As the Labour Party prepares for its annual conference Vernon Bogdanor asks what the ingredients are for good leadership.

It is easy for government to be successful. Mussolini, so it is said, made the trains run on time, although he appears to have been less good at making war. But any government that can control society and the means of communication has a good chance of proving successful, in the short term at least. The more interesting question is what makes for successful government in a democracy, a form of polity in which the political leaders depend on public support.

Bob Worcester, founder of MORI, Britain's leading polling organisation, is fond of saying that in a democracy public opinion is king.

This is even more so in the era of referendums and focus groups. Still, no country, not even Switzerland, governs entirely through the instruments of direct democracy. Governments everywhere must gauge public opinion for themselves.

ÖйúAƬ

But what is public opinion? How are popular attitudes to be distinguished from deeper trends of public sentiment? There can be little doubt that the British public was solidly behind Neville Chamberlain's policy of appeasement, as indeed the Gallup polls confirmed. This did not stop the public turning on Chamberlain when, in 1940, Britain found itself exposed to a greater danger than it had ever known. Similarly, before 1982, few British voters held strong views on the Falkland Islands. Nevertheless, had Margaret Thatcher's government been defeated in the attempt to retake them after the Argentine invasion, it might well have fallen, so great was public outrage against the Argentinians.

Successful government, then, does not consist simply in following popular attitudes wherever they may lead. It requires something more. It means discerning those deeper currents of opinion that lie behind the superficial manifestations. It means discerning what Isaiah Berlin once called "the hoofbeat of history". Sometimes, that hoofbeat can be heard by political leaders of a quite unintellectual, or even anti-intellectual, cast of mind.

ÖйúAƬ

George Schultz, former US Secretary of State, wrote in his memoirs that he could never understand how Ronald Reagan knew so little but achieved so much. The same could have been said of Franklin D. Roosevelt.

Yet it is not accidental that so many of the most successful political leaders in modern democracies - in the US, Theodore Roosevelt and Harry Truman; in Britain, Benjamin Disraeli and Winston Churchill; in France, Charles de Gaulle - have been keen students of history. What they learnt from history was not the drawing of superficial analogies - the supposed "lessons" of Munich or Suez - but rather an understanding of the main political and intellectual forces of the age. For the politician seeking to be a statesman, knowledge of history is priceless.

Knowledge of history can also give us clues as to which types of government are likely to prove successful or unsuccessful. There are in fact three main enemies of successful government: ideology, moralism and panic.

Ideological government is unlikely to prove successful since ideology lies in contradiction to historical understanding. It yields a closed system of thought, rather than a necessary openness to experience.

British Labour governments between the years of Clement Attlee and the advent of Tony Blair found themselves victims of what Samuel Beer, the American political scientist, called "compulsive ideologism". This was obviously so in the left-wing of the Labour Party, which avoided facing up to the realities of the mixed economy and the improvement in working-class living standards in the postwar world. "The masses," as Friedrich Engels once said, "have got damned lethargic after such long prosperity." But this compulsive ideologism was equally noticeable in the so-called Revisionist or Gaitskellite right-wing of the party, which was just as preoccupied with discovering the "true meaning" of socialism. As Beer noted: "The contestants, whether fundamentalist or revisionist, found themselves in agreement on a basic premise. All accepted the necessity for a social philosophy with programmatic consequences. The opposing sides were at sword's point with regard to their respective ideologies, but they were united in their ideologism."

It was this fixation on ideology that largely prevented the Labour governments of the Harold Wilson and James Callaghan era in the 1960s and 1970s from getting to grips with economic problems. Labour spent much of its time in opposition in the 1950s discussing the futile question of whether it should "accept" the mixed economy. It failed to consider the real question of how a mixed economy should be successfully managed. The ideologies of both Left and Right were like rigid searchlights. In focusing on the problems of yesterday, they turned away from the more difficult challenges of the contemporary world.

Labour had a much better chance of success after the removal in 1995, under the Blair leadership, of Clause Four of its constitution, which advocated nationalisation of the means of production, distribution and exchange, thus allowing it to develop into a practical social democratic party on the continental model. But to govern successfully it is not enough for a leader to lead. He must also find followers who will follow.

ÖйúAƬ

Just as a government that seeks to create a new heaven on earth is unlikely to prove successful, nor will a government that imposes its own values upon society. The most striking example of the failure of moralism in government is the attempt to prohibit alcohol in the US in the 1920s. Prohibition was characterised by Herbert Hoover, before becoming president in 1929, as "a great social and economic experiment, noble in motive and far-reaching in purpose". Yet, far from curing the evils of alcoholism, it encouraged them, since many who had previously been temperate in their habits began to imbibe freely in protest against the invasion of their "personal liberty" by the law. They demanded more liquor, not less; and, as John D. Hicks, the American historian of the period, wryly commented, "private enterprise, although in this instance unassisted by the law, never showed greater efficiency in meeting a consumer demand". The outcome was not temperance, but bootlegging and the growth of private armies of gunmen and thugs, of whom Al Capone was the most notorious example.

ÖйúAƬ

Successful government, then, is generally non-ideological government. It is also likely to be limited, rather than hyperactive, government. In 1993, Anthony King, a political scientist at Essex University, asked in The Daily Telegraph : "Will the Honourable Action Men please sit down at once?" He was referring primarily to the justly forgotten Tory minister, John Patten, and his Education Bill of that year, to which the Government had added 8 amendments during the House of Commons committee stage, a further 78 during the report stage, 258 during the House of Lords committee stage, 296 on report and 71 at the third reading. Did any of this make the slightest difference to educational achievement? Almost certainly not. We were governed, King complained, "by an entire tribe of hyperactive children".

Hyperactive governments are prone, moreover, to panic - Kenneth Baker's Dangerous Dogs Act of 1991 and Thatcher's poll tax being striking examples.

Much hyperactive government occurs in the first phase of a term of office, as with Wilson in 1964, Heath in 1970 and Thatcher in 1979. Whatever the merits of Franklin D. Roosevelt's 100 days in 1933, its repetition by less distinguished practitioners has led only to disaster.

A successful government, it may be thought, leaves a country feeling happier. By that criterion, the peacetime administrations of Churchill and Dwight Eisenhower were successful. Yet both, in pandering to the immediate wishes of voters, ignored more serious problems - Churchill failed to develop a sound European strategy, while Eisenhower refused to confront McCarthyism and segregation, leaving, in each case, difficult legacies for their successors.

Successful government involves confronting the problems of the future, not those of the past. At the beginning of the 20th century, Arthur Balfour, although Prime Minister of a decaying government, established a modern structure for secondary education, machinery for the organisation of defence, and ended Britain's splendid isolation. John Major, an equally undervalued prime minister, who also had to face a withdrawing tide, ensured, through negotiation at Maastricht and skilful parliamentary manoeuvring at Westminster, that Britain remained in the European Union.

Moreover, he introduced into the public services many of the techniques and disciplines of the private sector, believing, in contrast to Thatcher, that the public sector did not have to be equated with the second rate. It is thanks to him that we at least know whether or not the trains are running on time. Posterity will be kinder to Major than his contemporaries have been.

Successful government, then, is government aware of its limitations, quiet government, government by men and women with a highly developed sense of history. The trouble is that successful government, as the 19th-century journalist Walter Bagehot well knew, is almost always dull government.

Those who seek excitement will have to find other fields for exercising their talents.

ÖйúAƬ

Vernon Bogdanor is professor of government at Oxford University.