University of Helsinki

Political scientists have long been in a special position among the social scientists in having continuous conflicts and diverging opinions about the basis of their science, despite the surge and predominance of the "positivist" behavioral approach. This is due to the fact that political phenomena (e.g. those concerning democracy) necessarily also have philosophical aspects that steer the discussion to the basic issues of their science. This discussion, however, is too often isolated from the more general epistemological problems of the social sciences.¹

Perhaps the social scientist may feel uneasy when the discussion turns to the epistemological foundations of his science. "A serious economist hardly likes to be caught at the trivial occupation of discussing foundations," says I.M.D. Little. (1957, p. 4) Yet, at the present stage of the social sciences this kind of discussion seems relevant especially from the point of view of the methodology of these sciences.

The aim of this paper is 1o show that the presentday discussion about the methodology of the social sciences implies two different ideas of what amounts to a "social phenomenon" and what it is to "explain" or "understand" social phenomena. These two ideas, in turn, are part of two different conceptions of epistemology and philosophy in general.

An excellent introduction to what we mean by the epistemological problems in social sciences is Peter Winch's monograph, The Idea of a Social Science and its Relation to Philosophy (1965). The present paper is largely based on Winch's analysis of the basic ideas upon which the study of society is founded. Use is also made of Charles Taylor's, The Explanation of Behavior (1964), which contains a sharp criticism of the conceptual framework of the 'behaviourist' methodology as well as an analysis of the nature of teleological explanation. A third important work from the present point of view is G.E.M. Anscombe's Intention (1958), a study on the concept of intention, a concept that has been much discussed lately. An interesting historical background has been added to this discussion by the revived interest in Aristotle and Hegel.

Two Conceptions of Philosophy

Winch brings out clearly some of the philosophical problems of the social sciences. These problems are not 'philosophical* in the same sense as philosophy has generally been understood in the last few decades. The relationship between philosophy and empirical science has been understood in a way which Winch appropriately designates as "the underlabourer conception of philosophy", comparable to the medieval ancilla theologiae concept of the relationship between philosophy and theology. This conception was expressed already by John Locke:

"... in an age that produces such masters as the great Huygenius and the incomparable Mr. Newton, with some others of that strain, it is ambition enough to be employed as an under-labourer in clearing the ground a little, and removing some of the rubbish that lies in the way to knowledge." (Essay Concerning Human Understanding, Epistle to the Reader.)

The modern version of this conception can be perhaps expressed by saying that the task of philosophy is restricted to "eliminating linguistic confusions". Language, the scientist's necessary tool, has faults which handicap his work; the philosopher comes to his rescue as some kind of "language mechanic", who repairs with his analytical methods the faults found in the language. This view of philosophy — particularly in connection with the study of politics — is illustrated for example in T. D. Weldon's, The Vocabulary of Politics, An Enquiry into the use and abuse of language in the making of political theories (1955). Another good example of this is Margaret Macdonald's The Language of Political Theory (1955).

However, if we are not satisfied with this conception of philosophy (without contending value of the results it has produced), nor with the conception that a pr/o/7-speculation in philosophy can compensate for empirical research, what does actually remain?

Winch explains this "residue" as follows: "Whereas the scientist investigates the nature, causes and effects of particular real things and processes, the philosopher is concerned with the nature of reality as such and in general." (Winch, 1965, p. 8)

This is not a very good formulation because it leads to the assumption that in addition to all particular things there exists some universal essence that proceeds existence and must be revealed by philosophy. Perhaps a safer way to face these philosophical problems is to ask what is wrong with the 'underlabourer' conception of philosophy.

The difficulties in this conception lie in the assumption of clear and unequivocal separation between language (or the concepts expressed in the language) and reality. On the one hand, there is the "world of facts", reality and, on the other, man's description (his conception) of the world. For example, Weldon's aforementioned study is governed throughout by this separation.

Yet, no such simple separation can be made. Reality and our conception of

reality are in fact inseparably interwoven. An example of this is the so-called atomistic assumption, whereby reality is divided into separately identifiable "facts". Charles Taylor shows that this assumption — in itself a postulate on reality, although it seems to express reality's 'own' quality — is one condition in the concept of causality. "The notion is that the ultimate evidence for any laws we frame about the world is in the form of discrete units of information,

... each of which is separably identifiable from its connection with any of the others. Our knowledge of the world is built up from the experience between these units." (Taylor, 1965, p. 11) (He points out further that the Ideological explanation presupposes a partial rejection of this atomistic assumption.)

The 'intertwining' of our concepts and reality was expressed by Ludwig Wittgenstein in a lapidary way in his work, Tractatus Logico-Philosophicus (1922), using the unfortunate concept of "essence". "To give the essence of proposition means to give the essence of all description, therefore the essence of the world." (5.4711) Although Wittgenstein later partially rejected the cortceplion of language represented by his earlier work, even in his later work, Philosophical Investigations (1953) he clearly adheres to the idea of the 'inseparability' of language and the world.

This idea may be explicated in the following way. Let us suppose that we can peel off all the concepts which we actually have of the world around us. (This, of course, is impossible in practice.) What would be a typewriter, typing on paper, room, street, or a moving automobile on the street for me without the concepts which I have once learned or formulated? I'm inclined to say that the world without my conceDts of it would be for me an unintelligible mass of sense impressions without order or organization. To be sure, I no longer (as in my childhood) have to consciously exert myself to understand that this object is a typewriter (although my concept of a typewriter needs to be further refined); at this moment I want to understand what it is to understand the world. This understanding can be expressed by saying: the world is to me what is presented through my concepts.

If this proposition sounds too dramatic it can be toned down by saying that our concepts are like tenses through which we 'notice', 'comprehend', and 'understand' phenomena and their relationships. A good example of How a new concept like a lens helps us to notice and comprehend phenomena is the introduction of the concept of germ into the language of medicine.² Another example of this is the role of the concept of purpose (end) in Greek thinking. One could say (with certain reservations) that the Greeks 'noticed' and 'understood' only such phenomena to which the end concept was applicable. To other kinds of phenomena, they were more or less 'blind'.³

These brief remarks indicate how problematic the distinction is between language and reality. Understanding the problems involved in the concept of reality furnish so to speak a new lease on life for philosophy and, above all, for epistemology. The problem of what is true, what is knowledge, understanding, etc., are philosophical issues to which the sciences that study nature

and man do not answer, yet they imply that a certain answer has been given. Like a mole, philosophy bores into the labyrinths of the concepts implied by sciences in order to explicate what is meant within a given field of science by 'facts', 'phenomena', 'explanation', etc. This work is not without importance because from these notions — as we shall see particularly in the field of social sciences — depends the methodology of the science in question. The scientist's attitude towards this task is characterized by Winch in the following manner:

"Of course, this simply exasperates the experimental scientist — rightly so, from the point of view of his own aims and interests. But the force of the philosophical question cannot be grasped in terms of the preconceptions of experimental science. It cannot be answered by generalizing from particular instances, since a particular answer to the philosophical question is already implied in the acceptance of those instances as'real'." (Winch, 1965, p. 9).

Winch's understanding of the relationship between epistemology and science represents a new awakening to the problems which challenged Hegel and his students. (Very likely Winch himself would sharply dispute these roots of his ideas. On the basis of one of his remarks, he has the generally accepted one-sided view of Hegel as the culmination of a priori speculations.)⁴ The essential question of Hegel's epistemology was just the one developed by Winch: what is the relationship between concept and reality? From his earliest writings on, Hegel criticized the conception that the outer world (die Positivität) existed independent of man's (conceptual) activity. This is an illusion of common sense knowledge; it expresses man's alienation (Entäusserung, Entfremdung) from the results of his own conceptual activity.⁵ The outer world is essentially man's intellectual creation. Hegel's pompous grand term "absolute knowing' (das absolute Wissen) means awareness of the fact that man as a comprehending and thinking being constantly creates the outer world. But Hegel does not claim that the external world is merely the product of our comprehension; 'reality' is born when comprehension takes in that which is 'given' (das Gegebene). The 'given' is in the final analysis the content of our sense perceptions ⁴

One weakness in Hegel's philosophical architecture is that he never consistently developed its foundation, epistemology. Another weakness is Hegel's conviction that when the idea and that which is 'given' unite, this is 'objective world dialectics' and not merely the activity of a human mind.⁷ Nevertheless, those problems from which Hegel developed his epistemology are obviously again becoming timely.

Two Concepts of Human Behavior

The sciences that study human behavior presuppose — as was previously noted — some definite concept of human behavior. This concept implies what are the 'phenomena' or the 'facts' which these sciences study. At present a lively philosophical discussion is being carried on around the concept of human behavior (act, action).

Winch's basic remark on the concept of human behavior is that the conception of reality that we have learned or formulated is not separate from our behavior but belongs to it as its essential feature: "it is clear that men do decide how they shall behave on the basis of their view of what is the case in the world around them." (Winch, 1965, p. 21) A trivial example: We want to travel from place A to place B. We know that from A to B we can travel by train; we know where station A is situated and when the train leaves for B. On the 'basis' of this knowledge we act so that we catch our train on time.⁸ Another example from Winch: "A monk has certain characteristic social relations with his fellows and with people outside the monastery; but it would be impossible to give more than a superficial account of those relations without taking into account the religious ideas around which the monk's life revolves." (Winch, 1965, p. 23) Winch argues that to the description or explanation of an agent's social behavior necessarily belongs elucidation of the conception of reality he has learned or formulated and upon which his action is 'based'.

Winch's central thesis is that a person's social behavior is inseparably connected with his conception of reality and is to be understood only in terms of this conception; the development of this thesis has been the basis of all his analyses of the nature of social science. Before a closer examination of the particulars of this thesis and its limitations, let us review its historical background.

Winch does not seem to be aware that his thesis is a newly formulated Aristotelian (and Hegelian) theory about the relationship of knowledge and action. Aristotle's theory can be simplified by saying that the occasion (or cause) for an agent's action is the knowledge he has of the end the action, or that the action or resul: of the action expresses this knowledge. In his Metaphysics, Aristotle presents the following example: "The master builder has the idea of a house in his soul; he knows 'what being a house is'. In a way the house is born of the conception of the house; something immaterial gives occasion to something material." (VII, 7, 1032b 14) In medieval philosophy, Thomas Aquinas represents; essentially the same idea of the relationship between knowledge and action,' while in modern times we meet this idea in Hegelian psychology.¹⁰ To be sure, what Aristotle presents relatively clearly, later thinkers generally present more obscurely and in a way evasively. (Thus e.g. Winch speaks of 'the force of the concept of reality'.)

Winch's conception of the relationship between knowledge and action differs from Aristotle's, for example, in that he speaks principally of knowledge in terms of reality (also later about knowledge in terms of norms), while Aristotle speaks of knowledge in terms of purposes.¹¹ For some reason Winch seems to avoid the concepts of end and purpose. In a way Winch's thesis is amended by the earlier mentioned Anscombe study, 'Intention', which gives to the concept of intention the importance that belongs to it in the field of human behavior. But, Anscombe seems to be satisfied with rather ambiguous

expressions in explaining the relationship between intention-knowledge and
behavior.

Winch's view of the relationship between knowledge and behavior is clearly

a part of his epistemology. If we are of the opinion that "the world is for us, what is presented through our concepts", then we will also be inclined to think that our ideas and concepts are une quantité non-négligeable of our behavior, that our behavior in some sense is an expression or a realization of our ideas about reality. Therefore, the explanation of behavior must include those ideas, which the agent (Winch's use of the term) 'exercises' in his behaviour."

Although Winch is not aware of the historical background of his thesis, he does know that it conflicts with a certain, widely-held modern notion of behavior and explanation of behavior. This opposing thesis is expressed — in the field of social sciences — e.g. by Emile Durkheim:

"I consider extremely fruitful this idea that social life should be explained, not by the notions of those who participate in it, but by more profound causes which are unperceived by consciousness, and I think also that these causes are to be sought mainly in the manner according to which the associated individuals are grouped. Only in this way, it seems, can history become a science, and sociology exist." 13

Durkheim's unvoiced assumption is that the agent's notions are not an essential factor in his social behavior. These can be by-passed and his behavior explained "by more profound causes which are unperceived by consciousness"; these causes are in the social environment, "in the manner according to which the associated individuals are grouped". (The question arises, whether Durkheim does accept this theory as valid in the same way in relation to such a person as for example, a sociologist, who has become aware of those "profound causes".)

In order to elucidate the contrast between the above two ideas of social
science, we must examine certain viewpoints, which have been reflected in
recent discussions on the concept of behavior (action).

von Wright points out (1966, p. 14) that the concept of action has drawn little attention in modern philosophy. "But, if we return in time from Descartes to Thomas Aquinas, or to his master, Aristotle, the picture changes. Their thinking about action seems now uniquely modern..." I want to add to this list Hegel and Marx, both of whose thought on human action also seem so 'modern', perhaps for the reason that they have a common ancestry in the thinking of Aristotle.

This interest in the concept of action may be in part a reaction against that concept of behavior which has been developed from the conceptual apparatus of certain natural sciences. Based on a notion from the field of physics, human behavior is interpreted as mere motion, which can be explained by a conceptual apparatus analogous to the explanation of the movements of inanimate matter. One culmination of this endeavor is the movement of behaviorism, whose ideal it has been to reduce human activity to 'colorless motions'. (Already the term 'behavior' in the place of 'action' often reflects the tendency to interpret human activity in the manner of the natural sciences.)

Charles Taylor, whose study, Explanation of Behavior (1964), analyzes behaviorist concepts and theories, presents interesting comments on the consequences which follow when these are applied. The talk about freedom, choice, and responsibility loses its meaning, and thus collapses the foundation of the whole modern social structure, which is built upon these concepts.¹⁴ A comment about the notion of planning can be added here. According to a certain popular conception, the behaviorist study of man and society, which explains human behavior solely in terms of causal 'natural phenomena', then only makes possible efficient and rational social planning. This conception seems to be basedon an error in thinking. The idea that human activity is guided by concious (intentional) planning conflicts with consistent behavioristic conceptions. (One can perhaps be saved from this conflict through the aid of a 'practical compromise' by dividing society into two castes: on the one hand, there is a purposely acting group of planners, on the other, the causally behaving mass for whom plans are made. The idea of 'social engineering' which belongs to the belief that man finally is gaining the same control over human behavior that he has over natural phenomena, seems to be based unconsciously on two different concepts of human behavior.

Taylor's analysis — although we cannot go into its details here — shows, however, that modern behaviorist concepts and theories cannot stand up to detailed critical examination. As to the usual claim that the behaviorial sciences are in their infancy, Taylor remarks that "it begs the question. It may be that these theories are in their 'infancy' precisely because there is a fatal obstacle to their growing up, viz., that they are incorrect. The 'Galilean spirit' has been around in psychology for quite some time, and, if it hasn't produced anything very solid in experimental psychology, this may be because current approaches are wrong." At this time, according to Ta.ylor, there are grounds for the 'rational belief that animate organisms exhibite characteristics which can only be understood through concepts of intention and purpose. (Taylor, 1965, pp. 272—273)

von Wright, too aays that the causal explanation has not as yet displacod the teleological explanation, which uses purpose and end concepts. E>ut, he continues, the same phenomenon cannot be simultaneously explained as causal and teleological, and when there is 'competition' between the two, the causal explanation always 'wins' over the teleological one. (von Wright, 1966, p. 22) This claim must be more closely investigated, for according to it there will be a gradual abandonment of the (intentional) action concept.

According to von Wright, the following essential difference exists between causality and teleology. A causal relationship is characterized by the fact that it must be possible to observe and describe causes and events separately and independently of each other. But the relationship between purpose (or motive) and behavior is conceptual or logical. For the criteria of determining the purpose of some behavior are the traits of that behavior. "It would be an error in thinking to assume that something could be the causal reason of the occuring of its own criteria. ' (von Wright, 1966, pp. 17—19)

We can see that a statement about a teleological relationship cannot be 'translated" as a causal relationship, where "purpose" is the cause for behavior. On the basis of Taylor's analysis (1965, p. 12) it is at least very questionable whether such a "translation" can be done. But is a teleological relationship 'only' a logical one? We can point to certain conclusions reached by Anscombe in the previously mentioned study on the concept of intention. According to Anscombe we have two kinds of knowledge about our behavior: "knowledge by observation" and "knowledge in intention". (In forming the latter concept of knowledge, Anscombe notices that he has only reformulated the 'practical knowledge' concept of Aristotle.) This distinction means that we have - regardless of outwardly observed knowledge of our behavior - direct knowledge of our intentions. (Anscombe, 1958, pp. 49-57)" It seems that Anscombe's observations create a certain doubt about the proposition that the teleological relationship is merely a logical one. Neither can it be said — as both Taylor and Anscombe agree on this — that it is a causal relationship in which two clearly separately indentifiable facts are related. It seems that the character of the teleological relationship is for the time being insufficiently explicated. Apparently, we are so much accustomed to thinking in terms of causality that we find it very difficult to think that possibly there are other relationships, and it's even more difficult to form — particularly in the light of the previously mentioned atomistic assumption — an exact idea of such relationships. It may be argued that the teleological relationship is by its nature an expression relationship. The idea of the expression relationship originates from the understanding of the relationship between speech and thought: we say that a word or sentence (more or less, for better or worse) expresses a thought, we do not say that the thought causes the sentence, nor do we claim that speaking and thinking are the same thing.¹⁴ We understand another person's speech when we are able to associate it with the same thought as the one he is trying to express. Perhaps essentially the same way we understand another person's behavior. The agent may himself be conscious or not conscious of the purpose of his action; in the former case, he may also orally express to us his knowledge of this purpose. There is nothing mystical, 'going inside the other' in understanding action and speech in this way. It presupposes that we belong with the agent to the same cultural group, that we have learned to form the same concepts and the same rules of language expression. Winch affirms this in saying that "the concepts in terms of which we understand our own mental processes and behavior have to be learned, and must, theorefore, be socially established, just as much as the concepts in terms of which we come to understand the behavior of other people." (Winch, 1965, p. 119)

The idea of expression relationship is clearly related to the Aristotelian,
Thomistic, and Hegelian idea of 'essence and its realization'.

All in all we can note that the question of the relation of purpose to
behavior — at the same time that it has attracted keen philosophical interest

- is at the moment unclear and controversial. Two essentially different concepts of human behavior compete in the marketplace, and in addition we meet their various modifications. Ambiguity in these philosophical questions makes it uncertain to which direction the methodology of social sciences or generally of the» so-called behavioral sciences are to be developed. That the majority of the rocearchers in these sciences seem at the moment to hold up as their ideal in line with the natural sciences, does not give us a correct picture of the uncertainty which prevails on the philosophical level.

It seems to me that between the two competing concepts of behavior (and explanation) a state of 'peaceful coexistence' can be established. This presumes that both sides renounce 'imperialistic objectives' or the idea of 'world revolution'. The peace treaty will state that there are phenomena in human behavior whose adequate explanation is teleological, and phenomena whose adequate explanation is causal. At this moment, this kind of imaginary treaty (I don't, of course, imagine that will actually ever be signed) is possible only on the basis of a 'rational belief.

Two Ideas of the Social Sciences

The foregoing has characterized two different concepts of human behavior. (I have not separated social behavior from other kinds of behavior).¹⁷ My claim is that these two fundamental concepts imply two different types of explanation (teleological and causal) and furthermore two different methodological attitudes.

These two methodological approaches can be characterized by saying that the one which uses the teleological explanation or some variation of it tries to find out primarily what the agent's own ideas about the meaning of his actions are; the other approach, which uses the causal explanation or some variation of it, tends to by-pass the agent's own concepts about the meaning of his behavior and to explain it (in the words of Durkheim) "by more profound causes", i.e., by using the researcher's concepts.

One can scarcely point to any significant name in the history of social sciences who consistently represents one of these lines of thought or the other. (Max Weber would be, as Winch shows,' an inadequate example of the former approach, and, Karl Marx, an inadequate example of the latter approach, as we shall

see later.) A distinction between them, however, is fruitful from the point of
view of a methodological discussion.

The characteristics of the latter approach are presented clearly in the above quotation from Durkheim. Another classicist, M. Ostrogorski, point in the same direction; he emphasizes (1903) the similarity of social research with that of natural science and the researcher's outside 'observer' role to that which he is studying. This attitude is also reflected in the acceptance of the language of the natural scientist.

One of the most consistent efforts to replace the agents' concepts of their own actions with those of the researcher is Vilfredo Pareto's main work, Trattato di sociologia generate (1916; English title, The Mind and Society, 1935). He states that those ideas by which different societies understand and explain their activities differ greatly, although the activities themselves often are carried out in the same way. This statement is reflected in his famous distinction of 'residues' and 'derivations'. Those forms of behavior which seem to be relatively the same in different societies, he calls 'residues', and those which differ he calls 'derivations'. In the latter group belong ideas and theories by which people understand and explain their behavior. It represents "the work of the mind in accounting for (the residues). That is why (it) is much more variable, as reflecting the play of imagination." (Pareto, 1935, sect. 850) (Pareto's derivation-concept has obvious similarities with Marx's ideology-concept.) For example, Pareto points out (1935, sect. 863) that the ritual of baptism is the same in different cultures, but the ideas and beliefs "accounting for it" are different.

It is surely correct to say in the manner of Pareto that one task of the social science is to find common features of different societies susceptible to scientific generalization. But an essential philosophical question is how do we get the criteria of sameness. When we are dealing with purely physical phenomena, it is plain that the researcher forms the criteria of sameness or relevant differences. But when the question concerns human social behavior, the problem is not so clear and simple. If, like Pareto, we by-pass the agents' own concepts of the meaning of their behavior (and with these concepts their own criteria of sameness and dissimilarity), we adopt consciously or unconsciously the standpoint that these concepts have no essential significance for behavior - that they are in a sense 'unnecessary appendages' to actual behavior. At the same time we separate behavior from the social context into which the agent's own concepts place it. This standpoint and its consequences are clarified by two examples suggested by Erik Allardt:

"Suppose that we find that the outward forms, gestures, interaction frequency, motions, etc, are exactly the same in a game played by Scandinavian children and a religious ceremony performed among bushmen. It is still unreasonable to maintain that we are concerned with one and the same social behavior. We come to this conclusion because the ideas (conceptions) and rules of the acting individuals are quite different in meaning and their respective social relation. Let us take another example: When deciding whether a certain act is communistic, we consider the communist ideology and the meaning of communist ideas and rules in a communist environment. It may be possible to find some outward resemblance between, for instance, drinking groups and communist cells, but it is obvious that we then fail to understand something of the essentials of the communist cells." (Allardt, 1960, pp. 4-5)

Allardt's examples, or better, his interpretation refer to an idea of a social science diametrically opposite to that of Pareto. We can now return to Winch's idea of the nature of social science. He brings out the difference between the 'epistemological situations' of the natural scientist and the social scientist respectively as follows:

All scientific research requires a conceptual framework. When you notice something you identify its characteristics and this presupposes that you have concepts of these characteristics and that you have learned the rule how the concepts are used. The latter condition will bring you together with your fellow-scientists. The scientist develops and uses his concepts, on the one hand, in relation to the field of reality which he is trying to understand and, on the other hand, in his relationship with his fellow-scientists, "in participation with whom" he uses his concepts. (Winch, 1965, pp. 85-86)

"The concepts and criteria according to which the sociologist judges that, in two situations, the same thing has happened, or the same action performed, must be understood in relation to the rules governing sociological investigation. But here we run against a difficulty for whereas in the case of the natural scientist we have to deal with only one set of rules, namely those governing the scientist's investigation itself, here what the sociologist is studying, as well as his study of it, is a human activity and is therefore carried on according to rules. And it is these rules, rather than those which govern the sociologist's investigation, which specify what is to count as 'doing the same kind of thing¹ in relation to that kind of activity." (Winch, 1965, p. 87)

Here, I think, the difference in the 'epistemological situations' of the natural scientist and the social scientist is well exposed. It is worth-while to point out that the hidden assumption in this distinction is Winch's concept of meaningful behavior. One may also note that Winch here as well as elsewhere understands human behavior only as 'following a rule', although it would seem that often it is more natural to utilize the idea 'realization of a purpose (end, intention)' instead. A more serious objection to Winch is that he seems to assume that all human behavior is — in the aforementioned sense — meaningful. He seems to completely close out the concept of causality from the social sciences. It is not necessary to take such a resolute stand, although it can be recognized that Winch's characterization of the fundamental situation in which the social scientist finds himself is correct. It is also impossible to deny that the causal explanation (for example in the form of political ecology) and 'Durkheim's method' have proved fruitful for the social sciences; it deserves criticism only when it behaves 'imperialistically' and tries to rule the whole field of social research. (I: can be claimed that prevailing sociological thinking also influences human behavior; if man is seen merely as a 'reacting' being, who in certain circumstances always reacts in a certain way, he perhaps will behave in the predicted way more and more; if again, he is seen by nature to be an 'active being' who acts according to his own purposes, perhaps this characteristic is strengthened in his behavior.)

Winch offers a simple example from the sociology of religion, to illustrate his methodological approach. The Bible tells that a Pharisee and a Publican were praying. The former said, "I thank Thee, Lord, that I am not like other people"; the latter said, "Oh God, have mercy on a sinner

such as I". Did they both do the same thing? The answer depends on what is involved in the idea of praying, which is a religious question. The criteria which we use here to determine whether these two acts are the same, belong in the first place to religion, not to the sociology of religion. (Winch, 1965, p. 87) The relationship of the sociologist of religion to those who practice religious activity cannot be merely an observer's relationship to that which is being observed (as the natural scientist's relation is to natural phenomena); it can be said that he is in the first place in the position of a student in relation to that form of life which he is investigating. If we have a tendency to feel 'superior' towards those who practice religion, let us think about another case. We are investigating mathematical activity in some culture. What can we say about this activity, unless we first learn its rules from those who work in it. As a general rule-of-the-thumb, we can hold that the more 'intellectual' the activity which we study, the more we must take on the role of a student. It is clear that as scientists we are not bound to be satisfied with the ideas presented by the agents (nor need we believe in them in their way). Our task is to explicitly bring out the actually working system of values and norms. This is creative scientific activity; it has been demonstrated (Arrow, 1951) that there is no mechanical way of adding up individuals' different preferences in some society into a unified societal value system.

Perhaps the most significant methodological consequence of Winch's social science theory is that the reseacher must allow his object of study to speak first and only then and on this basis to develop his own concepts. Here it is not possible to undertake a detailed analysis of research methodology, but it can be argued that the order in certain existing practices is just the opposite: the scientist starts from (e.g. interview research) how those who are studied react to his ideas. Winch, in an answer to one of the criticisms of his book, stresses that his purpose was not to present a methodology of the social sciences: "any methodological inquiry would have to pay much more attention to the considerable differences between the procedures appropriate to different kinds of social study than I had deemed relevant to my philosophical purposes." (Winch, 1964, p. 202)

It may still be interesting to compare Marx's concept of human action with that of Winch. According to a certain popular view, Marx, if anyone, was a 'causality—oriented' thinker who tried to explain human behavior as totally determined by the economic and social environment. It is true that, e.g., Marx's concept of 'ideology' expresses this part of his thinking, and this has undoubtedly given a strong stimulus to that recearch method which presupposes as irrelevant man's own ideas of the meaning of his actions. But stressing this side of his thinking will give an erroneous view of Marx's philosophy. His thinking has also been influenced by the teleological mode of thought (inherited from Hegel); this is clearly evidenced in his way of using the concept of work:

"Wir unterstellen die Arbeit in einer Form, worin sie dem Menschen ausschliesslich gehört. Eine Spinne verrichtet Operationen, die denen des Webers ahneln, und eine Biene beschämt durch den Bau ihrer Wachszellen manchen menschlichen Baumeister. Was aber von vornherein den schlechtesten Baumeister vor der besten Biene auszeichnet, Ist, dass er die Zelle in seinem Kopf gebaut hat, bevor er sie in Wachs baut. Am Ende des Arbeitsprozesses kommt ein Resultat heraus, das beim Beginn desselben schon in der Vorstell ung des Arbeiters, also schon ideell vorhanden war. Nicht dass er nur eine Formevarenderung des Naturlichen bewirkt; er verwirklicht im Naturlichen zugleich seinen Zweck, den er w e i s s, der die Art und Weise seines Tuns als Gesetz bestimmt und dem er seinen Wil len unterordnen muss." (1932, p. 186.)

This is only another way of expressing the Aristotelian concept of the nature of human activity. (Marx even uses — by coincidence? — the same example of the builder as Aristotle.) The understanding of action is the understanding of the purpose which the action realizes. It is not possible here to examine more closely the 'dialectical' relationship, which Marx saw as existing between man's conscious and purposeful activity and its economic determination.¹' In recent discussions about Marx this teleological aspect of his thinking has come out more and more; this is a significant part of the revival of 'Aristotelian' thinking which seems to be presently taking place.

Returning again to Winch: the doubts, which he presents about the existing social science modes of thinking may add to the social scientist's 'self-consciousness', make him conscious of his basic concepts and habits of thought. On the basis of this kind of philosophical analysis he can make his methodological choices. If he selects the road Winch suggests by basing his investigation on the conceptions that are those of the object of study, we can expect that his results show the differences which exist between different societies and forms of life. It is not the task of the student of sociology and politics to try to advocate the acceptance of those differences; but I believe that this 'accepting attitude', whose necessity in the modern world is obvious, would be a consequence of the results of this kind of research. Neither, can anyone forbid us from using different modes of causal explanation (and aiming at a general 'globally valid' theory of social behavior.) But then we must be aware that using this conceptual apparatus we can hardly expect to explain all behavior — at least not our own scientific activity.

---

NOTES

1 For a discussion centering around these problems see, e.g. Herbert J. Storing, ed., Essays on the Scientific Study o/ Politics (New York: Holt, Rinehart, & Winston, 1 flfiOX *-.»•% *4 4-V»rt ni * Vv« Am i j-tv-t 4- v-hStl J3.XJ1. i JXJi

2) 1962) subsequent polemics.

3 See Hintikka. 1964, p. 62.

4 Winch's connection with Hegelian modes of thought comes perhaps through R. G. Collingwood, whose work he seems to value greatly. (The title of one of Collingwood's works is The Idea

5 From this Hegelian alienation concept (to become estranged from the results of 'conceptual labor'), Marx was influenced into his own alienation concept (to become estranged from the products of 'concrete labor*). See Lukacs, 1948, p. 680.

6 Hegel nowhere developed a unified theory of knowledge. Some of his central ideas are in the work Phänomenologie des Geist es (1807).

7 This is remarked on by Rudolf Steiner in his work Wahrheit und Wissenschaft

8 This example has been slightly changed from Winch's example. p. 21.)

9 Roa TVinmao Arminac Siimma TViAnlnuifa T_—TTnp nn VTTT_—XVTI.

10 See. e.g.. J. V. Snellman. Lärobok i Psvknlnei Rtnrkhnlnv 7. TT 1R37

11 See Hintikka. 1964. D. 66.

12 See Winch. 1964. n. 207.

13 Durkheim's review of A. Labrlola: "Essais sur la conception materiallste de 1'histoire" in Revue Philasnnhiaue. D*v»p>mh*»r 1R07 ffMtpH hv "Winrh 1QRR nn 23—24.^

14 Spe von

15 This statement is illustrated Conrad Aiken's famous novelulse.

16 Cf. Wittgenstein. 1958. DD. 217-218. 224.

17 See e.g. Allardt, 1962, p. 2.

18 •See Winch. 1965. DO. 111-120.

19 See Lukåcs, 1948, p. 431; Hintikka, 1964, pp. 80-81.

References

Allardt, Erik, Sosiologia, Research Reports from the Institute of Sociology,
Univ. of Helsinki no 12, 19(52. (Mimeo)

Allardt, Erik, Synpunkter pä teoribildningen inom sosiologin, Research Reports
from the Institute of Sociology, Univ. of Helsinki no 3, 1960. (Mimeo)

Anscombe, G.E.M., Intention. Oxford: Blackwell, 1958.

Arrow, Kenneth J., Social Choice and Individual Values, New York: John
Wiley & Sons, 1951.

Hintikka, Jaakko, "Päämäärä, sattuma ja välttämättömyys, Eräiden kreikkalalsten
ajatustapojen tarkastelua" (Purpose, Chance, and Necessity, Some ways
of thought in Ancient Greece), Ajatus XXVI (1964), 61-81.

Little, I. M. D., A Critique of Welfare Economics. 2 nd ed. Oxford: Clarendon
Press, 1957.

Lukåcs, Georg, Der /unge Hegel, Ueber die Beziehungen von Dialektik und
Oekonomle. Wien: Europa Verlag, 1948.

Macdonald, Margaret, "The Language of Political Theory", in Logic and
Language, First Series. Ed. by A.G.N. Flew, Oxford: Basil Blackwell, 1955.

Marx, Karl, Das Kapital, l. Hrsg. vom Institut fur Marxismus-Leninismus, Moskau,
1932.

Ostrogorski, M., La democratie et I'organisation des partis politiques. Paris:
Caiman-Levy, 1903.

Pareto, Vilfredo, The Mind and Society. Transl. by A. Bongiorno. New York:
Harcourt, Brace, 1935.

Steiner, Rudolf, Wahrheit und Wissenschaft. 4 th ed. Dornach: Verlag der
Rudolf Steiner-Nachlassverwaltung, 1958.

Taylor, Charles, The Explanation of Behaviour. London: Routledge & Kegan
Paul, 1965.

Weldon, T. D., The Vocabulary of Politics, An enquiry into the use and abuse
of language in the making of political theories. Harmondsworth: Penguin Books,
1955.

Winch, Peter, The Idea of a Social Science and its Relation to Philosophy.
4 th imp. London: Routledge & Kegan Paul, 1965.

Winch, Peter, "Mr. Louch's Idea of a Social Science", Inquiry 9 (Summer
1964), 202-208.

Wittgenstein, Ludwig, Tractatus logico-philosophicus. London, 1922.

von Wright, Georg Henrik, "Käyttäytymisen selittämisestä" (On Explanation of
Behavior), Psykologia, 1 (1966), 11-23.