UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Voting with care : women, men and moral reasoning : does difference make a difference? Bancroft, Wendy Ruth 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1994-0020.pdf [ 2.97MB ]
JSON: 831-1.0087310.json
JSON-LD: 831-1.0087310-ld.json
RDF/XML (Pretty): 831-1.0087310-rdf.xml
RDF/JSON: 831-1.0087310-rdf.json
Turtle: 831-1.0087310-turtle.txt
N-Triples: 831-1.0087310-rdf-ntriples.txt
Original Record: 831-1.0087310-source.json
Full Text

Full Text

VOTING WITH CARE:WOMEN, MEN AND MORAL REASONING - DOES THE DIFFERENCE MAKEA DIFFERENCE?byWENDY RUTH BANCROFT(Honours) B.A., The University of British Columbia, 1992A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR A MASTER’S DEGREE INTHE FACULTY OF ARTS(Department of Political Science)We accept this thesis as conformingto the required standardTHE UNiVERSITY OF BRITISH COLUMBIANovember 1993@ Wendy Ruth Bancroft, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of__________________________The University of British ColumbiaVancouver, CanadaDate____DE-6 (2/88)ilABSTRACTThe following thesis provides a theoretical and empirical treatment of theargument proposed in 1982 by psychologist, Carol Gilligan, that women and menemploy different orientations in their moral reasoning. Gilligan says men see moralconflict as a struggle between competing rights, while women see moral conflict as astruggle between competing responsibilities. Where the perspective associated withmen arises out of a valuing of individual achievement and sees fairness as equality,women value relational connections and see fairness as a response to need. Malesprionze universal principles; females pnonze attention to context.Beyond its assertion of difference, Gilligan’s argument challenges the privilegegiven to notions of justice espoused by liberal theorists such as John Rawls and JohnStuart Mill, and reflected in Lawrence Kohlberg’s stage sequence model of moraldevelopment. While Gilligan does not deny the value of this justice orientation, sheargues that it was formed in the absence of women and that as it stands, it lacks thenecessary, and equal, elements of compassion and connection that reflect an ethic ofcare.The argument has sparked philosophical and empirical debate across severalacademic fields. This thesis pays attention to that debate as well as contributing anempirical test of the hypothesis that women are more caring than men, in the contextof voting behaviour. One hundred and ninety-one students at the University ofBritish Columbia took part in an experimental survey in which the hypothesis wastested in two conditions: 1) subjects were asked to base a vote for either Candidate Xor Y in the absence of defining criteria other than electoral poll popularity ratings,and 2) with the addition of candidate issue positions on social welfare policy. Theexpectation was that while both males and females were subject to the socialinfluence provided by the opinion poii results, women, motivated by a careorientation, would be more likely to choose the underdog candidate than men. This111did not prove to be the case. Not only did more women bandwagon than men incondition 1, but in condition 2 where candidates were clearly associated with careversus rights positions, no sex differences emerged. Discussion of these findingsaddresses the impact of the political venue on moral orientation, while the conclusionfocuses on the implications of moral difference for women’s political behaviour andmodem society.ivTABLE OF CONTENTSAbstract iiTable of Contents ivList of Tables vList of Figures viAcknowledgement viiINTRODUCTION 1Chapter One The Ethic of Care 6Chapter Two The Difference Debate 24Observations in the Field 24Observations in the Laboratory 28Chapter Three Voting with Care -- The Study 48Design 52Method 57Total Sample Findings 61Cross-Cultural Findings 66Summary of Results 69Discussion 70Chapter Four Adding Care to the Moral Discourse --Implications for Women and Society 80Bibliography 91Appendix A Kohlberg’s Six Stage Model 99Appendix B Freidman et al: 12 Item Types 100Appendix C The Questionnaire 101VLIST OF TABLESTable 1: Vote for X or Y in Condition 1, by Sex 61Table 2: Condition 1--Why Vote for X, by Sex 63Table 3: Condition 1--Why Vote for Y, by Sex 64Table 4: Vote for X or Y in Condition 2, by Sex 65Table 5: Vote for X or Y in Condition 1 -- Cross Cultural Comparison,bySex 68Table 6: Voting for X or Y in Condition 2-- Cross Cultural Comparison,bySex 68LIST OF FIGURESFigure 1: Kohlberg’s Six Stage Moral Development Model 13Figure 2: The Movement of Votes from Condition ito Condition 2, by Sex 66Figure 3: Men, Women and Political Efficacy 78viviiACKNOWLEDGEMENTThere are a number of people who have contributed to the completion of thisthesis and whom I would like to thank. First must be my husband, Jay, who hassupported me not only financially but with love, time, effort, and constantencouragement throughout my entire six years of “going back to school.” Mychildren, Aaron and Leah, have also been tenific--Leah for never once complainingthat her mommy was, yet again, busy doing school work; and Aaron, who is about tograduate from university himself, for his pleasure in my successes and for beingobviously pleased (rather than embarrassed) to run into his mother on campus. Andthank you Karen, my other child/friend, for your enthusiastic interest and respect formy ideas.I also want to thank Dr. Avigail Eisenberg for her support as a thesis adviserand for conveying a genuine interest in this subject, and finally, Dr. Jean A.Laponce, my thesis adviser and mentor, for his support and guidance over this thesisperiod, and for opening intellectual doors for me over the past several years.1Introduction“I had this naive view that the first woman leader was going to be morevirtuous and more caring, the fantasy of the perfect mother. In the realworld, it doesn’t play out like that.” (Wallace et al. 1993, 19)Such were the sentiments expressed by Toronto novelist Susan Swan, in aMaclean’s article about then Prime Minister Kim Campbell. The sentences containthree pieces of information which have enormous relevance in the study of women’spolitical behaviour: 1) Canada has had its first female prime minister, 2) because oftheir maternal role in society, different moral values are expected from women thanare expected from men, and 3) the latter expectation may be more myth than reality.While the first fact may be significant, it is the latter two that hold the attention ofthis thesis. Myth or not, we do generally think of women as more nurturing than menand we expect that their political decisions will reflect this sentiment. But is this thecase? Kim Campbell aside, are women any more caring in their approaches topolitical decision-making than men?Those who study political behaviour refer to this difference between men andwomen as a “gender gap,” and they claim evidence of this gap exists in voting studiesundertaken since the late 1 970s. While by no means a homogeneous group, femalerespondents in these surveys reflect an overall humanitarian ethos known as “agape”- a virtue associated with charity, nurturance, love, and sense of community(Kopinak, 1987:26; Kohlberg, 1984:227). Women are more supportive than men ofgovernment efforts on behalf of disadvantaged groups, of tougher environmentalprotection legislation, of more generous foreign aid policies, and especially of strongpeace initiatives.’ Economically speaking, they are “sociotropic”--motivated more byIn a previous analysis of sex differences, using data from the 1988 National Election Survey, I comparedmale and female responses to the question, “All things considered, do you support or oppose buying nuclearsubmarines?” Coding responses to reflect either “oppose” or “support” I found that where 5 5.4% of maleswere opposed to such a purchase, 67.6% of female respondents indicated opposition, a 12.2% difference.(Reference from Bancroft, W. “Tracking the Gap: Women, Men and the Conservative Vote in 1988”Unpublished essay).Voting with Care 2considerations of social good rather than personal gain. (Weanng and Wearing,1991; Norris, 1988; Burt, 1986; Kopinak, 1987; Miller, 1988). Men, on the otherhand, tend to give more support to deficit reduction initiatives, to free enterprisepolicies, and to “tougher” foreign policy stances. Canadian political scientist,Thelma McCormack, suggests that perhaps men and women belong to differentpolitical cultures, and like any subculture, “the political culture of women has its ownway of understanding the world” (1975, 25). Tn some cases, even when the moraljudgment of men and women appears the same, the reasoning that led to the decisioncan be quite different, as the following case illustrates.Stimulated by McCormack’s ideas and by their own interest in the peacemovement, Froese and Nielsen carried out a study in 1984 in which undergraduatestudents were asked whether they agreed or disagreed with the peace movement andwhy they held this belief. While they found no sex differences in support for thepeace movement and nuclear disarmament--the majority of men and women agreedwith the peace movement and disagreed with nuclear arms proliferation--their reasonsfor doing so differed. Women’s reasons tended to be couched in “moral or ethical”terms such as the loss of human life; men tended to emphasize the “pragmatic issuessurrounding the disarmament process” for instance, that disarmament be bilateral andverifiable (132). When asked if they thought gender was related to political belief,most respondents thought it was not, but those who did offered reasons that reflectedthe sex stereotypes previously mentioned. One man said “I think that while men andwomen share a will for a common political end, women sometimes fail to appreciateor understand the mechanics of politics, while men are sometimes obsessed withthose mechanics” (133).According to psychologist, Carol Gilligan, what we see here is evidence thatwhen men and women think about moral issues they way they think about andrespond to these issues is quite different. While, for instance,both sexes may beVoting with Care 3aware of the actual process and implications of disarmament, women see that processin terms of lives saved; men emphasize the relative positions of the players involved.In Gilligan’s terminology, an “ethic of care,” resulting from a strong sense ofconnection to others, informs the moral judgments of women, while an “ethic ofrights” reflecting a notion of justice as equality informs the moral judgment of men(1982).In one respect, Gilligan is not saying anything new. Rather, she appears toreinforce long held behavioural stereotypes: women are nurturers, men are achievers.However, Gilligan is not only saying differences exist, she is challenging notions ofmoral development which reflect western philosophical values which subordinatewomen’s perspective, relegating it as appropriate only to the private, domestic andpersonal sphere, while pnorizing the male perspective as the ideal, and the one mostappropriate for the public world of business and politics.For those who study women’s political behaviour, the question of whether menand women use different moral reasoning when making their political decisions ishighly significant, particularly when one considers that the liberal democraticparadigm on which our political system is based is the product of male philosopherswith male experiences and male values. If a fundamental part of the way womenthink about the world is not reflected in the processes, policies, and institutions thatconstitute our political system, therein may lay much of the explanation for the factthat women continue to lag behind men in political interest, knowledge, and efficacy(Vickers and Brodie 1981; Kay et al. 1987).While the focus of this paper is political behaviour, interest in Gifligan’sargument extends far beyond political science into the disciplines of law, psychology,biology, sociology and philosophy. In all of these disciplines, the notion of women’scognitive-moral exclusion has ignited debates--most of it theoretical andphilosophical in nature. However, in the field of social psychology, the debateVoting with Care 4canied on by Gilligan’s proponents and opponents has centred on empiricalchallenges, many of which will be presented in this paper.This paper is a journey through many of the ideas and evidence that bothinform and challenge Gilligan’s thesis, a journey that ends with our owti empiricaltest of the argument. Chapter one lays out not only Gilligan’s thesis, but thetheoretical positions of the major players in this particular moral dilemma--inparticular the moral developmental models and theories of those Gilligan ischallenging--Lawrence Kohlberg, Jean Piaget, and Sigmund Freud--as well as theideas which inform Gilligan’s argument, in particular Nancy Chowdorow’s views onthe impact of the mother/child relationship on male/female perceptions of separationand connection. Chapter two takes us into the empirical debate mentioned earlier,including not only studies which investigate the existence of sex difference per Se,but those which focus specifically on manifestations of justice and care moralorientations. Chapter three is devoted to a treatment of the study undertaken for thisthesis, and which tests the hypothesis that women are more caring than men, anddoes so in the political context of voting behaviour. Following a discussion of theresults of this study, the thesis concludes in chapter four with an exploration not onlyof what an exclusion of this perspective in formal politics means for women’spolitical behaviour, but of the possible benefits for society should the voice beincluded.One final word before our journey begins. One evening, several months ago,my seven year old daughter and I were laying in bed reading together--she, achildren’s book I, some research material for this thesis. At one point my daughterasked me, “What is that you are reading?” So, I said, “Well, it’s for my thesis. Thisperson is talking about an argument made by a woman named Carol Gilligan thatmen and women think differently when they make decisions about some things.”And I went on to explain, in what I thought was very clear language, what my thesisVoting with Care 5was about. At the end of my explanation I said, “So, did that make sense?” Mydaughter said “no.” So, I made another attempt and this time really tried to make themeaning clear. At the end of this explanation I asked again, “Now, does it makesense?” My daughter said, “no.” So I thought Pd better find out just how much shedid understand before attempting this again. I said, “Well, what do you think I said?She said, “Blah blah blah blah blah blah Gilligan, blali blah blah blah blah. ...“ Isincerely hope that the reader is more informed by this thesis than was my daughter.6Chapter 1: The Ethic of Care“I deny that any one knows, or can know, the nature of the two sexes,as long as they have only been seen in their present relation to oneanother. If men had ever been found in society without women, orwomen without men, or if there had been a society of men and womenin which the women were not under the control of the men, somethingmight have been positively known about the mental and moraldifferences which may be inherent in the nature of each.” (Mill, J. S.1869. “The Subjection of Women.”)Over a century after Mill framed these sentiments, developmentalpsychologist Carol Gilligan expressed the same frustration. However, unlike Mill,whose opinions were framed in a time when psychological inquiry occurred in amostly ad hoc fashion, Gilligan was reacting to an extensive body of psychologicalstudy offered by the twentieth century’s three most renowned developmentaltheorists: Sigmund Freud, Jean Piaget, and Lawrence Kohlberg. All three hadpresented models of human cognitive and moral development in which women werefound lacking, and all based their studies on exclusively male subjects. In the early1970s, Gilligan had even co-authored one of these studies with LawrenceKohlberg.2 However, as she wrote in 1982:Over the past ten years, I have been listening to people talking aboutmorality and about themselves. Halfway through that time, I began tohear a distinction in these voices, two ways of speaking about moralproblems, two modes of describing the relationship between self andother (1).Gilligan challenged accepted paradigms of moral reasoning, especially that ofLawrence Kohlberg, and began her own investigations into moral thought, this timefocussing her attention on women. In Gilligan’s words, “Only when life-cycletheorists divide their attention and begin to live with women as they have lived withmen will their vision encompass the experience of both sexes and their theories2 In 1971, Gilligan and Kohlberg jointly published findings on a study of “The Adolescent as a Philosopher:The Discovery of the Self in a Post-conventional World.” Daedalus 100, 1051-1086.Voting with Care 7become correspondingly more fertile” (1982:23). Her research has led her toargue that because of differing developmental experiences, males and femalesdevelop different moral orientations. When it comes to moral decision-making,women see morality in terms of care and connection to others, where men seemorality as a question of fairness, or competing rights. While the male orientationforms the core of the dominant moral paradigm in western society, women’s moralorientation has been historically relegated to the personal sphere.The masculine emphasis in current developmental theory found its origins inthe theories of psychosexual development proposed by Freud in his Three Essayson the Theory of Sexuality, written in 1905. Freud theorized that the crucialdivergence in personality development between males and females occurred atpuberty, the time at which boys, because of castration fear, repress their libidinalattraction to their mother and transfer their identification to their father (1905, 93).This process, this mastering of what Freud referred to as the “Oedipus complex,”was mandatory for personal development. In a footnote added in 1920, Freudwrote: “Every new anival on this planet is faced by the task of mastering theOedipus complex; anyone who fails to do so falls a victim to neurosis” (92). Girlsdid not suffer castration anxiety and therefore, they never mastered this complex,and because of this, they did not develop the “superego,” or conscience, necessaryfor moral development. Nor did they experience the necessary “detachment fromparental authority” that young men experienced.At every stage in the course of development through which allhuman beings ought by rights to pass, a certain number are held baclçso there are some who have never got over their parents’ authority andhave withdrawn their affection from them either very incompletely ornot at all. They are mostly girls .In fact, Freud himself admitted he had very little understanding of femalebehaviour; in 1926 he still referred to women’s sexual development as a “darkVoting with Care 8continent” (Williams, 1987:34). Nevertheless, his belief in the link betweencastration fear and the development of the superego caused him to conclude in 1925that women “show less sense ofjustice than men, that they are less ready to submitto the great exigencies of life, that they are more often influenced in their judgmentsby feelings of affection or hostility” (Gilligan 1982, 7).One might expect that as women’s lives have evolved to a more independentstatus, and as some of Freud’s other ideas have diminished in efficacy, women’smoral status might enjoy a more elevated position within moral paradigms.However, while women are no longer thought to be ruled by their biological“shortcomings,” their moral reasoning is still seen to be generally inferior to that ofmen.When Jean Piaget pursued the theme of justice in children’s play in 1932, hetook no notice of biological factors in personality development beyond noticing thatgirls and boys played different games, and that that made a difference in their moraldevelopment. Believing that morality displayed by children throws light onunderstanding adult morality, and further that “all morality consists in a system ofrules,” Piaget believed that he could gain insight into an understanding of moraldevelopment by studying the thinking and behaviour of children playing marbles--agame involving an intricate set of rules. After watching many games played byboys of varying age groups, and talking with the players to detennine their attitudestoward the imposition and interpretation of the rules of the game, Piaget developeda dynamic model of moral development based on rule practice. Acccording toPiaget, progress in moral development is marked by an individual’s movementthrough four stages of rule orientation beginning from a position of egocentrism,where obedience is based on either fear of punishment or unquestioning acceptanceof rules as authority, and ultimately moving to an autonomous approach to ruleswhere laws become viewed as the product of mutual consent, and to be changedVoting with Care 9only through mutual consent. Piaget believes that the greater the knowledge of therules involved, the more the respect for those rules becomes based on rationalconsiderations rather than mystical acceptance (1965:28). At the highest stage ofmoral development, the code has been mastered, juridical discussions of principleand procedure become pleasurable in themselves, and the individual is able to“apply them to any case whatsoever, including purely hypothetical cases ...“ (47).At this level, cooperation and reciprocity are assured by a rational subordination ofthe individual to universally agreed upon laws.Girls do not play marble games and do not figure in Piaget’s moraldevelopment schema. He felt that girls’ games, such as hopscotch, never involvedthe complex codifications, nor complicated jurisprudence one sees in boy’scollective games, and because of this, girls did not attach the same importance torules as did boys (77-80). For Piaget, this was a serious developmental constraint,causing him to conclude that “the most superficial observation is sufficient to showthat in the main the legal sense is far less developed in little girls than in boys” (77).When, in the 1970s, Lawrence Kohlberg offered his ideas about morality andjustice, the discourse shifted from rules to principles. Kohlberg offers a prescriptiveparadigm of moral/cognitive development based, not only on his own extensiveresearch, but on Piagetian notions of moral justice and stage sequence ofdevelopment, and on the political philosophies of justice offered by J.S. Mill andespecially by John Rawls. Employing a self-described “neo-Kantian” definition ofmorality based on a “categorical obligation to act” in a way that is “universalizable,a point of view which any human being could or should adopt in reaction to thedilemma” (1984a, 224), Kohlberg retains justice as the central tenet of moraldevelopment, but incorporates notions of justice beyond rule codification andjurisprudence. Like Piaget, Kohlberg also viewed moral progress as a stagesequence, but Kohlberg saw this progress as an evolution toward a cognitive andVoting with Care 10behavioural ideal in which the morally autonomous individual at the highest levelcomprehends not only legal complexity, but seeks a Platonic type of ultimate truth.In KoMbergs paradigm, notions of justice as fairness come from Rawis; notions ofautonomous development are influenced by Mill:If, with equal virtue, one is superior to the other in knowledge andintelligence-- or if, with equal intelligence, one excels the other invirtue -- the opinion, the judgment, of the higher moral or intellectualbeing is worth more than that of the inferior (Mill 1861, 307).In KoMbergs model there are three major levels of moral reasoning: thepreconventional, conventional, and postconventional (see Figure 1 and AppendixA). At the preconventional level, the person bases his or her moral judgment onegocentric considerations of fairness--”what is right for me,” and exhibits morallyright behaviour primarily as a response to externally imposed sanctions. At theconventional level, the person looks beyond self to the needs of society. What isfair is what society deems to be fair; doing the morally right thing at this level ismotivated by a desire to be seen to be good by others, and out of genuine concernfor the welfare of others, often putting that welfare before ones own. At thepostconventional level, the person once again considers what is morally right fromhis or her own perspective, but this time fairness is determined by principles ofequality and reciprocity, and are grounded in a notion of what constitutes a justsociety (Kohlberg, 1976:32-36). Within each of these three levels are two stages ofmoral reasoning, with the second stage being a more advanced form of theperspective taken in the first. As in Piaget’s model, the individual moves fromegocentric considerations to ones that consider the social good, but in Kohlberg’sstage sequence, the emphasis shifts from rationally based cooperation to anawareness of rights and universal principles, seen in its most developed and morallyVotingth Care 11autonomous interpretation at stage 6, the highest stage of moral reasoning (seeFigure 1 below).Figure 1: Kohlberg’s Six Stage Moral Developmental ModelPRECONVENTIONAL CONVENTIONAL LEVEL POSTCONVENTIONALLEVEL LEVELno rights others have interpersonal societal needs objective rights universalconsidered rights but they perspective-- take take ethicalbeyond own are relative primacy of precedence precedence principles:others over self over any one over social equality ofperson’s attachments rights; respectand contracts for individualsIn fact, Kohlberg has not found empirical support for the existence of stage6; it is an ideal, a moral “ought.” Here we find J.S. Mill’s “being of higherfaculties,”a product of education, reflection and taste whose moral principles reflecta personal orientation where “the good of others becomes to him a thing naturallyand necessarily to be attended to, like any of the physical conditions of ourexistence (Mill 1861-3, 9,33). Here we also find the person able to make moraljudgments framed within a “veil of ignorance,” the device proposed by John Rawisto ensure that moral decisions are taken in a spirit of universalizability, notconditioned by personal circumstances or interest (Rawls 1985, 237).Kohlberg based his model on a longitudinal study of 84 male subjects askedto decide the morally right solution to a number of hypothetical dilemmas, with theHeinz Dilemma being the example most commonly referenced:“In Europe, a woman was near death from a rare disease. There wasone drug the doctors thought might save her. It was a drug that adruggist in the toi had recently discovered. The drug was expensiveto make, and the druggist was charging ten times what the drug hadcost him to make. He paid $200 for the materials and was chargingVoting with Care 12$2,000 for the prescription. The sick woman’s husband, Heinz, wentto everyone he knew to borrow the money, but he could only gettogether about $1,000, which was half of what he needed. He told thedruggist that his wife was dying and asked him to sell it cheaper or tolet him pay later. But the druggist said, ‘No, I discovered the drug andI’m going to make money from it” (Belenky et al. 1986, 236).Kohlberg, and others who have used his stage sequence and scoring system, rateresponses according to where they fall within his six stage sequence. The morally“best” responses are those which attempt to resolve a conflict of principles, in thiscase the right to property in conflict with the right to life.Those who have tested women as well as men have found that women do notfare well when it comes to their placement in Kohlberg’s moral continuum. Womenlook at this dilemma and want more contextual information, and their resolutionsaim for a harmony of interests rather than a priorization of principles. Most womenend up with a stage 3 scoring, defined as a stage which emphasizes “mutualinterpersonal expectations, relationships, and interpersonal conformity” (Kohlberg1976, 34-3 5).Gilligan (1982) does not deny that women find these values important; whatshe challenges is the inferior placement they receive in Kohlberg’s model, and sheblames this on research that has excluded women. Her o’wi conclusions reflect thefindings of in-depth interviews carried out with both mixed and all-female samplesin three major studies: the college student study, involving 25 randomly selectedstudents; the abortion decision study in which 29 women were interviewed aboutthe moral conflict they experienced and the decisions they reached when facing anabortion; and the rights and responsibilities study, involving 144 males and femalesmatched for a variety of demographic factors including age and education. Thecomments she heard have led her to conclude that what women’s responses reveal isnot a lesser morality, but a different one.Voting with Care 13When one begins with the study of women and derivesdevelopmental constructs from their lives, the outline of a moralconception different from that described by Freud, Piaget, orKohlberg begins to emerge and informs a different description ofdevelopment. In this conception, the moral problem arises fromconflicting responsibilities rather than from competing rights andrequires for its resolution a mode of thinking that is contextual andnarrative rather than formal and abstract (19).Gilligan distinguishes the two orientations as an “ethic of care” based on a“responsibility to discern and alleviate the ‘real and recognizable trouble’ of thisworld, and an “ethic of justice” based on “an injunction to respect the rights ofothers and thus to protect from interference the rights to life and self-fulfillment”(100). While the first is seen most by women, the latter by men, Gilligan isunwilling to attribute this to an actual gender difference, and notes that herassociation is based on empirical observation. She also states the scope of herconclusions as making no claim to the origins of this difference, nor to its“distribution in a wider population, across cultures, or through time.” However, asshe points out, these differences exist “in a social context where factors of socialstatus and power combine with reproductive biology to shape the experience ofmales and females and the relations between the sexes” (2). Central to her thinkingis the notion that justice and care perspectives arise out of differing experiences ofinequality and attachment. As children we all experience inequality in our relationswith our parents, which diminishes as we develop our own self-identity.Attachment and separation are also universally shared experiences. However, girlsdo not experience the same level of separation that boys do; their experiencesbecome grounded in their sense of connection to others. Boys experience higherlevels of separation in the process of identity formation, and then in theirrelationships with their father, a greater sense of inequality. These experiencesshape our awareness of ourselves and others: inequality leading to a perspective ofVoting with Care 14justice which privileges equality and fairness considerations, and attachment leadingto a moral perspective which privileges connection (1988, 114).Gilligan’s insights are heavily influenced by Nancy Chowdorow,Chowdorow’s insights by Freud. However, unlike Freud, Chowdorow findsdifferences emerging between boys and girls during the preoedipal experience.According to Chowdorow (1974), mothers treat their sons and daughters in adifferent manner from birth. Knowing that at some point a process of separationmust occur between themselves and their son if their son is to develop his ownmasculine identity, Chowdorow says mothers encourage their sons to seethemselves as different and separate from them much more than they do with theirdaughters. Boys not only experience greater differentiation but, forced to look totheir father as a model of masculine identity, they then find a father often physicallyand psychologically distant. Therefore, says Chowdorow, the boy’s male genderidentification becomes “positional” rather than “personal,” their masculinity definedby imagined criteria rather than by modelling through contact. Male identitybecomes defined by what is not feminine--a negatively defined identity based ondevaluing feminine characteristics. This is in contrast to the more “personal” andcontinuous identification girls have with the female role, a process which allowsthem to grow up feeling more connected to others, but less able to see themselves asindividuals in their own right. Citing a duality proposed by David Bakan in 1966,Chowdorow suggests that these differing mother-child experiences lead to sexdifferentiated world views in which males are preoccupied with “agency,” femaleswith “communion.” Agency sees the organism as an individual and manifests itselfin “self-protection, self-assertion, and self-expansion. Communion sees theindividual as part of a larger organism and manifests itself “in the sense of being atBakan, David. 1966. The Duality of Human Existence: Isolation and commumon in Western Man.Boston. Cited in Chowdorow (1974).Voting with Care 15one with other organisms” (5 5-6). While Chowdorow sees that, to some degree,these differences occur as a result of the differential treatment accorded to sons anddaughters by their mothers, she suggests that in western society, the differenceshave been exacerbated by the process of industrialization. The economicreorganization of the family, which saw the family income dependent on wagesearned primarily by men outside the home, meant boys spent far less time with theirfathers. As Chowdorow says, we are not simply “taught” our male and femaleidentities, but assimilate them through certain features of the social structure,supported by cultural beliefs, values and perceptions.Nowhere is this more dramatically illustrated than by Margaret Mead’sseminal study of three primitive societies in New Guinea. Between 1931 and 1933,Mead and her colleague, Dr. Fortune, lived with and studied the social behaviour ofthe Arapesh, Mundugumor and Tchambuli tribes, an experience she credits withopening her eyes to the role of social conditioning in human behaviour (1963, 279-282). Mead observed that in each tribe, while roles were organized around the factof biological sex difference, the social organization and expected sex-linkedtemperaments showed enormous variation. In the Arapesh, for instance, whilewomen carried out the everyday tasks of food production: the weeding, carrying,preparation, etc., and men oversaw ceremonial preparations, and the killing of pigsand growing of yams, both men and women were expected to care for the children.And, it is this latter occupation that is considered the tribes most important activity.As Mead describes it:Arapesh life is organized about this central plot of the way men andwomen, physiologically different and possessed of differing potencies,unite in a common adventure that is primarily maternal, cherishing,and oriented away from the self towards the needs of the nextgeneration. (15).Voting with Care 16Children are socialized to associate security and happiness with their tribalmembership, and no games are played that encourage aggressiveness andcompetition (57-62).The other two tribes, the Mundugumour and the Tchambuli, produced verydifferent behavioural norms. Where Arapesh personalities, male and female, werestandardized to conform with what could be called maternal values, theMundugumour encouraged opposite behaviour. Males and females were socializedto exhibit proud, harsh and violent behaviour--more typically described asmasculine. Here, children were brutally weaned, pushed from their mothers with“blows and cross words” (198). Children’s play, while unorganized, was highlycompetitive, as were interpersonal relations overall. Again, there were instances ofdifferential treatment accorded males and females, but according to Mead:behind this difference in the treatment of boys and girls lies notheory that women differ temperamentally from men. They arebelieved to be just as violent, just as aggressive, just as jealous. Theysimply are not quite as strong physically, although often a woman canput up a very good fight, and a husband who wishes to beat his wifetakes care to arm himself with a crocodile-jaw and to be sure that sheis not armed (210).In the Tchambuli, the characteristics we associate with males and femaleswere reversed. Here the women took charge of all domestic and business affairs,while the men concerned themselves with aesthetic activities of both a vain andartistic nature.But, while the level of sex difference varied as a correlate of cultural norms,certain patterns appear familiar to Gilligan’s (and Chowdoros) argument. Forinstance, in the Arapesh, a society which values nonviolence and nurturance, whilethe young boy experiences separation when his father must leave him to go hunting,it is not an experience analagous to males in western society because first of all, theVoting with Care 17Arapesh boy is not expected to develop a specific masculine identity that requires astrong contrast with the female identity and second, as soon as the boy is oldenough to go hunting, he regains the close contact previously enjoyed with hisfather. Because fathers and mothers are equally and positively involved inchildrearing, both boys and girls grow up with a sense of connection. In Gilhigan’sterms, both males and females in the Arapesh would likely make moral judgmentsthat reflect an ethic of care.While much of Gilhigan’s own research emphasizes in-depth interviews withwomen about real-life moral dilemmas, she has also studied groups involvingsubjects of both sexes and of varying ages, in which she has administered Kolberg’shypothetical dilemmas and used his scoring system. From this, she offerscontrasting perspectives on the Heinz Dilemma given by Amy and Jake, two elevenyear olds in one of her studies. After being presented with the dilemma, Amy andJake are presented with the moral question, “Should Heinz steal the drug?” Jake’ssolution, scored at a level falling between stages 3 and 4 is typical of a “justice’oriented response. For Jake the conflict is between the values of property and life;the dilemma for him is to choose between those two rights: the right of the druggistto sell his product at his chosen price vs the value of a human life. In Jake’s words,“For one thing, a human life is worth more than money, and if the druggist onlymakes $1,000, he is still going to live, but if Heinz doesn’t steal the drug, his wife isgoing to die”. His approach to moral reasoning is to look at the dilemma as “ sortof like a math problem with humans,” a problem capable of solving as long as youuse the right logic (26). Amy totally misses this concept of logic. Not only doesshe not see this problem as a conflict of rights, her attention becomes focused oncoming up with alternative solutions to the dilemma. She worries, for instance,about the impact that Heinz’s theft might have on the ongoing relationship betweenhim andhis wife:Voting with Care 18If he stole the drug, he might save his wife then, but if he did, he mighthave to go to jail, and then his wife might get sicker again, and he couldn’tget more of the drug, and it might not be good. So they should really justtalk it out and find some other way to make the money (28).In the end, Amy decides the best solution is to make the druggist understand thepossible dire consequences of his refusal to sell the drug and, failing that, to haveHeinz appeal to others not in the original dilemma, who might be in a betterfinancial position to help him. It is a very pragmatic solution which tries to find themost universally harmonious way out of the dilemma, but which ignores anyprinciples at stake.We see in Amy’s response a concern for the ‘real’ situation, and a desire forinformation obviously missing from this hypothetical description. This frustrationwas a common reaction from women working through Kohlberg’s hypotheticaldilemmas and because of this, Gilligan felt that discovering the true discourse ofwomen’s moral thought would require providing opportunities for women to speakof ways in which they had resolved real-life dilemmas. In the Abortion DecisionStudy, she interviewed twenty-nine women, ranging in age from fifteen to thirty-three, and coming from a variety of socio-econoniic backgrounds. While theirdilemmas were of a personal nature, the choices they made affected not onlythemselves but others, and contained consequences that could influence whetherthey thought of themselves as a morally right or ong person. Three moralperspectives emerged from these interviews, and provide the stages for Gilligan’sprescriptive model of moral development, centred on an orientation of care, thatdefines women.Survival, responsibility and interconnection mark the three levels of women’smoral development. An initial concern for self in order to survive passes through atransition phase when the woman begins to see this self-absorption as selfish, andVoting with Care 19begins to see herself more in connection with others in a relation of responsibility.At this stage, being good is synonymous with taking care of others - only throughselfless giving can self worth be attained. This is the stage where many womenremain. While they may do things for themselves, such actions provoke feelings ofguilt that are only assuaged if the action can be seen to be also benefiting others.Gilligan sees this as a crippling stage, unhealthy for the woman and for the others inher life. As Jean Baker Miller points out:As wives, mothers, daughters, lovers, or workers, women often feelthat other people are demanding too much of them; and they resent it.Frequently they cannot even allow themselves to admit that theyresent these excess pressures. They have come to believe that theyshould want to respond at all times and in all ways. Consequently,they cannot let themselves openly call a halt to the demands or eventake small steps to limit them. The hesitation to do this, to resistcontrol of their own lives in even ordinary ways, can result in manypsychological complications or even somatic symptoms (1976, 50).Gilligan would have these women move beyond seeing self-fulfillment asselfishness to a point where they can view their relationships with others asopportunities for mutual give and take, where the interests of self are alsoconsidered legitimate. This involves another transitional stage wherein goodnessbegins to translate into “truth” as the woman realizes that being morally goodinvolves care of self as well as others. From here it is only a short cognitive leap toGilligan’s third moral development stage, premised on a morality of nonviolencebetween self and other: “A moral equality between self and other is achieved byequally applying an injunction against hurting” (Brabeck 1993, 36). At this third,posteonventional, moral level, we see some of the characteristics of KoMbergsautonomous reasoner, approaching moral judgment from the perspective of self butin full knowledge of the position of others. It is the thinking manifested byBelenky et al.’s “constructivist woman.”Voting with Care 20Belenky et al. were interested in discovering whether women not onlyemployed a different moral orientation from men, but whether their ways oflearning were similarly different. Using questions employed in assigning WilliamPen-y’s learning development positions, as well as standard questions developed byGilligan and Kohlberg, they earned out intensive in-depth interviews with 135women. From these responses, they were able to outline five epistemologicalcategories, and we are able to see parallels between learning stage and moralreasoning employed. For instance, “received knowers” do not question authorities,be it husband or professor. Having no faith in their own intellectual abilities, theybelieve others must be more informed than they. Their moral judgments are alsobased on what others consider right. As Belenky et al. write, a persistent theme ofthese women is that “they should devote themselves to the care and empowermentof others while remaining ‘selfless,” (1986, 46) a characteristic of women inGilligan’s second, “responsibility” stage of moral development. This contrasts withBelenky et al.’s “constructivist woman” who approaches moral conflict from aninformed but compassionate position.Generally well educated, constructivists are aware of objective truths andforms of logical reasoning, but they also see that “all knowledge is a constructionand that truth is a matter of the context in which it is embedded.” Kohlberg wouldargue there is very little difference between this realization and his own higher stagereasoning where the person is aware “that most values and rules are relative to yourgroup” (1976, 34-3 5). Kohlberg has, in fact, argued that at higher education andoccupational levels, differences between men and women disappear, an argumentthat will be explored in the next chapter. But, Belenky et al. and Gilligan wouldargue that differences persist even at the level of higher order moral reasoning. Forinstance, Kohlberg’s stage 5 reasoner may recognize relative truth, but moraljudgment is based on rules arising out of an assumed social contract among equals,Voting with Care 21and moral conflict is seen as existing between abstract principles that can beobjectively solved. This contrasts with the approaches taken by the women inGilligan and Belenky et al.’s studies who tend to place much more emphasis on thecontext of the situation; and on the need to resolve inequitous relationships, and ona recognition of differences in need. While Belenky et al.’s “constructivists” arecapable of objective truth, their moral reasoning contains strong elements of a senseof connection to the players involved, resisting “premature generalization aboutwhat they would do or what should be done, particularly about matters of right andwrong” (149). Asked whether Heinz should steal the medicine, these womenwanted to know: “What does Heinz wish?’ ‘What is the condition of Mrs. Heinz’slife?’ ‘Why is the druggist behaving so?’ Does Heinz have children dependent onhim for care?’ ‘Who would care for the children if Heinz went to jail?” (149). LikeKohlberg’s higher order moral reasoners, women in Gilligan’s highest level of moraldevelopment also incorporate a notion of rights into their moral judgment, but it is anotion tempered by an awareness of the complexity of most moral dilemmas, and byan awareness that in many cases, the players are not equal.In 1984, Kohlberg revised his scoring system to include a care perspective atevery stage of reasoning, admitting that his concept of justice “does not fully reflectall that is recognized as being part of the moral domain” (1984, 227). However,Kohlberg sees “care” and ‘Justice” perspectives as two moral orientations with twodomains of application:From our point of view there are two senses of the word moral, andtwo types of dilemmas, each corresponding to these differingmeanings of the word... The ‘moral point of view’ stresses attributesof impartiality, universalizability, and the effort and willingness tocome to agreement or consensus with other human beings in generalabout what is right. It is this notion of a ‘moral point of view’ which ismost clearly embodied psychologically in the Kohlberg stage model ofjustice reasoning.Voting with Care 22There is a second sense of the word moral, which is captured byGilligan’s (1982) focus upon the elements of caring and responsibility,most vividly evident in relations of special obligation to family andfriends.The difference ... is captured by the distinctions that manyAmericans make between the sphere of personal moral dilemmas andchoices and the sphere of moral choice that is not considered personal,that is, the sphere captured by our justice dilemmas (230).Kohlberg has assigned morality as care to the personal, private sphere and moralityas rights to the public sphere.Gilligan says he is wrong, that women employ moral judgments that priorizea care perspective in their considerations of “public” moral dilemmas as well as inconsiderations of dilemmas closer to home. It is a debate that has stimulated aresponse from academics in law, psychology, sociology, biology, philosophy andpolitical science. While some have joined Gilligan in challenging Kohlberg’sparadigm, others point to the lack of empirical support existing for Gilligan’s claimof male/female difference in moral orientation. To some extent, the debate alsohinges on the necessity of social scientific experimentation to prove what seemsobvious to our eyes. For instance, in a review of In a Different Voice, offered bypsychologists Anne Colby and William Damon (1987), these authors criticizeGilligan for relying on anecdotal evidence, for failing to present empirical data tosupport her interview studies, and for failing to provide comparative data on malesto match the data on females from her abortion study. Nevertheless, these sameauthors allow that there is an intuitive appeal to Gilligan’s claim, and that certainlythe stereotypes which present males as more independent and objective, and femalesas more nurturing and acquiescing, remain strongly embedded in our consciousness.Voting with Care 23We are left with the question--is there a moral difference between men andwomen? Are women more caring in their moral response than are men? The debateon this question forms the substance of our next chapter.24Chapter 2: The Difference DebateIn his now classic study of gender identity, Robert Stoller4 tells us thatregardless of the combination of forces that influence our sense of gender identity--bethey biological or cultural--by the time we are three years old, that identity hasdeveloped (1964, 220). By this early age, we are aware that we are male or female--afact which colours not only the way we look at the world, but what the world expectsfrom us, and how the world treats us. Gilligan argues that these different experienceslead to different ways of thinking and she offers strong anecdotal evidence to supporther case, but her critics say this evidence is insufficient. This chapter looks atevidence--both challenging and supportive--that has emerged from studies on moraldifference conducted since 1982. While some of these studies involve (Jilliganherself, she has argued that social scientific research may be inadequate for a fullappreciation of the nuanced differences between male and female moral reasoning.Accordingly, before departing to the laboratory, we will spend some time in the field.Observations in the FieldWe need only look to the way children play to see differences in male andfemale behaviour. As Greenstein said in his seminal study of child socialization,“When sex differences emerge early in life it is likely that these differences reflectdeep-seated cultural themes” (1965, 112). Greenstein argues that the culturalassumptions learned in this period are particularly potent because it is a time whenthe child learns uncritically, is not conscious of alternatives, and lacks any standardsfor judging received information. And, because it is also a time when the child’spersonality is forming, what is learned during this period become unconscious,internalized values and provide a filter for future information (79). How the child’s‘ At the time, Stoller was Associate Professor of Psychiatry at the University of California School ofMedicine, Los Angeles, California.Voting with Care 25parents behave--individually and toward each other--what the child views ontelevision, reads in books, or learns in school or in church--all have an impact on thatchild’s sense of self and relation to the world. If, for instance, the televisionadvertising that accompanies children’s programming shows little girls quietly playingwith dolls or being nurses, and little boys aggressively playing with guns--as much ofit does--those images send a clear message to children of culturally ascribed roles.Janet Lever, following Piaget’s lead, felt that differences observed in playreflected not only early childhood socialization, but that play itself was a form ofsocialization. In 1974, Lever carned out an intensive study of 110 ten and elevenyear old boys and girls from one suburban, and two city schools in Connecticut,using a variety of research methods including observation, interviews, andquestionnaires. Children’s play was documented both at school and in their hoursaway from school. Lever’s study led her to a number of conclusions such as: boysplay in larger groups, girls in small; girls are more likely to take part in boys’ gamesthan boys are to take part in girls’; boys play more competitive games, girls play morecooperatively, with no explicit goal or winner; and, boys play tends to be more ageheterogeneous. When girls play with younger children, their play reflects motheringLever also agrees with Piaget that there are fewer moral lessons in girls’ play.Games like jump-rope and hop-scotch rely upon turn-taking in contrast with thecontingent rules of strategy that accompany sport games, therefore she argues thereare fewer opportunities for judicial experience (478-83).While I was reading Lever’s study, my seven year old daughter Leah, and hereight year old friend, Amanda, were playing beside me. Their presence provided anopportunity for me to see whether, nearly twenty years later, these differences in playstill existed. And, because Lever’s study had not focused on the question of moraldifference, I was also interested to find out more about sex specific play behaviourthat might shed light on this question in the context of Gilligan’s argument. The girlsVoting with Care 26agreed to answer a few questions about the way girls and boys play. Not wanting toinfluence their response, I was careful to keep the questions general.I began by asking the girls whether they perceived any difference between theway they and the boys they know play. It was not a question to which Leah hadgiven much thought, but Amanda had some very strong views on the matter. Shetold me that for one thing, “girls don’t like to kill their toys.” This seemed a novelresponse, so I pursued the matter. Amanda told me that when boys played with“things” like dinosaurs, the dinosaurs always kill each other. “What do girls play?” Iasked. “Girls play house.” “Don’t boys play house?” I asked. “Well yessometimes,” said Amanda but, “if boys play house, then there always has to besomething like a burglar who will kill somebody in the house.” I asked Amanda ifburglars never appeared when only girls were playing, and she allowed as howsometimes they did. So I asked, “if there is a burglar, what do the girls do and whatdo the boys do?” “TI there’s a burglar, we get them out of the house or they becomeour friend, but if boys are playing and there’s a burglar, they kill them.”While Amanda’s comments can only be considered as ad hoc support forGilligan’s thesis, they seem a vivid illustration of sex differences, at least in terms ofattitudes to violence and to solutions based on a harmony of interests rather than aclear set of rights. It would appear that where the boys see the solution in terms ofthe rights of those in the house, the girls can foresee a solution which includes theburglar, albeit a much nicer burglar. One could argue that by killing the intruder, theboys are also achieving harmony, this time for those who are rightfully in the house,but there seems an essential difference between a solution which achieves harmonyfor some and a solution which is able to envision a potential harmony for all.I decided to take a more empirical look at current manifestations of sexdifferences in children’s play. In August, a time when school daycares operate on aVoting with Care 27full day basis, I spent two days observing approximately 15 six, seven and eight yearolds at play at Leah and Amanda’s daycare, the Jericho Kids’ Club.At this daycare, indoor play is structured into activity centres and childrenmust sign up for the activity centre they prefer. If they wish to play outside, this isalso an agreement they make with staff My two days at the daycare led me to makea number of observations. For one thing, boys and girls still do not choose to playtogether, although some activities provided venues for sex mixing more than others.The art table, for instance, found boys and girls together although their conversationstended to be same sex. While organized sports activities seemed common for at leastthe boys children in Lever’s study, I observed few organized team games at thedaycare, although I know it does occur and that it normally involves boys (althoughsometimes girls play as well). Also, not all boys are violent and not all girls are non-competitive. There were examples of small groups of boys playing quietly, andthere were aggressive girls. And, girls tended to play outside just as much as theboys and when they were outside, they seemed equally boisterous. However,Amanda was right--girls do play house, and boys play “crash” games, and if therewas one feature that characterized their play, this would be it. Two groups provideexamples of this behaviour.In one case, a group of five girls were involved in designing and building a‘house’ in which there were to be two rooms: one for a hampster and one for a cat.There was much conversation about the materials needed to make proper homes forthese animals, a decision that seemed to call for understanding the situation from theanimal’s point of view. At one point, one girl said, “If you were a hampster, whatwould you want?” Other conversations focused on Sarah, the cat. “Close that atnight so she can’t get out.” “But what if she gets lost in our luggage?” “She canbreath under sawdust.” When design inadequacies presented obstacles to easy living,alternative solutions were sought. “Lacey (the hampster) could probably go in there.”Voting with Care 28“No, I don’t think Lacey can fit there.” “But Vanessa, one at a time could go.”“Well, if you change this, that would be good for Sarah.” When all was complete,one girl changed the dynamics of the situation by saying, “I’m going to be a dog.”In the other case, three boys were playing “dinosaurs” at the sand table. Theirplay was boisterous and seemed to call for various moves and positioning of thedinosaurs in the sand, followed by various crashes, death throes and verbal effects.The plot appeared to be that one of them was trying to get away from the others, andrules centred on where the search could take place. The conversation was punctuatedby comments like: “My guy drownded before you catched me!” “Oh, my leg, my legaaaah.” “He’s dead but when you guys were sleeping I came back.” “Boy, doyou snore.” “master, master!” “But you never found me cause I was dead.”While elements of strategy appear common to both groups, there do appear tobe differences. With the girls, much of the “game” is taken up with considerations ofthe animals’ needs; with the boys, the animals seem peripheral to the action.Nevertheless, although the sex differences observed in children’s play mayalert us to potential differences in moral cognition, the challenge to (Jilligan has beento produce causal evidence. Her empirical claim has, indeed, stimulated a number ofstudies that both support and challenge this notion of moral difference represented by“care” and “justice” positions. We find that the evidence presented clarifies thedebate in some areas and leaves us confused in others. Some evidence is directlycontradictory.5Observations in the LaboratoryTn 1984, Lawrence Walker used a metaanalysis procedure to produce astatistical analysis of some 79 studies using Kohlberg’s stage sequence and scoringWhen referring to biological difference, I will use the term “sex;” when the differences are attributed tosociailcultural Ihetors, I will use the term “gender.”Voting with Care 29system to see whether Gilligan’s charge of sex bias could be defended. Walker didnot find evidence of sex bias. Contrary to Gilligan’s claim that Kohlberg’s scoringtended to disadvantage females, Walker found that the studies actually produced fewsex differences in moral development, and he concludes that women and men aremore similar than not in terms of cognitive moral abilities. Separating the studiesinto three age groupings: childhood and early adolescence, late adolescence andyouth, and adulthood, Walker found that when sex differences did occur, they weremore often found at the adult level than at the other two age levels. However, hecontends that even these differences were not of the magnitude Gilligan hassuggested, and that these results are suspect because in studies where sex differencesfavored men, sex and educational andoccupational differences had been confounded.One of the studies included in Walker’s analysis was carried out by DianaBaunirind. Baunirind (1993) has in turn challenged Walker’s findings, saying theparticular metaanalysis procedure he used biased results toward the null hypothesis.6Again, we do not intend to pursue the debate over Kohlberg’s scoring system and theissue of sex bias, but Baunirind’s response to Walker includes elements pertinent tothe more general moral difference debate. Baunirind argues that Walker, in effect,underweighted the adult response, and since the charge is one of underrepresentationat the levels of higher order reasoning, this is the group of focus. Basing hismetaanalysis on the combined age groupings tended to produce results whichdiminished the importance of the sex differences favoring males in the adult group.Bauinrind further protests that the high correlation between educational attainment6 Baumrind says that the Kolmogorov-Sniirnov test used by Walker assumes a continuous distribution foranalysis, when in fact, KoMbergs stage sequence is based on discrete stages. Therefore, “it is moreappropriate to examine differences in frequency of men and women within a stage or stages than to comparethe differences between mean stage scores of men and women” (1993, 178). Walker’s results would thereforenot show whether women were overrepresented at a given stage.Voting with Care 30and stage 6 scoring produced by a small elite group tested by Kohlberg, has neverbeen demonstrated as an equally efficacious measure for women. In other words,that controlling for education may not even out the odds.Using findings from her study not included in Walker’s analysis, Baunuindreanalyzed the evidence using statistical techniques that treated Kohlberg’s stages asdiscrete, and controlled for both education and occupation. She found thatemployment per se did not make a difference in female moral stage scoring--therewere no significant differences between employed and unemployed women--andwhile educational level and stage score level was significantly correlated for men, thesame was not true for women. At the lowest educational levels, women obtain higherstage scores than men; at the highest educational level, men obtain the highest level.The results are particularly interesting for those subjects who actually scored at thestage 6 level. Here, while eleven of the twelve males scoring at this level hadpostgraduate degrees, this was true of only one of the nine women scoring at thislevel. Baumrind suggests that “more men than women may require the formalcognitive training provided by university education in order to apply principledreasoning to social-cognitive dilemmas” (188).Gilligan herself has questioned the efficacy of using Kohlberg’s dilemmas as a validmeasure of women’s moral orientation. Not only are they hypothetical, bearing littlerelation to the lives experienced by most participants in the studies, but they arebiased toward a justice perspective. Dilemmas, such as the Heinz Dilemma, areconstructed in a way to produce a conflict between two principles, in this case lifeand property. As we have seen from responses cited earlier in this paper, women donot necessarily conceive of the situation in the same framework of logic, andtherefore, argues Gilhigan, their answers do not necessarily reflect a ‘justice”orientation even though they may be a postconventional reasoner in Gilhigan’s careVotingwith Care 31orientation. Several researchers have attended to this argument and have employedmeasures designed to include a care perspective.While Friedman, Robinson and Friedman (1987) did not eliminate Kohlbergdilemmas from their study, their scoring system included both Kohlberg and Gilhigantype reasoning measures. One-hundred one undergraduate students (78 from aliberal arts college and 23 from a community college; 47 men and 54 women) wereasked to rate four dilemmas, including the Heinz dilemma, according to theimportance attached to each of 12 moral criteria, six items reflecting Gilligan type“care based” criteria (referred to a “G” items), and six Kohlberg type “justice based”rating items (referred to as “K” items). For instance, an example of a “G” itemapplied to the Heinz dilemma was “Is this likely to weaken or strengthen therelationship between Heinz and his wife?”; an example of a “K” item for the Heinzdilemma would be “Whether the druggist, in exercising his individual rights,infringes on the rights of others.” Subjects were asked to consider the rating item andthen say whether they would attach grea4 much, some, little or no importance to thecriteria (see APPENDIX B for the 12 rating items).Comparing means for the G and K scores, the authors found no significant sexdifferences, and looking at individual items showed only five of a possible 96produced significant differences with three of these in the opposite direction to thatpredicted. However, while their results would seem to dispute Gilligan’s thesis, theauthors suggest there may be possible weaknesses inherent in their methodology. Forinstance, because there was no existing standard by which they could evaluate theirchoice of G items, those items may not accurately represent a care orientation. Also,scores were based on fixed ratings rather than the “spontaneous productions” uponwhich Gilligan bases her findings. And finally, the use of traditional hypotheticalmoral dilemmas may not elicit care responses in the same way as subject initiatedVoting with Care 32real-life dilemmas. Gilligan has argued that “the moral judgments of women differfrom those of men in the greater extent to which women’s judgments are tied tofeelings of empathy and compassion and are concerned with the resolution of real asopposed to hypothetical dilemmas” (1982, 69). Tn other words, if one wants toactivate a substantive moral response on the part of women, one must present asalient dilemma.One unexpected finding in the Friedman et al. study showed that whilewomen from both the liberal arts college and the community college produced thesame mean on the G items, women from the liberal arts college scored higher on theK items. According to the authors, the latter result is likely a product of the highcorrelation between moral reasoning (of the Kohibergian sort) and level of education,but they are less sure of the implications of the former finding although they suggestthat these results may indicate that “this dimension may have some generality acrossdemographic groups” (1987, 45).Nona Plessner Lyons (1988) is credited with developing the first standardizedmethodology for assessing moral orientation. Using subjects’ self-descriptions andself-initiated real-life dilemmas, Lyons produced a coding scheme that allowed forcodings of real-life dilemmas based on whether the subject saw him or herself asconnected or separate from those in the dilemma, and their respective perspectivetowards others based on either “response” (care) or “rights” (justice). Those classedas “rights” tended to see their relations with others in a more objective and separatemanner, and tended to see moral issues as ones demanding a resolution of conflictingclaims, best solved by invoking impartial principles (35). Those classed as“response” saw themselves as connected to others, and tended to consider others “inVoting with Care 33their specific contexts and not always invoking strict equality” (34). Lyons offers asan example of a rights perspective, the self-description of a 14 year old boy whenasked “How would you describe yourself to yourself?”What I am? [pause] That’s a hard one ... Well, I ski--I think I’m apretty good skier. And basketball, I think I’m a pretty good basketballplayer. I’m a good runner ... and I think I’m pretty smart. My gradesare good ... I get along with a lot of people and teachers. And ... I’m nottoo fussy, I don’t think--easy to satisfy, usually, depending on what it is.As Lyons points out, this boy tends to describe himself in terms of his abilities. Shecontrasts this with the self-description offered by a 14 year old girl:I like to do a lot of things. I like to do activities and ski and stuff I likepeople. I like little kids and babies. And I like older people, too, likegrandparents and everything; they’re real special and stuff.... I have alot of stuff going on. I have a lot of friends in the neighborhood.Although this girl also begins her description by recounting her abilities, she soonpasses into a description of self in terms of her relation to others. Lyons goes on toshow how this way of seeing self is linked to the way one sees moral conflict andrelates a conflict volunteered by the boy, in which he clearly sees the issue as one ofcompeting principles: “Well, you have to think about what would be right ... and thenare you gonna stand up for what’s right and ong to your friends, or are yougonna let them get you into going.” She contrasts this with a dilemma produced bythe girl which involved a decision about which friend should be the beneficiary of apaper route job she was giving up. The girl decides in favour of the person who shefeels will be most responsive to her older customers. She decides she has made theright decision because “The person that was bad for the job finally realized that theperson [chosen] was going to be a good person to do it” and in the end, “everybody’shappy” (25-8).Having developed her coding scheme, Lyons then used it to analyze a studydesigned by Gilligan. Thirty-six people, two males and two females at each of aVoting with Care 34number of age groups from eight to sixty-plus, all of similar levels of highintelligence, education, and social class took part in intensive, Piagetian typeinterviews. Structured questions led to more unstructured probings and clarificationsdesigned to elicit the individual’s own “experience of self and domain of morality” asrevealed by their real-life dilemmas (37). Considerations were categorized as eitherResponse Predominating, Rights Predominating, or as Equal Response/RightsConsiderations. Following the suggestion by Kohlberg and Kramer (1969) thatwomen reaching the higher moral development stages are more likely found amongthose from higher education levels and engaged professionally outside the home, thestudy included a sample of professional women.Overall, Lyons found that men and women employ both considerations,response and rights, but that women use response more frequently and men use rightsmore frequently. Statistically, 75% of the women used predominantly response, 25%used predominantly rights. No women used both considerations equally. With themen, 14% used response predominantly, 79% used rights, and 7% (1) used both. Thefindings were significant at p<.OO1, although based on a relatively small sample size.Two unexpected findings emerged from this study. First, those who characterizedthemselves predominantly in connected terms--be they male or female-- tended to useconsiderations of response in constructing and resolving their dilemmas; those whocharacterized themselves in more separate/objective terms tended to useconsiderations of rights. Second, after age twenty-seven, women employed moreconsideration of rights in their moral judgments, reflecting perhaps either a life cyclechange or the higher educationloccupation level of this group.In a study that in some ways replicated Lyon’s, Gifligan and Attanucci (1988)used Lyon’s coding to test whether 1) there is evidence of both orientations inpeople’s discussions of real-life issues, 2) whether people represent both orientationsVoting with Careor privilege one, and 3) whether there is a relation between moral orientation andgender.Three subject samples were tested, each sample being matched in terms of educationand professional occupation. The first sample was composed of 11 women and 10men, varying in age from 15 to 77, but otherwise similar in education and race. Thesecond sample was racially mixed between white and minority students--19 whiteand 20 minority, of which 26 were men and 13 women--and the third sample variedonly by gender--b males and 10 females.Using Lyon’s basic coding system, Gilligan and Attanucci used the terms“justice” and “care” perspectives for scoring, with a justice perspective focusing onproblems of inequality and oppression, and valuing reciprocity and equal respect; andwith a care perspective focusing on problems of detachment or abandonment, andvaluing “an ideal of attention and response to need” (73). All participants were askedthe following questions (78):1. Have you ever been in a situation of moral conflict where you have had tomake a decision but weren’t sure what was the right thing to do?2. Could you describe the situation?3. What were the conflicts for you in that situation?4. What didyou do?5. Do you think it was the right thing to do?6. How do you know?The following dilemma is offered as an example of a justice perspective:The conflict was that by all rights she should have beenturned into the honor board for violation of the alcoholpolicy.... I liked her very much.... She is extremelyembarrassed and upset. She was contrite. She wishedshe had never done it. She had all the proper levels ofVoting with Care 36contriteness and guilt.... I was supposed to turn her inand didn’t.In this dilemma, the subject clearly believes the violator should be turned in, andjustifies not doing so on the basis that the friend was already being punished by herown feelings of guilt. The situation is seen in terms of the rightness or wrongness ofobserving the rules accompanying the no drinking regulation. This contrasts with thefollowing care perspective, in which the subject decides whether or not to turn aproctor in who appears to have a drinking problem.It might just be his business if he wants to get drunk everyweek or it might be something that is really a problemand that should be dealt with professionally.... Maybethere was just no problem there.... I guess in somethinglike a personal relationship with a proctor you don’t wantto just go right out there and antagonize people becausethat person will go away and if you destroy anyrelationship you have, I think you have lost any chance ofdoing anything for a person.In this case, the subject focuses on trying to understand the situation from theproctor’s point of view, and worries about losing an opportunity to help the person bysevering the relationship. Dilemmas were categorized according to whetherconsiderations employed one or other perspective exclusively, predominantly, or acombination of the two, resulting in five categories of response: Care Only, CareFocus, Care-Justice, Justice Focus, and Justice Only.7 Overall results, based on thetotal population of all three samples, revealed that:1) the majority of people used both moral orientations (69%) compared tothose who used exclusively one or the other (31%)A dilemma where 75% or more of the reasoning used either justice or care was labelled either a CareFocus or Justice Focus. Where both orienations were present, but neither accounted for more than 75% ofthe reasoning, the reasoning was labelled Care-Justice.Voting with Care 372) two-thirds of the results favoured one or other orientation (only one-thirdwas in the Care-Justice category)3) where men’s orientations fall either into a justice or care-justice orientation(n=15 and n=30 respectively, with only 1 male subject classed as Care Focus),women’s orientations tend to be spread fairly evenly across the threeperspectives: Care Focus (n:=12), Care-Justice (n=12), and Justice (nlO).Gilligan and Attanucci conclude that while most of us include both perspectives inour moral reasoning, most of us also privilege either justice or care, and that womenfavor the Care Focus, while men favor the Justice Focus. They also point out that ifwomen were not included in this study, the Care Focus perspective would virtuallydisappear.Finally, there is a group of studies which offer mixed support for Gilligan’sthesis, but which also suggest a number of intriguing corollaries.In 1988, Pratt et al. tested care and justice orientations in the context of anumber of factors which have been suggested as affecting orientation usage: dilemmatype (e.g whether it is considered personal or not), person’s age or life stage, andwhether the moral reasoning of males and females tends to converge as theireducational and occupational level advances. An earlier study by the authors8 hadshown that contrary to both Gilligan and Kohlberg, men and women may “movetoward more distinctive orientations in hypothetical moral reasoning with advanceddevelopment and education” (1988, 376). Tn the two studies undertaken by theauthors in 1988, they were interested in seeing whether this same pattern existed forreal-life dilemmas.Study 1 tested three hypotheses: whether sex differences in moral orientationwould be more apparent during 1) middle adulthood for both hypothetical and real8 Cited in Pratt et al (1988) as Pratt, M. and 0. Golding and W. Hunter. “Does Morality Have a Gender?”In Merrill-Palmer Quarterjy 3 (1984) 321-340.Votingwith Care 38life dilemmas, 2) for those reasoning at higher Kolbergian stages for both types ofdilemmas, and 3) when real-life dilemmas were further classified as either presentinga justice or care type of dilemma. The participants were 12 men and 12 women ateach of three age levels: 18-24, 30-45, and 60-75. Education was generally high,overall 62% had a university degree, although in the 18-24 group this percentage fellto 25%. Two Kolberg dilemmas constituted the hypothetical dilemmas, and subjectsvolunteered their own personal moral dilemmas. Personal dilemmas were scoredusing Lyon’s coding scheme. Pratt et al. offer as an example of a justice dilemma, adecision about fairness in deciding the punishment for a child who has stolensomething; and for a care dilemma, a man’s decision whether to counsel a formergirlfriend, now pregnant with someone else’s baby. AU dilemmas were also scored“relational” if the subject considered some aspect of an ongoing personal relationshipas part of the dilemma, and “nonrelational,” if the dilemma existed independent of anongoing relationship (e.g. filling out a tax form).9Results indicated that hypothetical dilemmas produced no significant sexdifferences, a finding that held true when controlled for age and education. However,differences did appear in real-life dilemmas, where men were significantly morelikely to exhibit justice-oriented responses than were women (p:.01). And, sexdifferences here were most marked for the middle-adulthood group (10 justice vs 0care for men; 1 justice vs 6 care for women). There was very little difference in the18-24 age group. Sex also made a difference in the type of real-life (personal)dilemma offered, with 39% of men (12 out of 31) presenting non-relational dilemmasversus only 7% of women (2 out of 27). While overall, relational dilemmas evokedWhile Kohlberg has argued for the priority of his justice type dilemmas as being ones which call for a trulymoral point of view, these authors argue that this is an empirical question, and suggest that relational moralissues may also exemplify moral reasoning, a challenge supported by others, such as Flanagan and. Jackson(1987), who suggest “there are many moral problems which have nothing to do with justice (83).Voting with Care 39care orientations more than non-relational dilemmas, it would appear that womentend to classify relational dilemmas as “moral’ more than do men. Finally, it wouldappear that women who give evidence of higher order reasoning (using Kolberg stagescoring), tend to use care reasoning in real-life dilemmas whereas men do not. Theauthors conclude that this evidence, combined with their 1984 study which also sawdivergence at this level, in orientation used to solve hypothetical dilemmas, indicatesthat “for both real-life and hypothetical dilemmas ..., the sexes tend toward somewhatdifferent orientations with advanced development” (1988, 382).Study 2 looked at the association between parental status, self-concept, andpersonal moral orientation. Based on research offered by D. Gutmann in 1985,’°which suggested that sex roles become more differentiated with the advent ofparenthood, and using Lyon’s coding scheme to determine “connected” and“individuated” self-concepts for justice and care orientations, Pratt et al. tested asample of married part-time university students: 10 male and 10 female parents; and10 male and 10 female nonparents. Each subject was asked to offer two personaldilemmas. They predicted that parents would show greater differentiation of self-concept and moral orientation than would nonparents. Results showed that indeed,there were significant sex differences among the parents, with women less justice-oriented than men, but no such differences occurred in the nonparent sample. And,in this study also, males produced more nonrelational dilemmas than did women(45% versus 22%), but when dilemma content score was used as a covariate againstsex and parental status, it was found that while not significantly related to being aparent, it was related to being male or female. With nonparents, dilemma content didmake a difference in moral orientation used, with no sex difference noted.10 Cited in Pratt et al. (1988) as Gutmann, D. “The Parental Imperative Revised” in The Family andIndividual Development (ed.) J. Meachain. (Basel: Karger, 1985) 31-60.Voting with Care 40Overall, these authors conclude that while sex differences do appear in moralreasoning, they are not as pervasive as Gilligan would suggest, their presence varyingwith a number of factors: life cycle, self-concept, type of dilemma, and stage ofKolbergian moral reasoning. However, the authors conclude that parenthood appearsto be a strong factor in increased gender differentiation (385), a conclusion supportedby an observation made by Belenky and her colleagues that there is a perceptualdifference between being a mother, and just having a mother. (1986, 177)Crown and Heatherington (1989) shifted the question of moral orientation to aspecific venue: competitive sports. A decision taken as to whether or not to sacrificepersonal achievement in order to give assistance to a friend in the competitionsuggests whether the subject’s desired outcome is centred in personal achievement--acharacteristic of an individuated sense of self and other (associated with a morality ofrights), or affiliation--a characteristic of a connected sense of self and other(associated with a morality of care).Two studies were carried out. In the first, 20 males and 20 females from anintroductory psychology class were given a script in which two friends, either twomen or two women, find themselves in a competitive situation in a basketball game.While the dilemma is not subject initiated, the authors argue that the situation is onecommon to participants and therefore can be considered a “real-life” dilemma. Also,subjects were scored according to “production” methods, whereby participantssupplied their own reasoning in their own words, thereby avoiding the possibledistortions of rating systems.According to the authors, these two friends “are paired in a drill that willaffect who starts in the first game of the season” The better athlete of the two mustmake a decision whether to adopt “an achievement orientation and beat the otherfriend, a medium affiliation orientation and allow the friend to partially succeed, or aVotingwith Care 41total affiliation orientation and allow the friend to win” (282). Using Lyon’s coding,the considerations used in reaching these moral decisions were coded as justice, careor justice-care. Results showed that 79.5% of the subjects chose an achievementorientation, the remainder choosing the medium achievement orientation. Nonechose the total affiliation orientation. There was an interaction between respondentgender and decision taken by protagonist only when the protagonist was female. Tnthis case, 60% (6) of the males felt “Jane” should follow an achievement orientedcourse of action; 40% felt she should make a medium-affiliation decision, contrastingwith virtually all of the females reasoning that “Jane’s” behaviour should bemotivated exclusively by achievement (283-4). Evaluations of the justice/careperspectives found 45% of the respondents using a justice predominant reasoning;40% using a care-predominant reasoning, and 15% using an equal justice-carereasoning. No significant sex differences were recorded.In the second study, 92 male and 92 female undergraduates read the samescript, only this time the protagonist had afready decided on which of the three moraldecisions he/she would make. Where the first script ended with the protagonist beinguncertain about which course of action to take, in this script, the protagonist decidesto either “go all out and beat” their friend, to “hold back” and allow their friend toregain some momentum, or to “hold back considerably” ultimately allowing theirfriend to “steal the ball” (284-5). Participants in the study were asked to rate thedecision taken for its correctness, its difficulty, whether it was a moral decision, andrelational considerations both on the team and away from the sports context.Crown and Heatherington found that in terms of correctness, subjects foundthe achievement decision most correct, followed by the medium affiliation and thenthe total affiliation. The situation itself was perceived as having a more negativeeffect in terms of relationship on the team for female athletes than male, and whenwomen rated female athletes, the medium affiliation decision was seen as mostVoting with Care 42problematic between ffiends, but when they rated males, it was seen as being theleast so. Female relationships, outside of sports, were also seen to be most negativelyaffected by the situation, and again when considering the medium affiliation decision,the friendship of women was seen to suffer more than that of males (285-6).The authors conclude that while support does not exist for sex differentiatedmoral orientations in a competitive sports venue, there does seem some suggestionthat women see relational dilemmas as moral where men do not, indicated by womenanswering “probably yes,” the decision was moral; men answering “probably no”(287). There also seem indications that women expect to act in a competitive way ina sports venue--that this behaviour is right and appropriate to the activity--but thatthis behaviour will hurt their friendships with other women more than similarbehaviour will affect male friendships.Several of the studies so far have presented evidence which shows that malesand females, while incorporating elements of both care and justice orientations intheir reasoning, tend to spontaneously pnorize one or the other. The conclusion hasbeen that the perspective priorized is the modal behaviour for that sex. D. KayJohnston (1988) challenges this assumption, suggesting instead that when prompted,males and females may be equally capable of using either orientation, depending onthe content of the dilemma. Johnston used fables as a context for eliciting rights andresponse orientations. Although the argument has been made that only real-lifedilemmas offer an appropriate context for the care orientation, Johnston defends hermeasure on the basis that they offer a specific and consistent context for comparison,and because they are not as personal as real-life dilemmas, she does not feelconstrained in challenging the respondent’s use of a particular orientation in solvingthe dilemma. Also, unlike Kohlberg’s hypothetical dilemmas, subjects were askedVoting with Care 43not only to provide a solution to the dilemma implicit in the fable, but to constructjust what that dilemma was.Sixty adolescents recruited from schools in a community north of Bostonbecame the subjects for this study, equally divided between boys and girls age 11 and15. These students were given two fables to read: “The Dog in the Manger,” and“The Porcupine and the Moles.” After each fable was read, the children wereinterviewed, with the interviewer using the Piagetian “clinical examination”previously described. The children were ultimately asked to produce two solutions: a“spontaneous” and a “best” solution. Lyon’s coding scheme was used to determinewhether constructions and solutions were best categorized as exhibiting a responseorientation, or a rights orientation. Without knowing the fables, it is difficult tounderstand the moral reasoning employed by the particpants and Johnston’sinterpretation of that reasoning, therefore I include the fables here.The Porcupine and the MolesIt was growing cold, and a porcupine was looking for a home.He found a most desirable cave but saw it was occupied by a family ofmoles.“Would you mind if I shared your home for the winter?” theporcupine asked the moles.The generous moles consented and the porcupine moved in. Butthe cave was small and every time the moles moved around they werescratched by the porcupine’s sharp quills. The moles endured thisdiscomfort as long as they could. Then at last they gathered courage toapproach their visitor. “Pray leave,” they said, “and let us have ourcave to ourselves once again.”“Oh no!” said the porcupine. “This place suits me very well.”The Dog in the MangerA dog, looking for a comfortable place to nap, came upon theempty stall of an ox. There it was quiet and cool and the hay was soft.Voting with Care 44The dog, who was very tired, curled up on the hay and was soon fastasleep.A few hours later the ox lumbered in from the fields. He hadworked hard and was looking forward to his dinner of hay. His heavysteps woke the dog who jumped up in a great temper. As the ox camenear the stall the dog snapped angrily, as if to bite him. Again andagain the ox tried to reach his food but each time he tried the dogstopped him.”For the Porcupine and Mole dilemma, Johnston used a rights coding for responsessuch as: “The porcupine has to go definitely. It’s the mole’s house.” She coded thesolution as a response orientation for such suggestions as: “Wrap the porcupine in atowel!” or “The both of them should try to get together and make the hole bigger.Solutions like, “They (moles) should help the porcupine find a new house” werecoded as indicating both orientations were used.As the interview proceeded, students were asked whether they could think ofanother way to solve the problem, a technique employed to test the interviewee’sability to switch orientations. An alternative solution instigated a repeat questioningprocess. If the subject was unable spontaneously to produce the alternative solution,they were questioned in a way that suggested the appropriate alternative. Forinstance, if a subject was unable to provide an alternate response orientation, theywere asked, “Is there a way to solve the dilemma so that all of the animals will besatisfied?” and so on. Finally, interviewees were asked which, of all solutionsdiscussed, they considered to be the best.Results indicated that while the moral orientation for spontaneous solutionsdiffered by sex for both fables, only in the Dog in the Manger Fable were thosedifferences significant and in the expected direction. Furthermore, response variationbetween the two fables occur most in the female population. While girls favoured aresponse orientation in the Dog in the Manger Fable, they favoured a rightsThe fables appear in the Appendix section of Johnston’s study and are adapted from Aesop’s Fables, retoldby A. McGovern, (Scholastic Book Company, 1963).Voting with Care 45orientation in the Porcupine and Moles Fable.’2 In the Dog in the Manger “best”solutions, males became split between rights and response, while females clearlyfavoured response. In the Porcupine and Moles “best” solutions, females remainedmore oriented to response,with males clearly favouring a rights orientation. In otherwords, thinking about the “best” solution, females responded in the expecteddirection, but males were much more influenced by dilemma content (the Porcupineand Moles Fable could more easily be construed as a conflict of rights). And,Johnston adds, “lii both fables if there is a change from the moral orientation usedspontaneously to that used for the best solution, it tends to be from the rights to theresponse orientation or to a solution using both orientations” (62). Johnstonconcludes that this suggests that when forced to reappraise a situation, a moreinclusive perspective is activated. This is so especially in the dog fable which, incontrast to the porcupine dilemma where differences appear untenable, lends itselfmuch more to compromise (62-63).Overall, Johnston suggests that rather than different orientations, it may bethat males and females perceive different strategies for conflict resolution. In herstudy, boys tended to invoke a response orientation only when they saw thepossibility of an ongoing relationship, and this could only occur when differencesmoved to the background. For instance, where boys see the differences between theporcupine and the moles as being so great as to minimize possible ongoing solutions,girls tend to assume that the relationship exists and can continue. So, says Johnston,“In contrast to a simplistic representation of the theoiy which holds that theimportance of relationships is more salient to females than to males, is the idea thatmales and females tend to negotiate conflict in relationship in different ways” (65).12 In the Dog in the Manger Fable, 50% of the girls used the response orientation; 40% used a rightsorientation. In the Porcupine and the Moles Fable, 50% favoured a rights orientation; 30% a responseorientation.Voting with Care 46Girls talk first, and then if talking fails, invoke rules. Boys invoke rules first, andthen when presented with the notion that rules might not be the best solution, theirchoice centers on whether to invoke power, or whether to begin to talk and addressspecific needs.And finally, Mary Brabeck, reviewing literature on sex differences in moraljudgment, presents evidence that shows little significant sex differences occurring inmanifestations of altruism and empathy--the attributes most commonly associatedwith caring behaviour. According to Brabeck, while the females of this society havea reputation for being helpful and caring, studies on children have shown girls to beonly slightly more altruistic than boys, and studies of altruism in adults areconfounded by the use of stereotypic measures. Males, for instance, are found to bemore altruistic in responding to a situation where a woman’s car has broken down(1993, 43-4). Studies on empathy also tend to reveal more similarities thandifferences. Males, it appears, are equally as capable of understanding the reactionsand needs of others as are females. However, citing Hoffman’s research on sexdifferences in empathy,13 Brabeck suggests there may be a distinction between“cognitive empathy” in which one has an awareness of another’s feelings, and“affective empathy” in which one is able to feel the emotions being experienced bythe other: “I feel what you feel” rather than “I know what you experience.” Studiesof affective empathy do produce significant sex differences, with girls obtaininghigher “vicarious arousal scores” than boys (44). Otherwise, says Brabeck, studieswhich focus purely on perspective taking and ability to recognize affect in others, donot yield such consistent results.13 Reference cited in Brthecks bibliography as S. Hoffman, “Sex Differences in Empathy and RelatedBehaviours,” in Psychological Bulletin 84 (1977) 7 12-22).Voting with Care 47We are left with some oveniding findings and a number of new questions. Itwould appear that evidence does exist for two different moral orientations, but that itis not as pervasive as Gilligan has suggested. When it is observed it is more likely tooccur among adults (although Johnston provides evidence of differences in youngerage groups); in parents more than in nonparents; and to be affected by other variablessuch as dilemma content (rights or response; relational or nonrelational), and context.Higher educational level seems to produce a Kohibergian type of higher order moralreasoning for men, but the evidence on this is conflicting for women. Walker, forinstance, says sex differences disappear when controlled for higher education, whilePratt and his colleagues say that if anything, differences become more manifest athigher education levels. Finally, in the one area so closely identified with womenthat we might expect a clear sex difference--altruism and empathy--it appears theverdict is mixed. However, even Brabeck, the author of this last challenge toGilligan’s thesis, admits there remains strong intuitive appeal that women are themore compassionate sex.In the following study, I offer a simple empirical contribution to this debate.While the experiment itself cannot claim the same level of justice versus carediscussion offered by the preceding studies, three elements of this study are worthyof attention: it addresses the veiy basic question of whether women are more caringin their moral reasoning than are men; it offers an opportunity to look for thisorientation in a cross-cultural sample; and finally, it provides an opportunity to testfor caring behaviour in the context of voting behaviour, complicated by the potentialinfluence of public opinion.48Chapter 3: Voting with Care--The StudyIn a democracy, when elections are fairly conducted, we assume that citizensvote not only because it is their right to do so, but also because they feel their votemay have some influence on an outcome about which they care. If this is so, then achoice between candidates becomes a moral choice, especially if those candidatesrepresent two moral points of view. Carol Gilligan has suggested that when men andwomen make moral choices, they may base their judgments on different criteria:where men, motivated by individualistic values, may see the issue as one involving aconflict of rights solved through the application of a universal principle; women,motivated by a sense of connection to others, may see the situation as one ofconflicting needs, best solved when no-one is harmed. But what happens when menand women enter the poffing booth? Does a justice type of moral reasoning influencethe way men vote? Do women vote for the candidate who will make the most caringdecisions? And finally, in modern election campaigns, subject to relentless electoralpolling, another variable enters the picture: the impact of public opinion on votingbehaviour. In what way does knowing how others think--not just our immediateacquaintances but the entire electorate--influence the way we vote?While there has been a resurgence of attention to the potential influence ofpublic opinion polls on voting behaviour, it is not a new issue. In British Columbia,from 1939 until 1983, electoral poils were prohibited during election campaignsprecisely because of this fear. George Gallup argues that the fear is unwarranted:Polls do have an influence on the conduct of government and theplanning of political campaigns for the obvious reason that no betterway exists to gauge the opinions and preferences of voters. But thisimpact ofpo1is on political decisions has little or nothing to do with thefear entertained by many that the publication of poll results leads anappreciable number of persons to switch sides in an electioncampaign... (1976, 33).Votingwith Care 49However this is a contention disputed by others like Jean Laponce who says, “... itwould be unreasonable to assume that knowing how others intend to vote would haveno influence on one’s vote” (1983, 2). Certainly, as long as one pays some attentionto media, it is difficult not to be aware of what others are thinicing. If nothing else,where poll results might once have been included as part of election news, today pollresults often are the news and form the headlines and lead stories in electioncoverage.’4 According to Richard Johnston, daily tracking during the 1988 nationalelection revealed that “polls penetrated deeply into the electorate.” Lookingspecifically at attention to polls occurring prior to the leaders’ debate in that election,70% of respondents surveyed in the 1988 National Election Survey, who indicated avote intention, had read or heard of a poll in the seven days prior to the debate (1992,206).But how do polls effect us? Do they make us want to vote like everybodyelse--a phenomenon known as bandwagomng? Do they change our vote from ourpreferred party to a strategic choice based on eliminating our least preferred choice?Or, do they make us want to vote for the lower rated candidate--a phenomenonknown as underdogging. Perhaps they have no effect beyond providing an interestingindicator of immediate political sentiment. Gallup would argue the latter is the mostaccurate scenario, that beyond the possibility of using a poli for strategic voting, andhe sees no problem with this, polls neither influence a person to bandwagon or tounderdog. He cites a study undertaken by his own polling company which matchedparty preferences of respondents in a city where there were no published polls to14 According to then British Columbia Liberal Party President, Floyd Sully, the event that actually launchedthe Liberal Party’s fortunes following the leaders’ debate in the 1991 provincial election, was the front pagepublication of a positive poll result in the Vancouver Sun, three days after the debate (Oct. 11), with theheadline titled, “Massive boost launches Liberals.”. According to Sully “until then, the print media had beenlargely ignoring the Liberals, and what they had to say wasn’t particularly positive. When that headlineappeared, it made people sit up and take notice.” (Comment made to author in Bancroft, W. “The DebateEffect and Liberal Fortunes in the 1991 B.C. Provincial Election.” Unpublished essay (April 1992) 11.Voting with Care 50party preferences of respondents in a city where polls were published in local papers.The study found the party preferences of those polled in the two cities to be almostidentical (1976, 36). However, an extensive experimental study carried out by Ceciand Kain in 1982 revealed massive shifting in response to poii information.During the Carter/Reagan presidential election campaign in the United States,Ceci and Kain used nine classes of undergraduate psychology students to testwhether dominance information contained in electoral poll results would changecandidate preference. Dividing the students into three test groups, conflictingelectoral polling information was given to each of the three groups in the morning,and then to a proportion randomly selected across the three groups in the evening. Inthe morning, one group was told that the most recent polling results put Carter in thelead; one group was told that Reagan was in the lead; and the final group was givenno polling information. Each group was then “polled” for their candidate preference.In the evening, polling confederates phoned the randomly selected group, and onceagain gave them conflicting poll results before soliciting their current candidatepreference. Ceci and Kain found “dramatic, but opposite findings” (235). Ratherthan using the information strategically, or siding with the winner, these respondentstended to react against whoever was cited as being dominant in the P.M. condition.And, as the authors point out, while this might at first appear as an underdog effect,the movement away from dominance did not mean a simple switching to the leastfavoured candidate as the underdog effect would suggest. In many cases, therespondent’s earlier stated preferred candidate would be the one he or she heard asbeing dominant in the evening, in which case, the respondent remained with theirinitial preference but scored them lower. Ceci and Kain call this behaviouran”oppositional reactivity hypothesis”--the dynamic which occurs in response todominance information which motivates “movement away from whoever wascurrently being touted as dominant” (240).Voting with Care 51There is also research which suggests that groups respond differently to pollinformation. A study by Navazio (1977)15 tested the impact of poll results by lookingat opinion responses of a control group (not aware of poll results) and anexperimental group (made aware of poll results). It was found that when poll resultsprovided a strong negative evaluation of Richard Nixon, while there were nodifferences between the two groups in terms of tendencies to bandwagon orunderdog, occupational attitudinal differences emerged. Blue-collar workers weremore favourable to Nixon than blue-collar workers in the control group; and clericaland white-collar workers were less favourable to Nixon than those in the controlgroup (Ceci & Kain 1982, 229-30). As Ceci and Kain say, “The opinion poll canplay a role as a reference group in itself’ (230). These authors, as well as Johnston,also suggest that both the level of party commitment and level of electoral politicalknowledge’6may make a difference to how much a person is affected by pollinginformation. According to Johnston, for respondents in the 1988 National ElectionSurvey, poll awareness had a direct impact not only on whether they expected theparty to win or lose, but also “dramatically tilted the balance toward information thatwas both current and in the public domain and away from information that was insidethe voter’s head (207). There is also some suggestion that the politically uninvolvedare more susceptible to “high-profile polls.” In other words, the recording of aparticularly dramatic popularity rating may receive undue weighting. Instead ofplacing the information within a campaign dynamic of change, those who arepolitically naive may fixate on this information. And if that information is15 Cited in Ceci and Kain as Navazio, Robert. “An experimental approach to bandwagon research.” PublicOpinion Quarterly 41(1977) 217-25.16 make a distinction here based on a definition of politics that extends beyond that traditionally defined asconcerning the formal political institutions of government, parties and elections. I prefer a broader definitionof politics, such as that offered by Harold Lasswell which defines politics as “Who gets what, when, andhow”.Voting with Care 52inaccurate, and not all polls are in fact accurate 19 times out of 20, that can haveserious effects for democracy.’7 Christopher Hitchens, writing in Harper’sMagazine, is especially critical of polls, seeing influence not just as a by-product ofpublished results, but as the insidious intent of pollsters intent on “shaping” publicopinion, especially that proportion of the public unable to analyze the informationcritically. According to Hitchens, “Fluidity,’ is what pollsters call the chaos andignorance that they seek to influence” (1992, 50).If we consider women to be a separate political culture, as ThelmaMcCormack has suggested, and as Gilligan’s thesis would support, then we mayexpect women and men might respond differently to information presented in polls--theoretically, we should expect women to respond to this moral decision in a mannermotivated by an ethic of care. The following study tests the hypothesis that if womenand men respond differently to electoral polling information, that difference will beobserved in voting choices that show that women are more caring than men.DesignThe study is an adapted replication of an earlier study by Laponce (1966).Theorizing that if citizens see an election as no more than a game, they will vote in away that will prolong the game--by restoring equibalance in this particular political“system” through a vote for the lower rated candidate--the underdog--Laponceconstructed a voting situation of “political near-vacuum” in which no informationexisted about the hypothetical candidates except their names, Smith and Jones. Fromthis no information condition, Laponce was able to insert additional candidate criteriato produce a more “war-like” election environment--one where the outcome matters17 In March and April of 1993, polls underaken by four polling companies: Angus Reid, Gallup, Environicsand ComQuest Research predicted three scenarios regarding the possible outcome of a future federal election.The Tories under the new leadership of Kim Campbell would 1) “sweep the next election” (predicted byGallup), or 2) “win a sizable majority” (predicted by Angus Reid and ComQuest), or 3) “lose to the liberals”(predicted by Environics) (Beltrame 1993:A4).Voting with Care 53more than the game--a situation that should see a tendency to vote more in line withpreference rather than dominance.Beginning with the basic scenario, Laponce carried out five differentexperiments involving approximately 1000 undergraduate psychology students. Avariety of hypotheses were tested including: whether the introduction of an issuemakes a difference to the voting decision; whether the size of the gap makes adifference in the tendency to rescue the underdog; whether in a three candidatecontest, the middle candidate would benefit from a combination of bandwagon andunderdog effects; whether stated party preference made a difference to the outcome;and finally, whether age makes a difference in the tendency to produce equibalance.While each experiment produced its own thematic variant, all were based onthe notion of a voting decision taken in response to the overall results of a previousvoting decision. For example, in the first experiment, students were given a ballot onwhich only the candidate names Smith and Jones appeared, one on top of the other.Students were asked to name their preferred candidate, and then the ballots werecollected, counted and a fake result was announced--that Smith had received 69% ofthe votes; Jones, 31%. With these first ballot results now in their head, students wereasked to vote again, and again the ballots were collected and counted. All in all fourballots were taken. On the third ballot, identifying candidate features such as age,sex, religion and ethnicity were included on the fourth, students responded toinformation revealing that the vote on the third ballot had produced a large gap of89% to 11% between Smith and Jones.Experiment number two dealt specifically with the question of whether thesize of the popularity gap makes a difference in the tendency to choose the underdog,while the third experiment introduced the notion of a third candidate to see whether aVoting with Care 54middle candidate might benefit from both bandwagoning and underdog effects.’8 Thefinal two expeiiments involved the introduction of party identification as anintervening variable, and of age, measured by testing two different age levels ofschool children.Results showed that overall, an issue does reduce the gaming nature of theelection (although the tendency reasserted itself after several ballots); that the largerthe gap, the greater the tendency to equibalance; that when one candidate clearlydominates and one candidate is clearly nondominant in a political near-vacuum, theunderdog is favored; that when party preference is activated, it does influence theoutcome; and that younger children are more likely to bandwagon than older agegroups. While Laponce attributes most movement to the underdog as a gamingattempt to equalize the system, the party preference experiment revealed a counterintuitive result. In this case, the third ballot in a three candidate competition providedsubjects with candidate party identification as follows:A (Independent) 45%B (Democrat) 44%C (Independent) 11%Votes for these three candidates were then cross tabulated with subjects’ o partyidentification, indicated in a follow-up questionnaire. It was expected thatRepublican voters, now motivated to respond in a more “war-like” fashion, wouldbandwagon for strategic reasons--voting for A to ensure that B did not win. Whilethe bandwagon effect was strongest in this condition, as Laponce says, the underdogvotes “were far from eliminated.” Of those who had voted for B on the second ballot,when party identification was unknown, 68% moved their votes to A on the third18 This did not prove to be the case. Testing three conditions: a near-equality; a near-equality of the top twocandidates; and a situation of clear dominance and even spread, Laponce found votes were randomlydistributed in response to the first two conditions, and in the last condition, while the underdog effect wasactivated, only the lowest rated candidate benefitted.Voting with Care 55ballot, but 32% went to the underdog. Laponce refers to these underdog rescuers as“floating voters” “whose function in the political system is in some respects akin tothat of the Red Cross on the battlefield” (991).In the current experiment, the focus is also on voting behaviour in response tofirst ballot news, however in this case, the “first ballot” is actually the results of apublished electoral poll. As well as the ballot design, this experiment also beginswith a political near-vacuum condition and then introduces candidate criteria, but thistime the intervening variable identifies the candidates by ideological position--adifference reflecting care versus justice orientations, rather than game versus war.As Laponce’s results indicate, the effect of polls is not systematic: “In somecultures, in some years, for some types of individuals, the bandwagon tendency willdominate, while in others, at other times, it will be the underdog effect” (1983, 2). Ifwomen and men respond differently to poll information, our intuitive expectation isthat women will be more likely to vote for the underdog than will men, reacting likeFlorence Nightingales rescuing the injured candidate/soldier in the “political RedCross.” However, in Gilligan’s moral construct, a caring orientation could bemanifest either by “getting on the bandwagon” or by “rescuing the underdog.”According to Gilligan, an injunction not to hurt, motivated by a sense ofconnection to others, can be seen not only in helping behaviour but also in apparentlyconfonning behaviour. Some women in her studies indicated a reluctance to takecontroversial stands that might cause offense to some; others a need to be part of aconsensus, exhibiting a reluctance to go against the status quo (1982, 65-6). Indeed,earlier work in conformity suggests that this is a trait associated with women, andmore with women than with men.’9 If this is the case, then we might expect thatwomen would also bandwagon more than men, and that if there is a sex difference inpolling effect, this would be it. However, more recent studies in this area,Several of these studies are cited in Pugh and Wahrman (1983).Voting with Care 56particularly those taking place within a branch of investigation known as statuscharacteristics theory, dispute the notion that conforming behaviour is a dispositionaltrait in women (Pugh and Wahrman 1983; Wagner, Ford and Ford 1986; Foschi1992; Stewart and Moore Jr. 1992).20 While most research involving mixed-sexgroups produces results in which females often defer to their male partner’s choice,while males do not defer to females, research in this field has also produced resultswhich indicate that conformity in women is not “dispositional,” or innate, but rather“situational”--a product of a) their gendered status, b) the nature of the task and c) thematerials used. For example, Stewart and Moore (1992), using same-sex dyads,tested levels of deference in relation to varying wage levels in three pay conditions:high, low, and no information. In the high pay condition, subjects were told that theywould be paid “on the basis of information we have received from you.” They thensigned a “contract” on which they could see that they would be paid $5.20 for theirparticipation in the task while their partner would be paid $2.50. In the low paycondition, the fee schedule was reversed, and in the no information condition,subjects were not given any particular amount but simply told they would be paid atthe end of the session. Results indicated no significant differences between men andwomen in their degree of resistance in the different conditions, and the authorsconclude that the findings “support the conclusion that women are not generally moredeferent than men” and that the results are consistent with previous findings that20 Status characteristics theory is concerned with the impact of status valued difference on performanceexpectations and task evaluations in task oriented groups. Most gender studies in this field employ astandardized laboratory experiment involving dyads, in which subjects are asked to make a number ofeither/or (binary) choices about information presented on slides (usually ambiguously grouped black andwhite squares)--a task that they tell subjects indicates spatial ability. While subjects work in pairs, they donot meet and cannot see each other. The only information they are aware of is that their “partner” is someoneof the opposite sex.Shown the slide pattern, for example the checkered pattern referred to above, subjects are asked toindicate whether there is a greater proportion of either white or black squares. After indicating their choiceby pressing a button which lights up on a console in front of them, the subject is then notified five secondslater, by another lit console button, of her/his partner’s choice. The subject can then change their initialchoice to agree with their partner, or stay with their own original choice.Voting with Care 57women’s more deferent behaviour in mixed-sex dyads is activated “at least in part” bygendered status differentials rather than any innate personality differences.In the current experiment, subjects worked alone in completing thequestionnaire, and their responses were anonymous. And, while any survey issubject to a certain amount of response bias and therefore it could be expected thatrespondents might “go along” with majority opinion, thinking this to be the desiredresponse, the research just outlined would suggest that this behaviour should be nomore true of women than men, especially since all respondents were aware that theresearcher was a woman. Therefore this study considers the question, “Are womenmore caring in their moral reasoning than are men?” through the tendency to rescuethe underdog.MethodAn experimental questionnaire was administered to undergraduate students intwo social psychology and two political science classes at the University of British.Columbia. 191 students participated, 76 were males and 113 were females. Therewere no refusals, however two incomplete responses were eliminated.2’ Studentswere given no information about the purpose of the questionnaire before filling it outother than it formed an important element in a master’s thesis, and that theirparticipation was voluntary. All classes were told that they were welcome to ask anyquestions as soon as all questionnaires were completed, and debriefings were heldwith all classes in which the purpose of the questionnaire was explained in thecontext of the central question for the thesis.21 In a number of cases, respondents refused to answer question #1--the hypothetical condition. While thesecases were eliminated from any analysis involving that particular condition, their usefulness in evaluatingother dimensions of analysis precluded their ultimate rejection from the study.Voting with Care 58In the questionnaire, students were asked to consider two voting situationsreferred to herein as condition 1 and condition 2. In condition 1, a situation of“political near-vacuum,” respondents were asked to imagine this scenario:You are about to make a voting choice between two candidates, let’scall them X and Y, to whom you feel equally committed. As you standin the poffing booth pondering your choice, you remember that justbefore you left your home, you heard a radio news broadcast reportingthe results of a public opinion poll in which X and Y were given thefollowing popularity ratings:CandidateX 47%CandidateY 35%Who will you vote for, X or Y?After answering this question, students then moved on to Condition 2, in whichinformation about the candidate’s positions on appropriate social welfare policy wasadded:Now let’s try another scenario. You’re once again in the polling boothabout to make a choice between Candidate X and Candidate Y, andagain, you’re aware that the most recent poil results have given X thegreatest share of the popular vote. However, this time you are awarethat X and Y take significantly different stands on the following issue:Candidate X 47% (Believes the poor get too manyhandouts, and must learn to be moreself-reliant. X suggests thegovernment should endeavor to cutback on welfare costs.)Candidate Y 35% (believes the government should do more tohelp the poor even if it means higher taxes.)Who will you vote for, X or Y?Voting with Care 59The gap size of 47:35 was chosen with the intention of having a large enough gap toactivate an underdog response and yet not so large that it becomes pure gaming. Apre-test of the questionnaire had shown this gap to be effective.22The first, hypothetical, condition of political near-vacuum was chosen with theintent of activating a care response based only on considerations of dominance andsubordination. As Johnston says, “the less access a voter has to external information,the more he or she resorts to introspection” (1992, 204). With no other informationto go on except the candidates’ standings in popular opinion, the person in the pollingbooth is forced to vote on internally generated reasons that may or may not be inresponse to others’ opinions. The expectation here is that the internally generatedresponse for women will be to rescue the underdog; for men, the more achievementoriented response of siding with a winner. The questionnaire included opportunitiesfor respondents to provide open-ended “spontaneous” reasons for their choice of X orY in Condition 1, and to express any frustration they may have felt with being forcedto make a choice in such a hypothetical situation.While the venue of the polling booth remains in the second condition, thistime Candidate X and Candidate Y represent ideologically different positions:Candidate X priorizes self-reliance; Candidate Y prionzes helping the poor even atthe cost of personal sacrifice. While the first condition is purely hypothetical--in reallife, no one would ever be asked to vote in the absence of any attendant informationabout the candidates--this is not case in Condition 2. Gilligan has argued that onlyreal-life dilemmas activate a care response, that women are very frustrated by theabsence of information in hypothetical dilemmas (1982, 69). Condition 2 presents a22 Two gaps had actually been tested: 47:42 and 47:3 5. The smaller gap tended not to produce more thanrandom choices. The pre-test also resulted in a rewording of the conditions to reflect greater clarity.Voting with Care 60real-life dilemma many of these subjects had faced before and, as they were all oldenough to vote, were likely about to face again in an upcoming federal election.23Condition 2 lends itself to an application of Lyon’s coding scheme. CandidateX values self-reliance--a principled position based on individual effort. Candidate Yadvocates altruism and self-sacrifice in a program of social welfare geared to thecommunity. Therefore, in condition 2, those choosing Candidate X will be seen asmanifesting a justice orientation; those choosing Candidate Y will be seen asmanifesting a care orientation.Beyond the questions directly related to the two conditions specified wereseveral soliciting demographic information such as age, sex, parent’s ethnic identityand personal ethnic identity. Students were also asked to state their academic major,their party preference, and a short scale of questions measuring levels of politicalefficacy, defined by Kornberg and Clarke (1992) as “a citizen’s belief that he or shecan influence the political process” (93)24 (see APPENDIX C for completequestionnaire). All open-ended questions were “blind coded”--in other words, carewas taken to ensure that no demographic characteristics of the respondent, such assex or ethnic identity, were known when coding spontaneous responses.The questions on ethnic identity were included with an awareness that UBChas an ethnically diverse population and that this was a variable that should beconsidered in order not to make cross-cultural assumptions. As it turned out, a largeproportion of the students taking part in the survey were Chinese, a fact that providedan opportunity to test the hypothesis in two population groups: an overall, ethnically23 The experimental survey was conducted in August, 1993. A Canadian federal election was held thefollowing October.24 A post-manipulation question was also included in which respondents were asked if they could identifythe information being sought in the questionnaire. While the majority of responses made reference to thepossible influence of polling, and a number suggested that it had something to do with the factors affectingvoting decisions, not one student correctly surmised that the variables of interest would be sex and moralreasoning.Voting with Care 61mixed population; and a sample which compared those who identified as Europeanwith those who identified as Chinese.Total Sample FindingsCondition 1: Political Near-VacuumAn overall frequency distribution of responses to the question for condition 1revealed that of the 189 respondents, 63% voted for X, 28% voted for Y, and 9%refused to answer the question. After assigning refusals as missing values, the votefor X or Y was cross tabulated with sex. These results, along with the overallrecoded distribution, appear in Table 1 below.Table 1: Vote for X or Y in condition 1, by sex.% Male % Female % TotalSample(67) (105) (172)X 58.2 76.2 69.2V 41.8 23.8 30.8x2==5.30 p.02In condition 1, over 76% of the females voted for X, the top-rated candidate,compared to just over 58% of the male respondents who did so, an actual differenceof 18%. In contrast, almost 42% of the males voted for Y, the underdog, comparedto just under 24% of the females. While the results of the crosstabulation aresignificant at p=.O2, the trend is in the opposite direction to that which is expected.In this political near-vacuum condition, it would appear that while both males andfemales are more likely to bandwagon than to underdog, men are proportionatelymore likely to vote for the underdog than are women.Voting with Care 62Since the purpose of the experiment was to gain an insight into potential sexdifferences in motivational forces influencing voting behaviour, the questionnaireafforded subjects an opportunity to state their reasons for voting for either X or Y inthis condition. An analysis of the open-ended responses to this question allows usgain an insight into why this apparent anomaly exists.When asked why they voted for X, those who did so (63% of the total sample)provided answers that fell into two clear categories: “conforming” if the subject votedfor X because he or she deferred to majority opinion, and “achievement oriented” ifthe subject voted for X because they thought X would be a winner. Typicalcomments in the “conforming” category were “If most people were voting for thecandidate, then he must be doing something right,” “TI I like both candidates equally,I would then vote for the more popular one,” “Since both candidates were equallycommitted and the majority or more people prefer Candidate X, Candidate X seemedto be more reliable.” Typical comments in the “achievement oriented” category were“He was leading and I wanted to be associated with a winning candidate since I knewno other facts about them,” and “If I voted for Y, X would still be more popular, butnot necessarily win. If vote for X, X may have greater chance to win.” Togetherthese two categories accounted for 107 of the 120 respondents voting for X, 89% ofthe total. Seven respondents refused to answer (6%) and the remainder of thecomments (5%) suggested no clear coding, typified by comments like “It was arandom choice,” and “I liked the higher number.” These were coded as “other.”After assigning refusals and “other” as missing values, “Why vote for X” measuredby “conforming” and “achievement oriented” were cross tabulated with sex.Voting with Care 63Table 2: Condition 1--Why Vote for X. by Sex% Male % Female % TotalSample(33) (74) (107)Conforming 72.7 90.5 85.0Achievement 27.3 9.5 15.0Orientedx24.38 p’=.04From the table, it would appear that while a vote for X, the top-rated candidate, ismotivated by conforming behaviour in both males and females, this tendency is mostclearly demonstrated by females, a difference that is significant at .04.Returning to Table 1, 53 respondents voted for Y in condition 1 (28% of thetotal sample). As expected, the most common reason indicated for voting for Y wassupport or identification with the underdog, typified by such comments as “Help theunderdog,” “Always go for the underdog,” and “Because I feel sorry for the guy.”(25 out of 53, or 47%). All other responses were categorized as “other.” Reasons inthe “other” category tended to fall into two camps: 1) those who chose Y because of awish to be nonconforming (9 of 53, or 17%), typified by comments like,”I believeothers are ignorant about real information and that a majority of people would votewithout knowing really what they are voting for,” and 2) those motivated by gaming(11 out of 53, or 21%). Typical of the latter group are comments like “To even theodds,” or “To make the race more interesting.” The remainder of the “other” categoryconsists of a number of unrelated comments that did not fall easily into the otherassignments (9 out of 53, or 17%). These responses included comments like, “I don’tVoting with Care 64believe in being surveyed by public opinion polls.”25 After assigning those whorefused to answer (3 out of 53, or 6%) as missing values, “Why vote for Y,”measured by “rescue underdog” and “other” was cross tabulated with sex, with theresults as follows.Table 3: Condition 1-- Why vote for Y, by Sex% Male % Female % TotalSample(27) (25) (52)Rescue 48.1 48.0 48.1UnderdogOther 51.9 52.0 51.9p=nsAs the information in the table illustrates, in a political near-vacuum condition, ofthose who vote for the underdog, the driving motivation for both males and femalesappears to be a need to “rescue.”26Condition 2: Social Welfare Issue Position AddedHowever, what happens when candidates become identified with an issueposition, particularly when the position deals with social welfare, a concernassociated with the ethic of care. If the lower-rated candidate takes a position thatreflects an attitude of responsibility and connection to the community, will we nowsee a clear movement of women toward the underdog? And, will we find moreunderdog supporters among women than among men? Table 4 below provides25 The decision to keep these “odd” conunents in the sample, and to group nonconforming and gamingcomments toether in the “other” categorys made primarily on the basis of providing a large enough samplefor statistically reliable results. Also, the category of interest here is the “rescue” category.26 And, since 6 of the 11 respondents who underdogged to keep the game going ‘re men, it also appearsthat women are just as motivated by gaming as men.Voting with Care 65results of the vote for X or Y in condition 2, cross tabulated by sex (refusals incondition 1 assigned as missing values).Table 4: Vote for X or Y in Condition 2, by Sex% Male % Female % TotalSample(76) (110) (186)X 67.1 63.6 65.1Y 32.9 36.4 34.9p=I’sWhile X remains the favored candidate by both men and women in condition 2, therehas been a clear movement toward Y by women. In contrast, men have tended tomove away from Y in this condition and toward X, the top-rated candidate,producing a pattern where slightly more women now favour Y, the underdog, than domen, although the difference is not statistically significant. Using Lyon’s codingscheme, we would say that when given a clear choice between policies reflectingjustice and care positions, 3 6.4% of the women in this sample have responded with acare orientation in their voting decision, compared to 32.9% of men. Howeveroverall, both men and women favour a justice orientation. The following chartprovides a diagram which allows us to see the actual pattern of vote switchingVoting with Care 66Figure 2: The Movement of Votes from Condition 1 to Condition 2, by SexMales Females(67) (105)x Y x YCondition 1 582j 141.8 762 23.8Condition 2 171.8 28.2 60.7 9.3 7.1 32.9 64.0 .0x y x y x y x Y(This diagram is based on a sample size of 172, the proportion who answered the question for the firstcondition, and does not reflect the votes of the 14 subjects who refused to respond to the first conditionbut did respond to condition 2, The breakdown of those votes are as follows: of the 9 males whorefused to respond to condition 1, 6 voted for X and 3 for Y in condition 2; of the 6 females, 1 votedforX and 5 forY.)Looking at the chart we see that while the same trend is manifested by maleand female Y supporters in condition 1--with just slightly more male Y supportersstaying with Y in condition 2--an opposite movement is manifested by condition 1male and female X supporters. Here we see that male X supporters in condition 1,the “no information” condition, become much more strongly attracted to candidate Xin condition 2, when X’s welfare policy position is known. In contrast, female Xsupporters in condition 1 become much less attracted to X when they become awareof X’s policy position.Cross-Cultural FindingsTwo open-ended questions in the survey dealt with ethnic identity. In onequestion respondents were asked with what ethnic group their family identified, whilethe other question asked with which ethnic group they personally identified. In thesecond question respondents either repeated their family identity or offered their ownVoting with Care 67as “Canadian” or hyphenated their family identity with “Canadian.”27 Reasoning thatthere may be a qualitative difference in cultural values between those Chinese whoself identify as Canadian or Chinese-Canadian--suggesting a possible internalizationof western values--and those who see not only their family’s but their own personalidentity as Chinese, the Chinese sample was formed by choosing only the latterrespondents. Those Chinese who had cited their own ethnic identity as Canadian, orChinese-Canadian, were eliminated from the sample. However, in the belief thatthere might not be quite such a difference between overseas European values andCanadian European values, and because otherwise the sample might have been toosmall to be usable, the “European” sample was formed from those who indicatedeither “British” or another European countiy as their family ethnic identity.28 Noattempt was made to ensure that their personal ethnic identity was identical to theirparents. While the European and Chinese male and female sample sizes do not allowfor subsample comparisons, we are at least able to see how these ethnic sexgroupings compare to each other and to the larger sample in their overall reaction tothe two conditions in the experiment.Condition 1: Political Near-Vacuum (Cross-Cultural)Again, assigning refusals in condition 1 as missing values, the European andChinese groups were each cross tabulated with the vote for X or Y. Two interestingpatterns emerge here. While in both European and Chinese populations, the male-female pattern follows the overall pattern established by the larger sample, thedifferences between males and females is larger in the Chinese sample than in theEuropean sample.27 Besides the ethnic groups identified in this experiment, a variety of other ethnic identities were citedincluding: Japanese, Iranian, Korean, Malaysian Chinese, African, Indian, Janpanese-Caucasion, andJamaican.28 These included those who cited their family ethnic identity as: Polish-British, Irish, German, andAustrian. Those who simply stated their family identity as Canadian were not included.Voting with Care 68Table 5: Vote for X or Y in Condition 1 - Cross Cultural Comparison, by Sex.European Chinese% Male % Female % Male % Female(16) (21) (28) (45)X 56.3 66.7 64.3 86.7Y 43.8 33.3 35.7 13.3pns x2=3.83 p=05In the Chinese sample there is a 22.4% difference between men and women insupport for the candidates in condition 1, a difference that is significant at .05.However, in the European sample this difference narrows to 10.4% and produces nostatistical significance (although significance is likely affected by the small samplesize). Again, it would appear that while the bandwagon effect is strongest in bothEuropean and Chinese populations, it is more evident with women than with men andthis is especially so in the Chinese population. Next, we look at the voting choices ofthese two sample populations when the candidates more clearly reflect justice andcare policy positions.Condition 2: Social Welfare Policy Position Added (Cross-Cultural)Table 6: Voting for X or Y in Condition 2- Cross Cultural Comparison, by SexEuropean Chinese% Male % Female % Male % Female(17) (20) (30) (46)X 58.8 60.0 70.0 71.7Y 41.2 40.0 30.0 28.3p=ns p=nsVoting with Care 69Here we see that the sex differences observed in condition 1 have been virtuallyeliminated in both European and Chinese sample groups, with the levelling out beingcreated primarily by the movement of women toward Y. This is the same patternobserved in the large sample, except that there is less movement away from Y in theEuropean male sample. The most striking cross-cultural difference observed occursin the differing male-female gap sizes.Summary of ResultsOverall, the bandwagon effect dominates in condition 1 and condition 2 forthe entire sample, and for both cross-cultural subsamples.29 While this effect holdstrue for both males and females, it is more dominant for females, especially incondition 1. In condition 1, over three-quarters of the females voted for X comparedto less than 60% of the males. However, the introduction of the candidates’ socialwelfare positions in condition 2 produces differential movements for males andfemales. In the large sample, proportionately more males voted for X, the “justicecandidate” in condition 2 than did in condition 1, while proportionately more femalesvoted for Y, the “care candidate” than did in condition 1. When the context movedfrom a contentless, hypothetical situation to a real-life moral dilemma with a choicebetween justice and care perspectives, women moved to the care perspective.However, while large and significant sex differences occurred in the political near-vacuum condition, in a direction opposite to that expected, in the care versus justicedilemma of condition 2, sex differences virtually disappeared, a finding that held trueacross both European and Chinese subsamples. It would appear that in a voting29 The fact that most respondents believed that the experiment was testing polling influence did not seem todeter this tendency, although some went to great pains to rationalize their decision as the following commentindicates: “Candidates were equal except for popularity, therefore there must have been something about themore popular candidate to make him more popular, such as style or charisma or a certain stand on an issuethat would influence my vote”Voting with Care 70context, subject to the possible influence of public opinion, there is not support forthe hypothesis that women are more caring than men.DiscussionTwo major questions emerge from these results: 1) when the voting situation isone where no information is known about the candidates other than majority opinion,why are women more conforming than men; and 2) why are there no significant sexdifferences when the voting situation changes to present a choice between a justiceand care orientation? The first question leads us into an area that touches on thebroader issue of women and politics, so we will save that discussion for last.However, the latter question has direct relevance for the empirical debate on moraldifference.The lack of sex difference observed in the care/justice condition in thisexperiment is consistent with many other studies in this area--both those which havedirectly addressed Gilligan’s argument of moral difference, and those which havelooked at the impact of public opinion polling. Laponce’s 1966 study did not find sexdifferences either.3°It may be that sex differences did exist in the current study but wentundetected because the condition did not provide salient justice and careperspectives. Greeno and Macoby have suggested that while there is no clearevidence supporting the argument that women are more altruistic in behaviour thanmen, this may be because tests have not employed situations which activate this kindof behaviour in women, that it may more likely be manifest in behaviour toward30 The only condition in Laponce’s study which produced any significant difference beten males andfemales occurred when infonnation on the candidate’s sex was introduced. In that case, while only 9% of themales transferred to the lower rated candidate, who was identified as a woman, 20% of the femalestransferred to this candidate. However, Laponce adds that 42% of the female subjects also left the womancandidate upon learning her sex which he offers as confinning proof of the saying that “women have, ascandidates, a limited appeal to their own sex” (1966, 992).Voting with Care 71friends and intimates, not strangers (1993). However, this contention would bedisputed by survey research which shows women more supportive of social welfareinitiatives than men (Kopinak 1987; Carroll 1988; Miller 1988) which, while notconclusive proof, certainly suggests that women extend care beyond immediatehorizons. Among the latter group of researchers, Susan Carroll has suggested that agender gap on this issue may reflect not only nurturing sentiments on the part ofwomen, but also socio-economic vulnerability. In this argument, those who wouldrespond most to policies advocating greater distribution would be those who canidentify with the need for these policies and who would most benefit by them. So,not only because they are motivated to care for others, but because 65% of all welfarerecipients are women (Kemp 1985, 143) it would be expected that women, more thanmen, would support more generous welfare policies. However, the subjects in thisexperiment were university students and it can be assumed that they feel little socioeconomic identification with welfare recipients. It might be that peace andenvironmental concerns would have been more accessible issues.Diana Baunwind has argued that educational level does not merely assessacademic skills or knowledge of subject matter, but is also “the best single index ofsocial niche, indicating at its higher levels acculturation into the dominant values ofthe intelligentsia in Western society” (1993, 188). This would suggest that not onlymay a response to the poor lack salience for these women through lack of classidentification, but also that women at this educational level and age may reflect statusquo thinking more than their own.31 Belenky and her colleagues (1986) describewomen at this level as “procedural thinkers.” Having learned the intellectual rightsand wrongs--the frameworks for analysis--they have yet to pass to a stage where theirown ideas and intuitions become merely informed by known knowledge, not drivenby it. While the scope of any study is limited by the sample base used, David Sears31 Over 88% of the population was 25 and under.Voting with Care 72(1988) has expressed strong reservations about the ecological validity of resultsobtained from experiments using college and university students, a population hebelieves are in no way representative of the more general population.32 Not only aretheir own opinions still unfonned, relying on the reference group for authority, butthey are also more oriented to self than to others, and are dominated by cognitiverather than affective processes. In the current study, it is possible that all threefactors might militate against a pure care response: the women may not have reacheda stage where their care encompassed others, they may not yet have formulated theirown value priorities, and in a setting that emphasizes cognition over affection,humanitarian concerns may not have the salience for this group that they might forsome. Both the findings and the limitations of this study suggest that further researchof this sort would benefit from using a broader cross-section of the population, andtesting for more variables.While Laponce found no sex differences, he did find that, except in theexperiments involving an issue and where children were tested, the underdog effectwas much stronger than the bandwagon effect. In the current experiment, it is thebandwagon effect which predominates for both males and females, and especially forthe females in condition 1. In the face of the larger societal opinion, subjects in thisexperiment have generally moved toward that opinion, not away from it. Becausethere was no control group, in which voting decisions would have occurred in theabsence of knowledge of public opinion, we cannot say for sure that the bandwagoneffect seen here is purely a result of the influence of published poll results. Butcertainly, since the main difference between this and the Laponce study is that in hisstudy, subjects are reacting to their classmate’s opinion and in this study subjects are32 Sears carried out a content analysis of articles published during 1980 in the three major journals ofsociopsychological research, the Journal of Personality and Social Psycholov, Personality and SocialPsycholov Bulletin, and the Journal of Experimental Social Psvcholoy. According to Sears analysis, 75%of the articles in these journals relied soley on undergraduate subjects (1988, 315).Voting with Care 73reacting to majority opinion generated by the greater society, it would appear that thiskind of polling infonnation has a stronger bandwagon effect. It may also be, as Ceciand Kain have suggested, that in an experimental situation, “poii reports may take ona greater salience and influence than they do in a natural context” (1982, 241). If thisis the case in the current experiment, this may have exaggerated the actualbandwagon effect, but would not account for the difference between males andfemales in this regard. We are left with the question, are males more caring or arefemales more conforming?It would appear from the findings in condition 1, that not only do males have astronger tendency to support the underdog than females, but that when Y supporterscited their reasons for choosing this candidate, both sexes were primarily and equallymotivated by a desire to rescue the underdog. It may well be that in a hypotheticalvoting situation, males respond to the underdog with more care than females, butthere may also be a weakness in the coding interpretation for this variable. A numberof the responses which simply said “I tend to vote for the underdog” might have beenmore properly coded as gaming or nonconformity rather than as rescuing. In thatcase, more of those responses might have been cited by males than females. It mayalso be that while my own sex might not have activated deference in female subjects,it might have prompted greater nonconformity in male subjects. In other words,males in the study might have chosen Y to show me, a woman, that they were notgoing to abide by my expectations, in a way that they might not have done had theresearcher been male. Finally, support for the contention that, in condition 1, maleswere not as motivated to vote for Y out of care as the data would suggest is providedby the movement of males away from Y in condition 2 (see Figure 1).And what of the differences observed between the Chinese men and women,and the European men and women. Why was the gap between European men andwomen so much less than the gap observed between Chinese men and women? AVoting with Care 74possible answer is suggested by Williams and Best’s (1990) cross-culturalexamination of sex differences in fourteen countries. These authors found that “incountries where the women are relatively ‘liberated,’ the affective meaningdifferences in the self-perceptions and ideal self-perceptions of men and women wereless [i.e. women and men perceived themselves in a more similar fashion]” (163).While we really have no idea whether the Chinese women in our sample are less‘liberated’ than the European women, if they are, this may account for some of thelarger gap size in the Chinese subsample. Another factor may have to do withrelative differences in self-esteem. For instance, where no great sex differences inself-esteem were found in the authors’ American sample, large differences werefound between Japanese males and females, with Japanese males exhibiting muchhigher self-esteem than Japanese females (110-1 1).33 However, as Williams andBest caution, one must be very careful in interpreting cross-cultural differences, asobserved differences may be merely the product of methodological flaws. Forinstance, in the present study, the European and Chinese codings are based solely onresponses to a question on ethnic identity and therefore cannot be taken as indicativeof held cultural values.And finally, there is the question of the greater conformity exhibited byfemales in condition 1. Anticipating that condition 1 might not activate a moralconsideration among female subjects because of its highly hypothetical nature, twoquestions in the experimental survey were designed to address this eventuality. First,those subjects who experienced frustration with the condition were asked to cite thenature of their frustration. A second question asked them to rate the level of theirfrustration on a Likert scale from one to five. We would expect that females wouldbe more likely than males to cite lack of contextual infonnation as the source ofThe greater support for X by Chinese students in both conditions is intriguing but beyond the scope of thisstudy to investigate. However, for further reading on this topic, I would suggest H.K. Mas “The ChinesePerspectives on Moral Judgment Development” in the International Journal of Psvcholov 23 (1988) 201-227.Voting with Care 75frustration, and to express higher levels of frustration than males. However, contraryto expectations, results showed that of those indicating frustration with this condition(29%), not only was the hypothetical nature of the situation cited equally by men andwomen, but of this group, men were far more frustrated by the lack of informationthan were women!34 It would appear that it is not the hypothetical nature of thesituation that is driving the higher conformity levels in women.If we accept the evidence that women are no more dispositionally deferentialthan men, then perhaps this apparent conforming behaviour may reflect notconsensus arising out of deference to majority opinion, but a greater sense ofconnection to the community. In which case, more women vote for X in condition 1because, having nothing else to go on except what the majority of people in societythink, they respond to the message of this great “other” by agreeing with their choice.However, this logic is disputed by the results of condition 2 in which proportionatelyless women abide by public opinion than men. If community connection is driving abandwagon response, why are women not bandwagoning in larger proportions thanmen in condition 2?I would argue that the conformity observed in condition 1 is, indeed, deferenceand not the product of connection, and I would argue further that the reason we see itmore in condition 1 than condition 2 is because there is no issue to mitigate theimpact of the abstract notion of politics. Let me be clear. Attributing thismanifestation of conformity to deference in no way implies that deference is aninnate behavioural trait for women, rather this draws upon research suggested earlierthat when women exhibit more deferential behaviour than men, it is the result ofsituational factors such as a gendered status differential or the nature of the task at34Having coded the 1-5 Likert rating into those who indicated lower ratings (1 - 3) and those who indicatedhigher ratings (3 - 5), this variable was then crosstabulated with sex. Of the 55 respondents falling in thisgroup, 63.6% of the males fell into the higher category, while 60.6% of the women fell into the lower levelcategory. The results were not significant (p=. 14) but this may be a factor of sample size.Voting with Care 76hand. In this case, both factors are operant. Women are being asked to vote--apolitical act--and they are being asked to do so in an information vacuum. All theyknow is what others have decided. And, when women think of others in a politicalcontext, those “others” are men.Since Aristotle first established public and private spheres of influence, andput men in charge of both, decreeing that “as between male and female the former isby nature superior and ruler, the latter inferior and subject” (325-323 B.C, 68), therehas been a virtual exclusion of women from the world of formal politics. WhenHobbes, Rousseau and Locke proposed their social contracts, they were contracts thatcould not include women because women were not considered citizens. Both Hobbesand Locke declared that family interests would be adequately represented by thehusband.35 Even Rousseau, who valued compassion and decried social and politicalinequality, saw no contradiction when he declared that “the man should be strong andactive; the woman should be weak and passive; the one must have both the power andthe will; it is enough that the other should offer little resistance” (Rousseau 1762,322).In Canada, women became federally enfranchised in 1918. Tn 1921, there wasone woman, Agnes MacPhail, in the House of Commons; in 1929, women becamelegally considered as “persons.” Even with these “dramatic” advances, politicsremained a man’s world as witnessed by the images portrayed in the 1972 movie, ThcCandidate, starring Robert Redford as a street front activist who becomes lured intorunning for political office. In this movie--a dramatic portrayal of backroom politics--women appear only as wives, secretaries and as political ‘groupies’. All candidatesand all main political players are male.As Brodie points out (1991, 12-13), the franchise was limited to male property owners. While this alsoleft a good proportion of men disenfranchised, it eliminated women. Theorists reasoned that by virtue of theirbiologically determined attributes and roles, women were unsuited for public life and therefore should not becitizens. not only were women the major theorists of classical liberalism,Voting with Care 77In the year that The Candidate was released, only 2% of the seats in the Houseof Commons were occupied by women. Since this time, there has been someimprovement. Tn 1984, the percentage of women in the House of Commons jumpedto 9.6%; in 1988 to 13.9% and in the 1993 federal election, the proportion of womenin the House soared to 17.9%, with 14% of the new Liberal government cabinetpositions given to women. While the increases do speak of dramatic improvement,the continued underrepresentation of women has prompted Janine Brodie to say that“Women remain governed rather than governors, legislated rather than legislators”(1991, 9). And, although it would appear that Canadians find the inclusion of womenin political elites to be an increasingly palatable notion, we have a way to go beforethe word “politician” conjures up images of women and men with equal frequency.Finally, should there be any doubt that certain assumptions about the sex-linkednature of this arena persist, I offer the evidence of the spontaneous commentspresented by the students in this study. Any reference to Candidate X or Candidate Ythat identified the candidate by sex, assumed that sex was male.People are generally more susceptible to social influence when they areuncertain about how to behave; if women are more deferential in a political venue, itis because it is a venue where they do not yet feel they can claim ownership. And,just as the women in Crown and Heatherington’s competitive sports venue felt lesssecure about acting on their own intuitive reasoning in that “masculine” venue, so domost women feel the same lack of security in pressing their views in a politicalvenue, where a “caring” orientation is not necessarily considered appropriate. AsPugh and Wahrman state, “Women are socialized to defer to the judgment of men incertain kinds of situations” (1983, 748). The political world is one of thosesituations.Examples of this deference to the assumed greater political knowledge of“others” are included in the reasons given by a number of women for voting for X.Voting with Care 78One woman said, “At my age and limited knowledge of politics and the factors orissues for which we choose a politician it has been so far easy or safe to go with themajority.” Another offered this reason for choosing X: “... having less knowledgeabout politics, I guess the ‘majority knows’.”36When people have a low sense of political self-worth they are said to possesslow levels of political efficacy. In this experiment, four questions measured politicalefficacy: two measured external efficacy; two measured internal efficacy. Whileinternal efficacy may be defined as the belief that one has the personal capacity toinfluence political decisions, external efficacy refers to the person’s belief that thepolitical system itself will be responsive to their needs. A crosstabulation of subject’ssex with their responses to the efficacy measures appears below.Figure 3: Men, Women and Political Efficacy70• Agree60 ü Disagree50040:.100•. IMales Females Males FemalesInternal Efficacy External Efficacyn=188 (p=ns for internal efficacy; p.05 for external efficacy)36 As the highest conformity ratings came from Chinese women, it may be that this group feels especiallyunempowered in this system.Voting with Care 79(Internal efficacy was measured by the questions: “Sometimes, politics and government inOttawa seem so complicated that a person like me can’t really understand what’s going on”and “People like me don’t have any say about what the federal government does.” Theexternal efficacy questions were: “Generally, those elected to parliament in Ottawa soon losetouch with the people” and “I don’t think that the federal government cares much whatpeople like me think.” The response categories were strongly agree, somewhat agree, don’tknow, somewhat disagree, and strongly disagree. The internal efficacy questions andexternal efficacy questions were regrouped into new variables and recoded to reflect thecategories agree, disagree, and botWdon’t know.))These results are consistent with similar analyses political efficacy undertakensince 1965 (Black and McGlen 1979; Kay et al. 1988; Brodie 1991). In all studies,while both males and females exhibit high levels of distrust toward political actorsand the political system, women have less confidence in their sense of political self-worth than do men. For some reason women, even highly educated women, do nothave as strong a sense of connection to the political system as men. While theevidence of this study does not offer proof that men and women live in differentpolitical cultures, the scope is after all limited to undergraduate university studentsattending the University of British Columbia. We cannot rule out the possibility thatGilligan may be right about moral orientations, and that the exclusion of women’smoral perspective from the formal arena of politics remains a barrier to women’spolitical knowledge, interest and efficacy. In the next chapter, we discuss theimplications of this continued exclusion, not only for women’s political behaviour butfor society as a whole.80Chapter 4: Adding Care to the Moral Discourse—Implications forWomen and SocietyThere is nothing, of course, in the form of liberal democratic theory that callsfor an exclusion of women’s political voice. Quite the contrary. It was expected thatonce women had the franchise and began to vote in the same numbers as men, theywould also become proportionately represented, and sex-linked barriers toinvolvement would disappear. That this has not occurred has been a source ofconsternation and inquiry for those who study women’s political behaviour. Ofspecial concern has been women’s continued lag in terms of political interest,knowledge and efficacy.Various factors, structural and cultural, are often cited as contributing to thissituation. For instance, those who focus on the impact of labour force participationcite the fact that most working women are grouped in low-paying and non-urnomzed“pink-collar”ghettoes34which offer few opportunities to gain political knowledge(Black and McGlen 1979; Burt 1986); others that women who work must alsoperform the dual role of wage work and home work (Vickers and Brodie 1982).Those who focus on cultural impediments emphasize the impact of socializingmessages which distance women from seeing their place in the formal world ofpolitics (Vickers and Brodie 1982; Lovenduski and Hills 1981; Vicky Randall 1987).However, a recurring theme underlays much of this discourse--that the arena ofgovernment, parties, politicians, and bureaucratic policy has little salience for mostwomen.Gilligan’s theory suggests that the arena may have little salience because itdoes not reflect the behaviour or values that represent women’s moral perspective.This description refers to an occupational segregation that sees 77% of working women grouped into fiveoccupational groups: clerical, service, sales, medicine and health, and teaching (1985 Statistics Canadainformation, cited in S.J. Wilson, “Gender Inequality”, in Understanding Canadian Society (Canada:McGraw-Hill, 1988) 537.Voting with Care 81Rather than emphasizing cooperation, politics is highly competitive. Rather thanresponding to need, politicians respond to power. And, rather than recognizing theconcerns that women have as important, the political world as we know it designatesthese concerns as less important, as the ‘soft’ issues. As Martha Ackelsberg haswritten “If what matters most to me is considered not to be appropriate to ‘politics,’ Iwill tend not to participate in [electoral] political activity (1988, 289). And, ofcourse, if I can find little connection between my so-called “private” interests and the“public” world of politics, Twill have “little incentive to acquire political information,develop a continuing interest in politics and public affairs” or to “try to influenceeither the content or implementation of public policy in the interim betweenelections” (Kornberg and Clarke 1992, 26).Liberal theory is presented as being gender neutral, but as Frazer and hercolleagues argue, it presents an ideal which is more representative of men:It is typical of male life-histories that one makes, or anyway dreams ofmaking, one’s way in the world in accordance with one’s own choice ofcareer; that one develops an autonomous capacity for business andpolitics; that one founds one’s own dynasty, however humble andsuburban; that one regards the outlay of money and work as aninvestment on which a return is to be expected (1992, 4).These authors point out that while these ambitions and values can also be present inwomen’s lives, they are not typical, they are “not the stuff on which a ‘normal’woman draws in constructing a sense of her own identity” (ibid). For instance,Ackelsberg argues that women are alienated by liberal democratic values thatemphasize individuality instead of community, an outlook that misses the reality ofthe concerns that occupy most women, working or domestic, on a daily basis:Women in industrial societies bear primary responsibility for thenurturance of both children and adult males within their households.But that responsibility means that women must be active in the urbanarena considerably beyond the boundaries of the so-called domesticVoting with Care 82sphere. They are the ones who negotiate with landlords, markets,welfare officers, health-care providers, and the like. They are the oneswho must make the adjustment when wages, prices, or rents fluctuate(1988, 302).As Ackelsberg argues, contrary to liberal notions of community, premised onthe combined voice of individuals, women often enter the public arena as members ofsocial networks. For many women, the political “problem” is not one of linking selfto the greater coniniunity, it is finding a way “to link the concerns, visions, andperspectives they share with their neighbors and coworkers to the ‘political system’that stands apart from them and seems to control their lives” (302-3). Denying thepolitical validity of their concerns denies the political nature of their acts and, formany women, acting on behalf of this “private sphere” requires very political acts:from the Filipino women who have become part of the armed resistance of thatcountry, to the “eco-feminists” who led this past summer’s environmental protests onVancouver Island (Bell, Bi) to the women in the End Legislated Poverty coalitionwho protested as part of the Solidarity Movement in 1983. While a common threadin these protests may be community connection, they are further linked by the factthat they all take place outside of the formal political arena. If Gilligan is right, thiswill remain the venue for women’s political involvement until the present systemacknowledges the worth of this other political culture--an acknowledgment that mustgo beyond mere quantitative representation to a blurring of the distinction betweenprivate and public.While we are seeing more women represented in Ottawa, the political agendaremains focused on traditional areas and approached in traditional ways. Manywomen who suggested a caring ethic before gaining office seem to lose sight of thisonce in office, where they become caught up in the world of competitive politics(Gray 1991). In 1985, Penny Kome wrote “lii the political context of the 1980s,feminism’s main impact may be that there’s more room for individual women toVotingwith Care 83succeed in the corporate and political worlds, as long as they conform to the currentstandards of those worlds” (1985, 192). If we recall the evidence from Crown andHeathenngton’s study of affiliation versus achievement orientations in a competitivesports setting, this behaviour is not surprising. In this arena, traditionally associatedwith men and competition, women accepted competition as normal and right--anacceptance that became easier with experience. In politics, while many players mayseek an ideal of cooperation, confrontation and competition remain the norm, andwomen, like their male counterparts, feel they must act in the appropriate competitivemanner. Many also feel that to “play the game” they must demonstrate an emphasison the “hard” issues in order to maintain power; that the important thing now, asformer Tory adviser, Jocelyne Cote-O’Hara said in 1989, is “to have a seat at thetable of the mighty” (Gray 1989). However, if the concerns that reflect a careorientation are considered superfluous to the “real” issues, as defined by those sittingat the table, substantive change is unlikely--a reality graphically illustrated in thefollowing comment by Liberal MP (and current Deputy Prime Minister), SheilaCopps:Picture, if you can, the traditional caucus meeting. There we would be,33 men and me, seated at a long, oblong table, arguing the issues of theday. Little wonder that tile drainage problems got much higher priorityon our agenda than daycare or family violence (Copps 1986, 28).Some suggest that change will come when women represent a critical mass inparliament, and in caucus. However, it is unlikely that numbers alone will make thedifference, nor will simple assertions that women’s voice should be included. A casemust be made that the political discourse will be improved by the inclusion of amoral orientation which emphasizes responsibility and connection to others, andwhich demands a contextual rather than an abstract understanding of the moralVoting with Care 84dilemma. Three such arguments come to mind. The first has to do with the notion ofcommunity implicit in a morality of care.Gilligan argues that the women in her studies present an overarching notion ofthe collective community. As one woman put it:By yourself, there is little sense to things. It is like the sound of onehand clapping, the sound of one man or one woman, there is somethinglacicing. It is the collective that is important to me, and that collectiveis based on certain guiding principles, one of which is that everybodybelongs to it and that you all come from it (1982, 160).Gilligan says that although the world described by men also includes relationshipsand attachments, “no particular person or relationship is mentioned, nor is the activityof relationship portrayed in the context of self-description” (160-1). Males think ofempowerment as individuals, and when autonomy is defined as individual autonomy,and individual achievement is presented as an ideal, concem with relationshipsweakens.While not employing the same conceptual framework as Gilligan, RalphDahrendorf (1979) also describes the modem liberal society as one which hasemphasized individual opportunity at the sacrifice of connection. Dahrendorf talksabout the tension between options and ligatures, and argues that in the liberal questfor expanding options, modem society has lost many of the ligatures, or bonds, thatformerly provided guidance and gave meaning to life. According to Dalirendorfwithout ligatures to provide meaning in the growing awareness that expandingopportunity is a myth, western liberal societies will see a crisis of politicallegitimacy. We are seeing that crisis now. We see it in the election of nontraditionalparties, and we see it in extemal efficacy measures, such as the one cited in thispaper’s care study, which show a lack of trust in political authorities and in thesystem. And, we hear it in populist demands to put the people back into politics andVoting with Care 85to decentralize political decision-making to communities. There seems a felt need fora moral orientation that emphasizes community and connection.Just as a morality of care may have something to add to a mending of themalaise of modern liberal society, so it may contribute to a liberal discourse ofuniversal justice. Gilligan argues for a morality of care as a counterweight to amorality of rights, because she sees an implicit failing in a notion of justice based onimpartiality. Impartiality, in Gilligan’s view, fails to take into account differences inneed. It is a theme also addressed by feminist writers like Iris Marion Young andSeyla Benhabib who argue that equal opportunity and equal treatment are notnecessarily fair.Young criticizes Rawl’s “veil of ignorance” which would remove anydifferentiating characteristics from those in the original position. While Rawis arguesthat doing so would ensure an impartial treatment of others, Young suggests that theother side to this coin is that a mutually disinterested position also “precludes any ofthe participants from listening to others’ expressions of their desires and interests andbeing influenced by them” (1990, 101). Young says the entire notion of impartialityis a myth, that it is impossible to adopt an unsituated moral point of view. Youcannot escape who you are and what your experiences have been. Withoutsubstantive understanding of what has also taken place for others, a designateduniversal principle can only represent the values of those who framed it in the firstplace. Therefore, the “ideal of impartiality” serves the ideological function ofmasking the way in which the particular perspective of the dominant group isperpetuated (97). As Young argues:Where social group differences exist, and some groups are privilegedwhile others are oppressed, this propensity to universalize the particularreinforces that oppression. The standpoint of the privileged, theirVoting with Care 86particular experience and standards, is constructed as normal andneutral (116).35She sees women’s demand for the particulars of the context, and their desire to findsolutions that attempt to satisfy all needs as a positive input into our current justiceparadigm. We are reminded of Johnston’s fable study in which the girls exhaustedattempts to negotiate difference between the porcupine and the moles before invokingniles, while the boys immediately resorted to principles because they saw so muchdifference it seemed to preclude the possibility of talk.Seyla Benhabib describes the view of others prescribed by universal notions ofjustice as that of the “generalized other.” In this view, every individual is seen “as arational being entitled to the same rights and duties we would want to ascribe toourselves” and justice requires that we accord this person equality and reciprocitywith ourselves. But, like Young, Benhabib argues that this is a blinkered view ofjustice in a plural society. This Rawlsian notion demands a consideration based onputting oneself in the other’s shoes and imagining what you would want but, saysBenhabib, without knowledge of the particular history of the other, this can only be a“definitional” understanding, an abstraction of needs. What is required, according toBenhabib, is a perspective that takes the standpoint of the “concrete other” whereinwe view “each and every rational being as an individual with a concrete history,identity and affective-emotional constitution” (281). Benhabib says women are ableto see the “concrete other” and, challenging Kohlbergs relegation of a morality ofcare to the realm of personal moral dilemmas, she argues that an inclusion of themorality of care as essential to a “moral point of view” will lead to a more maturemoral paradigm.And, as Benhabib argues, “liberal theory uses as its paradigmatic case, the experiences of a specific groupwho are invariably white, male adults, who are propertied or at least professional” (1992, 274). In a diversepopulation like Canada’s, this description omits a considerable portion of the population.Voting with Care 87Gilligan and Wiggins (1988) have referred to this as an “empathetic justice,”one which incorporates a sense of “co-feeling” for others. They say co-feeling“implies an awareness of oneself as capable of knowing and living with the feelingsof others, as able to affect others and to be affected by them” In chapter two of thisessay, Brabeck spoke of the distinction between cognitive and affective empathy withthe difference described as the difference between “I feel what you feel” and “I‘know’ what you feel” (1993, 45). In this case, one might conclude that what theRawlsian justice model suggests is a moral consideration based on cognitiveempathy, in contrast to the affective empathy found in women’s moral orientation.Gilligan and Wiggins argue that with this shift in conception of self to others, moralquestions change, leading to “a perspective that turns on questions of inclusion andexclusion, rather than respect for others while inequality persists (119). Theseauthors also challenge Kolberg’s contention that a morality of care is moreappropriate to the personal realm of moral conflict. They say “moral outrage can beprovoked not only by oppression and injustice but also by abandonment or loss ofattachment or the failure of others to respond” (120). In this view, one might supporthigher levels of social welfare assistance not because all children have a right to befed and clothed in our society (although one might also agree with that principle), butbecause these children must not be abandoned, and because one can feel the pain ofthe mother or father who must see their child suffer.Some feminists see a danger in any theory which reinforces stereotypes thatthey see as being the product of oppression. In this view, women are more nurturingand empathetic because, as part of a subordinated group, they are better able toidentify with others’ pain and need. Catherine Mackinnon objects to the verydiscourse of difference as one which perpetuates inequality. “Differences” she says,“are inequality’s post hoc excuse, its conclusory artifact. They are its outcomepresented as its origin, the damage that is pointed to as the justification for doing theVoting with Care 88damage after the damage has been done” (1990, 213). Tn this thinking, any time adifference exists there is the likelihood that a status value will be attached. It is astrong argument and one that is supported by other research,36 but Mackinnon’sargument is based on the notion that discussions of legal equality within a differenceperspective assume a male norm. The issue is not whether differences exist betweenmen and women, but rather how that difference is valued. This is, of course, theissue, and the question remaining is how to remove the status values that now areattached to rights and to care, and how to solve the tension between the two.Resolving this tension will not be easy. While, as the studies point out, manyissues can be viewed from either a “care” or “justice” perspective, the view itself canlead to quite different consequences and can indicate quite different priorities.As Gilligan and Attanucci (1988) point out:detachment, which is the mark of mature moral judgment in the justiceperspective, becomes the moral problem in the care perspective--the failure toattend to need. Conversely, attention to the particular needs andcircumstances of individuals, the mark of mature moral judgment in the careperspective, becomes the moral problem in the justice perspective--failure totreat others fairly, as equals (82).Gilligan and her colleagues argue that both perspectives are necessary for maturemoral decisions, and that “the capacity for love and the appreciation of justice is notlimited to either sex” (Gilligan and Wiggins 1988, 137). That not all females arenurturing and not all males are competitive can be seen in children’s play,37 and36 This argument would certainly be supported by much of the research that has been undertaken in a branchof expectation states theory known as status characteristics theory. Laboratory experiments, such as thatdescribed erlier in this paper, have shown that evidence of difference, especially what are known as diffusestatus values such as gender, class, or race become the basis of performance expectations in task orientedgroups (Berger et al. 1977)While “crash” games dominated boys’ play at the daycare, there were boys who certainly did not “kill”their toys. One boy, in a different dinosaur group than the one described earlier, asked his dinosaur, “Oh hi,dinosaur, what are you doing?” I am also reminded that when my son was small, he would play with G.I. Joewar toys, but he would spend an hour setting up the base, then there would be five minutes of battle, then thesoldiers would all have coffee after which he would put them to bed with blankets I had made for them.Definitely two orientations here.Voting with Care 89Gilligan suggests that if we wish to encourage an integrated moral orientation, this isthe age at which we should begin. Certainly as Flanagan and Jackson point out, andas Margaret Mead’s study would support, “there is nothing necessary (although theremay be biological and social pressures in a certain direction) about the way wearrange nurturance, or about the particular ways parents treat their male and femalechildren, and thus the story is not required to turn out exactly the way it does now”(1993, 77). The trick, of course, is having all members of society--males andfemales--agree that the proposed orientation is desired.In their study of sex differences in fourteen countries, Williams and Bestfound evidence that when it comes to sex roles, women in most countries tended tobe more “egalitarian” in their views about sex-role relationships than men, whotended to be more traditional. The authors suggest that “this is not a surprisingfinding in light of the fact that the traditional ideology, which assigns greaterimportance and/or power to men, would naturally be viewed as being more agreeableto men than to women” (1990, 157). This same pattern, in which males appear moreresistant to status change than females, emerges in some of the research on statuscharacteristics theory earlier cited (see pages 54, 55). It would appear that whilewomen may be ready to see themselves as equal, men have not quite reached thatpoint. Understandably, there is resistance to adjusting the current paradigms thatguide our societies as long as those paradigms maintain the status quo of those forwhom they serve.There is also, of course, a reluctance by some to alter a norm which appears tohave societal consent. Kohlberg, who has continued to priorize a rights perspectiveas being the only truly “moral point of view”, defends his position on the basis thatthe moral philosophical paradigm that informs his theory is, in fact, representative ofsociety (1984). However, Kohlberg points out that moral decision-making takesplace within the context of the group, and that changes in that structure can affect theVoting with Care 90content of moral choice. Our society has changed from the time when the currentparadigm was framed The “public sphere” is now one increasingly populated bywomen, and by other groups who are also demanding a voice; the private sphereincreasingly shared by men. And more and more, there is a demand for doing awaywith the distinction entirely. Flanagan and Jackson (1987) suggest not only that themoral voices argued by Gilligan and Kohlberg need further refinement, but that eventhese two voices may not adequately fill an existing moral philosophical gap. It maybe that the value of Gilligan’s argument lays as much in exposing the shortcomingsof the current discourse on moral ethics as it does in raising the issue of whetherwomen’s particular voice is included in the discourse.91BIBLIOGRAPHYAckelsberg, Martha. 1988. “Communities, Resistance, and Women’s Activism:Some Implications for a Democratic Polity.” Women and the Politics ofEmpowerment Ed. Ann Bookman and Sandra Morgen. Philadelphia: TempleUniversity Press, 297-3 13.Aristotle. 33 5-323 B.C. The Politics. Trans. T.A. Sinclair. Rev, and Re-presentedby Trevor J. Saunders. London: Penguin Books, 1981.Baron, Robert A., and Domi Byrne. 1991. Social Psychology: UnderstandingHuman Interaction. 6th ed. Boston: Allyn and Bacon.Bashevkin, Sylvia. 1989. “Political Parties and the Representation of Women.”Canadian Parties in Transition. Ed. Alain G. Gagnon and A. Brian Tanguay.Canada: Nelson, 446-45 9.Baumrind, Diana. 1986. “Sex Differences in Moral Reasoning: Response toWalker’s (1984) Conclusion that There Are None.” An Ethic of Care:Feminist and Interdisciplinary Perspectives. New York: Routledge, 177-192Belenky, Mary Field, and Blythe McVicker Clinchy and Nancy Rule Goldgerger andJill Mattuck Tarule. 1986. Women’s Ways of Knowing: The Developmentof Self, Voice, and Mind. USA: Basic Books, 1-100.Beltrame, Julian. 1993. “If they did a poli on polls, what would you say 19 times outof 20?” Vancouver Sun 16 April, A4.Benhabib, Seyla. “The Generalized and the Concrete Other.” Ethics: A FeministReader. (ed.) Elizabeth Frazer and Jennifer Hornsby and Sabina Lovibond.Oxford: Blackwell, 1992 267-300Berger, Joseph, and M. Hamit Fisek and Robert Z. Norman and Morris Zelditch, Jr.1977. “Introduction.” Status Characteristics and Social Interaction. NewYork: Elsevier, 3-11.Brabeck, Mary. 1993. “Moral Judgment: Theory and Research onDifferencesBetween Males and Females.” An Ethic of Care: Feminist andInterdisciplinary Perspectives. New York: Routledge, 3 3-48.92Brodie, Janine, Asst. Celia Chandler. 1991. “Women and the electoral Process inCanada.” Women in Canadian Politics: Toward Equty in Representation.Ed. Kathy Megyery.Burt, Sandra. 1986. “Women’s Issues and the Women’s Movement in Canada Since1970.” The Politics of Gender, Ethnicity and Language in Canada. (researchcoordinators) Alan Cairns and Cynthia Williams. Toronto: University ofToronto Press, 111-137.The Candidate. 1972. Dir. Michael Ritchie. With Robert Redford. Warner Bros.Carroll, Susan J. 1988. “Women’s Autonomy and the Gender Gap: 1980 and 1982.”The Politics of the Gender Gap: The Social Construction of PoliticalInfluence. Newbury Park: Sage Publications, 236-257.Ceci, Stephen J., and Edward L. Kain. 1982. “Jumping on the Bandwagon with theUnderdog: The Impact of Attitude Polls on Polling Behavior.” Public OpinionQuarterly 46, 228-242Chowdorow, Nancy. 1974. “Family Structure and Feminine Personality.” Woman,Culture and Society. (ed.) Michelle Zimbalist Rosaldo and Louise Lamphere.Stanford: Stanford University Press, 43-66.Colby, Anne, and William Damon. 1987. “Listening to a Different Voice: AReview of Gilligan’s In a Different Voice.” The Psychology of Women:Ongoing Debates. New Haven: Yale UP, 321-329.Copps, Sheila. 1986. Nobody’s Baby. Toronto: Deneau.Crown, Judith, and Laurie Heatherington. 1989. “The Costs of Winning? The Roleof Gender in Moral Reasoning and Judgments About Competitive AthleticEncounters” Journal of Sport and Exercise Psychology II, 28 1-289.Dahrendorf, Raif. 1979. Life Chances: Approaches to Social and Political Theory.London: Weidenfeld and Nicolson.“Election Sets Record for Female MPs.” 1993. Vancouver Sun 27 October, A6..Flanagan, Owen and Kathryn Jackson. 1993. “Justice, Care, and Gender: TheKohlberg-Gilligan Debate Revisited.” An Ethic of Care: Feminist andInterdisciplinary Perspectives. (ed.) Mary Jeanne Larrabee. New York:Routledge, 69-84.93Foschi, Martha. 1992. “Gender and Double Standards for Competence.” GenderInteraction, and Inequality. (ed.) C.L. Ridgeway. New York: SpringerVollag, 181-207.Frazer, Elizabeth, and Jennifer Hornsby and Sabina Lovibond. 1992. “Introduction.”Ethics: A Feminist Reader. (ed.) Elizabeth Frazer and Jennifer Homsby andSabina Lovibond. Oxford: Blackwell, 1-19.Friedman, William J., and Amy B. Robinson, and Britt L. Friedman. “SexDifferences in Moral Judgments? A Test of Gilligan’s Theory.” Psychologyof Women Ouarterly 11:37-46Froese, Patricia and Sharon Froese Nielson. 1986. “Gender and Attitudes Towardthe Peace Movement and Nuclear Disarmament.” Atlantis 12:1, fall.Freud, Sigmund. 1905. Three Essays on the Theory of Sexuality. (trans. and ed.)James Strachey. New York: Basic Books, 1962.Gallup, George. 1976. The Sophisticated Poll Watcher’s Guide. Princeton OpinionPress, 1-41.Gilligan, Carol. 1982. In a Different Voice: Psychological Theory and Women’sDevelopment. Cambridge: Harvard University Press.---, and Jane Attanucci. 1988. “Two Moral Orientations.” Mapping the MoralDomain: A Contribution of Women’s Thinking to Psychological Theory andEducation. Ed. Carol Gilligan and Janie Victoria Ward and Jill McLeanTaylor, with Betty Bardige. Cambridge: Harvard University Press, 73-86.and Grant Wiggins. 1988. “The Origins of Morality in Early ChildhoodRelationships.” Mapping the Moral Domain: A Contribution of Women’sThinking to Psychological Theory and Education. Ed. Carol Gilligan andJanie Victoria Ward and Jill McLean Taylor, with Betty Bardige. Cambridge:Harvard University Press, 111-138.Gray, Charlotte. 1989. “Beyond the New ‘F’ Word.” Saturday Night 104:4 April,17-20.Greeno, Catherine G. and Eleanor E. Maccoby. 1993. “How Different Is the‘Different Voice’? An Ethic of Care: Feminist and InterdisciplinaryPerspectives. (ed.) Mary Jeanne Larrabee. New York: Routledge, 193-8.94Greenstein, Fred I. 1965. Children and Politics. New Haven: Yale University Press.Hitchens, Christopher. 1992. “Voting in the Passive Voice: What Polling Has Doneto American Democracy.” Harper’s Magazine April, 45-52.Johnston, D. Kay. 1988. “Adolescents’ Solutions to Dilemmas in Fables: Two MoralOrientations--Two Problem Solving Strategies.” Mapping the Moral Domain:A Contribution of Women’s Thinking to Psychological Theory and Education.(ed.) Carol Gilligan, Jame Victoria Ward, and Jill McLean Taylor.Cambridge: Harvard University Press, 49-71.Johnston, Richard, and Andre Blais and Henry E. Brady and Jean Crete. 1992.Letting the People Decide: Dynamics of a Canadian Election. Montreal:McGill-Queen’s University Press.Kay, Barry J., and Ronald D. Lambert and Steven D. Brown and Janes E. Curtis.1987. “Gender and Political Activity in Canada, 1965-1984.” CanadianJournal of Political Science XX:4 December, 851-863.Kemp, Leslie. 1984. “Overview: Women and Welfare.” Women, the Law and theEconomy. (ed.) E. Diane Pask and Kathleen E. Mahoney and CatherineBrown. Toronto: Butterworths, 143-144.Kohlberg, Lawrence. 1976. “Moral Stages and Moralization: The CognitiveDevelopmental Approach.” Moral Development and Behavior: Theory,Research, and Social Issues. (ed.) Thomas Lickona. New York: Bolt,Rinehart and Winston, 31-53.---, and Charles Levine and Alexandra Hewer. 1984a. “The Current Formulation ofthe Theory.” in Kohlberg, Lawrence, The Psychology of Moral Development:The Nature and Validity of Moral Stages. Cambridge: Harper & Row, 212-319.---, and Charles Levine and Alexandra Hewer. 1984b. “Synopses and DetailedReplies to Critics” in Lawrence Kohlberg, The Psychology of MoralDevelopment: The Nature and Validity of Moral Stages. Cambridge: Harper&Row, 338-370.Kopinak, Kathryn. 1987. “Gender Differences in Political Ideology in Canada.”Canadian Review of Sociology and Anthropology 24:1, 232-3 8.95Kome, Penny. 1985. Women of Influence: Canadian Women and Politics. Toronto:Doubleday.Komberg, Allan and Harold D. Clarke. 1992. Citizens and Commumtv: PoliticalSupport in a Representative Democracy. Cambridge: Cambridge UniversityPressLaponce, J.A. 1983? “The Effect of Polls on Elections: Can British Columbia BeUsed as a Laboratory?” Unpublished essay.1966. “An Experimental Method to Measure the Tendency to Equibalance in aPolitical System.” American Political Science Review LX:4 December, 982-993.Lasswell, Harold. 1930. Psychopathology and Politics. New York: Viking Press,1960.Lawrence J. Walker. “Sex Differences in the Development of Moral Reasoning: ACritical Review.” 1984. Child Development 55:677-691.Lever, Janet. “Sex Differences in the Games Children Play.” Social Problems 23,478-487.Lickona, Thomas. 1976. “Critical Issues in the Study of Moral Development andBehaviour.” Moral Development and Behaviour: Theory. Research. andSocial Issues. (ed.) Thomas Lickona. New York: Holt, Rinehart and Winston,3-27.Lovenduski, Jone and Jill Hills. 1981. “Introduction.” The Politics of the SecondElectorate: Women and Public Participation. Ed. Jom Lovenduski and JillHills. London: Routledge and Kegan Paul, 1-7.Lyons, Nona Plessner. 1988. “Two Perspectives: On Self, Relationships, andMorality.” Mapping the Moral Domain: A Contribution of Women’s Thinkingto Psychological Theory and Education. (ed.) Carol Gilligan, Janie VictoriaWard, and Jill McLean Taylor. Cambridge: Harvard University Press, 21-47.McCormack, Thelma. 1975. “Toward a Nonsexist Perspective on Social andPolitical Change.” Another Voice: Feminist Perspectives on Social Life andSocial Science. New York: Anchor Books, 1-33.96MacKimion, Catharine A. 1990. “Legal Perspectives on Sexual Difference.”Theoretical Perspectives on Sexual Difference. (ed.) Deborah L. Rhode. NewHaven: Yale University Press, 2 13-225.McCormack, Thelma. 1989. “Examining the Election Entrails: Whatever Happenedto the Gender Gap?” This Magazine MarchlApril, 31-35.Mill, John Stuart 1859. On Liberty, in Utilitarianism, Liberty, RepresentativeGovernment. London: Everyman’s Library, 1972:69-188.1861. “On the Extension of the Suffrage.” Considerations on RepresentativeGovernment, in Utilitarianism, Liberty, Representative Government. London:Everyman’s Library, 1972:299-317.1869. “The Subjection of Women.” Feminist Frameworks: AlternativeTheoretical Accounts of the Relations Between Women and Men. Ed. AlisonM. Jaggar and Paula S. Rothenberg. New York: McGraw-Hill, 104-114.1861-3. Utiuitariamsm, in Utilitarianism, Liberty, Representative Government.London: Everyman’s Library, 1972:1-69.Miller, Arthur. 1988. “Gender and the Vote: 1984.” The Politics of the GenderGap: The Social Construction of Political Influence. Newbury Park: SagePublications, 25 8-282.Miller, Jean Baker. 1976. Toward a New Psychology of Women. Boston: BeaconPress.Morgen, Sandra, and Aim Bookman. 1988. “Rethinking Women and Politics: Anhitroductory Essay.” Women and the Politics of Empowerment Ed. AnnBookman and Sandra Morgen. Philadelphia: Temple University Press, 3-29.Norris, Pippa. 1988. “The Gender Gap: A Cross-National Trend?” The Politics ofthe Gender Gap: The Social Construction of Political Influence. NewburyPark: Sage Publications, 217-234.Pratt, Michael W., and Gail Golding and William Hunter and Rosemarie Sampson.1988. “Sex Differences in Adult Moral Orientations.” Journal of Personality56:2, 373-391.97Pugh, M.D. and Ralph Walirrnan. 1983. “Neutralizing Sexism in Mixed-SexGroups: do Women Have to be Better than Men?” American Journal ofSociology 88:4, 746-762.Rawis, John. 1971. “Justice as Fairness: Political not Metaphysical.” PhilosophyandPublic Affairs 14, 1985:223-251.Rousseau, Jean Jacques. 1762. Emile. London: Dent, 1963:321-341.Sears, David 0. 1988. “College Sophomores in the Laboratory: Influences of aNarrow Data Base on Psychology’s View of Human Nature.” Readings inSocial Psychology: Classic and Contemporary Contributions. (ed.) LetitiaAnne Peplau and David 0. Sears and Shelley E. Taylor and Jonathan L.Freedman. Englewood Cliffs: Prentice Hall, 3 13-337.Stewart, Penni A., and James C. Moore Jr. 1992. “Wage Disparities andPerformance Expectations.” Social Psychology Quarterly 55:1, 78-85.Stoller, Robert J. 1964. “A Contribution to the Study of Gender Identity.”International Journal of Psycho-Analysis 45, 220-226.Vickers, Jill McCalla. 1989. “Feminist Approaches to Women in Politics.” Beyondthe Vote: Canadian Women and Politics. Toronto: University of TorontoPress, 16-3 6.Vickers, Jill McCalla, and Jamne Brodie. 1982. “Canadian Women in Politics: AnOverview.” The CRIAW Papers No. 2. Ottawa: CRIAW.Wagner, David G., and Rebecca S. Ford and Thomas W. Ford. 1986. “Can GenderInequalities be Reduced.” American Sociological Review 51, 47-61.Wallace, Bruce, and E. Kaye Fultonand Nancy Wood and Christina Wolamuk. 1993.“Playing Gender Politics.” Maclean’s 106:40 4 October, 16-19.Wearing, Peter and Joseph Wearing. 1991. “Does Gender Make a Difference inVoting Behaviour?” The Ballot and its Message: Voting in Canada. (ed.)Joseph Wearing. Toronto: Copp Clark Pitman, 341-3 50.Williams, Juanita H. Psychology of Women: Behavior in a biosocial Context. NewYork: W.W. Norton & Co., 198798Williams, John E. and Deborah L. Best. Sex and Psyche: Gender and Self ViewedCross-Culturally. Newbury Park: Sage Publications, 1990.Young, Iris Marion. 1990. “The Ideal of Impartiality and the Civic Public.” Justiceand the Politics of Difference. Princeton: Princeton University Press, 96-12 1.99APPENDIX A: KOBLBERG’S SIX STAGE MODELPRECONVENTIONAL 1. Heteronomous MoralityAn egocentric point of view, notconsidering the interests of others orrecognizing that those interests might bedifferent from their own.2. Individualism, Instrumental Purpose, andExchangeAn awareness of rights beyond your ownthat might be conflicting. Rights arerelative.CONVENTIONAL 3. Mutual Interpersonal ExpectationsRelationships, and interpersonal ConformityA perspective based on sharedrelationships. Feelings, agreements,expectations of others take primacy overindividual interests. Personal notgeneralized in orientation.4. Social System and conscienceSees the societal point of view overinterpersonal agreements or motives.POSTCONVENTIONAL 5. Social Contract or Utility and Individual RightsA “prior to society” perspective - values andrights take precedence over socialattachments and contracts. Considers moraland legal points of view in a context ofobjective impartiality.6. Universal Ethical PrinciplesA moral point of view. Belief in thevalidity of universal moral principles ofjustice based on equality of rights andrespect for individuals.100APPENDIX B - FRIEDMAN ET AL.: 12 ITEM TYPESG ITEMS (Gilligan)Is Heinz likely to risk getting shot as a burglar or going to jail for the chancethat stealing the drug might help?Is this likely to weaken or strengthen the relationship between Heinz and hiswife?How old are Heinz and his wife, and do they have children who could helpthem raise money?Heinz’s willingness to substitute himsclf for his wife and bear the brunt ofsociety’s laws.Does Heinz have a responsbility to care for his wife?Which outcome will cause the least hurt for all of the people involved?K ITEMS (Kohlberg)Whether there is a moral code to which all individuals should adhere.What values are going to be the consistent basis for governing how people acttowards each other?The relative weights of life and property.Whether the value of life and the equality of human rights are prior to law.Whether the druggist, in exercising his individual rights, infringes on the rightsof others.Whether Heinz has a right to make a decision based on his own system ofvalues.101APPENDIX C: THE QUESTIONNAIREIn the following questionnaire we are interested in the political choices people make.We would ask you to answer each question before going on to the next, and please beassured there are no right or wrong answers and your responses are absolutelyconfidential. May we thank you in advance for your cooperation in completing thequestionnaire.#1You are about to make a voting choice between two candidates, let’s call themX and Y, to whom you feel equally committed. As you stand in the pollingbooth pondering your choice, you remember that just before you left yourhome, you heard a radio news broadcast reporting the results of a publicopinion poll in which X and Y were given the following popularity ratings:CandidateX 47%Candidate Y 35%Who will you vote for, X or Y?#2Now let’s try another scenario. You’re once again in the polling booth about tomake a choice between Candidate X and Candidate Y, and again, you’re awarethat the most recent poll results have given X the greatest share of a popularvote. However, this time you are aware that X and Y take significantlydifferent stands on the following issue:Candidate X 47%(believes the poor get too many handouts,and must learn to be more self-reliant. Xsuggests the government should endeavor tocut back on welfare costs.)Candidate Y 35%(believes the government should do more tohelp the poor even if it means higher taxes)Who will you vote for, X or Y?102We are also interested in how you feel about our current political system and yourplace in it. In general, how would you respond to the following statements?#3 Generally, those elected to parliament in Ottawa soon lose touch withthe people.strongly agreesomewhat agreedon’t knowsomewhat disagreestrongly disagree#4 I don’t think that the federal government cares much what people like methink.strongly agreesomewhat agreedon’t knowsomewhat disagreestrongly disagree#5 Sometimes, politics and government in Ottawa seem so complicated that aperson like me can’t really understand what’s going on.strongly disagreesomewhat disagreedon’t knowsomewhat agreestrongly agree#6 People like me don’t have any say about what the federal governmentdoes.strongly agreesomewhat agreedon’t knowsomewhat disagreestrongly disagree103Before we finish, we need to know a little bit about your background,#7 This class is (political science, social psychology, sociology, etc.)#8 Your major is#9 You are malefemale#10 Your ageis#11 With what ethnic group does your family identify? (e.g. Italian,Greek, Indian, First Nations, Chinese, British, Polish)#12 With what ethnic group do you identify?#13 Which federal political party do you prefer?LiberalConservativeN.D.P.ReformOtherNoneDon’t Know104And finally, we’d like to know what you thought of, and remember about, thequestions and your answers.#14 (Do not look back at the first page when answering this question).In question #1, what were the poil results forthe top rated candidatethe lower rated candidate#15 If you voted for the top rated candidate on question #1, can youremember why you did so? It is your very first reaction that we areinterested in. (you may look back at question #1). Please state yourreason.#16 If you voted for the lower rated candidate on question #1, would youstate your reason for doing so.#17 What, would you say, is the information we are seeking in this questionnaire?#18 Some people have expressed difficulty in answering question #1. In theinterest ofimproving our questionnaire, we would like to know if you share thisexperience. If you had difficulty answering #1, would you please state whatthat difficulty was?#19 On a scale where 1 represents “not at all difficult” and 5 represents“very difficult,” how would you rate your level of difficulty withquestion #1? (please circle the appropriate number)1 2 3 4 5


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items