The central problem of philosophical ethics is whether it is possible to provide rational justification for our moral values.¹ This problem typically comes to expression in the form of the relation between "facts", or "what is the case", and "values" or "what ought to be the case".² The problem of course looms large in any intended program of moral education. For as the etymology of the word indicates, education involves a drawing out or leading ("e" and "duco"), and moral education thus involves some sort of drawing out or leading in the area of morality. I leave aside here the question of whether or in what sense this entails some form of indoctrination.³ My concern bears rather on the nature of the substantive moral values one might wish to build into one's educational program. If and to the degree that moral values are intrinsically irrational or non-rational matters, it would seem that there can be no good reason for urging their appropriation in any such program. My question, then, is whether it is possible to affirm moral values as rational sorts of affairs. Is it possible, as a matter of principle and hence in any given instance, to distinguish rationally between moral good and evil? As this formulation of the question suggests, my intention is to consider, not which particular substantive moral judgments are to be taken rationally as good,⁴ but rather whether and in what sense substantive moral judgments as a matter of principle might be so taken.

It is a commonplace that much of contemporary educational practice and indeed of contemporary philosophy provides us with forms of a negative answer to this question. For example, the influential values-clarification approach to moral education, as the name itself suggests, tends to focus its concern on helping a student to clarify or elucidate the values which he or she already has, rather than on trying to move to a critical judgment about the content or substance of such values.⁵ The reason for this restriction of focus to the process of valuing rather than to the specific character of particular value judgments seems to be the assumption, adopted at least provisionally or for operational purposes, that there exists no objective, rationally justifiable, moral standard in terms of which such judgments might be assessed.⁶ The values-clarification approach to moral education, in other words, seems to instantiate a mixture of the types of ethical theory called naturalism and non-cognitivism.⁷ On the one hand, moral values are assumed, at least provisionally, to be a simple function of the contingent facts of one's culture or psychology. On the other hand and often simultaneously, moral values are assumed, again at least provisionally, to be matters of simple preference, and thus the sorts of things about which one cannot, strictly speaking, argue. The common import of such assumptions, then, is that they serve to remove morality from the arena of argumentation and justification.

One of the overarching concerns of the studies which make up this volume is to develop a philosophy which justifies some sort of education in the area of substantive moral claims. Such a concern commits one in principle to a rejection of the relativism and non-cognitivism carried in the philosophic assumptions of values-clarification as just noted, and hence to the task of defending some kind of objective standard for morality. But I should like to suggest at the outset that even the values-clarification theorists who adopt the above assumptions cannot so easily dismiss the foundational question about an objective standard for morality, that is, even from within their own assumptions. For as values-clarification theorists themselves have increasingly come to acknowledge, however much they may disparage efforts to make--or to evaluate critically--substantive moral claims in any direct or straightforward way, their very adoption of clarification as the "correct" mode of education carries within it, at least implicitly, an espousal of such values as tolerance, self-actualization, authenticity and the like, over their opposites.⁸

Of course, as Ellrod notes, the fact that values-clarification theorists increasingly acknowledge their own implied values does not mean that they have thereby justified those values.⁹ But this serves to focus on just the point I wish to make here: namely, that if in fact the very practice of value neutrality or value-relativity in one sense (for example, in terms of whether it is good to steal in this particular instance), carries within it a non-neutrality or non-relativity in another sense (for example, in terms of whether it is good to be tolerant of different views regarding the goodness of stealing in this particular instance), then it would seem to follow that the question of the warrants or foundations of morality does in fact arise even for one who adopts the assumptions of value-neutrality or relativity.

The possibilities are twofold: (1) the values-clarification theorist can deny that he or she implies in his or her practice the "correctness" of such values as democratic tolerance and the like. In such a case there is no demand for providing a warrant for one's practice, but only because one is thereby saying that one has exactly no reason, implicit or explicit, for subjecting students to the process of clarification--rather than, say, to a shootout--and for encouraging other educators to do the same. Since it is difficult to maintain such a position if one continues to defend the practice of clarification in preference to a shoot-out, (2) the values clarification theorist increasingly acknowledges his or her implied judgment of preference for the values of democratic tolerance and the like. In acknowledging such an implied value judgment, the values-clarification theorist is committed to showing the warrants for exempting such a value from the neutrality and relativity which he or she assumes regarding values generally.

It is important that this brief suggestion not be taken in a stronger sense than is intended. For the values-clarification theorist might respond simply by saying that selecting the value, say, of democratic tolerance for preferred status has no other warrant or "reason" than the fact that it is a value which happens to be shared in the society in which we live. The value of democratic tolerance, in other words, might be taken to be nonrelative only because it is as a matter of convention taken to be such by--and for-those who live in our society, and not because there are good reasons in the strict or strong sense for taking it, reasons, that is, which could be defended as binding in a cross-cultural or indeed universal sense. My initial suggestion, then, is only that, insofar as the practice of values-clarification implicitly involves taking exception in some form to the values clarificationist's professed value-neutrality, there is an internal demand for one who engages that practice to give some accounting for the exception.

On the one hand this should not, a priori, be taken to signify a demand that such accounting be made in terms of good reasons which hold trans-culturally or universally, for that would beg the question of whether there are such--a possibility which the values-clarificationist, in offering conventionalism as a response, precisely rejects. On the other hand, and this is the point of my initial suggestion, the demand for an accounting does seem to me to commit the values-clarificationist a priori to considering all the logically possible ways of explaining the exception involved in his or her performance. One cannot simply or a priori assume that the conventionalist response is the only, or indeed the most adequate, way of accounting for the exception, for this would also beg the question, now from the other direction: given some elements of non-neutrality and nonrelativity carried in any performance of values-clarification, the issue which demands adjudication is precisely that of the sort of "reasons" which suffice finally to account for these elements. That conventionalism is the most adequate explanation available must therefore be argued in light of other possibilities, notably the possibility that there are "natural" reasons or foundations for non-neutrality and non-relativity.

This demand for argument, then, sets both the context and limits of the present article. My purpose will be to show (a) that there is a basic philosophical claim which is operative in the conventionalist account of morality and indeed serves precisely to dictate that account, and to show (b) that there is available an alternative, contrary philosophical claim which makes possible a "natural"¹⁰ account of morality. My purpose, in other words, will be to show the logical possibility of a morality which is natural rather than merely conventional, and thereby to establish the gratuity of the claim that conventionalism is the only possible, and hence most adequate, response to the question of morality.

Among contemporary moral education theorists Lawrence Kohlberg has stood out as one who rejects the claim that we can adequately account for morality in conventionalist terms. He claims that good reasons, that is, reasons which bind cross-culturally or universally, can be given for our conception of morality and thus for our moral judgments. In a word, Kohlberg challenges the non-cognitivism and relativism operative in much of contemporary ethical theory. He does this in the name of a kind of "naturalistic" morality, that is, a morality which has a "natural" and hence both cognitive and universal foundation. Such a position is consistent with the intention of this chapter which therefore will proceed by examining the position of Kohlberg, both in terms of the criticisms he makes of contemporary trends in ethical theory, and in terms of the warrants he provides for his own alternative theory. The purpose of the chapter in so doing will be threefold: it will attempt to distill what I take to be the central philosophical claim operative in the intention of his argument, a successful defense of which is consequently necessary to sustain that argument. Secondly, it will attempt to show that this (implicit) philosophical position is in an important way in tension with the (Kantian/formalist) philosophical tradition on which he explicitly relies for support for his theory. Finally, the chapter will suggest that there is available another philosophical tradition (Aristotelian-Thomistic) which makes just the sort of claim which seems to me to be necessary to allow Kohlberg to realize the intention of his argument. The overarching aim, then, in engaging this threefold task, will be both to call attention to the (assumed) philosophical position which has dictated the rise of non-cognitivism and relativism, and to show the logical possibility of an alternative philosophical position which makes possible a view of morality which is both cognitive and normative.

There is no need to rehearse the details of Kohlberg's stages of moral development. The major point relevant to our topic is Kohlberg's claim that these developmental stages translate into a moral hierarchy, that is, his claim to "have successfully defined the ethically optimal end point of moral development."¹¹ Our concern bears on the warrants which Kohlberg provides for such a claim, and this concern leads us into consideration of what is called the "naturalistic fallacy".¹²

Stated broadly, the naturalistic fallacy is taken to consist in the claim to have established ethical conclusions beginning from factual premises. Or, to put it in the terms used at the outset of this article, the naturalistic fallacy as ordinarily understood consists in making values, or what ought to be the case, a function simply of facts, or what is the case. I have already noted above the assumptions connected with this fallacy in much of contemporary social science. On the one hand, insofar as contemporary social science subscribes to the mode of reasoning reflected in the naturalistic fallacy as stated, that science collapses into relativism. If it is not possible to make any distinction between facts and moral value, then a factual variety of moral values from person to person and culture to culture implies a relativism of moral value. On the other hand, insofar as contemporary social science rejects this position of naturalism, it typically does so by separating moral values from facts. That is, moral values are not taken to be matters of fact at all, but rather matters of preference. Strictly speaking, then, values are no longer taken to be relative, but only because they are not the sorts of things about which one can argue whether they are absolute or relative. Values, in short, are not cognitive matters at all. The upshot of these assumptions, then, as we have seen, is that values are either relative or cognitively meaningless. In either case, they are not susceptible of rational justification. Kohlberg claims to have developed a theory which warrants rejection of these assumptions. In order to see how and indeed whether his theory does so we turn to an examination of Kohlberg's position in relation to naturalism.

Kohlberg gives a summary statement of that position as follows:

To begin with, there are two forms of the `naturalistic fallacy' we are not committing. The first is that of deriving moral judgments from psychological cognitive-predictive judgments or pleasure-pain statements, as is done by naturalistic notions of moral judgment. Our analysis of moral judgment does not assume that moral judgments are really something else, but insists that they are prescriptive and sui generis. The second naturalistic fallacy we are not committing is that of assuming that morality or moral maturity is part of man's biological nature, or that the biologically older is the better. The third form of the `naturalistic fallacy' which we are committing is that of asserting that any conception of what moral judgment ought to be must rest on an adequate conception of what it is (FITO, p. 222).

The second form of ethical naturalism which Kohlberg is rejecting here is rather straightforward, and for our purposes can be disposed of quickly. For the kind of moral development Kohlberg suggests is not identified with growth in biological maturity. It is simply not the case that, as one becomes older, one necessarily moves to higher stages of moral reasoning. The move from stage to stage is a logical, not a biological sequence. Of course it is true that an ethically higher stage must come later, but this is precisely because it is ethically higher or more adequate; it is not ethically higher because it comes later (FITO, p. 181). In short, for Kohlberg it is philosophic adequacy, and ontogenesis, which determines moral hierarchy (FITO, p. 181).¹¹³ There are therefore two forms of naturalism with which we must concern ourselves here: the first one noted above, which Kohlberg rejects, and the third one, which he embraces. Why does Kohlberg reject the first and accept the third?

Kohlberg's criticism of the first or bad form of naturalism consists in pointing out what he takes to be a series of confusions among contemporary social scientists which locks them into this bad form. The first confusion is that between ethical and cultural relativism. The cultural relativist affirms that "moral principles are culturally variable in a fundamental way" (FITO, p. 156). The ethical relativist makes the further claim that such divergence is logically unavoidable, that is, that there are not rational principles and methods which could reconcile observed divergencies of moral beliefs" (FITO, p. 156). Stated in a positive form, then, the confusion is between the notion that everyone has his or her own values and the notion that everyone ought to have his or her own values (FITO, p. 156). Stated in negative form, the confusion is manifest in the "move from `There are no universal human values' to `There ought not to be any universal human values', every person or culture ought to do his thing." (FITO, p. 158) Kohlberg argues that such a move commits the naturalistic fallacy by identifying a factual judgment with a value judgment. More specifically, the collapsing of cultural relativism into ethical relativism involves an ignoring of what is called the "open question": granted that person or culture "x" values "y", we can sensibly ask, "But is that good?" or "Is it right?" To deny the legitimacy of this further question would be to continue to assert the very assumption for which some warrant is being sought.¹⁴

The second confusion which leads to a fallacious and relativistic form of naturalism is that between "ethical relativity" and "ethical tolerance", that is, between "the relativity of moral principles" and "the relativity of blaming or punishing persons or groups who do not act in accordance with those principles" (FITO, p. 159). The claim here, in other words, is that, to avoid sitting in judgment on others and assuming moral superiority for ourselves, we must accept an equality of moral principles, which is to say we must deny that there are any universal moral principles. Kohlberg's response to this position is twofold: first, the universal moral principles of obligation do not oblige one to blame the persons who deviate from these principles; and, secondly, the very urging of the principle of tolerance as applying to all human beings itself entails a denial of ethical relativity, for the principle itself is affirmed thereby as non-relative.

In short, then, Kohlberg rejects a kind of reverse of the "naturalistic fallacy" just discussed: namely, that wherein an "ought" (value) of tolerance, linked with an "is" (fact) of cultural relativism, generates ethical relativism. The "ought" of tolerance is assumed (falsely) to require an identification of ethical relativism and cultural relativism.

The third fallacy behind much of contemporary social-scientific thinking has been its "confusion of ethical relativism with `value neutrality' or `scientific impartiality'" (FITO, p. 161). Kohlberg cites as an example here the claim of Berkowitz that it is neutral to define moral values as "`evaluations of actions generally believed by members of a given society to be either "right" or "wrong"'" (FITO, p. 161). Kohlberg's response is that this is not neutral because it prejudges the facts. To put it another way, it fails to distinguish between "an a priori definition of morality in terms of cultural relativity, and the conclusion that morality is culturally relative" (FITO, p. 161). The point here is that Berkowitz's definition is not necessarily wrong but that, because it involves a theory which is adopted in advance of inquiry and hence informs the inquiry, it is not a neutral approach. On what grounds, for example, other than a priori definition, does Berkowitz rule out as an adequate understanding of moral value that of the Catholic priest who defines it in terms of belief in the catechism (FITO, p. 162)? In short, the point here is, once again, not that Berkowitz is wrong, but merely that he is not being value-neutral, that he is arbitrarily ruling out of court alternative understandings of morality, and that their thus being ruled out requires justification.

Furthermore, suppose a similar strategy were employed with respect to scientific beliefs. That is, suppose that scientific beliefs were defined as "beliefs about the world generally believed by members of a given society to be true or false" (FITO, p. 161). This would entail that we accept as equally "scientific" the most sophisticated forms of scientific theory and the crudest imaginable forms of superstition. But this simply belies what we do as scientists. We would recognize one who equalized matters in this fashion simply to be a poor scientist. Hence, argues Kohlberg contra Berkowitz, we must at least consider the possibility that the same situation obtains in the moral sphere (FITO, p. 161). A simply a priori rejection of this possibility will not suffice.

Kohlberg contends that Berkowitz's position derives from the view of men like Weber "who distinguish between a rational sphere of social-science methods and findings (`is'), and a sphere of value (`ought'), toward which a rational man or a scientist must take a stance of `value neutrality,' that is, recognize that his position is personal, arbitrary, and historically conditioned" (FITO, p. 162). Hence we are led to a fourth confusion, namely, that "between the `rational' as `the scientific or factual' and the `rational' as the value neutral." (FITO, p. 162). Kohlberg argues that this position once again assumes ethical relativity rather than justifies it. On what grounds does one rule out the possibility of establishing rational methods of coming to ethical agreement without a critical assessment of those methods? Indeed, Weber himself is taking a value stand in defense of the value of value-neutrality, and in doing so he attempts to support it with careful rational arguments. The point here, then, is twofold: first, the rejection of rational arguments in the area of ethical value should not be assumed, as it is in the case of men like Berkowtiz and Weber; rather such rejection demands a justification. Secondly, the defense of neutrality exemplified in Berkowitz and Weber itself involves taking a value stance, and doing so by an, at least implicit, appeal to the kind of rational argumentation they have already ruled out in such a context (FITO, p. 162).

What is common to these confusions then, which are linked with the first form of the naturalistic fallacy, is the assumption that moral judgment, as distinct from factual judgments finally have no cognitive status. Rational methods or modes of rational argumentation are identified with what is understood to be scientific method. The latter is limited to description of the facts of valuing among different persons or cultures. Since these facts of valuing are seen to vary, they are, given the assumed identification of the rational (hence "objective") with the scientific/descriptive, labelled mere preference, that is, something of a noncognitive sort.

Kohlberg's criticism of this form of naturalism, which makes moral valuing simply derivative from and finally reducible to psychological fact, is that at least as operative in much of the thinking of contemporary social scientists, it is advanced not as a conclusion, but rather is asserted in the form of a priori assumption. There is no justification for assuming in advance of any inquiry into the matter that we cannot find moral development of a genuinely hierarchical sort, and that it is impossible in principle for moral thinkers to come to agreement regarding moral matters on the basis of rational argumentation (FITO, p. 163). This assumption by social scientists is belied by their practice, which would seem both to instantiate some sort of ethical non-relativity and non-neutrality, and to carry at least an implicit appeal to some sort of rational argumentation on their behalf.

Kohlberg's criticism of the dominant form of naturalism then, is similar in many ways to the criticism I sketched at the outset of this article in relation to values-clarification. But his criticism would be incomplete and indeed innocuous without a second, more positive claim: namely, that careful empirical inquiry suggests findings which contradict the assumption of relativism built into the prevalent fallacious form of naturalism (FITO, pp. 163ff). As this suggests, Kohlberg at once understands himself to be involved in a form of naturalism, and claims that his form is of a sort which justifies non-relativism. This position claims both to involve a description of the facts of moral development, and that this very description in turn allows a prescription in terms of a hierarchy of kinds of moral judgment. Consideration of how Kohlberg advances this claim will situate us properly to begin reflection on the philosophical issue(s) which I take to be involved in the defense of his claim.

If Kohlberg is to be successful in his claim to offer a distinct alternative to the form of naturalism which he considers defective in its assumed relativism, then it seems he must, on his own terms, meet two requirements. First, he must show that moral judgments have some internal connection with facts in the ordinary sense, in order to be able to sustain the third or benign sort of naturalism which would seem necessary to maintain the cognitive status of moral judgments. Secondly, he must show that moral judgments at the same time are in some sense distinct from facts in the ordinary sense, in order to elude the (first or fallacious) sort of naturalism which would collapse moral judgments into relativism. How does he manage to walk this tight-rope?

To begin with, Kohlberg claims that empirical studies of moral development disclose as a matter of scientific fact that "there is a universal moral justice" (FITO, p. 223; see also pp. 163-180). It is important to be clear about the nature of this claim. This moral form about which Kohlberg intends to be making a factual claim assumes a distinction between fact and value. As Kohlberg puts it, and this is the second crucial element in his position, "the moral man assumes that his moral judgment is based on conformity to an ideal norm, not on conformity to fact" (FITO, p. 223). In other words, the form of naturalism to which Kohlberg thinks it necessary for any moralist to subscribe requires that judgments about what morality ought to be must begin with some characterization of what it is (FITO, pp. 222-223). If we attend carefully to the nature of morality as revealed in cross cultural studies, however, we find that in the minds of moral men morality is something distinct from what is taken to be fact in the ordinary sense. Moral men in various cultures consider their moral judgments to be founded on conformity to an ideal, rather than on conformity to facts, in the sense of what happens to be the case in their culture. To put it another way, moral men in various cultures consider what they ought to do in any given instance to be logically distinct from the empirical fact of their (or their culture's) doing it. Even if what one judges ought to be done coincides with what one does, empirical cross cultural studies of the facts of moral experience reveal that the moral man does not found his sense of obligation on fact in this sense. Rather, they show that the moral man experiences his moral judgment as prescriptive on its own terms, that is, precisely as distinct from his or his culture's psychological facts in the ordinary sense.

These two elements in Kohlberg's position, then, lead us to affirm the following regarding his understanding of the relation between facts and moral judgments (see FITO, p. 223). On the one hand, moral judgments are sui generis; hence what is ordinarily understood to be a scientific description of the facts can never justify them or pronounce on their worth. On the other hand, such a description of the facts does play an integral role in that it can tell us whether this distinctive concept of moral judgments accurately obtains. In other words, the claim that "x" is what morality is in the minds of moral men is a factual claim of the usual sort which can be tested through empirical study: morality either is this in the minds of moral men, or it is not. Only empirical qualitative observation of various cultures can determine this. But, secondly, the qualitative dimension of the claim made on behalf of one's conception of morality precisely as normative involves a mode of reflection distinct from empirical observation. Disclosure of what moral men do consider to be an adequate conception of morality leaves us in the domain of facts in the ordinary sense. That men do as a matter of fact claim that "x" is an adequate conception of morality is logically distinct from whether they ought to. The latter claim, therefore, requires a distinct justification.

The following quotations from three of Kohlberg's studies will I think serve to summarize both how he understands the relationship between fact and morality and how he justifies that understanding.

(The) isomorphism of psychological and normative theory generates the claim that a psychologically more advanced stage of moral judgment is more morally adequate, by moral-philosophic criteria. The isomorphism assumption is a two-way street. While moral philosophical criteria of adequacy of moral judgment help define a standard of psychological adequacy or advance, the study of psychological advance feeds back and clarifies these criteria. Our psychological theory as to what individuals move from one stage to the next is grounded on a moral philosophical theory which specifies that the later stage is morally better or more adequate than the earlier stage. Our psychological theory claims that individuals prefer the highest stage of reasoning they comprehend. This claim of our psychological theory derives from a philosophical claim that a later stage is "objectively" preferable or more adequate by certain moral criteria. This philosophic claim, however, would for us be thrown into question if the facts of moral advance were inconsistent with its psychological implications.¹⁵

However, we do hold a stronger position, claiming that while psychological theory and normative ethical theory are not reducible to each other, the two enterprises are isomorphic or parallel. . . . [We] have argued for a parallelism between a theory of psychological development and a formalistic moral theory on the ground that the formal psychological developmental criteria of differentiation and integration of structural equilibrium, map into the formal moral criteria of prescriptiveness and universality. If the parallelism were correct in detail, then formalist philosophers could incorporate an equilibrium concept as part of their normative ethical theory, and vice versa. The ultimate result would be a theory of rational moral judgment like that now present in economics, in which the theory of how people ought to make economic decisions and the way they do make decisions are very closely linked. What can warrant such a "parallelist" claim is only the fruitfulness of its results. I have argued that the fruitfulness of the parallelist assumption is revealed in the clear success of the psychological work based on it (FITO, pp. 224-5).

Epistemological and ethical principles guide psychological inquiry from the start. Thus, the strategy attempts to avoid the naturalistic fallacy of directly deriving judgments of value from judgments about the facts of development, although it assumes that the two may be systematically related. It takes as an hypothesis for empirical confirmation or refutation that development is a movement toward greater epistemological or ethical adequacy as defined by philosophic principles of adequacy (DAE, p. 484).

Let us move on, then, to our foundational question: given that Kohlberg's conception of the relation of psychology and morality is one of isomorphism, namely, that the criteria of moral development parallel the criteria of psychological development, what is his warrant for this conception? What warrants his adoption of these criteria of moral development, precisely as moral. If I have accurately described Kohlberg's position, there are three elements in the answer to this question: (1) Kohlberg assumes his principles of moral adequacy (prescriptiveness and universality) to be established on distinctly philosophical grounds, that is, in the context of a reliance on "the formalistic tradition in philosophic ethics from Kant to Rawls."¹⁶ (2) He then uses this distinctly philosophic assumption interactionally, or hypothetically, in relation to the facts of psychological development. That is, empirical study of the facts of psychology will either confirm or refute this original assumption. (3) Finally, Kohlberg claims that empirical study does confirm the original assumption. Hence his conclusion: the originally assumed and distinctly moral-philosophic criteria are justified.

The point I wish to introduce here is that this answer of Kohlberg seems to harbor an unresolved tension, and hence to need further unpacking. On the one hand, Kohlberg says that moral principles, precisely in their character as moral, must be justified on distinctly philosophical grounds, which he takes to be provided by the formalistic tradition stemming from Kant. In other words, the adequacy of these principles as moral is not dependent upon empirical inquiry. On the other hand, operating as a psychologist, Kohlberg takes over this philosophic justification, which then functions as a hypothesis subject to empirical confirmation or refutation. The second part of his argument thus seems to be that empirical studies play an intrinsic or internal role in determining what are finally to count as adequate moral principles. There results the following tension: namely, that morality is at once independent of facts and dependent upon facts.

Now it seems to me clear that some such position as this is necessary if one is to meet squarely the problem posed by the dominant forms of contemporary ethical theory. If morality is simply a function of facts in the ordinary sense relativism would seem to ensue; if morality is simply not a function of facts non-cognitivism tends to result. Kohlberg's intention is to elude these alternatives by--correctly in my judgment--entering a qualification. In one sense morality is a function of the facts, while in another sense it is not. But this serves to focus exactly the point I wish to make, namely, that, given his intention, the task incumbent upon Kohlberg is just that of justifying the sense of his proposed qualification, as the precise sense in which he takes morality to be at once internally related to facts and distinct from facts.

This statement of the task incumbent upon Kohlberg requires still further precision, for a third form of ethical theory, called non-naturalism¹⁷ has received support in recent thought, at least among philosophers. I take this theory to involve a threefold claim: first, that moral judgments are cognitive matters (in contrast to non-cognitivism); secondly, that moral judgments are nonetheless not cognitive matters of the ordinary sort, that is, the sort of factual matters which can be verified by empirical observation (in contrast to naturalism); and thirdly, that moral judgments are cognitive matters which involve a special sort of cognition, typically called something like "intuition" or "rational insight". Clearly, the whole weight of Kohlberg's argument is in agreement with non-naturalism in the first sense. My concern here bears on what the line of Kohlberg's argument entails in terms of its being situated relative to the second and third features of non-naturalism.

As we have seen, Kohlberg is explicit in his intention to embrace a form of naturalism, that is, to overcome a dichotomy between "is" and "ought" (FITO, pp. 154-55). He considers it possible, precisely within the context of this intention, to defend the claim that moral judgment possesses a distinctly necessary and categorical, and hence a normative, character. It seems clear, therefore, that Kohlberg is in agreement with nonnaturalism in ascribing to moral judgment a categorical or normative character. Nonetheless, at the same time he differs from non-naturalism regarding the possibility of defending such an understanding of moral judgment in "naturalistic" terms, that is, in terms of "what is" or "fact". In short, Kohlberg, in agreement with non-naturalism, is committed to defending the distinct categorical or normative character of morality. But, in contrast to non-naturalism, Kohlberg is committed at the same time to defending such a distinct character in terms of an internal or intrinsic connection with facts.

What I wish to suggest here then, is that Kohlberg's ethical theory, in sharing simultaneously the concerns of both naturalism and non-naturalism, is thereby distinct from either (as they are commonly understood). To put it more broadly in light of what I have written above, the theory which Kohlberg advances is an alternative to all three of the dominant forms of ethical theory. If I have correctly identified the central thrust of his theory vis-a-vis those dominant forms, then the philosophical task to which he is committed is that of justifying that/how morality is at once a fact, in contrast to non-cognitivism and non-naturalism, and a fact of a distinct sort, in contrast to naturalism of the usual variety. In a word, Kohlberg must justify the distinctness of morality precisely as a kind of fact.

In the face of this suggestion regarding the central philosophical claim to which the intention of Kohlberg's argument seems to commit him, my concern in what follows will be to make a threefold argument. First, I shall attempt to show that there has been in the tradition of Western modernity a common understanding of fact which has served to generate the three dominant contemporary forms of ethical theory. Secondly, I shall argue that the philosophical tradition on which Kohlberg expressly relies, namely the Kantian, does not break from this common understanding in a way which is sufficiently radical to provide a foundation for Kohlberg's intention of offering a distinct alternative to these dominant forms of ethical theory. Finally, I shall attempt to show that there is available another philosophical tradition, namely the Aristotelian-Thomistic, which does contain resources for breaking from that common understanding of fact in a way which is sufficiently radical to found the distinct alternative which Kohlberg intends.

First of all, then, as Kenneth Schmitz notes apropos of a contemporary dictionary account (Oxford English Dictionary, "Fact", n. 4) of "fact",

[t]he term designates a real occurrence ("actually the case"); it is "a datum of experience;" and it is what is "certainly known." This complex meaning of fact has been shaped within the problematic of human cognition, with attention to how realities can be known. It points, therefore, not only to the matter of evidence, to what is there, but also to the human conditions required to certify it. . . . In sum, then, the term "fact" has for its foreground, focus and surface what is actually the case, the evidence; but that evidence comes forward from a background of selective attention guided by an implicit understanding of what is significant for a distinctive kind of discourse. To speak of what is there as given fact is to speak within the circle of a discourse that directs attention to the matter insofar as the matter is capable of satisfying conditions that are determined a priori and in accordance with the demands of objective method.¹⁸

In this light I suggest that the salient features of the understanding of "fact" which set the context of the contemporary discussion regarding the nature of morality can be captured in what Schmitz terms an "objectification of nature".¹⁹ That is, a fact is typically taken to be whatever can be gotten at in terms of the external criteria of verification.²⁰ Such criteria involve an understanding of fact (nature) which can properly be termed at once mechanist, objectivist, and empiricist. Though there are of course numerous different ways in which these features come to expression in the thought of Western philosophers in recent centuries, the meaning I wish to ascribe to them can be indicated by turning first to Descartes.

Descartes, whom one might identify as a locus classicus for the distinctly modern understanding of fact, was certainly no empiricist. Nonetheless, his influence was decisive in terms of the development of what I wish to call mechanism and objectivism.²¹ That is, he adopts as his criteria for what can be affirmed as true--"real"--what can be gotten at clearly and distinctly. This, in turn, must be accessible in terms of the proportions and relations proper to mathematics (geometry),²² which is to say external spatial relations. And such relations are exactly the sort of relations which are proper to a machine.

Descartes' criteria for truth (what is to count as "real") do not really get transformed in the empiricist tradition. In other words, the empiricist tradition does not so much challenge Descartes' criteria as restrict the scope of their applicability to what is accessible to the senses. One need only refer here to Hume, who proceeds to give an account of fact ("matters of fact") exhaustively in terms of cause and effect, by which he means the external relations of discrete sense impressions or phenomena.²³ Hume cites as the perfect instance of the relation of cause and effect that of a billiard ball at rest which is then struck by another and thereby acquires a motion.²⁴

I do not at all intend to suggest with these brief remarks that the differences between Descartes and Hume are either few or unimportant, or in turn that there are not numerous ways of understanding fact (nature) in modernity which are in significant senses different from either Descartes or Hume. At the same time I do mean to suggest that in what one may call the tradition of Western modernity there has been a common assumption regarding the meaning of fact which is well illustrated by the central claims of Descartes and Hume as just noted.²⁵ That assumption is that reality or what is to count as real is what can be accounted for in terms of external (spatial) relations. The paradigm of causal activity is taken to be locomotion, or motion understood in terms of the displacement of physical bodies.²⁶ I call such an understanding of fact (nature) objectivist and mechanist. It is objectivist because it considers nature in terms of how it appears from without, and thus proceeds to give an account of the activities of nature in terms of the external influence of one thing on another (cf. Hume's billiard ball example). It is mechanist because external relations are best exemplified in, most proper to, the external behavior of machines. While Descartes represents a dualist variety of this mechanism and objectivism, Hume illustrates the empiricist variety which has been more common in the Anglo-American tradition. In a word then, I take the dominant understanding in the modern Anglo-American tradition to be one wherein what counts as a fact (nature) is what is both observable and sufficient to account for motion in the sense of the displacement of physical bodies.²⁷ In the language of the classical philosophical tradition, the causal activity of things is understood exhaustively in terms of a truncated view of efficient causality, to the exclusion of formal and final causes. Our question, then, is how this common Anglo-American empiricist variety of mechanism and objectivism shapes the problem of the relation between fact and morality.

The answer to this question lies simply in calling attention to the three results of excluding the formalizing and finalizing activities of an entity as proper causes. I suggest that those results, pertinent to the problem of the relation of fact and morality, are three. The removal of necessity in the sense meant here stems from the denial of form as a structural feature of each of the entities which make up nature. By form is meant that act of an entity in virtue of which it is identified as an entity of this or that kind. Form therefore is that which universalizes any given entity in the precise etymological sense ("universe": "turning toward one") of serving to unify or make into one the many aspects of an entity. The removal of form as an internal feature of what we mean by fact entails the elimination of the sort of necessity (necessary structural feature) of any given entity which would make it possible to universalize or make universal statements about that entity.

Secondly, the modern understanding of fact (nature) as described removes finalizing (or teleological) activity as a cause of an entity. By finalizing activity I mean that dynamic activity whereby an entity tends or seeks to be what it is. Finality is form in its dynamic, directional, purposive modality. The significance of such an exclusion of finality is suggested when we note the connection between finality or teleology and value. Whatever additional features we may wish to assign to what we mean by value,²⁸ a minimal and essential feature is value's connection with this tendency or seeking which we term finality or teleology. Since fact or nature (any given entity) becomes a value in itself or inherently valuable insofar as it is the object of its own finalizing activity, modernity's exclusion of finality from fact or nature makes it impossible to speak of any entities having intrinsic value, that is, value precisely in their character as fact or nature.

Thirdly, in understanding fact (nature) in terms of the external influence of one entity on another, modernity excludes form and finality as internal causes of an entity. In other words, modernity excludes any internal activity whereby that entity shapes itself and its environment in terms of its own immanent structure. As a result, insofar as modernity might wish to recoup the sort of necessity lost in its denial of form, it must do so in terms of forces outside any given entity. Thus, necessity becomes necessity of a deterministic sort. What an entity is, is not determined in any sense from within; it is in no sense self-determined. Rather what an entity is, is what is done to it, the sum of the forces acting upon it from without.

In sum, the dominant understanding of fact in modernity carries within it: (1) an inability to make universal statements about any of the entities which make up nature; (2) an inability to speak of the value of such entities in their character as factual or natural; and finally (3) an inability to speak of necessity with respect to such entities in anything but a deterministic way. For the problem of the relation of fact and morality this implies that one must break with the modern understanding of fact or nature if one wishes to speak of morality as both a matter of fact or nature and as requiring features such as universality, value, and non-determinism.²⁹

But let me be clear about the nature of this suggestion. I have made no claim as to what might count as an argument against the modern understanding of fact, or in favor of some other alternative which would incorporate the features I have noted as missing in the modern understanding. Nor do I mean to suggest that the sort of universality, value and non-determinism one might seek for morality need be, or is, simply the same as that sought for fact or nature generally. My point is simply that, if one wishes to affirm morality at once both as a fact and as requiring the features of universality, value, and non-determinism, then one must assign some universality, value, and non-determinism to what one means by fact.

To put it another way, if one continues to accept the modern understanding of fact, one must conceive morality in terms of the following alternatives. On the one hand, if one wishes to take morality as a kind of fact, one must understand morality as a non-necessary non-universal, and hence as a contingent and relative, matter. Any sense of value or obligation one might wish to assign to morality could be assigned only contingently and relatively. On the other hand, if one wishes to take morality as something universal, then, once again given the modern understanding of fact, one must understand morality not to be a matter of fact or nature. Any sense of value (teleology) or obligation (deontology) one might wish to assign morality could thus be found only outside fact or nature. In a word, given the modern understanding of fact or nature, the following dilemma arises: if naturalism (of the sort described by Kohlberg), then relativism; if non-relativism, then non-naturalism.³⁰

The point of the above excursus into modern philosophy's understanding of fact or nature has been to show how such an understanding dictates the rise of relativistic and non-naturalistic views of morality. My concern has been to show that, if one wishes to be neither a moral relativist nor a non-naturalist, that is, to be both a naturalist and a non-relativist, one must challenge just so far the modern understanding of fact or nature. The burden of my earlier argument with respect to Kohlberg was that such an intention of non-relativism and naturalism is fundamental to his undertaking. My question now is whether the Kantian tradition on which he expressly relies for philosophical support is able to help him successfully to carry out that intention. More precisely in light of the above discussion, my question is whether the Kantian tradition breaks with the modern understanding of fact in a way which would permit Kohlberg to found the universality and obligatoriness (deontology) of morality in nature as Kohlberg must do if he is to sustain his naturalistic intention.

My answer is that Kantian tradition does not do so, for a reason contained in the very starting point of Kant's philosophy. That starting point consists in an assumption and a question, to wit: given the Humean understanding of fact or nature, how can we rescue the sort of necessity required for an adequate view of both science and (in a different way) morality?³¹ Obviously the Humean context of the question shapes the direction of Kant's answer to this question: for in so far as the objects of nature do not, in accord with Hume's analysis, yield any features which are universal and hence necessary in the strict sense,³² such universality and necessity must be sought elsewhere than in these objects of nature in themselves. For Kant this "elsewhere" is in the human mind. In a word, then, Kant begins by accepting the Humean claim that nature has no necessity "in itself," and this in turn leads him to locate necessity in, and found it upon, the structure of the human mind which is something quite different from nature.

Further, while locating necessity in the human mind and in this way distancing himself from Hume, Kant nonetheless continues in a fundamental sense the mechanism of Hume. In other words, though now the causality proper to motion in the sense of the displacement of physical bodies (cf. Hume's billiard ball model) has necessity in the strict sense due to the structure of the mind, it nonetheless is still causality precisely in this mechanical sense. It is still the causality proper to external spatial relations. In a word, the study of nature for Kant does deliver necessity (due to the structure of the mind), but the necessity is nonetheless still characterized by external succession.³³ Given this mechanistic (if now idealistic) understanding of nature, there follow the sorts of consequences noted in a general way earlier: namely, any purposiveness (teleology) or obligation (deontology) one might be inclined to affirm must be located by Kant outside nature.³⁴

Thus, however much one might understand the objects of nature as having some purposiveness or finality, the point is for Kant that one can legitimately so understand them only as if they had such. To affirm finality as constitutive of nature in the same way as mechanical or efficient causality is constitutive of nature is precisely to go beyond what has been established as the limit of reason in its natural or empirical employment. In a word, insofar as one wishes to employ reason in a teleological sense, one can do so only insofar as one understands such employment to be merely regulative, rather than constitutive, of our knowledge of nature.³⁵

Secondly, the sort of necessity required by the "ought" of morality demands a kind of causality different from that of nature. The "ought" of morality demands freedom, which is "the power of a state beginning spontaneously",³⁶ and this sort of spontaneity runs counter to the situation of nature wherein one state of nature at any given moment follows of necessity from an immediately preceding state. In a word, the spontaneity and hence lack of determinism required by the kind of necessity proper to morality demands the positing of an order of freedom, which is other than the order of nature whose necessity is of a mechanical and hence deterministic sort.³⁷

The point of these brief considerations, then, is to suggest how Kant's understanding of the place of teleology and deontology (the "ought" of morality) in his system is a function of his assumed mechanistic understanding of nature. I say assumed because, in his starting point, Kant does not challenge but rather takes over Hume's mechanism. It is not a question for Kant of challenging the billiard ball model of causation; rather, having assumed that model as sufficient for the (constitutive) understanding of nature, the problem for Kant becomes one of finding a way of inserting a stronger sense of necessity into that model of causation than is provided by Hume. Of course, in locating necessity exclusively in the mind, Kant thereby affirms a dichotomy or dualism of nature and mind. But the mechanical character of that necessity in turn forces the two further dualisms which are directly pertinent to our concern. Given the mechanical character of the necessity of reason in its constitutive or natural employment, teleology and deontology--though in different ways--must ipso facto be extrinsic to reason in its natural employment. This means that, on a Kantian understanding of the matter, it is impossible to have any "natural" conception of morality (which involves teleology and/or deontology). There simply can be no intrinsic relation between the judgments of reason in its natural employment and the judgments of reason in its moral employment. In a word, a Kantian conception of morality is essentially non-naturalistic.

The implication of this for the argument of Kohlberg is clear: insofar as Kohlberg wishes to realize his intention of a naturalistic morality, he cannot rely on the Kantian tradition for his philosophic justification of that morality. The dilemma is this: if his conception of morality is naturalistic, then it must be non-Kantian; whereas if his conception of morality is Kantian, it just so far must be non-naturalistic. In other words, the Kantian position forces exactly the disjunction between nature ("is") and morality ("ought") which Kohlberg's naturalistic intention commits him to overcoming. In this sense, then, and for this reason, Kohlberg cannot seek the foundations for his conception of morality in Kant.

The overarching concern of this article requires that we press this Kantian issue further. Granted that Kohlberg faces a dilemma insofar as he tries to defend a naturalistic morality from within a Kantian perspective, why does he not resolve the dilemma simply by abandoning his naturalism? In other words, are there compelling reasons for Kohlberg's remaining a naturalist? It seems to me that there are, and that those reasons are two. First of all, as Kohlberg's research illustrates, people ordinarily do take morality to be connected with what we mean by fact or nature generally. When we defend our conception of morality, whatever the detail of that conception, it would seem that we finally make the claim, at least implicitly, that our conception of morality is adequate because "that is the way things are," or again because such a conception "fits the facts." Insofar as we claim that "x" is the most adequate conception of morality, that "x" therefore is what our conception of morality ought to be, it would seem that we mean to say, at least implicitly, that such a conception is warranted finally because that is the way morality is.

The issue raised here is of course profoundly difficult and delicate. For it is clearly the case that we can and do use terms such as "is", "facts", "the way things are" in different senses. Indeed the use of such terms in the context of morality provides an example of such a difference: though I use some such common phrases as "the fact is" in saying both that "the fact is that I ought to do `x'", and "the fact is that I do `x'", it seems clear that I mean something different in each case. Nonetheless this example serves to focus just the point I wish to make: namely that the meaning of "the fact is" in the two cases is not exhausted in the intention of difference. Rather, it seems to me, and the research of Kohlberg points to the same conclusion, that we mean what we say when we use the one phrase in both cases, and that there is thus (at least implicitly) some common/one meaning of fact which is simultaneous with, immanent within, the two distinct meanings of the term as used in the two cases. Insofar as one fails to allude to this commonness or oneness within the difference, insofar that is, as one attends only to the difference, one is just so far guilty of an equivocation which removes the warrant for continued use of that one term.

In short then, it seems to me, that the possibilities are two: on the one hand, we mean something equivocal when we speak of "fact" or "is" in the context of morality as distinct from the context of "nature". But if this is the case, then we should be able to find a way finally of justifying our conceptions of morality without appealing (at least implicitly) to some such common/one term as "fact" or "is" which we ordinarily use with respect to "nature". But Kohlberg's research shows, and indeed I believe our reflection bears out, that we do constantly--unavoidably--find ourselves making such an appeal. Insofar as this is the case, it becomes incumbent upon us to retrieve some sense of the unity of the meaning of "fact" ("is") within its distinctness as used apropos of nature and morality. This task of discovering some sense of unity between nature and morality commits one to the intention of naturalism. Nonetheless, it bears stressing that such an intention does not entail a rejection of the distinctness between nature and morality which is the concern of non-naturalism. The above criticism of Kant was not that he defended a difference between nature and morality, but that he defended a difference between the two in a way which neglected, which failed adequately to thematize, any unity between them. Kant thereby collapsed into a dualism which is inadequate, given the above argument regarding the implication of the way we use language.

A second reason for not abandoning some sense of naturalism at least in principle as a way out of the above dilemma is the lesson of history. Though it is impossible to rehearse the whole of modern and contemporary history relative to our problem,³⁸ it seems to suggest that a rejection of naturalism, a removal of morality from any anchoring in "facts" or "nature", leads almost inevitably--because it has in fact so led--to a non-cognitivist view of morality. However much one might want to defend the normative character of morality in non-naturalistic terms, that is, in terms which are different from what we ordinarily mean by "is" or "fact" or "nature", the lesson of history is that such an attempt inevitably gives way to the charge of gratuitousness. If morality is not accessible to us in the way in which "facts"--"reality"--generally are accessible to us, then it becomes increasingly difficult to claim that morality is "really" accessible to us at all. To the degree in which we wish to defend morality as something different from "facts" or "reality" in the ordinary sense, it becomes difficult to say that morality is nonetheless still cognitively meaningful. It seems to me that certain strains of both positivism and existentialism (though in different ways) provide eloquent testimony in support of this point.

For these two reasons, it seems that for Kohlberg the way out of the dilemma posed by his reliance on the Kantian tradition is not to abandon naturalism, but rather to abandon the Kantian tradition. More precisely, it is to seek a form of ethical theory which permits one to protect the distinctly normative or categorical character of morality, which is Kant's concern, but which does so while rejecting Kant's dichotomy between nature and morality which makes it impossible to meet the intention of naturalism. The burden of my argument above was that Kant's dualism of nature and morality was a function of his failing to challenge sufficiently radically the mechanism of Hume. Granted that Hume's nature receives in Kant a necessary character due to the structure of the mind, this necessity is exactly the necessity illustrated best by the operations of a machine. That is, the state of nature at any given instant is understood exhaustively in terms of the external influence of one state upon another. In a word, nature is viewed deterministically and as without purposiveness. It follows that, insofar as one wishes to insert into morality a distinctly purposive or teleological character,³⁹ or the deontological sort of necessity which requires freedom, one must seek the source of that teleology and deontology in something other than nature. In sum then, my contention has been that it is Kant's mechanistic understanding of nature which compels his dualistic understanding of the relation between nature and morality. If this is true, it follows that a necessary condition for overcoming a Kantian sort of dualism is to challenge that mechanistic understanding of nature.

It is just here then, that it seems to me, that the Aristotelian-Thomistic philosophical tradition is well equipped to help us. Of course there are different interpretations of that tradition in terms of the issue before us. Nonetheless, there is no need to engage here a discussion of the relative adequacy of these different interpretations. Rather, given the line of my argument thus far, it will suffice to show simply that the Aristotelian-Thomistic tradition can be shown to possess the kind of non-mechanistic view of nature needed to move beyond the dilemma of either a naturalistic relativism or a non-naturalistic dualism. Indeed, within this context, my intention is to show how that tradition must be interpreted if it is successfully to meet the necessary systematic conditions outlined above for overcoming that dilemma. My task, then, will be to show, first, how the Aristotelian-Thomistic tradition challenges a mechanistic understanding of nature, and secondly, how that tradition situates morality within a nature the mechanistic understanding of which has been so challenged.

The sense in which the Aristotelian-Thomistic tradition can be understood to challenge a mechanistic understanding of nature becomes readily apparent upon recall of our earlier description of mechanism. Very simply, the mechanistic view removes from nature the internal causes of the Aristotelia-Thomistic tradition, namely, form and finality. Consequently the states or activities of nature are understood exhaustively in terms of the effective influence from without of one entity or group of entities upon another. Successful understanding of any given natural entity thus hinges on understanding the forces outside that entity, given which that entity is (does) what it is (does). Successful understanding of an entity, in other words, requires no appeal to any internal activity of an entity whereby that entity actively determines itself to be what it is, or actively shapes its environment in terms of what it is determining itself to be. In short, nature as so understood is both without any inherent purposiveness and, insofar as it has necessity at all, that necessity can only be of a deterministic sort. The appropriate method for the study of nature so understood is the external one of empirical observation and verification.

Against the background of this mechanistic way of viewing things, then, the Aristotelian-Thomistic tradition affirms three basic metaphysical axioms of agency or effectivity, formality, and finality.⁴⁰ To begin with, these axioms are properly metaphysical, which means that they apply not simply to entities of this or that kind, but to all entities or instances of being, including those which make up what we call nature.⁴¹ But insofar as principles of agency, formality, and finality are operative in nature, the Thomist is committed to an inherently purposive and non-deterministic view of nature. Nature retains a sort of necessity due to the presence of form in the entities which make up nature, but the necessity is now seen as a dynamic one: what an entity is at any given instant is what that entity is actively shaping itself to be, in relation to the other entities which make up its environment. In a word, then, the three causes of agency, formality, and finality, understood as jointly present in any given natural entity, require a threefold affirmation. (a) Natural entities, possessing these internal causal activities, can no longer be properly understood in a deterministic way, that is, in terms of effective influence upon such entities from without. (b) Natural entities, possessing form, retain a mark of necessity and hence of universality: their activity is always of a specific sort. (c) Natural entities, possessing finality, always act in a way which bears a teleological character. Thus, if it is the case that, as I have argued above, a conception of morality which would be naturalistic, that is rooted in nature, requires a view wherein it is possible to speak of nature as bearing the features of universality, value, and spontaneity (non-determinism), I suggest that such a condition is met in the Thomistic understanding as here outlined. But of course realization of this condition, while necessary if morality is to be understood as natural, does not suffice if morality is taken at the same time to be in some sense distinct from nature. How then does Thomism meet this second or sufficient condition, which is the central concern of non-naturalism?

First of all, the causes which are operative in nature generally are for the Thomist operative also in human nature. It is simply the case now that those causes are given a human specificity due to the presence of a distinctly human form. That specificity we may call a rationality whose hallmark is self-consciousness. Thus a human being, like a natural being generally, is a locus of intelligible necessities which is at once dynamic and purposive. But as self-conscious, a human being is active and purposive in a distinct way which we call free. In other words, the non-determinism found in nature generally becomes in human nature, due to the presence of self-consciousness, the self-determination which is properly termed freedom.⁴² In a word, just as all natural beings actively seek an end or what is called good, so do humans, though humans do so consciously and freely. This assertion does not yet suffice to characterize human activity as moral: granted that humans do and indeed must act in this manner, does this warrant our saying that they ought to do so? Even granted that human activity is essentially or naturally characterized by an active seeking of the good, is there anything in such a claim that warrants our saying that humans therefore ought to seek the good?

To ask this question is to enter into a controversy which has plagued Thomistic ethical theory.⁴³ Relative to that controversy, it will suffice for my purposes to record an interpretation which I take to be both logically possible in terms of the texts of Aquinas, and satisfactory in terms of meeting the systematic requirements for an adequate conception of morality as those requirements have been outlined above. This interpretation makes two claims, which bear on Aquinas' statements that "the first principle of practical reason is one founded on the notion of the good, viz., that good is that which all things seek after," and that the primary precept of law is that "good is to be done and pursued and evil is to be avoided".⁴⁴ My twofold claim is simply that, while the truth contained in the first statement is the necessary condition for the note of obligatoriness contained in the second statement, that truth does not suffice to account for the note of obligatoriness.

On the one hand, then: were it not characteristic of human nature actively to seek the good, humans would not experience the obligation to do so, certainly not in any universal way. Human nature, in the sense of its active tendencies, is an inner condition of, and thus provides a necessary foundation for, morality. This is precisely what I take Aquinas to mean when he says that the first principle of practical reason is "founded upon (emphasis mine) the notion of good, viz., that good is that which all things seek after."

At the same time, this interpretation does not seem to entail the further claim that the judgment regarding human tendencies suffices to account for the note of obligation contained in the primary precept of law, which is to say, the first principle of morality. Indeed, that seems to me why Aquinas says that the principle which includes the note of obligatoriness "Good is to be done and pursued" (faciendum et prosequendum) is the first principle, which is to say a per se notum or self-evident principle of morality. The note of obligatoriness, or "ought," sounded first by Aquinas when he moves to the moral order establishes that order as moral. But if this is the case, then it follows that that note of obligatoriness is not reached simply by unpacking or reading off one's human tendencies. My awareness that I ought to seek the good is given immediately with, that is, as a distinct moment within, my tendency to seek the good; such awareness thus does not follow upon, in the sense of being inferred from, that tendency.

In sum, then, Aquinas says that the first principle of morality is founded upon man's natural tendencies. I take this to mean that such tendencies are the necessary foundation for morality: did we not have a tendency to seek the good, we would not experience an obligation to do so. On the other hand, Aquinas adds a distinct note of obligatoriness when he formulates the principle which he takes to be the first principle of morality. I take this to mean that there is a primitive experience of obligation which establishes human activity as moral: however much attention to our basic inclinations may disclose to us that we do seek the good (and indeed disclose the nature of the various goods we seek), the experience that we ought to seek the good (various goods) is given immediately with, and hence not inferred from, the awareness that we do seek the good.

The key to interpreting Aquinas correctly here and to summarizing the importance of his position relative to our earlier argument lies in recognizing that there is for Aquinas a primitive experience of the deon (being obligated) as well as the on (being). The use of such language to describe Aquinas' position seems to me both proper and instructive. It suggests that we have experiences of both "ought" and "is" which are irreducibly distinct (because given primitively), but which are nonetheless distinct within or as modalities of nature (on: being, what is). Insofar as this is the case, it seems to me, apropos of my earlier argument, that Aquinas is neither a naturalist nor a non-naturalist of the usual sort: the "ought" of morality is intrinsically related to (because a distinct modality of) the "is" of nature; but the "ought" of morality is at the same time irreducibly distinct from (because not inferred and hence not following from) the "is" of nature. Such a position, in cutting through the dilemma of either naturalism or non-naturalism, meets exactly, I suggest, the intention of Kohlberg's argument vis-a-vis contemporary ethical theories. In so doing it meets also the intention of this chapter.

In concluding, I think it is important to retrieve the main lines of my argument, and in so doing to focus once again what I take to be its scope. I began by noting that any program of moral education which intended to lead students in the area of substantive moral judgments was committed to rejecting a relativist or non-cognitivist view of morality. My further suggestion was that even those, for example the values-clarificationists, who claimed in theory not to be leading students in the area of substantive moral judgments belied such a theory in their practice: their practice seemed in fact to involve a non-neutrality and non-relativity regarding (say) democratic tolerance. My intention with regard to the values clarificationists, then, was to illustrate that even those who would most explicitly espouse value-neutrality and relativity thus could not successfully carry through such a position. My intention was thus to suggest that even would-be relativists and non-cognitivists must face up to the task of foundations, that is, of giving some accounting for the non-relativity which, however much they deny it in theory, they exercise in practice. Further, then, in this context my contention was that the conventionalist and hence relativist account which might be offered in support of their practice of non-relativity was precisely an account whose adequacies could not be assumed, but which required argument.

The demand for an argument from the conventionalist would be trivial were there not available, at least as a logical possibility, an account of morality which was not conventionalist. This led to what has been the burden of the chapter: to show that there is available such a logically possible alternative account of the foundations of moral value, one which might be properly termed a "natural" account. Since Lawrence Kohlberg has stood out among contemporary moral theorists as one who espouses the sort of "natural" account necessary to meet the challenge of conventionalism, my argument was worked out in relation to that of Kohlberg.

The substance of the argument then, at once in relation to conventionalism and Kohlberg, was to show how the modern Western understanding of fact had shaped the contemporary discussion concerning ethical theory. Specifically, my argument was that the modern understanding of fact or nature could be properly termed mechanistic, and that such mechanism was precisely what served to dictate the alternative ethical theories of naturalism and non-naturalism. On the one hand, insofar as morality is taken to be a matter of fact, morality just so far--that is, given the modern (e.g., Humean) understanding of fact--can only be a contingent and hence relative matter. Thus there results naturalism of the relativistic sort as described by Kohlberg. On the other hand, insofar as morality is taken to be non-relativistic, morality just so far--once again given the modern understanding of fact--can only be a non-factual matter. In other words, there results a non-naturalism which is characterized by a dualism of fact or nature and morality. The burden of my argument then, was to show that, because it was mechanism that dictated the alternatives of naturalism (of the relativistic sort) and non-naturalism, a challenge to mechanism was the necessary (though not the sufficient) condition for moving beyond these alternatives. In a word, my effort was to show that, insofar as one wished one's ethical theory to be neither relativist nor dualist, one must defend just so far some sort of non-mechanistic view of nature.

Within this context, the argument was twofold: (1) It was my contention that Kant accepted, in a fundamental way, the mechanistic understanding of fact and hence was forced, in seeking a distinctly normative conception of morality, to found it upon something outside fact or nature. In other words, Kant was seen to illustrate the second or non-naturalistic alternative noted above. My contention consequently was thus, that Kohlberg, insofar as he relied on the Kantian tradition for justification of his ethical theory, was just so far forced into a non-naturalism which belied his intention of naturalism, that is, of rooting his ethical theory in fact or nature. (2) Secondly, I suggested in this context that there was available another philosophical tradition, namely, the Aristotelian-Thomistic, which did in fact challenge the mechanistic understanding of fact of modernity, and which in so doing founded a possible way out of the impasse regarding ethical theory (that is, either relativistic naturalism or dualistic non-naturalism) dictated by that mechanistic understanding.

It is important to be clear about the scope of the argument as recapitulated here. I have not argued the preferability of a non-mechanistic understanding of fact or nature, such as that found in the Aristotelian-Thomistic philosophical tradition, over against the mechanistic understanding of modernity. The intention of this article is met in showing how one's assumptions regarding the meaning of fact or nature, that is, whether fact or nature is to be conceived mechanistically or non-mechanistically, bear upon one's ethical theory. If the argument has been sound, it follows that anyone who would defend the ethical theories of naturalism, non-cognitivism, and non-naturalism, all of which presuppose a mechanism, must provide an argument showing why such presupposed mechanism is to be preferred to non-mechanism as a way of understanding fact or nature. In view of the opening considerations of this article, this task is incumbent upon values clarificationists just so far as they wish to make anything other than a gratuitous claim upon others on behalf of their conventionalist ethical theory.

On the other hand, it follows that anyone who would defend an alternative to these three dominant forms of ethical theory, that is, who would defend an ethical theory which purports to be at once cognitive, non-relativistic, and natural, must provide an argument showing why non-mechanism is to be preferred to mechanism as an understanding of fact or nature. I would stress here, then, in light of my considerations of Kohlberg, that Kohlberg must either undertake this task, in order to maintain his intention of providing a distinct alternative to the three dominant forms of ethical theory; or he must be prepared to defend the mechanism which I have argued underpins Kantian ethical theory, and in so doing embrace non-naturalism.

Finally, then, I have suggested that an argument on behalf of non-mechanism is contained in principle in the Aristotelian-Thomistic philosophical tradition. Nonetheless I would stress that my suggestion was only that: My brief sketch of the central features of that tradition as it bears upon the problem of fact and morality requires fuller development, both in terms of its adequacy as an interpretation of that tradition, and in properly systematic terms.

Apropos of the requirement for such a fuller development, I must be content here in conclusion with re-focusing the precise issue I take to need an argument. The mechanistic understanding of fact consists essentially in considering things as they appear from without, that is, in their externality. Such an understanding involves both a content and a method. In accord with a method of observation and verification the mechanistic approach proceeds to give an account of things in terms of a truncated view of efficient causality (of effective influence from without). An argument on behalf of a non-mechanistic understanding of fact therefore must consist in justifying that/how things act also from within. Such justification must attend to both content and method. The idea of causality must be expanded to include internal causes (e.g., form and finality), and a method appropriate for the inclusion of such internal causes, that is, a method distinct from the objectivist method of observation and verification, must be worked out. In sum, then, if the argument developed in this article is sound, the need for such a justification of a non-mechanistic understanding of nature must be met, if not in the suggested Aristotelian-Thomistic terms then in some other. Otherwise, I submit, we are necessarily left in the contemporary impasse regarding ethical theory which consists in having to choose between relativism, non-cognitivism, and dualism.

University of Notre Dame

Notre Dame, Indiana

1. Alan Gewirth, Reason and Morality (Chicago: University of Chicago Press, 1978), p. ix.

2. W. D. Hudson, "Editor's Introduction: The `is-ought' problem", in The Is-Ought Question, ed. by W. D. Hudson (London: Macmillan, 1969), p. 11.

3. See Ch. X below.

4. Apropos of this issue, see Ch. IX above.

5. See the works by Sidney Simon, Howard Kirschenbaum and others, cited by Frederick Ellrod in nn. 1-2 in Ch. I above. Ellrod offers a brief description and assessment of the values clarification theory on pp. 10-19 of his article.

6. Cf. Ellrod, Ibid., p. 7 and pp. 4-17, passim.

7. For a description of naturalism and non-cognitivism, see Richard B. Brandt, Ethical Theory (Englewood Cliffs, N.J.: Prentice-Hall, 1959), pp. 151-153, and Ch. 7 and 9.

8. Ellrod, op. cit., pp. 7-11; p. 14.

9. Ibid., p. 14.

10. I use the term "natural" here in the sense of something which holds universally. The connection between "natural" and "naturalist" is of course a matter which is a major concern of this chapter.

11. Lawrence Kohlberg, "From Is to Ought: How to Commit the Naturalistic Fallacy and Get Away With It in the Study of Moral Development", in Cognitive Development and Epistemology, ed. by Theodore Mischel (New York: Academic Press, 1971), pp. 151-235, at p. 153. (Hereafter cited as FITO.)

12. It has been denied that the "naturalist fallacy" is in fact correctly termed a fallacy at all, and claimed that it is rather simply a mistake or at best but a species of the definist fallacy. See, for example, James M. Giarelli, "Lawrence Kohlberg and G. E. Moore on the Naturalistic Fallacy", Educational Theory, 26 (1976), 352. This distinction does not materially affect the line of argument taken in this paper.

13. Cf. also Lawrence Kohlberg and Rochelle Mayer, "Development as the Aim of Education", Harvard Educational Review, 42 (1972), 483. (Hereafter cited as DAE).

14. Cf. also in this context Kohlberg's rejection of what he terms "the psychologists's fallacy" and defines as follows: "the direct derivation of statements about what human nature, human values, and human desires ought to be from psychological statements about what they are." Kohlberg subjects this fallacy to the same criticism noted here, namely, that it ignores the "open question" of why a given fact of desiring or valuing should be called good. See DAE, pp. 466-67.

15. Lawrence Kohlberg, "Moral Development and Moral Philosophy", The Journal of Philosophy, LXX (1973), p. 633.

16. Ibid.

17. On naturalism, non-cognitivism and non-naturalism as the three prevalent forms of ethical theory, and for a brief description of each, see Richard B. Brandt, op. cit., pp. 151-153, and Chs. 7-9.

18. Kenneth L. Schmitz, The Gift: Creation (Milwaukee: Marquette University Press, 1982), pp. 40-41.

19. Ibid., p. 88.

20. Ibid.

21. Clearly Descartes is responsible in an important way for the turn to the subject characteristic of Western modernity. My point is simply that the criteria of knowledge which Descartes employs in his turn to the subject are criteria proper to objects. It is in just this sense that I take him to be objectivist.

22. Cf. inter alia, Discourse on Method, Part II, in: The Philosophical Works of Descartes, Vol. I (London: Cambridge University Press, 1973), pp. 92-93.

23. Cf. for example, Hume, An Inquiry Concerning Human Understanding, Section IV Part I, and passim (New York: Library of Liberal Arts, 1955); see also, "An Abstract of a Treatise on Human Nature", loc. cit., pp. 183-198; and the "Appendix to the Treatise on Human Nature", in A Treatise of Human Nature, Vol. II (London; New York; E. P. Dutton & Co., 1952), esp. pp. 317-320.

24. "An Abstract", pp. 186-187.

25. Cf. in connection with this claim: Alasdair MacIntyre, After Virtue (Notre Dame: University of Notre Dame Press, 1981), esp. Chs. 1-7; and Schmitz, op. cit., II, III, IV.

26. Schmitz, op.cit., pp. 120-21.

27. Ibid., p. 121.

28. On the additional features required for an adequate conception of value, see my "Whitehead's Inability to Affirm a Universe of Value", Process Studies, 13 (Spring 1983), 117-131, especially the remarks regarding the Aristotelian-Thomistic tradition in footnote 2.

29. My principle concern in this article is of course the foundation of the universality of morality as normative, and hence obligatory. For the sense in which this universal obligatory character involves a concern for value, or the good, see John Farrelly's "The Human Good and Moral Choice" in this volume. For the sense in which the universal obligatory character of morality presupposes non-determinism, or more precisely freedom, see my discussion of Kant which follows and ch. IV above.

30. There is of course a third possibility here, namely non-cognitivism, which holds that morality is, strictly speaking, neither relative nor non-relative, precisely because it is cognitively meaningless. In terms of the argument of the paper, this collapses into the two possibilities noted. For, on the one hand, non-cognitivism is often mixed with naturalism in the thinking of social scientists; and, on the other hand, insofar as non-cognitivism is strictly adhered to, it embraces a dualism of fact and morality which, as dualistic, is similar to non-naturalism in the sense pertinent to my argument.

31. Cf. especially in this connection Kant's Introduction to his Prolegomena to Any Future Metaphysics, edited by Paul Carus (Chicago: Open Court Publishing Co., 1933); see also the Preface to the Second Edition and Introduction, Critique of Pure Reason, trans. by Norman Kemp Smith (New York: St. Martin's Press, 1965). (Hereafter CPR).

32. Cf. e.g., CPR, p. 44.

33. Cf. e.g., CPR, p. 464.

34. Of course Kant himself does not ascribe a teleological character to morality. My point in what follows is simply that, insofar as he is concerned to find a place anywhere in his system for teleology, that place must be outside nature (in a constitutive sense). My further point, then, is that anyone who, in disagreement with Kant, might wish to ascribe to morality a teleological character must nonetheless, like Kant, locate that character outside nature if nature is interpreted in Kant's mechanical sense.

35. CPR, pp. 549-570, esp. pp. 560-61.

36. CPR, p. 464.

37. CPR, pp. 464-479, esp. pp. 464f and 472f.

38. See in this connection MacIntyre's After Virtue, loc. cit.

39. See n. 34.

40. Cf. Schmitz, op. cit., pp. 118-19 and 96-130. For the sense in which these metaphysical axioms bespeak a radical activity or energy (energeia) rooted in esse, which is act in the fundamental sense in Thomism, see especially the remarks by Schmitz on pp. 117f. For the sense in which these axioms bespeak an activity or energy (energeia) in Aristotle, see James P. Etzweiler, "Being as Activity in Aristotle: A Process Interpretation", International Philosophical Quarterly, XVIII (1978), 311-334, and Frederick E. Ellrod's response "Energeia and Process in Aristotle", IPQ, XXII (1982), 175-181. In what follows I am concerned principally with the Thomistic formulation of the issue relative to the problem of a mechanistic understanding of nature.

41. For studies pertinent to this claim see Schmitz, op. cit., and "Immateriality Past and Present", Proceedings of the American Catholic Philosophical Association, (1978), 1-15; my "David Bohm on Contemporary Physics and the Overcoming of Fragmentation", International Philosophical Quarterly, XXII (1982), 315-327, and my "Whitehead's Inability to Affirm a Universe of Value", cited in n. 28.

42. For the purposes of this paper, I simply abstract from the question whether, or in what sense, beings other than humans might be properly affirmed to be conscious. It suffices for the purposes of my paper to affirm that all beings including human beings are active and purposive, but that human beings are nonetheless active and purposive in a distinct way (a way which is called self-conscious and free).

43. See in this connection the following works: Germain Grisez, "The First Principle of Practical Reason: A Commentary on the Summa theologiae, 1-2, Question 94, Article 2", Natural Law Forum, 10 (1965), 168-201; John Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980); Ralph McInerny, "The Principles of Natural Law", American Journal of Jurisprudence, 25 (1980), 1-15; John Finnis and Germain Grisez, "The Basic Principles of Natural Law: Reply to Ralph McInerny", American Journal of Jurisprudence, 26 (1981), 21-31.

44. Aquinas, Summa Theologiae, 1-2, Question 94, Article 2.