Methodological Individualism

I recently dug out for an American friend a
section I had written for my Principles of
, but decided not to use. I publish
it here, in case it is of use to anyone else.
I have not checked the references.

In giving our account of groups, communities and the State, we have made great use of general terms: we have talked of arbitrators, judges, rulers, subjects, policemen, litigants; and the even more general term official. in a sense there is no difficulty in using these words: we all understand what we mean by a policeman; we all know one when we see one. Nevertheless there is a certain philosophic disquiet about the use of general terms in political theory, or, for that matter in history, or in the social sciences. After all, it is felt, cabinet ministers, Judges, officials of all sorts, are persons, individual persons, each with his own idiosyncrasies: in order to understand any one of then as an official, we need to understand him as an individual: and if we understand him as an individual, there is nothing more to know about him. History is about people, particular persons, what they said and felt and did, not about bloodless abstractions or paper officials: so too sociology; so too politics.

No one need deny the importance of personality in politics, sociology or history. Nevertheless the thesis of Methodological Individualism, 1 as it is called, that we shall be concerned with persons as persons and nothing else is seriously misleading, if really believed and really carried through.

The thesis consists of two parts: first that we cannot usefully talk of rulers, Judges, officials as such, because the holders of offices are always individual persons, and it is the individual person who decides, not the abstract idea of an official. Secondly that we can give a complete description in terms of individual persons, so that there is nothing left for anything phrased in terms of officials to describe.

The first part of the thesis has some truth in it, but fails to be completely true because of the positive aspect of conditions B(i) and B(11) of Human Nature. Men are to some extent rational, to some extent unselfish. A man can act in a rational and unselfish way, and to that extent in the same way as any other man acting rationally and unselfishly. Judges give, to a very large extent, the same decisions. Solicitors and barristers predict what the Courts will decide, not what Mr Justice Pennfarthing will decide. In the vast majority of cases--- the ones we do not hear about, because the outcome is so obvious that they never come to court--- it would not make any difference which Judge heard the case. It is only in cases which are unclear that the preconceptions, the idiosyncrasies, or the prejudices of the individual Judge will make any difference at all. And even then, thanks to the system of appeals, the latitude for individual enterprise on the part of any particular Judge is severely limited. Hence, for a large number of questions we do not need to look behind the wig and robe of the Judge to the man who wears them. And similarly with any other officials on a large number of occasions. Men are rational: they often agree: different individuals reach the same conclusions: and if our chief interest is in the conclusion, it does not matter which individual it was that reached it. But this is not always so. Men are not always rational, not always unselfish; and even when trying disinterestedly to reach the right conclusion, they sometimes disagree. We cannot expect that in all cases different officials will give the same decisions, or that their differences of opinion will not reflect differences of personality. And where the discretion is wide, and the assignment of responsibility not very definite---as in high political office---the influence of personality may be considerable, sometimes of decisive importance. Thus, while it is false to maintain that it is never enough to consider an official just simply as an official, it is true that it is not always enough to do so; and especially with overtly political offices. We need not often peer beneath the wig, but often must consider the man who wears the crown.

We have shown that an account of a community in general terms, such as ruler, Judge, official, is sometimes, but not always adequate. We now show that an account in particular terms, i.e. those of individual persons, is necessarily inadequate. This follows at once from what we have already said about the partial rationality and partial unselfishness of men. Men sometimes act not in accordance with whim, or in pursuit of their own interests, or idiosyncratically, but seriously, unselfishly and sensibly: and therefore, to understand their actions, we need to know not what is peculiar to their personalities, but what is common; in particular, if they hold official positions they will attempt to guide those of their actions which they undertake in an official capacity by rational and universal considerations which are intended to be general in their force, and not peculiar to one man rather than another. We cannot understand a Judge's actions in court if we discount the fact that he sees himself as a Judge, and that he is attempting to dismiss from his mind all personal considerations of advantage or interest, and is attempting to reach his decisions in a judicial frame of mind---in the same way as all his brother judges are. Although Judges are only human and are sometimes biassed, consciously or unconsciously, and therefore we may have to consider the personality of a particular Judge as well as the judicial arguments and ratiocination, nonetheless Judges are trying to be unbiassed, and to some extent succeeding. It is, indeed, naive to assume that they never fail: but it is ludicrous to neglect the fact that they do try, and seem sometimes to succeed. But if we are to give an account of what it is a Judge is trying to do, when, acting in a judicial capacity he reaches conclusions in a judicial frame of mind, we must have recourse to general terms, the idea of a Judge, what it is he thinks he is doing, the role that he is playing in his own and other people's eyes, as a Judge.

There may remain some residual discomfort. The historian in particular may feel that when he has recounted what each individual person has done and said and felt, there cannot be anything left still to say unless we believe in a Hegelian Geist or super-personal State, which acts over and above, and quite independently of, the individual human beings who happen to be members of the State.

Thus formulated, Methodological Individualism, with its reductionist message of ontological economy, still has some appeal; but a meretricious one. For the alternatives offered are not the only alternatives available, and we do not have to embrace Hegelian metaphysics as the only escape from a Namierite position of logical atomism.

If we are challenged to say what has been left out of an absolutely full account of what every individual person did or said or felt, we point to the subjunctive mood. To understand what people have done or did do or were doing, we need to know not only a bare description of their actual behaviour, but what else they would have done, had circumstances been slightly different. It is built into our concept of an action, that in ascribing actions to people we are not simply describing their overt behaviour, but also making claims about the corrections they would have made in their overt behaviour to make allowance for slight variations in conditions. This is what is meant in calling actions purposeful: it is clearest in those examples of actions which are literally goal-directed or ``homeostatlc''; the assassin who plunges his dagger into the victim's heart adjusts all his movements to those of his victim: it is this that distinguishes murder from the accidental running through that might occur if two friends were fooling about with daggers. Or, in a more modern vein, if a man shoots into a bush and kills a gamekeeper, it may be highly culpable, but is not murder: if he follows the man with his gun, and aims off to allow for the wind, etc., then his killing the man was clearly an action of his.

Thus, whenever we describe particular men as having acted in any way, we have not merely described their actual overt behaviour, but said something about their possible hypothetical behaviour. Even the simplest and ontologically most economical account of what particular persons did in a particular place on a particular occasion is not perfectly categorical, not entirely indicative in its mood. And when we come to give illuminating accounts of what people did, which attempt to explain why they did so, and are not bare catalogues of disconnected deeds, we go much more into depth, and become far more committed to the subjunctive mood. To say that Hannibal did not capture Rome because he did not march on it is to say that if he had marched on it he would have captured it. But when we are this far involved with subjunctive hypotheticals (or contrary-to-fact conditionals or counter-factual conditionals, as they are often called) we have lost all pretence of down-to-earth categorical indicative particularity. Subjunctive hypotheticals define ``open'' classes rather than ``closed'' classes, as the logicians call then. 2

Thus, while it is true to maintain that when we say ``Britain went to war with Germany in 1939'' we are talking about Britons and Germans, not some super-persons ``Britain'' or ``The British Nation'' or ``Deutschland'' or ``Die Deutsche Volk'', it is not true that we could replace the words `Britain', `British', `Briton' etc. with a list of 48 million names, Tom, Dick, Harry, etc. For the list leaves out important ``counterfactual facts'', such as if Tom and Marlene had had a son, he too would have been on the list, and that if Dick had emigrated to America, he would no longer have been on the list. Sir Winston Churchill, and the people whom he addressed and who responded to him, were the British Nation: but they were so not as a heterogeneous collection of individuals, but because they shared a common history, a common language, a common heritage, common sentiments, common institutions, common values. It is in virtue of these that Tom and Marlene's son, if they had had one, would have been in the list too, and it would have been by renouncing these and swearing allegiance to their American counterparts, that Dick, had he become an American citizen, would have ceased to figure in the British list.

In writing intelligible history we need to say not only what actually happened but something about what would have happened if circumstances had been slightly different. This is not to engage in wild speculations on big historical `if's, but only to give our account enough depth to bring the narrative into relief. If Tom and Marlene had had a son, or if Dick had gone to America, would it have made much difference? No: then we do not say that the story is particularly about them. If Sir Winston Churchill had died in 1939 or early 1940, or at any time before July 1945, that would have made some difference, perhaps a crucial one: therefore we need to mention him in particular. But with most people, in most historical accounts, for most historical purposes, the sudden death of one of the actors. even one in the title role, will not make a great difference to the outcome. If the Ambassador had had a heart attack half an hour before he was due to deliver his country's declaration of war, the war would have been declared nonetheless. If the Judge had had a stroke during the concluding speeches of counsel, there would have to be a new trial, but quite likely with the same result. And therefore when relating, the history of the war, or of the development of legal thought, we are interested in what the Ambassador did, or what the Judge decided, not what Sir Henry Fairweather did, or Mr. Justice Pennyfarthing said: for if Sir Henry Fairweather had not been the Ambassador, it would have been not his actions but those of whoever was Ambassador, which would have constituted the declaration of war, and if the case had come before a different Judge, it would have been that different Judge's decision which would have been cited in the legal textbooks.

In describing the activities of a Judge, it was necessary to say that he decided cases in a judicial frame of mind---itself a fairly abstract and non-particular concept. The same is true of all officials. Something comparable is true of many non-officials as well: they may not ratiocinate in a special frame of mind, but they see themselves as filling a certain role, and we cannot describe what it is that a particular man is doing unless we can describe the general type of man he is trying to be. The business man who sees himself as forceful and effective, the secretary who imagines she is destined to be a femme fatale, the workman who thinks of himself as downtrodden and exploited---they may not be rational, but to understand them or to or to male them intelligible, we must be able to enter into the private images that each projects of himself to himself. And even private images are too rich an ontological pabulum for the strict Methodological Individualist to allow himself to indulge in.

Even when an individual performs his part in the life of the community without thinking of himself in any special way, we often need general terms, referring to the community at large, in order to describe the particular activities of a particular individual. The postman or the grocer may not have ideals to which they are trying to conform, their patterns of behaviour: but their patterns of behaviour themselves remain unintelligible if we omit to say that they are delivering letters or selling groceries not to a certain ``closed'' class of people but to a chance selection of the public---an ``open'' class. A shopkeeper sets out to sell to any person who wants his goods and is prepared to pay his price. Merely to list the particular persons who happen to have been customers of a grocer in a particular week would be to miss out an essential part of the grocer's role, namely that if a hiker had come to that village, he could have bought two Oxo cubes, cheese and biscuits at the shop. To give a list of all the letters posted over a period, and from whom they were posted and to whom they were delivered, might be interesting: but the point of the postal service again would be missed unless it was further stated that any person could send a letter through the post to any other person he liked. And this, again, is something that the strict Methodological Individualist is debarred from saying.

We can sympathize with the Methodological Individualist, just as we did sympathize with Hobbes. 3 But just as Hobbes failed in his attempt to give a coherent account of sovereignty entirely in terms of persons who could be pointed at, and not of rules or principles or procedures which cannot be pointed at, so the Methodological Individualist fails to provide an intelligible account of social phenomena entirely in terms of particular persons, who are ontologically respectable, and not in terms of roles, official positions, ideals or images which are, to the purist, ontologically suspect. Whatever the attractions of nominalism to the metaphysician, they must be eschewed by the political philosopher, the historian and the social scientist. For society is covertly universal: communities, as the name implies, consist of all those people who have something in common, and who therefore, share some general characterization, in virtue of which they are members of the community. It is therefore impossible to talk about communities, or members of communities without using our stock of general terms defined with reference to some community or group. And so of men: for man is a social animal.

1. There has been much recent discussion: the issues at stake seem often to have been confused. The reader may like to follow up these references: A.J.Ayer, Language, Truth and Logic, 2nd ed., p.63.
K.R.Popper, The Poverty of Historicism, para.29-31.
J.W.N.Watkins, ``Methodological Individualism'', British Journal for the Philosophy of Science, III, 1952-1953, p.186.
May Brodbeck, ``On the Philosophy of the Social Sciences'', Philosophy of Science, 21, 1954, pp.140-56.
M.Mandelbaum, ``Societal Facts'', British Journal for the Philosophy of Science, 1955, 6.
E.A.Gellner, ``Explanation in History'', Proceedings of the Aristotelian Society, Supplementary Volume, XXX, 1956, p.157.
M.Ginsberg, ``The Individual and Society'', in On the Diversity of Morals, London, 1957.
L.J.Goldstein, ``The Inadequacy of the Principle of Methodlogical Individualism,'', Journal of Philosophy, 53, 1956, pp.801-813.
J.W.N.Watkins, ``Historical Explanation in the Social Sciences'', British Journal for the Philosophy of Science, VIII, 1957-1958, p.10.
M.Mandelbaum, ``Societal Laws'', British Journal for the Philosophy of Science, VIII, 1957-1958, p.211
L.J.Goldstein, ``The Two Theses of Methodological Individualism,'', British Journal for the Philosophy of Science, 9, 1958-1959, p.1.
J.W.N.Watkins, ``The Two Theses of Methodological Individualism'', British Journal for the Philosophy of Science, IX, 1958-1959, p.319.
L.J.Goldstein, ``Mr Watkins on the Two Theses'', British Journal for the Philosophy of Science, X, 1959-1960, p.240.
J.W.N.Watkins, ``Third Reply to Mr Goldstein'', British Journal for the Philosophy of Science, X, 1959-1960, p.242.
2. `All ravens are black' (about the open class of all ravens) means not only `'if anything is a raven, it is black' but if anything were a raven, it would be black': `All the engines at Bletchley have engine numbers that are prime means only 'if a number is the number of an engine at Bletchley, it is a prime number' but not `if a number were the number of an engine at Bletchley,, it would be a prime number'. The class of engines, and hence of engine numbers, at Bletchley is a "closed" class: we could replace it by a list of particular engines, or of particular engine numbers, without having to specify any class in general terms. A closed class to a class about which a Methodological Individualist can say all that there is to say.
3. Section 9 (of Principles of Politics).

Return to bibliography

Return to home page