The nature of American politics

In attempting to describe and explain the operation of a political system, it is necessary to arrange a vast amount of detailed information into a recognisable pattern that will give meaning and shape to the activities of those who live in it and make it work. In the case of the United States, the problem of identifying the major determinants of political behaviour is complicated by the enormous diversity of American life, and by the way in which constitutional structures and the patterns of political action are continually acting and reacting upon each other. Before we plunge into the detail of American politics, therefore, it is necessary to reflect for a moment upon the theories that have been proposed to explain the motive forces behind political systems, and the implications of these differing explanations for our understanding of the American system. These ‘models’ of political life will help us through the complexities of American politics at all levels of activity, in the electorate at large, in the structures of party and pressure groups, and in the workings of congressional and presidential politics.

Models of politics

One of the most powerful sources of political loyalty and action has always been the sense of attachment to a region or community. When this identification becomes so closely interwoven with the interests of a particular area or a particular group of people to the point where life ceases to have any real importance other than within that context, then people may be prepared if necessary to die to defend those interests. In countries with a very highly developed national consciousness it is the nation itself that becomes the sole focus for this sort of loyalty, but on the way to the realisation of such national solidarity there are many stages. Local and regional loyalties can be as important as the attachment to the country as a whole. The United States grew out of distinct colonial communities, extending gradually across the continent, in a way that tended to emphasise local loyalties. The constitutional structure of federalism that was evolved in 1787 gave opportunities for the continued expression of regional loyalties through the governments of the states. Thus American political history has been strongly characterised by sectional patterns of behaviour, in which the inhabitants of a particular geographical region, at all levels of society, have felt themselves united against the conflicting interests of other sections. The most dramatic confrontation of this sort was, of course, the Civil War in which North and South became for

a time distinct warring nations. But at a less dramatic level, sectionalism has been a moving force in American politics throughout its history. The unity of the section was dependent upon some common interest which set it off from the rest of the country and which was of sufficient importance to unite its inhabitants in spite of class or other internal divisions. Frequently this common interest was economic, a crop or product upon which the whole livelihood of the region depended, such as the importance of grain for the states of the Mid-West, and of cotton and tobacco in the South. Thus throughout the nineteenth century agricultural sectionalism deeply affected American political behaviour. The historian Frederick Jackson Turner described the sections of the country as faint reflections of European nations. The extreme example of sectional loyalty was provided by the presidential election of 1860, in which in the whole of ten Southern states not a single vote was cast for the candidate of the Republican Party, Abraham Lincoln.

In the last quarter of the twentieth century such extremes of sectionalism no longer exist, and indeed the United States has developed a sense of national identity and unity that in its own way is more cohesive than that of older nations in Europe. Yet sectional and regional factors continue to play a vital role in the working of American politics, a role that can be observed in the stubborn decentralisation of the party system, in the machinery of elections and in the working of congressional politics. It is in the interrelationship between this unique brand of nationalism and the reality of the decentralisation of political power that the special quality of the American system is to be found.

The second model of political motivation is that which looks to the class structure of society as the major determinant of political behaviour. Although a number of political thinkers, such as Locke and Montesquieu, have emphasised this aspect of political behaviour, it was Karl Marx who saw class as the ultimate explanation of people’s actions. Taken to extremes this is, of course, quite incompatible with sectionalism as a force in politics. If political loyalty is really a matter of social class, then regional loyalties will have no part to play in the political system, and, to the extent that these regional loyalties continue to exist, then class solidarity across the nation will be diminished. In fact, recent American political history is, in large part, the story of the complex interaction of these two political motivations, with sectionalism

declining as class-consciousness waxed. Each of these styles of political behaviour has, of course, very different implications for the type of party system one would expect to find. Indeed, if either sectional or class politics is taken to the extreme, then party politics as we understand it would be ruled out. There would simply be civil war, either between geographical regions or between classes. The working of the democratic system depends upon the fact that these extremes are never realised, and that political parties must appeal both to different sections of the country and to different classes of the population.

Our third approach to the political system we may describe as the pluralistic approach. This views the political system as a large number of groups each with a different interest, so that politics is a continually changing pattern of group activities and interactions. Economic, class and geographic factors are important parts of the pattern, but many other kinds of groups are also important: religious groups, ethnic groups and other social groupings. Furthermore, although economic groups play an important part in the political system, they do not coalesce into two or three big classes for purposes of political action. They are divided among themselves, union opposing union, one type of producer battling with his competitors, agriculture ranged against industry, small businessman against big businessman, the retailer against the manufacturer, and so on without end. Class and regional loyalties are fragmented, each group seeking for support to win its battles wherever that support is to be found. Thus we have a picture of the political system as a collection of a very large number of groups, of varying size and importance, battling for their interests in a society where no single group dominates. Since the membership of these groups overlaps considerably, there are Catholic businessmen and Protestant businessmen, Irish-American labour leaders and Italian-American labour leaders – there is a continual set of cross-pressures upon the leaders of these groups which helps the processes of compromise between them and moderates their demands. At the extreme, the role of government in such a society is simply to hold the ring, to act as referee between the groups to enable the necessary bargaining and compromise to take place. The political machinery becomes simply the mechanism through which equilibrium is achieved between the contending interests. As the government’s main autonomous interest becomes that of maintaining law and order there is little scope for active leadership to give direction to national policy, and political parties have little coherence or discipline, being merely organisational devices devoid of policy content. Pluralism is very much an American view of the political process and many accounts of the working of the system of government, and in particular the role of interest groups in it, are couched in these terms. It is essential to approach American politics from a pluralistic viewpoint but, as with the other models so far discussed, the temptation to push it to an extreme as the sole explanation must be resisted.

A rather different approach to the nature of the American system, but one closely related to both class and pluralistic theories of politics, is the belief that the United States is governed by a series of elites, or indeed by a single power elite. The latter view, associated with the name of C. Wright Mills, tends to place great emphasis upon the power and wealth of those groups in the population that control crucial areas of the economy. President Eisenhower, himself a great general, warned against the influence of the military-industrial complex when he came to the end of his term of office. Thus, certain relatively small groups of men cease to be just part of an internal bargaining process and become, behind the scenes, the real rulers of the country. The formal political machinery becomes less and less significant in the taking of the really important decisions, so that the electorate, and even the Congress, is bypassed. This interpretation of American politics, even though it can degenerate into a conspiracy theory that attempts to explain every important decision as the results of the secret manipulations of the power elite, must be given serious consideration. There are elite groups in the United States, as in any large industrial society, and they exercise great influence in certain circumstances, a fact that will become clear when we examine the way in which specific policy decisions are arrived at.

The final model of political behaviour we must employ in the analysis of American politics is individualism. In the other accounts of the political system to which we have referred, a class, a section or a group swallows up the individual. Political behaviour is ‘determined’ by class ideology, regional loyalty or group interest, and the individual has little or no significance in affecting the outcome of political situations. Such interpretations of political life seem to bear little relation to the mainstream of traditional democratic thought. For theorists such as John Stuart Mill the individual citizen was the central concern of writers on politics, and personality and individual choice were crucial elements in the way in which political decisions are taken. It is ironic that it is in America, the land of individualism par excellence, that students of political behaviour have demolished the classical description of the democratic political system as composed of rational, informed individuals making up their own minds. They have suggested that, in twentieth-century Western democracies at any rate, the influence of family, class, local community or other relevant social grouping is far more important in determining voting behaviour than knowledge of the issues that face the electorate. In reality, however, individualism plays a role of greater importance in America than in the political system of any other modern democratic state. To attempt to describe the working of the American political system without paying great attention to the importance of personal factors in the choice of candidates, or to the influence of personality on voting behaviour, would be to miss the very essence of American political life.

Each of these ‘models’ of political behaviour and motivation has, of course, very different implications for the type of party organisation that we would expect to find in systems of government in which they play a dominant role. They suggest very different opportunities for the exercise of leadership in the political system, very different attitudes towards ideology and ‘issues’ in the political process, and very different results in terms of party cohesion and discipline, particularly in the legislature. Taken to extremes, these models of the political system are mutually exclusive, each giving rise to a very different style of political life. The fascination of the American political system lies in the fact that it represents a complex amalgam of all these different patterns of politics, in a constantly changing kaleidoscope of sectional, class, pluralistic and individualistic styles of politics. None of these ‘explanations’ of political behaviour can be written off as insignificant, and, equally, none of them can be considered to be the dominant pattern of American political life. The significance of each of these elements differs from time to time, from issue to issue. At one point, because of economic circumstances, the class factor may become relatively more important in the understanding of the political situation. At another, as the result for example of an external threat, the ‘military-industrial complex’ may exercise considerable influence. When economic and external crises recede, personal and pluralistic factors may dominate the political scene. American politics are conducted at several levels and in many different arenas. The significance of one or other political style may alter at the level of presidential politics from that of congressional politics, or differ at state level from that of the federal government, or in the party system as compared with the structure of pressure groups.

It is in this spirit that we must approach the study of American politics, seeking out the elements of class, sectional, pluralistic and individualistic politics, putting them in perspective at the different levels of political life, and exploring their implications for political organisation. Only in this way can we hope to make sense of the complexity and diversity of American political patterns and to see the political system as a whole.

Sectionalism and nationalism

Sectionalism is the tendency of people in a particular geographical area, such as the South, New England or the Mid-West, to give their primary political allegiance to that region and its interests. Sectionalism has been a factor in American politics ever since the differing characteristics of the seventeenthcentury settlements in Virginia and Massachusetts began to interact with the differing climatic and economic conditions to be found on the southeastern and northeastern seaboards. The Constitution adopted in 1789 represented a bargain between the Northern and Southern states that provided the uneasy basis for American political life until the outbreak of the Civil War. The Republican Party, the party of Abraham Lincoln, was the party that led the Northern states to victory in that war; the Democratic Party was associated with the defeated South. This gave a twist to the distribution of political power in America that persists to the present day. It was the election of 1896, however, that set the high-water mark of sectionalism as a political force in modern politics. The populist supporters of William Jennings Bryan gained control of the Democratic Party’s Convention in that year, and secured his nomination as the party’s presidential candidate. Populism represented an attack upon privilege and upon the power of financiers and industrialists; it was a movement of the ‘common man’, and as such it would seem to mean an injection of a strong class element into the American political scene. But although the Democratic Party platform of 1896 pledged the party to support the interests of the farmer and the labourer, it was only able to capture the votes of the agrarians of the South and West. The industrial workers of the East voted strongly for the Republicans and the extreme sectional pattern of politics in that election can be gauged.

At the end of the nineteenth century there was a reaction against the populists and what they represented. Conservatives in North and South alike set about the creation of political machines in which they could maintain their ascendancy, working through the medium of the Republican Party in the North and through the Democratic Party in the South. Before 1896 the two parties had been relatively well balanced throughout the country, but by 1904 there remained only six states in which the parties were evenly matched. The Republican Party almost ceased to exist in the South, while the Democrats were almost as powerless in the North.

As the twentieth century progressed, however, the forces that had produced and maintained this sectional alignment were gradually eroded. The nationalising influences of the growth of communications and of America’s involvement in world affairs tended to diminish local patriotism, but perhaps the most important factor was the problem of an economy that was no

longer ‘open’ in the sense in which it had been in the expansive years of the nineteenth century. The years of economic depression were the background to the Democratic victories of Franklin D. Roosevelt in 1932, when the old sectional alignment was shattered, and in 1936, when he won every state in the Union with the exception of Maine and Vermont. Genuine competition between the two parties gradually spread into more and more of the states and even in the deep South a new kind of politics began to emerge.

Yet important as sectionalism undoubtedly was at the beginning of the century, we must not overstate its importance. Although one party might consistently win elections in a particular state or region over a long period oftime, this fact might mask the existence of a strong minority for the opposing party. As the century progressed the minority gradually began to achieve some sort of parity with the previously dominant group. V.O. Key has shown that even in the Deep South, that most distinctive of regions, it was only in matters concerning racial problems that the South differed profoundly from other regions in the make-up of public opinion. Much of the impression of Southern conservatism in economic and social matters is largely due to the way in which the one-party politics of the Southern variety distorted the operation of the machinery of representative government. However, even if public opinion does not differ radically in different parts of the country, the fact that the political system produces significant regional differences in the attitudes of Senators and Congressmen is of the greatest importance. It is in Congress that the effects of regional differences in political behaviour still have their greatest impact on the decisions of government.

Sectionalism declines in importance as nationalising forces develop, but the regional differences still remain as the bedrock of political behaviour. Thus since the 1930s there have been two kinds of presidential election. When, for one reason or another, tides of support for one candidate sweep across the country there are ‘landslides’ which may completely swamp any regional or sectional differences. But when the election is more closely contested, and popular support for the candidates is more evenly divided, then the regional differences re-emerge and can become decisive in determining the outcome of the election. Roosevelt’s landslide victory of 1936 was an example of an election in which regional differences were completely irrelevant to the outcome. Similarly in 1964 Lyndon Johnson achieved an overwhelming victory over Senator Barry Goldwater, winning majorities in forty-two of the fifty states. In 1972 Richard Nixon, with 61 per cent of the votes cast, won every state except Massachusetts, and in 1984 Ronald Reagan also won in 49 states, losing only in Minnesota, the home state of his opponent, Walter Mondale. Such sweeping electoral victories represent decisive national verdicts, but in more closely contested elections sectional differences may be critical. Thus, in 1960 the victory of the Democratic candidate, John F. Kennedy, was achieved by the slimmest possible margin of 0.1 per cent of the total votes cast. The significance of sectionalism in the result can be judged

Kennedy won a majority of the states of the South and East, whilst Richard Nixon almost swept the board in the West and Mid-West. For a number of reasons, particularly his support of civil rights and his Catholicism, Kennedy performed less well in some parts of the country than others. In the close election of 1976, between Jimmy Carter and Gerald Ford, sectional groupings again became apparent. With only 51 per cent of the vote going to Carter, his victory depended on his greater pulling power in the South and the industrial North, while Ford won every state in the West as well as a number of Mid-Western states. The election of Bill Clinton in 1992 also illustrates the underlying sectional nature of American politics. Clinton gained only 43% of the total vote as opposed to 38% for George Bush and 19% for Ross Perot. Clinton won every West Coast State, all the northeastern states except Maine (the home state of George Bush), and a strip of Mid-West states. Bush won most of the states in the South and a number of Mid-West and mountain states. In 1996 the sectional pattern was even clearer. Clinton won every northern state except Indiana, and the states of the Far West; Robert Dole won seven states of the old Confederacy plus a block of ten western states.

Sectionalism was particularly evident in the very close election of 2000. In the words of Gerald M. Pomper, ‘Not only two candidates, but virtually two nations confronted each other’. The Democratic candidate, Al Gore, won 20 states in the North and East and on the Pacific coast; George W. Bush won all the states in the South, the Border states, the Mid-West and the West, with the exception of New Mexico

The evidence suggests, therefore, that the United States has become progressively more of a nation in the political sense since the Second World War. Increasingly, the tides of opinion sweep across the country, swamping the regional differences that were once so dominant in American politics. Nevertheless, the extent of this ‘nationalisation’ of American politics should not be exaggerated. When election results are close and the nation is uncertain about its choice of a leader, it tends still to divide along regional lines, the old differences re-emerging. Furthermore, the nationalisation of politics is apparent at the presidential level, but it does not extend down to the congressional level. Popular Republican presidential candidates like Eisenhower and Reagan found that their popularity did not translate into victories for their party in Congress. Thus in 1980 the great landslide for Ronald Reagan  nevertheless left him facing a House of Representatives dominated by a majority of 242 Democrats over 193 Republicans, and again in1984 his popularity and his sweeping presidential victory could not secure a majority in the House of Representatives for the Republican Party. In 1996 President Clinton, the candidate of the Democratic Party, won the election for the presidency, but the Republicans secured majorities in both the House of Representatives and the Senate. However, the increasing polarisation of American politics in recent decades, which we will examine in Chapter 3, resulted in Republican majorities in the Senate and the House of Representatives in 2004, alongside the Republican President, George W. Bush.

 

Urban rural suburban politics

The relatively straightforward categories with which we began this chapter soon begin to look somewhat inadequate when we delve into the rich detail of the American political scene, for sectional, class and other aspects of the political pattern overlap and intermingle. The first major overlap is represented by the blending of geographical sectionalism and class politics owing to the uneven distribution of industry across the country, and to the consequent divisions between the urban and rural populations. During the nineteenth century sectionalism was largely the consequence of the differing crops and products resulting from different climatic and soil conditions. With the growth of industrial power, and the rapid concentration of population in urban centres, however, the clash of interest between city and country became a vital factor in the political scene. The greatest impact of the flood of poor immigrants at the turn of the twentieth century was felt in the cities, and at the same time there was an internal migration from the farms to the cities. The result was the concentration of the underprivileged, the poor and the less well educated in the great cities of the North and East. The proportion of the population of the United States living in cities of over 100,000 people rose from 12.4 per cent in 1880 to nearly 30 per cent in l930, during which time the total number of people living in such cities shot up from 6 million to 36 million. Here was the raw material for the transformation of the political system of the United States into something very different from that of the sectional alignment of 1896, but one in which geography continued to play a part.

The Great Depression of the 1930s was the catalytic agent that transformed this vast mass of human beings into what came to be called the ‘normal’ Democratic majority. Samuel Lubell has shown that it was Alfred E. Smith, the unsuccessful Democratic candidate for the presidency in 1928, before the Depression actually began, who first drew the political battle lines between the cities and the rural areas. It was the New Deal policies of Franklin D. Roosevelt, however, that fixed the urban masses of the North and East in the Democratic column. In 1924, in the twelve largest cities in the United States, the Republicans had an overall majority over the Democrats of 1,252,000 votes. Twenty years later, in the same twelve cities the Democratic majority over the Republicans was 2,296,000 votes. This enormous change of allegiance represented, in Lubell’s phrase, the revolt of the underdog. Economic, ethnic and religious factors combined to create a body of support for the Democratic Party that broke the hold that the Republicans had had upon the presidency, with short breaks, since the Civil War. The urban masses of the North became one prop of the Democratic Party, in uneasy alliance with the southern whites who used the Democratic Party to maintain white supremacy in the southern states. This remarkable coalition, the New Deal coalition, papered over the deep ideological cleavages between Northern and Southern Democrats. They came together because historical accident, and the already existing organisational structures, provided both wings of the Party with an unprecedented opportunity to exercise power. This was the basis of the Roosevelt system that set the pattern for American politics after 1932 for thirty years. The northern wing of the Party dominated the presidency, while the southern wing gained a strategically vital position in Congress, in particular by its control of committee chairmanships. The president and the Congress came to have different ‘constituencies’ because of the methods of their election. To gain election to the presidency a candidate had to woo California and the great populous urban states of the North and East, while Congress was more representative of, and more responsive to, suburban and rural interests. The resulting tension between president and Congress on a wide range of policies became, because of the coalition nature of both great political parties, a great source of internal strife within the parties as well as between them.

The years since the end of the Second World War have, however, brought a new complication to this pattern of urban–rural politics. The rapid development of suburbia has transformed America both visually and politically. These suburbs, spreading out many miles into the country around urban areas, represent a whole new way of life, and their impact upon politics is as great as was that of the Roosevelt revolution. They represent a new type of community, in which the old guidelines to political behaviour are no longer so reliable. The population of the suburbs is ethnically diverse while its economic composition is relatively homogeneous. Neither the old pattern of city politics based upon ethnic differences nor the urban–rural alignment is so relevant. As Robert C. Wood has pointed out, the suburbs fit neither into the class patterns of the early twentieth century nor into the sectional patterns of the nineteenth. Yet old political loyalties die hard, and the persistence of party allegiance is one of the facts of political behaviour, even when the original reasons for choosing one side rather than another have long become irrelevant. Perhaps the greatest significance of the rise of suburbia is to provide an overlap with another of our patterns of political behaviour, individualism. Suburbanites tend to think of themselves as independents in politics, discriminating between candidates rather than parties, paying attention to different issues at the various levels of government, and making use of all the opportunities for ticket-splitting that the American electoral system provides.

 

Pluralism in American politics

The group basis of politics became the subject of intensive study only in the twentieth century, and significantly it is two American works, The Process of Government by Arthur F. Bentley, published in 1908, and David B. Truman’s The Governmental Process of 1951, that most typify this approach. We shall look closely at this view of the political system in a later chapter when discussing interest group politics, but here we shall concentrate upon three aspects of group politics of particular significance at the level of electoral behaviour, the politics of class and of ethnic and religious groups. The overlapping group nature of the American political system can be gauged from the results of the 2004 presidential election

 

Class

That class is a factor in the voting behaviour of the American electorate is clear. The further up the income scale the more likely people are to vote Republican. At the end of the nineteenth century the triumph of sectionalism had as its corollary a relative lack of emphasis upon class in American voting behaviour; but as sectionalism declined in importance, the class alignments of American voters became more significant. The growth of an underprivileged urban working class during the first two decades of the twentieth century formed the basis of a transformation from the sectional politics of the earlier age. The economic depression of the late 1920s and the

1930s established a distinct relationship between class and party allegiance, though one that was subject to considerable variation from election to election. Franklin Roosevelt’s New Deal policies turned the Democratic Party into the champion of the working man. A majority of manual workers voted for the Democrats throughout this period and, with the exception of the elec tion of 1936, a majority of non-manual workers voted for the Republicans. Lipset has pointed out that, if these crude occupational categories are broken down, we find that, the further down the social scale a group is placed, the greater the percentage preference for the Democratic Party. In 1948 nearly 80 per cent of American manual workers voted for the Democrats, a higher percentage than most left-wing parties in European countries could achieve. In the 2004 election, the Democratic candidate, John Kerry, received 63 per cent of the votes of those electors with a family income of less than $15,000 per annum, but only 41 per cent of the votes of those with a family income of over $100,000 per annum. Thus a connection between social class, income and voting behaviour clearly exists, but it is by no means a simple one, and there is a considerable variation from one election to another in the extent of ‘class voting’. In 1948, for example, the issues before the electorate were largely of an economic kind, and the voting patterns reflected class interests to a high degree. The elections of 1956 and 1972, however, present a different picture. In 1956 almost as many manual workers voted for the Republican, Dwight Eisenhower, as for the Democratic candidate, and in 1972 the Republican candidate, Richard Nixon, received the support of 57 per cent of the manual workers, as against only 43 per cent voting for his Democratic rival, George McGovern. In 2004 61 per cent of members of trades unions voted for John Kerry, the Democrat, while only 38 per cent voted for his Republican rival, George W. Bush.

There are therefore wide variations from election to election in the extent to which voters are influenced by their perceptions of their class interests in their voting behaviour. Of course, the attribution of sectional or class motivations on the basis of the sort of statistics quoted above is a very difficult exercise. What is apparently class voting may be motivated in quite different ways, because religious, ethnic and regional groupings all overlap with class to a very considerable extent. Exactly why a low-paid Irish Catholic industrial worker in the North-East votes Democratic rather than Republican can hardly be explained by any simple formula. Angus Campbell and his co-authors in their study The American Voter found that a third of the American population was ‘unaware’ of its class position, and that social class played a significant role at a conscious level in the political behaviour of only a fairly restricted and sophisticated portion of the population. The relative volatility of the American electorate, the readiness to switch votes from one party’s candidate to another at successive elections, is also a measure of the limitations upon appeals to class orientation as a source of voting behaviour. American elections can produce ‘landslide’ results that would be unthinkable in a system where stable class voting is the norm.

These figures reveal that, from 1936 to1984 at any rate, there was a startling propensity for large sections of the electorate to switch their allegiance from the candidate of one party at a presidential election to the candidate of the other party at the next election, or at least to abstain from voting for the party that they previously supported. The tendency to switch votes from one election to the next reflected the declining significance of sectional influences, the increasing ‘nationalisation’ of American politics. The era of the ‘post-industrial society’ ushered in a political system in which the electorate was more independent, less committed to a particular party allegiance. Landslide victories by Roosevelt in 1936, Johnson in 1964, Nixon in1972 and Reagan in 1984 showed how waves of sentiment could sweep through the whole country, carrying all before them.

However, the above figures show a greater degree of balance between the two major parties in the period since 1988. This reflects the swing of many white voters in the South from their traditional allegiance to the Democrats, giving their support to the Republicans, at any rate in voting for the presidency. The change in the voting behaviour of Southern whites is the most significant change in the American political system in recent decades, making the South the pivotal section in presidential politics

Ethnic politics and multiculturalism

From the early 1930s, minority groups such as Italian-Americans, Irish - Americans, the Jewish community and blacks tended to vote for the Democratic Party. The tendency was particularly noticeable in the large cities, but in 1948 in the small city of Elmira, New York, Berelson and his collaborators found that, while 81 per cent of the white native-born Protestants voted Republican, only 33 per cent of the Jews, 19 per cent of blacks and 18 per cent of the Italian-American voters followed suit. Here again, there is an overlapping of our simple categories. Minority groups, particularly blacks and foreign-born immigrants, generally belong also to the lower-income groups. Nevertheless, Berelson found that the tendency of these groups to vote for the Democrats was not particularly affected by their socio-economic position. The significance of such group voting can be very great. In the nineteenth and early twentieth centuries blacks had traditionally been loyal to the Republicans, the party of Lincoln; but Roosevelt’s economic policies, and the commitment of the Democratic Party at the national level to civil rights, led blacks to support the Democratic presidential candidates in ever-growing numbers. Civil rights legislation made it possible for more and more blacks to get their names on the voting rolls, and as a consequence their electoral importance, particularly in Southern states, increased dramatically. Thus in two of the most closely fought elections since the Second World War – the victory of John F. Kennedy over Nixon in 1960 and of Jimmy Carter over Ford in 1976 – the allegiance of black voters to the Democratic Party’s candidates was a vital factor in delivering victory to them. In 2004 88 per cent of black voters cast their votes for the candidate of the Democratic Party, John Kerry. Other ethnic groups play less dramatic, but no less significant roles in electoral behaviour.

In the 1950s and 1960s Republican Senator Joseph McCarthy led a crusade against communists in government. Samuel Lubell, in his Revolt of the Moderates, found that McCarthy’s power base could be traced to those ethnic groups who were deeply affected by the cross-pressures they experienced as a result of America’s involvement in two world wars. Lubell showed that in McCarthy’s home state of Wisconsin there had been a considerable shift away from the Democratic Party by German-Americans, because that party was associated with the policy of war against Germany. In 1932 eight largely German Catholic counties in Wisconsin voted 74 per cent Democratic, but by 1952 the Democratic vote had dropped to 32 per cent. This change was reflected in German-American communities throughout the country. Lubell emphasised that, as the, then, second most numerous ‘foreign stock’, the German-Americans held the balance of power in many states, especially in the Mid-West. Such support provided a formidable reservoir of emotion upon which a man like McCarthy could draw, for these people wished to emphasise their American patriotism, and at the same time to give expression to attitudes towards communism that were related to their religious beliefs. In the same way the complex interrelationship between Catholic doctrine and the need to assert their Americanness led the Irish Catholics of New York to give overwhelming support to McCarthy.

At the lower levels of the political system ethnic divisions play a crucial role. The politics of New York City or Los Angeles present the extreme picture of ethnic diversity and its effects. In New York City in 2000, the total population was more than 8 million; over a quarter of the population were of Hispanic or Latino origin and a quarter were black or African-American, although there is some overlapping of these categories. Nearly 10 per cent were of Asian origin. In 2005 Michael R. Bloomberg, a liberal Republican, was re-elected as Mayor of New York, winning the support of 50 per cent of black voters and a third of the Latino voters. In the City of Los Angeles in 2000, 46.5 per cent of the population of nearly 4 million were of Hispanic or Latino origin, 10 per cent were of Asian origin and 11.2 per cent were black. 57.8 per cent of Angelinos reported that a language other than English was spoken in the home.

This complex of ethnic and class divisions not only has implications for local or city elections, but also may be a decisive factor in congressional politics, or even in the complicated processes by which the president of the United States is chosen. It is possible that differing attitudes towards such concepts as ‘the public interest’ may be affected by ethnic origins, although it is very difficult to demonstrate direct relationships between particular ethnic groups and specific attitudes towards governmental structure or policy. James Wilson and Edward Banfield have suggested that ‘Anglo-Saxon’ and Jewish voters in certain Ohio communities were more favourably disposed to increasing public expenditure than were Polish or Czech voters enjoying the same level of income. Some of the ethnic divisions within American society cut very deep, as is evidenced by the position of blacks, yet one might expect that, as groups of ‘hyphenated Americans’ become assimilated both culturally and economically, they would become indistinguishable in their political behaviour from the rest of the population. This ‘assimilation theory’ may well be correct in the long run, but it is important to remember that great waves of European immigrants were still flowing into the United States until well into the first half of the twentieth century, and the second half was characterised by a massive influx of Puerto Ricans, Mexicans, Vietnamese, Koreans and other Asians. It is one of the most frequently observed facts of political life that political loyalties tend to outlive the factors that created them. Local conditions and local leadership can give a quite remarkable persistence and coherence to ethnic political behaviour.

America has always been ethnically diverse, and as we have seen, in the last half of the nineteenth century and in the twentieth century, immigrants flooded into the United States on an unprecedented scale. But, although the origins of Americans were to be traced to countries all over the world, American culture was amazingly uniform. English was the language used in schools and public institutions. Immigrants were under pressure to learn English, and in many cases ‘foreign-sounding’ names were anglicised. Furthermore, the educational system was an overt instrument of Americanisation, to instil values and beliefs, in particular the tenets of the Constitution, into the children of immigrants. The pressures were all on the need for homogeneity, and for a very good reason. The fear of subversion, not necessarily in the sense of treasonable actions, but in the sense of undermining the consensus on which the political system depended, was ever present. The Vietnam War, and more recent immigration, legal and illegal, have brought about a change in these attitudes. The Vietnam experience and the way in which dissent was legitimised damaged the confidence of Americans in their cultural identity. The effects of this experience coincided with a large-scale immigration of a group of people – Latinos – who did not wish to be assimilated in the way in which earlier immigrant groups had been. By 2004 there were over 41 million people of Latino or Hispanic origin in the United States, and their numbers continue to increase. Federal government programmes of bilingual education began in 1968 to provide education for Spanish-speakers, but the question then arose whether this kind of programme was intended to help Latinos to assimilate to the English language culture of America or to be able to survive in that culture whilst retaining Spanish as their first language. In 1974 the Supreme Court ruled in Lau v. Nichols that teaching students in a language they did not understand was a violation of their civil rights, thus giving an impetus to bilingual programmes. Legislation was passed requiring that election registration forms, ballots and election materials should be made available in languages other than English in districts where a significant minority spoke a language other than English as their first language. The language problem took a new turn in 2006 with the launching of a Spanish version of the American National Anthem, entitled ‘Nuestro Himno’, bringing about a chorus of condemnation from conservative groups.

Predictably there was a backlash against this ‘Latinisation’ of America. Moves began to establish English as the ‘official language’ of the United States, and to try to limit immigration from Latin America. By 2005 twentyseven states had legislated to make English the official language of the state, and a Bill making English the official language of the federal government was passed by the House of Representatives in 1996, but did not reach the statute book. The nature of the Latino challenge to the established ‘Anglo’ culture of America has great political implications. The Spanish-speaking section of the American electorate is growing, and will continue to grow. Their influence at local level, such as in cities like Los Angeles, is already considerable and their voting power at state and eventually federal level will continue to increase. The long-run effect of Latino immigration on the American political system is difficult to predict, although it will certainly be profound. The Latino population is not as coherent as the African-American population; it is drawn from a number of different countries, with differing ethnic origins, white, black, American Indian. However, in 2006 the Latino community found a new basis for solidarity against the ‘Anglos’ with the development of an immigrants’ rights movement. Concern about the extent of illegal immigration led to attempts by the government to control the border with Mexico more effectively and to find a solution to the problem raised by approximately 11 million illegal immigrants in the United States. The possibility of large-scale deportations led to demonstrations by hundreds of thousands of Latinos in a number of cities, threats of strikes, boycotts and other action.

Religion and religiosity

Religion has been a factor in American politics ever since the Pilgrim Fathers made landfall on Cape Cod. Furthermore, the regional distribution of religious belief has served to complicate the sectional differences of American politics. In colonial times Quaker-dominated Pennsylvania, the Roman Catholics of Maryland and the Dutch Reformed Church in New York separated Puritan New England and Anglican Virginia. The mass immigration of the nineteenth and twentieth centuries transformed the pattern of religious affiliation, for a very high proportion of the immigrants were Catholic. There are 50 million Catholics in the United States today, more than in Italy. Because the waves of immigrants headed for the cities of the North and East, the pattern of sectional attitudes became further complicated, and because they were poor, Catholic immigrants contributed a further dimension to the pattern of sectional, class and religious influences that today form the fabric of American politics. Thus the electorate of the Southern states remains almost completely Protestant in composition, whereas the concentration of Catholics in Northern states such as Massachusetts gives to those states a quite distinctive political style. The Statistical Abstract of the United States lists thirty-five different Christian denominations and twenty-one ‘other religions’ to which Americans declared their adherence in 2001, in addition to the nearly 30 million Americans who stated they had no religion.

Like the other factors in American politics, religion plays a significant but varying role at the national level. In state and local politics its impact varies greatly from area to area and from issue to issue; but of its importance there can be little doubt. Religious factors have played an important role in presidential politics, in helping to influence the voting behaviour of Senators and Congressmen, or in the determination of political battles over contraception laws in Massachusetts or the closed shop in Ohio. The comparison between the candidacy of two Catholics for the presidency in the twentieth century with that of John Kerry in 2004 illustrates the way in which attitudes to Catholicism have changed. The candidacy of Al Smith in 1928 evidenced the extent of the bitterness that then existed against Catholics, although his defeat cannot be attributed primarily to religious motivations among the electorate. In 1956, when Eisenhower’s personal appeal was so great, Catholics split almost fifty–fifty between the parties, but four years later roughly 80 per cent of Catholics voted for the Catholic candidate, John F. Kennedy, and only 20 per cent for his Protestant opponent. In the election of 1960 Kennedy increased the percentage of the Democratic vote by nearly 20 percent in states with high proportions of Catholic voters compared with only 2 per cent in states where the number of Catholics was low. The following example, given by John H. Fenton, of the effect of religious divisions within a community, although it may not be typical, shows how religious affiliation can affect politics. In Nelson County, Kentucky, which was half Baptist and half Catholic, Kennedy received only 35 per cent of the vote in four predominantly Baptist precincts, but in five largely Catholic precincts he received 88 per cent of the total vote. Nationally, however, the tendency of Catholics to vote for a member of their own church was more than offset by those Protestant voters who switched to the support of Nixon on religious grounds. By contrast, the candidacy of John Kerry in 2004 seemed to provoke relatively little religious controversy. Although Kerry is a practising Catholic, in the campaign he adopted ‘liberal’ attitudes towards issues such as abortion. In fact he came under attack more strongly from some Catholic bishops who condemned his advocacy of a woman’s ‘right to privacy’. In the event a majority of Catholics voted for George W. Bush, in a slightly greater proportion than the electorate as a whole, perhaps because of their perception of his views on moral issues.

The most important recent manifestation of the significance of religion in American politics lies in the revival of fundamentalist Protestant ideas and the involvement of their proponents in elections and in attempts to influence government policy at all levels. The ‘born-again’ movement and the ‘Moral Majority’ demanded that fundamentalist religious values should be adopted as the guiding lines for action in all fields of government policy. In the election of 1976 the Democratic candidate, Jimmy Carter, received the support of many southern fundamentalists who desired a more conservative and ‘Christian’ approach to government. Carter was, however, a great disappointment to them, and in the run-up to the election of 1980 the Republican candidate, Ronald Reagan, assiduously courted this group. Political divisions along religious lines were particularly evident in the 1984 elections. While Catholic voters split relatively evenly between Reagan and Mondale, 73 per cent of white Protestants voted for Reagan and 80 per cent of those whites who described themselves as ‘born-again Christians’ voted for him. The leaders of the Moral Majority also conducted campaigns against ‘liberal’ Senators and Congressmen, and many of the latter were beaten in the election. In state and local politics fundamentalist groups battled to further their views on subjects such as abortion, women’s rights and the teaching of evolution in the schools.

By the time of the election of George W. Bush in 2000 the connection between the religious right and the Republican Party was firmly established. Bush, himself a Southern Methodist, received the support of fundamentalist Christians, particularly in the South, where 73 per cent of white evangelicals voted for him. But Bush also targeted Catholics, a majority of whom voted for him. In fact, since 1996, as Everett Carl Ladd has shown, it is ‘religiosity’,the extent of regular church attendance, rather than religious affiliation which is important in affecting voting behaviour. Voters who were not active church attenders voted overwhelmingly for Clinton, whilst a majority of regular church attenders voted for the Republican candidate, Bob Dole. In 2000 white Protestants who identified themselves as very highly committed to religion voted 87 per cent for Bush as opposed to 13 per cent for Al Gore, the Democratic candidate.

The role of individualism and personality

The patterns of sectional, class, ethnic and religious politics that we have looked at so far would seem to leave little room for the emergence of truly individualistic behaviour on the part of the electors or their representatives. Yet if we look at the trends of political behaviour in America we find that individualism in politics became increasingly important in the twentieth century. At the end of the nineteenth century, American political attitudes were strongly party-oriented; that is to say, voters identified themselves very strongly with a particular political party. It was in the nature of politics at that time that party identification was founded largely upon historical and regional loyalties rather than upon class but, as the twentieth century progressed and class voting increased in importance, the ties between the voter and his party progressively declined. That these two things happened at the same time was a reflection both of the extreme nature of the sectional alignment of 1896 and of the fact that, as the century progressed, much of the electorate became increasingly alienated from political life. Walter D. Burnham has pointed out that an increasingly large proportion of the electorate are peripheral voters, who are not closely tied to one party or another, and who participate in elections only when they feel strongly moved to do so by the impact of a personality or an issue sufficiently strong to make them reenter the political universe.

Voter turnout declined during the 1970s, and in 1976 only 54.4 per cent of the potential electorate turned out to vote in the contest between Ford and Carter; in 1980 the Reagan–Carter battle tempted only 53.9 per cent of the electorate into the polling booths. In 1996 a low point was reached when only 49 per cent of the electorate turned out to vote for the presidential candidates, Clinton, Dole and Perot. In 2004, however, turnout rose to 60.7 per cent, largely as a result of the public interest roused by the circumstances of the attack upon the World Trade Center and the war in Iraq.

Turnout for congressional elections in years when a president is not being elected is even lower, and voting for state and local elections is lower still. However, we should not be too hasty in our judgement of the American electorate in this respect. The number and frequency of elections is greater in America than in any comparable country, and ordinary people cannot be expected to be in a continuous political ferment. When they do feel strongly about any issue or candidate they have more opportunity than anywhere in the world to make their views known.

In recent decades voters have been increasingly ready to change sides from one party to another, and it has been estimated that as many as 40 per cent of the electorate consider themselves to be independents rather than firm supporters of either of the parties. One of the most important manifestations of this decline in close party identification is to be found in the phenomenon of ‘split-ticket’ voting. At each election the electorate is confronted by a ballot paper, or a voting machine, which allows the voter to cast ballots for a number of candidates for different offices at federal, state and local level. Each of the major parties, and some minor ones, will have candidates for all or some of these offices on the ballot. The voter may simply vote for all the Republican candidates, or all the Democrats; this is voting ‘the straight ticket’, and usually it is much simpler to do, requiring only a single mark on the ballot paper or the operation of a single lever on the machine. The voter is also free to ‘split the ticket’; that is, to vote for one or more Republicans for some offices, and for Democrats for the others – or indeed, where three or more parties appear on the ballot, to spread votes across all of them, voting for individual candidates, regardless of party. At the federal level voters can discriminate between candidates for the presidency, the Senate and the House of Representatives, voting Republican for two of these offices and Democrat for the third, or any combination they wish.

The complexity of the ballot paper and the trouble involved in making the necessary discrimination between candidates might suggest that splitticket voting would be relatively rare. Not so! Campbell and Miller found that in 1952 as many as one-third of the voters split the ticket, and in 1956 two-fifths of the electorate did so. The potential importance of this practice at the federal level was illustrated by the vote in 1952, when nearly 40 million Americans voted for the Republican candidate for the presidency, Dwight Eisenhower, but only 28 million voted for Republican candidates for the House of Representatives.

During the 1960s this pattern of voting Republican for the presidency and Democrat for Senate and House candidates – ‘presidential republicanism’ – became extremely important, particularly in the Southern states. As the ties of party loyalty weakened in the United States, and the behaviour of the electorate became more volatile and independent, the tendency towards split-ticket voting increased. In 1996, 65 per cent of voters reported that they had split the ticket at some time. One of the best ways to measure the propensity of the electorate to split its votes between the candidates of different parties is to compare the vote for the presidential candidates in a particular state with the votes for the senatorial candidates. Both offices, the presidency and membership of the Senate, are ‘national’ offices, and the vote for each is a state-wide vote for one office. In other words, as far as possible, the ‘constituencies’ are comparable. In the election of 1960, 82.7 per cent of the voters in Virginia supported the Democratic candidate for the Senate, but only 47.0 per cent voted for the Democratic presidential candidate, John F. Kennedy. Although partisanship has increased in recent years split-ticket voting is still an important factor in the outcome of elections. In the election of 2004 there were thirty-four Senate seats up for election, and the elections took place on the same day, and on the same ballot paper, as the election for president. In seven states the voters chose the candidate of one party for the presidency and the candidate of the opposing party for the Senate. In five states the Republican, George W. Bush, received a majority of the votes but a Democratic senator was elected. In two states John Kerry was successful at the presidential level, but a Republican won the Senate race. In some of these states the number of voters that split the ticket was relatively small, but in two states the extent of ticket splitting was remarkable: North Dakota and Indiana. In North Dakota, George W. Bush had a majority of 85,599 votes over the Democratic challenger, John Kerry, but in the election for the Senate the Democratic candidate, Byron L. Dorgan, who had been Senator from North Dakota since 1992, had a majority of 113,590 over the Republican candidate. In Indiana, Bush won by 510,427 over Kerry, but the Senate race went to the incumbent Democratic Senator Evan Bayh, who had a majority over his Republican opponent of 593,066.

One of the aspects of the phenomenon of split-ticket voting is that voters may be inclined to vote for incumbent candidates, those who are already in office, whom they judge to have been doing a reasonably good job, irrespective of the voter’s party identification – presumably on the basis of choosing the devil you know rather than the devil you don’t. The importance of being an incumbent is illustrated by the re-election to the Senate of Arlen Specter from Pennsylvania in 2004. In 1980 the Republicans won Pennsylvania for Ronald Reagan and Arlen Specter, Republican, was elected to the Senate for the first time. In 1992 Bill Clinton won the presidency for the Democrats and carried Pennsylvania by a large majority, but on the same ballot Specter was re-elected and, although the Democrats have carried Pennsylvania for the presidency in every succeeding election (although with diminishing majorities), Arlen Specter won the Senate seat for the Republicans for five successive terms. Specter’s success depended in part on having rather different views on policy compared with many of his Republican Party colleagues; he is pro-choice on abortion and has supported stem-cell research and family planning programmes.

Such examples could be multiplied by examining the votes for members of the House of Representatives, by looking at state offices such as the governorship, and so on. Much of this split ticket voting is to be attributed to sectional differences in attitudes within the party, so that the national presidential candidate may have policy views that differ considerably from those of the local party. Clearly there is, however, a very considerable personal element in these split-ticket voting situations, where a senatorial candidate may have a local reputation that brings in a much larger vote than the more remote presidential candidate can hope to generate. It is always difficult to assess the importance of the personality of a candidate upon voting decisions, and it may well be that personality alone is not enough, and will not sway the voters unless there is some other latent factor which the personal appeal of the candidate serves to bring into operation. The importance of personality will also vary according to circumstances. Thus, in areas where party organisation is minimal, personality may be decisive in some state and local elections. Even at the presidential level, the fact that Eisenhower could attract the votes of a quarter of those who normally voted Democratic was startling evidence of what personal appeal can achieve if the circumstances are favourable. Eisenhower would have won the election for whichever party he stood, and at one time or another he was considered as a potential candidate by both parties. Similarly, when he was active in Californian politics, the future Chief Justice of the US Supreme Court, Earl Warren, was able to win the nomination of both parties at the same election for the post of governor. Senator Wayne Morse of Oregon was able to change his party allegiance during his term of office and still gain re-election to the Senate. These are extreme examples, but they indicate that we ignore the importance of personality in American politics only at the risk of altogether failing to understand it.

The most recent expression of the role of personality in American politics, and perhaps the most questionable, is the part played by celebrities. Ronald Reagan, an actor who appeared in innumerable B movies, was elected governor of California in 1966 and again in 1970. He won the presidency in 1980 with a landslide vote, and was re-elected in 1984. Another ‘celebrity’, but not from the entertainment industry, to take up a political carer was John Glenn, the first American astronaut to orbit the Earth, who was elected Senator from Ohio in 1974 and served in the Senate until 1999. Perhaps the most extraordinary example is the election of a wrestler turned actor, Jesse ‘The Body’ Ventura, as Governor of Minnesota in 1998 representing the Reform Party. Ventura appeared in three films with Arnold Schwarzenegger, body builder and star of the ‘Terminator’ film series, who was elected Governor of California in 2003. It is expected that future elections for the governorship of the state will be contested by film stars. This development is perhaps not surprising in an era when television is so important in elections, but it has to be said that another profession still remains more important in providing recruits for elective office – lawyers.

Politics in California

The local sub-divisions of the states – the cities, counties, towns and myriad different elective bodies at local level – also exercise varying degrees of political power. The decentralisation of American political power can best be seen by examining the autonomy and the political role of a state such as California and the relationship between towns and counties and the state government that is their legal master. It becomes clear immediately that local government has significance in America that it has long lost in Europe. To understand the politics of the presidency, or the workings of the Congress of the United States, it is necessary to start by looking at the roots of American politics, at the characteristics of the electorate, at regional variations in political style and behaviour, at the way in which politics is conducted at the state and local level. Only then can we understand the nature of the presidency, its strengths and its weaknesses; only then can we understand the contradictions of a Congress that can be at one and the same time a parochial assembly and a body of national legislators. The study of American politics must build up from its local components in a way that is no longer so true of the study of most European countries, for the diversity that remains at the base of American life prohibits either the easy generalisation or the simple explanation of political behaviour. It is this pattern that we shall follow, constructing a picture of the American scene that will help to make explicable what happens at the more glamorous level of President and Congress, and without which the events that make the headlines in the newspapers of the world are often quite incomprehensible. Let us look, therefore, at the politics of an important state, California.

California, the ‘Golden State’, was admitted to the Union in 1850, as the thirty-first state. The indigenous home of a large number of Native American tribes, it was settled by Spain in the eighteenth century, becoming part of newly independent Mexico in 1821. American settlers started to arrive in the 1840s and began to agitate for independence from Mexico. In 1846 the United States declared war on Mexico and in 1848, following the end of the Mexican War, Mexico ceded California to the United States. Gold was discovered in the same year and the resulting gold rush brought many different nationalities to California to find their fortune.

California is often presented as a microcosm of the United States, but also as an extreme version of American society. It is a very diverse state in many ways. From the Mexican border, the southern desert changes to agricultural land, to the vineyards and forests of the north, up to the border with Oregon 800 miles away. The state has varied industries, including oil and gas production, mining, the computer powerhouse of Silicon Valley, tourism and of course Hollywood. There are great cities, Los Angeles and San Francisco, and innumerable smaller cities and towns. California has a population of over 36 million of which 15 million are white, 13 million are Hispanic, 4 million are of Asian origin, and 2 million are black. If it were a separate sovereign country it would be the world’s sixth largest economy, surpassing most of the world’s most populous countries.

The state constitution reflects the Constitution of the United States in embodying the separation of powers between the legislative, executive and judicial branches. The legislature is bicameral with a Senate and an Assembly. The governor has a veto over legislation, but as in the federal governmentthe veto can be overridden by a two-thirds majority of both houses of the legislature. In addition to the elected governor and legislature a number of state officials are also directly elected: the lieutenant governor, the attorney general, the secretary of state, the treasurer, the controller, the insurance commissioner and the school superintendent. In the nature of the Californian political system these officials can be drawn from different political parties, but are still required to work together. The justices of the state Supreme Court and the appellate courts are nominated by the governor, but then must be confirmed in office by the voters at the next election; the lower levels of the judiciary are also elected. The state constitution comes close to establishing federalism within the state by giving considerable autonomy to counties, cities and other local governments.

The governor has an important role, if an impossible one to carry out effectively. The governor must propose legislation to the legislature and prepare a state budget, which totalled $85 billion in 2005. The problem of persuading the legislature to accept these proposals is a daunting one. Political parties in California are weak, internally divided parties; to quote David G. Lawrence: ‘In the Democratic Party, liberals, moderates, and various caucus groups fight among themselves. In the Republican Party, fiscal conservatives, social issue conservatives, and more pragmatic moderates vie for influence over education, abortion, gun control, gender issues, and gay rights.’ In such circumstances party discipline is difficult to achieve and the situation is made even more difficult by the fact that the governor may be facing a legislature dominated by a majority of the opposing party. When Republican Arnold Schwarzenegger was elected governor in 2003 he faced a Senate and Assembly both with Democratic majorities. These difficulties are compounded by the fact that in 1990 the voters of California passed Proposition 140 which introduced ‘term limits’ restricting state legislators to a maximum of six years in the Assembly and to eight in the Senate.

To add to the complexity of the institutional structure there are three instruments of participatory democracy – the initiative, the referendum and the recall. The initiative allows citizens to propose legislation, which is then submitted to the voters and if approved becomes law, without the leg islature even considering it and without the governor’s signature. The state constitution requires the collection of signatures totalling 5 per cent of the vote cast in the last previous election for a proposition to be placed on the ballot for consideration by the electorate. Two of the very many initiatives to have become law in this way are Proposition 13 in 1978, which slashed the property tax level in half, with very serious consequences for local services, and Proposition 22, passed in 2000, which outlawed gay marriages in the state.

The referendum enables voters to prevent the implementation of state constitutional amendments, or laws intending to raise money through the issue of state bonds. Compared with the initiative, the referendum has been relatively unimportant, but recently the third of the participatory instruments, the recall, has acquired new significance. The recall is a procedure which allows voters to remove from office state and local elected officials, including judges, in the period between elections. In 2003 a movement began to remove the Democratic state governor, Gray Davis, who had gained reelection only in 2002 and still had three years of his term of office to run. In the ensuing election the recall was approved and Davis was replaced by Republican Arnold Schwarzenegger. The problem of these participatory procedures, particularly the initiative, is that they become yet another instrument for the manipulation of the policy process by pressure groups.

This diversity and complexity results in California’s particular brand of politics, the politics of ‘hyperpluralism’; the idea of majority rule in the state has given way to rule by minorities. To quote David Lawrence again, ‘The exercise of political power has become a highly competitive tug-of-war between institutions, policymakers, political parties, numerous interest groups, and voters.’ Because the role of the American states is still so important and the decisions that they take make a considerable impact on the lives of their citizens, this tortuous policy process has a real impact on the quality of government in California. The problem of coordinating the activities of government becomes almost insoluble; state finances are in a continuing crisis; what are intended as instruments of majoritarian democracy instead give to relatively small groups the opportunity to exercise undue influence to achieve their own self-interested ends. Thus vitally important policy areas, such as the supply of water and the generation of electricity, which are essentially matters for the state government, are not being adequately addressed.

The nature of politics within the state has considerable consequences also for the government of the United States. California has fifty-three members in the House of Representatives, 12 per cent of the total membership, as well as its two Senators. A number of the chairmen of congressional standing committees come from California. As the largest state in the Union, with fifty-five members in the Electoral College, its potential impact on the outcome of presidential elections is enormous. The disorganised politics of the state feed into the behaviour of its delegations in Congress and into the conduct of presidential elections.

We have tried to isolate four main strands of political behaviour in the electorate, and we shall see in later chapters how these strands continue right through the political system. Of course the patterns of political behaviour are more complicated than this, for in reality sectional, class, pluralistic and individualistic influences overlap and interact. Each election, each political situation, becomes a unique combination of these elements, and we are faced with an ever-changing panorama of political life, responding to changing social, economic and strategic forces, internal and external. However, though in constant change, and although the patterns never repeat themselves exactly, there are certain restraints and certain persistent structures that give continuity and shape to the political system. Constitutional forces, the electoral system, the party system: these provide the framework within which more transient political forces work themselves out, and it is to these that we must now turn our attention.