¹ See Pizzorno, Alessandro, ‘The Individualistic Mobilization of Europe’, in ‘A New Europe?’, Daedalus, Winter 1964, pp. 199–224.Google Scholar

² Berger, Suzanne, ‘Politics and Antipolitics in Western Europe in the Seventies’, Daedalus, Winter 1979, pp. 27–50.Google Scholar

³ Tilly, Charles, From Mobilization to Revolution (New York: Prentice Hall, 1978)Google Scholar and The Contentious French (Cambridge, Mass.: Harvard University Press, 1986).Google Scholar

⁴ See Crozier, Michel, Huntington, Samuel and Watanuki, J., The Crisis of Democracy (New York: New York University Press, 1975).Google Scholar

⁵ See Tarrow, Sidney, Democracy and Disorder: Protest and Politics in Italy, 1965–1975 (Oxford: Oxford University Press, 1989)Google Scholar, and Struggle, Politics and Reform (Cornell University: Western Societies Program, 1989).Google Scholar

⁶ See Hirschman, Albert, Shifting Involvements: Private Interest and Public Action (Princeton, NJ: Princeton University Press, 1982), p. 3.Google Scholar

⁷ Pizzorno, Alessandro, ‘Political Exchange and Collective Identity in Industrial Conflict’, in Crouch, C. and Pizzorno, A., eds, The Resurgence of Class Conflict in Western Europe since 1968 (London: Macmillan, 1978), Vol. II, p. 291.Google Scholar

⁸ See in particular Abromowitz, Stephen I. and Nassi, Alberta J., ‘Keeping the Faith: Psychosocial Correlates of Activism Persistence into Middle Adulthood’, Journal of Youth and Adolescence, 10 (1981), 507–23CrossRefGoogle Scholar; Fendrich, James M. and Krauss, Ellis S., ‘Student Activism and Adult Left-wing Politics: A Causal Model of Political Socialization for Black, White and Japanese Students of the 196O's’ in Kriesberg, L., ed., Research in Social Movements I (Greenwich, Conn.: JAI, 1978), pp. 231–57Google Scholar; Jennings, M. Kent, ‘Residues of a Movement: The Aging of the American Protest Generation’, American Political Science Review, 81 (1987), 367–82CrossRefGoogle Scholar; and McAdam, Doug, Freedom Summer (Oxford: Oxford University Press, 1988).Google Scholar

⁹ Jennings, , ‘Residues of a Movement’, p. 379.Google Scholar

¹⁰ See Galli, Giorgio and Prandi, Alfonso, Patterns of Political Participation in Italy (New Haven, Conn.: Yale University Press, 1970).Google Scholar

¹¹ See Dubois, Pierre, ‘New Forms of Industrial Conflict, 1960–1974’Google Scholar in Crouch, and Pizzorno, , eds, The Resurgence of Class Conflict in Western Europe since 1968, Vol. II, pp. 1–34.Google Scholar

¹² On the unions, see Lange, Peter, Ross, George and Vannicelli, Maurizio, Unions, Change and Crisis (London: Allen Unwin, 1982)Google Scholar; on social movements, see Melucci, Alberto, L'invenzione del presente: Movimenti, identita bisogni individuali (Bologna: Mulino, 1982)Google Scholar; on new parties and movements, see Tarrow, , Democracy and Disorder, chaps. 6–11.Google Scholar

¹³ See Hellman, Stephen, ‘The Longest Campaign: Communist Party Strategy and the Election of 1976’, in Penniman, H., ed., Italy at the Polls, 1976 (Washington, DC: AEI, 1977).Google Scholar

¹⁴ The term is Berger, Suzanne's, in ‘Politics and Antipolitics’Google Scholar. Other authors (for example, Tarrow, , Democracy and DisorderGoogle Scholar) question the extent to which the new movements can be understood as protests ‘against’ politics, except in an extremely figurative sense.

¹⁵ For the best overall summary of electoral trends since the early 1970s, see Mannheimer, Renato and Sani, Giacomo, Il mercato elettorale; L'identikit dell'elettore italiano (Bologna: II Mulino, 1987).Google Scholar

¹⁶ See Hellman, , ‘The Longest Campaign’, and the same author's recent Italian Communism in Transition: The Historic Compromise in Turin, 1975–1980 (Oxford: Oxford University Press, 1988).Google Scholar

¹⁷ Tarrow, , Democracy and Disorder, chap. 12.Google Scholar

¹⁸ See Lange, Peter, ‘Crisis and Consent, Change and Compromise’, in Lange, P. and Tarrow, S., eds, Italy in Transition (London: Frank Cass, 1980), pp. 110–32.Google Scholar

¹⁹ Barbagli, Marzio and Corbetta, Piergiorgio, ‘The Italian Communist Party and the Social Movements, 1968–1976’, in Zeitlin, Maurice, ed., Political Power and Social Theory: A Research Annual, Vol. 3 (Greenwich, Conn.: JAI, 1982), pp. 77–112.Google Scholar

²⁰ Hellman, Stephen, ‘Il PCI e la eredita ambigua dell'autunno caldo: Evidenza dal caso torinese’, Il Mulino (03–04 1980), 246–95.Google Scholar

²¹ The survey was conducted by CeSPE, a research institute for the study of economic and political affairs, with ties to the Italian Communist Party. For the basic analysis of these data, see Accornero, Aris, Mannheimer, Renato and Sebastiani, Chiara, L'identita comunisia: I militanti, le strutture, la cultura del PCI (Roma: Ed. Riuniti, 1983).Google Scholar

²² The delegates were representative of the vast majority of PCI members who take an active and regular part in grassroots party activity, along with a large number of party officials. Data were collected from the entire population of elected delegates to the party's 1979 provincial congresses. Questionnaires were distributed by provincial party officials, filled in by hand, and returned to the researchers directly, without passing through the party's national office. Of a potential 27,865 responses, 15,899, or 57.4 per cent of the delegates, filled in questionnaires. See Accornero, et al. , L'identha, pp. 51–61Google Scholar, for details on the surveys and the responses.

²³ Italiano, Partito Comunista, Dati Sull'organizzazione del partito (Roma: PCI, Sezione di organizzazione, 1976)Google Scholar and Italiano, Partito Comunista, Dati Sull'organizzazione del partito (Roma: PCI, Sezione di organizzazione, 1979).Google Scholar

²⁴ Our choice of periods is based upon years measuring significant turning points in party and political history. Period One (1921–46) collapses the years of party growth, clandestine activities and armed struggle, ending with the last full year of inter-party unity following the Liberation; Period Two (1947–56) is the period of pure cold-war politics, in which the PCI was besieged, excluded and isolated; Period Three (1957–66) is the period of disintegration of international Communist unity and of domestic PCI stagnation; Period Four (1967–69) is the period of decay of the centre-left governments, the beginning of the social mobilization which was to reach its apogee in the years following, and of significant, but at the same time largely invisible, internal evolution in the PCI; Period Five (1970–73) is the period of intense movement politics, the breakdown of the centre-left, and the growth and leftward movement of the PCI; Period Six (1974–76) is the period of greatest PCI growth, as the party moved towards power – and towards the centre – and assimilated the largest proportion of new members. It is marked by the victory of the divorce movement and major Communist electoral gains in administrative elections in 1975 and national elections in 1976. Period Seven is the period of PCI participation in various forms of parliamentary support for national governments still led by the Christian Democrats, and of growing internal party dissent with this policy and its failure to bring concrete results. It closes with the PCI's withdrawal from co-operation and with the elections of 1979 when the PCI suffered its first electoral losses since 1948.

²⁵ Each of the ‘pure’ types contains activists who came to the PCI after only one previous experience, with the partial exception of the ‘movement’ type, which may contain some who were active in more than one social movement (e.g., student, extra-parliamentary, feminist, Catholic or other groups).

²⁶ Even before the rise of the student movement in the late 1960s, the FGCI was sometimes at odds with the party. After 1969, these tensions were sharply exacerbated as the FGCI found itself competing for support with active and often confrontational youth movements, which often attacked not only the non-communist political parties for their conservatism but the PCI for its ‘reformism’. For evidence, see Tarrow, , Democracy and Disorder, chap. 6.Google Scholar

²⁷ In the 1970s, the trade unions – including the CGIL – increased their autonomy from all political parties, including the PCI. In addition, the more unified labour movement expanded its agenda to include a number of issues which had traditionally been left to the party system. It also became a competitor with the parties for the attention of, and influence on, the government as it sought to advance this agenda using tactics of mass mobilization and political pressure. As a result, those entering the PCI through this pathway in the 1970s might bring with them attitudes, and an appreciation for tactics, that were significantly different from the party's and from those who had followed this pathway in earlier years. For evidence, see Lange, , Ross, and Vannicelli, , Unions, Crisis and Change.Google Scholar

²⁸ Note that we exclude from both sets of figures those who joined a union simultaneously with, or only after associating with, the PCI.

²⁹ The PSIUP originally sought to gain new strength both from the PSI and the PCI by establishing its identity as a non-Communist, but ‘oppositionist’ representative of the left subculture. Failing to establish this role for itself, however, it drew closer to the Communists and formally dissolved itself into the PCI in 1972. In addition to former PSIUP members, the ‘other party’ category also includes activists who came to the PCI from other parties. Given the small numbers involved, however, it does not seem worthwhile disaggregating this group in detail.

³⁰ Combining the ‘pure’ and ‘mixed’ types (not shown in Table 3), the proportion of delegates reporting any experience in the FGCI also fell from 51.3 per cent to 33.5 per cent between the two time-periods. Thus, the pathway most directly controlled by the party declined dramatically in importance during the years of mobilization.

³¹ Some insights into the processes underlying the preceding findings are provided by Table 3. The table reveals that the changes between the two large periods are not simply a product of aggregation. Some of the shifts that we see after 1969 – the expansion of movement and union pathways and the decline of the FGCI and PCI family routes – were already appearing in less dramatic form in the 1967–69 period. This suggests that even during the much lower level of mobilization in the mid-1960s, many activists were already being fed into the PCI who had not passed through the party's socializing channels. This is consistent with what we already know about the ‘secularization’ of the PCI's mass base during the 1960s. For evidence, see Alberoni, Francesco et al. , L'attivista di partito (Bologna: II Mulino, 1964).Google Scholar

³² The changes in the Italian educational structure have been truly phenomenal since the war. Whereas in 1951, 13 per cent were illiterate, 59 per cent had only an elementary school degree and 4.3 per cent a high-school or college degree, by 1971 5 per cent were illiterate, 44 per cent had an elementary degree and 8.7 per cent had graduated from high school or college (ISTAT, Sommario di Statistiche Storiche, 1926–1985 (Rome: ISTAT, 1986), p. 27).Google Scholar

³³ The educational level of the delegates is, however, higher on average than would have been predicted on the basis of either their professions or their family income. In fact, although almost half the delegates had a secondary or a higher education, they tended overwhelmingly to have come from relatively impoverished backgrounds. Eighty-one per cent of their parents had only an elementary school education, and 57 per cent of the sample's fathers were either industrial workers or farm workers owning no land. Coming from relatively poor family backgrounds, the delegates in large measure were characterized by a good deal of educational mobility. They were thus an emerging social group, even if for many, particularly in the more recent generations, their income levels did not reflect their recent educational advantages. For greater detail on the delegates' socio-economic status and origins, see Accornero, et al. , L'identita.Google Scholar

³⁴ The relationship between university education and a social movement background among the delegates is further confirmed when we observe that the four pathways with the highest percentage of university-educated respondents all include an experience in social movements. These results show that the well-known relationship between participation in social movements and university education holds as well for those who participated in movements and then joined the Communist party.

³⁵ See Almond, Gabriel et al. , The Appeals of Communism (Princeton, NJ: Princeton University Press, 1954).CrossRefGoogle Scholar

³⁶ See Blackmer, Donald L. M. and Tarrow, Sidney, eds, Communism in Italy and France (Princeton, NJ: Princeton University Press, 1975)Google Scholar, and Alberoni, et al. , L'attivista, for evidence on this point.Google Scholar

³⁷ The exact question was: ‘Among the following objectives, which, in your view, are the most important (write 1 and 2 in the space next to the first and second in order of importance):

increase citizen participation;

defeat all forms of tax cheating;

defend public order against terrorism;

eliminate the “jungle of high salaries”;

(from ‘Questionnario per i delegati’, in Accornero, et al. , L'identita, p. 75).Google Scholar

³⁸ For example, independent strikes and strikebreaking are inherently neither illegal nor disruptive, but are politically objectionable to members of the PCI and its affiliated unions.

³⁹ It should be noted that this method of scoring biases our finding against openness to acts of unconventional political behaviour in so far as ‘non-comprehension’ does not necessarily imply that a specific act would not be tolerated.

⁴⁰ The correlations among the four indices are respectively: 0.40 (for ‘Dissent’ and ‘Collective Action’); 0.28 (for ‘Dissent’ and ‘Aggression’); 0.24 (for ‘Collective Action’ and ‘Aggression’); 0.19 (for ‘Dissent’ and ‘Delinquency’); 0.14 (for ‘Dissent’ and ‘Delinquency’); 0.09 (for ‘Delinquency’ and ‘Aggression’).

⁴¹ The logistic regression analyses were conducted on a random 20 per cent sample of the original data set yielding an N of approximately 3,000 respondents.

⁴² See Aldrich, John and Nelson, Forrest, Linear Probability, Logit, and Probit Models (Beverly Hills, Calif: Sage, 1984).CrossRefGoogle Scholar

⁴³ To compute the change in the probability estimate which indicates the increase or decrease in the likelihood that our categories of unconventional behaviour will be tolerated, we first determined the probability that the dependent variable would equal 1 when all dummy variables equalled 0 and all the other variables equalled their mean. We then computed the change in probability given a oneunit increase (based on each variable's range) in the value of a given variable. The value reported for each dummy variable is the change in probability when that variable goes from 0 to 1.

⁴⁴ Similar findings were obtained in models testing the relationships between different generations and the other categories of unconventional behaviour. Only in the case of ‘dissent’ was the coefficient for generation 5 (1970–73) positive and significant.

⁴⁵ With respect to education, we found that the probability for toleration increased with educational achievement (see Table 9). However, we found that university experience was not a prerequisite for toleration of unconventional political behaviour (see Table 10). This could help to explain why, despite the lower number of university-educated respondents in the post-1973 generations than in the 1970–73 generation, attitudes towards unconventional political behaviour remained tolerant.

⁴⁶ The only exception was that ‘pathway’ was insignificant in explaining tolerance towards the two illegal categories – ‘delinquency’ and ‘aggression’ – in the post-1970 generations, whereas it was positively and significantly associated with tolerance among the earlier generations. Distance from the Communist subculture, in other words, could explain tolerance towards illegal acts for the earlier generations but not for the later ones, a finding which is certainly not consistent with the stereotype, although it does fit with other results we have reported.

⁴⁷ See in particular Hellman, , Italian Comminism in Transition.Google Scholar