1. Korsgaard and Scanlon on normativity: an overview

Christine Korsgaard is familiar with contemporary discussions of normativity in non-moral contexts, but throughout her work, she is especially interested in moral normativity. Her analysis of normativity takes its starting point from the common experience of moral claims as necessitating, that is to say, imposing a necessity that has no parallels in the natural world: ‘… we find ourselves doing what we think we ought to do, in the teeth of our own reluctance, and even though nothing obvious forces us to do it. We toil out to vote in unpleasant weather, telephone relatives to whom we would prefer not to speak, attend suffocatingly boring meetings at work, and do all sorts of irksome things at the behest of our families and friends’.Footnote ³ If asked to explain this kind of behavior, we might say that we cannot do anything else, even if no one is forcing us to do it, and we are not facing injury or the prospect of loss. According to Korsgaard, such experiences of being necessitated stem from the normativity proper to a moral demand, which she identifies as ‘the grounds of its authority and the psychological mechanisms of its enforcement’.Footnote ⁴ Moral claims or laws are experienced as authoritative, and that implies that they are grounded in considerations that cannot be reduced to the agent’s well-being or to more general considerations of utility or expediency. In fact, they may require setting aside such considerations.

In order to respond to this challenge, Korsgaard develops an account of human action as self-constitution. She claims that we constitute ourselves as human beings, and at the same time as the specific individuals that we are, through our actions:

Korsgaard’s understanding of normativity is thus Kantian, at least in key respects. At the same time, she approaches the problem of normativity within a teleological framework, which she explicitly attributes to Plato and Aristotle, in general inspiration if not in detail. Her thesis is that ‘normative principles are in general principles of the unification of manifolds, multiplicities, or, in Aristotle’s wonderful phrase, mere heaps, into objects of particular kinds’.Footnote ⁶ She goes on to explain that ‘to be an object, to be unified, and to be teleologically organized, are one and the same thing… At the same time, it is the teleological organization or form of the object that supports normative judgements about it’.Footnote ⁷ Korsgaard initially illustrates this point with examples of artifacts, such as houses. We can identify this assembly of bricks and lumber as a house because we are aware of what a house is meant to be, and by the same token, we can judge whether this particular house is a good house, or in some way defective. However, the logic of her argument takes her beyond a consideration of artifacts, into the world of living creatures and especially rational men and women.Footnote ⁸ She claims that the forms of life and activities of living creatures are ‘defined by certain standards that are both constitutive of it and normative for it’,Footnote ⁹ the key point being that operations and actions are evaluated teleologically, by reference to the way in which they contribute to, or else undermine, a distinctively human way of life:

The normative standards to which Korsgaard refers are the principles associated with Kant’s categorical and hypothetical imperatives. She holds that the agent’s rational consistency depends on a willingness to act in such a way as to acknowledge and respect the rational agency of others; hence, ‘The necessity of conforming to the principles of practical reason comes down to the necessity of being a unified agent.’Footnote ¹¹ Korsgaard thus holds that the rational agent has ineluctable, self-regarding reasons for acting in accordance with these principles, since, otherwise, she undermines her identity as a unified agent.

In his Being Realistic about Reasons, T. M. Scanlon offers a strikingly different account of normativity. In his own words, he is a reasons fundamentalist. As he explains,

Reading further, we find that when Scanlon refers to reasons, what he has in mind are simple, everyday reasons such as ‘The fact that a person’s child has died is a reason for that person to feel sad’, or ‘the fact that it would be enjoyable to listen to some very engaging music…is a reason to do this, or to continue doing it’.Footnote ¹³ He anticipates responses that attempt to account for the normativity of reasons by analyzing them in terms of something that reasons have in common, such as the satisfaction of desires. The fundamental difficulty with this line of thought, he argues, is that it does not account for the critical element of normativity: ‘The question, however, would be whether identifying facts about reasons with non-normative facts would explain reasons or eliminate their normativity. The action guiding force of reasons, on such a theory, would seem to be purely causal and explanatory’. He goes on to say that ‘I myself believe that this claim is refuted by the evident lack of intrinsic normative significance of facts about desires’.Footnote ¹⁴

This line of argument recurs in Scanlon’s remarks on Korsgaard. After summarizing her constructive account of reasons, he says, ‘I do not see how such a constructivist account of reasons can succeed. Although it seems true that individuals have different reasons depending on the ends and practical identities they have adopted, these reasons depend on their having good reasons to adopt those ends or identities in the first place, and not to revise or reject them’.Footnote ¹⁵ In other words, Korsgaard’s claim that moral imperatives are grounded in a particular understanding of rationality presupposes an appeal to the normativity of rational self-constitution and, so, cannot serve to generate such an account.

How, then, can we account for the normativity of reasons? Scanlon claims that we cannot do so and that it is a mistake to try. Reasons are foundational to normative judgments, and if we attempt to analyze them in simpler terms, or much less to justify them, we have misunderstood what the reasons fundamentalist is claiming. The notion of reasons and the practices associated with it are so basic to human life, and so intimately bound up with our judgments across a range of domains, that the cost of rejecting this notion is prohibitively high. Fortunately, we have no occasion for doing so.

At the same time, Scanlon is not a non-cognitivist. He believes that claims about reasons can be true or false, and he thinks that we can reason about reasons, in such a way as to affirm, reject, or modify our initial views about the reasons that we have. In his words, he offers a ‘qualified defense of a realistic cognitivism about reasons: a view that is cognitivist in holding that claims about reasons for action can be correct or incorrect, but realistic also in recognizing that there may be limits to the range of cases in which such claims have determinate truth values’.Footnote ¹⁶ Reading further, we see that for him, a realistic cognitivism relies heavily on such basic considerations as consistency, reflection on the implications of one’s views, and in the case of moral reasoning, a consideration of the impact that these views will have on others. He resists attempts to move from these kinds of reflections to a theory of practical reasons tout court, and above all, he resists any claim to the effect that reasons can be identified, defended, or rejected on the basis of some supposedly more fundamental non-normative basis.Footnote ¹⁷

It will be clear by now that Korsgaard and Scanlon offer strikingly different accounts of normativity. Korsgaard attempts to reformulate a Kantian notion of normativity in terms of an account of the teleology of action, broadly Aristotelian in its main lines, but more specifically Platonic in its account of the unifying effects of rational and morally good action. Scanlon, for his part, takes a more radical approach, arguing that normative judgments rest on fundamental reasons which cannot be justified by appeal to non-normative considerations. He accordingly rejects Korsgaard’s claim that moral actions are normative because they promote the rational self-integration of the agent. She may be right, empirically, but we may still ask why this quality constitutes a reason why we should be moral.

It would go well beyond the scope of this paper to attempt to work through this debate in greater detail. My main concern at this point is to show that Aquinas has points of contact with both positions, and he suggests a way to move beyond certain difficulties that they generate. Here is the issue as I see it: first, it seems to me that Scanlon is right to say that reasons – more specifically, reasons for acting or reacting in certain ways, or more broadly, normative claims – are fundamental to normativity. We cannot justify a normative claim on the basis of non-normative claims alone, although these may play an important role in specifying such claims. At the same time, I think Korsgaard’s account of the teleology of action offers what is in effect a challenge to the way in which Scanlon distinguishes between normative and non-normative considerations. I want to suggest that they do not disagree about the foundational status of normative claims, but, rather, on the more basic question of what counts as a normative claim. So we might wonder whether it is possible to bring together the acknowledgement of something like Scanlon’s reasons fundamentalism with Korsgaard’s robust metaphysical account of normativity.

I believe that it is possible to do so, drawing on Aquinas’ account of the logic and metaphysics of evaluative notions. In the next section, I will set out the main lines of Aquinas’ account of the fundamental evaluative concepts of good and bad, comparing these with the theories of normativity offered by Scanlon and Korsgaard. In the third section, I will argue that this three-cornered comparison suggests a way to harmonize the theories of normativity proposed by these two philosophers, at least up to a point.

2. Aquinas on the good

According to Aquinas, the fundamental evaluative concept is good, bonum, and its contrast, bad, malum, expressed in the well-known first principle of practical reason (hereafter, FPPR), ‘Good is to be done and pursued, and bad is to be avoided’ (Summa theologiae I-II 94.2). We are told at different points that this is the first principle of the natural law and the first principle of the virtues, thus implying that it is the first principle for all moral evaluations (ST I-II 94.2, II-II 47.6). What is more, it is the FPPR, generally and comprehensively considered. In the ST, Aquinas first makes this point in a discussion of synderesis, an obscure term long associated with conscience, and moral judgment more generally. The relevant article occurs in his discussion of the intellectual powers of the soul, and Aquinas is arguing that synderesis is neither a faculty nor an act, but a habit. In order to make his point, he needs to identify the object of the habit, and this calls for a brief discussion of the FPPR:

Aquinas goes on to say that just as the first principles of speculative reason are habitually known, so the first principles of operation are known through the habit of synderesis. This text thus establishes a parallel between the first principles of speculative reason and the FPPR. In each case, the relevant principles are naturally known, foundational, and (by implication) underivable. They express the conceptual structures of reasoning itself, in general, or with respect to some domain. In the case of the FPPR, the domain is operations, broadly considered, and the claim would be that reasoning about operations is necessarily grounded in concepts of the good, and its contrary, the bad.Footnote ¹⁸

To put the point in another way, once we begin to reason about what is to be done or not done, we are already making use of the FPPR. There can be no normative considerations that would lead us to conclude that we ought to pursue the good, because the only reason we can have for pursuing the good is that it is worth pursuing, that is to say, good. In particular, truths about God or human nature cannot lead us to conclude that we ought to pursue the good – rather, these truths only take on normative significance insofar as they point to something that is worthy of pursuit, that is, good. Thus, the FPPR does not rest on any religious, metaphysical, or natural truths, and by the same token, it is neutral with respect to particular conceptions of what is good and what is evil. If this were not the case, the first principle could not function as such, that is to say, a first principle necessarily employed in all reasoning, whether sound or faulty, that takes place within a given domain.Footnote ¹⁹

At this point, it will be apparent that Aquinas’ account of good is similar in at least one key respect to Scanlon’s account of fundamental reasons;, that is, they apparently agree that normative claims are necessarily grounded in more fundamental normative considerations of some kind. At the same time, they seem to have very different ideas of what these normative considerations look like, so to speak. For Aquinas, all normative considerations are ultimately grounded in an abstract and general first principle, formulated in terms of the basic notions of good and bad.Footnote ²⁰ For Scanlon, in contrast, normative considerations are grounded in reasons, which he understands in very specific terms. Aquinas might respond that this disagreement is verbal rather than substantive, since Scanlon’s reasons could be reformulated in terms of the FPPR. Would Scanlon accept this move? I suspect that he would not, on the grounds that it implies that we must first grasp some notion of good in order to recognize and to act on reasons. Hence, Scanlon might say, Aquinas’ appeal to a FPPR is just one more attempt to deny the status of reasons as fundamental to normative judgments.

But this objection, I believe, rests on a misunderstanding of Aquinas’ claim that all practical reasoning is necessarily grounded in a first principle of practical reasoning. On this view, Aquinas would be saying that practical reasoning cannot get underway until the individual learns the meaning of good, as that which is desirable and worthy of pursuit, which he then applies by identifying those things which are worthy of pursuit, and pursuing them. This way of construing the FPPR would indeed represent an attempt to analyze reasons in terms of some more general quality that they all exemplify, thus denying the fundamental status of reasons as such.

However, this line of interpretation is not plausible, neither on its own terms, nor as a reading of Aquinas. On this interpretation, the concept of the good expresses whatever it is that renders all and only good things desirable, or worthy of pursuit. We learn the meaning of ‘good’, somehow, and then we apply it to every candidate for desire and pursuit, in order to determine which candidates are truly good and which are not. This line of interpretation raises two difficult questions: What is the concept of the good, and how do we learn it? Aquinas does not attempt to answer either of these questions, and I would suggest that he does not need to do so, because he has no stake in defining ‘good’. On the contrary, Aquinas’ view implies that anyone who thinks that we can define the good has not understood the notion of good and its place as a starting point for practical reason. Let me explain.

According to Aquinas, the notion of the good is expressed in the formula, ‘Good is to be done and pursued, and bad is to be avoided’. Admittedly, this looks like a definition of ‘good’ as, ‘that which is to be done and pursued’. But in order to serve as a definition, this formula would need to express ‘good’ in terms which are more simple, better understood, and more familiar to us than ‘good’ itself. In order to see the difficulty here, try to think of a normative term that is more basic than ‘good’. Aquinas’ formula itself suggests one possibility: Perhaps ‘doing’ and ‘pursuing’ provide us with more basic notions in terms of which to define ‘good’, as that which is worth pursuing and doing. But this line of analysis simply raises a further question, namely, how do we know what it means to seek and to do? The answer to this question might, again, seem obvious, but that is because we normally think of doing and pursuing in the first-person, as operations that we perform. We typically engage in goal-directed behavior in a conscious, self-reflective way, in such a way that we know that we are doing something with some end in view. But considered in the third person, as motions that we observe, it is not always clear that these are instances of seeking and doing, as opposed to, for example, random motions or operations which are in some way directed by another, through hypnotism, perhaps. At the level of theory, we understand the difference between random motions and operations, because we understand the difference between aimless movements and goal-directed operations. And fortunately, we can usually apply this distinction in practice, not through an analysis of one set of terms correlated with the other, but through the simple apprehension of some motion as goal-directed activity, implying an intrinsic orientation toward some good.

The upshot is that the two sets of terms correlated by the FPPR are logically on the same plane. If the FPPR were a definition, it might just as well be taken to be a definition of doing and pursuing, as a definition of good and bad. Clearly, it is neither. The FPPR, like other first principles of reason, is not a definition at all. It is a formulation of the fundamental correlation between two basic notions, neither of which can be understood independently of the other. These correlations are fundamental to thought and activity within the domain of goal-directed activity, because they express what it means to be operating in that domain. This is why Aquinas can say that the FPPR is known to all, because all know the meaning of the terms. We know what it means to act and to seek some end through our experiences of goal-directed activity. This does not mean that we bring a predetermined concept of good to every activity. Rather, we come to understand good together with the operations directed to the good, that is, doing and pursuing, through the apprehension of good as the fundamental principle informing diverse experiences of pursuit and action. As we acquire linguistic skills, we learn how to express this fundamental principle verbally, in such a way as to come to know the meaning of the terms of the FPPR.Footnote ²¹

With this in mind, let us turn to another text in which Aquinas elaborates on the FPPR and its relation to practical judgments, namely, I-II 94.2. In this familiar article, he sets out to make the case that the natural law is one unified law, even though it contains a multitude of precepts. He begins by observing that the precepts of the natural law are related to practical reason in the same way as the first principles of demonstration are related to speculative reason, for in each case, the principles are per se nota. He goes on to draw a distinction between principles that are self-evident in themselves and those that are self-evident to us, on the grounds that we may not be able to recognize a self-evident proposition if we do not know the meaning of the terms. He goes on to say, following Boethius, that certain propositions are self-evident to all, because the meaning of the terms is known to all, for example, ‘every whole is greater than the part’, or ‘Those things which are equal to one and the same thing are equal to each other’. Other propositions are self-evident only to ‘the wise’, that is to say, those who are capable of grasping the meaning of more recherche terms. He continues:

Aquinas thus claims that the FPPR, like other indemonstrable principles, is self-evident, that is to say, manifestly true by virtue of the meaning of the terms. However, the FPPR is not foundational because it is self-evident. After all, there are some self-evident principles which are not foundational, even for those who are wise enough to understand them. Rather, the self-evidence of the first principles, and more specifically, their self-evidence to everyone, indicates why it is that these principles are not only universally known but also necessary to reasoning in a given domain. This is so, because the first principles of speculative or practical reasoning are grounded in the apprehension of concepts which are both logically fundamental and metaphysically necessary. The principle of noncontradiction reflects the rational apprehension that every existing finite thing is something, a particular instantiation of an intelligible form, and not anything else, excluded by the form. In the same way, the FPPR is grounded in a rational apprehension that desires are grounded in, or in some way reflective of, a fundamental quality of desirability which we identify with good.Footnote ²² This apprehension need not – and I suspect, cannot – be generated through instruction. Rather, the rational agent grasps the meaning of good through experiences of desire and pursuit, leading to the emergence of foundational concepts of good and its contrary, together with some understanding of pursuit and avoidance as the fundamental modes of response to good and evil.

Once again, we see that Aquinas’ account of practical reason and Scanlon’s reasons fundamentalism are more similar than we might have thought. The most apparent difference between them lies in the level of specificity at which reasons for action are formulated. Aquinas offers an abstract formulation of a general reason for action, whereas Scanlon’s examples of reasons for action are formulated in specific terms. But now that we have dealt with the worry that Aquinas is attempting to define ‘good’ and ‘bad’ in his FPPR, we can plausibly construe Aquinas’ general FPPR as a statement of the logical form for Scanlon’s specific reasons.

At this point, we turn to a consideration of Korsgaard’s theory of normativity and its relation to Aquinas’ account of the good. It might seem that by defending the similarity between Scanlon’s reason fundamentalism and Aquinas’ account of practical reason, I have ruled out any meaningful comparison between Aquinas and Korsgaard. After all, Aquinas’ claim that the FPPR is self-evident, and Scanlon’s claim that reasons are fundamental, both entail that the starting points for practical reason cannot be grounded in psychological or metaphysical starting points. This would seem to rule out Korsgaard’s claim that moral norms stem from our need to constitute ourselves as agents of a certain kind. But this conclusion would be too quick. I think that Aquinas’ account of practical reason does indeed rule out the possibility of deriving the most fundamental normative principle from psychological or metaphysical truths about human existence. This is what it means, logically, to be a first principle. Nonetheless, he also offers an account of the way in which our actions, which are structured by these principles, are grounded in our nature as creatures of a certain kind. And as we might expect, he develops this account in terms of Aristotelian philosophy, reminiscent of Korsgaard’s philosophical approach. I do not want to overstate the similarities here – to mention only one point, Korsgaard’s account of self-constitution as rational integration makes use of Plato in a way that Aquinas would not accept.Footnote ²³ But there are striking similarities between Aquinas and Korsgaard nonetheless. In order to identify these, let’s return to his account of the FPPR, seen in relation to the natural law.

As we have already observed, Aquinas holds that the FPPR expresses the intelligibility of the good as that which all desire. As such, it provides a unifying principle, through which diverse moral precepts are brought together in an intelligible way into one natural law. As he explains,

Aquinas goes on to extend this analysis to those things (laws? – the text does not say) that stem from inclinations that we share with other living things, and those that are proper to our rational nature. These inclinations are not rational insights into what is good and worthy of pursuit. Rather, they are operations of the rational appetite, that is, the will, inclining it toward those objects and states of affairs that preserve and perfect the natural form of human existence (ST I 80, especially 80 ad 1; I-II 10.1). This line of analysis presupposes that there really is such a thing as a natural form of human existence, we can know what it is, at least imperfectly, and we can choose to act in such a way as to express it fully and appropriately – or not. Aquinas would affirm all three of these presuppositions. He is a metaphysical realist, not only with respect to human nature, but also with respect to the fundamental structures of formal and final causality that inform all finite forms of existence. Let me elaborate on this point.

In his initial treatment of the natural law in the ST, Aquinas claims that the first principle of practical reasoning is a principle through which the rational creature participates in the eternal law (I-II 91.2). As such, it reflects both the distinctiveness of the rational creature, and its commonality with every other kind of creature: its distinctiveness, because the rational creature can only participate in God’s creative wisdom through its characteristic form of operation, namely, reasoning and rational desire; its commonality, because practical reasoning represents a distinctively human way of doing what every creature does, namely, seek its full development in accordance with its proper principles of operation. My point is that the FPPR gives expression, through the characteristic structures of the rational intellect, to the general metaphysical principles structuring everything that exists. Every creature participates in God’s eternal law by pursuing its final end, that is to say its perfection in accordance with the principles of its form (I-II 1.7,8). In order for this to be possible, each creature needs principles of operation that are both aimed at the full development of the creature, and themselves grounded in the causal principles proper to the kind of thing that it is. The FPPR is thus the intellectual representation of a general metaphysical truth, namely, that all creatures express and in some sense seek an appropriate good. Through the FPPR, the intellectual creature is able actively to express that truth by doing and seeking its appropriate good through its own species-specific form of activity, namely, rational activity. The FPPR, as grasped by the intellect and expressed through action, is a general, formal principle, and as such, it does not depend upon any awareness of its metaphysical grounding. But the metaphysics of human existence, as apprehended through theoretical reflection, does imply normative standards. It would go well beyond the scope of this paper to try to spell these out in any detail, but the point is that on Aquinas’ account, we do have some access to objective standards of natural goodness, even though these are not perspicuous in the structure of practical reason itself.

When we turn from Aquinas to Korsgaard, we find that the two authors take remarkably similar views on fundamental questions. As we have just seen, Aquinas correlates the form of a finite creature with its goodness and perfection, in such a way that its activities are understood in terms of an ongoing pursuit of perfection, that is to say, an ever-fuller realization of the kind of thing that it is. Korsgaard, for her part, claims that ‘every object and activity is defined by certain standards that are both constitutive of it, and normative for it. These standards are the ones that the object or activity must at least try to meet, insofar as it is to be the object or activity at all’.Footnote ²⁴ A standard that is both constitutive, that is, conferring identity, and normative, that is, evaluating operations in the light of identity – this sounds remarkably similar to the criteria for existence and perfection set by the formal and final cause. A little later, this suspicion is confirmed: ‘A living thing is a thing so designed as to maintain and reproduce itself; that is, to maintain and reproduce its own form. It has what we might call a self-maintaining form. So it is its own end; its ergon or function is just to be—and to continue being—what it is.’Footnote ²⁵ A little further, she offers a ‘defense of teleology’, which is important to her project, she explains, because ‘A teleological conception of the world is essential to our functioning as agents’.Footnote ²⁶ We are reminded here of Aquinas’ claim that every rational agent acts for an aim, and so does every other creature, claims that come at the beginning of his treatment of the general principles of morality and are clearly meant to set out the logical underpinnings of moral reasoning (I-II 1.1,2).

Admittedly, Korsgaard holds back from fully endorsing her teleological claims: ‘Teleological thinking need not be grounded in a claim about the world. It may be grounded in a claim about how human beings conceptualize the world. The idea, of course, a Kantian one, is that human beings are faced with the task of carving the sensible manifold into objects. The claim is that we pick out objects by identifying functional unities…’.Footnote ²⁷ Korsgaard explains how we do this, and then she then goes on to qualify her qualification: ‘I do not mean to imply here that living things are merely human constructs, or anything of that sort’.Footnote ²⁸ Clearly, Korsgaard is ambivalent about her endorsement of a kind of Aristotelian teleology, and in another context, it would be fascinating to ask why she holds back on this point and what might persuade her to change her mind. At this point, however, the salient point is that she does endorse an Aristotelian approach to teleology, with whatever reservations, and her key claims about identity and self-constitution are framed in those terms.

At this point, let’s return to the question with which we began.

3. Scanlon, Korsgaard, and Aquinas

This paper took its starting point from the observation that Aquinas’ account of good shares some elements of each of the accounts of normativity offered by Korsgaard and Scanlon. That being so, his account may serve as a bridge between these two very different accounts. At this point, we are in a position to confirm that. At the beginning of this paper, I claimed that Aquinas’ account of good said that Aquinas suggests a way in which the very different approaches to normativity taken by Korsgaard and Scanlon might be reconciled or at least brought into fruitful conversation with one another. Of course, the point of this exercise is not really to bring Scanlon and Korsgaard into agreement with each other, or much less with Aquinas. But it is interesting to consider the possibility that they might agree, up to a point at least, because by doing so, we may learn something about normativity and the good.

So what might we learn from Aquinas’ account of good that might suggest a way in which different contemporary approaches might be brought together? In the first place, he distinguishes between the logical status of first principles and the metaphysical framework within which these principles function, and in terms of which they can be interpreted and applied. Seen from this perspective, the accounts of normativity offered by Scanlon and Korsgaard can be construed as two accounts of the same complex phenomenon, each of which isolates a different aspect of the phenomenon without taking account of the way in which each implies something about the other.

In the first instance, Aquinas agrees with Scanlon that normative judgments are foundational to the exercise of practical reasoning and evaluative judgments. The most noticeable difference between Scanlon and Aquinas comes down to this: Aquinas offers a general formulation of the fundamental normative concept, ‘good’, whereas Scanlon does not. This may or may not reflect a deep disagreement about the possibilities of formulating a general normative concept in terms of which we could analyze reasons, as Scanlon presents them. At any rate, I want to suggest that this disagreement, important though it may be, should not lead us to overlook the key point of agreement, namely, that practical reasoning depends on fundamental and underivable normative claims or concepts, reasons in Scanoln’s case, the notion of the good, in Aquinas’ case. Aquinas expresses this point through his account of the FPPR, which, as we have seen, is the foundational principle for practical reasoning as such.

At the same time, however, Aquinas agrees with Korsgaard that normative claims are ultimately grounded in the teleological structures of existence and activity, including most fundamentally our own actions as self-determining agents.Footnote ²⁹ He agrees, in other words, with her fundamental realism concerning the metaphysics of goodness and the teleological structure of action.Footnote ³⁰ The upshot, for both Korsgaard and Aquinas, is that for living creatures, including human beings, and correlatively for human action, normativity extends further than we might have initially expected. In Korsgaard’s words, ‘every object and activity is defined by certain standards that are both constitutive of it and normative for it. These standards are ones that the object or activity must at least try to meet, insofar as it is to be that object or activity at all’.Footnote ³¹ Human action is always subject to normative criteria, insofar as it ought to be aimed at preserving one’s practical identity.Footnote ³² Aquinas expresses this point by identifying the desirable, by implication the good, with the expression and development of the principles of existence proper to a specific kind of creature, that is to say, its form (again, see I 5.1, 5). While his language is different from Korsgaard’s, it seems to me that they are making essentially the same point: the principles that structure the existence of any particular creature are also, inextricably, the ideals of goodness and perfection toward which the creature strives.

At the same time, Aquinas does not claim that the capacity to grasp and use the FPPR depends on any kind of rational insight into the metaphysical grounding of the FPPR. He makes it clear that this principle is innate, known to all rationally competent persons, and general enough to be applied to practical reasoning in every sphere of life. Hence, it can be employed in every instance of practical reasoning. Nonetheless, given a correct understanding of the metaphysics of human existence, we can see how this principle fits within that account. Scanlon and Korsgaard need not disagree about the nature of normativity, if we construe their claims as pertaining to two different levels of explanation, namely, the logic of normative concepts, in Scanlon’s case, and the metaphysics of goal-directed desires and operations in Korsgaard’s case.

It might seem that the specific metaphysical approach that Aquinas and Korsgaard share, more or less, opens up a further possibility of rapprochement between them and Scanlon. But this is admittedly not clear to me. Scanlon objects, rightly, I believe, to any attempt to ground normative claims in non-normative considerations of any kind. And for Korsgaard, as for Aquinas, the kinds of claims that are relevant to our reasons for action will not be purely non-normative; rather, they will include some kind of reference to the agent’s good, seen in reference to the constitutive standards of her existence, or alternatively, her specific form. The difficulty here is that Scanlon does not just object to grounding reasons in non-normative claims, he also objects to grounding them in normative claims that are more general than the reasons themselves. He notes that ‘Korsgaard believes that in order to explain the force of this “must,” we have to find something about the agent in virtue of which she must accept that p is a reason for her to do a’.Footnote ³³ He then goes on to a general discussion of internalism and externalism, leading to the conclusion that reasons for action need not be grounded in attitudes that the agent already has, or standards of rationality generally considered. He has already rejected the strategy of grounding reasons in desires, and so he concludes that, if his approach is sound, ‘Reasons Fundamentalism will be left as the only available position’.Footnote ³⁴

At this point, I have to admit to some confusion. At some points, Scanlon seems to be saying that normative claims cannot be justified by some appeal to purely non-normative claims. That seems to me to be a point at which he and Korsgaard and Aquinas would all agree. But at other points, he seems to be saying that reasons cannot be justified by an appeal to any reasons, normative or non-normative, that are more general than the reasons with which one begins. In other words, at some points he seems to be objecting to the project of justifying reasons at all. After all, he describes his project as a defense of Reasons Fundamentalism, and, so, we should not be surprised to find that he offers a thoroughgoing rejection of any attempt to understand particular reasons in general terms.

But if this is the case, what is left of the first point of potential agreement, namely, the distinction between the logical and the metaphysical in the analysis of reasons? I think that Scanlon could accept the formal analysis of reasons in terms of good and bad, correlated with action, pursuit, and avoidance. This kind of analysis, in itself, is not meant to explain anything; rather, it displays the logical form of practical reasoning itself. Scanlon would have no stake in rejecting this element of Aquinas’ approach, and he might even welcome it, as supporting his analysis of the relative strength of reasons.Footnote ³⁵ Aquinas himself would perhaps add that unless Scanlon is willing to admit at least this much, it will be difficult to see how reasoning about reasons can count as reasoning at all.

Of course, it is an open question whether Scanlon and Korsgaard would in fact accept any of these points. But, after all, the purpose of this kind of analysis is not so much to bring about a reconciliation between two individuals as to explore the possibilities of interpreting theoretical trajectories in an expansive way. By placing Aquinas in conversation with our contemporaries, we expand our understanding of Aquinas’ views and his continuing significance, as well as learning something about our own contemporaries’ views.