moral) and when the dissenting beliefs and acts were part of a cultural context (as opposed to when they were not). First graders were, overall, less tolerant than fourth graders, seventh graders, and undergraduates, ps < .05. However, as indicated by significant dimen- sion X belief type, F(3,435) = 2.97, p < .032, dimen- sion X age, F(9,435) = 6.85, p < .001, and dimension X age X belief context, F(9, 435) = 2.49, p < .009, interactions, these effects varied for the different di- mensions. Therefore, separate repeated-measures ANOVAs by age, belief type, and belief context (with belief context as a repeated measure) were performed on the judgments for each dimension.
As shown in Table 1: the majority of participants (between 40% and 58% of first graders, and between 75% and 90% of all other participants) made tolerant judgments of the holding of dissenting beliefs. Their judgments were not affected by the type of belief or by the context of the belief; however, participants' tolerance for belief was significantly affected by age, F(3, 150) = 10.48, p < .001. Post hoc contrasts in- dicated that first graders were less tolerant than fourth graders, seventh graders, and undergraduates, ps < .05.
As also shown in Table 1, participants were more divided in their judgments regarding public speech about dissenting beliefs. Whereas the majority of first graders (between 65% and 79%) made nontolerant judgments, the majority of seventh graders and un- dergraduates (between 55% and 95%) made tolerant judgments (fourth graders were about equally di-
3. Although mean judgments were used in the analyses, per- centages were also included in Table 1 and reported in the text, as they cast additional light on the distinction between tolerant and nontolerant responses, on the one hand, and mixed re- sponses, on the other, and on the connection between partici- pants' judgments and justihcations (presented in Table 3).
Wainryb, Shaw, and Maianu 1547
vided between tolerant and nontolerant judgments). Tolerance for speech about dissenting beliefs was sig- nificantly affected by age, F(3,150) = 19.83, p < .001. Post hoc contrasts indicated that first graders were less tolerant than all other participants, and that fourth graders were less tolerant than undergradu- ates, ps < .05. Overall, participants were slightly more tolerant of the expression of dissenting infor- mational (63%) than of dissenting moral (52%) be- liefs, F(l, 150) = 4.03, p < .046.
In contrast to their relatively tolerant judgments regarding the holding of dissenting beliefs and their public expression, the large majority of participants expressed nontolerant or mixed views about the ac- tual practices based on such beliefs. Interestingly, the type of belief underlying the practice made no sig- nificant difference. Also of interest is that age did not affect participants' tolerance along this dimension: Younger and older children were equally intolerant of the acts themselves. Their judgments, nevertheless, were affected by the context of the practice; as shown in Table 1, more participants made nontolerant judg- ments of practices that were not part of a cultural tra- dition (between 60% and 100%) than of identical practices that were part of a cultural tradition (be- tween 40% and 75%), F(1, 150) = 34.41, p < .001.
Participants' judgments of the person engaged in the act were affected by the type of dissenting belief underlying the practice and by the context of such beliefs. Overall, participants were more tolerant of persons engaged in potentially harmful or unfair practices who held a dissenting informational belief, F(1,150) = 14.38, p < .001, and were also more toler- ant of persons engaged in potentially harmful or un- fair practices in the context of a cultural tradition, F(l, 150) = 14.14, p < .001. Tolerance for persons was also affected by age, F(3, 150) = 6.10, p < .001: seventh graders and undergraduates made more tolerant judgments than did first and fourth graders, ps < .05. A significant interaction between age and belief con- text, F(3, 150) = 7.62, p < .001, further indicated that seventh graders and undergraduates made more tol- erant judgments than did first and fourth graders only of persons engaged in practices that are part of a cultural tradition, ps < .05; tolerance of persons en- gaged in similar practices that were not part of a cul- ture remained fairly stable (and low) across age groups.
Relations between Dimensions
Correlations among judgments in the four dimen- sions were conducted for each age group. As shown in Table 2, among first graders, tolerance for speech, act, and person were all significantly correlated, ps <
Table 1 Percentage of Tolerant, Mixed, and Nontolerant Judgments, and Mean Judgments by Dimension, Belief Context, Belief Type, and Age
|No cultural context:|
|No cultural context:|
|No cultural context:|
|No cultural context:|
Note: M =dissenting moral belief; I =dissenting informational belief. Means are based on a 3 point scale (1 =tolerant; 2 =mixed; 3 =nontolerant).
.01; in addition, tolerance for belief and for speech speech and for act were significantly correlated, ps < were also correlated, p <.05. By contrast, there were .05; among undergraduates, there was a significant fewer significant correlations among the judgments correlation between tolerance for act and for person, of older participants. Among fourth graders, toler- p <.01; no significant correlations were found among ance for belief and for speech and tolerance for the judgments of seventh graders.
Wainryb, Shaw, and Maianu 1549
Table 2 Correlations between Judgments in Four Dimensions, by Age
Note: The four correlation coefficients reported for each pair of variables are for first, fourth, and sev- enth graders, and for undergraduates, respectively; ns range from 37 to 40.
* p < .05; **p < .01.
Participants' justifications for tolerance and in- tolerance of belief, speech, act, and person are pre- sented in Table 3. Multivariate analyses of variance (MANOVA) by age, belief type, and belief context (with belief context as a repeated measure) were per- formed on the proportional use of justifications for each dimension (proportional data were subjected to an arcsine transformation to normalize the dis- tribution). Unelaborated justifications and justifi- cations with small frequencies (<5%) were not in- cluded in the MANOVAs or in subsequent analyses.
The analysis of the justifications of Belief yielded significant effects for age, p < .001, belief context, p < .001, and an age X belief context interaction, p < .001. The analyses of the justifications of Speech and of Person yielded significant effects of age, ps < .005, belief context, ps < .001, and belief type, ps < .019. The analysis of the justifications of Act yielded significant belief context, p < .001, and belief type, p < .001, effects. Subsequently, separate ANOVAs by age, belief type, and belief context (with belief context as a repeated measure) were performed on the pro- portionate use of the main justification categories for each dimension.
Tolerance of belief. The category of harmlessness ac- counted for the majority of participants' reasons for why it is all right to hold different beliefs. The appeal to harmlessness was significantly affected by belief context, F(1, 152) = 27.74, p < .001, and age, F(3, 152) = 6.51, p < .001. Participants used this category more often when dissenting beliefs were not part of
a culture than when they were; overall, first graders justified their tolerance for belief on the basis of harmlessness less often than all other participants, p < .05. A significant age x belief context interaction, F(3,152) = 8.32, p < .001, indicated that this category was used less often by first graders than by other par- ticipants when dissenting beliefs lacked cultural grounding; when the dissenting beliefs were part of a cultural context, first graders and undergraduates used this category less frequently than did the other participants, ps < .05. The category of tradition was also used by participants to justify tolerance for belief when discrepant beliefs were part of a cultural con- text but not at all when the beliefs were not part of a cultural context, F(1, 152) = 49.70, p < .001. The appeal to tradition was affected by age, F(3, 152) = 6.63, p < .001. Undergraduates appealed to this type of reasoning more often than did fourth and seventh graders who, in turn, used it more often than did first graders, ps < .05. As indicated by a significant age X belief context interaction, F(3, 152) = 6.63, p < .001, the age difference in the use of the appeal to tradition was significant when justifying beliefs that were part of a cultural context; no participants used this cate- gory when justifying beliefs that were not culturally grounded. Participants' nontolerant judgments of dissenting beliefs were justified mainly with refer- ence to the assumption that believing something is intrinsically connected to behavior (referred to as "leads to action"). As indicated by a significant age effect, F(3, 152) = 6.88, p < ,001, the use of this type of reasoning was common among first graders but rare among all other participants, ps < .05.
Tolerance of speech. The categories of harmlessness and intellectual exchange accounted for the majority of participants' justifications for why it is all right to publicly express dissenting beliefs. In general, the use of both reasons was affected by age, Fs(3,152) = 14.07 and 7.13, respectively, ps < .001. Undergraduates ap- pealed to harmlessness more often than did seventh graders and fourth graders who, in turn, evidenced this reasoning more often than did first graders, ps < .05. With regard to intellectual exchange, seventh graders used this justification more often than all other participants, ps < .05. In addition to those two main reasons, the category of tradition was used by participants to justify tolerance for the public expres- sion of beliefs that were part of a cultural context but not when such context was lacking, F(1,152) = 26.40, p < .001. Participants' nontolerant judgments of the public expression of dissenting beliefs were justified mainly in terms of the assumption that speaking about something leads people to change their minds and to act on those beliefs (referred to as "leads to
|First Grade||Fourth Grade||Seventh Grade||Undergraduate|
|M I||M I||M I||M I||M I||M I||M I||M I|
|Leads to action|
|Leads to action|
|Leads to action|
|Leads to action|
Unelaborated Note: M = dissenting moral belief; I = dissenting informational belief.
action"). This type of reason was used to justify intol- erance for the expression of dissenting moral beliefs more often than dissenting informational beliefs, F(1, 152) = 11.13, p < .001. As indicated by a significant age effect, F(3,152) = 9.39,p <.001, first graders used this type of reasoning more often than did fourth and seventh graders who, in turn, used it more often than undergraduates, ps < .05.
Tolerance of act. The most common reason for justi- fying nontolerant judgments of the practice itself was that the practice had harmful or unfair consequences for others. This reason was used more often to justify intolerance for acts based on dissenting moral beliefs and for acts that are not part of a cultural tradition, Fs(1, 152) = 43.64 and 18.42, respectively, ps < .001. Interestingly, age did not affect the usage of this type of reason, as participants of all ages focused on the acts' harmful or unfair consequences in justifying their nontolerant judgments. Another reason used with some frequency to justify nontolerant judg- ments of acts was the lack of factual validity of the belief on which the acts were based ("falseness"); this reason was used only with reference to acts based on dissenting informational beliefs, F(1, 152) = 50.14, p < .001. The relatively few tolerant responses for acts, as well as the positive aspects of the mixed re- sponses, were based on one of two reasons. Tradition was used more commonly to justify acts that were common in a culture, F(1, 152) = 24.64, p < ,001, and acts based on dissenting moral beliefs, F(1,152) =5.07, p < .026; the positive intent of the person engaged in the act was referred to more often when justifying acts based on dissenting informational, but not moral, beliefs, F(1, 152) = 15.08, p < .001.
Tolerance of person. The most common reason used for justifying nontolerant judgments of persons was the harmful or unfair consequences of their acts; this reason was used more frequently to justify nontoler- ant judgments of persons engaged in acts based on dissenting moral beliefs, F(1, 152) = 10.83, p < .001, and persons engaged in acts that were not part of a cultural context, F(1, 152) = 8.85, p < .003. The use of this type of reasoning was also affected by age, F(3, 152)= 6.37, p < .001; overall, first graders appealed to this reason more often than did fourth graders who, in turn, used it more often than did seventh graders and undergraduates, ps < .05. Tolerant judgments of persons were justified on the basis of one of two rea- sons. Tradition was used more often to justify toler- ance for persons engaged in practices common in a culture, F(1,152) = 25.84, p < .001; the inferred posi- tive intent of the person engaged in the act was re- ferred to when evaluating persons motivated by dissenting informational, but not moral, beliefs, F(1, 152) = 31.93,p < .001. It should be noted that 14% of justifications regarding tolerance and intolerance of persons were unelaborated (as opposed to 3% un- elaborated responses for Act and 8%-9% for Belief and Speech, respectively). Further inspection of these responses indicated that, whereas the majority of unelaborated justifications in the other dimensions was given by participants in the two younger age groups (83% for Belief, 93% for Speech, and 83% for Act), 61% of the unelaborated justifications in the Person dimension were given by seventh graders and under- graduates. Furthermore, whereas unelaborated justi- fications in the other dimensions were distributed equally between tolerant and nontolerant judgments, 78% of unelaborated justifications in the Person di- mension were given in the context of tolerant judg-
Wainryb, Shaw, and Maianu 1551
ments; participants reasoned that in spite of engaging in a "bad" act, the person was nevertheless a "good" person, but could not elaborate further on why the person was a good person. A repeated-measures ANOVA performed on the proportion of unelabo- rated justifications by age, belief type, and belief context (with belief context as a repeated measure) yielded a significant belief context effect, F(1, 152) = 11.46, p < .001; unelaborated responses were more frequent with respect to persons engaged in acts that were not common in a culture.
In this study, we examined children's and adoles- cents' tolerance and intolerance for different types of dissenting beliefs, speech, practices, and the people engaged in those practices. Overall, the findings demonstrated that patterns of tolerance and intoler- ance coexist at different age levels, and that the dis- tinctions children and adolescents make between what they tolerate and what they do not tolerate are a function of a number of parameters (in addition to age), including the specific dimension of tolerance being examined, the type of dissenting beliefs being tolerated, and the cultural context of such beliefs. We conclude, therefore, that judgments of tolerance and intolerance hinge on what children and adolescents are asked to tolerate and on the sense in which they are asked to tolerate it.
The sense in which participants were asked to be tolerant (what we labeled "dimension" of tolerance) was found to be a major source of variance (partial q2 = .42). Across all age groups, participants were more tolerant of the holding of dissenting beliefs than of the public expression of such beliefs, were more tolerant of the expression of dissenting beliefs than of the persons engaged in acts based on those beliefs, and were more tolerant of the persons than of the acts themselves. Not only were there significant mean dif- ferences in judgments among the four dimensions of tolerance, but very few participants (5%) gave all tol- erant or all nontolerant responses across all dimen- sions (interestingly, this was not equally distributed among age groups: whereas 20% of first graders gave all nontolerant responses across dimensions, no fourth and seventh graders and only 2% of under- graduates did; more about this below). Furthermore, correlational analyses suggested that, at least by early adolescence, responses to the different dimensions of tolerance are differentiated. As will be discussed be- low, we suggest that even what appears to be a global tendency toward intolerance among young children might not be such, but rather a function of the specific
assumptions young; children make about the relation between beliefs and behaviors.
Results also indicated that judgments along the various dimensions of tolerance were differentially constrained by different parameters of the situation. In general, tolerance for both the holding of dis- senting beliefs and their public expression varied with age; tolerance for actual practices was affected by whether or not those practices were grounded in a cultural context; tolerance for people was affected by the type of dissenting beliefs people espoused. The examination of the effects of age, the type of dis- senting beliefs, and the context of those beliefs for each dimension of tolerance thus further illustrates the complex coexistence of tolerance and intolerance.
For the most part, participants in this study indi- cated that it is all right for people to hold beliefs with which they themselves disagree on the basis that sim- ply believing something does not carry with it nega- tive consequences for others even if the beliefs held are immoral or untrue. The only exception to what seems to be a generalized tolerant attitude with re- spect to the holding of dissenting beliefs was the re- sponses of the youngest participants. Although nearly half of first graders in this study did make tol- erant judgments of dissenting beliefs, the other half stated that it was wrong for people to hold such im- moral or untrue beliefs. These youngest participants reasoned that there is an intrinsic or inevitable con- nection between believing something and acting in accord with that belief. Although this type of reason- ing was systematically probed when it occurred (e.g., "Suppose he doesn't do X, suppose he just thinks X; is it okay or not okay for him to just think that?"), it nevertheless persisted among a large number of first graders, who stated that people naturally act out their beliefs.
The majority of first graders and nearly half of the fourth graders revealed a similar type of reasoning when justifying their nontolerant judgments of the public expression of dissenting beliefs. In this case, the argument was that publicly expressing immoral or untrue beliefs leads other people to agree with and act on such beliefs. By contrast, the majority of older participants (seventh graders and undergraduates) made tolerant judgments of the public expression of dissenting beliefs on the grounds that expressing a belief in public does not, in and of itself, harm others and that the ensuing exchange of opinions often helps identify inaccurate ideas and pursue the truth.
Notably, it was only with respect to these two di- mensions of tolerance, namely, the holding of dissen- ting beliefs and the public expression of dissenting beliefs, that significant age differences were found.
Independently of whether the dissenting beliefs were informational or moral or of whether or not they were part of a cultural context, young children were quite intolerant and older participants were in- creasingly more tolerant of both these forms of dis- sent. Although these data can be taken to mean that young children display a global tendency toward in- tolerance, their intolerant judgments appear to be related to the way they construe the relation between believing or speaking, on the one hand, and behavior, on the other. Our data also suggest that age-related changes in the assumptions about the connection between beliefs and behavior (rather than changes in a global social-cognitive structure, as suggested by Enright & Lapsley, 1981) may be at the basis of the observed increase in tolerance for dissenting beliefs and their expression.
In contrast to the increasingly tolerant attitude ob- served with respect to believing and expressing im- moral or untrue beliefs, no age differences were found on the Act dimension, as the majority of partic- ipants of all ages made nontolerant judgments of the practices based on dissenting beliefs. In other words, whereas with age participants increasingly thought that it was all right for people to believe something that they themselves thought untrue or immoral and to publicly express those ideas, most participants at all ages also thought that it was wrong for people to act on such beliefs because of the harm or injustice that ensues from such practices.
It is important to note that within the context of what appears to be a generalized intolerant attitude with respect to actual practices, participants in this study were relatively more tolerant of practices that were part of a cultural tradition. In some respects, this finding might be thought to be inconsistent with a large body of moral development literature which has demonstrated that individuals judge acts entail- ing harm to others or injustice negatively irrespective of the cultural context in which the acts take place (Turiel, 1983; Turiel et al., 1991). The present results, however, do not stand in conflict with this literature, as the large majority of participants in this study did not simply judge harmful or unfair acts in other cul- tures positively, but rather made mixed judgments about such acts. The analysis of participants' justifi- cations further elucidated the meaning of this intriguing finding. Whereas practices not embedded in a cultural context were judged according to the ensuing harm or injustice, (identical) practices em- bedded in a cultural tradition were judged in terms of both the ensuing harm or injustice and the impor- tance of cultural traditions and belief systems. In other words, when judging practices in other cultures, participants (across the age range) did not sim- ply overlook the practices' negative consequences but rather attempted to coordinate their concerns for welfare and justice with their concerns for the integ- rity of cultural systems.
As opposed to their somewhat generalized lack of tolerance for the acts themselves, participants were relatively more accepting of the persons engaged in those acts. In general, children and adolescents were
relatively more accepting of people engaged in po- tentially harmful or unfair practices who held dis- senting informational beliefs than of those who held dissenting moral beliefs. As expected, they reasoned that persons holding dissenting informational beliefs, although misinformed, were well-intentioned. Re- sults also indicated that older participants were rela- tively more tolerant than younger participants of per- sons engaged in potentially harmful or unfair practices in the context of a cultural tradition; their reasoning in this context was often labeled "unelabo- rated," as they judged that those persons are good people but could not explain why. This type of rea- soning is, to some extent, reminiscent of the tendency of college students (as observed by Enright & Laps-ley, 1981) to suspend judgment about dissenting oth- ers (although it was clearly circumscribed to the di- mension of tolerance for persons, and did not emerge in the context of other dimensions). Further research on tolerance of persons is needed to examine in more detail the judgments of blame and responsibility chil- dren and adolescents make about persons whom they see as operating in accord with the constraints of cultural belief systems and long-held traditions.
As a whole, the findings of this study suggest that even in American society, where tolerance can be thought to be the official norm as well as a valued goal, individuals wrestle with the conditions under which certain ideas or behaviors should or should not be tolerated. As they try to locate the fine line between tolerance and intolerance, children and ado- lescents weigh and account for different aspects of ideas, behaviors, and their contexts and consequences. This study also indicated, however, that children and adolescents do not settle for one "fine line" between tolerance and intolerance. Although we are accustomed to thinking of tolerance and in- tolerance as though they were polarities, this study demonstrates that individuals' decisions over whether a particular idea or behavior should be toler- ated involve a variety of considerations and may fall on either side of the "tolerance-intolerance" balance. Furthermore, this study also made it apparent that judgments that certain ideas or behaviors should not be tolerated are not necessarily indicative of narrow-
Wainryb, Shaw, and Maianu 1553
minded intolerance. Although it is often thought that intolerance is easier than tolerance, that it involves a less complex decision-making process and does not require as much deliberation (McClosky & Brill, 1983), the findings of this study indicated that, to a large extent, participants' nontolerant responses re- flected a discriminating moral position that coordi- nates the various grounds of legitimate dissent with the perceived ensuing harm or injustice of beliefs and practices. This pattern of results, of course, is bounded by the type of dissenting beliefs and prac- tices examined in this study, which entailed moral issues over which there is wide cultural consensus. One of the challenges of future research is to delin- eate the patterns of tolerance and intolerance as they emerge with reference to other types of dissent (e.g., dealing with religious beliefs, political beliefs, beliefs about conventions, and so on), including issues over which cultural consensus does not exist.
As depicted in this study, tolerance emerges as a product of deliberation. This should clearly not be taken to mean that cognitive and personality biases play no role in individuals' judgments of tolerance and intolerance. Indeed, this study has not looked into this important question. We can only conclude from this study that although personality or cogni- tive characteristics may dispose individuals to toler- ate or suppress the expression of dissent by people whose ideas or behaviors they themselves disagree with (Adorno, Frenkel-Brunswick, Levinson, & Sanford, 1950; Rokeach, 1960), in making those judg- ments children and adolescents take into account various parameters of what they are asked to tolerate and the sense in which they are expected to be toler- ant, refer to different bases for justifying their deci- sions (including the truthfulness and morality of ideas, people's socialization history and intentions, and the cultural contexts in which people operate), and make differentiated judgments. No single global construct (of tolerance or intolerance) has emerged from this study as guiding all decisions; we thus sug- gest that personality or cognitive biases, where they exist, will be constrained accordingly.
The complex social interactions characteristic of life in a multicultural society have underscored the need to examine the kind of education necessary for dealing responsibly with pluralism of belief and con- duct. Within the last 20 years, substantial work has been conducted in multicultural education (Banks, 1992; Lynch, 1986; Modgil, Verma, Mallick, & Modgil, 1986; Nieto, 1996; Vogt, 1997), but its implementa- tion within elementary and secondary schools has been less than successful. Among other reasons, this may be due to the fact that debates about multicul- tural education have focused on sorting out desirable social and political goals, while neglecting to examine the nature and constraints of children's and adoles- cents' understanding and acceptance of diversity and dissent. Consequently, the task of educating children and adolescents to appreciate the scope of variation in beliefs and practices has often been reduced to ei- ther exposing children to the customs of different cul- tural groups, or directly inculcating the need for mutual tolerance. If, however, as suggested by this research, the appreciation and acceptance of personal and cultural disagreements over matters of value and belief involve processes of reflection and interpreta- tion, including drawing distinctions among different dimensions of tolerance and among different types of dissenting beliefs and the contexts in which such beliefs are embedded, then prescriptions for educa- tion that do not account for such processes of inter- pretation and conceptual distinctions are not likely to succeed. This study provides a foundation for considering some of the psychological constraints operating on children's and adolescents' thinking about cultural and personal diversity of belief and conduct.
This research was supported by a Spencer Postdoc- toral Fellowship from the National Academy of Edu- cation to the first author. The authors wish to ac- knowledge the cooperation of the Salt Lake City School District and, in particular, the principals, teachers, and students in the following schools: Em- erson Elementary, Franklin Elementary, Meadowlark Elementary, and Bryant Intermediate.
Adorno, T., Frenkel-Brunswick, E., Levinson, D., & San
ford, N. (1950). The authoritarian personality. New York:
Harper. Asch, S. E. (1952). Social psychology. Englewood Cliffs, NJ: Prentice-Hall.
Avery, P. G. (1988). Political tolerance among adolescents.
Theory and Research in Social Education, 16, 183-201.
Banks, J. A. (1992). Multicultural education: Approaches,
developments and dimensions. In J. Lynch, C. Modgil, & S. Modgil (Eds.), Cultural diversity and the schools. London: Falmer.
Baumrind, D. (1998). From "Ought" to "Is": A neo-Marxist perspective on the use and misuse of the culture con- struct. Human Development, 41, 145-165.
Bishop, J. C. (1989). Cultural relativism and the virtue of tolerance. In D. M. Topping, D. C. Crowell, & V. N. Ko- bayashi (Eds.), Thinking across cultures. Hillsdale, NJ: Erl- baum.
Bumiller, E. (1990).May you be the mother of a hundred sons: A journey among the women of India. New York: Ballantine.
Davidson, P., Turiel, E., & Black, A. (1983). The effect of stimulus familiarity on the use of criteria and justifica- tions in children's social reasoning. British Journal of De- velopmental Psychology, 1, 49-65.
Davis, J. A. (1975). Communism, conformity, cohorts, & categories: American tolerance in 1954 and 1972-73. American Journal of Sociology, 81, 491-513.
Dennis,J., Lindberg, L., McCrone, D., & Stiefbold, R. (1969). Political socialization to democratic orientations in four Western systems. Comparative Political Studies, 1,71-101.
Enright, R. D., & Lapsley, D. K. (1981). Judging others who hold opposite beliefs: The development of belief-discrepancy reasoning. Child Development, 52,1053-1063.
Enright, R. D., Lapsley, D. K., Franklin, C. C., & Steuck, K. (1984). Longitudinal and cross-cultural validation of the belief-discrepancy reasoning construct. Developmental Psychology, 20, 143-149.
Goodwin, J. (1994). Price of honor: Muslim women lift the veil of silence on the Islamic world. New York: Penguin. Hatch, E. (1983). Culture and morality: The relativity of values in anthropology. New York: Columbia University Press.
Helwig, C. C. (1995). Adolescents' and young adults' con- ceptions of civil liberties: Freedom of speech and reli- gion. Child Development, 66, 152-166.
Helwig, C. C. (1997). The role of agent and social context in judgments of freedom of speech and religion. Child Development, 68, 484-495.
Herdt, G. (1990). Sambia nosebleeding rites and male prox- imity to women. In J. W. Stigler, R. A. Shweder, & G. Herdt (Eds.), Cultural psychology: Essays on comparative human development. Cambridge: Cambridge University Press.
Jones, R. S. (1980). Democratic values and preadult virtues: Tolerance, knowledge, and participation. Youth and Soci- ety, 12, 189-220.
Lawrence, D. (1976). Procedural norms and tolerance: A re- assessment. American Political Science Review, 70,80-100. Lynch, J. (1986). Multicultural education: Principles and prac- tice. London: Routledge. McClosky, H. (1964). Consensus and ideology in American politics. American Political Science Review, 58, 361-382.
McClosky, H., & Brill, A. (1983). Dimensions of tolerance: What Americans believe about civil liberties. New York: Russell Sage Foundation.
Mill, J. S. (1956). On liberty. Indianapolis: Bobbs-Merrill.
Modgil, S., Verma, G. K., Mallick, K., & Modgil, C. (1986). Multicultural education: The interminable debate. London: Falmer.
Wainryb, Shaw, and Maianu 1555
Nieto, S. (1996). Afirming diversity: The sociopolitical context of multicultural education (2d ed.). White Plains, NY: Longman.
Nunn, C. Z., Crockett, H. J., & Williams, J. A. (1978).Tolerance for nonconformity. San Francisco: Jossey-Bass. Owen, D., & Dennis, J. (1987). Preadult development of po- litical tolerance. Political Psychology, 8, 547-561.
Patterson, J. W. (1979). Moral development and political thinking: The case of freedom of speech. Western Political Quarterly, 32, 7-20.
Prothro, J. W., & Grigg, C. W. (1960). Fundamental princi- ples of democracy: Bases of agreement and disagree- ment. lournal of Politics, 22, 276-294.
Rokeach, M. (1960). The open and closed mind. New York: Basic.
Sigelman, C. K., & Toebben, J. L. (1992). Tolerant reactions to advocates of disagreeable ideas in childhood and ado- lescence. Merrill-Palmer Quarterly, 38, 542-557.
Stouffer, S. (1955). Communism, conformity, and civil liberties. New York: Doubleday.
Turiel, E. (1983).The development of social knowledge: Morality and convention. Cambridge: Cambridge University Press.
Turiel, E., Hildebrandt, C., & Wainryb, C. (1991). Judging social issues: Difficulties, inconsistencies, and consis- tencies. Monographs of the Society for Research in Child De- velopment, 56(2, Serial No. 224).
Turiel, E., Killen, M., & Helwig, C. C. (1987). Morality: Its structure, functions, and vagaries. In J. Kagan & S. Lamb (Eds.), The emergence of morality in young children (pp. 155-244). Chicago: University of Chicago Press.
Vogt, W. P. (1997). Tolerance and education: Learning to live with diversity and dlfierence. Thousand Oaks, CA: Sage.
Wainryb, C. (1991). Understanding differences in moral judgments: The role of informational assumptions. Child Development, 62, 840-851.
Wainryb, C. (1993). The application of moral judgments to other cultures: Relativism and universality. Child Devel- opment, 64, 924-933.
Wainryb, C., & Ford, S. (1998). Young children's evalua- tions of acts based on beliefs different from their own. Merrill-Palmer Quarterly, 44, 484-503.
Williams, B. (1972). Morality: An introduction to ethics. New York: Harper & Row. Zellman, G. L., & Sears, D. 0. (1971). Childhood origins of tolerance for dissent. lournal of Social Issues, 27,109-136.