Although most incumbent members of Congress succeed in being re-elected, American politicians are in fact more subject to the vicissitudes of electoral politics than are the politicians of any other major democracy. They face re-election more often. They face the possibility of having to contest primaries. They have to raise most of their own money. They are not given a large measure of protective cover by strong political parties. The consequences for American government and politics are profound. They include such familiar and well-documented phenomena as particularism, ‘pork’ and the power of special interests; but they also include the high incidence in America of purely symbolic politics, the drastic foreshortening of American politicians' time horizons and the difficulty that the American system often has in taking tough decisions.

Theories ascribing democracy to inclusive contestation imply that the speech of authoritarianrulers will differ from the speech of electoral politicians. Authoritarian rulers will use an officiallanguage that insulates them from populations under their control. This official language insultssubject populations by implying that the vernacular in which they formulate their thoughts isinadequate for the discussion of political ideas. Electoral politicians, praising the competence ofcitizens to decide political questions, take care to frame their ideas in ordinary language, asotherwise the politicians would contradict their message that bearers of ordinary language arepolitically competent. If rulers implicitly insult and politicians flatter, citizens should respondwith disaffiliation to rulers' language and affiliation to the language of politicians, at least of thosepoliticians whom the citizens favour. This hypothesis is tested using an experimental paradigmin three Russian cities at the end of 1993. The prediction that Russians will affiliate to the textsof some electoral politicians contrasts with the claim that Russians evaluate political ideas mainlyby contemplating change in their standard of living, since Russia represents an unusual case inwhich authoritarian speech is associated with greater affluence, while electoral speech isassociated with increasing impoverishment for most people.

The Northern Irish party system is plausibly represented as two systems, in which partycompetition occurs within nationalist and unionist blocs. Social and ideological divisions withinthese blocs constrain parties' electoral strategies and thus facilitate or inhibit cross-communalcompromise. Using Northern Irish Social Attitudes data, various accounts of these intra-communal divisions are tested and their political implications assessed. Contrary to expectationsconcerning the effects of cross-communal contact, neighbourhood integration is found to haveno relationship with partisanship although, consistent with political socialization theory, youngpeople on both sides of the divide are more likely than those who are older to support the morerecently introduced parties. Most significantly, however, there are noticeable asymmetries in thepatterns of cleavage within the unionist and nationalist blocs. Among Protestants, left–rightideology has a far stronger impact than constitutional position on patterns of partisanship; andsocial class has considerably stronger effects than does denomination. Most Protestants, whatevertheir partisanship, also express strongly unionist constitutional preferences. In clear contrast, onthe nationalist side party support is polarized along constitutional lines, there is no cross-cuttingideological division over economic inequality, and a majority of Catholics adopt a moderatestance on nationalism. It is argued in consequence, that within the unionist bloc the pattern ofintra-communal party competition militates against constitutional compromise as a solution to‘the troubles’, whereas among nationalists the unidimensional structure of competition forelectoral support and the distribution of attitudes towards the constitutional issue are likely to haveinfluenced the adoption of compromise strategies by Sinn Féin.

If the automobile business had developed like the computer business, a Rolls Royce would cost $2.75 and run for 3 million miles on one gallon of gas.

Forrester, Hight-Tech SocietyTom Forrester, High-Tech Society (Oxford: Basil Blackwell, 1987), p. 18.

The computer power available to political scientists has increased at least one thousandfold since 1970. We can now make millions of calculations in a few seconds, generate beautiful and meaningful graphics, and communicate instantly around the world, all at virtually no cost. But in what ways has this computer power helped us make more accurate statements about political phenomena? Beyond simply doing faster and more cheaply that which we had done previously, this computer revolution has made possible a qualitative leap in our analysis of political behaviour.

We are flattered that Ken Newton considers our research to be the only serious attempt to test Tiebout in Britain, but we would like to correct the errors of his fairy tale as well as refute his criticisms.Kenneth Newton, ‘Residential Mobility in London: Rational Choice Fairy Tale, Utopia, or Reality?’ British Journal of Political Science, this issue, pp. 148–51; Peter John, Keith Dowding and Stephen Biggs, ‘Residential Mobility in London: A Micro-Level Test of the Behavioural Foundations of the Tiebout Model’, British Journal of Poltical Science, 25 (1995), 379–97.

Margaret Thatcher's dominance of British politics during the 1980s is well known. A self-proclaimed ‘conviction politician’, she was likely to seize any opportunity to shape the public's values. Surprisingly, however, her efforts seem to have been ineffective. Ivor Crewe has labelled Thatcher's attempt to modify the public's beliefs ‘the crusade that failed’. He concluded that ‘much of Thatcherism will die with Thatcher. Its permanent legacy at the level of the mass public will be very limited.’ So unexpected is this finding that it invites further investigation. Did Crewe find little evidence of Thatcher's impact simply because of the generality of his analysis? Might a more sharply focused study, one dealing with the delimited area of public policy, reach a different conclusion?Ivor Crewe, ‘Values: The Crusade that Failed’, in Dennis Kavanagh and Anthony Seldon, eds, The Thatcher Effect: A Decade of Change (Oxford: Oxford University Press, 1989), p. 250. See also Ivor Crewe, ‘The Thatcher Legacy’, in Anthony King et al., Britain at the Polls 1992 (Chatham, NJ: Chatham House Publishers, 1993), pp. 1–28, especially pp. 8–11.

Once upon a time, and a good time it was too, there was a faraway country ruled by a wise economist-king called Tiebout who realized that the way to happiness was to organize local public services in market-analogous manner. He did this by dividing local government in Tieboutiana into many competing units, each with its own package of public goods and taxes. In this way citizens (in Tieboutiana they were called consumer-voters) could move between local authorities in search of the public services they wanted at tax rates they thought reasonable.

A basic question in the analysis of the economic voting of the individual is whether the decision to vote for the government is sociotropic or determined by the pocketbook? It is sociotropic if the voter considers the whole of the economy. It is pocketbook voting if people look at their personal economic circumstances. The question has remained at the centre of research in economic voting since it was first raised by Kinder and Kiewiet in 1979. Over the years, it has generated a sizeable number of empirical studies, mostly based on American data, as well as a fair share of disagreement. From a theoretical viewpoint it is not obvious whether economic man is a sociotropic or a pocketbook voter. There is a strong argument for both assumptions: (α) the pro-pocketbook argument suggests that homo economicus pursues his own interest or that of his household, and votes with his pocketbook; (β) the proto-sociotropic argument suggests that voters consider what governments try to accomplish. They do not try to steer the economy of any individual, but the economy of the country.The word sociotropic literally means society-directed. The corresponding concept for self-directed would be egotropic, as we write in other papers. The Editor has resolved that this is not in the dictionary. Moreover, several writers prefer the term egocentric. This is too loaded with other meanings for our taste. So we shall use the term pocketbook voting in the present Note.Donald R. Kinder and D. Roderick Kiewiet, ‘Economic Discontent and Political Behavior: The Role of Personal Grievances and Collective Economic Judgements in Congressional Voting’, American Journal of Political Science, 23 (1979), 495–527, and ‘Sociotropic Politics’, British Journal of Political Science, 11 (1981), 129–61. See further the cross-national study by Michael S. Lewis-Beck, Economics and Elections (Ann Arbor: University of Michigan Press, 1988, 1990), and the surveys by Peter Nannestad and Martin Paldam, ‘The VP-Function. A Survey of the Literature on Vote and Popularity Functions after 25 Years’, Public Choice, 79 (1994), 213–45; and Douglas A. Hibbs, Solidarity or Egoism? (Aarhus: Aarhus University Press, 1993). These recent surveys allow us to restrict these references to a minimum.

Who gets what, when and how is affected by the power of different groups in society to affect the political process. Many feminists see increasing women's political representation, via elected office or interest-group involvement, as a necessary condition for improving the status of women. Many studies have examined whether the numerical representation of women translates into legislation or legislative behaviour that seeks to improve the status of women. Although we have evidence that women's legislative representation matters, the passage of legislation is only half of the story in public policy making. After policies are enacted, the bureaucracy must implement them and little research has examined the impact of women's political representation on policy outputs. The question of women's influence on bureaucracies is especially important, because bureaucracies deliver services that have a direct bearing on the status of women. It is an important issue for representative democracy whether women have an influence on the bureaucracies that affect them.Beth Reingold, ‘Concepts of Representation among Female and Male State Legislators’, Legislative Studies Quarterly, 17 (1992), 509–37; Michelle A. Saint-Germaine, ‘Does Their Difference Make a Difference? The Impact of Women on Public Policy in the Arizona Legislature’, Social Science Quarterly, 70 (1989), 956–68; Sue Thomas, ‘The Impact of Gender on Activities and Priorities of State Legislatures,’ Western Political Quarterly, 44 (1990), 445–56.