What Is Naturalism?

Harvey Siegel
University of Miami

I want, first, to thank Fred Ellett and Dave Ericson for the serious attention they have paid to my book, and for the many compliments they have paid it.

I find myself in the unusual position of replying here to an essay to which I have already replied elsewhere,¹ as Ellett and Ericson's essay also appears, with some changes, as a review² of my book.³ I have replied to many of their critical points - including those concerning the epistemic roles of trust and intellectual authority, the nature of truth, the regarding of plausibility as a truth value and the character of multi-valued logics, the (il)legitimacy of all forms of necessity, the universality of concepts and norms, the unavailability of either a God's-eye point of view or a necessary framework for inquiry, the (mis)use of van Fraassen to discredit the possibility of conceptual truth, the "miracle argument" for realism, replies to skeptics concerning reason, and others - in that earlier exchange; I will not repeat those replies (much) here. Fortunately, their essay is a rich, wide-ranging one, which raises many fundamental issues. Rather than simply repeat my earlier reply here, I shall try to address other aspects of their view, and ask questions which I hope will lead to further clarification of it, in what follows.

Ellett and Ericson's main aim is to present and defend a view that they call "naturalism" (and to criticize me for being insufficiently naturalistic). The naturalism they espouse rejects "first philosophy," "necessity in all its guises," "the quest for certainty based upon meaning," and any "God's-Eye View" or "Godlike necessary constraints (or frameworks) for all inquiry"; but it accepts a sharp distinction between "first-order" and "second-order" discourses and questions, the latter being discourse aimed at providing normative "legitimation" or justification of first-order claims,⁴ and insists that there is no "external authority" which trumps the dictates of reason.⁵

I too reject and accept these things (with the possible exception of the rejection of "necessity in all its guises," to which I will return and request clarification below); this broad agreement between our two views heartens me. But I wonder why this view should be called "naturalistic," since in the philosophical literature naturalists reject the sharp first/second order distinction which is central to Ellett and Ericson's positive view. For example, C.A. Hooker - cited by Ellett and Ericson as a fellow naturalist - explicitly rejects this sharp distinction, arguing that normative notions and judgments must be treated theoretically in the same way that first-order scientific judgments about "the way the world is or appears to us to be" are to be treated; he wishes to unify first- and second-order discourse under the banner of naturalism.⁶ Hilary Putnam, on the other hand, despite being identified as a fellow naturalist by Ellett and Ericson, rejects naturalism on the basis of his embrace of something very like Ellett and Ericson's first/second-order distinction.⁷ Hooker and Putnam both reject as incoherent the joint embrace of naturalism and the first/second order distinction. Ellett and Ericson embrace them both. Obvious questions arise: Is Ellett and Ericson's position rightly regarded as "naturalistic?" Are they right to regard themselves as fellow travelers with Hooker and Putnam? Does their first/second-order dichotomy constitute a "necessary framework for inquiry" of the sort which they eschew? Can they coherently embrace that distinction alongside their espousal of naturalism? I would welcome clarification from Ellett and Ericson on these matters in the context of a more detailed account of their "naturalism."⁸

There are also unclarities, I believe, in Ellett and Ericson's rejection of "necessity in all its guises." They take that rejection to entail the rejection of certainty, a prioricity, analyticity, universality, and what might be called "conceptuality," but they neglect to draw crucial distinctions among these.⁹ They seem also to think that rejecting necessity requires rejecting deduction, because deduction (necessarily?) yields necessary truth: the "claim that there is no FIRST philosophy is an inductive conjecture...[which] does not itself have the status of a necessary truth."

Consider then the traditional example of a sound (hence valid) deduction concerning Socrates and mortality. The premises are true, although not necessarily so. But the premises "guarantee" the conclusion: If the premises are true, the conclusion must be true as well. As we tell our introductory logic students, it is impossible that in a deductively valid argument the premises can be true and yet the conclusion false. Notice two points. First, the conclusion in question is not a necessary truth; it is just an ordinary, contingent truth. (After all, it's possible that Socrates is immortal.) Deduction involves a kind of necessity, but it does not typically yield necessary truths. So one question for Ellet and Ericson: In rejecting "necessity in all its guises," do they reject deduction?

Rejecting the legitimacy of deduction is sufficiently beyond the pale (in the context of this discussion) that I will not pursue it further here. Assuming that Ellett and Ericson accept deduction, the second point worth noting is that the "Kantian," "transcendental" elements of my view which they find problematic involve nothing more than deduction. I do rely upon deduction, and upon the "weak" sort of necessity exemplified by the "Socrates is mortal" example: for example, I argue that if one wishes to call rationality into question on philosophical grounds, one will have to use reason to do so; therefore one cannot "defeat the rationalist" on philosophical grounds.¹⁰ I conjecture that this seems problematic to Ellett and Ericson because they lump all "guises of necessity" together into one undifferentiated mass, whereas if they drew some relevant distinctions among these various guises, they would, I believe, find my view and theirs quite close. Is this conjecture correct? I look forward to Ellett and Ericson's reaction to it, and to learning of their attitude toward the drawing of such distinctions in the further articulation of their brand of "naturalism."¹¹

Space precludes detailed discussion of many other issues raised by Ellett and Ericson's challenging essay. But I would welcome their comments on the following:

1. Ellett and Ericson interestingly relate naturalism and postmodernism, arguing that both share a rejection of metanarratives "which involve...unconditional necessities." But this seems unfair to postmodernism, which rejects all metanarratives, not just those involving "epistemically inaccessible" necessities. Is reference to necessity a necessary condition of something's being a metanarrative? Why?
2. Why do Ellett and Ericson equate "objectivity" and "second-order legitimation?" These seem to me to be far from equivalent notions. On the contrary, objectivity is one attribute which such legitimations (and judgments, more generally) may or may not have; judgments are objective when they are acknowledged to admit of, and are subjected to, evaluation in terms of relevant, fair and independent standards. Second-order legitimations might or might not be objective, so the two notions are hardly equivalent.¹²
3. Why do Ellett and Ericson equate "the subjectivist about values" and "the skeptic about reasons?" These too seem to me to be far from equivalent philosophical position-holders.
4. Ellett and Ericson criticize my view as follows: "[T]o talk about 'fallibilism' without offering a positive account is to be left with a vacuous view. One must develop a positive theory of inquiry which coordinates epistemology, a theory of truth, and ontology." The bulk of their essay offers what they take to be an alternative epistemological view, which has been the focus of these comments. But I see little in their paper which constitutes either a theory of truth or an ontology. I would welcome their further articulation of these aspects of their view, and of the sort of coordination they have in mind which might avoid the alleged vacuity of my own. In addition, I would welcome further clarification of both their characterization of rationality and their second-order legitimation of it.
5. Finally: Ellett and Ericson and I are agreed that rationality stands in need of a "second-order legitimation," and that that need can be met; we each try to meet it. They criticize my proffered legitimation, and take themselves to be offering an alternative. But I am unclear, after these several rounds of discussion and reply, just how different our alternative accounts are. I hope that they will summarize their present view of the differences, if any, remaining between us.

---

1. Harvey Siegel, "Replies to Reviews," Paideusis: Journal of the Canadian Philosophy of Education Society11, no. 2 (Winter 1998): 27-37.

2. Frederick S. Ellett, Jr. and David P. Ericson, "The Complexities of Reasons: A Critical Response to Harvey Siegel's Rationality Redeemed?" Paideusis: Journal of the Canadian Philosophy of Education Society11, no. 2 (Spring 1998): 3-12.

3. Harvey Siegel, Rationality Redeemed?: Further Dialogues on an Educational Ideal (New York: Routledge, 1997).

4. Although second-order discourse is also said to involve "questions about what...we should mean...about the right use of the concepts of reality and knowledge" (emphases added), which seems to presuppose a sharp fact/meaning distinction which their rejection of analyticity precludes, and thus to smuggle back in what their version of naturalism claims to eschew. In any case, the first/second-order distinction seems clearly enough to be a distinction between discourses concerning fact and norm. Any such sharp fact/norm distinction sits uneasily with naturalism.

5. Ellett and Ericson suggest that I disagree with their claim that there is no "external authority" overseeing reasoned inquiry. This surprises me. The allegedly external authority they find in my book is "epistemology," but that just is the domain in which the general theory of reason and inquiry resides. We are agreed that there is no authority which governs reason but reason itself.

6. Though he would not put the point in terms of distinct sorts of discourse. See C.A. Hooker, Reason, Regulation, and Realism: Towards a Regulatory Systems Theory of Reason and Evolutionary Epistemology (Albany: SUNY Press, 1995). For extended discussion of Hooker's view, see my "Hooker's Revolutionary Regulatory Realism," Studies in History and Philosophy of Science 29, no. 1 (March 1998): 129-41; and my "Naturalism and Normativity: Hooker's Ragged Reconciliation," Studies in History and Philosophy of Science 29, no. 4 (December 1998): 639-52; sandwiching Hooker's "Naturalistic Normativity: Siegel's Scepticism Scuppered," Studies in History and Philosophy of Science 29, no. 4 (December 1998): 623-37.

7. Hilary Putnam, "Why Reason Can't Be Naturalized?" Synthese 52, no. 1 (1982): 3-23.

8. I suspect that what they call "naturalism" is not much more than what I call "fallibilism." I look forward to Ellett and Ericson's (dis)confirmation of that suspicion. I also suspect that all Ellett and Ericson need for the success of their view is just that questions about justificatory status are legitimate; that the normative evaluation of scientific and other beliefs and claims is epistemically respectable, important, and not reducible to scientific claims about "the way the world is or appears to us to be." They need not posit two fundamentally distinct types of discourse, questions, and inquiries - in effect, a rigid fact/norm distinction - which seems contrary to the spirit of naturalism.

9. See my "Replies to Reviews," esp. concerning Goldman, Kripke, and van Fraassen. Putnam, Realism and Reason: Philosophical Papers, vol 3 (Cambridge: Cambridge University Press, 1983), chaps. 3, 5-7, offers a much more nuanced discussion of the relationships among these various alleged "guises of necessity." I should also note that Ellett and Ericson's suggestion that necessary truths/falsehoods are contingent in "actual inquiry" because "one can always choose to reinterpret the terms" is problematic, since in reinterpreting the terms of a proposition one changes the proposition.

10. See my "Can Reasons for Rationality Be Redeemed?" in Philosophy of Education 1996, ed. Frank Margonis (Urbana, Ill.: Philosophy of Education Society, 1997), 74-76; also Rationality Redeemed? chap. 5, in which I make clear that this is on my view a fallible conjecture and that, more generally, we are never in a position to regard any arguments (more precisely, any argumentatively derived conclusions) as certain. On the general point concerning fallibility and "guises of necessity," see Putnam, Realism and Reason, 96, 136.

11. As explained in "Replies to Reviews," I am in agreement with much of Ellett and Ericson's version of naturalism; I too reject certainty, God's-eye perspectives, necessary frameworks, and so forth. I reject what is regarded as "naturalized epistemology/philosophy of science" in the literature for reasons given in my "Hooker's Revolutionary Regulatory Realism" and "Naturalism and Normativity: Hooker's Ragged Reconciliation"; see also my "Naturalized Epistemology and 'First Philosophy,'" Metaphilosophy 26, no. 1 (1995): 46-62; "Naturalism and the Abandonment of Normativity," in The Philosophy of Psychology, ed. William O'Donohue and Richard Kitchener (London: Sage, 1996), 4-18; "Instrumental Rationality and Naturalized Philosophy of Science," Philosophy of Science 63, no. 3 Supplement (PSA Proceedings, Part 1, 1996): 116-24; and "Naturalism, Instrumental Rationality, and the Normativity of Epistemology," Proto Sociology 8 (1996): 97-110.

12. See Alven Neiman and Harvey Siegel, "Objectivity and Rationality in Epistemology and Education: Scheffler's Middle Road," Synthese 94, no. 1 (1993): 55-83, esp. 59-61. I should note here that Ellett and Ericson's related remarks concerning "referential import" and "'initially credible' beliefs" badly misconstrue Neiman's and my (and Scheffler's) discussion.