Abstract
This article asks, ‘how do practitioners understand the relationship between the prime minister, ministers and the rest of Westminster and Whitehall?’ We focus on three topics. First, we review tales of a Blair presidency. Second, we explore the governance paradox in which people tell tales of a Blair presidency as they recount stories of British governance that portray it as fragmented with several decision-makers. Finally, we argue that this paradox reveals the distorting influence the Westminster model still exerts on many accounts of British politics. It acts as a smokescreen for the changes in executive politics.
This article focuses on the debate about the ‘Blair presidency’ between 1997 and 2005. We ask the deceptively simple question, ‘how do practitioners understand the relationship between the prime minister, ministers and the rest of Westminster and Whitehall?’ We focus on three topics. Firstly, we review tales of a Blair presidency. Our main concern is not how our academic colleagues understand the relationship. 1 We draw on the work of academic colleagues when they cite interviews, but we concentrate on the views of practitioners. We rely on the obvious sources of prime ministerial, ministerial and civil servant autobiographies, diaries and memoirs as well as official publications. 2 As Anthony Mughan (2000, p. 134) remarks, ‘for every “insider” … assertion that prime ministerial government has arrived in Britain, it is possible to find the counter-assertion that cabinet government remains the order of the day’. Such inconsistencies are the puzzle. We compare the several stories and show there is much inconsistency and contradiction. This divergent evidence highlights the paradox between presidential claims and the governance narrative.
So, secondly, we explore the governance paradox – even as people tell tales of a Blair presidency, they recount also stories of British governance that portray it as fragmented and multipolar. In particular, we argue that New Labour appears to accept key tenets of the governance narrative. Innovations like joining-up, the reforms at Number 10 and the switch to improved service delivery recognise the weakness of the centre and fuel claims of the Blair presidency. But New Labour ignores the other half of the governance narrative that stresses interdependence and cooperation, not command and control. So, claims of a Blair presidency founder on policy-making and implementation deficits.
Finally, we argue that this paradox reveals the distorting influence the Westminster Model still exerts on many accounts of British politics. It acts as a smokescreen for the changes in executive politics. We conclude that Blair is locked into complex patterns of domestic and international dependence with the prime minister winning, losing and drawing as one might expect given the volatile nature of high politics.
Presidential Tales
Journalists have repeatedly described Tony Blair as presidential from the moment of his election as prime minister. In Britain, The Independent ran a news item by Anthony Bevins under the headline ‘Blair Goes Presidential’ on 6 May 1997 (see also Rawnsley, 2001, pp. 292–4, p. 379). Political scientists too argue that Blair has manipulated his personal resources and expanded his institutional power to achieve a degree of predominance unmatched in British history. 3 For our purposes, the key point is that such views are shared by insiders. At the start, Jonathan Powell (Number 10 chief of staff) had famously warned senior civil servants to expect ‘a change from a feudal system of barons to a more Napoleonic system’ (Daily Telegraph, 8 December 2001, cited in Seldon, 2004, p. 437). Blair's Number 10 aides assert, ‘Cabinet died years ago’, claiming ‘we want to replace the Department barons with a Bonapartist system’ (quoted in Kavanagh and Seldon, 2000, p. 291).
Blair's ministerial critics do not demur. Mo Mowlam (2002, p. 356, p. 366), former Secretary of State for Northern Ireland, claimed ‘more and more decisions were being taken at No. 10 without consultation with the relevant Minister or Secretary of State’. She criticises ‘the centralising tendency and arrogance of No. 10’, especially that ‘their lack of inclusiveness of the cabinet, MPs, party members and the unions leads to bad decisions. Try as I might, I got no indication that their views or behaviour would change’. Similarly, Clare Short (2004, p. 272, p. 278), former Minister for International Development, talks of ‘the concentration of power in No. 10’, criticising Blair's ‘informal decision making style’ with ‘his personal entourage of advisers’ because it ‘enhances the personal power of the Prime Minister and reduces the quality of decision-making’. However, ‘President Blair’ asserts that such claims have been made ‘about virtually every administration in history that had a sense of direction … Of course you have to have Cabinet Government (The Observer, 23 November 1997; see also the citations in Hennessy, 2000c, p. 11, and n. 70).
So, we assess the three main claims made to support the contention that Blair has transformed his role as prime minister into that of a president; namely, there has been a centralisation of coordination, a pluralisation of advice and the personalisation of party leadership and elections. 4
Centralisation
Structural changes at Number 10 and the Cabinet Office are the way in which Blair has strengthened the centre of government. The Policy Unit mutated into the Policy Directorate when it merged with the Prime Minister's Private Office. From day one Blair surrounded himself with a network of special advisers. Their numbers rose from eight under John Major to twenty-seven under Tony Blair (Blick, 2004, appendix; and on the growth of advisers see next section). Total staff employed at Number 10 rose from 71 in 1970 under Heath, to 107 under Major and to over 200 under Blair (Kavanagh and Seldon, 2000, p. 306), creating ‘the department that-will-not-speak-its-name’ (Hennessy, 2002c, p. 6). Initially the focus was on improving communications, with Alastair Campbell heading the Strategic Communications Unit (SCU). Latterly the emphasis fell on policy advice. The Cabinet Office was reformed to improve central coordination. Several new units were created: for example, initially, the Social Exclusion Unit and the Performance and Innovation Unit, latterly the Strategy Unit, the Office of Public Services Reform and the Delivery Unit. As Peter Hennessy (1998, p. 15) observes, ‘Number 10 is omnipresent’. The Cabinet Office has always been a ragbag of functions bequeathed by former prime ministers. Now it groans under its own multiplying units posing the question of ‘who will coordinate the would-be coordinator?’. Blair seeks to control government functions without bothering himself with too many operational details.
In presidential tales, the prime minister's department in all but name allows Blair to remain on top of several projects if not in detailed touch. It checks the problem of prime ministerial overload. As Anthony Seldon (2004, p. 630) observes, ‘however distracted Blair might be by other events, domestic and international, the work of monitoring … went on regardless (“The [Delivery] Unit never sleeps”, Blair was told)’ (see also Hennessy (2000a, p. 390).
Pluralisation
In the Westminster model, the civil service has a monopoly of advice and this advice is collated and coordinated by the Cabinet through its ministerial and official committees and the Cabinet Office. This neat and tidy picture has given way to one of competing centres of advice and coordination for which, allegedly, Blair is the only nodal point. The Cabinet Office now serves the prime minister rather than the Cabinet collectively. Blair cut back on collegial decision-making, ‘reducing most meetings of the Cabinet to just forty minutes of approving decisions already taken elsewhere, parish notices and short speeches either delivered by the Prime Minister or vetted by him in advance’ (Rentoul, 2001, p. 540; see also Hennessy, 1998, p. 11; Kavanagh and Seldon, 2000, p. 278; Rawnsley, 2001, p. 33; Seldon, 2004, p. 437). Seemingly it is a commonplace that Blair rarely chairs cabinet committees. There are fewer committees, meeting less often and not always reporting to full Cabinet. Most decisions take place in ‘bilaterals’ – agreements struck in ad hoc meetings between Blair and ministers directly – a style favoured by both the prime minister and the chancellor (Rawnsley, 2001, p. 53). In his first three years of office, Blair held 783 meetings with individual ministers compared with John Major's 272 for the same period (Kavanagh and Seldon, 2000, p. 279). As Blair said, ‘I think most Prime Ministers who have got a strong programme end up expecting their Secretaries of State to put it through; and you've always got a pretty direct personal relationship’. Also, he would not expect ministers to raise matters in Cabinet: ‘look I would be pretty shocked if the first time I knew a Cabinet Minister felt strongly about something was if they raised it at the cabinet table’ – ‘I would expect them to come and knock on my door’ (cited in Hennessy, 2000c, p. 12).
The list of decisions never even reported to Cabinet includes: independence for the Bank of England; postponement of joining the euro; cuts in lone-parent benefit; and the future of hereditary peers (Rentoul, 2001, p. 540). Robin Butler, former Cabinet Secretary and Head of the Home Civil Service, reports that ‘during the late 1940s, cabinet met for an average of 87 times a year, with 340 papers being circulated; in the 1970s, 60 times a year, with 140 papers; and by the late 1990s, no more than 40 times a year, with only 20 papers’ (cited in Hennessy, 2000b, p. 5). Both the frequency and content of Cabinet meetings are said to have diminished significantly under Blair. Bilateral agreements have replaced collective government, and Blair is the only person able to oversee the work of the government.
Blair is supported in this role by the new machinery of the centre and by sources of advice other than the civil service. The decline of the civil service monopoly of information and rise of more varied sources of advice has deep roots. Thatcher accelerated these trends. Blair took them further. Derek Scott (2004, p. 14, p. 17, p. 206) was Blair's economics adviser at Number 10. He argues that Blair paid less attention to his policy advisers and civil servants than to ‘the occasional outsider or those members of his inner circle who had little grasp or real interest in policy’ (Scott, 2004, p. 16). Moreover, Blair's circle was not the only, or even the most important, source of advice on social and economic policy. Gordon Brown had his own coterie, and his pre-eminent consigliore was Ed Balls, Chief Economic Adviser to the Treasury and a key Brown supporter. So, pluralisation of advice also meant competing centres of advice and the competition between Blair and Brown's teams was intense.
Personalisation
Although Blair sought to institutionalise central capacity, nonetheless, like his predecessors, personalism characterises his ‘presidency’. The professional management of media relations and the use of spin doctors are harnessed to two bigger purposes – continuous electioneering and personalising the campaign, policy issues and indeed the government, by an almost exclusive focus on Tony Blair (see Seymour-Ure, 2003). Andrew Rawnsley (2001, p. 488) amusingly illustrates the point with his anecdote about the Labour election manifesto and its seven pictures of Blair. When journalists queried this exclusive focus, ‘Brown's features were a study in granite … [and] … the Deputy Prime Minister [John Prescott], wearing what his mother called his “ugly face”, looked like a man one provocation away from a detonation’.
Blair did not invent media management as a way of sustaining the pre-eminence of the prime minister. However, his ‘public communications, from the designer leisure wear to the designer accent and the designer press conferences probably attracted more public interest than those of any previous British government’ (Seymour-Ure, 2003, p. 7). Managing the media, or ‘spin’, is a game of chance and Blair's gambler-in-chief, his ‘spin doctor’ managing the media, was Alastair Campbell, Director of Communications and Strategy (see Oborne and Walters, 2004). The key organisation was the Strategic Communications Unit, created in 1997. Its job was to monitor the news and provide a rapid response, expounding the government's position and, where necessary, rebutting any criticisms of government policy. Campbell was the prime minister's voice. His job was to ensure that the prime minister's voice was also that of the government. He was the spin doctor who used his daily lobby briefings to control government links with the media. Also, this prime ministerial centre extended its role to commanding the press relations of all ministers. Early in 1997 he even ‘informed all departmental press chiefs that media bids for interviews with their ministers must be cleared first with him’ (The Independent, 6 May 1997). In this way, Blair allegedly got an advanced news management service akin to that of an American president (see also Scott, 2004, pp. 15–8; Seldon, 2004, ch. 22). Managing the media was also a central element in policy formulation. The strategy is called ‘triangulation’. It involves packaging policies so they conflict with the left wing of the Labour party, thus winning support from the right-wing press.
Blair's premiership is also said to have been marked by a significant increase in the personalisation of power. Present-day media create an environment in which a politician's ability to attract publicity is crucial to electoral success. Indeed, Blair's office helps to create this environment by personalising policy initiatives. For example, when Blair spoke of a rise in the rates of cancer, he publicly mentioned the death of both his own mother to throat cancer and his wife's aunt to breast cancer. Blair personalised policies with this public mix of sincerity and personal experience. As Seldon (2004, pp. 432–6) documents, whenever Blair thought he was not getting the results he wanted, he took personal charge. He identified himself personally with policy initiatives in, for example, crime, education, health, immigration and transport. In the pungent phrase of the (then) leader of the opposition, Michael Howard, when he takes charge he has ‘more summits than the Himalayas’.
Governance Stories
Even as journalists, political scientists and practitioners tell tales of a Blair presidency, so they continue to recognise many limitations to Blair's ability to get his own way. Rawnsley (2001, pp. 292–4) initially subscribed to ‘the command and control’ view of Blair. But, by June 2003 he wrote of ‘a prime minister who is not looking in the least bit presidential’ at the head of ‘a government displaying signs of drift’ (The Observer, 15 June 2003). In similar vein, Peter Riddell (2001, p. 40) commented, ‘If Mr Blair has been a Napoleonic figure, he has been a frustrated rather than a commanding one’. So, there is a second story that focuses on the problems of governance and sees Blair as perpetually involved in negotiations and diplomacy with a host of other politicians, officials and citizens.
The governance narrative provides an alternative to the Westminster model. This narrative highlights the place in British politics of networks, the informal authority of which supplements and supplants the formal authority of government. It stresses the horizontal and vertical networks of interdependence in which the core executive is embedded. Britain is seen as a differentiated polity characterised by a hollowed-out state, a core executive fumbling to pull rubber levers of control, and a massive growth of networks. The commonplace version of the governance narrative emphasises a shift in British government from government of a unitary state to governance in and by networks, often understood by practitioners as a shift from hierarchy to markets to networks. 5
In this section, we use the widely-held distinction between the horizontal networks of Westminster and Whitehall and the networks beyond Westminster and Whitehall (as found, for example, in Cm 4310 (1999) and its discussions of joined-up government). As the story of the rival courts of Brown and Blair demonstrates, the core executive can be seen, not as a single decision centre focused on Blair, but as a set of overlapping networks. As the story of government policy-making shows, central intentions are all too often confounded by central fragmentation and the Blair reforms of the centre seek to impose the desired degree of coordination. Add the simple fact that service delivery is disaggregated to a multiplicity of networks and the explanation of the gap between rhetoric and reality is obvious. The implementation gap is ubiquitous. Unintended consequences are inevitable. So, Blair is just one actor among many interdependent ones in the networks that criss-cross Whitehall, Westminster and beyond. So, now we tell the story of the Blair government from the standpoint of Whitehall governance and governance beyond Whitehall.
Whitehall Governance: Blair and Brown
Even political scientists who support the notion of a Blair presidency typically mention the Treasury, under Gordon Brown as Chancellor of the Exchequer, as ‘a great crag standing in the way of a thoroughly monocratic government’ (Hennessy, 2002, p. 21). Brown and the Treasury have come to influence an ever-growing range of activities. In particular, Brown implemented a new system of Public Service Agreements (PSAs) that define and direct the activities of government departments by setting agreed targets and then monitoring them. This control of public expenditure shows Brown's reach throughout government. Blair helped to increase the scope of Brown's authority by appointing him to chair the main economic committee of the Cabinet – a post historically occupied by the prime minister.
Recognition of Brown's authority requires us to shift from tales of a Blair presidency to stories of at least a dual monarchy: ‘Brown conceived of the new government as a dual monarchy, each with its own court’ (Rawnsley, 2001, p. 20). This notion has its roots in the ‘infamous’ Granita restaurant story – a meeting between Blair and Brown in Islington on 31 May 1994 (but see Peston, 2005, p. 57, p. 58, p. 66). There is much disagreement about, and little documentary evidence on, the degree of control ceded to Brown, ‘But there is no doubt that substantial if imprecise control was granted to Brown’ (Seldon, 2004, pp. 193–4) over economic and social policy. James Naughtie (2002, p. 71) believes command over economic policy and ‘significant chunks’ of social policy were conceded (as do Keegan, 2003, p. 124; Peston, 2005, p. 58; Rawnsley, 2001, p. 20, p. 11). While there is no documentary evidence to support a deal on handing over the prime ministership to Brown, there is some evidence on the policy deal (The Guardian, 6 June 2003). Michael White, political editor of The Guardian, concludes that ‘Blair had effectively ceded sovereignty to Brown in the economics sphere’ (cited in Seldon, 2004, p. 669; see also Peston, 2005, p. 67).
There have been several occasions on which Blair has found his authority checked by Brown. Such checks have occurred most often and dramatically over Blair's European ambitions and the budget. For example, Brown frustrated Blair's wish to join the euro (Peston, 2005, ch. 6; Keegan 2003, ch. 12; Seldon 2004, pp. 682–3). Brown also controlled the budget by withholding information. As Scott (2004, p. 24) comments, ‘getting information about the contents of Gordon Brown's budget was like drawing teeth’ (see also Peston, 2005, p. 99, pp. 226–77; Seldon, 2004, p. 674). And, it mattered because ‘Brown always put his “poverty” agenda above Blair's “choice” agenda’ (Seldon, 2004, p. 688; and on the choice agenda see Blair (2004, ch. 43). Thus, Brown ‘viewed the big increases he achieved in NHS spending as a huge moral victory against Blair’ while he thought Blair's policy on hospitals was a ‘distraction from his achievement in increasing expenditure’. Blair's policy on tuition fees for universities was also deemed a distraction from the real achievement of Brown increasing education expenditure (Seldon, 2004, pp. 682–3).
It may be accurate that in the second term, ‘while Blair aimed … to limit Brown's authority over domestic policy, Brown fought to increase it (Seldon, 2004, p. 627). But the result was two men presiding over territory ever more jealously guarded. Brown was ‘immovable’, ‘dominating his own territory’ with ‘jagged defences designed to repel any invader, including the Prime Minister’. Not only was Downing Street left ‘wondering on the latest thinking about the Euro’ but ‘unthrifty ministers’ found him ‘unrelenting in his pursuit of his own strategy’. Brown's role was that of ‘social engineer who was redistributing wealth’. So, ‘they were not interested in submerging their differences in outlook, but in making an exhibition of them’ (Naughtie, 2002, p. 352). Brown was reported as saying to Blair that ‘There is nothing you could ever say to me now that I could ever believe’ (Peston, 2005, p. 349). By 2005, Brown was now ‘the official opposition to Blair within the very heart of the Cabinet’ (Peston, 2005, p. 13, p. 353). Their oscillating relationship is a fine example of the politics of political space. Brown commanded much domestic political space forcing Blair almost by default into overseas adventures.
A key characteristic of the period 1997–2005 is this shifting of fortunes, the contingency of the court politics and the duumvirate. Hennessy (2000b, pp. 493–500) has conscientiously mapped Blair's inner circle and its changing membership. Many commentators discuss its influence (see for example Rawnsley, 2001, p. 292; Rentoul, 2001, pp. 542–3; Seldon, 2001, p. 407). Francis Beckett and David Hencke (2004, ch. 14) describe the ‘oestrogen-fuelled’, ’Girl's Own, comic book’ view of life at the Number 10 court (see also Oborne and Walters, 2004; Price, 2005). We do not need to accept any account of life at Number 10 to make the observation that court politics are an important feature of the British executive.
Court politics were not confined to Blair and Brown. Ministers remain like medieval barons presiding over their own policy territory (Norton, 2000, pp. 116–7). Thus, the rivalry between Brown and Mandelson is a constant: ‘one of the great laws of British politics … is that any action by Mandelson causes an equal and opposite reaction by Brown’ (Peston, 2005, p. 223; see also Rawnsley, 2001, p. 20; Seldon, 2004, p. 162). There have been other major running conflicts; for example, between Brown and Alan Milburn, Secretary of State for Health, over Foundation Hospitals. Other ministers struggle to become heavy hitters. David Blunkett's frank if injudicious comments on the abilities and progress of his Cabinet colleagues are a public example of a conversation that Westminster and Whitehall conducts all the time in private. For example, Alan Milburn, Health, had ‘grown in competence and ability’; Margaret Beckett, Environment and Agriculture, is ‘just holding the ring’; Charles Clarke, Education, ‘has not developed as expected’; Patricia Hewitt, Trade and Industry, does not think strategically; and Gordon Brown throws his weight around (Pollard, 2005, pp. 27–8). Of course his colleagues reciprocate. John Prescott, deputy prime minister, is said to hold Blunkett in a mixture of contempt and suspicion while others grit their teeth at his ‘idiotic indiscretion’ (The Observer, 12 December 2004). Such gossip is the currency of court politics and the judgements are markers in the endless ministerial jockeying for position and recognition.
Not only are Blair's presidential tendencies constrained by court politics but the trends to presidentialism are overstated. There has been a growth in special advisers but the total number remains small compared with 3,429 members of the senior civil service. The effectiveness of such advice is also moot. Blair knows the general direction in which he would like government to move, but not how to get there. Officials protest that ‘the Prime Minister jumps around … It's a succession of knee jerks’ (Official, Cabinet Office, cited in Hennessy, 2000c, p. 9). The result is a frustrated civil service and special advisers. Thus, Scott (2004, p. 14, p. 17, p. 22) was clearly frustrated by what he saw as Blair's limited grasp of economics.
It may also come as a surprise to learn that Cabinet and its infrastructure of committees continues. As John Rentoul (2001, p. 544) observes, ‘a lot of the business of government continued to be done in cabinet committees’. So, during the second term of government, there were some 66 Cabinet committees and Tony Blair chaired 10 of them. Cabinet ‘stock takes’ are now in vogue and the deputy prime minister, John Prescott, is ‘keen’ on Cabinet committees (Hennessy, 2005, p. 10, p. 12). Similarly, ministers play their traditional roles. David Blunkett rationed his contributions to key issues. He did not interfere in the affairs of other departments. However, he brought highly political issues such as introducing identity cards to Cabinet where they were fully ventilated. The policy was also run through Cabinet and interdepartmental committees (Pollard, 2005, p. 26, pp. 305–66). If the decline of Cabinet government refers to the meetings of full Cabinet, then that specific meeting is no longer the forum for policy-making, if indeed it ever was. If Cabinet government refers to the Cabinet system then it is still active, even thriving, and desuetude is not yet Cabinet's fate. Political power is not concentrated in either prime minister or Cabinet, but more widely dispersed. It is contested, so the standing of any individual, prime minister or chancellor, is contingent.
Governance beyond Westminster and Whitehall
This argument is illustrated by the several studies of policy under Blair. 6 Of course, there are policy successes; for example, devolution to Scotland. At the end of Blair's first term, Polly Toynbee and David Walker (2001, p. 40) confessed that a ‘deep-dyed cynic’ would be impressed by Labour's commitment to a fairer society and conclude they have improved the lot of the poor, although the driving force of this poverty agenda was, of course, Gordon Brown, not Tony Blair. In the second term, despite the war in Iraq and its aftermath, ministers just got on with their jobs. The results make ‘pretty impressive reading’ and ‘by 2005 Britain was a richer and fairer society than in 1997’, especially for children and the elderly (Toynbee and Walker, 2005, p. 320, p. 327). Seldon (2005) is less sanguine. He concedes the government's success in tackling health and poverty but points to the failures in transport, Europe, House of Lords reform, local government and regionalism. Others would add the cock-ups, such as privatising air traffic control and tax credit payments, and the disasters, such as the millennium dome, and the invasion of Iraq and the failure to find weapons of mass destruction. Seldon (2005, p. 429) concludes that the second term ‘will be remembered as much for its opportunities lost as for its achievements’, and the opportunities were squandered because of the Iraq War and the perpetual discord with Brown. At best Blair leaves a ‘bitterly contested legacy’ (Riddell, 2005, p. 208).
Of course, Blair cannot claim credit for every policy success. Brown would lay claim to economic management and health service funding. Equally, not every failure lies at the door of Number 10. But there are the policy initiatives in which Blair took a personal interest. They include specific issues like social exclusion, teenage pregnancy, drugs and on-the-spot fines as well as more general initiatives like improving service delivery. For example, the crusade against drug use was led by his drugs Tsar, Keith Hellawell, former chief constable of West Yorkshire, who was given little extra money and no staff to a spectacular lack of effect; none of his four targets were met. The politics of presentation triumphed over the politics of substance. Similarly, Blair signalled that improving service delivery was one of his top priorities, setting ever more demanding targets for measuring and evaluating performance. However, Tony Wright, Labour Chair of the Select Committee on Public Administration, commented perceptively: ‘it is just not technically feasible, never mind desirable, to have that much centralization. If everything is a target, nothing is a target’ (cited in Rawnsley, 2001, p. 292). The emphasis on greater choice for users of public services is welcome but, as Short (2004, p. 279) points out, ‘public sector reform cannot succeed on the basis of headline-grabbing slogans’.
Finally, there is the rest of the world. Events such as 9/11, Northern Ireland, Kosovo, the Afghan war and Iraq divert prime ministerial attention from domestic policy. Over Iraq, for example, not only did Blair have to persuade international leaders on the case for war, which he conspicuously failed to do, he also had to maintain support at home, which he did, but at the price of eroding his authority in the party and with the electorate. The war presented Blair with the embarrassing resignations of two of his Cabinet colleagues, Robin Cook (formerly foreign secretary, at the time Leader of the House of Commons) and Clare Short. The resignation of Cook and the ensuing fallout increased Blair's dependence on his Cabinet colleagues. John Kampfner (2003, pp. 161–2, pp. 225–6, p. 272, p. 277, p. 315) describes the extent of the opposition to the invasion of Iraq in the parliamentary Labour party. The rebellion by 139 Labour MPs was the largest ever and the public demonstration in London was the biggest in decades. Even the Cabinet was uncertain, verging on divided. In the understated phrases that are employed at times of stress and conflict, Cabinet support moved from ‘rock solid’ to ‘broad’ and ‘fears were being expressed with uncharacteristic candour’ (Kampfner, 2003, p. 294, p. 255). As Cook (2003, pp. 271–2) pointed out, ‘this political damage to the Labour government was a self-inflicted wound’ and ‘it could have been avoided by listening to the majority who were opposed to the war’.
All governments fail some of the time. All governments are constrained by world events. All prime ministers intervene. Few control and then only for some policies, some of the time. The test of success in politics is elusive and shifting. Maybe, as Enoch Powell said, all political careers end in failure. Maybe, as George Orwell (1968 [1944, p. 156]) said, ‘any life when viewed from the inside is simply a series of defeats’. It is not that the Blair government differs greatly from other governments, but that so much more was expected. It is not that his ‘presidential’ style uniquely fails, but that its failures are no different to those of more collegial styles. Blair's failures stand in stark relief to the early promise, to the exaggerated claims and to the expectations of his supporters.
The problems the Blair government shares with all others have been compounded by two problems of his making: conflicts at the centre and his management style. Blair's initiatives have depended on Brown's support – for example, top-up fees for students where Brown called off the dogs at the last moment (Peston, 2005, p. 55; Seldon, 2004, p. 648; Stothard, 2003, p. 83). Although improving public services lies at the heart of the modernising agenda, ‘there were few signs that Blair was winning over his critics on public service reform’ (Seldon, 2004, p. 634, p. 636). Blair's weaknesses included a lack of follow through. He intervenes, persuades and then forgets. He lacks ‘policy making and management skills’ (Seldon, 2004, p. 692). So, although he wants results, ‘he finds it hard to understand why things can't happen immediately’ and he is frustrated when ‘waiting for the pay-off and he doesn't have time’ (Official, cited in Hennessy, 2000c, p. 10).
However, although ‘the machinery of government was in a state of permanent revolution at the centre after 1977 … he never succeeded in finding a structure that suited him’. In effect, the reforms were a sign of weakness, not strength (Seldon, 2004, p. 694). In 2001 Riddell (2001, pp. 38–9) talks of a ‘beleaguered centre’ and a prime minister weak on detailed policies. By 2005, he talks of ‘central flaws’ such as ‘inexperience’, ‘lack of clarity about both means and ends’ and ‘confusion about the role of central government’ (Riddell, 2005, p. 41).
Westminster Smokescreens
We have told stories about the dependence of the prime minister on the court politics of the core executive and on the networks of service delivery. We have also pointed to the importance of party support, and the impact of political adventures in the international arena on domestic politics. To compare Blair pre- and post-Iraq is to see that prime ministerial pre-eminence comes and goes; to witness the transition from President Blair to the ‘unfulfilled prime minister’ (Riddell, 2005).
Some of the claims about the changing pattern of political leadership in Britain are accurate. It helps to distinguish between the electoral, policy-making and implementation arenas. First, personalisation is a prominent feature of media management and electioneering in Britain. As Michael Foley (2000, p. 98, p. 11) observes, Blair dared to be Thatcher and ‘raised the concept and application of spatial leadership to unprecedented levels of development and sophistication’. If we must use presidential language, it is here in the electoral arena that it is most apt. Blair is the figurehead. However, this statement must be qualified immediately because the court politics of the duumvirate fits uncomfortably with the notion of monocratic leadership. Brown played a pre-eminent role in the 2001 election (Seldon, 2004, ch. 31).
In the policy-making arena, there is some truth to the claim that Blair centralised policy-making on Number 10 and the Cabinet Office and eschewed Cabinet government. However, this claim applies to selected policy areas only, with the equally important proviso that the prime minister's attention was also selective. The continuous reform of the centre speaks of the failure of coordination, not its success.
The prime minister's influence is most constrained in the policy implementation arena, so it is conspicuous for its absence in most accounts of presidentialism. Here, other senior government figures, ministers and their departments, and other agencies are key actors. Similarly, although personalisation can affect implementation, that effect is intermittent. Too often, the presidential thesis treats intervention as control. There is much that goes on in British government about which the prime minister knows little and affects even less. And, all these arenas are embedded in dependence on domestic and international agencies and governments, making command and control strategies counterproductive.
So, we have a paradox. On the one hand, journalists, political scientists and practitioners are telling tales of a Blair presidency characterised by centralisation, personalisation and pluralisation. On the other, the same people recount governance stories in which British politics consists of fragmented policy-making and policy implementation networks over which a core executive maintains a fragile – and increasingly fraught – influence. We want to draw attention to two ways of interpreting this paradox.
First, all the discussion about a Blair presidency is a counter both in the court politics of the duumvirate and in wider party politics (and on the analysis of such rhetorical games see Bevir and Rhodes, 2005, pp. 178–80). For Blair's supporters, it is a way of promoting his standing in the party and in the country. For opponents inside and outside the Labour party, it is a way of expressing hostility to Blair in particular and the Labour government in general. So, it matters not that the presidential analogy is misleading because the game is not about empirical accuracy. The critics have several specific targets.
Foley (2004) argues that the epithet can refer to Blair's personal characteristics, to claims that he is too powerful, to the consequences of Blair's command and control style of government, to his international adventures and attendant disregard of domestic politics, to his flouting of constitutional conventions, to the influence of the USA on British politics and to the failure to understand the shift from government to governance. So, the term is a smokescreen behind which lurk several criticisms of Blair and the Labour government.
Conversely, when critics bemoan the demise of Cabinet government, what exactly has been lost? Patrick Weller (2003, pp. 74–8) distinguishes between the Cabinet as the constitutional theory of ministerial and collective responsibility, as a set of rules and routines, as the forum for policy-making and coordination, as a political bargaining arena between central actors and as a component of the core executive. Blair's critics single out Cabinet's policy-making and coordination functions, yet it has been clear for over a quarter of a century that these functions have been carried out by several central agencies including but not limited to the Cabinet. To suggest that Blair has abandoned the doctrine of collective responsibility is nonsense. Unity is essential to electoral success, so dissenters go. To suggest that any prime minister in the post-war period has adhered to anything but a pragmatic view of individual ministerial responsibility is equally foolish. Ministerial responsibility is alive and well, although not in its conventional formulation. It is no longer the prime minister and the political standing of the minister alone that decide a resignation – but the media maelstrom (Woodhouse, 2004). David Blunkett, home secretary, had high personal political standing in the party and the full support of the prime minister but the pack brought him down (see Pollard, 2005, ch. 12, ch. 13; Woodhouse, 2004, p. 17). It would seem that only fox-hunting among blood sports is to be banned.
In short, and again, key terms about British government act as smokescreens. But what are they acting as a smokescreen for? Why do so many people, who describe British governance as multipolar, nonetheless constantly talk about a Blair presidency? We want to suggest that the paradox arises because of the bewitching effect of the Westminster model of British politics. In the need to preserve Westminster fictions, the tales of presidentialism are a smokescreen behind which we find a widespread acceptance of the governance narrative. If a commentator accepts any version of the governance narrative, with its stress on interdependence, then any tale of a Blair presidency will be undermined. Command and control mix with interdependence and cooperation like oil and water.
So, how does the Westminster model infuse talk of a Blair presidency? Of course, there is no agreed version of the Westminster model. There are at least three possible versions: Tory, Whig and Socialist.
Philip Norton is a Tory and a combative defender of the UK constitution against all comers. He believes the Blair presidency is ‘dangerous’ because it centralises power in Number 10, adopts a principal—agent relationship with departments ‘that is likely to be difficult to sustain’, relies on goodwill for implementation ‘that may not be forthcoming‘ and ‘ignores parliament’. These problems are compounded by ‘the lack of experience and, indeed, understanding of government by the prime minister and many of those around him’ coupled with a ‘leadership … obsessed with power’ and ‘no understanding … of relationships within the system’ (Norton, 2003, p. 277). Underpinning this critique is a governance interpretation of British government that believes ‘interdependency is a necessary feature of government’. However, ‘the more the prime minister and senior ministers have sought to centralise power in their own hands then … the more fragmented British government has become’. Norton worries that ‘the glue of government has started coming unstuck’ (Norton, 2003, p. 276).
What to do? We need to end the ‘institutionalisation of fragmentation’ by returning to the ‘party-in-government’ as the body ‘responsible for public policy’ that ‘can be held accountable by electors at a subsequent general election’ (Norton, 2003, p. 278). In other words, Norton uses the governance narrative to urge a return to the eternal verities of the Westminster model. He criticises the notion of the Blair presidency to resurrect the Westminster model.
Hennessy (2000b, p. 535) is a Whig: ‘history is a discipline that sobers up its practitioners’. He rejects the command and control model of the prime minister as chief executive for two reasons. First, ‘command models sit ill with open societies’. Second, ‘British political culture reflects the compost in which it is grown’. It is a parliamentary not a presidential compost. So, he defends the ‘deep continuity’ of the constitutional side of the job – relations with the monarchy, accountability to parliament, collective government and a career civil service (Hennessy, 2000b, p. 539). However, he too recognises that Britain must change to meet the challenges of an interdependent world. He foresees prime ministers ever more entangled in international affairs, an expanding ‘hybrid arena’ where international and domestic mingle, relentless media pressure, ‘the avalanche of information’ and a reconfigured British state because of, for example, devolution (Hennessy, 2000b, p. 538). In sum, he describes a world of complex interdependencies.
To meet these demands, he envisages, for example, Number 10 distancing itself from the hurly-burly and developing both a plurality of analytical capacities and a greater capacity to provide risk and strategic assessments. All such changes would be within the context of collective government. Or, to rephrase, to meet the challenges posed by the governance narrative, Hennessy envisages a return to Cabinet government with reinforced analytical and strategic support. His notion of the British presidency is less that it is dangerous, although it may well be, but that to institutionalise it is to plant an alien invention in British soil.
The socialist tradition in the guise of New Labour has its own conception of how British government should be run. 7 In Peter Mandelson and Roger Liddle's (1996, ch. 10) ‘shadow’ manifesto they argue that, to succeed, Blair needed ‘personal control of the central-government’. They describe with approval Margaret Thatcher's ‘focus on a clear set of goals’ and ‘strength of will’, claiming it ‘says a lot about leadership in government’. Tony Blair should follow her example ‘in getting control of the centre of government’. In particular there should be a ‘more formalised strengthening of the centre of government’ so it can ‘give much-needed support to the prime minister’ and ‘provide a means for formulating and driving forward strategy for the government as a whole’. So, the Number 10 Policy Unit should be ‘beefed-up’, and the Cabinet Office needs to be more ‘pro-active’. When New Labour came to power, therefore, it should have been no surprise that ‘there was never any intention of having collective Cabinet government’. Blair was ‘going to run a centralised government, with a commanding Policy Unit which was solidly New Labour’ (insider, cited in Seldon, 2004, p. 437).
There are two features of New Labour's approach worth noting. First, it is strongly influenced by the example of Margaret Thatcher's leadership style. Second, it consigned Labour traditions, many of which are more democratic, to the dustbins of history (Bevir, 2005, pp. 128–37). Even the contrast with Jim Callaghan or Harold Wilson is marked. For example, Callaghan (1987, p. 408) observes that a prime minister ‘is able to provide himself with his own sources of information, he can send up a trial balloon or fire a siting shot across a Ministerial bow without directly involving his own authority or publicly undermining that of the Minister’. Deserting Labour traditions for Thatcherite dynamism had its costs. It provoked criticism for eroding the ‘traditional norms of democracy and administration in favour of a model that rested more on central diktat’. No previous Labour leader ‘had adopted such a personal style of control’ so, once again, Blair ‘showed himself to be a leader lacking empathy with the traditions of his party’ (Seldon, 2004, p. 694).
Yet, Blair and his entourage consistently deny that they have abandoned collective government, arguing that their reforms are consistent with present-day constitutional conventions. In part, such a defence is mere conventional convenience. If policy-making is presidential, then only the president is to blame when things go wrong. However, when the government faced its many policy-making and implementation problems, it blamed those long-standing whipping boys of the Westminster constitution – the civil service – said to lack both ideas and drive (Seldon, 2004, p. 436). Others saw a problem with Blair's policy-making and management style and the mistaken belief that running the government was like running the Labour party writ large. Such autocritique was not on the central agenda.
Of course, the government could see that policy success depended on others cooperating – hence the drive to ‘joined-up’ government (see for example Cabinet Office, 2000; Cm 4310, 1999). The ubiquity of networks was drawn to the government's attention by its own think tanks (see for example, Perri 6, 1997; and, for comment, Bevir, 2005, pp. 29–53). They did not translate this recognition of dependence into a new leadership style. The governance narrative conflicted with their view of a strong centre. Command and control remained in vogue for running services built around many governments and organisations. But, whatever the attractions of command and control, it did not work. New Labour's beliefs about the best way to run government positioned Blair between the rock of presidential critiques and the hard place of governance. Only the Westminster model obscured the dangers of such a position.
Finally, there is one characteristic of the Westminster model that is present in every tradition – it is inward looking. Once we look at the role of the prime minister beyond the confines of Westminster and Whitehall, any assessment of their presidentialism must be tempered. For Britain, the post-war years saw the end of empire and a loss of influence in the world. September 11th and Iraq rubbed salt into the wounds of dependence. British political leaders never ceased to hanker for a return to world prominence. So, parliamentary sovereignty and the Westminster constitution live on as emblems of a past age. The debate about presidentialism is a false debate, a smokescreen obscuring the frailty of the eternal verities of a tattered constitution.
Conclusion
When commentators focus on Westminster and Whitehall, the prime minister can appear pre-eminent. When their focus shifts beyond Westminster and Whitehall, to the rest of the UK and beyond, then any ‘presidential’ pretensions must be tempered by recognition of dependence. The inescapable fact is that Blair has to work in, with and through a complex web of organisations, governments and networks with his power constrained by ever more pervasive and complex patterns of dependence. The more we look outside the Westminster model, the more we find that centralisation, pluralisation and personalisation represent, not a concentration of power, but an endless search for effective levers of control by a core executive less powerful than many commentators and insiders claim. And, while the notion of a core executive can encompass the duumvirate and even baronial government, the ideas of prime ministerial power or presidentialism cannot. We can think of no clearer example of how the language of Westminster obscures our understanding of trends in British governance. We live in a land where barons vie for favour in the court of a would-be president as dependent on them for support as they are on him for favours.
We have contrasted the presidential and the governance narratives to show that recent trends in British government do not provide certain evidence of prime ministerial power. Tales of the Blair presidency can be retold as tales of the unfulfilled prime minister. There are two major limitations to the focus on presidentialism. First, when used as a smokescreen for attacks on the prime minister and government, the term is but a flag of convenience. Better by far to focus on the specific criticisms. If used as an analogy to identify leadership changes, it is potentially misleading because the differences between a parliamentary and a presidential system far outweigh the likenesses by some margin (see Rose, 2001, pp. 236–44). Better to talk of changing patterns of leadership. Second, a focus on presidentialism is too narrow; it is excessively preoccupied with Westminster and Whitehall.
The analysis of changing patterns of leadership should start with the beliefs and practices of politicians and civil servants (see Bevir and Rhodes, 2006). Giovanni Sartori (1997, p. 102), in his comparative analysis of parliamentary systems, suggests that prime ministers can be first above equals, first among unequals and first among equals. The British prime minister can be each of these, but not at the same time, and by exploring the contingencies of political life and the ways in which individuals modify their inherited beliefs and practices when they confront the dilemmas of governance, we will understand how patterns of leadership change.
Footnotes
We would like to thank Andrew Gamble, Anthony Mughan, David Richards, John Wanna, Martin Smith and the Journal's three referees for their helpful advice and comments on the draft of this article.
1
For reviews of the academic literature in the UK see Heffernan (2003); Rhodes (2003); Smith (2003). For useful collections of articles see King (1969; 1985); Rhodes and Dunleavy (1995). For the equivalent debate in Canada see Punnett (1977, ch. 1; and compare Savoie (2003) with Bakvis (2000). For Australia, see Aulich and Wettenhall (2005); Weller (1985; 1992; 2003). For a cross-national comparison of the presidentialisation trend see Helms (2005); Poguntke and Webb (2005). A useful way of understanding the link between this literature and our focus is Tivey's (1988, p. 3) notion of ‘the image’. It refers to ‘a set of assumptions about “the system” … and how it works’. It contains ‘operative concepts’ or ‘operative ideals’. So, such notions as the Westminster system have become operative; ‘they have gained currency among those who study politics, and diluted and distorted they have reached the practitioners’ (Tivey, 1988, p. 1; see also Beer, 1982 [1965, p.] xiii, p. 404). We explore the diluted understandings of the Westminster system, drawn in part from the academic literature, that provide the operative concepts for practitioners.
2
Most quotes are from practitioners but we also draw on distinguished journalists and biographers because of their sources. In particular, we use Rawnsley (2001), Hennessy (2000b), Peston (2005) and 
because their insider sources seem as impeccable as they are limitless. There is a further complication; there is often no clear-cut distinction between academic commentators and elite actors. So, for example, Lord Crowther Hunt was both a member of the Fulton Committee on Civil Service Reform and a Fellow of Exeter College, Oxford University. Subsequently, he became a political adviser to Prime Minister Harold Wilson, whom he advised on implementing the recommendations of the Fulton Committee. Individuals can be academics, authors of official documents and political actors all at once or at different times in their lives. Thus, distinguished present-day commentators such as Lord Norton of Louth wear many hats. The key point is that we seek to tap the considered views of practitioners whether reported by our academic colleagues or by themselves in whatever capacity.
3
Foley (1993; 2000; 2004) is the most prolific academic contributor. Others who identify a trend to presidentialisation, even while criticising it, include: Allen (2003); Heffernan (2003); Hennessy (2005, and citations to earlier work in his note 3); Mughan (2000); Kavanagh and Seldon (2000); Poguntke and Webb (2005); Pryce (1997); .
4
On the several definitions of the presidentialisation thesis see: Foley (1993, ch. 1); Mughan (2000, pp. 9–10); Poguntke and Webb (2005, p. 5, pp. 8–11); , p. 37, p. 66).
5
For the original accounts of the governance narrative and the differentiated polity see Rhodes (1988; 1997; 2000). For criticisms see Heffernan (2005); Holliday (2000); Marinetto (2003); Marsh et al. (2003). For replies see Rhodes (2003); Rhodes (2007). More recent versions of the governance narrative decentre networks by analysing them as the products of situated agency, explaining shifts in government by reference to contingent contests over meanings and suggesting that power has always been more dispersed than many political scientists allow – the state is never monolithic and it always negotiates with others (see Bevir and Rhodes, 2003; ).