1. Introduction
In this paper, I aim to provide an analysis of moral obligations that accommodates several features that I take to be essential to any plausible account of morality. In sections 2 and 3, I develop an analysis of moral obligations that seeks to reconcile (a) Harry Frankfurt’s claim that considerations regarding our most intimate cares are those that are most important in structuring our practical lives with (b) the common platitude that moral obligations are overriding, i.e., that moral considerations cannot be justifiably subordinated to non-moral considerations in practical deliberation.[1] In order to accomplish this, I will argue for the thesis that the act φ that is morally required of an agent in circumstances C is identical to that which the agent is constrained by practical rationality to perform in C. Practical rationality, as I treat it here, is “a holistic enterprise, properly concerned not merely with identifying means to the realization of individual ends, but with the coordinated achievement of the totality of an agent’s ends.”[2] As such, it functions to determine what one really wants at the most fundamental level, where this object is identical to whatever best conduces toward, or is in greatest part constitutive of, the wellbeing of the agent. Its conclusions, established via acts of judgment, are hypotheses concerning the content of this value and are fallible.
In section 4, I speculate concerning the origin(s) of moral obligations and argue that the theoretically complex character of obligations can best be explained by supposing them to derive from multiple sources. Specifically, I will argue that morality properly generates imperatives – and, consequently, obligations – of both objective and subjective varieties depending on whether the object upon which the agent’s wellbeing depends in a particular set of circumstances are natural or transcendent.
Finally, in section 5, I explain how the above analysis bears on the debate between ethical internalists and externalists. Here I will defend the internalist thesis that, because every genuine moral judgment is ultimately grounded in the agent’s beliefs regarding what he or she wants at the most fundamental level, it is psychologically impossible to make a genuine moral judgment without being motivated to some degree to act in accordance with one’s judgment. This thesis borrows from Sander Voerman’s so-called “volitional account” of morality, and so this section’s discussion will rely heavily on comments he himself provides in its defense.
2. Volitional Necessity
2.1 Volitional Necessity
In The Importance of What We Care About, Harry Frankfurt argues that “the question concerning what is most important is distinguishable from the question concerning what is morally right,”[3] and that “the subordination of moral considerations to others might be justified.”[4] He elaborates by saying, “there are many important decisions with regard to which moral considerations are simply not decisive, and which must accordingly be based, at least to some extent, upon considerations of non-moral kinds.”[5] The considerations that Frankfurt ultimately identifies as being most important in structuring our practical lives are those concerning our most intimate cares. Following William James, I will refer to these as passional considerations, so as to distinguish them from evidential considerations, which concern facts external to the agent’s psychology.[6] From the former type of considerations derive what Frankfurt calls “volitional necessities,” which are essentially psychological constraints that prevent the agent from willing otherwise than he in fact does under the influence of his various cares.[7] We might formulate Frankfurt’s thesis as follows:
Frankfurt’s thesis: The act φ that is (all things considered) rationally required of an agent in circumstances C is identical to that which the agent is constrained by volitional necessity to perform in C, where the term “volitional necessity” connotes a psychological constraint that prevents the agent from willing otherwise than he in fact does under the influence of his various cares.[8]
According to Frankfurt, we can infer from a number of cases in which passional considerations play the most decisive role in our practical deliberations that moral considerations (which are supposed to be definitionally distinct from the former) must sometimes occupy a role subordinate to them. Frankfurt’s argument can be reformulated as follows:
(1) The act φ that is (all things considered) rationally required of an agent in circumstances C is identical to that which the agent is constrained by volitional necessity to perform in C, where the term “volitional necessity” connotes a psychological constraint that prevents the agent from willing otherwise than he in fact does under the influence of his various cares.
(2) Some actions that an agent is possibly constrained by volitional necessity to perform conflict with the prescriptions of morality.
(3) Therefore, some actions that an agent is possibly constrained by volitional necessity to perform are justifiable exclusively on the basis of non-moral considerations.
(4) Therefore, moral considerations must sometimes occupy a role subordinate to non-moral considerations in rational deliberation.
My primary objection to Frankfurt’s account concerns his designation of these rationally subordinated considerations as moral and the rationally privileged considerations as non-moral (an assumption required to support several of the premises above). Classifying the two respective types of considerations in this way entails the denial of the highly plausible (if not conceptually necessary) premise that moral obligations are overriding, i.e., that moral considerations cannot be justifiably subordinated to non-moral considerations in practical deliberation. This consequence of Frankfurt’s view makes it questionable whether or not his position is even coherent. Therefore, in order to make Frankfurt’s claims consistent with the premise that moral obligations are overriding, I propose that we identify these rationally privileged considerations (in whatever form they present themselves) as the genuinely moral considerations, and identify the rationally subordinated considerations (again, in whatever form they present themselves) as the genuinely non-moral considerations.
2.2 Moral Modality
One justification I might offer for treating passional considerations as properly moral appeals to the widely accepted principle that “ought” implies “can.” This principle states that a normative system can only generate obligations for an agent that the agent in principle can fulfill, where “can” is interpreted relative to the appropriate modality (e.g., metaphysical, conceptual, nomological, etc.).[9] In the case of moral oughts, we might think that the relevant modality is volitional, such that an agent S can do φ only if φ, inter alia, is not excluded by volitional necessity.[10] Let us refer to any such φ as a live option for S.[11] From this principle, we can proceed to construct the following argument: If the principles of practical rationality were able to determine an agent’s will to act[12] in accordance with his passional considerations (via the mechanism of volitional necessity),[13] then this would effectively incapacitate that agent for moral obedience in such cases as the latter’s requirements conflicted with the requirements of practical rationality. But because treating these two possible sources of obligation as conceptually distinct virtually guarantees that such conflicts will occur, the only sure way to preserve ought implies can is to conclude that moral obligations and the obligations of practical rationality are conceptually equivalent.
Thus construed, it should be apparent that the plausibility of a volitional analysis of obligation depends to a large extent on the plausibility of treating volitional possibility as a legitimate modal category. There are reasons, however, to be suspicious of such a category. First of all, it just doesn’t seem true that someone under the influence of volitional necessity (henceforth VN) is genuinely unable to fulfill his moral obligations. One way to flesh out this objection is to argue that motivation is not an all or nothing affair, but rather a matter of allocation. Even if an agent S is most strongly motivated with respect to passional consideration x, she may nonetheless possess some lesser degree of motivation with respect to moral consideration y. Then, as long as this latter amount of motivation was sufficient to enable S to perform the action prescribed by y, there would be no violation of ought implies can. Although the strongest motive (i.e., the reason or prospect that garners the greatest share of the agent’s motivation) is often that which is given expression in action, few would maintain that this connection is indefeasible. The argument above, therefore, only goes through if we assume a stronger internalist thesis, such that the strongest motive necessarily is that which is expressed in action. But this latter premise is highly implausible.
This objection calls for a slight refinement of our position. On Frankfurt’s account, being motivated to φ is not sufficient to will φ. Rather, willing involves the satisfaction of a further condition: namely, that the agent be wholehearted about – i.e., wholeheartedly endorse, or identify with, – an option.[14] Applying this consideration to the present discussion, we might reason that an agent’s moral potency consists not just in his being motivated to φ, but, more essentially, in his will to φ. When we adjust our premise above to incorporate these further considerations, the result is a new premise, which states that S can φm, where φm denotes a moral performance, just in case S wills φm, where willing requires (at minimum) that S more strongly identify with φm than any other present alternative. (Though this formulation is significantly weaker than Frankfurt’s wholeheartedness condition, it suffices for our present purposes.)
Our new premise entails that for an action to be genuinely moral requires, inter alia, that it be willed by the agent who performed it. Thus construed, it would indeed be impossible to fulfill one’s moral obligations in such a case as the prescriptions of morality conflicted with the prescriptions of VN. Having arrested the greater share of the agent’s motivation, VN would effectively prevent an agent from willing what morality prescribed, thus incapacitating the agent for genuine moral action.[15] If the agent nonetheless proceeded to perform this action, then, “though the letter of the law be fulfilled, he would have failed to fulfill its spirit.”[16] It’s not that VN simply prevents an agent from acting as morality prescribes when conflicts of the two occur, but rather that it renders any such performance under these circumstances morally inert.
Having dispensed with the objection above, I see no further reason to reject “volitional possibility” as a legitimate modal category. With this foundation for the mean time secure, we may tentatively move on to developing a volitional analysis of moral obligations.
3. A Volitional Analysis of Moral Obligations
In order to accommodate the considerations discussed above, I propose that we adopt an analysis of moral obligations such that the act φ that is morally required of an agent in circumstances C is identical to that which the agent is constrained by practical rationality to perform in C, where what is “practically rational” in C is a function of whether the objects upon which the agent’s happiness depends in C are natural or transcendent.[17] Roughly construed, an object is natural just in case it is something that nature is both equipped and naturally disposed to provide. The desires that correspond to these objects I will accordingly call natural desires. The paradigm examples of this type correspond to the appetites (e.g., food, sex, sleep, etc.). Conversely, an object is transcendent just in case it is something that nature is either not equipped, or not naturally disposed, to provide.[18] The desires that correspond to these objects I will accordingly call transcendent desires. The paradigm examples of this type correspond to desires for the realization of intersubjective possibilities, i.e., states of affairs the existence of which depend on the cooperation or shared interests of multiple agents (e.g., love, friendship, significance, purpose, etc.).[19] The fact that human happiness does in fact depend on objects belonging to both of these respective types explains why it is that morality often prescribes imperatives in which passional considerations play the most important or decisive role.[20] (This latter point will be explained in greater detail in section 4.)
In order to support the above analysis, I propose further that we adopt a moral theory in which (a) the proper goal of morality is eudaimonia, where this is understood in terms of possessing that particular set of practical habits that enables an agent to satisfy his or her fundamental interests; (b) the principles of morality issue in practical imperatives that are instrumental in reinforcing certain practical habits and attenuating others so as to produce an agent of the aforementioned quality;[21] (c) when an agent makes a moral judgment, he endorses – with a certain degree of conviction – a hypothesis about what he wants at the most fundamental level;[22] and (d) it is conceptually impossible to make a genuine moral judgment without being motivated to some degree to act in accordance with one’s judgment (for the aforementioned reasons). Theses (a) and (b) will be discussed in the following section concerning the origin(s) of moral obligations, while theses (c) and (d) will be discussed in section 5.
4. The Origin(s) of Moral Obligations
4.1 Objective Versus Subjective Imperatives
Perhaps the most vital question regarding any analysis of moral obligations concerns their origin(s). Determining the origin(s) of moral obligations will have far-reaching implications in the development of our larger moral theory, specifically with regard to our analysis of the moral concepts right and wrong. The terms right and wrong are, of course, ambiguous between a plurality of normative systems. The specific value that each term assumes with respect to any particular normative system is a function of whatever goal is proper to that system. Once such a goal has been identified, the term ‘right’ is used to designate those performances that accord with this goal in the appropriate way, while the term ‘wrong’ is used to designate those performances that fall short of this standard to some degree.[23] Thus understood, the function of normative evaluation is to describe a performance in relation to whatever value is internal to that system, where that value is apparent in the goal to which the system is oriented. My purpose in this section is to identify the particular normative standard that is proper to morality, and subsequently to provide an analysis of moral rightness and wrongness that follows suit.
I suggested above that we should adopt a moral theory in which, inter alia, the proper goal of morality is eudaimonia, where this is understood in terms of possessing that particular set of practical habits that enables an agent to satisfy his or her fundamental interests. Morality facilitates this process by prescribing imperatives that, when obeyed, lead to the reinforcing of advantageous practical habits and the attenuation of disadvantageous practical habits. Let us call these advantageous habits virtues, the disadvantageous habits vices, and the process itself moral formation.[24]
Now it should be obvious that the fixing of different virtues requires that morality prescribe different procedures according to the unique relationship that virtue bears to the world (or, more specifically, the selective pressures in the world). If, for example, an agent’s happiness were to depend entirely on material goods such as food, drink, sleep, sex, bodily security, etc., then his moral faculties would presumably record and canonize those actions and behaviors that were observed (on average) to result in the acquisition of these items. Over time, the practical strategies that proved most advantageous to the agent would become the most deeply entrenched in that agent’s psychology as these were continually rewarded with the pleasure that their success occasioned. In this first case, since the agent’s ability to satisfy the relevant desires depends primarily on being able to recognize, and willing to conform his behaviors to, the natural laws and processes that govern the dispensation of these items, morality would prescribe imperatives with a more strongly objective quality. Let us therefore call these objective imperatives, and the obligations that correspond to these objective obligations.
If, on the other hand, an agent’s happiness were to depend also on the realization of certain intersubjective possibilities – i.e., states of affairs the existence of which depend on the cooperation or shared interests of multiple agents (e.g., love, friendship, significance, purpose, etc.)[25] – then the intellectual and practical habits that his moral faculties would presumably record and canonize would be those that the agent either had some reason to believe would result (as by instinct or testimony), or otherwise came to discover could result (as by experience), in the realization of such states of affairs. In this case, in contrast to the former, the agent’s ability to satisfy the relevant desires depends primarily on his ability to recognize, and willingness to entrust his wellbeing to, the instincts and subjective inclinations of his own nature – even, at times, in disregard of certain evidential considerations that might discourage such practice.
As a last word on these latter interests, let me cite James’ brief apologetic from his 1881 essay, “Reflex Action and Theism.” Here he writes,
Man’s chief difference from the brutes lies in the exuberant excess of his subjective propensities, – his pre-eminence over them simply and solely in the number and in the fantastic and unnecessary character of his wants, physical, moral, aesthetic, and intellectual. Had his whole life not been a quest for the superfluous, he would never have established himself as inexpugnably as he has done in the necessary. And from the consciousness of this he should draw the lesson that his wants are to be trusted; that even when their gratification seems farthest off, the uneasiness they occasion is still the best guide of his life, and will lead him to issues entirely beyond his present powers of reckoning. Prune down his extravagance, sober him, and you undo him. The appetite for immediate consistency at any cost, or what the logicians call the ‘law of parsimony,’ – which is nothing but the passion for conceiving the universe in the most labor-saving way, – will, if made the exclusive law of the mind, end by blighting the development of the intellect itself quite as much as that of the feelings or the will.[26]
4.2 Beliefs, Desires, and Direction of Fit
The principle sketched above bears in important ways on our present enquiry concerning the origin(s) or moral obligations. Specifically, it suggests that, with regard to such circumstances as the vital object is transcendent (i.e., based on the irreducible subjective interests of the agent), we should understand morality’s program to be to bring the world into conformity with the agent’s subjective interests (i.e., with the instincts and subjective inclinations that ground the agent’s fundamental wants and desires).[27] On the other hand, it suggests that, with regard to such circumstances as the vital object is natural (i.e., based on the objective interests of the agent), morality’s program will be precisely the opposite: namely, to bring the agent’s behaviors into conformity with the world (i.e., with the natural laws and processes that govern nature’s provision of material goods).[28] My proposal is that we take these observations as evidence in support of the claim that moral obligations correspond to both objective as well as subjective imperatives, where the former derive primarily from evidential considerations and the latter derive primarily from passional considerations.
A final argument in support of the above proposal appeals to the observation common in the contemporary literature that beliefs and desires differ most significantly in their different “directions of fit.” Michael Smith explains this distinction in the following terms:
[T]he difference between beliefs and desires in terms of direction of fit can be seen to amount to a difference in the functional roles of belief and desire. Very roughly, and simplifying somewhat, it amounts, inter alia, to a difference in the counterfactual dependence of a belief that p and a desire that p on a perception with the content that not p: a belief that p tends to go out of existence in the presence of a perception with the content that not p, whereas a desire that p tends to endure, disposing the subject in that state to bring it about that p.[29]
From these considerations, it seems reasonable to conclude that objective imperatives are grounded in beliefs, subjective imperatives are grounded in desires, and that both of these are properly moral. I have a few qualifications to make regarding this conclusion, but I will reserve these for the following section’s discussion.
The last thing to do in this section is to provide an analysis of the moral concepts right and wrong that is rightly aligned with morality’s proper goal. Therefore, based on our discussion above, I propose that we adopt an analysis of moral rightness such that it is morally right for an agent S to φ in circumstances C only if, by φ-ing in C, S will bring about a result that will serve to reinforce a generally advantageous practical habit (or attenuate a general disadvantageous practical habit); and, conversely, an analysis of moral wrongness such that it is morally wrong for S to φ in C only if, by φ-ing in C, S will bring about a result that will serve to reinforce a generally disadvantageous practical habit (or attenuate a generally advantageous practical habit).
5. Voerman-Style Internalism
5.1 Moral Judgments
Following the account that Sander Voerman develops in his 2006 paper “Volitional Morality: A Theory of Free Agency and Moral Judgment,” I propose that we understand judgments as “acts that can establish intentional attitudes.” Thus, for example, when a person makes a judgment about a matter of fact, we can say that this judgment establishes an attitude of belief. Following this pattern, explains Voerman, we can proceed to distinguish various types of judgment in terms of what kind of attitudes they serve to establish.[30] When it comes to moral judgments, Voerman proposes that we define the moral in terms of the volitional, such that to make a moral judgment is to establish a belief concerning what one really wants.[31]
Because every genuine moral judgment is ultimately grounded in the agent’s beliefs regarding what he or she wants at the most fundamental level, it is psychologically impossible to make such a judgment without being motivated to some degree to act in accordance with one’s judgment. This follows from the fact that one is necessarily motivated with respect to the prospect of one’s wellbeing (as will be explained further below). When we combine this thesis with Voerman’s analysis above, the result is a Voerman-style analysis of moral judgments such that, when an agent makes a moral judgment, he endorses with a certain degree of conviction a hypothesis about what he wants at the most fundamental level, thus determining the agent to will to act in conformance with this hypothesis.[32]
Sometimes, however, the object of one’s moral judgment is sufficiently remote from one’s present decision that engaging the natural motivational mechanisms ceases to be a straightforward affair. In the following section, therefore, I discuss some of the complexities involved in making genuine moral judgments. By addressing these issues, I hope to be able to clarify my position somewhat, and also to defend this section’s analysis against amoralist objections.
5.2 Answering the Amoralist Challenge
All moral judgments yield conclusions that are propositions of the form ‘φ-ing is right’, where “right” roughly connotes having the property of promoting eudaimonia. In what we might call simple cases, i.e., cases in which the present decision’s relation to eudaimonia is readily apparent, this judgment will be as immediate as perception and give rise to violent passions, which are immediately motivating. In more complex cases, however, i.e., cases in which the decision’s relation to eudaimonia is remote or otherwise not apparent, the judgment will be burdened with theory and generate motivation only as long as the agent is able to preserve a perception of this connection in his imagination.[33] These perception-like judgments we might call thin,[34] while the theory-laden judgments we might call thick. Moral judgments of both thin and thick varieties, however, involve a complete tracing from one’s present decision back to morality’s ‘first principle,’ which we have defined as the wellbeing of specific agents, or eudaimonia. The perception of this relation necessarily causes the agent to be motivated with respect to the content of his judgment.
One may object along the lines of Smith that the thesis above is unnecessarily strong, and, indeed, implausible as presently formulated. Such a thesis, argues Smith, “commits us to denying that, for example, weakness of the will and the like may defeat an agent’s moral motivations while leaving her appreciation of her moral reasons intact.”[35] According to this line of reasoning, if the final product of a successful moral judgment is just a proposition of the form ‘φ-ing is right’, then why shouldn’t an agent be able to preserve the propositional integrity of his judgment even through a spell of depression or moral apathy?
At first glance, Smith’s objection seems a quite reasonable one. Upon closer inspection, however, it may be seen to rest on a category mistake. Specifically, it conflates the proposition that is yielded by an act of judgment with the act of judgment itself. Indeed, it is highly implausible to think that a change in an agent’s motivational state would threaten the integrity of any proposition – just as it is implausible to think that an agent could be motivated with respect to any proposition the deeper significance of which was not adequately grasped.[36] But neither of these claims is implied by the thesis above. Rather, our thesis concerns the connection between the act of judgment and motivation; and I insist again that such acts do necessarily result in the agent being motivated with respect his judgment. This is so precisely because it not just the proposition that is being entertained in such circumstances, but also – and more significantly to the present point – the relation of that proposition’s content to the agent’s standing desire to have his fundamental interests satisfied.
A final consequence of my view that is worth mentioning before concluding is that when the relation between a present decision and eudaimonia is very remote, it will be much more difficult to make a genuine moral judgment than it will be when this relation is more proximate. Also, and for the same reason, our view suggests that it will be more difficult for individuals of weaker intelligence to make genuine moral judgments than individuals of more able intelligence – especially insofar as this bears on their capacity for theoretical reasoning, which is so essential to making successful thick judgments. These observations lead us to predict a higher occurrence of akratic-like behavior with respect to situations involving exceedingly complex moral dilemmas and exceedingly simple-minded deliberators. I think both of these predictions are consistent with our experience, and provide further support in favor of my view.
[1] Considerations, as I treat them in this paper, are facts that are suitable for incorporation into practical syllogisms. These are to be understood as distinct form reasons, which are considerations actually incorporated into a practical syllogism in support of an imperative. The distinction between considerations and reasons, therefore, roughly mirrors that of William’s distinction between external and internal reasons, respectively (see Williams 1981).
[2] “[This] holistic approach finds its most sophisticated and influential expression in the maximizing principle of practical rationality[, according to which] the fundamental task of practical reason is to determine which course of action would optimally advance the agent’s complete set of ends. Thus it is widely accepted that the rational action for a given agent to take is the one whose subjective expected utility – reflecting both the utility of possible outcomes, from the agent’s point of view, and the agent’s beliefs about the probability of those outcomes – is the highest.” Wallace, R. Jay, "Practical Reason", The Stanford Encyclopedia of Philosophy (Summer 2009 Edition), Edward N. Zalta (ed.), URL =
[3] Harry Frankfurt, The Importance of What We Care About (Cambridge: Cambridge University Press, 1988), 82.
[4] Ibid., 81.
[5] Ibid. David Brink refers to this claim, which he also endorses, as the Antirationalist Objection. See David O. Brink, “Moral Motivation,” Ethics, Vol. 108, No. 1 (Oct., 1997), 19-21, 31-32. Christian Miller defends a similar view in his 2008 paper “Motivational Internalism.” The argument I develop here, therefore, is offered as an alternative to each of these respective views.
[6] William James, The Will to Believe and other essays in popular philosophy (New York: Dover Publications, Inc., 1956), 20. Frankfurt appears to be treating moral considerations here as a species of evidential consideration.
[7] “In certain instances…[a] person is susceptible to a familiar but nonetheless somewhat obscure kind of necessity, in virtue of which his caring is not altogether under his own control. There are occasions when a person realizes that what he cares about matters to him not merely so much, but in such a way, that it is impossible for him to forbear from a certain course of action…I shall use the term ‘volitional necessity’ to refer to constraint of [this] kind…An encounter with necessity of this sort characteristically affects a person less by impelling him into a certain course of action than by somehow making it apparent to him that every apparent alternative to that course is unthinkable” (Frankfurt 1988, 86-87).
[8] From the limited textual support I have so far provided, it may be unclear whether or not Frankfurt would endorse the thesis that I’ve formulated here. We might think, rather, that Frankfurt would only support a weaker thesis like the following: The act φ that is all things considered psychologically required of an agent S in C is that which is volitionally necessary for S in C. It might be, however, that in certain circumstances we are psychologically compelled to act in a way that is practically irrational. The following passage, however, shows that Frankfurt does consider volitional necessity to be a matter of rationality: “There is a mode of rationality that pertains to the will itself. Like the mode of rationality that is articulated in the necessary truths of logic, it has to do with the inviolability of certain limits. Logical necessities define what it is impossible for us to conceive. The necessities of the will concern what we are unable to bring ourselves to do” (Frankfurt 1988, 190).
[9] One reason to reject universalist-objective accounts of moral obligations is that these construe this principle too broadly, interpreting “can” relative to something like conceptual possibility. They then define this domain with reference to a hypothetical idealized agent (often possessing such features as “perfect rationality,” etc.), effectively discounting many morally relevant considerations that are irreducibly agent-specific.
[10] “…inter alia”: e.g., is also not conceptually or nomologically impossible, etc. This premise is weaker than Frankfurt may have intended by the doctrine of volitional necessity. The stronger formulation would state that “an agent S can do φ only if φ is volitionally necessary,” thus limiting what the agent can do in C to a single determinate action. The weaker formulation above merely states that “an agent S can do φ only if φ is not volitionally impossible,” thus preserving a range of possible actions that S can do in C. There is some evidence to suggest that Frankfurt intended the stronger formulation. For example, he writes, “An encounter with [volitional necessity] characteristically affects a person less by impelling him into a certain course of action than by somehow making it apparent to him that every apparent alternative to that course is unthinkable” (Frankfurt 1988, 86-87; italics mine). I, however, find this stronger premise highly implausible, and so have chosen only to endorse the weaker claim here.
[11] James’ definition of a live option is “one that appeals as a real possibility to him to whom it is proposed” (James 1956, 2-3). Cf. ibid., 26 and Miller 2008, 6-9.
[12] I am not saying that principles of practical rationality determine the agent to act in such and such a way, but rather that the principles of practical rationality determine the agent’s motivation to act. I am treating motivation as identical to the response of the will (i.e., as a volitional event) when presented with a stimulus of the appropriate type (either a direct perception or an imaginative representation of a percept.)
[13] I will revisit this antecedent claim in section 5, where I will also provide arguments in its defense.
[14] See Frankfurt 1988, 166, 170-172, 174-176. Cf. Voerman 2006, 6-24. Voerman explains this condition in its larger theoretical context when he writes, “To be free in virtue of having a will of your own can be understood as involving two important criteria: reflection an unity…Frankfurt has tried to capture the reflexive requirement into the concept of higher order desires, and the unity requirement into the concept of wholeheartedness” (ibid., 10-11).
[15] Of course, according to the view I develop in this paper, it is absurd that an agent should fail to will what he believed morality prescribed. I am speaking this way now merely to ward off possible objections to my view.
[16] It is therefore possible to accomplish what morality prescribes without thereby satisfying one’s moral obligation, though the latter is in greater part given by the former. Two acts may be identical in their external expression, but nonetheless diverge in that feature that is essential to morality: namely, that the performance proceed from a willing spirit.
[17] My view endorses an objective account of practical rationality, according to which “An agent S is practically rational to the extent that S’s practical thought and action are guided by what in fact are S’s reasons for action” (Miller 2008, 22). For a contrast with subjective accounts of practical rationality, see ibid. Christian Miller, “Motivational Internalism,” Philosophical Studies, Vol. 139, No. 2 (May, 2008), pp. 233-255.
[18] See James 1956, 131-132.
[19] See James 1956, 24-25, 59.
[20] According to the account I am providing, it is still true that considerations regarding our most intimate cares of the most important in structuring our practical lives (Frankfurt’s thesis), but only on the condition that cares are taken to refer to both natural as well as transcendent objects. There is some evidence, however, to suppose that Frankfurt only had the latter objects in mind when he formulated his account.
[21] Such practical habits are constituted by psychological states and are to be understood in terms roughly equivalent to Aristotle’s dispositional account of virtue and vice.
[22] Cf. Voerman 2006, 30.
[23] Some normative systems incorporate more “flex room” than others, such that there are values intermediate between right and wrong. If morality were to incorporate such flex room, these intermediate values would presumably yield various degrees of moral obligation. Our common sense notion that some actions are obligatory while others are supererogatory, as well as our notion that some actions are forbidden while others are simply permissible, lends prima facie support in favor of conceiving morality in this way.
[24] Another way to characterize the analysis I’m developing here is in naturalistic terms, i.e., in terms of morality’s fitness-enhancing function. Thus construed, the present account conceives of morality as aiming to enhance the fitness of specific agent’s, where “fitness” is indexed to some particular standard of living that is deemed intrinsically valuable. Therefore, insofar as it prescribes objective imperatives, we may understand morality to be concerned with the survivability of the agent, i.e., the individual’s fitness qua organism. Insofar as it prescribes subjective imperatives, however, we may understand morality to be concerned with the flourishing of the agent, i.e., the individual’s fitness qua transcendent being (in whatever terms this should be fleshed out). I will not speculate here concerning which of these two objects is most integral to eudaimonia.
[25] See James 1956, 24-25, 59. James sometimes describes these as “truths dependent on human action” (ibid., 24).
[26] James 1956, 131-132.
[27] This requires, inter alia, that morality provide the appropriate sanction for the behaviors that enable the agent to satisfy these desires, which is required in order to fix the requisite habits.
[28] It might be doubted whether this latter category is properly moral, since the ends at which is aims are so mundane (e.g., is learning to hunt a proper part of moral formation?), and, taken in isolation from the latter interests, so amenable to abuse and perversion (e.g., if I can acquire more material goods by killing my neighbor, should killing my neighbor be considered a moral virtue?).
[29] Michael Smith, The Moral Problem, (Malden: Blackwell Publishing, 1994), 115.
[30] Voerman 2006, 30.
[31] “The idea is simple,” writes Voerman, “why not reduce the moral to the volitional, and claim that to morally approve of something means to really want it?” (ibid., 28).
[32] Voerman 2006, 30. As explained above, an agent’s fundamental wants might correspond to either natural objects, transcendent objects, or any proportion of the two. Whether they do in fact correspond to either object, or in what proportion they might be divided between the two, are further questions upon which I do not presently wish to comment.
[33] Moral judgment works to yield motivation in this latter case by imaginatively representing the desired object in conjunction with the present decision so as to stimulate the will’s natural disposition to exert itself in that direction, i.e., to pursue said object. Descartes proposes a mechanism similar to this in Passions of the Soul. Here he writes, “Our passions...cannot be directly aroused or suppressed by the action of our will, but only indirectly through the representation of things which are usually joined with the passions we wish to have and opposed to the passions we wish to reject” (CSM I, 354). René Descartes, The Philosophical Writings of Descartes, Volume I, translated by John Cottingham, Robert Stoothoff, and Dugald Murdoch, (Cambridge: Cambridge University Press, 1985).
[34] The perception-like quality of thin moral judgments perhaps explains the appeal of intuitionism as a moral theory. For a similar defense of intuitionism, see Prinz 2006, 37.
[35] Smith 1994, 61.
[36] This point provides a possible explanation of moral fetishism. An agent may, over time, come to discover that doing what is right is constantly followed by an enhancement of his wellbeing. Eventually, the two ideas may become so closely associated in the agent’s mind that he no longer has to engage in the process of moral judgment in order to excite his motivation: like Pavlov’s dogs, the very intimation that ‘φ-ing is right’ is sufficient to excite the agent’s natural disposition to conform. That the agent experiences motivation toward φ, however, is not sufficient to prove that his judgment is genuine, since it may not have successfully related φ to morality’s first principle.