This is an attempt to operationalize and test the basic proposition of Thomas Hobbes's Leviathan: namely, that Leviathan (or an all-powerful government) makes the life of man less solitary, poor, nasty, brutish and short.

In his celebrated study of American democracy written in 1888, Lord Bryce reserved his most condemnatory reflections for city government and in a muchquoted passage asserted: ‘There is no denying that the government of cities is the one conspicuous failure of the United States. The deficiencies of the National government tell but little for evil on the welfare of the people. The faults of the State governments are insignificant compared with the extravagance, corruption and mismanagement which mark the administration of most of the great cities'sangeetha.

The overall conclusion to be drawn from the discussion contained in Part I of this article is that, even in respect of what are usually regarded in the literature as being some of its more distinctively democratic characteristics, the American system of urban government is lacking. It may even be in these respects less democratic than the British. Whether it is agreed that the attributes discussed in Part I are those of a democracy or not, by its own lights American city government is deficient.

Can there really be much doubt who governs our complex modern societies? Public bureaucracies, staffed largely by permanent civil servants, are responsible for the vast majority of policy initiatives taken by governments. Discretion, not merely for deciding individual cases, but for crafting the content of most legislation has passed from the legislature to the executive. Bureaucrats, monopolizing as they do much of the available information about the shortcomings of existing policies, as well as much of the technical expertise necessary to design practical alternatives, have gained a predominant influence over the evolution of the agenda for decision. Elected executives everywhere are outnumbered and outlasted by career civil servants. 1 In a literal sense, the modern political system is essentially ‘bureaucratic’ – characterized by ‘the rule of officials’.

III THE PATTERN EXPLAINEDIn part I of this paper we described the gross pattern of public policy in our five countries. In part II we looked at how the pattern developed in each of the countries. We noticed that the countries have pursued policies that diverge widely, at least with respect to the size of the direct operating role of the State in the provision of public services. We also noticed that the United States differs from the four other countries far more than they do from each other. These findings will not have come as a great surprise to anybody, although some readers may have been surprised – in view of the common assumption that all major western countries are ‘welfare states’ – to discover just how much the countries differ and what different histories they have had.

Systems theorists introduced the concept of ‘support’ to permit explanations of political stability and instability. Yet most attempts to verify the existence of a relationship between support and stability empirically have dealt with wellestablished political systems, and have relied on data collected at one point in time. This paper reports an initial effort to examine the growth of support for a new political regime using a series of sample surveys providing data on changes in the level of support over time.

In their study of the ‘affluent worker’ in Luton,1 Goldthorpe and his colleagues reached a number of important conclusions about the political behaviour of the ‘traditional’ and ‘new’ working class in post-war Britain. They rejected the belief, commonly held by the late 1950s, that a growing proportion of manual workers was beginning to support the Conservative Party as a result of attaining a middle class level of income and material possessions (the ‘embourgeoisement’ theory). In their sample, which was elaborately designed to ensure the most favourable conditions for confirmation of the embourgeoisement theory, they found (i) that the level of stable Labour support was higher than the national average for manual workers; (ii) that there was no evidence of any gradual, long-term shift of support towards the Conservatives or away from Labour; and (iii) that the small minority of Conservatives was distinguished not by a higher than average standard of living, but by a relatively large number of white collar workers among their kin. The notion that there was a necessary connection, among manual workers, between growing material prosperity and increased support for the Conservative Party was therefore decisively rejected.

This paper is about the things governments do and why they do them. It is written in the belief that, while we know quite a lot about decision-making processes in individual countries, we do not know nearly enough about why the governments of different countries make different decisions and pursue different policies. The countries of North America and western Europe are often described as ‘welfare states’, the implication being that the governments of all of them do broadly similar things in broadly similar ways. As we shall see, however, these broad similarities conceal important, wide divergences. These divergences deserve to be explained.

Several theorems have been established recently on the existence or location of equilibrium policies in collective decision-making in the case when the alternatives are points in a multi-dimensional space. One of these theorems is founded on an assumption about individual preferences which seems intuitively plausible in the context of political decision-making. We have conducted a number of experiments designed to test this theorem directly. In addition, the resulting data were used to compare the accuracy of this assumption about individual preferences with that of certain alternative assumptions, some of which have also been used in theories of collective decision-making. Before the experiments and the analyses of the results are described, we first briefly present the theorem and the various alternative assumptions to be tested.

In the ten years or so since the publication of Hyman's seminal work, 1 students of politics have given increasing attention to political socialization. There has been a proliferation of works utilizing the concept. 2 The notion of socialization has perhaps been most rigorously applied in studies of the development of the political attitudes of children and adolescents, 3 but it has also been employed, if rather more loosely, in the study of adult electors. 4 There is, however, a dearth of material relating to the socialization of party activists at local level. This is not to say that the social and political background and the recruitment patterns of party activists have not been investigated, 5 but the concept of political socialization has not been explicitly or very rigorously employed. The importance of local party organizations and their memberships has not always been self-evident to students of politics and we do not propose to argue the point here. We merely assert that British parties and British politics derive much of their ‘style’ from the character of party activists. This being the case, we feel that the relative lack of information about the socialization of activists represents a considerable gap.

All the theories of coalition formation known to us incorporate an assumption (or a set of assumptions which imply) that coalitions which are formed must at least be ‘winning’. 1 In the present context, a coalition is said to be winning if the sum of the seats held by its members is at least a simple majority of all the seats in the parliament. Most of these theories were intended to apply to government coalitions: indeed, several of them were specifically designed for this situation. Yet out of the total of 207 governments which have formed in twelve western democracies since the war, 2 seventy-four of them have not been winning.

Comparativists interested in the political dynamics of policy development, and students of public policy desirous of studying national programs in crossnational contexts, face similar dilemmas at present. There is a literature focusing on decision-making processes which often pays little regard to policy substance. There are statistical analyses of expenditure levels, but these are difficult to relate to the ample literature bearing on how particular substantive problems have been handled in specific national and local situations. As a result, political scientists interested in the cross-national study of broad policy areas - such as those studying social policy development in Western Europe and America - are now engaged in the creation of historically based frameworks which will permit more fruitful analyses of differences in the patterns of development.

In 1919, three years after the Easter Rising in Dublin, a revolutionary government was established in Ireland. This government, the Dail Eireann, was the forerunner of the Irish Free State that ultimately resulted from the Anglo-Irish Treaty of 1922. Slightly more than fifty years have elapsed since the union of Ireland and Great Britain was formally dissolved and, in that period of time, many changes have occurred in both the social and political systems of Ireland. Not the least of these changes has been the passing of one elite generation from the political scene and the emergence of another.

In electoral research the examination of the effect on voting behaviour of personal interaction has frequently had a subsidiary place in investigation and an ambiguous status in theory – a situation that has arisen despite evidence suggesting that personal interaction may often be a central factor in structuring voting choice. Typical of this neglect has been the approach to the phenomenon to be discussed here – the effect of neighbourhood structure on voting; for although this was first described at an early stage in voting research and subsequently re-examined several times, it has still to be adequately explained and integrated with other research findings.

Axelrod has demonstrated the equivalence of a measure of structural propensity to conflict, based on game-theory and applicable to certain types of social interaction, and spatial conceptions of policy agreement. However his spatial model rests on assumptions of interval status not commonly met by data from which consensual interpretations have evolved and against which they have been tested. The generality and rigour of consensual explanations of democratic stability can be enhanced by integration with Axelrod's framework. At the same time consensus hypotheses focus on a question left unanswered by game-theoretic formulations: why do democratic political games continue to be played by existing rules? Following Axelrod's recommendation to integrate conflict of interest with other bodies of theory, 1 this paper assesses points of contact between Axelrod's discussion and one line of consensual reasoning, extends the equivalence between conflict of interest and policy agreement to all levels of measurement, and details the contributions which game-theoretic and consensual formulations can make to each other.

One of the complaints most frequently raised against modern empirical political science is that it overemphasizes political inputs (such as voting behavior) and pays insufficient attention to political outputs (such as allocation of resources). In response to this complaint, a number of political scientists in recent years have focused their attention on the question of whether variations in political inputs are much related to variations in outputs. A considerable literature has grown up around this question, particularly focusing on relationships between variations in electoral competitiveness and variations in public spending patterns across a range of different political systems. Because they are different yet comparable, American states have been the setting for much of this research. In addition to the state policy outputs literature, investigations have been undertaken of variations in outputs among differing jurisdictions within a metropolitan area and between variations in outputs and differing types of city organizational structure. The focus of the research reported on here is somewhat different. Rather than looking at variations between political systems, we focus here on variations within political systems. Without dealing with all of the literature alluded to above, and the myriad criticisms that each attempt has spawned, it seems fair to say that the general conclusion of most researchers has been that political characteristics account for little of the variation in output between political systems, and that variation in levels of wealth is usually a far more powerful predictor. Such a conclusion, however, does not relate to the possible impact of politics on allocational decisions within a single jurisdiction. While the City

In spite of all the research stimulated by the English and Scottish Royal Commissions on Local Government, we still have no clear idea of why people go into local politics, or of what sorts of people become local councillors. A recent study of Glasgow politics threw some light on these questions.1 One of the things it did was to draw attention to the central importance of the political party. Both Royal Commissions ignored the part played by parties, despite the fact that in virtually every large town in Britain the council is dominated by parties.2 One of the major functions the parties perform is the recruitment of candidates and councillors. Indeed, in connection with recruitment the parties are almost the only active agencies.

Mathematical languages and computer algorithms are becoming important modes of analysis for theory evaluation in political science. Typically, the process involves (1) translating the major theoretical relationships into a ‘formal’ language for the logical analysis of internal consistency, or, (2) empirically interpreting the formal language in order to make specific predictions, which, in turn, allow evaluation of external consistency with theoretically significant real world phenomena. In this paper I wish to discuss, first, some basic concepts and, second, some aspects of the technology and methodology of these modes of theory evaluation.

I wish to consider what kind of force attaches to the fact that a certain policy or course of action is in the public interest. How, for example, does it compare with the consideration that a policy is in the interests of a majority, or the consideration that a policy is a fair one? Such questions about the weight to be attached to the fact that a policy or course of action is in the public interest arise naturally if we accept the view, put forward by Barry, that just as to say that a policy is in the interests of farmers, for example, is to say that it is in the interest of each farmer as a farmer or qua farmer, so to say that a policy is in the public interest is to say that it is in the interest of each member of the public as a member of the public or qua member of the public.1