Abstract
Moral intuitionism, once an apparently moribund metaethical position, has seen a resurgence of interest of late. Robert Audi, a leading moral intuitionist, has argued that in order for a moral belief to qualify as intuitional, it must fulfill four criteria: it must be non-inferential, firmly held, comprehended, and pre-theoretical. This paper centers on the fourth and seemingly most problematic criterion: pre-theoreticality. The paper begins by stipulating the defensibility of the moral cognitivism upon which moral intuitionism turns. Next, the paper develops the distinction between semantic and epistemic pre-theoreticality, and goes on to explore and reject the putative ubiquity of the theoreticality of first-order moral discourse: it argues that on a defensible understanding of theoryhood, both semantic and epistemic pre-theoretical moral belief is not only possible, but in fact frequently realized. The paper then briefly explores and rebuts objections to (epistemic) pre-theoreticality issuing from (i) the “Cornell Realists” and (ii) considerations of the epistemic relevance of the epistemic/valuational background of moral belief. The paper concludes that, so far as the pre-theoreticality criterion is concerned, moral intuitionism remains in the running as a viable thesis regarding foundationally justified first-order moral belief and knowledge.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Audi (2004, pp. 33–35): this is not direct quotation.
Shaffer-Landau (2003).
Sinnott-Armstrong (2009).
Huemer (2003: p. 102): “An initial, intellectual appearance is an intuition” [his emphasis]. Also see Kagan (2001, p. 45 n1).
I provide a more extended defense of the doxastic interpretation of moral intuitions in Kulp (2014).
Notice that (i) with the exception of the “firmly held” criterion, Audi’s criteria appear applicable to “intellectual seemings”—the conception of moral intuitions endorsed by Huemer et al.—and (ii) it appears that in particular, we typically construe moral intuitions doxastically. (I owe these points to Robert Audi.)
I assume that in order to either “firmly hold” or “comprehend” a proposition \(p\), it is not necessary to firmly hold or comprehend all of \(p\)’s entailments. Perhaps some of \(p\)’s entailments, yes, but surely not all as a general requirement. There may be indefinitely, perhaps infinitely many such entailments. (The same applies mutatis mutandis to \(p\)’s presuppositions.) To require otherwise would be a reductio ad absurdum of the possibility of belief and understanding.
Strictly, matters are more complicated than this. For ‘proton’ may be meaningful relative to any of a variety of particular atomic theories, although it may, or it then again it may not, have the same meaning relative to theory \(H_{1}\), but not to theory \(H_{2}\). We can safely ignore such complications here.
See Audi (2004, pp. 35ff) for useful discussion.
One might suspect that pre-theoreticality follows from non-inferentiality, since if \(S\) believes that \(p\) and a theory \(H\) is part of \(S\)’s evidence for believing that \(p\), then \(S\) inferred that \(p\) in whole or in part from \(H\). But then why think that pre-theoreticality is distinct from non-inferentiality? This, however, seems to me mistaken. \(S\) could believe that there is a hand in front of his face, which is grounded in his perceptual experience, even though \(S\) performed no inference. Similarly, \(S\) could believe that it is prima facie wrong to kill his dog, which is grounded in his belief-set about, say, the morality of treating higher mammals. Perhaps \(S\)’s belief-set even includes a “Dog Killing Theory” \(H\), yet \(S\) made no inference from, or that included, \(H\). In this case the non-inferentiality requirement would be fulfilled, even if the pre-theoreticality requirement is not. The two criteria are not the same.
In this case semantic dependence on theory is also evident, e.g., in the very meaningfulness of ‘alpha particle’. I suspect that where there is epistemic dependence on theory, there is very often semantic dependence on theory as well. I will not pause to sort this out here.
This is nearly exactly as Audi (2004, p. 36) puts it.
See, e.g., Audi’s discussion (2004, p. 36f). To hold that being well informed in a given theoretical context precludes having intuitions seems quite implausible. The theoretical physicist cannot have intuitions about theoretical physics, or the moral philosopher cannot have intuitions about moral philosophy? Surely she can. In fact, the better informed one is about a given subject matter, the more likely it is that one will have intuitions about that subject matter, and the greater the likelihood of their truth. I argue this in detail in Kulp (2014).
Audi (2004, p. 36).
But not all belief is propositional. For example, objectual belief is belief about an object which, though conceptual, does not rise to the level of cognitive sophistication requisite for propositional belief. Thus, \(S\) may perceive a red object and consequently possess a belief of its redness, but not possess a belief that it is red. The operative concept of belief in moral contexts, however, is propositional, and this shall be my sole concern throughout this paper.
See, e.g., Brink (1989, p. 1).
A classic example being Ayer (1952).
The intuitionist is perfectly free to insist that the two examples just given are inferential beliefs.
A moral intuitionism that entails that all moral beliefs, or even all justified moral beliefs, are intuitions is implausible on its face.
Moral intuitions do not arise ex nihilo; they arise out of an “epistemic/valuational background,” which may contain both true and false beliefs. Falsehood in this background can undermine epistemic justification. For a discussion of the concept of an epistemic/valuational background, and its effects on the justification of moral intuition, see Kulp (2011, pp. 65ff.),; and I discuss these matters in substantial detail in Kulp (2014).
Mackie (1977).
I discuss some of these same issues in Kulp (2011).
Ayer’s emotivism as a classic statement of such a view.
A debate with a long history: Emotivists like Ayer (1952) and Stevenson (1944), and expressivists generally—Blackburn (1993) is a contemporary example—have long squared off against moral realists: recent examples are Brink (1989) and Huemer (2005)—and even against those who are no friends of moral realism, such as error theorists like Mackie (1977). Contributors to this debate are legion: an enormous literature.
See Kulp (2011).
I would also point out that non-cognitivists should consider that it isn’t clear that all first-order moral locutions are alike as pertains to any putative non-propositionality. Consider the following:
-
L1: It is morally impermissible to enslave people
-
L2: A normative ethical theory is morally incorrect insofar as its entailments fail to treat like moral cases alike
-
L3: Kantian deontology is a more valuationally coherent normative ethical theory than rule-utilitarianism.
-
L4: Murder is prima facie morally wrong.
One may make a case for the absence of cognitive import of L1—not a convincing case to my mind, but a serious case nonetheless. Emotivists, and expressivists generally, for example, have tended to center their arguments on such locutions. But if L1 can be regarded as a “primary” form of first-order moral discourse, and thus a prime candidate for non-cognitivist explication, matters become less clear as we pass on to L2 through L4. L2 still appears non-cognitive to many—certainly not to me—although here we are talking not about the putative impermissibility of a social policy or act-type, but about the requirements that any normative ethical theory should fulfill. The operative term ‘morally wrong’ is present, so on the non-cognitivist interpretation L2 is non-descriptive or expressive in its primary sense. But it seems that the putative absence of cognitive significance is even less plausible here than in the case of L1. Just what, for example, is being emoted? It certainly appears, rather, to be a general statement asserted regarding the adequacy conditions of any candidate for acceptance as a normative ethical theory. Similar considerations apply a fortiori to L3 and L4. A locution like L3, which is still “within” morality (i.e., still a first-order moral locution), seems to be about the logical and/or epistemological status of two normative ethical theories: one is hard put to see this locution as non-cognitive in any primary sense. And on at least one prominent interpretation of L4, also clearly within morality, L4 is tautologous, hence clearly propositional. Why, however, should the moral intuitionist be limited to holding that we can have intuitions only about locutions like L1? Why can’t we also have intuitions whose objects are propositions belonging to moral theory? If L3, for example, is (i) a genuine first-order moral locution, which (ii) refers to a kind of epistemic, logical, or normative adequacy regarding two contrasting normative ethical theories, and which (iii) possesses the cognitive significance, the truth-candidacy that it certainly appears to have, then (iv) it would seem that L3 is a candidate for pre-theoreticality.
The point of these brief remarks is to suggest that there is a range of first-order moral discourse, some forms being more, some less amenable to non-cognitivist analysis. Thus, if one rejects the possibility of fulfilling the pre-theoreticality requirement for moral intuitions because of presupposition failure, that is, on grounds of the falsehood of the cognitivism moral intuitionism presupposes, then one may find one’s case far more plausible, if plausible at all, with regard to some spheres of first-order moral discourse than others. But all this suggests that the pre-theoreticality criterion is not so easily swept aside.
-
Pragmatists like John Dewey and Richard Rorty are paradigm cases. For example, Dewey: A term, proposition, or theory \(x\) is meaningful insofar as it has relevance to the solution of a “problem situation” confronting an epistemic agent \(S\). \(x\) is an “instrument” that \(S\) uses toward this end of problem-solving, and \(x\) can function in this role only insofar as it is embedded in a matrix of other instruments which confer meaning. That \(x\) is only meaningful in such contexts is obvious in the advanced physical sciences—What would the term ‘electron volt’ mean outside the relevant theoretical context?—but because there is continuity between the advanced scientific and the mundane solutions of everyday problem situations, all meanings are theory dependent. See, for example, Dewey (1910, 1938).
I take up some of these same issues in Kulp (2011).
The infant may be responding instinctually, and the the adult perceiver may be responding with objectual, but not propositional belief. See (Audi (2011b), p. 21f), regarding objectual belief.
A point made in a similar context by Tropman (2012).
Robert Audi brought this point home to me in private correspondence.
Of course we may well have theories about myths.
One of the great motivators of logical positivism’s verificationist theory of meaning was to discriminate between empirical propositions, analytic propositions, and pseudo-propositions. Unless analytic, any genuine position must be verifiable (later “confirmable”) through observation. And any purported empirical “theory”—a scientific theory, for example—devoid of observationally verifiable entailments (e.g., predictions) is not an empirical theory, hence not a scientific theory. In fact, it would not be a theory at all—a mere collection of pseudo-propositions. The literature generated by this perspective, pro and con, beginning in the late 1920’s and ‘30s with Schlick, Neurath, Carnap, and Ayer et al. is voluminous, and continues into current debates. In broadly similar, but importantly different fashion, Popper in (1962) and other works sought to distinguish science from pseudo-science (“metaphysics”) in terms of falsifiability. A scientific theory entails predictions which may be observationally falsified, thus falsifying the theory; a pseudo-scientific “theory” does not entail such predictions, and is not falsifiable. Lack of progress has also been urged as a demarcator between science and pseudo-science: see, e.g., Thagard (1978a). A substantial literature emanates from these perspectives as well. I might also note here that issues regarding the pre-theoreticality (or “theory-ladeness”) of observation has been prominent in the philosophy of science and the philosophy of mind. For example, in a well-known exchange between Jerry Fodor and Paul Churchland, Fodor (1984) has argued against the theory-ladeness of observation, and Churchland (1988) in favor of it [cf. Fodor (1988)]. These are specialized matters, however, that I cannot pursue here. And in any event, I would join Audi in cautioning against construing the analogy between perception and intuition too strongly: see Audi (2004, p. 36f.).
Consilience, i.e., how much a theory explains, is likely another general desideratum, distinguishable, it would seem, from explanatory force; for how well a theory explains, and how much it explains are not equivalent. Perhaps one desideratum can be maximized only at the expense of another, e.g. simplicity at the expense of consilience, and vice versa. The proper weighting of theory desiderata is a complex business, in part dependent on specifically what one wants the theory to do. See, e.g., Thagard (1978b). See also Leiter (2001).
A desideratum is not the equivalent of a necessary condition: it is a criterion of judgment that must, or at any rate should, be fulfilled insofar as possible. Thus theory \(H\) could be lacking in internal consistency or explanatory power and still qualify as a theory. However, to the extent that it lacks these qualities, this counts as a deficiency. Specifying exactly what conditions must be fulfilled for something to qualify as a theory, especially across all possible conceptual domains, is very difficult best. It is in any event beyond the scope of this paper.
Huemer (2005, p. 103) makes a similar point in a different context : “...moral intuitions are not in general caused by antecedent moral beliefs, since moral intuitions often either conflict with our antecedently held moral theories, or are simply unexplained by them”: my emphasis.
I regard this stipulation as relatively safe; for many current philosophers have written convincingly in defense of the possibility of foundational belief and knowledge, e.g., Audi (2011b) and Moser (1985); and there are many others. Even Laurence BonJour, once a leading critic of foundationalism, has backed off: cf. BonJour (1985) with (1999).
Unconscious inference seems quite implausible to me. But cf. Huemer (2005, p. 267, n4).
I am not claiming that it is inevitable that I or anyone else would claim to know, or even assent to, P1. For the valuational background may be such that we would intuit the falsehood of P1. This, however, I would argue is an erroneous intuition, and indeed a very unlikely intuition. So far as I can tell, it is a universally held cultural norm that killing under such circumstances is wrong, and it is exceeding unlikely that an individual would be inclined to judge counter to this norm. In fact I think that we would be strongly inclined to consider anyone who judged against this norm to be suffering from a psychological pathology.
But even the geologist, the astronomer, and the nuclear physicist will need to rely on testimony at some point, for none can be expected to have non-testimony-based, first-hand knowledge of all the theoretical elements involved in grounding her theoretical knowledge: no contemporary scientist does, or could in today’s complex scientific world. Science depends centrally on the epistemic respectability of testimony. In some of these cases I understand the relevant theories well enough to have some theoretical grasp of the truth of these propositions; but as a layperson, I’m largely taking it on faith—on testimony.
This is a somewhat Deweyan point, although Dewey would disagree with my general view. The aspect that Dewey would approve of is that physical operations, such as pouring liquid from a beaker, are also part of science; there is no sharp break between the “scientific” and the “non-scientific.” The aspect with which he would disagree, however, is my view that some belief/knowledge is pre-theoretical. For Dewey, it is all theoretical.
An issue of semantic pre-theoreticality here.
An issue of epistemic pre-theoreticality here. Deep familiarity with normative ethical theories can yield intuitions of this sort. One just “sees” that a theory, being of a preferred type, is better than a different type of theory.
I owe this point to an anonymous reviewer for this journal.
I have argued for the supervenience of first-order moral properties on physical properties in Kulp (2011). On my view our epistemic access to these properties is not empirical, at least not in any primary sense.
I believe, but will not defend here, that first-order moral truths are, for the most part, synthetic a priori.
Dewey would agree that beaker-pouring, metal-counting and heating are part of the conduct of science. Also, see Post (1986) for the metaphysical implications of the integral scientific role of mathematics in much of contemporary science.
My colleague Mane Hajdin advanced this objection to me.
One may also construe this objection as applying specifically to moral claims. But then we are owed an account of why moral claims are especially problematic. I discuss associated matters in detail as applied to moral intuitions in Kulp (2014).
P3 also fails the non-inferential criterion for intuitions.
The challenge may, for example, issue out of a tricky case of illusion, requiring sophisticated perceptual theory to mount a satisfactory account of what is being perceived. Broadly analogous complexities may be present in moral cases.
See, e.g., (Audi (2011b), pp. 229f, 352f.). My example: I may know that I am thinking about my wife; but how can I show it, other than by repeating my claim—which is not to “show” anything?
I wish to thank Robert Audi, Mane Hajdin, and two anonymous reviewers for this journal for their invaluable comments on previous versions of this article.
References
Armstrong, A. H. (1981). An introduction to ancient philosophy. Totowa: Rowman & Allenheld.
Audi, R. (2004). The good in the right. Princeton: Princeton University Press.
Audi, R. (2011a). Intuitions, intuitionism, and moral judgment. In J. G. Hernandez (Ed.), The new intuitionism (pp. 171–196). London: Continuum.
Audi, R. (2011b). Epistemology (3rd ed.). New York: Routledge.
Ayer, A. J. (1952). Language, truth, and logic. New York: Dover.
Blackburn, S. (1993). Essays in quasi-realism. Oxford: Oxford University Press.
BonJour, L. (1985). The structure of empirical knowledge. Cambridge: Harvard University Press.
BonJour, L. (1999). The dialectic of foundationalism and coherentism. In J. Greco & E. Sosa (Eds.), The blackwell guide to epistemology (pp. 117–144). Malden: Blackwell.
Boyd, R. (1988). How to be a moral realist. In G. Sayre-McCord (Ed.), Essays in moral realism (pp. 181–228). Ithaca: Cornell University Press.
Brink, D. (1989). Moral realism and the foundations of ethics. Cambridge: Cambridge University Press.
Cornford, F. M. (1957). From religion to philosophy. New York: Harper.
Dewey, J. (1910/1978). How we think. In J. Boydston and B. Graubner (Eds.), John Dewey: The middle works, 1899–1924 (Vol. 6, 1910–1911, pp. 177–356). Carbondale and Edwardsville: Southern Illinois University Press.
Dewey, J. (1938/1985). Logic: The theory of inquiry. In J. Boydston (Ed.), John Dewey: The later Works, 1925–1953 (Vol. 1938, pp. 3–527). Carbondale and Edwardsville: Southern Illinois University Press.
Churchland, P. (1988). Perceptual plasticity and theoretical neutrality: A reply to Jerry Fodor. Philosophy of Science, 55, 167–187.
Fodor, J. (1984). Observation reconsidered. Philosophy of Science, 51, 23–43.
Fodor, J. (1988). A reply to Churchland’s ‘perceptual plasticity and theoretical neutrality’. Philosophy of Science, 55, 188–198.
Huemer, M. (2005). Ethical intuitionism. New York: Palgrave MacMillan.
Hyland, D. A. (1975). The origins of philosophy: From myth to meaning. New York: G. P. Putnam’s Sons.
Jones, W. T. (1969). A history of western philosophy: The classical mind (2nd ed.). New York: Harcourt, Brace & World.
Kagan, S. (2001). Thinking about cases. In E. F. Paul, F. D. Millar, & J. Paul (Eds.), Moral knowledge (pp. 44–63). Cambridge: Cambridge University Press.
Kulp, C. B. (2011). Moral facts and the centrality of intuitions. In J. G. Hernandez (Ed.), The new intuitionism (pp. 48–66). London: Continuum.
Kulp, C. B. (2014). “Disagreement and the Defensibility of Moral Intuitionism,” an invited paper presented at a conference, “New Moral Intuitionism: Its Possibilities and Challenges,” at the University of Gdansk, Poland, June 6–7.
Leiter, B. (2001). Moral facts and best explanations. In E. F. Paul, F. D. Millar, & J. Paul (Eds.), Moral knowledge (pp. 79–101). Cambridge: Cambridge University Press.
Mackie, J. L. (1977). Ethics: inventing right and wrong. New York: Penguin.
Moser, P. (1985). Empirical justification. Dordrecht: D. Reidel.
Popper, K. (1962). Conjectures and refutations: The growth of scientific knowledge. New York: Basic Books.
Post, J. F. (1986). The faces of existence: An essay in nonreductive metaphysics. Ithaca: Cornell University Press.
Shaffer-Landau, R. (2003). Moral realism. Oxford: Oxford University Press.
Sinnott-Armstrong, W. (2009). Moral Perception and Heuristics. The Modern Schoolman, LXXXVI (March and May): 327–347.
Stevenson, C. L. (1944). Ethics and language. New Haven: Yale University Press.
Sturgeon, N. (1988). Moral explanations. In G. Sayre-McCord (Ed.), Essays in moral realism (pp. 229–255). Ithaca: Cornell University Press.
Thagard, P. (1978a). Why astrology Is a pseudoscience. Proceedings of the Philosophy of Science Association, 1, 223–234.
Thagard, P. (1978b). The best explanation: Criteria for theory choice. Journal of Philosophy, 75(2), 76–92.
Tropman, E. (2012). Can cornell moral realism adequately account for moral knowledge? Theoria, 78(1), 26–46.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kulp, C.B. The pre-theoreticality of moral intuitions. Synthese 191, 3759–3778 (2014). https://doi.org/10.1007/s11229-014-0497-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-014-0497-z