Most of us have been faced at some time with demands for an essay on ‘The causes of the French Revolution’, or some other great event. Yet the notion of ‘cause’, whether for the historian or the plain man, is one of the most baffling that the student of History ever encounters. We all think we know what it is, yet the more closely we look the more puzzling it becomes. In the following 10 sections I will try to show that finding causes in history is certainly necessary, apparently simple, and almost impossible.

1. In our everyday life we take causation for granted. We know that a heavy night’s drinking will cause a hangover, that a critical remark about her dress sense will enrage your girl-friend. It always does, so we think we understand causation. We depend upon this knowledge for all our plans and projects. We put on a waterproof to keep out the rain, we buy a ticket in order to get into a concert. In hundreds of acts every day we achieve the desired effects by selecting and activating the correct cause. What could be simpler?

2. Yet the problem of what a cause really is remains a mystery. Let us take a very simple example. When I play billiards or snooker I strike a white ball with my cue. The ball rolls forward and in turn strikes a red ball, hitherto at rest, which then moves across the table and drops into a pocket. Why did the red ball move? Because it was struck with a certain force and at a certain angle by the white. It is a simple problem in mechanics: a sphere of radius r and mass m rolls across the table t with a velocity v and strikes the other ball at an angle of d degrees. Is not that a full description of the event? No, you say. You have not shown the cause. I saw one ball strike another, but I did not see anything else. Was the cause too small to see? Do I need a microscope? No, comes the reply. You cannot see, hear, taste, smell or feel a cause. Then how do I know it exists? You know that there was a cause; otherwise the second ball would not have moved.

This seems to bring us into deeper trouble. When we ask for evidence that a cause existed, the justification points to precisely what did not happen. What did happen showed no trace of a cause. Can what did not do better?

3. David Hume (1711-1776) was the first to point out the non-appearance of causes. He found the solution of the puzzle in our expectations. We have seen cause and effect conjoined so frequently (ball moving ball, key opening lock, etc) that we expect the one to follow from the other and so accustom ourselves to belief in the operation of something wholly unreal. Causation is thus reduced to a form of self-delusion.

In Hume’s own words:

4. Yet it is hard to believe Hume’s challenging remark. We simply do not accept that, for all we know, ‘anything may produce anything’. Our whole understanding of the world - science, technology, art, history - is based on our beliefs in the regularity of things. Water does not run uphill nor the sun set in the east. Moreover, we think we can distinguish regular connection from causation. Wednesday regularly follows Tuesday, but we do not think that Tuesday causes it. One way to find a cause is by comparing two similar events. Let us strike matches. On one occasion the match lights; on the other it does not. We examine all the conditions: phosphorus on the match, oxygen in the air, a brisk friction against a rough surface. We find that on the occasions when the match failed to light it had been rubbed on the smooth side of the box, not the rough. That, we conclude, was the cause of its failure to light. We shall come back to this when we consider causation in history.

5. For this we have to consider the human factor. The philosopher-historian, R.G. Collingwood, argues that the primary sense of ‘cause’ refers to what brings about a free human action. Russell agrees that it is ‘the analogy with human volition which makes the conception of cause such a fruitful source of fallacies’.

To cause a person to do something is to give them a motive for doing it. (Example: My wife’s illness caused me to send for the doctor.) But motivation does not guarantee the action. Circumstances may favour or may hinder it. Strictly, our only experience of causation is the experience of our own willed actions in response to some motivation. When we observe the motion of billiard balls we can imagine the first ball motivating the second, that is, giving it a cause to move. This is an illusion. Is not Hume right to insist that there is no evidence of causation in nature - only in human actions?

6. Bearing these arguments in mind, we turn to causation in history. The historian is required both to give an account of what happened in the past and to account for it. This leads to a consideration of causes - as Carr said.

In history men and women are doing or trying to do, things. Their intentions are helped or frustrated by circumstances. The circumstances are human and non-human. The first consist of other people’s thoughts, words and deeds. The second are the setting in which human actions may take place - for example, the weather, the soil, the presence or absence of tools, machines, buildings, weapons, means of transport and communication. These may assist us or hinder us in carrying out our aims.

Let us, by way of example, ask for the causes of the Italian renaissance. First, what was it? It was the coming together at a certain time and place of the actions of a number of scholars, financiers, poets, philosophers, scholars, painters, sculptors and architects. To explain their actions we need to know, first, what motivated them, and second, what were the circumstances that favoured or hindered their aims. These circumstances included the presence of Roman remains - literary, architectural, etc - in Italy and the invention of printing (both for); also the tendency to conflict among the Italian states and dependence on princely patronage (both against). Similar lists can be drawn up for other events. The chief thing is to distinguish the causes (what motivated the actors) from the surrounding circumstances. These only helped or hindered their intentions. Circumstances are often described as the conditions. To take another example: the cause of the First World War was the decision of various European governments to go to war. The favourable conditions included longstanding mutual distrust among nations and stockpiles of armaments. The unfavourable conditions included the fact that very few people actually wanted war, and the high cost and destructive power of modern weapons. It is clear that circumstances (or conditions) cannot themselves start a war or a renaissance; these require human intentions. But the circumstances can be powerfully influential on human action.

7. Unfortunately, this clear distinction between cause and conditions is frequently muddled. It comes about like this. When we wish to understand why something happened - a road accident, for example - we tend to look for that factor in the situation which, as it were, gives us a handle on it; something that we can alter at will. Among the conditions favouring a road accident include the existence of roads, the production of motor cars, the darkness at night, and many others. An accident enquiry will ignore these, for they are nearly always present. It is likely to pinpoint the cause as the presence of black ice at a dangerous corner. Now, strictly, black ice is just ice; it cannot bring about anything. But the cause in this case (a human desire to drive along that road) was normal and seen as unexceptionable. The best ‘handle’ on the situation is to deal with the black ice - perhaps by early gritting of the road. What is confusing is everyone saying that black ice was the cause, when it was only a non-human circumstance; in this case, one unfavourable to that action (driving) which was the true cause. To prohibit driving altogether in winter would certainly deal with the true cause, but that solution is not desirable. However, it is confusing to label as cause what was no more than a condition.

8. The historian does not seek a ‘handle’. Rather, as in No 4 above, he tries to isolate the significant factor. It is the unusual (a war, a revolution, a great discovery) that he seeks to account for. In doing so he looks for the unusual precedent. Why, after a thousand years, did a very unusual event occur in France in 1792: the French monarchy fell? Among the favourable circumstances were the ideas of the Enlightenment, the bankruptcy, the high price of food, the threat from neighbouring monarchs. All these, commonly cited as causes, were actually circumstances rather than moving causes. The true causes were the actions of Danton and other revolutionaries. So long as this confusion exists the debate about the French Revolution need have no end. Of course, to understand why the men and women in Paris at this time acted as they did is not easy. But to concentrate on them alone at least simplifies the question about the causes of the Revolution.

9. However, what most bedevils the historian’s search for causes is not the confusion between true cause and favourable circumstances. It is the assumption that he knows what did not happen. Most would agree with this definition: a cause in history is ‘an event, action or omission but for which the whole subsequent course of events would have been significantly different’, as the philosopher W.H. Walsh says. This, surely, is the sort of thing we seek when we look for causes in history. For example, the great French historian, Georges Lefebvre, wrote: ‘There would have been no French Revolution - such as actually took place - if the King, "handing in his resignation", had not convoked the Estates-General.’ This is called ‘a counter-factual condition’, because it runs counter to the facts. The fact is that he did convoke them. We cannot know what might otherwise have happened, for the simple reason that we cannot know what never existed. We can only guess. Logically, this is alarming: historians are professing an impossible knowledge. In practice it is not so alarming, for we do that sort of thing all the time. ‘If there had been no black ice there would have been no accident’, we say. Can we be certain of that? Of course not. We can only say ‘probably not’ about the accident. Human life is full of such guesses and statisticians calculate probabilities.

10. However, this is not altogether comforting for the historian. If we cannot know the non-existent, we can guess at it and put our guesses into numbers. Cars have taken that corner many thousand times without accident. On the one occasion there was black ice an accident occurred. That is enough for the ice to be seen as the cause. But what of Lefebvre? How many French revolutions have there been? Three or four at most. The numbers are statistically insignificant. In short, the historian has a far weaker case for his belief in the ‘might have beens’ of history than in ordinary life. When he flies to a conference in the United States he can smother any qualms by recalling that planes have crossed the Atlantic safely perhaps a million times. He cannot know his future safety, but his belief in it has a fairly sound base; something like a million to one. But the 1789 revolution was unprecedented. What basis had Lefebvre in his belief? Nothing statistical, but perhaps it was a shrewd guess.

I conclude by returning to my earlier remark that finding causes in history is necessary, apparently simple and nearly impossible. To find a cause in anything but human intention is an error. Causes in nature (billiard balls, chickens and eggs) do not exist. Moreover, the counter-factual guesses on which historians base their various guesses are also strictly non-existent. The true business of history is the study of the men and women who made it. As Marx said: ‘Men make their own history, but they do not make it under circumstances chosen by themselves.’ If you want to know why things happened in history, the best thing is to seek out the intentions of the people concerned and to examine the circumstances, favourable or unfavourable, in which they had to act.

Dr Michael Stanford is the author of The Nature of Historical Knowledge, Blackwell, 1986, A companion to the Study of History, Blackwell, 1994, and An Introduction to the Philosophy of History, Blackwell, 1998.