Ian Kershaw examines the extent to which strong leaders can effect sweeping political change.

How did Hitler's narcissism influence his dominance of German politics during the Third Reich? How far was Nazi policy dictated by the obsessional hatred of Jews that was unquestionably a vital feature of Hitler's personality? These apparently simple questions demand complex answers, even for one of the most studied individuals in history.

The guessing game about Hitler's personality and psychological make-up has not let up since it began soon after he first began to attract attention in the Munich beer halls. One recent speculation finds the key in his presumed homosexuality, another in the syphilis he allegedly contracted from a Viennese prostitute. Was the young Hitler drawn into Vienna's gay scene? Or did he frequent brothels? Or neither? No one ever asked him, so it's all intellectual guesswork, prompting the reductionism beloved of psycho-historians.

If we are honest, we have to admit that it is impossible to be sure where, how, when and why Hitler acquired such obsessional hatred of Jews. So it's perhaps best to answer the more important question of how this obsession helped Hitler's rise to power then translated itself into genocidal policy once he ruled Germany. But this then turns into a question ranging far beyond Hitler's personality.

ÖйúAƬ

And however much the personality of the individual in this one - self-evidently important - issue needs to be taken into account, it seems obvious that no deductions can be drawn from it about the effect of personality on politics generally. Each individual's personality is unique.

Even if the psychology and motivational forces that shape a personality could be fully established, this would have significance only for that singular case, and even then it would be only one part of an explanation of the individual's political effectiveness. As the example of Hitler shows, the psychological underpinnings of political motivation can often be grasped only imperfectly. All that can be said is that where individuals play a significant role in politics, their specific imprint is shaped in part by the traits of their personality - hardly an earth-shattering conclusion.

ÖйúAƬ

A more promising approach would be to move away from concentration on "personality" to consider the related but separable issue of the effect of the individual on the political process, quite specifically on the shaping of major political change. Here, it is worth bearing in mind Karl Marx's dictum on Louis Bonaparte: "Men do make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past."

The maxim can be applied to dictators as well as to democrats, to absolute monarchs as well as to revolutionary leaders. Particular cases, whatever their differences, might conceivably here suggest a number of generalised conclusions.

As an example from a pluralist democracy, take Margaret Thatcher, in personality terms the most assertive postwar British prime minister. It would be absurd to deny that she had a substantial effect on late 20th-century politics in the UK. Her single-minded pursuit of objectives, overriding all opposition, was legendary and is central to her self-styled historical image. It did, of course, demand strong will and great tenacity to combat the then powerful unions. And Thatcher unquestionably left an indelible personal imprint on the style of her government - a "handbagging" has become popular parlance.

But seen in terms of the longer run of British politics and economic development in the second half of the 20th century, Thatcher's ability in 11 years of rule, backed by huge majorities, to bring about fundamental change seems more limited - shaped, but also constrained, by objective structural determinants that even she was unable to control or master.

Given an alternative Conservative prime minister, the confrontation with the unions may have been less brutal. But the unstoppable changes in the world economy - including falling demand for steel and coal, the revolution in information technology and globalisation - would have necessitated under any government the systematic reorientation of the UK economy, which Thatcher at best accelerated by drastic methods.

And, for all the emphasis she placed on radical change, the hallmark of her legislative programme (as of almost all governments this century) was continuity. Most of the legislation she inherited went unrepealed.

Thatcher, in other words, had an effect on British politics. But, in terms of her personal ability to bring about significant long-term change in Britain, it would be as well not to exaggerate it. The same point could be made of Tony Blair. What might be said of both is that as "presidential" styles of running the government have apparently replaced the role of Cabinet - to my mind, an exaggerated assertion - external constraints on prime ministers' room for manoeuvre of have grown, not diminished.

In authoritarian regimes, there are fewer constraints on regime leadership than under democratic-pluralist systems of government. It would be perverse to suggest that an individual dictator does not have a profound influence on his regime's policy. To try to reduce Hitler to no more than a cypher for the interests of big business always was an absurdity. Under another leader of Germany in the 1930s, it is doubtful whether national assertiveness would have been taken so far that general European war was the consequence, and almost certain that discrimination against Jews would not have led to the "Final Solution".

ÖйúAƬ

Yet even dictators are not free agents. Here, too, the personal role is conditioned by an array of internal and external determinants that the individual can influence to only a minor degree, if at all. Most authoritarian leaders are bound by the interests of the power groups they represent - normally the army and a ruling monopoly party. Given the invariably accompanying repression, successful resistance from below is rare. But if the interests of the power elites are not upheld, other contenders for leadership emerge, and a mass basis of support for a new claimant is engineered. The result is usually that the person of the dictator is changed or the regime is replaced.

In exceptional circumstances, however, usually after a period of most profound crisis or revolutionary upheaval, a dictator can build such a level of personalised power that he is able, for a relatively short period, to override the sectional interests of the elites that hoisted him to power, break the constraints that these normally impose and even threaten to undermine or destroy those interests. Here, the effect of the individual is extremely far-reaching. But it is usually of fairly short duration. The obvious examples are the two most infamous dictators of the 20th century, Hitler and Stalin.

ÖйúAƬ

Hitler, like all modern dictators, ruled by a combination of manipulated acclamation and terroristic repression. But down to the middle of the war, his decisions were not out of step with the interests of the major power elites in Germany. And until the war began to turn sour, he enjoyed a high level of genuine popularity in Germany, based on what were widely perceived to be his "successes" and "achievements". This enabled him gradually to establish a personal dominance unusual even for dictators and to break free of the constraints that the traditional power elites might otherwise have imposed.

By the time the power elites recognised that Hitler was leading them to national disaster, it was too late. Dissident groups failed to overthrow him from within. Nothing was left but to await the destruction of the regime from outside. When all account is taken of structural determinants in the course of the Third Reich, Hitler's personal imprint on what happened is undeniable. But that personal role has still to be located in the constellation of forces that created and sustained such a destructive and inhumane regime. Politics, even in the Third Reich, cannot be reduced to the role of the individual.

In Stalin's case, the personal imprint on politics is also self-evident.

There were alternatives open to the Soviet leadership when Stalin gained power. The path taken through brutal collectivisation was Stalin's choice.

And the extreme levels of terror directed at his own people reflected Stalin's paranoid personality. But the violent nature of Stalin's regime cannot be reduced simply to an expression of the dictator's paranoia. Not only did it mirror the social and political forces that had produced such a monster; the terror was itself possible only because it accorded with the interests of, and was backed by, the dominant forces within the Communist Party.

But as terror spawned terror, the party increasingly became little more than a vehicle of the elaborated Stalin personality cult, while the army leadership was decimated in the purges. Stalin, like Hitler, but in this case more through terror and fear than plebiscitary acclamation, had largely freed himself from internal political constraints. The war, in which Stalin (unlike Hitler) gradually acknowledged the need to concede strategic decisions to military professionals, then raised him to a new pedestal of national hero. Until his death in 1953, the dictator was untouchable, his personal steerage of Soviet politics indisputable. Whether an internal coup would have been attempted at some point had he lived much longer is purely a matter of speculation.

What was remarkable, nonetheless, was the rapidity with which party dominance, undermined while he was alive, reasserted itself as soon as he was dead. The structurally damaging aspects of Stalinism, especially the extreme insecurity under his rule of all echelons of government and administration, were with remarkable speed converted into orthodox repressive authoritarianism under party domination. Between Khrushchev and Gorbachev, the leader of the Soviet Union was no dictator in the Stalin mould but rather the representative figure of party interests. Dramatic change came under Gorbachev because traditional party interests were no longer compatible with economic realities. Fundamental alteration was recognised as inevitable by parts of the Soviet elite. Once introduced, the change was unstoppable. The system collapsed. The individual, Gorbachev, had been the agent, or catalyst, of change. But his role had been in essence to acknowledge that the impersonal forces demanding change could no longer be held back, Canute-style.

These and other examples perhaps suggest a number of generalised hypotheses about the individual's effect on politics.

First, in pluralist-democratic systems, where political decisions are reached by rational processes of deliberation, the role of even powerful personalities in bringing about major political change is limited and, for the most part, subordinated to the determining influence of wider, more impersonal forces.

Second, in authoritarian systems, the scope for determining individual influence is much greater, though it is more dependent than might at first be thought on satisfying the interests of the power elites that had created the authoritarian framework of rule and subject to external influences and determinants. Ultimately, the individual's role is still circumscribed, though less so than in pluralist systems.

And third, some authoritarian systems can nevertheless be described as "exceptional regimes". The background here is framed by an extraordinarily far-reaching crisis of legitimacy of the previous system in which the "heroic" status of the authoritarian leader who emerges from the political turmoil elevates him to a position in which, over time, his personal power becomes a factor of decisive importance, even to the extent that it can begin to erode, or conflict with, powerful sectoral interests. This "exceptional rule" can, by definition, last for only a limited time before it collapses, is defeated by external forces, is replaced by a more pluralistic system or subsides into more conventional authoritarianism.

Other than in extraordinary circumstances, therefore, the effect of the individual on politics is of accelerating or retarding, to some extent, processes that are, in the main, shaped by impersonal forces. Marx's adage, formulated more than a century and a half ago, seems the most appropriate encapsulation of this relationship. So in one thing, at least, Marx was right.

ÖйúAƬ

Ian Kershaw is professor of modern history, University of Sheffield.