For other entries in the series, please see most especially this post as well as this post
I. Dichotomies pose problems for philosophy and the social sciences
In “The Rationality of Science is Partial” in Towards a Rational Philosophical Anthropology (1977) TRPA Joseph Agassi points out the two of the key dichotomies in philosophy, namely between nature and convention and between the “utterly universal” and the “utterly particular.” It is possible to view any”specific society as merely arbitrary” … “because although from the outside a custom in a given society may look quite arbitrary, from within it may look quite rational” (263). Many social institutions, such as organized religions, contain both rational and supposedly irrational elements. They are a mix of the universal and the particular. Because they appear to be a mix of dichotomies, social scientists and philosophers are at a loss to explain them. They then explain one and explain away the other. Explanations typically end in an arbitrary manner.
Thus, many have concluded that religion is rational, but not rational enough (Ludwig Feuerbach). Because social institutions are rational, but not rational enough, various solutions have been applied. Relativism more or less declares the debate useless: everything is particular (266-7). For relativists, one can only describe and not engage in causal reasoning. For functionalism, customs, like religious institutions are “natural” and perfectly reasonable, “though only from within.”
Functionalism presents every institution as 100% conducive to good order. Opposite to relativism, everything may be explained. Functionalism also promotes a kind of quietism. If customs and institutions are rational and perfectly reasonable, “natural,” even Hegelian, then how they are to be improved remains a mystery. Their naturalness speaks against their reform. “How can natural things have errors in them?” one would ask. Of course, any modern biologist would tell you that nature is full of errors. Some of them quite interesting.
II. Functionalism and Relativism are Connected and Suffer from the Same Failing
If relativism is gleeful in its rejection of explanation (Noam Chomsky in his Aspects of a Theory of Syntax in Agassi TRPA, pg. 264) functionalism delights in its low standards of explanation and in the high standards to which it holds cultural institutions (they work as intended.)If relativism decrees “to each his own faith” functionalism says, “to each his own customs.” “Each custom, functionalism says, should be viewed from within, and indeed seen as natural” (266). Here, functionalism attempts to explain the particular existence of a custom (as well as its persistence) through its “naturalness”, rooting the particular in the universal without really explaining the connection or the relationship in the least.
On the lack of explanation inherent in functionalist discussions, Agassi complains (rightly) “the most incredible thing about the classic functionalist view of conflict … is the absence of any discussion on or even explanation of the point at hand.” Agassi underscores that in order to really prove (or better provide) for the existence of functionalism as a really existing thing in society, society must be demonstrated as a harmonious entity, as exhibiting a real organicism. However, “the view of society as a whole was never achieved. And so even functionalists could feel the fragmentary nature of their functionalism. They did try to offer more general theories of aspects of primitive life, from magic to economics. This however is quite problematic” since all theories of this type lead to overdetermination.” As problematically, although kinship can be described according to functionalistic premises (sociobiology and biosocial anthropology both do this with almost annoying consistency), the language describing kinship terms can not be described using functionalism (in fact in many cases, kinship terms are non-functional survivals which point to long-dead social roles.) (pg. 270ff., as with many Agassi works, the best material is in his footnotes.)
In his explanation of the drawbacks of functionalism, Agassi also has in mind Ernest Gellner’s qualms with “tolerance.” Gellner considered tolerance “facile” as it is “amounts to admitting that a system is criticizable yet refusing to see it criticized.” Tolerance “leads thinkers to explain away a lot- whatever a system is observed to be unable to digest” which necessitates a slight of hand, where the vastly more important material to be explained is that which is exempted from the argument (275).
Functionalism and relativism are both guilty of explaining too much and too little respectively and of taking a rather dogmatic view of explanation, which is a fear of contradictory evidence as outright refutation. Many philosophers have held (contrary to traditional rationalism) that explanation is partial, as is refutation. Gustav Hempel noted this and this is Agassi’s position as well.
III. Agassi’s Resolution of Dichotomies Reexplained and Applied
A unifying theme throughout Agassi’s work is the resolution of dichotomies and his “anti-all-or-nothing attitude.” He is against monism in all its forms. Characteristically, he wishes to find a middle ground between nature and culture and between the particular and the universal. In Agassi’s mind, the perceived polarity between nature and culture and between the particular and the universal are linked. This is right.
In the philosophy of law and history of ideas, there has always been much made of the distinction between laws made by men (and followed by convention) and those which can be intuited from nature or which are reflective of the natural order (it is never really clearly stated as to how this works—John Finnis gives natural law jurisprudence its most interesting modern formulation, but one can really get the gist by reading Thomas Aquinas). Laws by convention and reflective of some man’s will are not really seen as laws at all, but as mere reflections of power, which is not law. Man-made laws are particular and arbitrary, while laws from nature are reflective of some universality.
This was among the reasons why in the 20th century, legal positivists got into all kinds of trouble and were blamed for all sorts of nasty things. Of course, John Austin, Hans Kelsen and H.L.A Hart never really said that law needed no grounding in intuited natural order or morality and that a sovereign could just one day tell all to “butter bread side down”, but this is what they were imputed to argue. And if being a historian has taught me anything, its imputation rather than argument which is important. Austin and Hart in particular did no one any favors by more or less not writing in response to critics.
Likewise, discussions of nature and culture have always been defined by the polarity between the social and the natural, with particularity defining man’s social life. Social life (unlike nature) has also been resistant to explanation. Social scientists have always considered the natural realm to be the space of universality and impute this view onto natural scientists.
When addressing human behaviors and culture more generally, we (meaning social scientists like Marshall Sahlins) always tend to resist generalization and explanation, or provide explanations so unsatisfying that what is recalled is the description of the behavior, rather than any explanation of the behavior itself. Any generalization and any explanation in the description of culture is viewed (by anthropologists and others) as reductive reasoning. This is why description is favored by social scientists, as description is not reductive.
Of course, reductionism is not necessarily bad, it is merely a form of explanation. What matters is whether that reductionism becomes dogmatic, whether the explanation stifles debate and discussion. Reductionism as a form of explanation is useful, because it explains (and almost any explanation is better than no explanation at all, as non-explanation can not be subject to tests and criticisms).
Here I part a little with Agassi, who is much more concerned about reductionism tout court because of the historical tendency of reductionists to explain away rather than use reductionism to explain (TRPA, 50ff). And of course, all explanations are partial. And so frankly is all reductionism. Reductionism is embarassing to the reductionist because any person can point to an example of something which the reduction does not explain. Free will and all that. This is one of the chief failings of the greatest of modern reductionism, neo-Darwinian anthropology (if cultural relativism adjures explanation, neo-Darwinism tries to explain everything).
Now, it depends on which neo-Darwinian personage and theory cluster you are trying to explain, say Chagnon vs. Lee Cronk. If reductionism is deployed heuristically and with a light touch than it harms no one and can serve as a useful explanation for certain things. If reductionism is taken too seriously or is deployed in reference to too many topics, then the reductionism becomes problematic. Cheap example, but illustrative: “Amounts spent on engagement rings reflect aspects of male and female mate quality.” Personal factors are also quite important here.
Example of this: Napoleon Chagnon and Robin Fox are both neo-Darwinian biological anthropologists (more or less). Both are in the National Academy of Sciences. Chagnon is the world’s most bickered-about anthropologist. Robin Fox writes poetry and spends time with his grandchildren. Why is Chagnon controversial and Fox not? Fox is funny, Chagnon not. Fox writes about Robert Browning and the incest taboo (and kinship). Chagnon about primitive violence, status and differential fitness (social selection, more or less); all dicey topics for generalization to biology. Humorless and dogmatic writing doesn’t help either.
To return to the main argument, nature is invested with explanatory power and is the realm of generality (and legitimacy), while the social is the space of the particular and merely descriptive. The social moreover is arbitrary and a figment of will rather than law. Now this narrative is neither true nor productive. Although there are an infinite number of social mores, there emerge with some work, some generalities. Not everything which is natural and universal is of necessity good. Finally, we tend to think of the particular and the universal as opposed because of our notion that evidence either universally affirms a theory or by pointing out a contradiction, refutes it.
IV. Evidence and Refutation Are Taken Too Seriously, Our Knowledge Is Frequently Imperfect
Of course, when a “seeming conflict is declared real enough then the theory must go” but in most cases there should be an advocacy of tolerance for difficulties. In science as in politics, tolerance is to be observed. The other side of this is Agassi’s early point (emerging in Science in Flux) about the nature of our claims of knowledge: that they are not ideal and that they are mostly imperfect. Though sometimes they can be perfect. We can aspire to perfect knowledge, as we can to unvarnished rationality (it is a nice sentiment).
He begins, “Let us assume that claims for knowledge are often made which do not amount to claims for perfect or for demonstrable knowledge” (338). As not everyone can all the time be considered perfectly rational (or act in a perfectly rational way), in the same way evidence rarely outright confirms or denies a theory. Human beings live with a great deal of contradiction and even outright difficulty, scientists (and social scientists) as humans live with this too. Conversely, positive evidence does not usually satisfactorily confirm a theory. We are frequently left in a muddle of universals and particulars. Such is life.
In Science in Flux (1974), Agassi was keen to provide tests for the grading of imperfect knowledge. Perfect knowledge, he notes, is totally unqualified. Less than perfect knowledge can be graded according to qualifications. Qualifications are more or less what a community considers to be a qualifications: all knowledge claims are socially conditioned as to their believably. So, “When someone promises something which, our accepted views imply, is impossible, say a new invention, he thereby undertakes not to use our accepted views. This very instance indicates that whatever we all share with a claimant for knowledge may ordinarily be construed as a condition for his claim” (349). In this sense, knowledge is testimony and the reporting of witnesses, though we very much wish for such claims and reports to be impartial, we must take care to underscore whom is speaking.
Is this Wittgenstein? Close but no, because we can still say that less than perfect knowledge claims and less than perfect knowledge claimants are in error and we can grade the believably of their knowledge claims. They can still be and are often wrong. We can choose to not believe them and consider ourselves perfectly rational for not doing so.