A hundred years ago, we seemed about to witness the strange death, not of Liberal England, as George Dangerfield suggested in his book, but of Conservative England. The hold of the Liberals on government, buttressed as it was by the Irish nationalists and the infant Labour Party, seemed unshakeable. It was not easy to see how the Conservatives could ever displace them. After the 1914-18 war, however, everything changed. The Conservative century began. In every election between the wars, the Conservatives secured more votes than the Liberals or Labour. Indeed, until the advent of Tony Blair in 1997, there were just two governments of the left with comfortable majorities - in 1945 and 1966.

How did this transformation come about? Scholars have disputed the causes endlessly. Many years ago, the Oxford historian Ross McKibbin found the answer in the huge increase in the franchise generated by the fourth Reform Act, which gave the vote to men over 21 and most women over 30 in 1918. He now says that this answer was too deterministic, and suggests that we should look instead at the changing political culture of England during this period. Parties and People is a model of careful scholarship that serves to summarise recent research, rather than suggesting any controversial reinterpretation of accepted views.

Liberal predominance before 1914 was based, McKibbin now thinks, on the notion of a "Progressive Alliance". This alliance was grounded socially in the culture of nonconformity that united the Liberals with Labour, and ideologically in a commitment to free trade, social reform, free collective bargaining and constitutional change.

The 1914-18 war shattered this alliance - perhaps it was always destined to be impermanent. The war split the nonconformists, weakening if not destroying them as a political force. Ireland achieved independence and a new socio-economic agenda displaced constitutional issues. The Liberal social programme receded into the background. Labour became the dominant party of the left.

Before 1914, most Labour supporters had been prepared, in the absence of a Labour candidate, to vote Liberal. After 1918, however, Liberals proved unwilling to reciprocate. They were too frightened of the social threat that they thought Labour represented. The Liberal vote in the 1920s swung roughly 2:1 in favour of the Conservatives, making any renewal of the Progressive Alliance impossible.

Proportional representation might have remedied this situation, but it was defeated in 1917 - not solely by the Conservatives, as Mc-Kibbin suggests (indeed the Tories in the Lords favoured it), but, paradoxically, by the Liberals and primarily by the Liberal prime minister Lloyd George, who later regretted it. Indeed, the Liberals did not come to support proportional representation until 1922, when they were in opposition, and by then it was too late.

The culture that sustained the Progressive Alliance collapsed with startling suddenness during the First World War. The culture that sustained its successor, Labourism, has been in steady decline since 1951. In a series of essays in the London Review of Books, McKibbin has revealed his nostalgia for that Old Labour culture. Yet, as he points out, Old Labour, unlike the Liberals, accepted the institutions of state quite uncritically, and it was for this reason that "the decline of the Liberals vitiated England's political culture. The Liberals, as the party of nonconformity and Celtic Britain, to some extent stood outside the hierarchies of the English state and eyed them in a rather jaundiced way."

The Conservatives and Labour, on the other hand, "disagreed about the economic functions of the state, not about the state and its institutions as such". Old Labour never thought seriously about constitutional reform. Indeed, Herbert Morrison was happy with an upper house based almost exclusively on a hereditary peerage, on the grounds that "we should try to maintain continuity and not set up something new and different from the past".

Part of the reason for Labour's decline after 1951 was, McKibbin argues, that its "postwar electoral base, though much larger than in 1939, was not a broad-based democracy but one over-reliant on trade union membership, 'heavy' manufacturing and mining, and council housing, a base which, with the exception of council housing, was shrinking, even in the 1940s". Labour's failure to reform the state created a legacy of "a society with powerful democratic impulses but political structures and habits of mind which could not adequately contain them. It was an unresolved tension which was to dog England for the rest of the century." That tension remains. It is no doubt unfair, though sometimes irresistible, to draw "lessons" from a work of historical scholarship. Perhaps the lesson of Parties and People is that the unresolved tension to which McKibbin draws attention can be resolved only through the creation of a new Progressive Alliance.

Parties and People: England (1914-51)
Ross McKibbin
Oxford University Press, 202pp, £20

Vernon Bogdanor is professor of government at Oxford University and the editor of “From New Jerusalem to New Labour: Prime Ministers from Attlee to Blair", published by Palgrave Macmillan (£20)