UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The determinants of working-class conservatism : a cross-national comparison Bakvis, Herman 1972

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1972_A8 B33.pdf [ 5.18MB ]
JSON: 831-1.0064618.json
JSON-LD: 831-1.0064618-ld.json
RDF/XML (Pretty): 831-1.0064618-rdf.xml
RDF/JSON: 831-1.0064618-rdf.json
Turtle: 831-1.0064618-turtle.txt
N-Triples: 831-1.0064618-rdf-ntriples.txt
Original Record: 831-1.0064618-source.json
Full Text

Full Text

THE DETERMINANTS OF WORKING-CLASS CONSERVATISM: A CROSS-NATIONAL COMPARISON by HERMAN BAKVIS A., (Honours), Queen's University at Kingston, 1971 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of POLITICAL SCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1972 In present i-ng-this thesis in parti at fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Political Science The University of British Columbia Vancouver 8, Canada Date September, 1972 ABSTRACT The study focusses on the phenomenon of 'working-class conservatism', support for right-wing or centre parties on the part of certain segments of the working-class. This mode of voting behaviour is analysed in three West European countries; Great Britain, West Germany and Italy. In the introduction a number of explanations are examined which could conceivably account for the phenomenon. It is suggested that explanations centred around the notion of 'embourgeoisement' of the working-class (the process whereby 'affluent' workers come to identify with the middle-class thus voting 'conservative') contain certain weaknesses. It is argued that immediate subcultural influences such as the family, the church and the work-place are still most important in explaining the working-class vote for 'conservative' parties. An explanatory model is developed which emphasizes the influence of a worker's social environment as the chief determinant of his voting behaviour. Hypotheses derived from this model, as well as alternative hypotheses related to the 'embourgeoisement' argument are tested through secondary analysis of survey data collected in the late 1950's and early 1960's. The results of the study suggest that in all three countries 'working-class conservatism' can be explained largely in terms of the subcultural influences outlined in the model. Level of income attained by workers may have some independent effect in West Germany and Great Britain. In these two countries conservative parties are somewhat more successful in retaining the loyalties of conservative workers in the low and high income categories compared to those in the medium categories. However this does not mean a confirmation of the 'embourgeoisement1 argument. Most 'affluent conservative workers', it is argued, arrived at their conservative voting identity via early parental socialization. There are few defectors from left-wing parties to conservative parties among 'left' workers who attain a high level of income. iii TABLE OF CONTENTS Chapter Page I Introduction Background and Propositions .. .. .. 2 Methodology .. .. .. .. .. 7 II Embourgeoisement and Working-Class Conservatism .. 10 III The Social Determinants of Working-Class Conservatism .. . . . . .. .. .. 32 Subcultural Influences: The Neighbourhood . . .. 37 Parental Voting: .. .. 37 Religion .. . . . . 38 Rural-Urban Migration .. .. .. 47 Traditionalism and Working-Class Conservatism .. .. 50 Subcultural Influences .. .. .. .. 55 The Work-Place Occupation .. .. 57 Skill Level 61 IV Conclusion .. .. 79 Implications .. .. 92 Bibliography .. .. .. 106 iv LIST OF TABLES Table Page (Chapter 2) 1 Income Frequencies of Classes 15-16 2 Party Vote by Level of Income Among European Workers . . .. 18 3 Party Vote by Level of Income 20 4 Party Vote by Economic Satisfaction Among Workers .. .. .... .. .. 24 5 Party Vote by Interviewer Rating of Socio-Economic Class .. 25 6 Party Vote by Self-Placement in Class Among Workers .. .. 28 7 Party Vote by Income Level for Middle-Class Identifiers 28 (Chapter 3) 1 Party Vote by Parental Voting Behaviour Among British Workers 39 2 Party Vote by Religious Affiliation: West Germany 41 3 Party Vote by Church Attendance: West Germany and Italy ' 42 4 Party Vote by Church Attendance for German Catholics 44 5 Party Vote by Religion for British Workers 44 6 Percentage of Workers Voting Conservative by Attendance Controlling for Religion .. 47 7 Party Vote by Migration: West Germany .... 49 8 Mean Scores for Conservative Workers and Left Workers .. 53 V LIST OF TABLES (Continued) Table Page (Chapter 3) 9 Party Vote by Sex Among Workers 58 10 Party Vote by Occupation .. 59 • 11 Party Vote by Income Controlling for Occupation 62 12 Party Vote by Skill Level . 64 13 Party Vote by Income for German Unskilled Workers .. .. . 65 14 Party Vote by Skill Level 66 15 Percentage Voting Conservative by Income Controlling for Skill 67 16 Party Vote by Union Membership .. .. .. 68 17 Party Vote by Union Membership for British Workers .. .. .... .. 75 18 Party Vote by Income Controlling for Union Membership .. .. 76-77 (Chapter 4) 1 Party Vote by Income for British Workers Who 'Feel Close to their Union* 87 2 Party Vote Income Controlling for Parental Voting 89 3 Percentage Voting Conservative by Income Controlling for Age 90 CHAPTER I Introduction What do you say to the elections in the factory districts? Once again the proletariat has discredited itself terribly... It cannot be denied that the increase of working-class voters has brought the Tories more than their simple percentage increase; it has improved their relative position. Letter from Engels to Marx, 1867.^  According to Karl Marx Great Britain was possibly one of the few places where the blood and violence, usually accompanying proletarian revolutions, could be averted. The working-class, with its newly gained franchise and by virtue of its numerial superiority, would merely vote the bourgeoisie out of office and install itself as the dominant class. Unfortunately, as Marx and Engels discovered, this peaceful transition to socialism and hence communism, failed to occur. And, as Robert McKenzie points out, "Ninety years and thirty-three elections later a considerable 2 section of the proletariat is.... still 'discrediting itself terribly'." The phenomenon that proved to be so vexing to Marx and Engels has continued to persist to the present day, providing many social scientists with a focal point for their research. Aside from its apparent uniqueness this segment of the electorate, approximately thirty percent of the working-class vote, regularly provides the Conservative party with about 3 one-half of its total support at election time. According to Jean Blondel most of the work which has been done on voting behaviour in Great Britain 4 has been concerned with the problem of the Conservative working-class vote. 2 •fet in treating this question as a separate case, unique to Great Britain, many social scientists seem to neglect the fact that in nearly every West European country a substantial segment of the working-class population regularly supports centre or right-wing parties. In Italy the industrial working-class (approximately 32% of the population) regularly gives 26 percent of its vote to the Christian Democratic Party (DC) and 7 percent to Monarchist and neo-Fascist parties. And in West Germany, which has basically a two party system, the Christian Democratic Party (CDU), like the Conservative party in England, receives about one-third of the working-class vote. Similar examples of 'deviant' voting behaviour on the part of members of the working-class, may be found in France, the Scandinavian countries, and elsewhere."* There have been few, if any, comparative studies in this area, although some writers have touched on the problem.^  It is the aim of this thesis, therefore, to study, in some detail and in cross-national fashion, this one sector of the voting public in three West European countries. The three countries selected are Great Britain, West Germany and Italy. Although displaying differences in terms of party systems, political culture and levels of economic development they all have in common the phenomenon I shall nominally refer to as 'working-class conservatism', working-class support for centre and right wing parties. Background and Propositions In the literature there appear to be three possible explanations 3 or modes of explanation for 'working-class conservatism'. Two of these explanations can probably be placed under the rubric of the 'end of ideology' thesis.^  The first explanation, one that was especially popular in England during the late 1950's, claims that the 'affluent' worker, who had formerly seen himself as both worker and left-wing, was gradually being engulfed into an expanded middle-class, a 'middle-majority'. Kurt Mayer, for example noted that in Western societies an increasing number of workers were achieving levels of income that overlapped with those of many white-collar workers, allowing such workers to indulge in consumer goods access to which, until the post-war period, had been restricted to g the middle and upper-middle classes. This, argued people like Mark Abrams and Rita Hinden, led the worker to adopt a middle-class life style which 9 in turn predisposed him to vote for the Conservative party. The worker, so the argument claims, did not so much cross class lines but rather, because of his increased power on the consumer market, these lines became blurred. The second approach offers a somewhat different and at the same time more plausible explanation of the impact of affluence. Instead of workers switching to conservative parties, S. M. Lipset in 1964 argued, more workers are actually voting for left-wing parties. Working-class conservatism, in England and other European countries, is seen as a traditional holdover which will gradually decline as most workers realize that moderate left parties will best protect and advance their eclectic, 4 economic interests.This interpretation however, is in large part still derived from the end of ideology thesis. Like others, Lipset contends that the impact of economic development in industrial societies has led to a decline of "ideologically intransigent politics...""'""'" at the mass level as well as the elite level. What people like Abrams and Lipset seem to have in common are certain assumptions that economic affluence and a particular type of voting behaviour are directly and necessarily related. They seem to imply that increased affluence can be translated by workers into 'middle-majority' status. However in stressing the deterministic influence of economic development and the instrumental, rational and calculative nature of workers, other factors which may have equal if not greater effect on voting behaviour are often neglected. The third approach, the one I propose to use in my analysis, is related to what is sometimes known as the consensus and cleavage model. In terms of this model, according to Janowitz and Segal: Politics and political behaviour are still seen as derived from the conflicts of social strata, but political affiliations are more than a by-product of social strati-fication. Political institutions and political leadership are more autonomous elements in the process of change... Likewise religious, ethnic and linguistic differences can persist or emerge as bases of political cleavage which includes ideological elements.12 The determinants which I feel are still most important in explaining working-class conservatism are those subcultural groupings known as primary or reference groups. These groups, such as family, church 5 and work place, I would argue, are most important in shaping and reinforcing an individual worker's political attitudes. In arguing that the social pressures inherent in the immediate environment are of special relevance in the determination of political attitudes I have developed a model, incorporating five basic reference groups. The model takes the form of five hypotheses. They are as follows: (1) Early socialization via parents is crucial in establishing a conser-vative voting identity or in providing basic attitudes that will make a worker receptive to such an identity; (2) this tendency to vote conservative will be maintained or even reinforced if a worker's neighbourhood environ-ment is heterogeneous or middle-class, that is lacking a sort of left-wing, working-class "Gemeinschaft'; (3) regular church attendance is likely to increase the probability of voting for a conservative party; (4) the likelihood of a worker voting conservative will be increased if his place of employment is relatively small and traditional; (5)finally union membership is likely to decrease the probability of voting conservative. In following the spirit suggested by the consensus and cleavage model, I should add the following qualifications: the importance of the above named subcultural influences is contingent upon special qualities which may vary from country to country. One of the more important of these variables is the manner in which parties and related organizations attempt to alter the conditions of these reference groups in order to enhance their support base. Indeed it will be argued later that, in certain cases, the organizational efforts of a centre, conservative or right-wing party 6 or the lack of such efforts on the part of left-wing parties, is a crucial factor in explaining the persistence of older cleavages as a basis for conservative voting among worker's. At the same time my acceptance of the consensus and cleavage model does not imply an outright rejection of the 'embourgeoisement of the working-class' argument or the denial that the post-war economic boom has had an effect on working-class politics. On the contrary, as I will note later, the impact of affluence has had some effect, though perhaps not necessarily bringing about the changes postulated by some of the 'end of ideology' theorists. Thus in my analysis an attempt will be made to assess the importance of economic, cultural and leadership factors, and to place them in proper perspective. This aim can in part be achieved by testing the various hypotheses offered to account for working-class conservatism through the secondary analysis of economic, cultural and demographic survey data on workers. In addition, questions raised can be further explored by examining the attitudes of conservative workers towards their respective polities. For example a number of writers on working-class Tories in Britain, in stressing the influence of past traditions, contend that such voters tend to be deferential in their attitudes towards political authority. I should point out that the propositions examined in this study do not exhaust the range of possible explanations to account for working-class conservatism. There may well be other important factors related to ethnic, linguistic or religious differences within the three 7 countries. The problem is that almost no survey includes these variables or, if they are included, attempts to ascertain their importance in a rigorous manner. The goal then, is to focus on those explanations which can be adequately tested, thereby leaving the construction of more complete models to those who have access to greater resources. In addition it should not be thought that the factors delineated are necessarily involved in other examples of 'deviant' voting, middle-class support for left parties for instance. Similar sociological forces (i.e. primary group influences) disposing workers to vote conservative may also cause 'middle-class left' voting. This, however, is a separate question; again, although deserving attention, it cannot be adequately handled in this study. Methodology The major source of evidence for this study will be data collected via a number of surveys. These surveys were carried out in the three countries in various time periods. In the main, for analyzing variables such as class, religion and so on, simple tabular analysis has been used. In order to tap abstract attitudinal dimensions more complex measures have been devised utilizing various scaling techniques. The data most heavily depended upon is in the Almond and Verba 13 survey of 1959-60 covering all three countries. Also used are parts of the Butler and Stokes survey of the British electorate, carried out 8 in a series of waves from 1963 to 1969. The major aim of the Almond and Verba study was to determine respondents' basic political attitudes and modes of political participation. In addition the survey contained questions concerning demographic variables, class, occupation,group affiliation and church attendance. This survey is probably most useful in that it covers the three countries in question. Unfortunately there are also a number of drawbacks involved in using the Almond and Verba data. Questions pertaining to parental voting behaviour and plant size were not included in the survey. Furthermore the relatively small sample of workers in each country limits the number of variables that can be controlled for and the degree to which relationships apparent in the data can be claimed as being representative of the universe being sampled. This problem is compounded by the fact that in Italy there is an extremely high refusal on the part of respondents, especially when asked to reveal their party identification. A further problem which, in some respects, mitigates the usefulness of the Almond and Verba data, is that the survey is over 12 years old. One could argue that developments in the political, social and economic spheres in the intervening period have altered the patterns of political cleavage in Europe. This may well be true. My analysis,therefore may not be directly applicable to the Europe of 1972. Some of these problems however, have been alleviated by the use of the Butler and Stokes data. This survey has allowed me to cover gaps in the Almond and Verba and to explore certain relationships, 9 particularly those concerning economic influences, in greater depth. The study is structured into two major chapters and a conclusion. The chapter immediately following will be devoted to discussing and testing some of the more specific tenets of the embourgeoisement argument. The third chapter will be concerned with the social influences inherent in the immediate environment of the working-class conservative, neighbourhood influences such as the family and the church, and the influences of the work-place, trade union membership, occupation and so on. The concluding chapter will offer a possible explanation of puzzles that remained unresolved in the main body of the thesis. 10 CHAPTER II Embourgeoisement and Working-Class Conservatism The notion of embourgeoisement is not a new one. In fact it was among the early Marxists that the problem was first raised; Engels being the most notable among them. In accounting for the fact the British working-class failed to exploit their newly granted franchise during the 1870's and 1880's, Engels argued that the then current wave of economic prosperity in Britain was holding the revolutionary potential of workers in abeyance.^  Once this economic bubble collapsed and the worker was once again subject to the harsh economic realities of capitalist society, Engels felt the proletariat would fulfill its historic mission.^  This line of argument was pursued further by other Marxists in the twentieth century. V. I. Lenin, in his famous tract entitled "Imperialism, the Highest Stage of Capitalism", claimed that the worker was 'bought off, as it were, by profits gained by the exploitation of foreign colonies.^  The post World War II version of embourgeoisement, however, differs somewhat in that it incorporates more complex refinements concerning the manner in which the impact of affluence, or at least a higher standard of living, is felt by the worker. The emphasis is still on material consumption but a number of writers, like Ferdinand Zweig in Britain and Wilensky in the United States, seem to imply that affluence 18 has led to a deterioration of working-class norms and cultural values. 11. Workers are no longer satisfied with simple material goods 'per se' but wish to emulate and adopt the life-style of white-colllar workers. There is of course no single, post-war, embourgeoisement model. There is still the relatively simple prosperity argument, quite similar to earlier Marxist theories, postulating that the voter choses a conservative party in order to preserve or enhance his standard of living, 19 without regard to middle-class status. The other theories are put forward in a more speculative fashion and therefore are less amenable to clarification. For example it is sometimes difficult to say whether the pre-occupation with middle-class status on the part of workers occurs prior to their satisfaction of those aspirations. In such a case one might assume that the worker votes conservative in order to ensure continued national prosperity and thereby his own economic well being. Alternatively the process may occur in a somewhat more subtle fashion. The worker gradually, perhaps even in a quite unconscious manner, picks up the accoutrements of a middle-class life-style. Once having arrived in a particular position in the status hierarchy, he may decide to change his vote in favour of a conservative party in order to complement or to conform to this new life-style. It would be misleading to say that the emphasis of theorists has been purely on the consumption sphere. Many writers on the theme of 'consensus in the age of affluence', have also cited changes in the nature of the work place as being associated with the post-war economic 12 boom. The worker, it is argued, has greater job security, the hours are shorter, and the working conditions themselves have been rendered more attractive. Thus Ralf Dahrendorf, in his Class and Class Conflict in Industrial Society, claims that tension between worker and management 20 is no longer a major source of political cleavage. Industrial management-labour conflicts have lost their intensity; what conflicts remain have been institutionalized and thereby insulated from the 21 political sphere. This in turn, others have argued, has aided in what is perceived by many as a decline in working-class solidarity; the worker has lost interest in his trade union and associated political activities and become more concerned with his own personal interests and 22 those of his family. This restructuring of the relationship between the worker and his place of employment, one could argue, would undoubtedly affect his complex of attitudes towards life and politics. Thus in a setting whereby the worker has the financial ability to develop a middle majority life-style, and where the problems concerning his work, if they exist, are isolated from the other areas of his life, a vote for a conservative party would not be out of place. There is no doubt that all three countries, Britain, Germany and Italy, have, to a greater or lesser degree, experienced a relatively high degree of prosperity as measured by per capita income. The economic 23 growth rates of Germany and Italy have been especially phenomenal. The per capita income of Italy, however, was still below that of Great 13 Britain as of 1968 and the gap was probably somewhat greater in 1960. Nevertheless one must keep in mind the notion of 'two Italies'. Most of the industrial development has occurred in the Northern half and this is also the area in which most members of the industrial working-class are 25 located. Thus although the wage levels of industrial workers in Italy are lower in comparison to those of other West European nations, the differences are probably closer than per capita figures would suggest. In testing some of the more specific tenets of the embourgeoise-ment hypothesis with regard to working-class conservatism, one would first of all expect that the working-class has been sharing in the largesse created by the post-war economic boom. As Robert Lane notes, the notion of the "'Age of Affluence' refers... to more than higher per capita national income...", it also embraces "a relatively equalitarian 26 distribution of income." Thus if members of the working-class are to attain a middle-majority status via the consumption market, they must have the financial ability to indulge in the variety of consumer goods offered. A rough test of this can be to see to what extent working-class incomes overlap with those of the middle-class. This has been done in Table I by comparing frequency distributions of income levels for different socio-economic classes using the Almond and Verba data. The three classes utilized are: (1) upper-middle, those persons occupying professional or managerial positions, (2) the middle-class, constituted mainly of those who are self-employed (i.e. small storekeepers). 14 lower level civil servants and rank and file white-collar workers; (3) persons located in the working-class are those defined by Almond and Verba as being either skilled or unskilled, or working in 'domestic service' (i.e. maids and butlers). Farmers and farm-workers have been eliminated from the sample. Included in the sample, however, are housewives who have been defined as belonging to one of the three' categories by virtue of their husbands' occupation. The income categories are based on reported family income rather than individual income. As one can see in the following tables for each country, the bulk of the three social classes fall into quite distinct income categories. But at the same time there is some overlap between the income distribution of workers and that of the middle-class, especially in the case of West Germany. In the latter country it appears that nearly one-fifth of the workers have a family income equal to that of the median income of middle-class people. In interpreting those graphs, however, it should be borne in mind that the degree of overlap may be exaggerated by a number of factors. A considerable amount of the overlap may be due to the comparison of high-income industrial workers, living in large cities, with less well off middle-class people living in less urban areas where the wage levels may be lower. Age may also confound the relationships plotted. For example the workers located in the higher income categories could well be fairly senior in years and therefore at their optimal earning power. Many of the middle-class people in the lower income categories may be just starting their careers and have yet to reach their full income potential. Fre-quency 40 of Income in % TABLE I Income Frequencies of Classes 15 30 " 20 : 10 Italy Under 200,000- ' 400,000- 700,000-! 1 millionl ±h mill-! 2h million 200,000 399,000 699,000 999,000 Ik million ion- plus lire lire lire 2^  million Annual Income = Working-class = Middle-class = Upper Middle-class Under 150-250 250-350 350-500 500-750 750-1000 1000 plus 150 DM DM DM DM DM DM DM Monthly Income = Working Class = Middle-class = Upper Middle-class TABLE I C o n t i n u e d 16 A n n u a l Income = W o r k i n g - c l a s s = M i d d l e - c l a s s = Upper M i d d l e - c l a s s 17 Such low income middle-class people may not be able to partake of a truly middle-class life style, but the future prospect of doing so may dispose them in a conservative direction. Nevertheless, even if there still is a considerable gap in income between the working and middle classes, one could still argue that in the 'affluent' society even workers are able to buy the basic 'middle-class package', so to speak, consisting of consumer items. In Great Britain, for example, the Conservatives emphasized this theme in the late 1950's by running advertisements showing an average family with a vast 27 array of luxury goods. Thus it is conceivable that there is at some point an income threshold which, once crossed by members of the working-class, would enable them to become part of the 'middle-majority'. If this is true, and if members of the working-class do indeed also vote conservative upon crossing this threshold one would expect that the higher the level of reported income the greater the chance that a worker will vote conservative. Table 2 reports the relationship between income and party vote for workers in the three countries. The category of working-class is the same as used in Table 1. For Great Britain the Labour party is defined as the Left party. In Germany the Social Democratic Party (SPD) has been placed in that category. For Italy the following parties have been defined as being left of the DC by virtue of their ideologies and programs; the Communist party (PCI), the Socialist party (PSI), the 28 Social Democratic party (PSDI) and the Republican party (PRI). Included under the label of 'Right' for Italy is the right-wing Liberal party (PLI), 18 TABLE 2 Party Vote by Level of Income Among European Workers Great Britain INCOME (Annual) Party Under £300 L300-L650 L650-il000 L1000 plus Total Sample Cons. 36.1% 33.2% 28.0% 53.8% 32.3 Labour 58.3 64.2 67.1 38.5 63.9 Liberal 5.6 2.6 4.9 7.7 3.8 Total per cent (N) 100 (36) 100 (232) 100 (143) 100 (13) 100 (424) West Germany INCOME (Monthly) Party Under 250 DM 250-350 DM 350-500 DM 500 DM plus Total Sample CDU 55.6% 38.9% 32.3% 50.0% 40.9 SPD 44.4 61.1 67.7 50.0 59.1 Total per cent (N) 100 (18) 100 (36) 100 (62) 100 (38) 100 (154) Italy INCOME (Annual) Party Under 200,000 Lire 200,000-399,999 Lire 400,000-699,999 Lire 700,000 Lire plus Total Sample DC 71.9% 56.7% 51.1% 47.6% 56.9 RIGHT 15.6 13.3 8.5 0.0 10.0 LEFT 12.5 30.0 40.4 52.4 33.1 Total per cent (N) 100 (32) 100 (30) 100 (47) 100 (21) 100 (130) Source: Almond and Verba Survey, 1959-60, 19 the Monarchist party (MON) and the neo-fascist Social Movement party 29 (MSI). The results are rather interesting. For our English sample, comparing the differences between income levels, there appears to be a decline in the tendency to vote conservative as one moves across the columns with the exception of the fclOOO plus income category where there is a sudden upswing in the percentage of workers who vote Tory (53.8% vs an average of 32.5%). This last category of voters, however, contains a total of only 13 voters. Thus aside from the fact that, in terms of the Almond and Verba data, the Conservative Party draws only 5.1% of its working-class support from this group, the small number might indicate that some caution is required in interpreting these results. Nevertheless the same relationship appears in the case of West German workers. The strength of the CDU decreases, as one goes up the income scale, from 55.6% to 32.3%, but then jumps up again to 50.0%. Moreover, the proportion of respondents in the high income category is larger than is the case of Great Britain. In Italy the relationship is more linear, albeit in the direction opposite to that posited by the 'embourgeoisement' argument. It appears that the greater the income the higher the probability that the Italian worker will not vote for either the DC or a right-wing party. If one turns to Table 3 the same variables, party vote and income, have been cross-tabulated, this time using the 1963 wave in the Butler and Stokes survey for Great Britain. Employing three 20 TABLE 3 Great Britain Party Vote by Level of Income Under L550- Over Total Party- 6500 £750 &750 Sample Cons. 32.8% 18.1% 30.8% 26.8% Labour 62.5 76.8 64.9 68.5 Liberal j 4.7 5.1 4.3 4.7 Total per cent 100 100 100 100 (N) (253) (276) (211) (740) Source: Butler and Stokes Survey, 1963 Wave income categories one can see that the curvilinear relationship is even more pronounced, workers in the low and high income categories are much more likely to vote Conservative than those in the medium income category. Perhaps the time element, the larger sample, different cutting points for income or different sampling techniques might explain why the relationship appears to be more pronounced in the Butler and * Stokes data. * The Butler and Stokes definition of 'working-class' is similar to the one used in the Almond and Verba survey. Included in this category are servants, foremen, skilled, semi-skilled and unskilled workers. In the Butler and Stokes survey there are nine income categories. In order to present the data in the most economical fashion without unduly distorting the results, I have collapsed the nine income categories into three categories. These income categories for the Butler and Stokes data, as well as the socio-economic class category, are retained throughout this study. For the Almond and Verba data the definition of the working-class and the income categories, as delineated earlier, are also retained for the remainder of this study. 21 Thus so far it appears that if the notion of 'embourgeoisement' has any validity it does so only in the case of West Germany and in Great Britain. In any case affluent workers who vote for conservative parties appear to be in a distinct minority among the total number of workers who vote conservative. As I will attempt to demonstrate in the next chapter the majority of conservative workers in all three countries are influenced in their voting behaviour by a more traditional subculture. For example, in Great Britain, servants, usually seen as being employed in a more traditional occupation, tend to vote Conservative and at the same time are concentrated in the lowest income category. However this does not necessarily negate the value of the embourgeoisement argument. If one looks back at Table 1 it appears that of the three countries, at that particular time period, the overlapping of workers salaries with those of the middle-class is probably most pronounced in West Germany. In England the distinction between the salaries of the working-class and those of the middle-class, in terms of Almond and Verba data, is somewhat sharper. And in Italy, recalling our earlier discussion of differences in levels of economic development, the per capita income was below that of Great Britain. One could argue, then, that the notion of an economic threshold still holds. In West Germany, compared to the other three countries, a larger proportion of workers have been in receipt of what could be interpreted as 'middle-majority' incomes. In England, it might be the case that considerably fewer workers had 22 crossed this threshold at the time the Almond and Verba survey was carried out. Meanwhile in Italy, perhaps because of lack of economic development, none or few of members of the working-class would be of 'middle-majority' economic standing. Thus in terms of the above evidence one could conceivably revise the embourgeoisement argument as follows: the least affluent workers would be influenced by what Lipset has referred to as traditional hold-30 overs from the past. As the workers become more affluent they realize that it is in their best interests (perhaps economic ones) to vote left. However at a certain point (the threshold) the embourgeoise-ment process sets in. The worker has achieved an income of a sufficient size that allows him to become 'middle-class'. At this stage the worker votes conservative again. In view of the small samples and perhaps also because of the simplistic nature of the argument, it would be wise to regard the above in a rather speculative vein. There are however, a few more tests that can be carried out. In the Almond and Verba survey respondents were asked how satisfied they were with their incomes. If voting conservative in part reflects satisfaction with the status quo (and in all three countries the Conservative party and the two Christian Democratic parties were in power) then this should be demonstrated in their response to the economic satisfaction question. The results of cross-tabulating party vote with income satisfaction among workers, shown in Table 4, vary from country to country. 23 In Italy, the degree to which workers are satisfied with their income seems to bear little relationship to their choice of party. In Great Britain workers satisfied with their income are more likely to vote Conservative compared to those disatisfied with their income (33.5% versus 24.4%). However, British workers most likely to vote Conservative are those who consider their level of income to be merely adequate (41.1% versus an average of 32.5%). For Germany the differences are more clear cut. Workers who are satisfied with their income are more likely to vote for the CDU than either the 'so-so' or the disatisfied workers. Thus workers in Great Britain and Germany who are satisfied with their income are more likely to vote Conservative than workers who are unsatisfied with their income. Nevertheless income satisfaction is not necessarily related to the actual income received by workers. This became apparent when the relationship between party vote and income satisfaction was controlled for level of income (tables not shown). In Great Britain most of the 'satisfied' workers were in the higher income categories. In West Germany, however, a considerable number of 'satisfied' workers were in the low income category and they were just as likely to vote CDU as high income 'satisfied' workers. 24 TABLE 4 Party Vote by Economic Satisfaction Among Workers Great Britain  Level of Economic Satisfaction Total Satisfactory So-so Unsatisfactory Sample Cons. 33.5% 41.1% 24.4% 32.5 Labour 62.4 57.8 70.7 63.8 Liberal 4.1 1.1 4.9 3.7 Total per cent (N) 100 (221) 100 (90) 100 (123) 100 (434) West Germany Level of Economic Satisfaction Party Satisfactory So-so Unsatisfactory Total Sample CDU 50.5% 22.7% 40.9% 41.6 SPD 49.5 77.3 59.1 58.4 Total per cent 100 100 100 100 (N) (95) (44) (22) (161) Italy Level of Economic Satisfaction Party Satisfactory So-so Unsatisfactory Total Sample DC 54.1% 58.3% 54.9% 56.3 RIGHT 10.8 9.5 7.0 8.9 LEFT 35.1 32.1 38.0 34.9 Total per cent 100 100 100 100 (N) (37) (84) (71) (192) 25 The Almond and Verba survey does not contain questions pertaining to self-placement in class. Interviewer assessment, however, may serve as an indirect measure of possible middle-class aspirations. If workers perceive themselves to be middle-class this should be apparent in their attitudes and lifestyle, thus perhaps leading the interviewer to rate them higher on the socio-economic scale than their more working-class cohorts. This hypothesis is only partially confirmed. In Table 5 party vote has been cross-tabulated with interviewer rating of socio-economic class, controlling for class as defined by occupation. One can see that in England for those in the higher socio-economic rating category there is a slightly higher probability of voting Conservative. In Germany however there is little in the way of a significant difference. In the case of Germany, in view of possible embourgeoisement effects there, we might have expected a greater difference in the direction of a greater probability of CDU voting among the more highly rated workers. For Italy the data seem to suggest that workers receiving a higher rating tend to vote for a left party rather than the DC or a right-wing party. But in view of the previous evidence concerning income this result is not unexpected. Thus if conservative workers tend to be more middle-class, or at least think of themselves as more middle-class, this was not always readily apparent to the interviewers. The Butler and Stokes study contains questions designed to tap whether or not a respondent feels himself to be part of any 26 TABLE 5 Party Vote by Interviewer Rating of Socio-Economic Class Great Britain Rating of Class Medium Total Party High Low Low Sample Cons. 61.9% 32.3% 29.2% 32.3 Labour 33.3 62.4 69.3 64.0 Liberal A.8 5.3 1.6 3.6 Total 100 100 100 100 per cent (N) (21) (226) (192) (439) West Germany Rating of Class Total Party Medium Low Low Sample CDU 42 .6% 39.6% 41.7 SPD 57 .4 60.4 58.3 Total per cent 100 100 100 (N) (115) (48) (163) Italy Rating of Class Medium Total Party High Low Low Sample DC 41.7% 57.6% 61.9% 56.1 RIGHT 13.9 7.1 9.5 9.1 LEFT 44.4 35.4 28.6 34.8 Total 100 100 100 100 per cent (N) (36) = (99) (63) (198) Source: Almond and Verba Survey, 1959-60. (Note: For Germany only two interviewer rating categories are used since there were too few workers in the higher categories.) 27 particular class. Since the curvilinear relationship between Conservative voting and income appearing in the Almond and Verba data is even more evident in the Butler and Stokes sample the latter survey is probably a valid source of evidence for testing whether high income, middle-class identification and conservative voting are related in any way. Cross-tabulating party vote with middle-class identification for British workers (Table 6) does suggest that workers who see themselves as being 'middle-class' are much more likely to vote Conservative than those who consider themselves to be 'working-class' (48.6% versus 19.9%). At the same time it should be noted that the Conservative party draws only a small minority of its working-class support (19.3%) from 'middle-class' identifiers. Thus British workers who identify themselves as middle-class are more likely to vote Conservative. However when using the variable of middle-class identification to control for the relationship between party vote and income it becomes obvious that those workers who see themselves as middle-class and Conservative are not necessarily among the most affluent. As can be seen in Table 7 those who call themselves middle-class in the lowest income group are more likely to vote Tory than 'middle-class' workers in the high income category (70.0% versus 48.0%). From this it almost appears that middle-class identification among workers combined with low income is a stronger force in impelling workers to vote Tory than is middle-class identification combined with high income! At this point in the analysis there is some evidence to lend credibility to the argument that a worker receiving a high level of 28 TABLE 6 Party Vote by Self ^-Placement in Class Among Workers Self-Placement in Class Party Middle Working Don't Know Total Sample Cons. 48.6% 19.9% 33.3% 25.5 Labour 45.9 76.0 59.2 69.6 Liberal 5.4 4.1 7.5 4.9 Total 100 100 100 100 per cent (N) (74) (513) (147) (734) TABLE 7 Party Vote by Income Level for 'Middle-Class' Identifiers Party Low Income Level Medium High Total Sample Cons. 70.0% 23.8% 48.0 47.0 Labour 30.0 66.7 44.0 47.0 Liberal 0.0 9.5 8.0 6.1 Total 100 100 100 100 per cent (N) (20) (21) (25) (66) 29 income is more likely to vote conservative compared to medium income workers. The notion that the influence of high income is mediated by middle-class identification is only partially borne out. There are, moreover, still several problems that remain unresolved. For example implicit in the embourgeoisement argument, even as restructured in this chapter, there is still the notion that a left-wing worker will switch to a conservative party once an income 'threshold' is crossed. There is nothing in our evidence so far however, to indicate that this has actually occurred. Workers in the upper income groups in Great Britain and Germany may have voted Conservative/CDU most of their lives regardless of their income or life-style in the past. The notion of middle-class identification also deserves closer scrutiny. One suspects that the workers in the lowest income category who call themselves middle-class are not the ones most likely to be endowed with the cultural and material accoutrements of a middle-class life-style. Eric Nordlinger, in his book Working-Class 31 Tories, has also noted this peculiarity. Probing this phenomenon in greater depth he points out that this sort of identification on the part of low income Conservatives was purely subjective. It was induced neither through affluence nor aspirations for higher status. Nordlinger's sample of 'middle-class' workers were more likely to have below average incomes and seemed unconcerned with the social status aspects of middle-class membership (i.e. education for their offspring, 32 occupational mobility and so on). 30 On the question of life-styles Richard Hamilton has argued that in the case of the German working-class, even when the income of affluent workers overlapped with those of the middle-class, their 33 consumption patterns were markedly different. German workers, according to Hamilton, regardless of income preferred "home entertain-ment devices" (i.e. radios and televisions), placing much less value on status symbols. Members of the middle-class are much more likely to purchase luxury goods such as high quality automobiles and to find their entertainment outside of the home. Using survey data collected in the early 1960's, Hamilton claims that the results, even when educational levels were controlled for, still demonstrated strong differences in cultural habits; workers prefer radio or television 34 in lieu of reading or play going. Nonetheless, in spite of the above, one could still make the argument that subjective middle-class identification, even if largely unrelated to any sort of middle-class standing or life-style, may still be an important factor. Put simply, a worker of low income, who obtains a television set on hire-purchase, might well think that he has at last entered the realm of the middle-class and thus vote for a conservative party. Thus a worker may perceive himself to be affluent and middle-class objective conditions notwithstanding. But again there is no evidence that this sort of process is actually involved in predisposing a worker to vote conservative. Such a revision of the embourgeoisement argument would still impute to the worker a high degree of conscious concern with consumption of goods 31 and middle-class status. Furthermore it again involves the crucial notion that such a worker was a supporter of a left-wing party prior to his switching to a conservative party. Keeping the various aspects of embourgeoisement in mind I would now like to pursue some alternative hypotheses explaining working-class conservatism. In the next chapter I will examine whether certain sociological influences, especially those relating to the workers' immediate environment, have a greater bearing on conservative voting behaviour than the direct impact of affluence. 32 CHAPTER III The Social Determinants of Working-Class Conservatism It has sometimes been said that Marx was a better sociologist 35 than he was an economist. Marx as an economist may be faulted for his unfulfilled predictions concerning the demise of capitalist society. Nevertheless there is no doubt that Marx would have made a first rate industrial sociologist. In several of his writings Marx focussed not only on the economic but also the social conditions under which the proletariat would arrive at some sort of class-consciousness as to their historical mission. Outlining the process of the industrial revolution Marx noted that the enormous steelmills and textile factories in England required large numbers of workers who in turn had to be housed nearby in huge tracts of terraced houses. This contiguity of worker with worker both in the factory and the neighbourhood created conditions facilitating social and political interaction. Workers could easily discuss their problems, becoming aware that their grievances were shared collectively and perhaps could even be acted upon collectively. Such a working-class, according to Marx, could generate and sustain values at odds with those of capitalist society, bring pressures to bear upon the more equivocal members of the proletariat and act to insulate the class as a whole from bourgeoise . , . 36 social influences. The idea that a strong, homogeneous working-class subculture in-creases the potentiality for radical activity has been illustrated many 33 times over. Clark Kerr and Abraham Siegel, for example, in their study, 37 "The Interindustry Propensity to Strike - An International Comparison", note that certain industries are much more strike prone than others. They found that in industries such as mining and fishing the number and duration of strikes was much greater in comparison to industries related to agriculture and trade. These differences in propensity to strike were related not only to differences in industrial conditions but also to dissimilarities in community settings. The miners, the sailors, the longshoremen, the loggers, and...the textile workers form isolated masses, almost a "race apart". They live in their own separate communities: the coal patch, the ship, the waterfront district, the logging camps, the textile town. These communities have their own codes, myths, heroes, and social standards. There are few neutrals in them to mediate the conflict and dilute the mass.38 By implication then, it would seem that "conservative workers" in all three countries are more likely to be located outside of such homogeneous subcultures. Frank Parkin, following Baldwin's dictum, "Whichever party may be in office the Conservatives are always in power", argues that since there is a close affinity between the dominant values surrounding social and political institutions in English society and those implicit in Conservatism, many workers who do not have access 39 to 'deviant' left-wing subcultures will naturally vote Tory. This argument to some extent dovetails with those made by writers favouring the embourgeoisement hypothesis. It is argued that increased affluence enables workers to move into middle-class neighbourhoods thus exposing 34 them to the conservatizing influences of the middle-class, whether or not there has been a mass migration of workers to middle-class suburbs in recent times in England or elsewhere is open to question. Nevertheless evidence to support the argument that workers who live in middle-class areas are more likely to vote conservative has been assembled by a number of English researchers. Wilmott, for example, found that in one English city, Dagenham, an overwhelmingly working-class town, only 13% of workers assessed themselves as middle-class and less than 40 10% of all workers voted Conservatives. He contrasts this finding with the degree of self-assessed middle-class status expressed by workers in more heterogeneous communities. Thus in Greenwich, a predominantly middle-class town, 48% of the workers classified themselves 41 as middle-class. From this Wilmott concludes that: "The rule suggested by these four places is that the more the middle-class predominates in a district the more working-class people identify them-42 selves with it and, incidently, the more they vote Conservative." 43 The evidence presented by Wilmott and others, does not seem to be in contradiction with the results I presented earlier concerning self-placement in class by British workers. Middle-class identifiers were indeed more likely to vote Tory compared to those workers who called themselves working-class, (48.6% vs 19.9%). However it should also be recalled that this proportion, according to the Butler and Stokes data, represents only 19.3% of the total working-class support base for 35 the Conservative party. Furthermore, as previously noted, approximately half of these middle-class identifiers were in the bottom income category, the ones least likely to be able to afford a home in a middle-class area. Thus living in a middle-class area may indeed dispose workers towards the Tories, but this would be true only of an extremely small minority. The British Conservative party still draws most of its working-class support (54.5%) among workers from those who see themselves as working-class, and a somewhat smaller proportion (26.8%) from those who are unaware that they are working-class. There are no data available concerning the degree of 'middle-classness' of workers' neighbourhoods. Nevertheless on the basis of the data on the degree of 'middle-class' indentification, the fact that the majority of Conservative workers judge themselves to be working-class might indicate that they are still located within some sort of working-class subculture. The findings on Germany presented by Hamilton, and to some extent the data from the Almond and Verba German sample, would lead one to infer that CDU workers have not removed themselves from working-class subcultural influences. In Italy, needless to say, the fact that DC and right-wing workers tend to be concentrated in the low income groups compared to left workers, probably indicates that these workers as well do not reside in middle-class areas and are still within the cultural parameters of the working-class. The question remains then, how such workers can maintain their ability to vote for a conservative or a right-wing party in spite of subcultural influences to the contrary. One would imagine that there 36 are certain structural supports within the working-class subculture which will aid in the maintenance of a conservative voting identity. The aim in this chapter, therefore, is to see if there might be a conservative subculture within the working-class subculture. The alternative hypothesis might be that such a "sub"-culture, so to speak, does not exist. This in turn would suggest that the left-wing working-class subculture has very little in the way of a negative influence on conservative workers. Such a finding could also put the findings concerning the effects of income in a more favourable light. Finally one could make the argument that a lack of subcultural influences of any sort would indicate that the conservative vote is dependent entirely on the organizational talents of the various parties. I will now proceed to test the propositions in the reference group model outlined in the introduction. The first part of the chapter will be concerned with examining the social influences in a conservative worker's neighbourhood environment. The variables relevant to such a discussion are parental voting behaviour, church affiliation and church attendance, and rural-urban migration. The second part of the chapter will be devoted to an examination of the impact of the work-place to see what conditions may promote or mitigate conservative voting. 37 Subcultural Influences: The Neighbourhood Parental Voting: Unfortunately no data on parental voting behaviour is available on either West Germany or Italy. One could argue of course, that a discussion of parental voting behaviour is not particularly germane in the case of these two countries. Both the CDU and the DC did not come into being until after the second World War. Thus parental voting patterns would not have developed to a degree to be significant for the socialization of offspring. On the other hand it should be remembered that in Germany during the Weimar period there was an active Catholic 44 Centre Party to which the present day CDU can trace part of its lineage. The pre-war 'Partite Popolare Italiano' was somewhat shorter-lived and not as well known as the German 'Zentrum', but for a period after World War I (1919-1923) it did receive the approval of the Papacy and was steadily gaining electoral support until elections were 45 abolished by Mussolini. Since both pre-war parties were closely related to the Catholic Church, as are the present day CDU and DC, and since the family is an important link in the transmission of religious values, we can perhaps draw certain inferences from the data on religious behaviour to be presented later. Party preference, as transmitted via parents, is often rooted in some past historical cleavage. In the case of Great Britain, McKenzie and Silver and others have pointed out that large numbers of workers voted Conservative in the nineteenth century because in many ways the Tories, often under Disraeli, represented the interests of the 38 working-class in much better fashion than the Liberals under leaders 46 such as Gladstone. Although no longer relevant, this holdover from the past still seems to have some force in holding workers to the Tories, writers claim. Examination of Butler and Stokes data for Great Britain. (Table 1) reveals that the influence of working-class Conservative parents still appears to have some effect on the voting hehaviour of their children. However compared with Labour parents, the defection of offspring to the opposing party is considerably higher, 45.8% from the Tories to Labour and only 7.7% from the Labour party to the Conservatives. If one of the parents is Labour and the other Conser-vative, the odds are almost 70% in favour of offspring voting Labour. The probability of offspring remaining Tory is marginally higher if the mother is classified as 'don't know'. One might expect a large proportion of those having Conservative parents and still remaining Conservative to be affected by the influences of a more traditional subculture. Following the sections on religious behaviour and rural-urban migration we will briefly examine some data on the attitudes European workers have toward their political institutions. Such an examination might indicate whether or not conservative workers in the three countries are more traditional in their cultural and political values. Religion: In terms of the literature on European voting behaviour, we can expect the religious variable to have the greatest effect in 39 TABLE 1 Party Vote by Parental Voting Behaviour  Among British Workers Party Vote Parental Voting Behaviour Father One parent Father: Both Cons. Cons.: The Labour Both Both Parents Mother: other Mother: Parents Parents 'Don't 'Don't parent 'Don't Cons. Labour Know' Know' Labour Know' Total Sample Cons. Labour Liberal 45.8% 45.1 9.2 Total 100 per cent (N) (273) 7.7% 91.2 1.1 100 (181) 25.7% 68.0 6.3 100 (175) 48.6% 51.4 0.0 100 (35) 31.4% 68.6 0.0 100 (35) 11.4% 87.6 1.0 100 (105) 27.9% 67.3 4.0 100 (804) Source: Butler and Stokes Survey, 1963 Wave. 47 countries like West Germany and Italy. However it should be borne in mind that often it is not religious affiliation per se which is crucial but church attendance. This is especially the case in Italy where virtually the entire population is at least nominally Catholic. But even in the case of Great Britain it is argued, by Glock and Stark for example, that working-class Tory supporters are more inclined to attend church than Labour workers.^  Table 2 examines the relationship between party vote and religious affiliation for German workers. It is evident that in Germany Catholics tend to vote CDU while protestants prefer the SPD. 40 Furthermore the CDU draws nearly 68% of its working-class support from among this Catholic subculture. For Italy, since there were only four non-Catholics out of our sample of 373 workers, it is of course meaningless to draw inferences about party choice from religious affiliation. In Table 3 party vote and church attendance has been cross-tabulated for Germany and Italy. One can see that in Germany the higher the rate of church attendance the greater the probability of workers voting for the CDU. The case is even more clear cut for Italy. Whether or not one votes for the DC seens largely dependent on just one criterion, regular attendance at Mass. Over 85% of Italian workers who attend weekly or more vote DC, compared to 13.2% for workers who attend only on major holidays. The probability of voting DC declines precipitously if Church attendance drops below that of at least once a week. If one looks at Table 4 below for Germany it can be seen that when religion is controlled for in the relationship between regularity of Church attendance and party vote there is some difference with regard 49 to German Catholics. As in Italy German Catholic workers who attend Mass regularly are more inclined to vote CDU. However this disposition to vote CDU does not decline with lack of Church attendance as drastically as in Italy. This may be due to differences in the. nature of the Catholic Church in the two countries. During the latter part of the nineteenth century the Church in Germany and related organizations were heavily involved in the social welfare of Catholic 41 Party Vote TABLE 2 Party Vote By Religious Affiliation  West Germany Religious Affiliation Total None Protestant Catholic Other Sample CDU 11.1% 25.3% 61.3% 50.0% 41.7% S P D 88.9 74.7 38.7 50.0 58.3 Total 100 100 100 100 100 per cent (N) (9) (75) (75) (4) (163) Source: Almond and Verba Survey, 1959-60. workers.This may have imparted an image of the Church, and the closely related 'Zentrum', similar to that of Disraeli's Conservative party in England. In recent times the Church in Germany has made some attempts to mobilize the vote for the CDU but this has been more in the way of efforts by individuals and cannot be construed as a concerted effort on the part of the hierarchy. Furthermore Catholic workers in Germany have not been threatened with excommuniation if they chose to vote for the SPD. Unlike Italy the Church in Germany has never been delegitimized among many sectors of the working-class. In Italy the religious appeal of the DC has been, and still is, quite different. In contrast to the German hierarchy, which was TABLE 3 Party Vote by Church Attendance  West Germany  Church Attendance Major Weekly or Once in Holidays Total Party Vote more a while or less Sample CDU 75.0% 42.9% 20.9% 42.5% SPD 25.0 57.1 79.1 57.5 Total 100 100 100 100 per cent (N) (44) (49) (67) (160) Party Vote Italy Church Attendance Weekly or more Once in a while Major Holidays or less Total Sample DC RIGHT LEFT Total per cent (N) 85.8% 4.7 9.4 100 (106) 28.6% 14.3 57.1 100 (49) 13.2% 15.8 71.1 100 (38) 57.0% 9.3 33.7 100 (193) Source: Almond and Verba Survey, 1959-60. 43 relatively innovative with regard to the plights of workers, the social conscience of the Italian Church was considerably less well developed. Since the second World War the Italian Church has given more than passive support to the DC; the ecclesiastical hierarchy has intervened in most elections since the war. "Papal statements and Cardinals' dec-larations have urged Catholics to vote, on the well founded assumption that non-voting is most common among potential supporters of the DC."^  It is at the parish level, however, where the most direct pressure is applied to vote DC. The influence of the Church appears to be most pervasive among women voters. For example according to the Almond and Verba data, over 70% of those who attend weekly or more often are women. Thus for many members of the Italian working-class, religion, far from being an opiate is in fact a positive straight jacket. Murrary Edelman writes: ...(I have) heard enough accounts of the practice to be able to reach the confident conclusion that the threat of hellfire has been held before many a pious and uneducated woman while the more educated Church goer was likely to hear a lecture on the moral and political evils of Socialism and Communism. In spite of the much publicised "opening to the Left" in 1963 it seems that the Church still has a great deal of antipathy towards 53 socialism in general and the PCI in particular. Especially at the local level, the priest (and to some extent the lay Catholic organization, Catholic Action) still continues to invoke the threat of hellfire on behalf of the DC. The Italian Church, therefore, performs a crucial role TABLE 4 Party Vote by Church Attendance for  German Catholics Church Attendance 44 Party Vote Weekly or more Once in a while Maj or Holidays or less Total Sample CDU SPD Total per cent (N) 81.1% 18.9 100 (37) 45.8% 54.2 100 (24) 38.5% 61.5 100 (13) 62.2% 37.8 100 (74) Source: Almond and Verba Study, 1959-60. Party Vote TABLE 5 Vote by Religion for British Workers Other Church of Church of Non Con-England Scotland Methodist formists Catholic None Total Sample Cons. Labour Liberal Total per cent (N) 30.5% 64.1 5.4 100 (557) 23.9% 76.1 0.0 100 (71) 16.1% 80.4 3.6 100 (56) 27.3% 60.6 12.1 100 (33) 22.2% 11.8% 27.7% 74.6 82.4 67.4 3.2 5.9 4.9 100 (63) 100 (17) 100 (797) Source: Butler and Stokes Survey, 1963 Wave. 45 in moulding and reinforcing the political attitudes of a substantial number of Italian workers. In conjunction with the family, the Church is probably the major agency through which workers are inducted into a DC subculture. Table 5 represents the relationship between party vote and religious affiliation among British workers. In contrast to Catholic workers in the other two countries, the British Catholic worker is much less inclined to vote for the Conservative party. However this mode of party preference in Britain probably has very little of a religious basis. Catholicism among British workers is, in the main, related to the immigration of Irish labourers in the late nineteenth and early twentieth century. Most of these people, mainly because of their low socio-economic position, were among the earliest supporters of the Labour party. Thus in Great Britain Catholicism in itself is probably of little importance in disposing workers toward one party or the other. This may not be true of Protestantism, however. In comparison the values of the various protestant religious may be of somewhat greater consequence in determining the voting behaviour of British workers. The Church of England, for example, has sometimes been called the 'Tory party at prayer'. Non-conformist sects such as the Methodists and Baptists, on the other hand,have historically been linked with the development of the Labour party.^  Thus one can see in Table 5 that workers affiliated with the Church of England are more inclined to 46 vote Tory while Methodists are least likely to vote Conservative. If one takes into account differences in rates of church attendance (see Table 6) it becomes obvious that workers who attend services at the Church of England at least once a week are much more likely to vote Conservative, 60% versus about 27% for the working-class as a whole or 39% for all workers who attend church regularly. One can see that for Methodists the probability of voting Tory increases with high attendance. One might have expected the Labour party to be favoured by workers who regularly attend these places of worship. It should be remembered however that for most members of the working-class religious affiliation is purely nominal.This is, I think, evident in the data concerning rates of participation. Religious affilia-tion probably reflects the influence of tradition, such attachments being transmitted through the family over time. Since non-attendance is the norm those workers who do attend may be upwardly mobile, or have status aspirations. This might account for the slight increase in the tendency to vote Tory."*^  Thus although British workers who attend church more often tend to vote Conservative, no undue causal significance should be attached to this variable. Aside from the relatively small number of workers involved, neither the Anglican Church nor any other church makes any effort to influence voting behaviour. If church attendance in Great Britain does have any influence it is probably in the way 47 of exposing workers to middle-class influences or perhaps the dominant societal values Parkin referred to, those broad values congruent with the organic values espoused by the Conservative party. TABLE 6 Percentage of Workers Voting Conservative  by Attendance Controlling for Religion (N's in Parentheses) Church of Church of Other Non- Catholic Marginals England Scotland Methodist Conformist High 60.0% 23.1% 38.5% 31.3% 26.8% 39.0% Attendance (N=40) (N=13X (N-131 (N=16l (N=4D Medium 33.3% 36.4% 9.1% 18.2% 14.3% 30.2% Attendance (N=180) (N=22) (N=22) (N-ll) (N-7) Low 25.5% 16.7% 9.5% 33.3% 13.3% 23.1% Attendance (N=337) (N=36) (N=21) (N=6) (N=15) Rural-Urban Migration: This study is not directly concerned with the values and behaviour of rural peasants. Nevertheless a discussion of the socialization process at the rural level becomes important insofar as there has been a considerable amount of migration of workers from rural to urban areas in both Germany and Italy. In Great Britain the question is not really relevant. The British working-class is several generations beyond its rural ancestry; there has been no significant 58 exodus of farm workers to the cities since the last century. In Germany, although no precise figures are available, a large proportion of workers were originally born and raised in rural 48 communities. Richard Hamilton, in contrasting the German peasantry with the French peasantry, suggests that the German peasant is considerably more conservative and upon arrival in an industrial 60 environment will continue to vote for the CDU or a right-wing party. Although the sample is rather small the data presented in Table 7 below provides at least a partial confirmation of this hypothesis. Those workers who were born in a rural area but currently residing in a city 100,000 in size or larger, are much more likely to vote for the CDU than workers who were born (and currently residing) in large urban areas, 64.3% versus 27.3%. In Italy the impact of rural-urban migration assumes even greater importance. There has been a constant stream of migrants from rural hinterlands to industrial centres; well over four million left for Northern cities between 1950 and 1964.^  In contrast to Germany, however, there appears to be a singular lack of uniformity in the sorts of attitudes that those migrants bring to the cities. According to Franco Alberoni, southern migrants to the North seem to have strong drives to become assimilated into their new Northern environment 62 and political conversion is one way of conforming. Thus if southern peasants fall within the organizational ambit of the Communist party, they are easily converted into Communist voters, in spite of the fact that in the south many of them voted for the DC or neo-fascist parties. However, Fried points out, if they are easily converted their allegiance 63 is not particularly strong. Their newly acquired political identities 49 TABLE 7 Party Vote by Migration West Germany Born in rural Born and currently area, currently residing in city All other living in city 100,000 workers 100,000 plus plus CDU 41.7% 64.3% 27.3% SPD 58.3 35.7 72.7 Total 100 100 100 per cent (N) (127) (14) (22) Source: Almond and Verba Survey, 1959-60 are still largely based on 1clientisme'. In the slum areas of Rome for example, the MSI has captured the eclectic allegiance of numerous 64 southern migrants. In contrast peasants from the North-East, areas such as Veneto, tend to retain both their allegiance to the Church and their DC voting identity. Fried notes that North-East peasants in urban areas have tended to settle in colonies, in this manner thus insulating their traditional Catholic culture from anti-clerical leftist sentiments in the cities. 50 Traditionalism and 'Working-Class Conservatism': One of the concepts that crops up repeatedly in discussions of British working-class Conservatism is the notion of deference. The term originated with Bagehot in the nineteenth century who defined it as the abnegation of authority on the part of the lower classes to those of higher social standing, the former perceiving the latter to be better fitted for governing the country. "Certain persons are by common consent agreed to be wiser than others, and their opinion is, by consent to rank for much more than its numerical value.Deferential attitudes on the part of British working-class Tories are usually seen as forming the basis of a broader complex of attitudes of which the predilection for voting Conservative is but one part. Along with a Conservative voting identity, parents also transmit traditional, deferential attitudes to .. . . 66 their offspring. In the other two countries it would be false to say that workers who vote CDU/DC are also 'deferential' in their attitudes. Nevertheless, given the conservatizing influences from the Catholic Church and other sources we might also expect such workers to be more 'traditional' in their views on political leaders and institutions. Unfortunately the data available do not allow me to test directly whether or not conservative workers in the three countries are more deferential or traditional in their attitudes. However the Almond and Verba survey does contain a number of attitudinal items concerning political efficacy and political trust. This has permitted 51 the .construction of two attitude scales which might serve as indirect measures of 'traditionalism'. The two scales are the political efficacy scale and the administrative trust scale. The former scale is designed to tap how respondents evaluate their capabilities to affect governmental operations and policies. A high score on this scale would tend to indicate not only that a respondent felt competent to approach governmental agencies but also that the govern-ment is perceived as being willing to listen and, if possible, to respond in accordance with his wishes. This scale includes such items as, "What would you do if a harmful law was passed by the national ,,67 government. The items in the administrative trust scale pertain not to specifically political objects but rather to civil service organizations such as the tax office and the police. A high score on this scale might reflect an individual's acceptance of and trust in the basic adminis-trative institutions of his country. Such an individual might conceivably be distrustful of politicians but at the same time still 68 have a high regard for the above named institutions. It seems plausible to argue that the social.pressures from a left-wing working-class subculture might lead to the withdrawal from many social and political interactions on the part of conservative workers. Furthermore in a country like Italy, where there has been considerable antipathy between Church and State, it is possible that the Church has imparted those malevolent feelings to members of the 52 working-class who support the DG. Since a high score on the efficacy scale implies that the respondent tends to be a more active and aware citizen we might expect conservative workers in all three countries to score lower on this scale compared to left workers. In Great Britain deference on the part of workers is supposedly directed at persons of high social standing, political leaders and so on. Nevertheless deference may also be generalized to more mundane public institutions such as the tax office and the police. Thus we might expect British working-class Tories to score higher on the administrative trust scale than Labour workers. For Germany we might also predict CDU workers to score highly on administrative trust but not necessarily for reasons directly related to the British notion of deference. There has always been traditional authoritarian strain in German political culture which emphasizes respect for bureaucratic officialdom, a respect often psychologically rooted in a sort of Germanic 'Angst'.^  In Italy institutions such as the tax office and the police are probably more closely identified with government and party politics.^ Thus it might seem reasonable to say that the relation-ship between DC voting and administrative trust would tend to approximate that of the one between DC voting and efficacy. For each country mean scores on the two scales have been computed for conservative workers and left-workers. The results are shown 53 72 below in Table 8. As can be seen, xn Great Britain Conservative workers are slightly more 'efficacious' than Labour workers. One might have thought that such workers would score lower. However this does not necessarily mean that Conservative workers are more active. Their relatively high mean efficacy score might be a reflection of the high esteem in which they hold British political' institutions and the Conservative party. The prediction concerning mean scores on TABLE 8 Mean Scores for Conservative Workers and Left Workers Political Efficacy Administrative Trust Political Efficacy Administrative Trust Political Efficacy Administrative Trust U.K. Conservative Labour 2.014 1.961 3.296 2.989 West Germany CDU SPD 1.441 1.747 3.088 2.642 ^aly DC Left 1.117 1.928 2.775 2.406 (Maximum Possible  Score) (5) (4) (5) (4) (6) (4) 'administrative trust' for British workers seems to hold true. Conservative workers are more inclined to trust their police and tax offices. It is open to argument of course, whether or not this indicates 'deference'. At the minimum however, it does show that most Conservative 54 workers have a great deal of trust in, and are probably highly supportive of, the basic administrative organs of the British state. German CDU workers tend to score lower than SPD workers in terms of political efficacy. This is probably a reflection of their lack of interest in political activities. The hypothesis concerning the degree of 'administrative trust1 among CDU workers appears to be partially confirmed; CDU workers have a higher mean score compared to SPD workers. I say partially confirmed, however, insofar the items in the scale do not necessarily indicate that this higher level of trust is based in certain of the authoritarian cultural traits discussed earlier. This question, like the notion of deference in Great Britain, would require further research with tools (i.e. a 'deference' scale and an 'authoritarianism' scale) which, at the moment,are not available. As expected in Italy DC workers are less efficacious than left-workers. This is probably due to their distrust of Italian political institutions and their disinclination to take part in 73 political activities. The anti-state influence of the Church pro-bably has had some effect in moulding such attitudes. Contrary to expectations however, DC workers appear to have a high degree of trust in their tax offices and police compared to left-workers. Thus their lack of interest in and alienation from the political system (defined in terms of the legislature and the party system) is not generalized to the above named bureaucratic institutions. It is difficult to specify in the three countries what cultural norms influence a worker when he gives either a negative or positive response to the items in the two scales. As mentioned earlier these scales at Best are indirect measures of traditional attitudes that might be associated with working-class conservatism. Nevertheless the difference in mean scores on 'political efficacy' and 'administrative trust', especially in Italy and Germany, does provide some evidence that conservative workers have acquired more traditional cultural norms as well as a conservative voting identity during their socialization period. In the next section we will examine what conditions at a conservative worker's place of employment may sustain this voting identity and associated attitudes. Subcultural Influences: The Work Place The European worker, or any worker for that matter, spends the major portion of his waking hours at his place of employment. The work-place then would seem to be a crucial structural component in either mitigating or promoting conservative voting among workers. Contacts with his work-mates, union officials and supervisors are among the various social interactions which might influence a worker's mode of thinking and behaving. The factory environment is usually seen as being at the centre of any left-wing 'Gemeinschaft' that might exist and therefore 56 militate against the development of conservative attitudes. The working-class housewife in contrast is concerned with matters related to her family and the immediate neighbourhood environment. If one can assume for the moment that a woman's domestic environment is in large part responsible for her social and political attitudes (i.e. that such-attitudes are not purely a function of her sex or other psychological characteristics associated with that sex) a crude indicator of the influence of the work place might to control the relationship between voting and class for sex. As can be seen in Table 9, in all three countries Conservative parties seem to draw a disproportionate number of women voters compared to left parties, although the difference is not as pronounced in Great Britain. From this one might assume that the work place influences the worker in a left-ward direction. However there still remains a question as to what aspects of the work place actually affect the worker. Furthermore since a large proportion of male workers, especially in Great Britain and Germany, still vote conservative there must be certain plants or occupations which do not unduly affect a worker's preference for a conservative party. One variable which is usually thought to be associate with working-class voting behaviour is plant size. In smaller plants there is greater probability that the worker has face-to-face contact with his employer opening the worker up to conservative influences. 57 TABLE 9 Party Vote by Sex Among Uorkers  Great Britain Total Party Male Female Sample Cons. 31.3% 33.5% 32.3% Labour 66.5 61.3 64.0 Liberal 2.2 5.2 3.6 Total 100 100 100 per cent (N) (227) (212) (439) Germany Total Party Male Female Sample CDU 37.1% 61.3% 41.7% SPD 62.9 38.7 58.3 Total 100 100 100 per cent (N) (132) (31) (163) Italy Total Party Male Female Sample DC 43.3% 68.3% 56.1% RIGHT 11.3 6.9 9.1 LEFT 45.4 24.8 34.8 Total 100 100 100 per cent (N) (97) (101) (198) Source: Almond and Verba 58 Larger plants are more homogeneous, more alienating thus facilitating 74 the propagation of left-wing ideas. Juan Linz, for example, in his study of West Germany claims that plant-size has an effect independent of all other possible variables, the larger the plant, the higher the probability of left voting.This finding, however is not always replicated in other studies. Richard Hamilton notes that in France the medium sized plants are the ones most prone to left-wing radicalism.^  Anomalous findings have also been reported for Great Britain?'' Plant size may well have an independent effect as Linz argues. However this relationship is probably also affected by other factors such as the nature of occupations and perhaps regional differences. But more important the plant must also be seen as the context in which the worker is exposed to cross-pressures arising from organizational efforts on the part of politically oriented unions. Thus in the case of France, Hamilton claims that the high degree of radicalism in medium sized plants is due to the influence of the communist dominated CGT. Occupation: In none of the surveys is there a question pertaining to plant size. However since plant size is closely linked with occupation, a detailed breakdown of the latter category might be even more useful. The Butler and Stokes survey contains this information for Great Britain. Unfortunately the Almond and Verba survey does not have a detailed breakdown for occupation, thereby restricting our focus to Great Britain for the moment. / 59 TABLE 10 Party Vote by Occupation Glass  Chemical, Furnace Wood- Leather-Party Vote Mining and Mill Electrical workers workers CONS 4.2% 21.2% 33.2% 31.6% 50.0% LABOUR 95.8 75.0 61.3 63.2 45.0 LIBERAL 0.0 3.8 5.5 5.3 5.0 Total 100 100 100 100 100 per cent (N) (48) (52) (199) (38) (20) Source: Butler and Stokes Survey, 1965 Wave As one would expect workers in more traditional occupations tend to be more conservative. A representative sample of the numerous occupational categories is presented in Table 10. Leather workers, plying a trade which is presumably more traditional, tend to be the most conservative. In contrast very few working-class conservatives are to be found among miners. Conservatives are under-represented in heavier industries, steel, chemical, glass and so on. Wood workers are somewhat more inclined to vote Conservative as are electrical workers. In the case of the latter category, the definition 'electrical worker' includes such people as electronics workers, radar technician and similar workers. The somewhat higher probability of 60 these workers voting conservative might be due to the following: the firms tend to be smaller thus facilitating interaction betv/een employer and employee; the nature of the work itself may entail a great deal of contact with white-collar personnel; and finally these occupations are probably fairly well paid. This may be interpreted as providing some evidence for the embourgeoisement hypothesis. But it would be the social influences associated with the work place and not necessarily income by itself which is of importance in disposing such workers to the Tories. Thus occupational differences might account for the curvili-near relationship between conservative voting and income outlined in the second chapter. For example,leatherworkers, employed in small firms would still be influenced by traditional influences from the past. Workers in, say steel mills, earning a medium sized income would vote labour mainly because such occupations have historically 78 been related to strong trade union solidarity. Workers in electronics industries might vote Conservative because of their exposure to a more white-collar milieu. To test this more thoroughly occupation was used as a control in the relationship between Conservative voting and income. Theoretically the income curve should disappear. We should also find leatherworkers to be concentrated in the lower income category while electrical workers are more likely to be in the high income bracket. 61 Our expectations are partially borne out (see Table 11). Nevertheless within three of the four occupational categories the curvilinear relationship for income is still there. Only for steel and chemical workers does the curve flatten out. Although the cell sizes are somewhat low, income still seems to have an independent effect. However it should be noted that the occupational categories used make no distinction between different sorts of workers within the same category. There is the possibility that socio-economic hierarchies within the various occupations may have some effect in explaining away the income curve. Skill: An alternative way of looking at occupations is in terms of differences between skilled and unskilled workers. The Butler and Stokes data set again provides the most detailed breakdown, differen-tiating between domestic servants, unskilled, semi-skilled, skilled and foremen. The Almond and Verba survey makes distinctions only between skilled, and unskilled and those employed in domestic service. Foremen and similar workers in directive or supervisory positions come into contact more often with their white-collar superiors. The expectation then, would be that they tend to vote conservative. As for skilled workers the possession of specialized knowledge or skills might confer upon such a worker a special status. i He may therefore feel himself above less skilled workers, consequently voting conservative. On the other hand, he would still be very much TABLE 11 Leatherworkers Low Cons. Labour Liberal Total per cent (N) 60.0% 30.0 10.0 100 (10) Income Medium 20.0% 80.0 0.0 100 (5) High 50.0% 50.0 0.0 100 (4) Vote by Income Controlling for Occupation Electrical Workers Total Sample 47.4 47.4 5.3 100 (19) Cons. Labour Liberal Low 28.6% 67.9 3.6 100 Total per cent (N) (28) Income Medium 22.7% 72.0 5.3 100 (75) High 39.2% 53.2 7.6 100 (79) Total Sample 30.8 63.2 6.0 100 (182) Cons. Labour Liberal Total Glass, Chemical, Furnace, Mill Workers Income Low Medium 23.1% 13.3% 61.5 86.7 15.4 0.0 100 per cent (N) (13) 100 (15) High 10.0% 90.0 0.0 100 (20) Total Sample 14.6 81.3 4.2 100 (48) Cons. Labour Liberal Low 50.0% 50.0 0.0 100 Total per cent (N) (10) Woodworkers  Income Medium High 15.4% 35.7% 76.9 57.1 7.7 7.1 100 (13) 100 (14) Total Sample 32.4 62.2 5.4 100 (37) Source: Butler and Stokes Survey, 1963 Wave ON Is) 63 part of the working-class subculture. Furthermore in a country like Germany, as Linz points out, skilled workers tend to be protestant, 79 work in larger firms and live in larger cities. The results for Germany and Italy are presented in Table 12. One can see that in both countries it is the unskilled worker who tends to favour the CDU/DC. Since skill level is probably related to income in some manner it might be wise to examine income differences within the skill categories (Tables not shown). For the unskilled category in Italy there is a straight decline in DC support as income rises, 90.0% of the low income unskilled workers supporting the DC vs 50.0% for the high income unskilled. A similar relationship is evident for the skilled workers except for the top income category where the level of DC voting rises somewhat. It might be that a number of skilled workers at this income level have a supervisory role which would account for their DC support. In Germany there is a slight decline in CDU support among the skilled workers as income rises. The interesting finding however is for the unskilled category (see Table 13). Again CDU support declines as income rises except for the top income category where the probability of voting CDU rises to 64.3%. Since these are unskilled workers it is unlikely that they are in supervisory positions. The embourgeoisement argument might perhaps apply to these workers. Unfortunately due to the rather small overall sample it is impossible to institute further controls. Nonetheless it should be noted that they represent 64 approximately 5.0% of the total number of workers sampled and nearly 15% of the CDU's working-class support. Table 14 derived from the Butler and Stokes data gives a good indication of the relationship between skill level and the probability of voting Conservative. Servants and related workers, probably most archetypical of the classic deferential worker described by Bagehot, are most likely to vote Tory. Foremen and supervisors also tend to vote Conservative. As one moves down the skill categories we can see that unskilled workers are least likely to vote Tory. TABLE 12 Party Vote by Skill Level Germany Personal Total Skilled Service Unskilled Sample CDU 36.4% 75.0% 44.8% 41.7% SPD 63.6 . 25.0 55.2 58.3 Total per cent 100 100 100 100 (N) (88) (8) (67) (163) Italy Personal Total Skilled Service Unskilled Sample DC 42.1% 28.6% 67.0% 56.1% RIGHT 18.4 28.6 1.7 9.1 LEFT 39.5 42.9 31.3 34.8 Total per 100 100 100 100 cent (N) (76) (7) (115) (198) Source: Almond and Verba Survey, 1959-60. 65 TABLE 13 Vote by Income for German Unskilled Workers Income Medium Medium Total Low Low High High Sample CDU 50.0% 37.5% 40.0% 64.3% 46.2% SPD 50.0 62.5 60.0 35.7 53.8 Total 100 100 100 100 100 per cent (N) (10) (16) (25) (14) (65) Source: Almond and Verba Survey, 1959-60. The next question is of course whether or not income still has an independent effect. Are foremen also the most affluent? In that case one might be able to argue that their propensity to vote Tory is due not so much to their high income but to the fact that they are partially removed and alienated from the working-class subculture. Looking at Table 15 below one can see that foremen are indeed among the better paid workers. The more significant finding, however, is that the curvilinear relationship between income and Conservative voting seems to persist within the various skill categories. The differences involved in some categories (i.e. the skilled) are not that pronounced and also certain cell sizes are rather small. But the fact that the same relationship reoccurs over all five categories seems to lend further credence to the argument that high income has an 66 TABLE 14 Vote by Skill Level for British Workers CONS Semi- Total Servants Foremen Skilled Skilled Unskilled Sample 52.7% 40.9% 26.1% 22.9% 19.0% 27.9% LABOUR 38.2 52.7 69.3 73.4 76.7 67.3 LIBERAL 9.1 Total 100 per cent (N) (55) 6.5 100 (93) 4.5 100 (352) 3.7 100 (188) 4.3 100 (116) 4.9 100 (804) Source: Butler and Stokes Survey, 1963 Wave 67 independent effect in disposing workers to vote s Conservative. TABLE 15 Percentage Vot :ing Conservative by Income Controlling for Skill (N's in Parentheses) Low Income Medium Income High Income Total Servants 66.7% 28.6% 50.0% 52.1% (24) (141 (10) Foremen 40.0% 20.7% 47.9% 40.9% (10) (29) (48) Skilled 33.7% 20.6% 24.0% 26.1% (86) (136) (104) Semi-Skilled 27.0% 16.9% 23.7% 22.9% (74) (59) (38) Unskilled 23.7% 5.3 27.3% 19.0% (59) (38) (11) Union Membership: The Almond and Verba data are rather ambiguous with regard to union membership. This variable is incorporated into a multi-response category concerning membership in a wide range of organizations. Below in Table 16 party vote is cross-tabulated with union membership for Italy and Germany. In Italy non-union workers are somewhat more likely to vote for the DC. Unfortunately for Italy the total number of unionized workers in Table 16 is rather low. This may be largely an artifact of 68 TABLE 16 Party Vote by Union Membership Among Workers DC RIGHT LEFT Total per cent (N) Italy Union Membership? Yes 41.7% 8.3 50.0 100 (12) No 58.0% 9.2 32.8 100 (174) Total Sample 57.0% 9.1 33.9 100 (186) West Germany Union Membership? CDU SPD Yes 28.1% 71.9 No 51.7% 48.3 Total Sample 41.7% 58.3 Total 100 per cent (N) (64) 100 (87) 100 (151) Source: Almond and Verba Survey, 1959-60 69 the way the question was asked in the Almond and Verba survey. But at the same time this low number may also be due to the nature of industrial relations in Italy. There is no such thing as a closed 80 shop therefore union membership is strictly optional. This does not mean that unions have little effect. Workers in virtually all industries above the small firm level vote in elections for one of three unions. The winning union then acts on behalf of workers in collective 81 bargaining for a whole series of plants. Of the three unions the Communist controlled CGIL is the strongest; it appears to have the most members and is usually most successful in plant elections. g The CISL, the Catholic supported union, also draws considerable support. During plant elections, and elections for supposedly neutral plant grievance committees, a considerable amount of organizational force is brought to bear upon workers by the various unions. "Election campaigns are hard fought, and often entail the expenditure of subs-tantial amounts of money..." A similar amount of energy and money is expended on the mobilization of workers for a variety of political acts ranging from mere voting in national elections to massive street 83 demonstrations. The CGIL is of course notorious for the latter type of activity. Although there are only a few formal connexions between the two, it is through the CISL that the DC attempts to retain the loyalty of the Catholic worker. There is also the "Christian Association of 70 Italian Workers" (ACLI) but its membership is based mainly in the lower middle-class. The CISL may have some success in insulating the serious Catholic worker from counter-pressures in the work-place thus bolstering its comparatively weak support base among urban workers and, as noted, has been successful in a number of plant elections. However, according to Edelman, support for the CISL in plant elections may be deceiving; many workers vote for the CISL in plant elections 84 but continue to vote for left parties in political elections. The short range appeal of the CISL probably rests on its ability to promise specific concessions from management, who for the most part, are also associated with the DC. The situation may vary from plant to plant and in different industries but on the whole the strongly Catholic worker who votes DC is likely to be exposed to fairly strong left-wing influences. As shown in Table 16 union membership is more pervasive among West German workers. A larger proportion of SPD workers are unionized compared to those voting CDU. However undue importance should not be attached to the possible left-wing effects of German trade unions. In fact, in certain circumstances union membership may reinforce a Catholic worker's disposition to vote CDU. During Weimar there were separate Catholic and Protestant trade unions, the former often cementing and reinforcing the relationship between the worker, the Catholic Church and the Catholic Centre Party. The post-war DGB (German Trade Union Federation) however is an amalgamation 71 of many of these religious trade unions. The current leadership, constituted of both Catholics and Protestants sometimes belonging to opposing parties, is prevented (both by law and mutual self-restraint) from making political announcements. This, however, does not always prevent individual DGB unions from giving a political orientation to 85 their members. What frequently happens is that power is dispersed and brought down to the individual factory level. Union leadership at the factory may try and politicize workers, the direction and intensity of such politicization depending upon who has control of 86 the union hierarchy within any given factory. It is quite possible that a plant with a predominantly Catholic work force is likely to elect a Catholic leadership who may wish to proselytize for the CDU. This might occur in regions like North-Rhine-Westphalia. In other cases a mixed union hierarchy would prevent such activity. Overall,the impression seems to be that local leaders, even if they are in a position to do so, take the edict of official neutrality seriously, and are not prone to begin mobilizing their constituencies on behalf 87 of their political party. This unwillingness to take on political duties may be due to the unusual role that union leaders have in management of the factory. Industrial management in Germany is structured quite differently in comparison to other countries. Instead of separate shop union stewards and company managers as in England, the management of industries 72 in Germany is 'co-determined' by works councils composed of management and labour. However, as Dahrendorf notes, co-determination often has the effect of co-opting union leadership into the goals and values of 88 management. This in turn has the result of isolating the union leadership from the members at large often leading to a de-politicali-zation at the factory level. Because of the DGB's official neutrality and the policy of co-determination an alternative way of politicizing workers is by the formation of party organizations within factories outside of the union structure. However, as Chalmers points out, both political parties, especially the SPD, are cautious in not overplaying the use of these 'Betriefsgruppe' largely to avoid disunifying tendencies within the 89 DGB or do damage to the policy of co-determination. The end result is often a vacuum at the factory level. Thus for the German CDU worker there might still be some informal pressure from his work mates but on the whole his work situation is not always apt to expose him to the cross-pressures of a strong left-wing subculture. Of all three countries British trade unions are potentially the most powerful in instilling left-wing attitudes among workers. Although the closed shop rule is in force in only a minority of firms, 90 many industries still have 100% union-membership among workers. There has always been a stong historical relationship between trade unions and the Labour party. Currently officials of the Trade Union 73 Council (TUC) sit on the council of the Labour party and many Labour party MP's are trade unionists. Even in Italy the DC and the CISL are supposedly independent organizations. Therefore we would expect that British workers belonging to a trade union would be much less likely to vote Conservative. In Table 17 one can see that this is the case. Nevertheless since the majority of British workers are unionized, the Conservatives still draw nearly half of their support from union members. If Conservative union members are strongly pressured by a sort of left-wing "Gemeinschaft' one would imagine that such persons take some sort of evasive action or develop a state of mind to insulate themselves from their unionized industrial environment. Eric Norlinger, for example, has noted that working-class Tories interact less with their 91 fellow workers. One of the questions asked by the Butler and Stokes survey was: "How close do you feel to your union?" Of all Conservative union members 63.0% reported that they did not feel close to their union. In contrast only 38.7% of Labour union members replied in the negative. This would indicate then that Conservative unionists feel themselves to be alienated from their union and perhaps their fellow workers. However as in Germany the effects of union membership should not be exaggerated. Although the TUC has strong formal links with the Labour party in reality there is often a great deal of antagonism between the two bodies. It would be patently false to say that the 74 party in any sense controls the TUC. The latter organization may still give financial support and man polls at election time but ther"e is some question as to the degree to which unions attempt to politicize and mobilize their own constituencies, namely the workers. Butler and Stokes claim that unions' communication networks, from the point 92 of view of achieving political aims, are at best neutral. Union publications aimed at workers deal only indirectly with political affairs and union members seem to have little cognizance of articles that are political in nature. Surprisingly enough it is within the walls of the factory, however, where there is the least evidence of union political activity. In the 1964 wave of the Butler and Stokes survey; carried out just after the election of that year, only eight out of a sample of 484 workers responded in the affirmative when asked if they had been approached by a trade union official on behalf of a political party. Butler and Stokes claim that the majority of workers join unions on the basis of self-selection which in turn 93 is based on tradition passed on from father to son. This suggests that union membership is mainly a function of a persisting, underlying Labour subculture. Therefore, as in West Germany, the Conservative worker who, for a variety of reasons, is forced to join a union (i.e. closed shop rules), may to some degree be cross pressured but not unduly so. The next question is again related to embourgeoisement. What happens when we use union membership as a control in the relation-75 TABLE 17 Vote by Union Membership for British Workers Belong to Union? Total Yes No Sample CONS. 22.4% 36.8% 28.0 LABOUR 73.3 57.4 67.1 LIBERAL 4.3 5.8 4.9 Total 100 100 100 per cent (N) (487) (310) (797) Source: Butler and Stokes Survey. ship between party choice and income. The results are shown in Table 18. They suggest that the lack of membership in a union greatly increases the probability of voting conservative among the top Income workers. However there is still that anomalous middle income category. Even among non-union workers only 20.0% of respondents in this particular income bracket vote Conservative. Union membership (and probably the subcultural influences associated with it) tends to flatten the curve considerably but it is still there to a slight degree. So far we can only say that the relationship between conservative voting and income posited in Chapter II, the revised version of the embourgeoisement argument, cannot be unequivocally rejected. For Italy and Germany it was suggested that subcultural 76 influences, mainly related to religion but also to region and city size, could still explain a considerable amount of the variance in 'predicting' working-class conservatism. The phenomenon of the 'affluent' conservative worker in Italy is extremely rare. In West Germany, however, there is a relatively small minority of workers whos tendency to vote CDU might be due to their comparative affluence. Unfortunately the limitations of the data make it difficult to impose the controls necessary to test the embourgeoisement argument in more rigorous fashion. TABLE 18 Party Vote by Income Controlling for Union Membership Among British Workers Union Members Income Total Low Medium High Sample CONS 24.4% 17.3% 23.5% 21.4% LABOUR 71.5 79.8 71.9 74.8 LIBERAL 4.1 2.9 4.6 3.8 Total 100 100 100 100 per cent (N) (123) (173 (153) (449) 77 TABLE 18 (Cont'd.) Party Vote by Income Controlling for Union  Membership Among British Workers Non-Union Members Income Total Low Medium High Sample CONS. 41.3% 20.0% 50.0% 35.6% LABOUR 53.2 71.0 46.6 58.1 LIBERAL 5.6 9.0 3.4 6.3 Total 100 100 100 100 per cent (N) (126) (100) (58) (284) Source: Butler and Stokes Survey For England the availability of the Butler and Stokes data has allowed a more intensive investigation of the embourgeoisement question. However, in spite of various controls, income still seems to have a pronounced effect on Conservative voting among workers. There appear to be two sorts of working-class conservatism. One group, by far the majority, seems to be located in a lower income, traditional type of subculture, the social influences inherent in such a subculture disposing them towards the Tories. The other group, the minority, appear to be influenced in their voting behaviour by a relatively high level of income. How this high level of income actually affects such a voter, whether it affects him at all or whether 78 it is associated with a certain type of social milieu, is a puzzle which remains unresolved. Many of the criticisms that were raised against the embourgeoisement argument at the end of Chapter II can still be regarded as valid ones. The notion that a Labour voter will switch to the Conservatives upon crossing a hypothetical income threshold has not yet been proven. In the next chapter I will offer some tentative explanations that might clear away some of the mystery surrounding our 'affluent' conservative worker. . 79 CHAPTER IV Conclusion It was argued in the introduction that subcultural influences were most important in explaining the working-class vote for the Conservative party in Great Britain, the CDU in West Germany and the DC in Italy. There is still some question as to the precise manner in which subcultural primary groups affect a worker's choice of political party. For example a worker who attends church regularly may vote conservative in order to conform to his social environment. Alternatively, such a worker may vote conservative because he genuinely values his religious beliefs. As was noted in the first chapter, a great deal more detailed information on individual workers is required before such questions can be dealt with adequately. Nevertheless we can say that in all three countries the five proposi-tions outlined in my model, concerning parental socialization, neighbourhood environment, church attendance, occupation and union-membership have been confirmed to some degree. That is not to say that a variable like church attendance is of equal weight in explaining working-class conservatism in each country. For one thing the limitations of the data prevented the proper evaluation and direct testing of some of the propositions. The lack of a question regarding 80 parental voting behaviour in the Almond and Verba survey is one such example. The more significant point, however, emphasized throughout the thesis, is that the importance of each of the five variables varies from country to country. In Germany CDU support among workers appears to be mainly based in a Catholic subculture. Workers who are devout Catholics, non-unionized and unskilled are most likely to vote for the CDU. The rural origins of several CDU workers may also have a role in predisposing them towards the CDU. The religious values of many CDU workers may have some effect in alienating them from the working-class subculture at large. The lower rate of unionization among CDU workers might be a reflection of this. Union membership represents part of a larger left-wing subculture in which they are unwilling to join. This might be further indicated by their low feelings of efficacy. Many of these characteristics associated with working-class CDU voters, low participation and so on, may be explained, or at least predicted by, one single factor, namely sex. Working-class women, who constitute a substantial segment of CDU working-class support, are most likely to be beyond the influence of the work place and at the same time to be strongly bound to the Church. It seems probably that a number of male Catholic workers are likely to succumb to left-wing pressures prevailing in their place of work. But this cross-pressuring in their work-place need not be overriding. 81 Unions are by law politically neutral. Furthermore, on the whole, the policy of co-determination, agreed to by both the CDU and the SPD, prevents party activities of either the official or the un-official type. Thus if religious beliefs or other feelings strongly dispose a worker towards the CDU, he need not be unduly alienated from his work-place. In all this the role of leadership should not be neglected. The stable, fatherly image projected by Adenauer may have had great appeal to many German workers. More importantly the deliberate effort by the CDU to recruit leaders from the pre-war Catholic Centre party and Catholic Trade Unions, probably helped to enhance their support in the Catholic working-class subculture. For Italian workers the polarization between the Church and the influence of the work-place is much greater than in Germany. Over 60% of DC support in the Italian working-class is derived from women, most of whom are strongly religious. Of the total number of DC workers, over 80% attend mass once or more per week. The probability of a left worker attending mass weekly or more is extremely low in comparison. Due to the militancy of left wing Italian trade unions, such as the CGIL, and the strong left-wing traditions among many workers, the potential DC worker is probably subjected to a great deal of cross-pressure and may conform accor-dingly. In some cases the DC worker may find some structural support in the DC oriented trade union, the CISL, if it is active in his place 82 of work. Like CDU workers in Germany, DC workers compared to left-workers, are much less inclined to feel efficacious. Again as in Germany this is probably a reflection of their alienation from the working-class subculture. These attitudes may also be engendered by the methods used by the DC to mobilize DC workers. The reactionary modes of operation of the Parish priest, Catholic Action and many DC leaders is probably not conducive to the creation of positive feelings toward the Italian political system. In both Germany and Italy the difference between left and Conservative voting among workers is structured mainly around the religious cleavage. This is not the case in Great Britain. What we can point to, however, are two traditional subcultures, pne older and the other of more recent origin, yet both residing within a wider working-class subculture. The Labour subculture, the newer one, is concentrated mainly in areas where there is a heavy concentration of workers. These people work in the medium or heavy industries and tend to be unionized. The Conservative subculture is more deeply rooted in past historical traditions, traditions dating back to 1867 when Disraeli and others made a strong and successful effort to capture the loyalties of the working-class. This historical attachment to the Conservative party is rather inefficiently trans-mitted via parental socialization and, as Butler and Stokes point out, is therefore being steadily eroded. A Conservative worker has a greater likelihood of retaining his voting identity, perhaps 83 even have it reinforced, if he works in a smaller, more traditional firm. There is also some evidence that Conservative workers tend to live in more heterogeneous neighbourhoods. Thus Conservative workers are to some extent insulated from the Labour subculture. Nevertheless the Conservative party still draws a considerable proportion of its support from workers employed in factories whose social climate is usually considered to be Labour in orientation. However, as in Germany, the left-wing ethos in these places need not be unduly oppressive. British trade unions appear to make little effort to mobilize workers on behalf of the Labour party. Thus one would be safe in saying that in all three countries a large proportion of the variance in the working-class vote can be explained by referring to a variety of subcultural influences. Unfortunately there is still the problem of the 'affluent' working-class conservative. In all three countries, workers voting CDU, DC or Conservative tend to be less well off compared to their left-wing counterparts. This is clearly the case in Italy where there is a straight linear * relationship; the higher the income the lower the probability of voting DC. In Germany and Great Britain, however, the relationship is curvilinear. The probability of voting conservative decreases as income increases with the exception of the top income category where the tendency to vote conservative increases again. Using the Almond 84 and Verba data, it appeared that only a relatively small number of 'affluent' workers were involved in Great Britain while in Germany there was a somewhat larger number. However, in terms of the Butler and Stokes data collected in 1963, it was apparent that the curvilinear relationship was also quite pronounced for Great Britain. Tests were made to see if this income curve was related to the embourgeoisement process. The results of these attempts, however, were inconclusive. Using the Butler and Stokes data it seemed that Conservatives in the top income bracket were more likely to perceive themselves as 'middle-class' compared to Conservatives in the medium income bracket. But this finding was partly obviated by the fact that Conservatives in the lowest income bracket were most likely to see themselves as middle-class. For Germany the overall size of the Almond and Verba sample and the lack of necessary controls, made it difficult to further explore this curvilinear relationship. Using the Butler and Stokes data, when occupation and skill level were controlled, though there was variation in overall levels of support (i.e. foremen were more likely to support the Tories than unskilled workers), the curvilinear relationship persisted. High income workers were more likely to vote Conservative than medium income workers. Union membership seemed to have the greatest effect in explaining away the curve. Non-union, 'affluent' workers were by far 85 the most likely to vote Conservative. However, even among unionized workers there was still some evidence that 'affluent' workers were more inclined to vote Conservative. Thus as a variable income has some predictive power. By knowing a worker's occupation, parental voting behaviour, union-membership and income one can make an excellent prediction as to his likelihood of voting Conservative. For example, on running a multiple regression analysis using the Butler and Stokes data it was found that these four variables would explain over 20% of the variance in 94 Conservative working-class voting in Great Britain. Yet although we can give a satisfactory explanation why the combined effects of occupation, union and parents should influence a worker towards one party or the other, we cannot do the same thing for income. It is still unclear in what manner a high level of income affects a worker's party preference; whether or not high income in effect 'causes' Conservative voting. It is a question which is worth pursuing a little further. During the discussion of trade unions, it was pointed out that union membership or non-union membership, since it is in the main voluntary, probably indicated the existence of deeper underlying subculture, a Labour subculture and a Tory subculture. For the sake of argument let us say that high income is in some way related to a subculture. By taking into account the high income category one could say that there are in effect three subcultures. There is the 86 traditional Tory subculture, dating back to the days of Disraeli, concentrated mainly in the lower income bracket. Secondly there is the traditional Labour subculture represented by the union membership category. It is concentrated mainly in the medium income bracket but cuts into both the low and high income bracket. For example in Table 1, I have cross-tabulated party vote with income for all workers who reported themselves to be 'close to their unions'. Note that the curvilinear relationship disappears. It is this group of Labour supporters, 'close to their union', which probably constitutes the core of the Labour subculture. This subculture probably also contains a large number of non-union members in the medium income group, who, by virtue of neighbourhood environment and early socialization, would vote Labour. Finally there might be a newer, more fluid Tory subculture. This is represented mainly by the high income category. It is of course difficult to describe the type of worker occupying this subculture. 95 Sociologically one might want to call it the "new working-class". Under its rubric one could place workers employed in more service oriented industries, auto-mechanics, technicians and similar occupations, and possibly workers elsewhere whose income allows them to move to a middle-class area. The social influences involved are no doubt variegated and multifarious to say the least, but probably have the overall effect of insulating, or at least counter-balancing, the worker from the Labour subculture. Thus high income would in part be a 87 function of occupation, but in turn could have the independent effect of allowing such workers to remove themselves from the Labour subculture. To some extent this notion of a newer, 'affluent' subculture runs against the grain of much of the socialization literature. It is TABLE 1 Vote by Income for British Workers Who 'Feel Close to their Union' Low Income Medium High Total Sample CONS 5.7% 16.5% 16.1% 13.5% LABOUR 90.6 82.4 83.9 85.0 LIBERAL 3.8 1.2 0.0 1.5 Total per cent (N) 100 (53) 100 (85) 100 (62) 100 (200) Source: Butler and Stokes Survey, 1963 Wave. still difficult to envision a worker who after voting Labour for a long period of time would, upon crossing the 'affluent' threshold, begin voting Conservative, possible conservatizing effects of a new social milieu and aging effects notwithstanding. Paul Abramson, for example, has noted that even upwardly mobile Labour workers tend to retain their Labour voting identity.^  The question is then, would there be any 88 particular type of worker whose socialization experiences might make them likely candidates for the 'affluent' subculture? If one looks at Table 2 where party vote and income are controlled for parental voting it can be seen that income has little effect on the defection of Labour offspring. From this I think it is safe to assume that workers socialized into the Labour subculture aire quite stable in their voting behaviour from one election to the next. Furthermore it is unlikely that this type of worker will cast aside his union membership and other associations and vote Conservative upon attaining a high income. From Table 2 it would seem evident that workers who are apt to switch to the Conservatives Assuming that some of them did actually vote Labour at one point) would be those in the low and high income categories having Conservative parents. One might expect that the propensity of a worker to stabilize his Conservative voting identity would occur at a relatively early age. Furthermore, since this 'affluent' subculture is probably more of a post war phenomenon, we would expect it to have more of an effect on younger workers. A rather crude test for this might be an age cohort analysis, controlling for age level in the relationship between Conservative voting and income. Although the sample is rather small one can see in Table 3 that the curvilinear relationship is most pronounced for the youngest age cohort (The youngest age cohort capable of attaining a high income level is 25-29 years in age). The other age cohorts undoubtedly TABLE 2 Vote by Income Controlling for Parental Voting Both Parents Conservative Low Medium High Sample CONS 51.1% 31.6% 48.1% 44.1% LABOUR 40.2 57.9 43.0 46.6 LIBERAL 8.7 10.5 8.9 9.3 Total per cent 100 100 100 100 (N) (92) (76) (79) (247) Both Parents 'Don't Know' Income Total - Low Medium High Sample CONS 26.3% 18.8% 36.4% 25.2% LABOUR 68.4 73.9 57.6 68.6 LIBERAL 5.3 7.2 6.1 6.3 Total per cent 100 100 100 100 (N) (57) (69) (33) (159) Both Parents Labour Income Total Low Medium High Sample CONS 2.0% 8.5% 9.8% 7.0% LABOUR 96.9 90.1 90.2 91.9 LIBERAL 2.0 1.4 0.0 1.2 Total per cent 100 100 100 100 (N) (50) (71) (51) (172) 90 TABLE 3 Percentage Voting Conservative by Income  Controlling for Age (N's in Parentheses) Income Total Age Cohorts Low Medium High Sample 25-29 years 50.0% 7.1% 42.9% 27.6% (16) (28) (14) 30-39 years 13.3% 14.5% 23.5% 18.0% (30) (69) (68) 40-49 years 26.0% 16.5% 32.9% 24.6% C50) (79) (70) 50-59 years 30.3% 20.3% 33.3% 27.3% (76) (69) (42) 60-69 years 36.2% 29.2% 35.3% 34.1% (47) (24) (17) have been affected by past events such as the depression in the 1930's, the emergence of the Labour party in the 1920's and so on, which might confound the results somewhat. For the youngest age cohort, however, one might' assume that the Conservative party has been particularly successful in retaining the vote of younger workers in the High and Low income brackets. This hypothesis does not necessarily vitiate the argument made by Butler and Stokes, namely that working-class Conservatism is gradually on the decline and that the Labour party is occupying more 91 and more of its natural class base. As is evident even among affluent workers the defection rate of those with Conservative parents is considerably higher than those with Labour parents. It does suggest however, that the rearguard action being fought by the Conservatives is more successful in the high income category. The key then to the puzzle of the 'affluent' Conservative seems to lie in either the retention or the retrieval of workers whose parents voted Conservative, and to a lesser extent those whose parents were neutral. The above hypothesis is of course somewhat tenuous. It is an overview of a very complex phenomenon. A more definitive assessment would require an extensive and extremely detailed analysis. Assuming, however, that my argument has some validity, how would this illuminate the West German phenomenon? One may recall that 'affluence' and CDU voting appeared to be most closely related among unskilled workers. As Linz points out unskilled German workers tend to be Catholic, and, as our data shows unskilled workers are also more prone to vote CDU than skilled workers. I have no proof for this but it may be that the SPD working-class subculture, like the Labour one in Britain, tends to be fairly stable with a very low defection rate. Since CDU workers would be subject to more cross-pressuring there might be a higher defection rate. Perhaps CDU workers less likely to defect would be those who attain an income that would be sufficient for them to insulate themselves from the SPD subculture. 92 Does the embourgeoisement process as outlined above, have any relevance for Italy? If, for example, in the future workers attain an affluent income or an occupation associated with high income, allowing them to move beyond the left subculture, would this increase their probability of voting DC (assuming they were disposed that way because of religion and so on)? Such a prognosis is rife with complications. One must keep in mind the differential impact high income could have on the various regions. The fragmented nature of the Italian party system also complicates matters. For one thing neither the DC nor the PCI (the major left party) have been very successful as catch all parties. Furthermore given the manner in which the Church attempts to mobilize workers for the DC, there is an equal probability that an 'affluent' Italian worker would insulate himself from the Catholic subculture as well as the working-class subculture. Thus for Italy I think at this time it is best to suspend judgement. Implications The societal cleavages on which working-class conservatism is based are certainly not new ones. In all three countries the phenomenon is rooted in a fairly traditional subculture. Thus it is not a strictly post-war phenomenon, nor is it due to the wholesale defection of left workers to conservative parties. Post-war economic affluence, in Great Britain and Germany may have had the effect of cementing the loyalty of Conservative/CDU workers. Even among the low income conservative workers, the rise in their real income may have reinforced their long standing connexion to the Conservative party or in Germany, the Catholic subculture. On a broader level, economic development has no doubt wrought changes in the European social structure. The number of people employed in heavy industry may have declined relative to other 97 employment sectors. There appears to be a new more affluent working-class subculture. Workers in this group seem more inclined to vote conservative than workers in medium income categories. For Great Britain there is some evidence that many of these workers were exposed to some Conservative influence during their formative years This might be an example then of the new interacting with the old to form a new source of cleavage. However the size and significance of this group of voters is problematic. Due to their ambiguous location in the social structure they may well hold or come to hold a set.of beliefs similar to those held by Dahrendorf's "service class", an ideology which is individualistic in orientation and not 98 readily amenable to the acceptance of collectivist ideas. At the' moment there is no direct evidence to either confirm or refute this hypothesis. There is some evidence however, that the defection rate of Conservative offspring to Labour, even in the high income category, is higher than the rate from Labour to Conservative. Of course one needs to look at the absolute number of defectors involved as well as the percentages. But it seems quite possible that this group 94 of Conservative offspring will eventually be swallowed by the Labour subculture. The same process might be occurring in West Germany, though this prediction is based more on intuition than evidence. Numerous writers have documented the decline of partisanship among working-class voters as well as the electorate at large. The data available did not allow me to test for this but findings by other researchers would suggest that this is the case for our sample 99 of 'working-class conservatives'. Is this then an affirmation of Lipset's notion that "ideologically intransigent politics" is on the decline?^ ^^  My answer is yes and no. For a particular generation the age of affluence is undoubtedly the age of consensus. But it must be remembered that in our Butler and Stokes sample, for example, even our youngest age cohort was born before the war. All respondents in both the Almond and Verba and the Butler and Stokes surveys must have been affected in some way by the depression, the war and other events. The relative affluence and stability of the post-war period was probably a welcome relief, which in turn mitigated the appeal of more radical ideologies. However for newer generations, brought up in conditions of comparative comfort and insulated from the harsh realities of past events, notions of economic security and political stability may be of very little relevance. A host of writers on the New Left, have testified to the fact that this process has occurred in impelling many 95 offspring of the middle and upper-middle classes towards radical ideologies. And in his excellent article, "The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies", Ron Inglehart points out that this process seems to be affecting younger members of the working-class as well.^ ^ Thus the offspring of our working-class conservatives might well be even more inclined to defect to left-parties, or more importantly, create sources of cleavage for a more radical politics. The discussion of political cleavages, both present and future, brings us to a point that has so far been left 1sui generis', namely the implications of working-class conservatism, and societal cleavages generally,on party politics. It would be false to assume that ideological cleavages (of both the moderate and radical kind) at the party level have their cause directly in society at large; or alter-natively that there exists among electorates deeply rooted ideological 102 cleavages which are transmitted directly into the party system. If there has been any one unstated assumption underlying this paper it is this: the various abstract ideological cleavages on which parties are based, often as they are outlined by their leaders, do not necessarily have their counterparts in society at large or are of relevance to members of the working-class. As Richard Hoggart has written in his Uses of Literacy, a study of British working-class culture: 96 ...(A) great many ...(workers), though they may possess a considerable amount of disconnected information, have little idea of an historical or ideological pattern or process. Their minds rarely go back beyond the times of their own grand parents; before that is a darkness out of which one or two items emerge, not usually in a proper order or with a supporting background - Guy Fawkes and the Gunpower plot, the French Revolution....King Alfred and the cakes. With little intellectual or cultural furniture, with little training in the testing of opposing views... judgements are usually made according to the promptings of those group apopthegms which first come to mind.103 On a more empirical level Philip Converse has pointed out that the belief systems of mass publics are usually crude and unsophisticated, 104 'lacking in constraint'. Butler and Stokes found that less than ten percent of respondents in Great Britain were able to use the left-right dimension in proper fashion and few of these were in the working-class.'^ '' Similar findings for Italy have been reported by Samuel Barnes. This is not to say that the opinions and behaviour of working-class conservatives are irrelevant or offer no constraints on the behaviour of party leaders. However unless the linkage between parties and their electoral support is clearly specified a rather misleading image of a given cleavage structure may result. For example in Italy the behaviour and attitudes of DC leaders, their lack of commitment to democratic norms and the party system may well be a reflection of the attitudes held by their electorate.^ '' But this need not necessarily be so. As Di Palma, Lijphart and others have 97 pointed out, apathy, disaffection or radicalism at the mass level may have a variety of consequences depending upon the use party 108 leaders wish to make of these potential sources of cleavage. Unfortunately the direction in which the effects of working-class conservatism may be channelled by the parties concerned is a question, albeit an interesting one, which is beyond the scope of this thesis. It will have to await study elsewhere. 98 FOOTNOTES 1. Quoted in McKenzie and Silver, "The Delicate Experiment: Industrialism, Conservatism and Working-Class Tories In England", in S. Lipset and S. Rokkan (eds.) Party Systems and Voter Alignments (New York: The Free Press, 1967), p. 118. 2. Ibid., p. 118 3. J. Blondel, Voters, Parties and Leaders (Harmondsworth: Penguin, 1966), p. 58 4. Ibid., p. 58 5. J. Linz, "Cleavage and Consensus in West German Politics: The Early Fifties", p. 200; M. Dogan, "Political Cleavage and Social Stratification in France and Italy", p. 135-136; and E. Allardt and P. Pesonen, "Cleavages in Finnish Politics", p. 346; all in Lipset & Rokkan, Party Systems and Voter Alignments 6. J. Chariot, The Gaullist Phenomenon (London: Allen and Unwin), p. 71; R. Dahrendorf, "Recent Changes in Class Structure", in S. Graubard (ed.) A New Europe (Boston: Beacon, 1967), p. 323; J. Goldthorpe et al. The Affluent Worker in the Class Structure (Cambridge: Cambridge University, 1969), p. 7 7. D. Bell, The End of Ideology (New York: Free Press, 1962) 8. K. Mayer, "Recent Changes in the Class Structure of the United States", in Transactions of the Third World Congress of Sociology, vol. 3 (London: International Sociological Association, 1956), p. 75 9. M. Abrams and R. Rose, Must Labour Lose? (Harmondsworth: Penguin, 1960) 10. S. M. Lipset, "The Changing Class Structure and Contemporary European Politics", in Graubard, pp. 337-351. 11. Ibid., p. 361 12. M. Janowitz and D. Segal, "Social Cleavage and Party Affiliation: Great Britain, Germany and the United States", (May, 1967) p. 602 American Journal of Sociology, Vol. 72, No. 6 13. For details of the sampling procedure etc. see the major publication that resulted from this study; G. Almond and S. Verba, The Civic Culture (Princeton: Princeton University Press, 1963) 99 14. For details of this survey see; D. Butler and D. Stokes, Political Change in Britain (New York: St. Martin's, 1969) 15. Goldthorpe et al., The Affluent Worker in the Class Structure, p. 1 16. Ibid., p. 2 17. H. Christman (ed.), Essential Works of Lenin (New York: Bantam, 1966) , pp. 177-264 18. F. Zweig, The Worker is an Affluent Society (London: Allen and Unwin, 1961); H. Wilensky, "Work, Careers and Social Integration", International Social Science Journal, Vol. 12, No. 4 (Dec., 1960) pp. 543-560 19. M. Abrams, "New Roots of Working-Class Conservatism", Encounter, Vol. 14, No. 5 (May, 1960), pp. 57-59. 20. R. Dahrendorf, Class and Class Conflict in Industrial Soceity (Stanford: Stanford University Press, 1959), p. 268 21. Ibid., p. 268 22. Lipset, "The Changing Class Structure and Contemporary European Politics", p. 340 23. The Economist, Sept. 5, 1970, p. 69 24. M. Salvadori, Italy (Englewood Cliffs: Prentice-Hall, 1965) p. 10 25. P. Ammassani, "The Italian Blue-Collar Worker", in M. Dufty (ed.), The Sociology of the Blue-Collar Worker (Leiden: Brill, 1969) p. 5 26. R. Lane, "The Politics of Consensus in an Age of Affluence", American Political Science Review, Vol 59, No. 4 (Dec, 1965) p. 874 27. M. Abrams and R. Rose, Must Labour Lose? p. 89 28. - G. Galli and A. Prandi, Patterns of Political Participation (New Haven: Yale University Press, 1970), p. 36 29. Ibid. , p. 36 30. Lipset, "The Changing Class Structure", p. 346 31. E. Nordlinger, Working-Class Tories (London: MacGibbon and Kee, 1967) , p. 172 100 32. Ibid., p. 175 33. R. Hamilton, "Affluence and the Worker: the West German Case", American Journal of Sociology, Vol. 71, No. 2 (Sept., 1965), p. 146 34. Ibid., p. 148 35. T. Bottomore, Classes in Modern Society (London: Allen and Unwin, 1970), p. 28 36. Ibid., p. 21 37. In C. Kerr Labour and Management in Idustrial Society (Garden City: Anchor, 1964), pp. 105-147 38. Ibid., p. 109 39. F. Parkin, "Working-Class Conservaties; a theory of political deviance", British Journal of Sociology Vol. 18, p. 278, No.3 (June, 1967) 40. Quoted in Parkin, p. 282 41. Ibid., p. 282 42. Ibid., p. 283 43. W. Runciman, "Embourgeoisement, Self-Rated Class and Party Preference", British Journal of Sociology, Vol. 12, No. 2 (July, 1964) pp. 137-154; H. Tingsten, Political Behaviour (Totowa: Bedminster, 1963), pp. 120-181. 44. W. Narr, CDU-SPD: Programm und Praxis Seit 1945 (Stuttgart: Kohlhammev Verlag, 1966), p. 66 45. M. Lyon, "Christian-Democratic Parties and Politics", The  Journal of Contemporary History. Vol. 2, No. 4 (Oct. 1967), p. 73 46. R. McKenzie and A. Silver, Angels in Marble (Chicago: University of Chicago Press, 1968), pp. 18-73. 47. S. M. Lipset and S. Rokkan, "Cleavage Structures, Party Systems and Voter Alignments: An Introduction", in Party Systems, p. 21 48. Glock and Stark, Religion and Society in Tension (Chicago: Rand McNally, 1965), p. 188 101 49. Protestent workers who attended Church regularly were not more inclined to vote CDU (16.7% vs an average of 19.4%). The CDU draws the majority of its support among Protestant workers from those who attend 'once in a while'. 50. M. Fogarty, Christian Democracy in Western Europe (London: Routledge and Kegan Paul, 1957), pp. 211-231. 51. M. Edelman, "Sources of Popular support for the Italian DC in the Post-War Decade", Midwest Journal of Political Science, Vol.2 No. 2 (May, 1958), p. 156 52. Ibid., p. 158 53. N. Kogan, "Italian Communism, The Working Class, and Organized Catholicism", Journal of Politics Vol. 28, No. 3 (Aug., 1966), p. 555 54. T. Parker, "Religion and Politics in Britain", Journal of Con- temporary History, Vol. 2, No. 4 (Oct., 1967), p. 131 55. Ibid., p. 133 56. Butler and Stokes, Political Change in Britain, p. 133 57. J. Bochel and D. Denver, "Religion and Voting: A Critical Review and a New Analysis", Political Studies. Vol. 18, No. 2 (June, 1970), p. 217 58. For a discussion as to how 'peasant' values of the eighteenth and nineteenth centuries have affected the beliefs of present day British workers see R. Hoggart, The Uses of Literacy (London: Chatto and Windus, 1967), pp. 24-25. The data available do not permit the direct examination of the effects of these traditional attitudes on voting behaviour. 59. A. Gerschenkron, Bread and Democracy in Germany (Berkeley: University of California Press, 1943), p. 28 60. R. Hamilton, Affluence and the French Worker in the Fourth Republic (Princeton: Princeton University Press, 1967), p. 258 61. R. Fried, "Urbanization and Italian Politics", Journal of Politics Vol. 29, No. 3 (Aug., 1967), p. 519 62. Cited in R. Fried, p. 521 63. Ibid., p. 521 102 64. Ibid., p. 522. Unfortunately the data available do not allow me to test the theories outlined above. An attempt to probe the behaviour of Northern workers who were born in the South using the Almond and Verba data yielded a total of four such workers. This does not mean that rural migration has not been involved in this sample. One must remember that in the Fascist era rural migration to urban centres was punishable by imprisonment. Since this law was not officially repealed until 1961 many respondents may have been hesitant in revealing their place of birth. 65. Quoted in McKenzie and Silver, Angels in Marble, p. 6; The concept of deference has been redefined a number of times, For a review of the various definitions see; R. Jessop, "Civility and traditionalism in English political culture", British Journal  of Political Science, Vol. I, No. 1 (Jan., 1971) pp. 1-24. 66. E. Nordlinger, Working-Class Tories, pp. 34-43 67. The efficacy scale for all three countries contains the following items: (1) Suppose a regulation were being considered by the local government which you considered harmful, what do you think you could do? (2) Would you have any success in changing this law? (3) If such a case arose how likely is it that you would actually do something about it? (4) If you made an effort to change a bad law at the national level, how likely is it that you would succeed? (5) In such a case, how likely is it you would actually try to do something about it? In addition the efficacy scale for Italy also contains this item; "Suppose a law were being considered by the national government you considered to be harmful, what would you do?" Each item in the scale was scored on a pass-fail basis. A positive response by a subject to all five items would result in a score of five (or six in the case of Italy). The items for each country were selected on the basis of the Guttman scaling technique. Guttman scalogram analysis allows one to test for the reliability and uni-dimensionality of the items in the scale. The Guttman reproducibility coefficients for the efficacy scale in each of the three countries is as follows: Great Britain, .8896; West Germany, .9335; Italy, .9094. A reproducibility coefficient of .9 or higher is usually considered to be adequate. 103 68. As with the efficacy scale, the items in the administrative trust scale are scored on pass-fail basis in a positive direction. The items for all three countries are as follows: (1) Suppose there were some questions that you had to take to a government office, for example a tax question or housing regulation. Do you think you would be given equal treatment? (2) If you explained your point of view to the officials, would they give your point of view serious consideration? (3) If you had some trouble with the police, do you think you would be given equal treatment? (4) If you explained your point of view to the police, what effect do you think it would have. Would they give your point of view serious consideration? Guttman reproducibility coefficients for each country are as follows: Great Britain, .8949; West Germany, .8608; Italy, .9208. 69. Galli and Prandi, Patterns of Political Participation, p. 181 70. Dahrendorf, Society and Democracy in Germany (Garden City: Anchor, 1969) pp. 156-166 71. L. Hazelrigg, "Religious and Class Bases of Political Conflict in Italy", American Journal of Sociology. Vol. 75, No. 4 Part I (Jan., 1970), p. 508 72. It should be noted that the two scales are not of the interval type. Therefore differences in the magnitude of scores are somewhat ambiguous. For example the difference between a high score and a low score does not necessarily indicate that a high score is three or four times as likely to be efficacious. Nevertheless in terms of the face validity of the items a higher score probably does indicate a perceptible increase in that persons's level of efficacy or administrative trust. 73. This finding is similar to that of other researchers, Hennessey, for instance, found using a sample of Italian youth, that persons identifying with the Communist party were more efficacious, more involved and more supportive of the party system compared to those who identified with the DC. T. Hennessey, "Democratic Attitudinal Configuration among Italian Youth", Midwest Journal of Political Science Vol. 13, No. 2 (May, 1969), p. 189. 104 74. S. M. Lipset, M. Trow and J. Coleman, Union Democracy (Glencoe, Free Press, 1958), p. 172 75. J. Linz, The Social Bases of West German Politics (Ph.D. diss., Columbia University, 1959), p. 373. 76. Hamilton, Affluence and the French Worker, pp. 206-208 77. G. Ingham, "Plant Size and Political Attitudes and Behaviour", The Sociological Review, Vol. 17, No. 2 (July, 1969), p. 238 78. Butler and Stokes, Political Change in Britain, p. 158 79. Linz, The Social Bases of West German Politics, p. 300 80. W. Galenson, Trade Union Democracy in Western Europe (Berkeley: University of California Press, 1957), p. 2 81. Ibid., pp. 3-4 82. A. Ross, "Prosperity and Labour Relations in Western Europe: Italy and France", Industrial and Labour Relations Review, Vol. 16, No. 1 (Oct., 1962), p. 66 83. Galenson, Trade Union Democracy in Western Europe, p. 5 84. Edelman, "Sources of Support for the Italian DC", p. 157 85. H. Spiro, The Politics of Co-Determination (Cambridge: Harvard University Press, 1958), p. 56 86. Ibid., p. 58 87. Dahendorf, Society and Democracy in Germany, pp. 156-166 88. Ibid., p. 163 89. D. Chalmers, The Social Democratic Party of Germany (New Haven: Yale University Press, 1964), p. 178 90. Butler and Stokes, Political Change in Britain, p. 153 91. Nordlinger, Working Class Tories, p. 195 92. Butler and Stokes, Political Change in Britain, p. 162 93. Ibid. , p. 158 105 The multiple regression analysis was run with a dummy variable structure. 3ummy variables were used in order to take into account the curvilinear relationship between Conservative voting and income. In the table below the dummy variables represent every value but one for each of the four variables (income, parental voting, skill level and union-membership). The values not shown in the table are the residual categories for each of the four variables. The predictive power of each dummy variable is assessed relative to the residual category. Multiple Regression for Conservative Voting (Multiple R^  = ...2644) Income (Residual = Medium Income) Variable Low Income High Income B. Coefficients 0.10667 0.08619 8.121* 4.874* Skill Level (Residual = Unskilled) Servants Foremen Skilled Semi-Skilled 0.35517 0.20403 0.10926 0.5233 22.613* 10.255* 5.206* 1.024 Parental Voting (Residual = Father, Labour, Mother Don't Know) Both Conservative 0.03357 Both Labour -0.04228 Both Don't Know 0.11611 One Conservative 0.16870 One Labour Father Conservative, 0.32132 Mother Don't Know 43.749* 0.660* 4.698* 4.303* 14.646* Union Membership -0.11305 11.987* * Significant at .01 level One can see that in all cases the coefficients are in the direc-tion one would predict for these variables. For example both 'high' and 'low' income variables have positive coefficients while 'both parents labour' and 'union-membership' have negative coefficients. 106 95. Dahrendorf, "Recent Changes in Class Structure", p. 325 96. P. Abramson and J. Books, "Social Mobility and Political Attitudes", Comparative Politics, Vol. 3, No. 3, (April, 1971), pp. 403-28 97. G. Mackenzie, "The Economic Dimensions of Embourgeoisement", British Journal of Sociology, Vol. 18, No. 1 (Jan., 1967), pp. 29-44. 98. Dahrendorf, Class and Class Conflict, pp. 295-301 99. P. Abramson, "Social Class and Political Change in Western Europe", Comparative Political Studies, Vol. 4, No. 2 (July, 1971), pp. 131-156. 100. Lipset, "The Changing Class Structure", p. 361 101. R. Inglehart, American Political Science Review, Vol. 65, No. 4 (Dec., 1971) , pp. 991-1017 102 For an example of a researcher who makes these types of assumptions see V. McHale, "Religion and Electoral Politics in France", Canadian Journal of Political Science. Vol. 2, No. 3 (Sept., 1969), pp. 292-312. 103. Hoggart, The Uses of Literacy, p. 158 104. P. Converse, "The Nature of Belief Systems in Mass Publics", in D. Apter (ed.) Ideology and Discontent (New York: The Free Press, 1964) 105. Butler and Stokes, Political Change in Britain, p. 208 106. S. Barnes, "Left, Right and Italian Voter", Comparative Political ' Studies, Vol. 4, No. 1 (July, 1971), pp. 157-176. 107. Hennessey, "Democratic Attitudinal Configurations Among Italian Youth", p. 192 108. Di Palma, "Disaffection and Participation in Western Democracies: The Role of Political Oppositions", in A. Lijphart (ed.) Politics in Europe (Englewood Cliffs: Prentice-Hall, 1969); A. Lijphart, The Politics of Accommodation (Berkeley: University of California Press, 1968) 107 BIBLIOGRAPHY M. Abrams, "New Roots of Working-Class Conservatism", Encounter, Vol. 14, No. 5 (May, 1960) pp. 57-59 M. Abrams and R. Rose, Must Labour Lose? Harmondsworth: Penguin, 1960 P. Abramson, "Social Class and Political Change in Western Europe: A Cross-National Longitudinal Analysis". Comparative  Political Studies, Vol. 4, No. 2 (July, 1971), pp. 131-156 P. Abramson and J. Books, "Social Mobility and Political Attitudes", Comparative Politics, Vol. 3, No. 3 (April, 1971), pp. 403-428 G. Almond and S. Verba, The Civic Culture .'Political Attutudes and Democracy in Five Nations. Princeton University Press, 1963. D. Apter (ed.), Ideology and Discontent. New York: The Free Press, 1964. S. Barnes, "Ideology and the Organization of Conflict: On the Relationship Between Political Thought and Behaviour". Journal of Politics, Vol. 28, No. 31 S. Barnes, "Left, Right and the Italian Voter", Comparative Political  Studies, Vol. 4, No. 2 (July, 1971), pp. 157-176 S. Barnes, "The Legacy of Fascism: Generational Differences in Italian Political Attitudes and Behaviour" Comparative  Political Studies, Vol. 5, No. 1 (April, 1972), pp. 41-58 D. Bell, The End of Ideology. New York- Free Press, 1962. R. Bendix, Work and Authority in Industry. New York: John Wiley and Sons, 1956. J. Blondel, Voters, Parties and Leaders. Harmondsworth:. Penguin, 1966 J. Bochel and D. Denver, "Religion and Voting: A Critical Review and a New Analysis", Political Studies, Vol. 18, No. 2 (June 1970), pp. 205-219. 108 T. Bottomore, Classes in Modern Society. London: Allen and Unwin, 1970 D. Butler and D. Stokes, Political Change inBritain: Forces Shaping . Electoral Choice. New York: St. Martin;s Press, 1969 F. Cancian, "The Southern Italian Peasant: World View and Political Behaviour", Anthropological Quarterly, Vol. 34, No. 1 (Jan., 1961) pp. 1-18 D. Chalmers, The Social Democratic Party of Germany. New Haven: Yale University Press, 1964. J. Chariot, The Gaullist Phenomenon. London, George Allen & Unwin, 1971. H. Christman (ed.) Essential Works of Lenin. New York: Bantam, 1966 R. Dahrendorf, Class and Class Conflict in Industrial Society. Stanford: Stanford University Press, 1959 R. Dahrendorf, Society & Democracy in Germany. Garden City: Anchor, 1969 G. Di Palma, Apathy and Participation. New York: The Free Press, 1970 N. Dufty (ed.) The Sociology of the Blue-Collar Worker. Leiden: Brill, 1969. H. Eckstein, Division and Cohesion in Democracy: A Study of Norway. Princeton: Princeton University Press, 1966 M. Edelman, "Sources of Popular Support for the Italian Christian Democratic Party in the Post War Decade", Midwest Journal of Political Science, Vol. 2, No. 2 (May, 1958) pp. 151-163. M. Einaudi and F. Goguel, Christian Democracy in France and Italy. Notre Dame: University of Notre Dame Press, 1952 M. Fogarty, Christian Democracy in Western Europe 1820-1953. London, Routledge and Kegan Paul, 1957 R. Fraser (ed.) Work. Harmondsworth: Penguin, 1968 109 R. Fried, "Urbanization and Italian Politics, Journal of Politics Vol. 29, No. 3 CAug., 1967) pp. 505-534 W. Galenson, Trade Union Democracy in Western Europe. Berkeley: University of California Press, 1957 G. Galli, and A. Prandi, Patterns of Political Participation in Italy. New Haven: Yale University Press, 1970 A. Gerschenkron, Bread & Democracy in Germany. Berkeley: University California Press, 1943 J. Goldthorpe, D. Lockwood, F. Bechhofer, J. Piatt, The Affluent Worker in the Class Structure. Cambridge Cambridge University Press, 1969 The Affluent Worker: Political Attitudes and Behaviour. Cambridge: Cambridge University Press, 1968 S. Graubard (ed.), A New Europe? Boston: Beacon, 1967 F. Greenstein and S. Tarrow, "The Study of French Political Socialization: Toward the Revocat ion of Paradox", World Politics, Vol. 22, No. I (Oct., 1969) pp. 95-138 C. Glock and R. Stark, Religion & Society in Tension. Chicago: Rand McNally, 1965. R. Hamilton, Affluence and the French Worker in the Fourth Republic. Princeton: Princeton University Press, 1967 R. Hamilton, "Affluence and the Worker: The West German Case", American Journal of Soeiology, Vol. 71, No. 2 (Sept. 1965) pp. 144-152 L. Hazelrigg, "Religious and Class Bases of Political Conflict in Italy", American Journal of Sociology, Vol. 75, No. 4, Part I (Jan., 1970) pp. 498-512 S. Henig & J. Pinder (eds.), European Political Parties. London: Allen and Unwin, 1969 T. Hennessey, "Democratic Attitudinal Configurations Among Italian Youth", Midwest Journal of Political Science, Vol. 13, No. 2 (May, 1969) pp. 167-183 110 R. Hoggart, The Uses of Literacy. London: Chatto and Windus, 1967 D. Horowitz, The Italian Labour Movement. Cambridge: Harvard University Press, 1963 G. Ingham, "Plant Size and Political Attitudes and Behaviour", The Sociological Review, Vol. 17, No. 2 (July, 1969) pp. 235-249 R. Inglehart, "The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies", American  Political Science Review, Vol. 65, No. 4 (Dec, 1971) pp. 991-1017 M. Janowitz & D. Segal, "Social Cleavage and Party Affiliation, Germany, Great Britain and the U.S.", American  Journal of Sociology, Vol. 72, No. 6 (May, 1967) pp. 601-618. M. Janowitz, "Social Stratification and Mobility in West Germany", American Journal of Sociology, Vol. 64, No. 1 (July, 1958) pp. 6-24 R. Jessop, "Civility and Traditionalism in English Political Culture", British Journal of Political Science, Vol. I, No. 1 (Jan., 1971) pp. 1-24. C. Kerr, Labour and Management in Industrial Society. Garden City: Anchor, 1964 N. Kogan, "Italian Communism, The Working Class, and Organized Catholicism", Journal of Politics, Vol. 28, No. 3 (Aug. 1966) pp. 531-555 J. La Palombara, Interest Groups in Italian Politics. Princeton: Princeton University Press, 1964 J. La Palombara, The Italian Labour Movement: Problems and Prospects. Ithaca N.Y.: Cornell University Press, 1957. R. Lane, "The Politics of Consensus in an Age of Affluence", Vol. 59, No. 4 (Dec, 1965) pp. 874-895 R. Leonhardt, This Germany: The Story Since the Third Reich. Harmondsworth: Penguin, 1961. Ill A. Lijphart (ed.) Politics in Europe: Comparisons and Interpretations. Englewood Cliffs; Prentice-Hall, 1969 , The Politics of Accommo dation. Berkeley: University of California Press, 1968 J. Linz, The Social Bases of West German Politics, Ph.D. diss., Columbia University, 1959 S. Lipset and S. Rokkan, Party Systems and Voter Alignments: Cross- National Perspectives. New York: The Free Press, 1967 S. Lipset, Political Man: The Social Bases of Politics. Garden City: Anchor, 1963 S. Lipset and R. Bendix, Social Mobility in Industrial Society. Berkeley: University of California Press, 1959 M. Lyon, "Christian Democratic Parties and Politics", The Journal of Contemporary History, Vol. 2, No. 4 (Oct., 1967) pp. 69-88 V. McHale, "Religion and Electoral Politics in France", Canadian Journal of Political Science, Vol. 2, No. 3 (Sept., 1969) pp. 292-312. H. McClosky and H. Dahlgren, "Primary Group Influence on Party Loyalty", American Political Science Review, (Sept.1959) pp. 757-776. G. Mackenzie, "The Economic Dimensions of Embourgeoisement", The British Journal of Sociology, Vol. 18 No. 1 (Jan., 1967) pp. 29-44. R. MacKenzie and A. Silver, Angels in Marble: Working Class Conservatives in Urban England. Chicago: University of Chicago Press, 1968 K. Mayer, "Recent Changes in the Class Structure of the United States", Transactions of the Third World Congress of Sociology, Vol. 3, London: International Sociological Association, 1956, pp. 68-80. K. Mayer and W. Buckley, Class and Society. New York: Random, 1955 R. Milne and H. Mackenzie, Marginal Seat, 1955. London: Hansard Society, 1958. F. Myers, "Social Class and Political Change in Western Industrial Systems", Comparative Politics, Vol. 2, No. 3 (April, 197U) pp. 389-412. 112 W. Narr, CDU-SPD Programm und Praxis Seit 1945. Stuttgart Kohlhammer Verlag, 1966 M. Neufeld, Italy: School for Awakening Countries. Ithaca: Cornell University, 1961 N. Nie, D. Bent, C. Hull, Statistical Package for the Social Sciences. New York: McGraw-Hill, 1970 E. Nordlinger, Working Class Tories: Authority, Deference and Stable Democracy. London: MacGibbon and Kee, 1967 J. O'Kane, "Economic and Non-Economic Liberalism, Upward Mobility Potential of Catholic Working Class Youth". Social Forces, Vol. 48 No. 4 (June, 1970) pp. 499-506 T. Parker, "Religion and Politics in Britain", Journal of Contemporary History, Vol. -2, No. 4 (Oct., 1967) pp. 123-136 F. Parkin, Class Inequality and Political Order. London: MacGibbon & Kee, 1971 F. Parkin, "Working Class Conservatives: A Theory of Political Deviance", British Journal of Sociology, Vol. 18 No. 3 (June, 1967) pp. 278-290 H. Popitz, H. Bahrdt, E. Jures, H. Resting, Das Gesellschaftsbild des Arbeiters, Tubingen, Mohr, 1957 J. Raffaele, Labour Leadership in Italy and Denmark. Madison: University of Wisconsin Press, 1962. R. Rose (ed.) Studies in British Politics. London: Macmillan, 1967 A. Ross, "Prosperity and Industrial Relations in Western Europe: The Case of West Germany", Quarterly Journal of Economics, Vol. 76, No. 3 (Aug., 1962) pp. 331-359 A. Ross "Prosperity and Labour Relations in Western Europe: Italy and France", Industrial and Labour Relations Review, Vol. 16, No. I (Oct, 1962) pp. 63-85 W. Runciman, "Embourgeoisement, Self-Rated Class and Party Preference", Sociological Review, Vol. 12, No. 2 (July, 1964) pp. 137-154 M. Salvadori, Italy. Englewood Cliffs: Prentice-Hall, 1965. 113 D. Segal, "Classes and Parties in West Germany and the U.S.", Comparative Studies in Society and History, Vol. 10 No. 1 (Oct., 1967). pp. 66-87 K. Shell (ed.) The Democratic Political Process. London: Blaisdell, 1969 N. Smelser and S. Lipset (eds.) Social Structure and Mobility in Economic Development. Chicago: Aldine 1966 H. Spiro, The Politics of German Co-Determination. Cambridge: Harvard University Press, 1958. R. Stark, "Class, Radicalism & Religious Involvement in Great Britain", American Sociological Review, Vol. 29, No. 5 (Oct., 1964) pp. 603-619 S. Surace, Ideology, Economic Change and the Working Classes: The Case of Italy. Berkeley: University of California Press, 1966. H. Tingsten, Political Behaviour: Studies in Election Statistics. Totowa: Bedminster, 1963. S. Verba, N. Nie, J. Kim, The Modes of Democratic Participation: A Cross-National Comparison, Beverley Hills: Sage, 1971 R. Webster, Christian Democracy in Italy 1860-1960, London: Hollis and Carter, 1961. H. Wilensky, "Work, Careers and Social Integration", International Social Science Journal, Vol. 12, No. 4 (Dec. 1960) pp. 543-560 F. Zweig, The Worker in an Affluent Society. London: Allen and Unwin, 1961 


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