MAY I BEGIN WITH A DASH OF HISTORICAL PERSPECTIVE? I DO SO AS A health warning, to myself as much as to anyone else. Just as the great Lord Salisbury once said that too much poring over maps tended to drive nineteenth-century ministers and military chiefs mad, an excessive concentration on prime-ministerial styles can propel scholars and commentators smack bang into the Tommy Cooper school of analysis – here you have collective cabinet government then, just like that, when the premiership changes hands, you switch to overbearing prime-ministerial government. There has been a good deal of this since Tony Blair entered No. 10 exactly 214 days ago.

IN HER SUBMISSION TO THE OPSAHL HEARINGS ON THE NORTHERN Ireland problem in 1993, the literary and cultural critic Edna Longley made a simple point about the ‘Anglo-Irish Agreement’. This term for what was a pact between the UK and Republic of Ireland governments is, she argued, a misnomer: ‘[It] obscures the contested area, and panders to the belief – in both London and Dublin – that the UK is coterminous with England.’ Longley later urged that the Northern Ireland problem should be viewed more widely, in its proper context as part of the ‘melting pot’ of cultures of the two islands, Britain and Ireland.

‘POLICY DETERMINES POLITICS’ IS A FAMILIAR APHORISM AND ITS durability has been tested time and again. Lowi's initial typology showed the normal trinitarian design (distinguishing regulative, distributive and redistributive policy). Later a fourth type was added constituent policy. Lowi's original three elements are, however, not on the same logical level. If we distinguish between the regulative and distributive levels of decision-making we end up with a six-fold typology (Matrix 1).

A SIGNIFICANT DEVELOPMENT IN RECENT SOUTH AMERICAN POLITICS has been the re-emergence of populism. More interesting still has been the unexpected combination – in some countries – of traditional populist appeals (successfully made), the determined application of free market policies, and the successful re-election of the market-reforming populists. This does not mean that populist politicians can succeed to order. Many populist candidates have sought election and only a few have secured it. Of those elected as (essentially) independent candidates, only some have succeeded. Presidents Bucaram in Ecuador and Collor in Brazil were removed from office by Congress and subjected to legal proceedings for corruption. However, where populists have succeeded, they have done so on a far more impressive scale than most people originally predicted.

A BASIC QUESTION IS HOW POLITICS ARE TAMED AND CEASE BEING A deadly, warlike affair. The most dramatic way is through sudden, deliberate and lasting compromises of core disputes among political elites – what we think of as ‘elite settlements’. Prior to settlements elites disagree about government institutions, engage in unchecked fights for dominance, and view politics as winner-take-all. After settlements, elite persons and groups continue to be affiliated with conflicting parties, movements, and beliefs, but they share a consensus about government institutions and the codes and rules of political competition. Settlements tame politics by generating tacitly accommodative and overtly restrained practices among competing political elites.