ABSTRACT
Albert rejects as unrealistic. Albert does not, however, provide us with an interesting alternative, realistic approach to institutions. In order to develop and defend his own, updated version of the traditional approach to social science going back to Bentham, Hume and Smith, Albert has (1) sharply criticized various perspectives which argue that the classical program has to be replaced by some alternative because the quest for laws in social science is misguided. He has (2) explained how and why the methods used in the natural sciences may also be used in the physical sciences. He has also (3) explained how the construction of models in the social science may lead us closer to the truth with a method of ‘decreasing abstraction’ which for all intents and purposes is the same view offered by Popper. On the other hand, he has argued that the neoclassical program has failed because it is not realistic. It needs to be significantly revised in order to make it so. He has (4) pointed out that many theories in this tradition may be characterized as model-Platonism-they fail to be empirical; (5) that the use of a logic of decision has led the program away from both the construction of realistic models and from the desirable and traditional integration of psychological considerations into economic theory; (6) that it has not adequately taken account of institutions, because it has neglected the sociological approach found in the work Smith, Bentham and Hume which laid the groundwork for it; (7) that it cannot yield any plausible social function; (8) that its assumption that the attempt to maximize personal utility by individuals yields without further ado the attempt by managers to maximize profits is mistaken; and (9) that one cannot presume that individuals are capable of estimating the probabilities of the outcomes of their action in any realistic way. Albert (10) recognizes the importance of Schumpeter but stumbles over the same problem Schumpeter did: How can one reconcile, on the one hand, growth due to creative ideas which change systems with, on the other hand, the assumption that there are economic systems which obey economic laws? The inclusion of the former is realistic but does not yield laws; the latter yields laws but is not realistic. Albert’s criticisms of alternatives In order to defend the classical program, Albert has criticized at least two alternative approaches to explanation in the social sciences. He has hoped to show that these two approaches cannot and/or should not be used in the social sciences, either because they cannot achieve their goals or because they set their goals too low. The first approach, which Albert finds in sociology, hopes to use the inductive accumulation of observations in combination with statistical studies to produce interesting social scientific theories. The second is an historist approach, often combined with hermeneutical methods, which seeks to replace the search for nomological theories with interpretative ones (Albert 1964, pp. 38ff; 1968, pp. 131ff; 1972b, pp. 193ff; 1994). Historism is the theory that social scientists can only achieve historical explanations of particular events; they must abandon any hope for the discovery and/or use of nomological theories. Historism comes close to the view of Weber and Popper according to which the central aim of the social
sciences is to use trivial laws to explain particular events rather than to discover laws. Popper does not deny, however, that social scientific laws may be discovered. This means that his view cannot be refuted by showing the existence of some, even trivial, social scientific laws, as some other views may be. Albert apparently finds his own proposals in agreement with Popper, because both think that that the social sciences can construct models of social situations which describe their structure. Insofar as the social sciences simply seek to gather facts and to make inductive inferences with the help of statistical methods from them, we may accept Albert’s criticism of the first alternative without further ado. It is a restating of Popper’s well-known criticism of inductivism. The determination of the extent of social scientific research which may be described in this way requires a survey of this research which Albert has not undertaken. This criticism is also a protest that generalizations of statistical results do not lead to understanding of social processes. Albert seeks a realist theory of the social sciences. An empiricist view which does not include a theory of social scientific theories and of how one gets nearer to the truth in the social sciences is, he holds, defeatist and barren. Albert offers three criticisms of historism and/or an hermeneutical approach, which, he notes, are often found together. He also provides an interesting analysis of the metaphysics of thinkers who defend historism such as Heidegger or Gadamar, which I will not discuss here. In his first criticism he reminds us that the lack of success in finding laws is no reason not to seek them and no demonstration that they do not exist-we may always seek deeper explanations of empirical generalizations. Secondly, he argues that the methods of the natural sciences can also be used in the physical sciences, that the logic of research is no different in the two fields. Thirdly, he argues that the historist position itself is incoherent, because it must presume the existence of laws, whose existence it denies, in order to proceed with its own methods. In his first point in defence of the pursuit of laws in the social sciences Albert points out that the fact that a ‘law’ is known only to apply to some historical period does not mean it cannot be explained as a special case of a universal law not bound by time and place. We may thus always seek laws in the social sciences (Albert 1964, pp. 38ff). Albert is no doubt correct. But this fact still leaves open whether pursuing laws to explain generalizations is the best way to proceed or not. It does not explain the possibility of laws in the social sciences, that is, how individual rationality and social scientific laws may be reconciled. I will discuss his second argument, in which Albert claims that because the logic of research in the social sciences is no different from that of the physical sciences the same goals may be pursued, in the following section. In regards to the third point we should note, that it does not follow even from the existence of social scientific laws that law-like explanations of all or even of most individual social events may be given in the social sciences. Albert does not, at least not explicitly, make this jump. But without it, it is difficult to know how significant this argument is. Albert’s discussion leaves unresolved the issue of the which aim-or aims-the social sciences should pursue. If one has a law, say, the law that no society which has been open can return to a closed state without the use of barbaric methods, and
suppose, as I think we do, that we know well enough what counts as ‘closed’, ‘open’ and ‘barbaric’, we will be able to make some predictions about existing societies with a law-like nature. We can explain Mugabe’s use of barbaric methods to control society as a consequences of his attempt to roll back the openness of Zimbabween society and predict that if one should try to do that in Costa Rica a like result would follow. But this example of a simple social scientific law which can predict specific consequences may lead to an unwarranted optimism in regard to the possibility of finding social scientific laws which may be used to provide nomological explanations of a wide variety of social events. Popper’s law describes a situation in which the changing aims, plans, and knowledge of individuals do not increase the outcome of events in one important respect, that is, the need to repress individuals to turn an open society into a closed society. But various aspects of the development of societies-perhaps most aspects-do indeed depend on these factors. Individuals can steer events by inventing new plans and theories. And, insofar as they do, no laws will be found which may predict these events as consequences of them alone. We will not be able to predict them or to do so only with the proviso that the (rational) actions of individuals do not lead events in some other way. In such cases the task of the social sciences has to be conceived of quite differently than that of the physical sciences (Wettersten 1982b). Albert notes that rational action cannot be predictable and emphasizes that the failure to take account of the cognitive activity of individuals is a grave weakness of the neoclassical program. But he does not notice that remedying this difficulty will have an effect on the aim of the social sciences, which should remain that of finding laws. Natural scientists presume that no events of this kind exist. Surely there must be some difference in our aims when in one case no creative and therefore unknowable responses to problems are at hand and in the other case they are. Can we presume that nomological theories by themselves could explain all social events? Or only some? If so, which ones? I do not find answers to these questions. These questions are important for Albert’s criticisms of historism and pure hermeneutics. He criticizes historism, that is, the view which says that there should be or can be no nomological theories in the social sciences. Historical explanations, he says, require the assumption of at least some nomological theories even if they do happen to be trivial (Albert 1979, pp. 111-132; 2001b, pp. 140ff). Although Albert is right in pointing out that we need to presume nomological theories in order to give historical explanations, it does not follow from this fact that we should seek law-like explanations of social events. Albert only shows that we need to presume that some such laws are true, even if we offer only historical explanations. Outside of history, in which he notes that the tasks are different from those of the theoretical sciences, he does not ask further whether their role in the social sciences could nevertheless be quite different from their role in the physical sciences. How wide can we expect to explain events as instances of their application? The existence of nomological theories in the social sciences and the ability to predict social events as consequences of them may mislead one to thinking that the central aim of the social sciences should be the discovery of such laws, just as in
the natural sciences. Although it is obvious that a social scientific law has to be able to predict events or we could not claim that it was an empirical law, it does not follow that we may use such laws to predict the particular events we want to explain as simple consequences of them or of any law whatsoever. We may, for example, quite trivially presume that all humans have an emotional need to love and be loved. But, even though we may predict that we will find expressions of this need in any individual, and even though we have sufficient agreement on what counts as such an expression to say that this theory is testable, we still may not be able to predict how cultures will integrate this need into a social structure or how this structure might change to account for it. This depends on how individuals think about this need. And we cannot explain with any nomological theory how and why people think about this need the way they do. We may nevertheless use such a law to explain social events, in the sense that we identify problems individuals have (rationally) solved and explain how they have done so. This is quite different, however, from applying laws to predict future events as instances of them. Neoclassical economics as a social scientific research program The question of which aim the social sciences should pursue arises both in Albert’s defense of the use of the methods of the natural sciences in the social sciences and in his critique of hermeneutical and historist views of the social sciences. He portrays classical economics as a social scientific research program and suggests that the kinds of problems which arise for this program are no different from the kinds of problems which arise for research programs in the natural sciences (Albert 1978a, 1998 first chapter). There is, then, no reason to think that there should be any difference in the aims of the social and the physical sciences. This makes his defense of the use of the methods of the natural sciences in the social sciences stronger than Popper’s, who sharply distinguished between the aims of the natural and the social sciences, even though he also said that the same methods could be used in both disciplines. On Popper’s view natural scientists seek new laws whereas social scientists seek above all to use trivial laws to explain events. On Albert’s view social scientists, like natural scientists, should seek deeper and deeper laws. He correctly observes, ‘Gerade wenn man von der Wandelbarkeit der sozialen und natürlichen Bedingungen-und damit auch von der Geschichlichkeit sozialer Strukturen-ausgeht, ist es meines Erachtens besonders naheliegend, nicht nur nach der Möglichkeit struktureller Relativierung von Quasi-Gesetzen, sondern darüber hinaus auch nach ihrer Erklärung durch Rekurs auf tiefere Theorien zu suchen.’ (Albert 1976, p. 144).1
