INSTITUTE F*Q8 SOCIAL

Transcription

1 8341 INSTITUTE F*Q8 SOCIAL US3ARY RELIGION AND CLASS IN ITALIAN ELECTORAL BEHAVIOR Samuel H. Barnes Professor of Political Science and Program Director, Center for Political Studies, Institute for Social Research The University of Michigan FINAL DRAFT chapter for Richard Rose (ed.), Comparative Electoral Behavior: An Introduction New York: Free Press. DRAFT: Please do not quote without explicit permission from the author. December 1970

2 Religion and Class in Italian Electoral Behavior I. Historical Background A. Social Structure since 1861 B. The Electoral System in the Twentieth Century C. The Party System in the Twentieth Century D. Competition for Seats E. Continuities and Discontinuities I I. The Act of Voting Today A. Elections and Ballots B. The Legal Facilitation of Voting C. Elections Since the Fall of Fascism I I I. Data for Analysis A. Sample Surveys B. Secondary Analyses C. Ecological Analyses D. Non-Voting and Its Measurement E. Voting and its Measurement IV. Social Structure and Voting A. Sex B. Age C. Educa tion D. Religion E. Ethno-Linguistic Differences F. Region G. Urban-Rural H. Residential Mobility

3 i i I. Mass Media J. Organizational Memberships K. Occupational Class L. Income M. Consumption Patterns arid Life Style N. Conditions of Employment 0. Subjective Class Assessment V. Social Psychological Influences A. Political Efficacy B. Interest in Politics C. Religious Influences VI. A Tree Analysis of Determinants of Italian Partisan Choice A. Some Contradictions in Italian Political Behavior B. Tree Analysis and the Variables Employed C. The Tree Analysis D. An Interpretation E. Behavioral Group Partisan Preferences VII. Conclusions A. Variables of Comparative Significance B. Countries Approximately Compared Tables Bibliography

4 RELIGION AND CLASS IN ITALIAN ELECTORAL BEHAVIOR The fundamental political divisions of Italy have existed in their present mold for at least half a century. Their origins are even more remote, for the social cleavages of an industrial society are superimposed on previous lines of stratification and religion blurs further the political impact of class. Only recently has the impact of industrialization begun to erode premodern fissures, hence their traces are likely to linger into the future. The result is a series of cleavages that originate in different ages and lead to contemporary structures of political conflict that sometimes seem to contradict and sometimes to confirm the conventional wisdom about political behavior. Thus according to the conventional wisdom Italy is badly fragmented politically. Cited as evidence is the existence of the largest nonruling Communist party in the world, a small neofascist movement, and a ruling Christian Democratic party that is based on religious values in an increasingly secular age and society. Italy also has Monarchist, Republican, Liberal and, at the time of writing, three Socialist parties represented in parliament. Political fragmentation and polarization coexist with low levels of mass education, slight interest in politics and one of the highest rates of electoral turnout in the world. Italy is also considered to be poorly integrated socially. Class divisions are sharp, differences between North and South remain highly salient, and the religious question divides otherwise similar groups. The Italian citizen is viewed as alienated and despairing, the government

5 perpetually on the verge of collapse, and the administration chaotic. But Italy is also a European, industrial, parliamentary democracy. Compared with the universe of polities existing today i t ranks very high on indices of political and economic development. I t is only when compared with the countries of Northern Europe and North America that i t appears behind in industrialization and tardy in effectuating a secularization of politics, national integration, and a homogeneous national political culture. Contemporary politics s t i l l reflect regional, religious, social and cultural differences of the past to a remarkable degree. This is of course true of a l l polities, but they are probably quantitatively somewhat more significant in Italy than in the other countries considered in this volume. 1 Certainly the conventional wisdom suggests this. Our goal in this chapter is to subject the conventional wisdom concerning Italian politics to a c r i t i c a l analysis in the light of empirical research. We w i l l be concerned in particular with assessing the impact of religion and class on contemporary political attitudes and behavior. I. HISTORICAL BACKGROUND In determining what is the proper historical perspective needed for an analysis of contemporary Italian politics i t is necessary to keep a number of dates in mind, for they serve as alternative cutoff points for different kinds of analyses. The first important date is when modern Italy begins (or, for some purposes, 1870, when Rome was incorporated). The next date of great significance for political behavior is the achievement of universal manhood suffrage in The third is the March on Rome on October 28, The final date is 1943, when military defeat and partisan

6 -3- activity led to the overthrow of Fascism. Each of these dates is useful for some analyses and not for others. For most purposes, social scientists concerned with the political and especially electoral behavior of modern Italy w i l l begin with 1861 or later. While statistical and other records exist for most of the states that were joined to form Italy, problems of access and comparability render them d i f f i c u l t to use; in general, since unification records have been excellent. In 1870 Rome was incorporated into the new Kingdom and made its capital. Territorial changes since that date have been marginal; the principal administrative subdivisions of the country have also not changed greatly. There has been considerable political continuity since unification, despite the change in regime and the experience of Fascism. However, the strength of different political forces in the country has altered greatly. During the nineteenth century the polity was dominated by the politics of trasformismo. This was the practice of putting together a government by dealing on an individual basis with members of parliament, securing their support by favors provided them and their constituents. Parties in the modern sense did not exist; factions developed around personalities and, to a lesser extent, vague ideological tendencies. In present-day political terms these elites were on the whole liberals in the European sense. Catholics were forbidden by the papal Non expedit (1874) to participate in politics because the Papal States were incorporated over the objections of the Pope and he in turn refused to cooperate with the new regime. In this heavily Catholic country the Pope's actions deprived the system of

7 legitimacy in the eyes of many of its citizens, including groups that could h we been very important for its stability. Socialism and its rival, anarchism, became important in the later years of the century. The development of unity on the left then as now was d i f f i c u l t to achieve, with the socialist tendency finally dominating over anarchism. The Italian Socialist Party was bounded in 1892, and even with the electorate restricted by property and educational qualifications socialists began to be elected to the lower house. As the electorate gradually expanded due to relaxation of the requirements as well as the growing ability of the electorate to meet them, the prospects of socialist electoral success led f i r s t to a relaxation of the prohibition of Catholic political activity and then to its repeal by the time universal manhood suffrage was achieved in The 1919 elections were the f i r s t in which a l l adu.it males voted without property, educational or age restrictions. I t was alco the f i r s t in which the clear outlines of the present party system are visible. Italian socialism was badly divided by the Russian Revolution. A general strike in Italy failed to bring down the government and the party was gravely weakened. I t was further damaged in 1921 when its revolutionary wing broke away to form the Italian Communist Party. The Popular party, the forerunner of the present Christian Democratic Party, was a potential source of support for stability and democratic reforms; but i t was ambivalent about cooperating with the Socialists, and governments remained weak and unstable. Although the Fascists had had only modest electoral success, Mussolini was invited to form a government following the "March on Rome" on October 28, (Mussolini arrived by trainj)

8 -5- The f i r s t Fascist government was a coalition. The members of other parties were soon weeded out and opposition parties had largely been destroyed and the dictatorship established by Whatever measure of broad support that Fascism received in its early years seems to have disappeared in the 1930's; though internal opposition was never very effective, Fascism could not survive the war. Mussolini had been appointed head of the government by the King and he was dismissed by the King in 1943, after the Fascist Grand Council had voted to oust him. Arrested, he was freed by the Germans and set up a Fascist Republic under their close control. When they retreated from Italy he was caught and executed by a partisan band. The Italians rejected the monarchy in a referendum in 1946 and at the same time elected a constituent assembly that drew up a new constitution. I t is the basic document that governs Italy today. Several aspects of Italian history since unification require detailed attention, as they are necessary to an understanding of contemporary political behavior. We begin with a review of changes in the Italian social structure since A. Social Structure Since 1861 The four gross indicators of the social environment of the political system used in this analysis are the size and distribution of the population, employment in industry, urbanization and literacy. The population of Italy (using i t s present borders) almost doubled between 1861 and 1961, from 25,756,000 to 49,877,000. In addition, between 1869 and 1962 more than 24,000,000 people emigrated to the United

9 -6- States, Canada, Australia, Argentina, and other countries, and at least 8,000,000 of these never returned. (Ciaurro, 297) Moreover, internal migration has been very great as well, and the gross changes in population of various parts of the country often mask even larger shifts in population. The mortality and birthrates today vary greatly from one region to another. The national average birthrate, for example, was 37 per thousand in 1871 and 18.4 in 1961; Piedmont and Calabria had rates of 36.5 and 36.7 per 1000 in 1871 and 13.4 and 24.1 in (Annuario Statistico, 1878, 45; 1962, 28) Nevertheless, the overall portion of the population living in the South has declined slightly, from 39 percent in 1860 to 36.1 percent in (Tarrow, 41; Annuario dell Economia, della Politica, della Coltura, 1968, 103) A distinctive feature of Italian history has been its retarded and incomplete industrialization. Modern methods of manufacturing came relatively late to this cradle of the joint stock company, modern banking, and superb craftsmanship. Although some pockets of industry are longestablished, at the time of unification the population was 68.6 percent agricultural; in 1950 the agricultural sector comprised 42.4 percent and in percent of the population. But even within the modern sector much of the activity was in fact artisanal, and in manufacturing small family-oriented firms were s t i l l very important in the postwar period despite the existence of some very large companies. Italy has always been an urban civilization; its Renaissance cities elevated local particularism to a very high level in art and politics. Relatively few people live in the countryside, though villages are often without many of the amenities of urban l i f e. According to the census of

10 only 174,996 Italians lived in communes with less than 500 persons, 765,850 lived in the 501 to 1,000 category, and 2,717,080 in the category. The total living in communes with fewer than 5,000 people (12,292,371) was similar to the number in communes above 100,000 (12,530,221). (Annuario Statjstico Italiano, 1962, 26-27) Illiteracy has been extensive in Italy relative to northern European countries. In 1871, 72.9 percent of the population was I l l i t e r a t e. In the postwar period the percentage has declined from 14.1 percent in 1951 to 8.4 percent in 1961, and is today disappearing. Functional illiteracy, however, is more serious than these figures indicate. B. The Electoral System in the Twentieth Century From unification until 1882 the Italian electorate was extremely restricted. The average number of votes needed to elect a deputy was 500, and sometimes slightly more than 100 sufficed. (Capecchi, et al., 23) From 1882 until 1919 the electorate was gradually expanded to include the entire adult male population. Literacy and property requirements for men thirty and over were dropped in 1913; universal manhood suffrage for men over twenty-one was adopted for the 1919 elections, and women f i r s t voted in national elections in Until 1882 the single member district system was used; from 1882 until 1892 a multimember system was in effect; fcr the period the single member district system was again utilized. Proportional representation was adopted in 1919 and today seems to be firmly installed as the system that divides Italians least. Unlike in France and Germany, there is no weighty section of opinion that advocates switching to a single

11 -8- member system. Attempts to tamper with the proportional features of the law in order to insure a working majority in parliament, as with the so-called "swindle law" of 1953, offer excellent ammunition to the opposition. Adopted for the elections of that year, this law would have guaranteed a comfortable absolute majority to the party or parties that secured 50 percent plus one of the votes. Memories of a somewhat comparable Fascist electoral law were called up in opposition to the law; when the DC coalition failed by a fraction to achieve 50 percent, the law was quietly repealed. Against this background, democracy and the present form of proportional representation tend to be equated in Italy, at least at the national level and for the Chamber of Deputies. In the half century following unification, elections were won by careful tending of the small electorate, intimidation, governmental interference when necessary, and other practices so corrupt that the Minister responsible was labeled by Gaetano Salvemini the "Minister of Evil." These practices declined with the rise of mass parties, except for their extensive use by the Fascists in the last elections under the Monarchy. Postwar elections have been free of irregularities. The secrecy of the ballot is assured by the constitution (Article 43) and by practice. Since women acquired the vote fox the 1946 elections a l l adults have been able to vote except for the usual restrictions relating to felons, and so on. Voting age differs, however, for the two houses; it is twentyone for the Chamber and twenty-five for the Senate. Voting is defined as a "civic duty" and nonvoters with inadequate explanations have the fact stamped on their identity papers. Although civic duty is seldom a principal determinant of political behavior in Italy, turnout is high.

12 Voting has become an established norm of the culture and great efforts are made to comply with the norm. The competitiveness of the party system, the effortless registration, the financial inducements in the form of reduced fares on trains, and the holiday atmosphere combine to insure a high turnout. Registration is easy. The authorities keep records of inhabitants; when one reaches voting age he is added to the rolls automatically and before an election is sent a certificate. He remains on the rolls until he is removed by death or by acquiring residence elsewhere. Fascist laws that long remained in force made i t d i f f i c u l t to change one's legal residence, so that voters returned to their "home" commune to vote. The law has been declared unconstitutional and the practice is now breaking down, but the large number of people who live and work in one commune while maintaining their legal and hence voting residence in another remains large, though declining. This was a serious handicap for survey research in the past, as electoral rolls are used for sampling; and i t is s t i l l an inconvenience, since such voters are d i f f i c u l t to trace. The subjects lost in this manner are likely to be emigrants of possible unusual electoral v o l a t i l i t y. Today, however, the numbers involved are small. Election to the Chamber is by party l i s t in constituencies ranging from four to forty-seven seats. These constituencies generally group together the provinces of a small region or a part of a large region. They follow closely the administrative divisions of the country. However, the constituencies themselves seldom reflect fully the structure of normal party organization, which is largely by province, so that drawing up the l i s t and managing the campaign in the constituency often present problems

13 -10- of coordination between coequal party units. In practice there is often a pragmatic partitioning of the constituency; candidates may campaign in only part of the constituency and in fact may feel that their "real" constituency is a limited part of the whole. The election of senators is also in fact largely by proportional representation, in the following manner: each region of Italy is alloted a number of senators proportional to its population with a minimum of seven except for the sparsely populated regions of Molise and Valle d'aosta, which are guaranteed minima of two and one senators. Each region is then divided into a number of districts of roughly similar size equal to the number of senators to be elected. Each party runs one candidate in each district. Anyone with sixty-five percent of the vote in that district is declared elected. Otherwise, the votes for a l l candidates of a party or group of parties in the region are pooled and seats are assigned by proportional representation to the parties and then to the candidates who received the highest percentage of votes within their individual districts. This system permits parties to draw up joint regional l i s t s. Thus the Social Proletarians and the Communists might agree that in district X the latter would put up no candidate while the former would run none in other districts. In this manner the Social Proletarians could obtain representation in the Senate and the Communists would be assured of additional votes in the region. This practice is of marginal importance. Mos t senators are elected by proportional representation with the region as the constituency; for example, in 1968 only two were elected by receiving sixty-five percent of the vote--one from the Christian Democratic party and one from the Sudtiroler Volkspartei.

14 -11- The Chamber and the Senate have similar powers and are elected at the same time, so that not surprisingly they virtually duplicate one another. As one is elected by party lists in large constituencies and the other in single member districts-^at least ostensibly i t may be tempting to use the Senate returns so as to have units comparable to British or American constituencies. For most purposes this would be unwise. The voting age differs for the two. And despite their constitutional equality, the Chamber is of greater political importance than the Senate. Most party leaders s i t there and most ministers are chosen from the Chamber. The Chamber also receives more attention in elections. And most Senators are in fact elected by proportional representation. The 1948 constitution provided for one deputy for every 80,000 electors. The number of seats in the Chamber consequently fluctuated upward, from 574 in 1948, 590 in 1953, to 596 in The number was permanently fixed at 630 for the Chamber and 315 for the Senate by constitutional amendment in The number to be elected in each constituency is adjusted for every election according to the latest population figures and i t is thus not necessary to redraw the constituency boundaries. The major parties present lists throughout the country. This is not the same thing as mounting effective campaigns, and the smaller parties, especially the Monarchists and Republicans, often are barely visible outside of their areas of concentrated strength. The nomination of candidates is a party matter, unregulated by law and unstudied by scholars. Technical problems of securing signatures make i t d i f f i c u l t for casual lists to get on the ballot but the sixteen on the Roman ballot in 1968 (including the Pensioners' party, dissident

15 l e f t i s t s, local l i s t s, etc.) suggest that the difficulties are not overwhelming. The counting of votes involves a discussion of the system of preference votes (D'Amato, 1964). The elector votes for the party and may, i f he wishes, l i s t particular preferences. In constituencies electing fewer than sixteen he may mark up to three; in others, four. The number of candidates elected on a l i s t depends on the total vote for that l i s t, while the particular candidates elected are determined by the number of preference votes received. Thus position on the l i s t means l i t t l e. There are great differences in the extent to which the right to express a preference is exercised. Overall, for example, the rate varied in 1958 from percent in the constituency of Milan to percent in that of Palermo (Schepis, Table 26). But this masks considerable differences among the parties and i t also means different things in different areas and parties. In much of the South, for example, the local organizations of the Right and Center parties, including the Christian Democratic, make arrangements with candidates to deliver them a certain number of preference votes. In this case the large number of preference votes cast reflects not only the greater personalization of politics but also the clientelistic nature of party organization. Certainly the politics of preference votes is an inadequately studied subject in Italy. In distributing votes among the party l i s t s, a quotient is established for the constituency by dividing total vote by seats to be f i l l e d plus two, and seats are assigned to parties that meet multiples of the quotient. I f this process would result in more deputies being seated than have been allotted the constituency, the quotient is computed again on the number of

16 -13- seats plus one, then plus none, until seats are assigned. Votes that do not contribute to the election of a candidate are transferred to the national electoral office, i f the party received 300,000 votes nationally and elected at least one candidate in a constituency. Likewise, seats not distributed even with the quotient of the number of seats plus two are transferred to the national electoral office. A national quotient is then established and parties are assigned the remaining seats. However, there is no national l i s t. Instead, parties rank unsuccessful candidates according to their number of preference votes. Thus a candidate who receives a large number of preference votes but is not elected on the constituency l i s t s t i l l has a chance of being elected on the national college. Moreover, candidates who were elected in two constituencies, or to both the Senate and the Chamber, must opt for one or the other (they can run in up to three Chamber constituencies); the candidate with the next highest number of preference votes on the vacated l i s t is then declared elected. Subsequent vacancies are f i l l e d in a similar fashion, avoiding byelections. In the Valle d'aosta, i f no one receives a majority in the competition for the single Chamber seat, a runoff election is held two weeks later between the two leading candidates. For the Valle d'aosta Senate seat a plurality suffices. Participation as a percentage of those eligible to vote varied between 45.5 percent in 1870 and 65 percent in 1909 in the period of restricted suffrage.(capecchi, et al., p. 23). With the suffrage extended, in the hotly contested election of 1919 only 56.6 percent of those eligible to vote did so, and this rose only to 60.9 percent in Under the republic i t has been consistently high:

17 U 92.2% 93.8% 93.8% 92.9% 93% The true rate of participation is actually higher than this, because these figures are based on the total electoral rolls. In order to vote, electors must have certificates that are delivered to them by municipal authorities, and the percentage not delivered because the elector could not be found is revealing concerning the turnout rate (the certificates not delivered are retained by the authorities until the polls close). In 1946 the c e r t i f i cates not delivered were 5.4 percent of the total; in 1948, 3.5 percent; in 1953, 2.7 percent; in percent; and in percent. Regional differences are significant: the rate in Abruzzo in 1963 was 8.3 percent and in Molise in the same year i t reached 14.9 percent. These are regions of heavy emigration, suggesting that many voters simply are no longer resident there; and, further, that the lists are inadequately updated. Thus the rate of participation for Abruzzo goes from 86.5 percent to 94.8 percent and Molise from 79.5 percent to 94.4 percent i f i t is computed on the basis of electoral certificates delivered (Capecchi, et al., 1968, 69-70). Scholars analyzing turnout in Italy must take such factors into account. The phenomenon of the non-resident voter w i l l not be as important in the future but i t w i l l remain, because those who emigrate abroad to work w i l l be carried on the rolls and, indeed, many of them w i l l return to vote. Blank ballots and invalid ballots are of some importance in the final results. Between 1948 and 1963 blank ballots varied between.6 percent and 1.8 percent of the total; during the same period invalid ballots ranged from 1.3 percent to 2.0 percent of the votes cast. For an analysis

18 -15- of the correlates of blank ballots and invalid ballots see Capecchi, et al., 1968, Complete electoral returns are published in several volumes by the Ministry of the Interior. However, they are not commercially available and hence are d i f f i c u l t to obtain. More limited but s t i l l useful sources are the Annuario Parlamentare and, for elections since 1963, the Annuario Politico. Neither of these is truly satisfactory, nor is any other single source available. C. The Party System in the Twentieth Century A simplified overview of the Italian party system in this century would identify at least four principal historical tendencies, with each supported by particular social groups and subcultures. These are not always represented by single parties, however; and the overlapping and exchanges through time between parties appealing to the same groups and subcultures provide much of the dynamics of the Italian party system. These tendencies are the secular bourgeois tradition, the l e f t i s t tradition, the Catholic tradition and the Fascist tradition. The formation of modern Italy was the achievement of secular elites who came to terms in a highly selective fashion with many traditional regional elites. However, they were actively opposed by Catholic elites, who were forbidden by the church to participate in the affairs of the new state that had been formed in part at the expense of the Papacy. The Catholic masses were unmobilized; the emerging urban proletariat, attracted by anarchism and denied the vote, was not an important political force until late in the century. In the absence of Catholic and l e f t i s t participation,

19 -16- the restricted electoral system left politics in the hands of the liberal elite. With the extension of the suffrage, however, the rise in votes for the left led to a strategy of incorporating the Catholic masses into the system as a conservative force. But the Italian party system was denied the stabilizing influence of the Church, and the Liberal politicians were unable to incorporate the new voters into their system of mutual favoritism. The result was a long-term decline in their electoral strength and the rise of two mass movements, one socialist and one Catholic. The Catholics began to organize prior to the First World war, and combined for the 1919 elections to form the Partito Popolare, which received the second highest number of votes and seats. This party, greatly expanded in conception and support, formed the basis for the Christian Democratic party after the Second World War. The socialist movement has chronically suffered from disunity and frequently from disarray as well. I t has always contained a wide spread of opinion ranging from reformist to revolutionary, and this opinion has usually found political expression in at least two parties. Thus prior to the rise of Fascism there were sporadically at least two socialist parties and in the 1920's a communist party as well. However, they a l l shared a common subculture, and the socialist tradition is remarkably durable and of particular strength in some areas, of which more below. The fourth tradition is the Fascist. Fascism is an Italian invention, and its "real" meaning continues to be the subject of analysis and research in Italy and abroad. Within the context of a discussion of the party system i t is sufficient to note that i t was a movement that secured the support of a number of elite groups and at the same time was able to mobilize, or

20 -17- simulate the mobilization of, a substantial mass following. As a result of its years in power there has been a blurring of distinctions between who had and who had not been a Fascist, with only the peasantry, the working class and its leaders, and some Catholic elements largely free of contamination. This fact had considerable ramifications for the party system after There is another well-established political practice in Italy that merits attention, though i t is not an organized point of view as much as i t is a state of mind, or, more properly, a set of behaviors. That is clientelism, in which local notables construct a network of followers through mutual favoritism, pork barrel politics and intimidation. I t is sometimes said to antedate even the classical Greek and Roman political traditions of the peninsula. The nineteenth century party system relied almost exclusively on these practices to the neglect of party organization and ideology. As an organizational slice of the party system i t is no. longer of significance, but as a source of votes and as an explanatory variable i t s t i l l retains considerable importance. The Christian Democratic party, the Liberals, and the Monarchists rely heavily on these methods, especially in the South, and i t is an important element in a l l parties. But i t is today a technique for getting votes, and not a separate tradition. I t is only with the rise of the l e f t i s t parties that parties as organizations acquired meaning in Italy. The previous parties were aggregations of notables, each of whom tended his own fences with the assistance on a personal basis of the prime minister and cabinet. Cabinets were formed on the basis of these personal arrangements (trasformismo) with few considerations of ideology or party organizational needs. I t was a system

21 -18- that worked well only with a restricted electorate, so that the rise of the left and subsequent emergence of the Popular party left the notables without a mass following. This weakness added to the attractions of Fascism for the traditional political elites of the country, and i t also contributed to the discrediting of many of them at the end of the Second World War, These traditions were a l l represented in the last elections before the rise of Fascism and they were also present at its downfall. They thus provide a background for the analysis of the party system since The Italian party system at the time of writing has at least nine parties that can claim realistically to be national parties. In 1970 these were the following, as conventionally listed from left to right: Italian Political Parties in 1970 PCI Partito Comunista Italiano--Italian Communist Party PSIUP partito Socialista Italiano di Unita Proletaria ltalian Socialist party of Proletarian Unity *PSI Partito Socialista Italiano--Italian Socialist Party *PSU Partito Socialista Unitario--Unitary Socialist Party PRI Partito Repubblicano Italiano Italian Republican Party DC Democrazia Cristiana--Christian Democratic Party PLI Partito Liberale Italiano--Italian Liberal Party PDIUM Partito Democratico Italiano di Unita" Monarchica--Monarchist Party MSI Movimento Sociale Italiano--Italian Social Movement (Neofascist Party) These two formed a single party in 1968, the Italian Socialist Party--Italian Social Democratic Party (PSI-PSDI).

22 -19- Each of these parties runs candidates in a l l the thirty-two constituencies. In the Valle d'aosta the single member system leads to two competitive coalitions. In addition to these national parties there are three major regional parties, the Sardinian Action Party, the Union Valdostaine, and the Sudtiroler Volkspartei. Moreover, there is a third category of party that sometimes assumes marginal significance, the true minor parties. These include dissident socialist groups, nonmarxist radical l i s t s, protest groups such as the pensioners, etc. Their importance is limited to certain elections and constituencies in which they have had a marginal impact on seats assigned to particular parties. The l i s t of major parties on the ballot has varied very l i t t l e since 1946, The MSI and others absorbed votes that went to the "Everyman's Party" in 1946, the Monarchists have split and reunited during the interim, the socialists have split, merged (with a faction refusing to merge), and split again, so that in 1969 there were three major socialist parties in addition to the PCI. The actual number of parties on the ballot in a constituency varies from election to election and from constituency to constituency. However, the national nature of the major parties and their near monopoly of votes means that for practical purposes the party system is remarkably stable and national. Electoral returns reinforce the contention that Italy is a system of "Polarized Pluralism." (Sartori, 1966) The two largest parties have received at least 62 percent of the vote between them since In 1946, the Christian Democrats received 35.2 percent, the Socialists 20.7 percent, and the Communists 18.9 percent. In 1948 the Socialists and Communists supported a joint l i s t, which received 31 percent of the vote;

23 -20- however, more Comrnunists than Socialists were elected. Since that time the Communists have been the number two party and the PSI third. The vote of other parties has been fragmented, with the PSDI's 7.1 percent in 1948 and the PLI's 7 percent in 1963 being the highest levels reached by fourth parties during the postwar period. What has shifted, however, is the vote for parties within these categories. The Communists have received an increasing percentage of the l e f t i s t portion, and the extreme right has declined in favor of the moderate right of Christian Democrats and Liberals. Taking changes of name, mergers, and splits into account, there has been l i t t l e change in the party system. The nine parties listed above (along with their predecessors) accounted for 97.7 percent of the votes in 1948, and 98.6 percent in 1968, for example; and the only other parties of importance were those representing ethnic minorities and particular regions. Although each parliament lists a few independents, almost a l l were elected on a party l i s t and subsequently broke with the party. Few of these are reelected, unless they join a party. The fragmentation of the vote gives rise to the perpetual problem of a majority in parliament. The electoral system reproduces quite faithfully in parliament the division of the vote, with only a slight tendency to favor the larger parties through the distribution of remainders. The location of the Christian Democratic party near the center helps i t to dominate a l l governments, as i t has since 1946, but the importance of marginal deputies to a cabinet gives smaller parties such as the Republicans and, formerly, the Social Democrats a crucial role in the parliamentary system despite their meager representation.

24 -21- The historically important tendencies in Italian political parties have been previously described as secular bourgeois, l e f t i s t, Catholic, and Fascist. Richard Rose has suggested that the following groupings seem relevant in a European context: Conservative-Bourgeois, Religious, Liberal-Secular-Bourgeois, Socialist, Communist, Nationalist, Fascist, Agrarian and Protest. A l l of these forms have been of significance in Italy, and most of them are s t i l l represented in the party system. One must, however, point out that parties are usually both more and less than is suggested by the label of the tendency. The PCI, for example, is the largest Communist party in the world after the Chinese and Russian. Yet, i f the Chinese and Russian parties represent the ideal type, then the PCI might better be placed in the Socialist category. I f we accept the a t t i tudes and expectations of most of its leaders, militants, and voters as expressed in interviews, i t has l i t t l e in common with these two ruling communist parties. Italy has a strong socialist tradition: the Socialist tendency is the f i r s t in voting strength and in organizational effectiveness in the country. The strongest organizational base within that tendency is occupied by the PCI, which is also the largest in electoral strength. I t is important in understanding Italian communism that i t be viewed as a part of the Socialist tendency rather than as merely the clever and cynical agent of a foreign power.

25 -22- The Republicans are more d i f f i c u l t to classify. They are a part of the left and many of them are perhaps nonmarxian socialists. At the same time they are also secular and bourgeois. Only in the Romagna do they have a mass following. The PRI f i t s poorly in contemporary schemes of classification. The Christian Democratic party is obviously a religious party, except that i t is so much more than that. As the party of order and continuity in Italy as well as the governing party, i t attracts many followers for nonreligious reasons. In fact, i t attracts different groups for different reasons. However, the Christian Democratic party is f i r s t and above a l l else the party of the Church and thus must be classified as a religious party. The Liberal party was traditionally a Liberal-Secular-Bourgeois party, but in the postwar period i t has become increasingly the party of constitutional conservatism. The PLI could perhaps equally well today be labeled a Conservative-Bourgeois party. The Monarchists are a traditionalistclientelistic party that f i t s none of the categories. Although most of its strength--which is in severe decline--was Southern and rural, i t is not an Agrarian party. I t is the dying representative of a type of party now extinct in most of Europe. I t does, however, have characteristics that are common to other Italian parties of the Right and Center as well as to parties in other countries. The MSI is a Fascist-Nationalist party, with perhaps a greater emphasis on the second than the f i r s t. I t is also in practice a protest party in its appeal to the dissatisfied. Single theme protest parties such as the pensioners exist but without electoral success. The Sudtiroler Volkspartei is the party of German-speaking particularists; in its socioeconomic outlook i t is a Conservative Bourgeois party.

26 -23- An analysis based on tendencies thus must carefully specify the dimensions involved (see Barnes, 1970A). The major historical dimensions are left and right, but this is a complex issue involving socioeconomic, religious and other dimensions. On most of these the Communists, Socialists, and Republicans are on one side and the rest on the other. The major political dimensions relevant to early postwar cabinet formation would be l e f t, right, and center, with the Christian Democrats divided between center and right and the rest in the left or right. In cabinet formation in the late 1960's the relevant divisions would be l e f t, center-left, and right. Thus in making the leap from historical tendency to cabinet formation care must be taken to sketch in the connections. Indeed, conventional notions of left and right are useful for some but not a l l aspects of contemporary political analysis. There is a strong association between party identification and left-right self placement and between ID and an index of programmatic position. The tau beta for the former is.63 and for the latter.40 (Barnes, 197OA). However, much of the Christian Democratic vote is l e f t i s t on socioeconomic issues and that party has seemingly moved to the left in recent years. This ability to cover both its flanks has been a constant source of strength for the DC. But i t also is the cause of much of its internal factionalism and the increasingly d i f f i c u l t task of putting together a government. D. Competition for Seats Italian parties are highly articulated by American or even British standards, and the multi-member constituency, preference votes, and party control over the composition of the lists place the party in virtually

27 -24- complete control over who is and who is not elected. While the role of the party apparatus is crucial in a l l of the parties, i t varies according to the strength of the central bureaucracy. In the PCI, for example, the party organization nominates what is in effect a "balanced ticket" assuring representation to relevant t e r r i t o r i a l and socioeconomic categories, party militants are then instructed as to whom to give their preference votes and voters are encouraged to vote for the party rather than to express preferences. Although selection of the persons actually elected is greatly influenced by the organization, i t cannot, of course, control the total number of votes received. Thus i t is only the marginal candidate, the nth one in the party's scale of importance, who cannot be certain of election. For the rest, election is no surprise once they have been nominated. The national leaders receive many preference votes; others have only a small spread between them. The same system works, though less effectively, within the socialist parties. The better known party officials are usually assured the preference votes of the militants, whereas the candidates on the l i s t who have no close organizational ties have l i t t l e chance of election. The Christian Democratic party works somewhat the same way throughout most of Italy, though there are party notables who possess such a strong local, clientelistic, or affiliated organizational base that they are not effectively controlled by the organization. The Liberals and Monarchists are much more parties of notables, and the role of the organization is limited or nil. Little has been written of a scholarly nature about the organization and selection of candidates in the MSI.

28 -25- In a system such as the Italian there is l i t t l e competition between individual candidates of different parties. I t is the party that counts, and the parties that are closest together in subcultural networks and policy preferences are the ones that compete most in elections. Within parties, there is a subtle competition for preference votes but the process has not been studied. Turnover of personnel between one parliament and the other thus represents different things in different parties. For the Communists i t generally represents a deliberate attempt to keep the delegation young and representative of the diverse groups within the party. In the socialist parties i t more likely represents shifting prestige of leaders within local and national party organs. In the DC i t represents a combination of these factors plus some particularly Christian Democratic considerations. What i t does not usually represent in comparison with, for example, the United States--is the rejection by the public of particular individuals. I t is true that important public figures are sometimes not reelected, but the reasons are likely to be found in intraparty politics; the most important party figures can be assured of election by being placed on the ballot in constituencies in which the party is very strong and being guaranteed the preference votes of the party militants. E. Continuities and Discontinuities In comparing electoral returns for Italy in the twentieth century i t is important to consider the effects of the twenty years of Fascism. Optimists view this as a parenthesis in the democratic evolution of the country; pessimists suggest that i t was the necessary culmination of tendencies within the liberal prefascist regime. Political tendencies within Italy have been

29 -26- remarkably constant throughout this century. Labels, personnel, and situations, however, have changed so much between prefascist and postfascist Italy that the u t i l i t y of general comparisons between the party and electoral systems of the two is open to question. On the other hand, there are important continuities such as the persistence of radicalism in certain areas, the continuing relationship between the strength of the church and of the Popular and Christian Democratic parties in some areas, the traditionalism of parts of the South, and so on. The postwar period forms a natural unit of analysis. Extending the analysis further into the past requires particular attention to definition of concepts, variables, and units of analysis. Undertaken with care, i t is likely to be quite rewarding. I I. THE ACT OF VOTING TODAY A. Elections and Ballots There are two major types of elections in Italy, the national parliamentary balloting, called elezioni politiche, or political elections, and local elections, labeled elezioni amministrative, or administrative elections. In the former, citizens vote for lists and express preferences among candi dates for the Chamber of Deputies and for a single candidate for the Senate. The latter are for representatives to the communal and provincial councils. Citizens of the five special regions (Sicily, Sardenia, Valle d'aosta, Trentino-Alto Adige, and Fruili-Venezia Giulia) have been voting for representatives to the regional assemblies as well, and the other regions elected regional representatives for the f i r s t time in Although referenda are provided for in the constitution, parliamentary elections are

30 -27- the only national elections that have been held under the 1948 constitution. A referendum on the monarchy was carried out before the constitution took effect. This constitution gave the Chamber a five year term and the Senate a six year term, but permitted the dissolution of either house under certain conditions. In practice, the Senate was always dissolved with the Chamber so that both have been replaced in the same elections. A constitutional amendment has regularized this practice by establishing a maximum five year term. All parliaments to date have lasted the f u l l five years. The actual date of the election, however, is set only at the time of dissolution and can fluctuate over a period of several months, a practice that can be frustrating to scholars planning surveys. The ballot for a parliamentary election contains the symbol or symbol and name of each party presenting a l i s t in the constituency. The party's position on the ballot in each constituency is determined by the date of f i l i n g its l i s t in that constituency. Parties that have difficulty in agreeing on a l i s t thus are penalized by being lower on the ballot, though the importance of position is unknown. The ranking of names on the l i s t is determined by the party. Often, a well known national leader or two w i l l head the l i s t with others following in alphabetical order, but this is at the discretion of the party. Although the lists are widely publicized, the ballot itself contains only the party symbol or symbol and name. Preferences must be written in; one technique is to provide voters with a tiny stencil containing the name; the candidate's number on the l i s t may be used instead of his name.

31 -28- B. The Legal Facilitation of Voting Elections are administered by the Ministry of the Interior. Voting is from 6:00 a.m. to 10:00 p.m. on Sunday and from 7:00 a.m. to 2:00 p.m. on Monday. These are legal holidays and reduced fares facilitate travel. Voting booths are widespread and access is seldom a problem. There are no provisions for absentee voting, a factor that causes considerable disruption to those Swiss, German, and other foreign industries that employ large numbers of Italian workers. Special polling booths are set up in hospitals and rest homes, and servicemen and sailors may vote in communes other than those of their legal residence. C. Elections Since the Fall of Fascism In a political system that dates only from 1943 i t is not easy to determine which political situations are normal and which are non-normal. However, changes from election to election are very slight and the current voting patterns show no signs of altering dramatically in any five year period. Longer range trends are another matter. There are several aspects of the elections in the postwar period that should be clarified, i f only to make reading of the overall returns more meaningful. The f i r s t is the stability of the l e f t i s t vote combined with the great changes that took place in its distribution in the f i r s t three elections. In the elections to the Constituent Assembly that drew up the new constitution, the Socialists outpolled the Communists. The former party soon split, however, and the majority faction entered into an electoral alliance with the Communists for the 1948 election in which far more Communists than Socialists were elected from the joint l i s t. By 1953 the