¹ Unlike most political strategies, separatism has never been formally adopted; there are no auth oritative structures in the women's movement that could do so. It has spread, by means of argument, imitation and trial and error. For Britain, see Coote, Anna and Campbell, Beatrix, Sweet Freedom (Oxford: Basil Blackwell, 1982), pp. 33–5.Google Scholar

² Dennis, Jack, ‘Future Work on Political Socialization’ in Dennis, , ed., Socialization to Politics (New York: John Wiley, 1973), p. 494.Google Scholar

³ Hyman, Herbert, Political Socialization (Glencoe, Ill.: The Free Press, 1959), p. 46Google Scholar. See also Jennings, M. Kent and Niemi, Richard, ‘The Transmission of Political Values from Parent to Child’, American Political Science Review, LXII (1968), 169–84.CrossRefGoogle Scholar

⁴ Hyman, , Political Socialization, pp. 74–5Google Scholar; Jennings, and Niemi, , ‘The Transmission of Political values’, p. 183.Google Scholar

⁵ Almond, Gabriel and Verba, Sidney, The Civic Culture (Princeton, N.J.: Princeton University Press, 1963), p. 324.CrossRefGoogle Scholar

⁶ Dennis, , Socialization to Politics, p. 494.Google Scholar

⁷ Pateman, Carole, ‘The Civic Culture: A Philosophic Critique’, in Almond, and Verba, , eds., The Civic Culture Revisited (Boston, Mass.: Little, Brown, 1980), p. 72.Google Scholar

⁸ See, for example, Easton, David and Dennis, Jack, ‘The Child's Acquisition of Regime Norms: Political Efficacy’, American Political Science Review, LXI (1967), 25–38, pp. 36–8CrossRefGoogle Scholar; Iglitzin, Lynne, ‘The Making of the Apolitical Woman: Femininity and Sex-Stereotyping in Girls’, Jaquette, Jane, ed., Women in Politics (New York: John Wiley, 1974), pp. 25–36Google Scholar; Dowse, Robert and Hughes, John, ‘Girls, Boys and Polities’, British Journal of Sociology, XXII (1971), 53–67CrossRefGoogle Scholar; Welch, Susan, ‘Women as Political Animals? A Test of Some Explanations of Male-Female Political Participation Differences’, American Journal of Political Science, XXXI (1977), 711–30CrossRefGoogle Scholar; Welch, Susan, ‘Sex Differences in Political Activity in Britain’, Women and Politics, 1, no. 2 (1980), 29–46.CrossRefGoogle Scholar

⁹ Sapiro, Virginia, The Political Integration of Women: Roles, Socialization and Politics (Urbana: University of Illinois Press, 1984).Google Scholar

¹⁰ Chapman, Jenny, ‘Marital Status, Sex and the Formation of Political Attitudes in Adult Life’, Political Studies, XXXIII (1985), 592–609.CrossRefGoogle Scholar

¹¹ Dunleavy, Patrick, ‘The Urban Basis of Political Alignment: Social Class, Domestic Property Ownership and State Intervention in Consumption Processes’, British Journal of Political Science, IX (1979), 409–43, pp. 412–13.CrossRefGoogle Scholar

¹² Franklin, Mark and Page, Edward, ‘A Critique of the Consumption Cleavage Approach to British Voting Studies’. Political Studies, XXXII (1986), 521–36, p. 528.Google Scholar

¹³ Parsons, Talcott, Politics and Social Structure (London: Collier-Macmillan, 1969), p. 13.Google Scholar

¹⁴ Dawson, Richard and Prewitt, Kenneth, Political Socialization (Boston, Mass.: Little, Brown, 1969), pp. 63–80Google Scholar. These authors cite three indirect and four direct methods of political learning; see also Dennis, , Socialization to Politics, pp. 2–27Google Scholar, who suggests nine agencies of socialization, including governments.

¹⁵ Dawson, and Prewitt, , Political Socialization, pp. 15–24.Google Scholar

¹⁶ Mead, George Herbert, Mind, Self and Society (Chicago: University of Chicago Press, 1934).Google Scholar

¹⁷ See McWilliams, Nancy, ‘Contemporary Feminism, Consciousness-Raising and Changing Views of the Political’Google Scholar, in Jaquette, , Women in Politics, pp. 162–4.Google Scholar

¹⁸ See, for example, Verba, Sidney and Nie, Norman, Participation and America: Political Democracy and Social Equality (New York: Harper and Row, 1972), p. 12 and passim.Google Scholar

¹⁹ Githens, Marianne and Prestage, Jewel, eds., A Portrait of Marginality; The Political Behavior of the American Woman (New York: David MacKay, 1977)Google Scholar; Sapiro, , Political Integration of Women, pp. 5–8.Google Scholar

²⁰ For sympathetic discussion of the pressures involved, see Vallance, Elizabeth, Women in the House (London: Athlone Press, 1979), pp. 83–96Google Scholar and Bozeman, Barry, Thornton, Sandra and McKinney, Michael, ‘Continuity and Change in Opinion about Sex Roles’Google Scholar, in Githens, and Prestage, , eds., A Portrait of Marginality, pp. 38–65.Google Scholar

²¹ On which the dichotomy and incompatibility of gender roles was based in the first place, forcing contemporary women to choose between motherhood (female sex role) and career (male gender role), the latter presupposing the existence of low-cost domestic and childcare services (female gender role). The classic account of out-group assimilation, in Dahl, Robert, Who Governs? (New Haven, Conn.: Yale University Press, 1961)Google Scholara, is suggestive of two other obstacles to women's assimilation. Like blacks, the least assimilated of the out-groups he considered, women wear the visible badge of difference. Unlike most other out-groups, blacks included, they are dispersed among the in-group rather than concentrated in ghettoes as a convenient constituency for mobilization by ambitious politicians.

²² Such as Christine de Pizan, Mary Wollstonecraft and the early American feminists with atypical education. See Warner, Marina's ‘Introduction’Google Scholar to de Pizan, Christine, The Book of the City of Ladies (London: Pan Books, 1983), p. xixGoogle Scholar and Rossi, Alice, The Feminist Papers (Toronto: Bantam, 1974), p. 16 and passim.Google Scholar

²³ The emergence of the women's movement itself (and with it the theory of consciousnessraising) illustrates this thesis. It originated in the United States, among suburban groups of ‘coffeeklatching housewives’ and ‘in the ferment of the early New Left, where activist women began to share their perceptions of a common situation … In both places the realizations that one's problems were shared, that individual dissatisfactions had social bases … struck many with the force of a revelation’; see McWilliams, , ‘Contemporary Feminism’, p. 159Google Scholar. Informal groups had also existed among women civil rights activists for several years before 1967, when they developed ‘a determined, if cautious, continuity and began to expand’; see Freeman, Jo, ‘The Women's Liberation Movement: Its Origins, Structure, Activities and Ideas’, in Freeman, Jo, ed., Women: A Feminist Perspective (Palo Alto, Calif: Mayfield, 1984), p. 544.Google Scholar

²⁴ One of the male strategies of disempowerment analysed in Barth, Else, ‘Women and Empowerment’Google Scholar, Keynote Address, Second Interdisciplinary Congress on Women, Groningen, 1984Google Scholar. The contemporary use of this strategy is by no means confined to left-wing socialists, as some might think. My interviews with Scottish politicians of both sexes have found it currently being adopted enthusi astically by men of all shades of opinion within the Labour party: the attempt to set up black sections in the Labour movement in England is usually given as an example of how ‘divisive’ it would be for women to meet on their own. A few SNP women say they have encountered the strategy in their party too, used to devastating effect.

²⁵ Parry, Geraint and Day, Neil, ‘Political Participation in Britain: Issue Agendas and Political Action’, paper presented to the Political Studies Association, Nottingham, 1986.Google Scholar

²⁶ With a response rate of 85 per cent overall and 87 per cent among women candidates.

²⁷ A more sophisticated typology using a third variable, the perception of subgroups of women with particular political needs, to make a subtle distinction within the unpoliticized Type B, has been reported elsewhere (see asterisked first footnote) and will be used in a forthcoming analysis.

²⁸ For a review of these findings see Randall, Vicky, Woman and Politics (London: Macmillan, 1982), pp. 35–68CrossRefGoogle Scholar and Sapiro, , Political Integration of WomenGoogle Scholar, which offers additional data as well.

²⁹ Feldberg, Roslyn and Glen, Evelyn Nakano, ‘Male and Female: Job Versus Gender in the Sociology of Work’, in Silthanen, Janet and Stanworth, Michelle, eds., Women and the Public Sphere (London: Hutchinson, 1984), p. 35.Google Scholar

³⁰ A connection between attitude and awareness of the women's movement in the United States is established by Soule, John and McGrath, Wilma, ‘A Comparative Study of Male-Female Political Attitudes at Citizen and Elite Levels’Google Scholar, in Githens, and Prestage, , A Portrait of Marginality, p. 190.Google Scholar

³¹ On such key variables as range of community type, level of council house tenancy, political history and percentage Labour vote.

³² Two cases which occurred in both samples have been weighted by 100 per cent to preserve the representative character of the main sample; the effect on the coefficients reported below is negligible. It should be pointed out that the representative character of this sample precludes comparison with some earlier studies of female elites. The most often cited American study is Kirk-patrick, Jeanne's Political Woman (New York: Basic Books, 1984)Google Scholar, but this work is not based on a systematic sample. Its subjects were fifty women who had been nominated by branches of three well-known women's organizations and who additionally had both the resources and the motivation to attend a three-day conference on women and politics at Rutgers University.

³³ Unless otherwise stated, all coefficients cited in this article are significant at 0.05.

³⁴ To avoid multicollinearity one of the party dummy variables must not be included in the regression analyses. None of the four major party affiliations was significantly correlated with either experience or orientations, and experiment showed negligible effects whichever was excluded. In all the regressions reported in this article the party affiliations included are Labour, Conservative and minor party (the only significant correlate of the dependent variables). ‘Minor party’ refers to candidates for the Ecology party, a Progressive and, for the sake of convenience, the only Independent in thisdataset.

³⁵ Since some of these women's role-models included very rare cases of successful women politicians, prominent but very isolated in a man's world, this is perhaps not as surprising as it might seem.

³⁶ The proportions were 53.0 per cent in the single-district study and 54.7 per cent in the random sample.

³⁷ This proportion is slightly higher than given in an earlier, unpublished paper (see asterisked first footnote) due to the recoding of one case. This was a women intending to give up ‘polities’ and concentrate on improving the position of women in her profession (architecture). Having no self-styled ‘political’ aim for women, she was originally coded as Type B but this coding has been aban doned here as pedantic. This is the only case of a coding decision running counter to the respon dent's own conclusion.

³⁸ The ‘women’ questions were not raised until well into the interview when the respondent was likely to be at ease. They were paired where possible with similar questions not specifying women and were divided into two batches to give the interviewee a second bite at the cherry.

³⁹ Who got round the inconsistency by defining a ‘women's group’ as something other than the one(s) they belonged to.

⁴⁰ The exception being a woman of so marginal a state of mind that she proffered and retracted her aim no less than three times and was finally asked to code her own response.

⁴¹ Nevertheless, the Young Sample contained one of the two most acute cases of marginal cross-pressures encountered in this research. This was a candidate who could not bring herself to reveal that feminism was her main motivation for being involved in politics at all until nearly the end of an increasingly uneasy interview.

⁴² Housing issues included the problems of elderly women, battered women and single parents, questions of joint tenancy and homelessness and the repair and modernization of council houses. The latter overlapped with concern about the exclusion of women from the planning and design processes and from related professions; this was connected also to the issues of town planning, industrial zoning and transport. Other issues ranged from national policies of taxation, welfare and disarmament to very local problems such as hospital closures, the need for neighbourhood watch committees and the discriminatory employment practices of specific District Councils. General aims among the Young Sample focused on power rather than representation and on equality in education as much as employment. Specific aims were presented in greater numbers and were more likely to be formulated in recognizably feminist terms. The proportion of the young with political aims for women was not only greater but seemingly more strongly motivated as well, although this could of course be due simply to the greater optimism and lesser experience of youth.

⁴³ Several, in particular politically-conscious councillors of all parties, noted the peculiar relevance of District Council functions to the everyday lives of most women, a point that was often made by male councillors too. The women (but not the men) tended also to stress the irony that it is men and male values that dominate local politics, a view that was usually related to the belief that the political perspectives and/or interests of women are different from those of men.

⁴⁴ The only significant ‘political’ correlate of politicization was membership of the tiny group of minor-party affiliates; this was negative and it was revealing that three out of four cases had no experience of women's groups.

⁴⁵ This woman, a trade-union official, did not want to see women's sections revived in the Labour party precisely because they would benefit women; discussion revealed that her self-image was too bound up in her own successful struggle against the odds to favour equality even in the distant future!

⁴⁶ See fn. 34.

⁴⁷ Some readers may care to know that the method used in the analyses reported here and in Tables 2 and 4 is stepwise regression. Extensive experiment with different regression methods produced highly convergent results in each sample, irrespective of the method used. In every case, the results were dominated by the separate experience variable, in terms of its path coefficient, slope, significance and proportion of variance explained. Within samples, there was very little variation even in the exact size of the coefficients. However, the stepwise method produced the most comparable set of equations. This is because the Young Sample is so small that other methods calculate statistically significant coefficients for variables that affect only one or two people, reducing both the comprehen-sibility and the comparability of the results.

⁴⁸ Franklin, Mark, The Decline of Class Voting (Oxford: Clarendon Press, 1985), p. 88, Figure 4.3.Google Scholar

⁴⁹ Political groups included women's sections, committees, etc., traditional or new, formally or informally constituted, in political parties and trade unions; political action and pressure groups; overtly feminist groups and any other groups connected however loosely with the women's movement. Women were coded as having political experience if they had ever belonged to any (all-women) group of this kind, irrespective of whether they had also belonged to a ‘non-political’ group or groups. Women with ‘non-political experience only’ had, as the label suggests, never belonged to any group categorized as ‘political’.

⁵⁰ It does not reflect any difference in the effect of prior feminist socialization on the two kinds of separate experience, for regression analysis found that there was no significant effect in either case.