3. There is a more sophisticated argument for the view that the social sciences cannot be value-free. It maintains that the distinction between fact and value assumed in the preceding discussion is untenable when purposive human behavior is being analyzed, since in this context value judgments enter inextricably into what appear to be “purely descriptive” (or factual) statements. Accordingly, those who subscribe to this thesis claim that an ethically neutral social science is in principle impossible, and not simply that it is difficult to attain. For if fact and value are indeed so fused that they cannot even be distinguished, value judgments cannot be eliminated from the social sciences unless all predications are also eliminated from them, and therefore unless these sciences completely disappear.
For example, it has been argued that the student of human affairs must distinguish between valuable and undesirable forms of social activity, on pain of failing in his “plain duty” to present social phenomena truthfully and faithfully:....
Moreover, the assumption implicit in the recommendation discussed above for achieving ethical neutrality is often rejected as hopelessly naïve —this is the assumption, it will be recalled, that relations of means to ends can be established without commitment to these ends, so that the conclusions of social inquiry concerning such relations are objective statements which make conditional rather than categorical assertions about values. This assumption is said by its critics to rest on the supposition that men attach value only to the ends they seek, and not to the means for realizing their aims. However, the supposition is alleged to be grossly mistaken. For the character of the means one employs to secure some goal affects the nature of the total outcome; and the choice men make between alternative means for obtaining a given end depends on the values they ascribe to those alternatives. In consequence, commitments to specific valuations are said to be involved even in what appear to be purely factual statements about means-ends relations.--Ernest Nagel (1961) The Structure of Science: Problems in the Logic of Scientific Explanation, pp. 490-491 (emphasis in original).
It is hard to imagine how influential Nagel's Structure was at one point. (Now it has about 10k in citations on scholar.google. Quine's Word and Object, which appeared the same year, has about 50% more citations.) By the time I was a student, it was associated with 'the received view' and a misguided form of reductionism in science, overshadowed by Kuhn's more influential Structure (which appeared a year later) and by all the exciting, technical work in philosophy of special sciences and the burgeoning movement looking at the metaphysics of science. But when it came to presenting the received view, Hempel's lucidity and brevity was preferred. As a consequence I never read Nagel's Structure as an undergraduate or graduate student, but I did frequently encounter discussions of Nagel reduction (see here for an intro).
The passage I omitted from the text I quoted from Nagel, is Nagel quoting (as the accompanying footnote states), Leo Strauss, “ The Social Science of Max Weber,” Measure, Vol. 2 (1 9 5 1 ), pp. 211-14. To the best of my knowledge it's one of the few instances where a bona fide analytic philosopher engages directly with an argument by Strauss while he was alive. Strauss (1899-1973) is rarely mentioned by political philosophers of the age, but if he is it is mostly in the context of his claims about esotericism (see Nozick here).* To the best of my knowledge, Rawls and Gewirth never mention Strauss.+
To have a sense of how unique Nagel's engagement is, I offer the following illustration. When specialist, analytic ancient philosophers of the age mention Strauss they do so (see Vlastos) by praising a 1974 review, in Philosophical Review, by Irwin of Strauss' (1972) Xenophon's Socrates. In the review Irwin devotes about half a dozen lines to Strauss:
Strauss's "interpretation" consists of a tedious paraphrase with comments or questions on major and minor issues. No coherent line of interpretation emerges from his enigmatic asides. His paraphrase merely reminds us how unexciting Xenophon can be, and even reduces the amusing episodes to a uniform level of dullness.
The rest of the review is Irwin describing the questions Strauss should have engaged in. While Irwin's summary is not strictly false -- there is something tedious about Strauss' Xenophon's Socrates -- it is completely uninformative. (More about that some other time.)
As I have discussed, Nagel could be a fierce polemicist, including a public, political polemicist. But the response to Strauss in Structure, does not fit that mold. Rather the section -- it's about five pages -- attributes to Strauss a family of arguments that [A] 'social sciences cannot be value free.' As can be seen from the passage quoted, Nagel treats this as entailing or nearly identical to the thesis [B] that there is no "compelling reasons" that "an ethically neutral social science is inherently impossible." (Structure 495). I am a bit surprised by this conflation in Nagel because ethical neutrality may well be a significant 'value'! To be sure, Nagel himself derives this formulation from Strauss, who attributes the identity of "a "value-free" or ethically neutral social science" to Weber (Natural Right and History (hereafter NRH), p. 40).
Before I get to Nagel's arguments two observations: first, Nagel responds to a version of Strauss's argument published in 1951 in Measure. To the best of my knowledge this journal did not have a long life or wide circulation. (I have requested a copy of it from the British Library.) But Strauss reprinted a version of the article in NRH. The passage that Nagel quotes can be found there in chapter 2, pp. 50-53. The omissions in Nagel are basically all the places where Strauss is criticizing Weber directly. (I return to that below.) Natural Right and History was published in 1953, and famous, so it is a bit weird that Nagel doesn't cite it. Second, a few years ago, Anna Alexandrova published a paper, "Can the science of well-being be objective?." The British Journal for the Philosophy of Science (2020) that has quickly become a classic in its own right. This paper engages with the material I am about to discuss (and even mentions that Strauss was Nagel's target), and expresses my own substantive disagreements with Nagel better than I could.
Now, the discussion by Nagel of Strauss occurs in a chapter 13 on the "methodological problems of the social sciences" in section (V) on "The Value-Oriented Bias of Social Inquiry." This is a rather long, ambitious chapter (over 50 pages). It made Nagel a recognized authority in the philosophy of social science, and got him invited (as I have documented elsewhere) to debates internal to (say) economics. The chapter also includes, for example, Nagel's criticism of Hayek's The Counter-Revolution of Science, which (recall) Nagel partially recycles from a polemical 1952 review in Journal of Philosophy. This section (V) is devoted to the following issues:
Since social scientists generally differ in their value commitments, the "value neutrality” that seems to be so pervasive in the natural sciences is therefore often held to be impossible in social inquiry. In the judgment of many thinkers, it is accordingly absurd to expect the social sciences to exhibit the unanimity so common among natural scientists concerning what are the established facts and satisfactory explanations for them. Let us examine some of the reasons that have been advanced for these contentions. It will be convenient to distinguish four groups of such reasons, so that our discussion will deal in turn with the alleged role of value judgments in (1) the selection of problems, (2) the determination of the contents of conclusions, (3) the identification of fact, and (4) the assessment of evidence. Nagel Structure 485
I quote it for two reasons: first, because it shows that the way Nagel operationalizes 'value neutrality' is in terms of unanimity, or consensus. (To readers of Tom Kuhn's Structure this should be familiar!) The broader context of Nagel's argument shows that the relevant class is "competent workers in the natural sciences" (Nagel Structure, 448). In addition, the question for social science is articulated in terms of the purported "unanimity so common among natural scientists concerning what are the established facts and satisfactory explanations for them." Nagel is aware, of course, that natural scientists don't always agree, but this occurs for him either at the research "frontier" (448) or in contexts where the disagreement is an effect of "alternative formulations" that are "mathematically equivalent" (158).
Second, the response to Strauss falls under what Nagel calls (3) "the identification of fact." In fact, Nagel identifies Weber's position with (1). He doesn't identify a single author with (2), but in it he treats S. F . Nadel (an anthropologist, who was "a pioneer of multi-sited ethnography,"), his own teacher Morris R. Cohen, and A.E. Burtt (whose work on Newton is important) as typical representatives of it (Nagel's own discussion draws on work by Gunnar Myrdal). Finally, Karl Mannheim is treated as the exemplar of someone who holds (4). What's neat about this list is that it is both politically diverse and reflects views of different kinds of social science(s).
Nagel attributes to Strauss two distinct criticisms: the first is that "the distinction between fact and value.. is untenable when purposive human behavior is being analyzed, since in this context value judgments enter inextricably into what appear to be “purely descriptive” (or factual) statements." The second is that purported "factual claims about means-ends statements" are themselves infected with values. It's not hard to see that they are similar in kind.
In his response to the views he attributes to Strauss, Nagel emphasizes that Strauss is right about three features: "(a) that a large number of characterizations sometimes assumed to be purely factual descriptions of social phenomena do indeed formulate a type of value judgment; (b) that it is often difficult, and in any case usually inconvenient in practice, to distinguish between the purely factual and the “evaluative” contents of many terms employed in the social sciences; and that (c) values are commonly attached to means and not only to ends." (Nagel Structure, 491--[letters added to facilitate discussion]).
But Nagel insists that Strauss equivocates on two notions of 'value judgment: first, "the sense in which a value judgment expresses approval or disapproval either of some moral (or social) ideal, or of some action (or institution) because of a commitment to such an ideal; and the sense in which a value judgment expresses an estimate of the degree to which some commonly recognized (and more or less clearly defined) type of action, object, or institution is embodied in a given instance." (Nagel, Structure 492) Nagel adds that these notions are often conflated in the social sciences and that sometimes it's not so easy to sort distinguish them. But he concludes his argument that "there are no good reasons for thinking that it is inherently impossible to distinguish between the characterizing and the appraising judgments implicit in many statements, whether the statements are asserted by students of human affairs or by natural scientists." (Nagel, Structure 494).
Now, rhetorically, it's important to see that Nagel's argument -- and this is characteristic of his argument in the whole chapter -- generally has the form 'an apparent problem X in social science occurs also in natural sciences and when X occurs in natural science, X does not prevent the development of consensus in the natural sciences (and, thus has been tamed in the natural sciences) and so X does not pose an in principle obstacle to consensus in social science.' So, for example, Nagel illustrates the two senses of “value judgment” with an example from biology.** And the effect of this move is to turn the critic of value neutrality in social science into a critic of the value neutrality of natural science. Nagel assumes nobody (not even the sociologist of knowledge he discusses under (4)) will go that far.
Today's post won't be able to evaluate Nagel's arguments, not even as arguments against Strauss. But it is worth saying something about what Strauss is up to. Now, Strauss' argument is directed against Weber, and in particular, because Weber is taken to be a spokesperson for the following position: "Natural right is then rejected today not only because (i) all human thought is held to be historical but likewise because it is thought that (ii) there is a variety of unchangeable principles of right or of goodness which conflict with one another, and none of which can be proved to be superior to the others." (Strauss NRH 36) That is to say, Strauss associates Weber with features of historicism (viz, i) and with value pluralism (viz, ii). To be sure, Strauss recognizes that Weber is not a pure historicist, because Weber recognizes the historical situatededness of the sciences alongside the "trans-historical" nature of its "findings regarding the facts and their causes. More precisely, what is trans-historical is the validity of these findings." (Strauss NRH 39). And because Strauss wants to argue for natural right, or at least, the inherent possibility of the recovery of natural right, he is critical of Weber.
So, Strauss himself is not especially interested in doing social science (although I qualify that below). But he is interested in a kind of self-limitation Weber put on social science: "the absolute heterogeneity of facts and values necessitates [for Weber] the ethically neutral character of social science: social science can answer questions of facts and their causes; it is not competent to answer questions of value." (Strauss NRH 40) But unless values are self-contradictory, the social scientist must be silent about them. Now this position is coherent, and it is not obvious why Strauss cares about social science method at all.
The answer is that Strauss (correctly I think) discerns that Weber is a kind of skeptic about value: "his belief that there cannot be any genuine knowledge of the Ought." (Strauss NRH 41) And so observed value pluralism is a kind of effect of this Weberian skepticism (which according to Strauss naturally leads to a nihilism that Weber obscures from himself). To be sure, Strauss attributes to Weber a kind of formal possession of the norm, "Thou shalt have ideals," (Strauss NRH 44); so on this view Weber is not a radical (second order) skeptic about value. For Weber the content of this norm is, as Strauss notes, "Follow thy god or demon," which Strauss reinterprets as "devotion to a cause." (NRH 46).
As an aside, the really significant observation on Weber's philosophy by Strauss is that "Weber's thesis that there is no solution to the conflict between values was then a part, or a consequence, of the comprehensive view according to which human life is essentially an inescapable conflict. For this reason, "peace and universal happiness" appeared to him to be an illegitimate or fantastic goal. Even if that goal could be reached, he thought, it would not be desirable; it would be the condition of "the last men who have invented happiness," against whom Nietzsche had directed his "devastating criticism."" (Strauss NRH 65) That is to say, Strauss diagnoses how Weber's Nietzscheanism anticipates Carl Schmitt's position. (Schmitt is not mentioned by Strauss in this context.)
Now, the objection to Weber by Strauss that Nagel cites occurs in the discussion of Strauss' analysis of Weber's sociology of religion, which "presupposes a fundamental distinction between "ethos" and "techniques of living " (or "prudential" rules)." And Strauss suggests that "the sociologist must then be able to recognize an "ethos" in its distinctive character; he must have a feel for it, an appreciation of it, as Weber admitted." (Strauss, NRH 50) Then occurs the passage Nagel (selectively) quotes on 490-491 in Structure as follows:
Would one not laugh out of court a man who claimed to have written a sociology of art but who actually had written a sociology of trash? The sociologist of religion must distinguish between phenomena which have a religious character and phenomena which are a-religious. To be able to do this, he must understand what religion is. . . . Such understanding enables and forces him to distinguish between genuine and spurious religion, between higher and lower religions; these religions are higher in which the specifically religious motivations are effective to a higher degree. . . , The sociologist of religion cannot help noting the difference between those who try to gain it by a change of heart. Can he see this difference without seeing at the same time the difference between a mercenary and nonmercenary attitude? . . . The prohibition against value-judgments in social science would lead to the consequence that we are permitted to give a strictly factual description of the overt acts that can be observed in concentration camps, and perhaps an equally factual analysis of the motivations of the actors concerned: we would not be permitted to speak of cruelty. Every reader of such a description who is not completely stupid would, of course, see that the actions described are cruel. The factual description would, in truth, be a bitter satire. What claimed to be a straightforward report would be an unusually circumlocutory report. . . . Can one say anything relevant on public opinion polls . . . without realizing the fact that many answers to the questionnaires are given by unintelligent, uninformed, deceitful, and irrational people, and that not a few questions are formulated by people of the same caliber—can one say anything relevant about public opinion polls without committing one value-judgment after another?
In Nagel's hands it is completely unclear that Strauss is offering an immanent critique of Weber.
And if one reads Nagel then it is natural to think that Strauss's rejection of ethical neutrality of social science is an obstacle to consensus or unanimity in social science. And so that lurking in Strauss' argument is a kind of denial of social scientific objectivity. But it is worth noting that for Strauss the whole point of recognizing values in social science is to make social scientific and historical objectivity possible. This is actually completely explicit in Strauss. For, Strauss concludes his own argument against Weber (at least the present one that Nagel is focused on) as follows:
The rejection of value judgments endangers historical objectivity. In the first place, it prevents one from calling a spade a spade. In the second place, it endangers that kind of objectivity which legitimately requires the forgoing of evaluations, namely, the objectivity of interpretation. The historian who takes it for granted that objective value judgments are impossible cannot take very seriously that thought of the past which was based on the assumption that objective value judgments are possible, i.e., practically all thought of earlier generations. Knowing beforehand that that thought was based on a fundamental delusion, he lacks the necessary incentive for trying to understand the past as it understood itself. (Strauss NRH 60-61)
That is to say, establishing the possibility of establishing natural right just is the possibility of establishing a trans-historical consensus/agreement or unanimity in social science. (To avoid confusion: all agree that history is at least partially part of social science.) So, somewhat ironically Nagel and Strauss agree about what we might call the formal aims of social science, that it involves a species of objectivity that makes fundamental consensus possible. This fundamental agreement is completely obscured by Nagel's presentation. To be continued.
*Interestingly enough, elsewhere in Philosophical Explorations, Nozick mentions "followers of Strauss?" in passing.
+Outside of analytic philosophy, it is clear that Berlin and Arendt were familiar with Strauss, but there is almost no engagement. By contrast, Burnham -- by then in his conservative phase -- praises Strauss in The Suicide of the West.
**To what degree Nagel can presuppose the value-neutrality of biology is actually an interesting topic.
Comments