1. Commonsense Morality

Intuitionism is supposed to draw its materials from commonsense morality. Proponents and allies of intuitionism consider this a virtue of the theory, for who does not believe that lying is generally wrong? Opponents of the theory consider this a defect. J. S. Mill claims that intuitionists merely ‘assume the ordinary precepts of morals as of à priori authority’ (Mill Reference Mill and Miller2002: 235). If that is what intuitionists do, that would certainly set intuitionism apart from most moral theories. Moral theory, we think, is supposed to determine systematically what is right and wrong in such a way that we have made some explanatory advance over what we learned as children. As MacIntyre says, ‘all intuitionist writers suffer from one difficulty: they are, on their own view, telling us only about what we all know already’ (MacIntyre Reference MacIntyre1966: 254).

Simply going with commonsense morality also produces problems for practice. For one thing we have expected from moral theory is to provide us action-guiding information. If a theory like intuitionism merely reminds us of what we already know, then we will not find the very thing that initially drew us to moral theory. Perhaps on a day-to-day basis, we can get by with intuitive principles such as ‘Murder is wrong’ and ‘Rape is wrong’. But difficult moral situations will have us looking toward other moral theories. Moral theories are supposed to provide practical support. As Mark Timmons states, ‘The main practical aim of a moral theory is to discover a decision procedure that can be used to guide correct moral reasoning about matters of moral concern’ (Timmons Reference Timmons2002: 3). Kantianism and utilitarianism, equipped as they are with supposed supreme principles of morality, seemingly provide action-guiding information outside our everyday knowledge set. Intuitionism has nothing comparable to offer agents in vexing moral situations.

Even Ross's theory of prima facie duty will, we are told, be of no practical avail. Although duties can conflict, he claims, in every case one prima facie duty prevails. That is our actual duty. Ross holds that such overriding is due to the stringency of the prevailing prima facie duty. It is simply more of a duty than the others. But critics charge that agents in tough moral cases should not seize upon these details of Ross's theory as promising action-guiding information. For as McNaughton states: ‘It turns out after all to have nothing general to say about the relative stringency of our basic duties’ (McNaughton Reference McNaughton and Stratton-Lake2002: 91). Most theorists have taken this to indicate that nothing of value can be said about the stringency of prima facie duties. That is one central point I challenge here.

2. Moral Theory Critique

Decision procedures have been at the center of moral theories. They have played a number of roles. First, they are supposed to be based on the foundations of morality. In the past, moral theories characteristically based their decision procedures on supposed single supreme principles of morality, such as the categorical imperative and the principle of utility. More recently, decision procedures have been proposed that provide thought experiments to test rules or actions for their moral acceptability, such as Firth's ideal observer and Rawls's original position. Second, decision procedures are supposed to provide action-guidance for agents, especially in tough moral cases. Without such procedures we might be unable to determine what is right. The advantage of a moral theory is that with one we can solve very hard moral conundrums.

On the basis of their confidence in their own decision procedures, moral theorists have challenged intuitionism. They have especially targeted intuitionism's lack of such a procedure. Mill holds that if there are plural fundamental moral principles, ‘there should be a determinate order of precedence among them; and the one principle, or the rule for deciding between the various principles when they conflict, ought to be self-evident’ (Mill Reference Mill and Miller2002: 235). Rawls echoes Mill's request: ‘Once we reach a certain level of generality, the intuitionist maintains that there exist no higher-order constructive criteria for determining the proper emphasis for the competing principles of justice’ (Rawls Reference Rawls1971: 34). About intuitionists, Rawls continues: ‘They include no explicit method, no priority rules, for weighing these principles against one another: we are simply to strike a balance by intuition, by what seems to us most nearly right. Or if there are priority rules, these are thought to be more or less trivial and of no substantial assistance in reaching a judgment’ (Reference Bales1971: 34). This paper provides intuitionist priority rules. As we will see, these rules are already implicit in our moral thinking.

3. Common Sense and Self-evidence

What relation does intuitionism have to commonsense morality? That it centers on commonsense moral principles is widely acknowledged as a strength. One apparent downside of this theoretical choice is that intuitionism is said merely to privilege the commonsense principles at the expense of considered theoretical ones (Daniels Reference Daniels1979: 257). It has been claimed on this basis that intuitionism is thereby unsystematic. These points, in part, help explain why intuitionism is unable to guide action, especially in difficult situations.

Intuitionism does not privilege commonsense moral rules at the expense of others. Among moral theorists intuitionists are among the most epistemically cautious. We ought to affirm only the moral propositions with the highest epistemic credibility. The intuitive principles such as lying is wrong, murder is wrong, stealing is wrong, and so on seem to meet that standard. What I call the intuitive principles differ from what Ross apparently considers the fundamental prima facie duties. Aside from having very high epistemic credibility, one value of identifying the intuitive principles is that they enable us to determine the principles of stringency, as we will see later. Not only do intuitionists give proposed supreme principles a fair hearing, but they have incorporated such principles they believe to meet the standards of intuitionism, as has been seen in Ross (Reference Ross1930), Johnson (Reference Johnson1957), and Audi (Reference Audi2004). Each did so after putting such principles to the test of continual and severe extensive reflection. At every turn Ross bases his conclusions on ‘what we really think’ of morality. Just as Ross attempted vigorously to deny that promise-keeping has a bindingness of its own but failed to do so (Ross Reference Ross1930: 40), thus confirming it, we ought to continually question, deny, argue against, and apply any proposed moral principle until it appears to be undeniable. All this is done, bearing in mind there might be reasons, such as well-supported psychological reasons (Huemer Reference Huemer2009: 232), for denying some moral propositions.

The intuitive principles are commonsense principles and are, according to intuitionism, self-evident. It is important, however, that the relation being commonsense has to being self-evident is properly understood. Intuitionists do not merely grab a bunch of commonsense moral principles and slap the label ‘self-evident’ on them. Being self-evident is an explanatory hypothesis used for moral propositions that are intuitively persistent, even after rigorous scrutiny. Since belief and intuition can come into conflict, as Huemer (Reference Huemer2005: ch. 5) has effectively shown, mere doxastic persistence is not enough. On the intuitionist view, a major reason that the intuitive principles become items of common sense is that they are self-evident. That means that the evidence within a proposition, such as that murder is wrong, is itself sufficient to justify one in believing it (Kaspar Reference Kaspar2012: 52–59). Examining the matter further, our understanding of the very nature of murder justifies us in believing that it necessarily has the property wrongness.

4. Normative Factors

The tasks of normative ethics can be understood in various ways. Shelly Kagan offers a particularly helpful way neutral between them. He states that normative ethicists attempt to articulate ‘various normative factors’ and to discover ‘how they interact so as to determine the moral status of an act’ (Kagan Reference Kagan1998: 17). Normative factors include what good will be produced by a given action, whether someone will be harmed by a given action, what relation another person bears to the agent in question, and so on. Ethicists characteristically disagree on the crucial normative factors in moral situations. Some ethicists identify what they think the essential normative moral factors are, while others merely go to work using different factors without explicitly identifying them.

Normative factors have roughly two uses in ethics. The practical use of them is for agents in situations to determine the moral status of a given act. These factors help us to ‘figure out what the moral status of a given act is’ (Kagan Reference Kagan1998: 18). Obviously, if all else is equal between two moral theories, the one that can enable us to figure out what is right in real moral situations is theoretically preferable. Intuitionism's criticism of such a practical device has not been that it would not be helpful. It has merely been that we do not know of any principle or procedure that genuinely, accurately, and helpfully provides such action-guiding information. And proponents of every decision procedure will agree to this for every other theory but theirs.

The second use of a normative factor is explanatory. It has to do with ‘making a claim about what makes the given act have the particular status that it does’ (Kagan Reference Kagan1998: 18). In this capacity, normative factors have been labeled criteria of rightness, or right-making factors. Using a factor in this way is not primarily intended to enable an agent to solve a practical moral problem. Instead, its use is to help the theorist identify and explain just what it is that makes this act, possibly all acts, right or wrong. To employ language commonly used in ethics, the two uses to which normative factors may be put are as part of a decision procedure and as a criterion of rightness, or right-making factor.

Intuitionism's normative factors set it apart from all other theories. First, intuitionism holds that there is but one set of factors for both uses. In my view, moral kinds, moral relations, and values, integrated in well-structured ways, determine the rightness of actions. We ought to consider them in moral situations to figure out how we ought to act. In fact, we already do this, though not always effectively. Moreover, such entities, in my view, explain what makes right acts right. My hypothesis is that such entities, the core deontic grounds of duty–moral kinds, moral relations, values–are real, and the fact that we so naturally persist in thinking of them in our own situations only attests to their reality. In brief, what I aim to do here is identify the real normative factors of morality–intuitionist ones.

However, the main immediate benefit we will see of identifying such factors is explanatory. Our understanding of morality will be improved by considering such factors and following out all the implications of their being real. Where decision procedures are concerned, all that we can offer is a glimpse of how an intuitionist decision procedure might work. For the time being we must be content with having the elements of such a procedure that can guide agents more effectively and more faithfully to moral realities than any extant decision procedure. That does not imply that any future intuitionist decision procedure will meet the inflated expectations that have been set for such a procedure by several generations of moral theorists.

5. The Nature of the Act

Intuitionists have long made claims about the moral natures of acts. People tend to think that there is something about lying itself, stealing itself, murdering itself, that is wrong. Such claims have had two properties: they are quite appealing and very puzzling. We simply do not know what such natures could be. Ralph Cudworth stated centuries ago that, ‘nothing is morally good or evil, just or unjust, by mere will without nature, because every thing is what it is by nature, and not by will’ (Cudworth Reference Cudworth and Raphael1991: 108). Some might think that such considerations have a limited range of application. Promises, it might be thought, are made obligatory not by their nature, but by our promising. Cudworth responds by stating that it is not ‘as if the mere will or words and breath of him that covenanteth had any power to change the moral natures of things’ (Cudworth Reference Cudworth and Raphael1991: 110) Rather, it is ‘because natural justice and equity obligeth to keep faith and perform covenants’.

Richard Price also claimed that moral actions have distinctive natures: ‘Let it be considered, that all actions, undoubtedly, have a nature. That is, some character certainly belongs to them, and somewhat there is to be truly affirmed of them. This may be, that some of them are right, others wrong’ (Price Reference Price and Raphael1991: 146). Although we have a vague understanding of what he might mean, Price does not provide us with enough material to be able to say anything informative about what the moral nature of any act is. Ross also employs the language of moral natures. Speaking of different particular acts, Ross states that, ‘we see the individual act to be by its very nature wrong’ (Ross Reference Ross1939: 173). However, he nowhere states what a moral nature is, what is included in one, or how such natures determine the rightness or wrongness of acts.

6. Moral Kinds

When we speak of prima facie duties we are necessarily speaking of moral kinds. Lie, steal, promise, life-save, murder, rape, help-another are moral kinds. They are act-types. They are complex right-making properties we can consider and can instantiate by our actions. They are the primary objects of our moral intuitions. A situation in which a person cannot identify any moral kind being involved includes no prima facie duty. That means that moral kinds are essential to prima facie duties. Nonetheless, as we will see, moral kinds are not all there is to prima facie duties. Other real normative factors, such as moral relations and values, are involved in them. In general, though, when people give a minimal description of what is morally involved in a situation, they are identifying a moral kind, such as lie. Moral kinds in my view are the chief moral explanatory factor. As I have shown elsewhere, such kinds enable intuitionism to deal with challenges that are supposed to dog it, such as the problem posed by moral supervenience (Kaspar Reference Kaspar2012: ch.6; Reference Kaspar and Taddio2017). Next, our knowledge that lying is wrong, and self-evidently so, has its basis in our understanding of the moral kind lie and the property wrongness that necessarily belongs to it. Lastly, moral kinds, as we will see, enable us to provide an internal explanation of how goods determine what is right in the acts themselves. An internal explanation remains within the range of concepts in the intuitive principles or remains within other normative concepts (see Hurka Reference Hurka and Hurka2011: 20).

Moral kinds are also of central practical explanatory value. In moral deliberation the acts being considered are possible prospective acts. Being nonexistent, they are merely in the offing. For example, if you invite me for a hike on Saturday afternoon, and I previously told Jones I would help him move to a new dwelling on that day, I recognize the kind promise when I consider your offer. That makes me hesitate to accept it. Ethics has greatly underestimated the significance of this regular feature of our moral experience, assuming it to be obvious. But daily experience shows the importance of the recognition of moral kinds in our moral choices. Before I decide, the moral kind promise is the clearest thing in my mind. I understand my promise to Jones puts me in a special moral relation to him. My knowledge of it is more secure than my knowledge of any other situational feature, whatever my choice may be. The look of the hiking trail, the tone of my voice as I tell Jones, ‘I can't make it’, what Jones's reaction to my utterance will be, what words he uses to express how he feels about my choice–none of these things are as well known as the fact that I will, indeed, certainly break my promise by not showing up at Jones's place to help him move. Thus, if we wish to know the ways in which knowledge is involved in moral situations, we must look to moral kinds.

Moral kinds are the integral grounds of all duties. They are integral in two senses of the word. First, kinds are indispensable to there being prima facie duties at all. Second, kinds integrate the other factors that determine what our duties are, as I will show. Moral kinds are the keys to identifying and understanding the moral natures of acts. Without kinds there is no duty. With them there is. When we speak of a prima facie duty, we are primarily identifying a kind of act, even though there is more to the natures of moral acts. So we must first look to moral kinds to understand the primary factor determining the stringency of a duty. And we can see this at work in Ross's own stringency claims. He states that ‘it seems to me clear that non-maleficence is apprehended as a duty distinct from that of beneficence, and as a duty of a more stringent character’ (Ross Reference Ross1930: 21). At the same, however, Ross himself is skeptical of the possibility of laying down rules of stringency: ‘For the estimation of the comparative stringency of these prima facie obligations no general rules can, so far as I can see, be laid down. We can only say that a great deal of stringency belongs to the duties of “perfect obligation”’ (Ross Reference Ross1930: 41). I will show that we can, indeed, lay down rules of stringency and even rules that are self-evidently true.

Moral inquiry has us repeatedly return to right acts themselves to explain what makes them right. One would think that if we wished to explain why, say, murder is wrong, we ethicists would look to the nature of murder itself. But we do not. And that calls for explanation. Without moral kinds like murder to examine, we lack the tools to investigate the nature of murder properly. Since ethics has until recently lacked moral kinds as explanatory tools, it has sought other ways to explain why the proposition that murder is wrong is so incredibly convincing. One explanation has been that the proposition that murder is wrong is analytically true or true by definition. But my claim that the very nature of murder is wrong and that we adequately understand that is an equally good explanation of the high epistemic credibility of the proposition that murder is wrong (I have dealt with the analytic claim previously, see Kaspar Reference Kaspar2012: 68 and Reference Kaspar2015b: 60). To test which explanation is better, consider which one makes the most sense of how we really regard murder. Do we consider instances of murder so abhorrent because of a semantic property of the truth-bearer expressed by utterances and inscriptions of the sentence ‘Murder is wrong’? Or do we sentence people who commit murder to life in prison because of the nature of the act they committed?

7. Principles and Kinds

Most discussions of Ross's theory have described it as consisting of a fundamental plurality of moral principles (see Timmons Reference Timmons2002: 191; Shafer-Landau Reference Shafer-Landau2012: 232; Burnor and Raley Reference Burnor and Raley2011: 153). It might appear that what I call moral kinds simply are prima facie duties. So what advantage, it might be asked, is to be gained by speaking of moral kinds instead of prima facie duties? The reason to speak of both is that they are intimately connected, but they are not the same. Separating moral kinds from prima facie duties allows us to analyze the concept of prima facie duty. From this we can provide internal explanations of prima facie duties, something many have thought Ross's intuitionism is constitutionally incapable of providing. Burnor and Raley state:

Moral kinds are crucial to giving intuitionism explanatory power. Furthermore, in analyzing prima facie duty, we find other fundamental elements involved, such as moral relations and values. Bear in mind, then, that moral kinds are integral grounds of duty; they are not everything. But by considering moral kinds and other intuitionist normative factors, we might find that we have covered the essential grounds of duty.

A close look at Ross's The Right and the Good (Reference Ross1930) supports this theoretical strategy. Moral principles are not nearly as important to Ross as most references to his work would have us think. In fact, he holds that agents can and do determine what to do without consulting any principle, as apparently only Jonathan Dancy (Reference Dancy and Singer1991: 225) and myself (Kaspar Reference Kaspar2015a: 70) have recognized. Ross's favored way of speaking of prima facie duties is in terms of moral realities. He uses the phrase ‘principles of prima facie duty’ but once (Ross Reference Ross1930: 30), and ‘principle of prima facie duty’ just once (Ross Reference Ross1930: 33). In what way, then, does Ross speak of moral realities? He most often speaks of characteristics in virtue of which actions are right or wrong. He defines ‘prima facie duty’ in the following terms: ‘the characteristic … which an act has, in virtue of being of a certain kind (e.g., the keeping of a promise)’ (Ross Reference Ross1930: 19, emphasis added). Most of the characteristics Ross is speaking of can be considered moral kinds, but he also speaks of certain relations as being ‘the foundation of a prima facie duty’ (Ross Reference Ross1930: 19). He presents his fundamental categorization of prima facie duties as being distinct by the different facts the duties rest on (Ross Reference Ross1930: 21). So anything we speak of as a Rossian moral principle must capture the moral realities that ground our duties.

8. Principles of Comparative Stringency

Ross makes an uneasy compromise on stringency. Some kinds of act seem clearly more stringent than others; yet, apparently no general rules concerning stringency can be laid down. Unsurprisingly, this has led to skepticism about Rossian stringency. McNaughton states, ‘Ross fails to sustain his claim that some duties are more stringent than others’ (McNaughton Reference McNaughton and Stratton-Lake2002: 77).

The major obstacle to firm stringency claims has been methodological. Although kinds are integral to duty and to the stringency of duty, kinds are not the only factors involved in determining our duties. Basic to the idea of prima facie duty is that any kind of prima facie duty can be overridden by some other prima facie duty. This is true. But no analysis of prima facie duty yet given can explain why this is so. For example, it is conceivable that some possible property theft is worse than some possible kidnapping. What has been overlooked is this: when a duty of a less stringent moral kind overrides a duty of a more stringent moral kind, it is some other normative factor that accounts for the overriding. In other words, claiming that a kind of action K is more stringent than a kind of action L does not rule out the possibility that an act of the kind L is in some case more stringent than a K act. What this means is that insofar as an act is of the kind K it will always be more stringent than an act insofar as it is of the kind L.

Consider lying and stealing. I would claim that insofar as an act a is a theft, it is morally worse than an act b insofar as it is a lie. If there is any overriding of L to K, it is due to some normative factor other than moral kinds or the normative force of another kind besides L. An act b that is a lie can be worse than an act a that is a theft, but only based on other normative factors involved. A vicious lie about an innocent person is a morally worse wrong than a petty theft of a chocolate bar. Only further information about the nature of the particular lie b and the nature of the particular theft a makes us confident of this. That is why it is critical to discuss not only moral kinds but also the other normative factors separately if we are to determine the relative stringencies of prima facie duties.

Taking a step-by-step procedure will help us understand what accounts for differences of stringency in particular moral situations. This approach happens to be Ross's: ‘It is only by recognizing these different features one by one that we can approach the forming of a judgement on the totality of its nature; our first look reveals these features in isolation, one by one: they are what appears prima facie’ (Ross Reference Ross1939: 84). A question might now be raised. If Ross had the procedure that would enable him to analyze and explain differences of stringency in situations, why did he not use it and thereby settle his doubts about stringency? The main reason is that Ross did not identify and sort out the different normative factors his theory commits him to, as we are now doing. Moreover, he did not see how one factor in particular, moral kinds, would enable him to establish principles of comparative stringency.

Let us lay down some principles of comparative stringency. They are self-evident moral truths. The locution used to isolate kinds of act, ‘in so far as it is an X’, is one of the locutions Ross takes to be equivalent to the use of the concept of prima facie duty:

1. (a) Insofar as an act A is a murder, it is morally worse than an act B insofar as it is an assault;

2. (b) Insofar as an act B is an assault, it is morally worse than an act C insofar as it is a theft;

3. (c) Insofar as an act C is a theft, it is morally worse than an act D insofar as it is a lie.

Principles (a) through (c) are just a sample of principles of comparative stringency. A much larger set of principles is undoubtedly needed. Inspection of the ones just stated reveals points worth further examination. Evidently, several inferences based on transitivity can be made. Furthermore, these principles can be stated more succinctly, with an implicit understanding that prima facie duties are involved. These are Principles of Comparative Stringency:

1. (a′) The duty not to murder is more stringent than the duty not to assault.

2. (b′) The duty not to assault is more stringent than the duty not to steal.

3. (c′) The duty not to steal is more stringent than the duty not to lie.

As with other principles of intuitionism, there is bound to be wide agreement about these principles of comparative stringency. However, there will likely be doubts about them until we provide an explanation of why they hold. One part of the explanation will come next. Before that, though, we should note, as Ross did, that more than one moral kind can be involved in a duty (Ross Reference Ross1930: 27). Likewise, more than one moral kind can be instantiated in a single act, making the act better or worse by its presence. Thus stringency can be impacted by what multiple moral kinds are involved in the act. A lie is wrong, but telling that lie followed by an insult is worse. Murdering is wrong, but murdering someone after torturing them is worse. Helping someone in need is right, but helping someone, then cheering them up is better. Sometimes what qualifies one act to override another is simply the combination of kinds of one of the acts.

Moral kinds are the only factor we have so far considered. We can easily think of scenarios in which these stringencies no longer hold, including cases in which no combination of moral kinds decides what is right. Principles (a), (b), and (c) are, at least for ethicists, given their plentiful stocks of unusual moral scenarios, quite ripe for cases of overriding. I contend, however, that in order for such convincing overriding scenarios to be offered we must move our thinking outside of moral kinds. Putting the matter differently, if we stick with the moral kinds of acts involved, the comparative stringencies of such kinds will remain securely in place. Even if we agree with the principles of comparative stringency above, the question as to why they hold will remain. Although we are likely to agree with self-evident propositions concerning stringency, we want something that previous intuitionist theories lacked: explanations for our self-evident claims. In order to understand why one moral kind is more stringent than another, we must turn to goods that are internal to different kinds of moral acts. That is, we must turn to moral goods.

9. Moral Good

Intrinsic moral goods have been considered an extinct species of moral entity for ages. In Kant's quest to find what kind of action was of the highest moral goodness, he mistakenly claimed that only those moral acts that people perform out of a purely moral motive are morally good at all. Yes, acting for the sake of duty itself might be at the height of moral goodness for human beings. But that does not imply that other actions that are morally right are excluded from some level of moral goodness. If we hold that every fact in the world is either good, bad, or indifferent in value, then, since actual moral acts are inhabitants of this world, they must be either good, bad, or indifferent in value. As I will show, considering what goods and bads are intrinsic to acts of different moral kinds will enable us to do two things. One, it will provide us a way of explaining what it is about acts themselves that make them right or wrong. Two, on that basis, we will have the explanatory resources for explaining why an act of one moral kind is more stringent than an act of another moral kind.

H. A. Prichard (Reference Prichard1912) and Ross (Reference Ross1930) mistakenly excluded the category of morally good acts, with one exception. They accepted Kant's claim that only a good motive can make an act morally good. In other words, they excluded intrinsically morally good actions, right actions that are good regardless of motive, from their frameworks. I call this their colossal blunder (Kaspar Reference Kaspar2012: 123) and argue that if Prichard and Ross consistently followed their thinking with regard to moral actions, then they would accept intrinsically morally good actions. Their blunder has handicapped deontological theories to this day. As a result, it has become commonplace to say things such as that moral value ‘is something only agents can possess’ (Timmons Reference Timmons2002: 10).

Moral kinds, in my view, are sufficient to explain why a great many acts are right. However, deeper explanations of what makes acts right are available if we turn to moral goods. These two explanatory tools are connected in this way: Moral kinds have components. Each component of a kind is good, bad, or indifferent. In relation to kinds, the distinct components that constitute them, along with their values, result in a value for the whole kind. This allows for another level of internal moral explanation. Values are internal to acts of all kinds. One reason moral kinds are instructive is that the values contained in the kinds are invariant, just as the components of kinds involved are invariant. And if we can figure out what values are internal to what moral kinds, we will have made an advance in our comprehension of the relative stringencies we all see to hold between different kinds of act. Importantly, as I will show, certain critical values are contained a priori in moral kinds.

10. Explaining Comparative Stringencies: Moral Kinds

Most will agree with the principles of comparative stringency. Suppose, as I claim, that we know them. What I am attempting here is to explain such knowledge. Let us consider what kind of knowledge and what kind of explanation would be involved. Knowledge of principles of comparative stringency is knowledge of fundamental self-evident propositions. A self-evident proposition is one for which all the evidence needed to be justified in believing it is within the proposition itself. Such a proposition can become evident to an agent if the agent has adequate understanding of it (Audi Reference Audi2004: 48–49). Take a single self-evident principle of comparative stringency. Given the nature of such a proposition—it contains all the evidence needed to be justified in believing it and it is fundamental—no more fundamental substantive moral truth implies it. That means no proof can be given for it. Couple that with the fact that such a proposition is evident to an agent who adequately understands it and it is clear that no explanation of the proposition can be more convincing than the proposition itself. What we can do is provide an explanation to make explicit what is implicit in our understanding of such a proposition. An internal explanation can draw out what is internal to our understanding.

Take the principle that the duty not to murder is more stringent than the duty not to assault. Now consider the moral kinds involved. A major bad within a murder is death. The murder victim ceases to live. Since life is of paramount value for human beings, murder is not only wrong, but extremely morally bad. Within assaults there are serious bads as well. A successful assault will result in the assailant inflicting bodily damage on the victim. Such physical damage is a moral bad the assailant instantiates that would not have been instantiated if it were not part of an assault. If a natural accident produced the same damage to the victim's body, it would have been bad, but not a moral bad. The fact that the assailant intended and executed this infliction of the bad each contributes to making the act of assault morally bad. Having identified key components of two kinds, murder and assault, and having identified their respective values, we are able to ask which of the two moral values involved is worse. Death is worse than bodily damage. Since death is an essential component of murder and bodily damage is, likewise, an essential component of assault, this helps explain why murder is much worse than assault. Identifying such bads within moral kinds helps us to understand their comparative stringencies. Let us now turn to other real normative factors.

11. Moral Relations and Stringency

Prichard and Ross claimed our duties are based on relations we have. Prichard states, ‘The rightness of an action consists in its being the origination of something of a certain kind A in a situation of a certain kind, a situation consisting in a certain relation B of the agent to others or to his own nature’ (Prichard Reference Prichard1912: 27). He also claims that while there are special relations we can bear to others, to any agent we bear some relation merely by virtue of being human; let us call this the human-human relation. Ross states that other agents may ‘stand to me in the relation of promisee to promiser, of creditor to debtor, of wife to husband, of child to parent, of friend to friend, of fellow countryman to fellow countryman, and the like; and each of these relations is the foundation of a prima facie duty, which is more or less incumbent on me according to the circumstances of the case’ (Ross Reference Ross1930: 19).

Just as we will learn much about our duties by considering moral kinds as entities, we will learn more about them by considering moral relations as entities. The duties we are most apt to recognize as such involve others in some way. Lying necessarily involves lying to another. Murder necessarily involves murdering another. If we inquire into the many duties we currently have, it will not take us long to discover that we are at the center of a web of moral relations to others. Special relations explain why I owe duties to my children that I do not owe to yours, and unmysteriously the same goes for you. We will continue the program of explanatory expansion by positing moral relations and seeing how employing them matches up with what we really think about morality.

Moral kinds are the integral grounds of our duties. However, moral relations are also grounds of our duties. And for two acts of the same moral kind, an agent's being in two different moral relations can make one prima facie duty more stringent than the other. Between two lies that are the same, that are identical in terms of the speech act performed, one can violate a more stringent duty simply in virtue of the moral relation one has to another. Suppose you are chatting with a stranger on a bus. Out of the blue, she asks what your income is. You hesitate and consider lying to her. You think of a figure much lower than your income and tell her p: ‘I make $30,000’. It would certainly be wrong to do so, unless an overriding prima facie duty warrants it. Let us say none does. Consider now a situation in which you are at the dinner table with your wife. Merely out of curiosity she asks what your income is. You have not discussed your income for years as your two incomes have been serving your family adequately. You hesitate to answer. You consider lying to her. You think of a figure much lower than your income and tell her p: You tell her what you told the stranger. We would agree that this lie would be wrong, but we would immediately add it would be a much worse lie p to tell your wife. It seems, then, that we really think that the spouse-spouse relation is such that telling the lie p in violation of this relation is much worse than telling the very same lie p when you are only in a human-human relation.

As human beings you and your wife bear the human-human relation to one another, the very same relation you bear to the stranger on the bus. But the spouse-spouse relation you and your wife bear to one another puts many additional moral burdens on each of you. In the case of the lie p, clearly the spouse-spouse relation is more binding than the human-human relation alone. Similar examples can be multiplied. Stealing from a coworker is wrong, but stealing from a coworker who is a friend is worse. Murdering a stranger during a heated verbal dispute is wrong, but murdering a spouse during a heated verbal dispute is worse. The moral relation one is in can make an act of the kind L worse than the same act of that kind L when one is in a less binding relation. That is why telling a lie to one's spouse is morally worse than telling the same lie to a stranger. Unless there are other details that will make a difference between the two cases, this is something we all accept.

12. Axiological Factors and Stringency

The normative factors that determine what is right and what is wrong are kinds, relations, and values. They are the core deontic grounds of duty. Kinds are the integral grounds of duty. Moral relations can determine one prima facie duty to be more stringent than another. I hold that axiological factors also play a determinative role. Since I have identified moral goods and posited them to explain what about acts of distinct kinds make some kinds worse than others, I must emphasize that here I am speaking of nonmoral goods. Such goods are familiar to ethicists, for the vast majority of ethicists currently assume that nonmoral goods are the only goods.

Acts in virtue of being a lie are wrong. A lie such as p told to a spouse is worse than when it is told to a stranger. How do axiological factors determine what is right? Consider a husband telling his wife a lie about eating a cookie. He claims he did not have the last one, suggesting that one of the kids or the babysitter ate it. Since the lie is about such a trivial good, although it is wrong, we would say it is not a lie of great weight. However, if he lies to his wife about his income, that lie is of much greater weight. Since money is central to his family's welfare, that lie is a much more serious lie. What makes the difference between these cases is the value of the good involved. And we can think of goods of greater value than money. Life is more valuable than money. If a husband lies about a matter that has some likelihood of being fatal to his wife, that is worse than a lie he tells her about their money.

Stealing as a kind is comparatively worse than lying, when we consider acts of the two kinds in isolation from all else. It is a matter of common sense and a feature of our legal systems that stealing something that is worth a great deal is worse than stealing another thing worth much less. Consider two scenarios involving the stealing of a backpack. Jones comes out of his house and rests his backpack on the front steps. He takes some notebooks out of it, and walks back into the house. He comes back out, and inserts a number of jewelry boxes into the backpack. Remembering that he forgot something, he returns inside. Smith is a landscaper a few houses away who is watching Jones carefully. He runs over, opens the backpack, finds many pieces of jewelry inside, and runs away with it. In the second scenario, Jones removes jewelry boxes from his backpack and goes into the house to replace them with some notebooks. Smith waits until Jones is inside his house, then runs over, opens the bag, finding it empty, then runs off with it. It is evident that knowingly stealing a backpack with contents of great worth is much worse than stealing one whose only worth is the backpack itself.

That the comparative values involved in two similar situations can determine what is our duty is something we all accept. What I am arguing for here is that such goods must be understood in relation to moral kinds to understand properly what role they play in grounding our duties.

There is strong evidence that we already think this way even though ethicists do not currently recognize this in their theories. That evidence can be found in our penal codes. In New York State law, for example, larceny is divided in to petit and grand. Petit larceny is the charge of stealing an item worth $1,000 or less. Commission of it is considered a misdemeanor. Grand larceny, on the other hand, is a felony. The different grades of grand larceny go from grand larceny in the fourth degree, which is theft of an item worth $1,001 to $3,000, all the way to grand larceny in the first degree, which is theft of an item worth over $1,000,000. The penalties meted out are proportionate to the degree of larceny. The values involved in each act are only considered in relation to the act performed. If no theft were committed, the values of different possessions would be of no legal relevance. What I am claiming is that similar considerations hold for moral acts: moral kinds determine whether any value claim is or is not of moral relevance.

Our moral thinking is the same in moral and legal contexts. In each case considered, the value is involved in the kind of act being performed. Granted, the goods and the bads involved in the consequences of the acts have a role in determining the degree of wrongness of the different acts. Ross's account of duty includes the prima facie duty to maximize the good. What makes his view distinctive is that consequence considerations are important, but they are not all-important, as the consequentialists insist. My account does not exclude such matters as how Jones's life will be impacted by the two backpack scenarios. Here, the clear focus is on the act itself. And when we take a moral kind such as lie or theft, the values involved in these kinds of acts will contribute to determining the degree of wrongness of the acts themselves. But note that we are not simply making judgments of goods involved. We are making judgments of goods through and in relation to given moral kinds. This indicates that moral axiology is deontically dependent. Any role that good or bad plays in an internal explanation of what determines the rightness or wrongness of a given action is dependent on the kind of act. So goods and bads as structured by moral kinds contribute to determining what is right.

13. The Natures of Acts

We are in a position now to state what the nature of a moral act is. The first and often determinative factor is moral kinds, the integral grounds of duty. Moral relations are a secondary factor that can also contribute to the stringency of our prima facie duties. Lastly, values that are involved in the kind of act can strengthen or lessen its stringency. Moral kinds, moral relations, and values—these real factors combine to form the moral natures of acts.

$$\hbox{The nature of a moral act} = \hbox{moral kind} \lt \hbox{moral relation} + \hbox{value}\gt $$

A moral kind, as an integral factor, governs other factors within its scope. Often the moral kind of an act alone can determine the stringency of its corresponding duty and the severity of its consequent violation. These three factors are often enough to be decisive, but other factors might influence the degree of rightness or wrongness of the act. They can be roughly covered by how and when the act is done. But it must be stressed that the bulk of the weight is not on such factors.

14. Agential and Circumstantial Factors?

The core deontic grounds of duty are moral kinds, moral relations, and values. Following Ross's lead, I make no claim to completeness of the grounds of duty. However, I hold that often speaking of agential and circumstantial factors merely helps us pinpoint what kinds are involved. When we consider an immoral act of a given kind, the intention with which it is done can determine its degree of wrongness. Take a given lie (p): ‘I have no money’. If p is told out of fear, say during a mugging, it will be a lie-in-self-defense. In such a case, saying p is less wrong than if p is told for gain—a lie-for-gain—say, to get spare change from others. When the intention to do wrong is immediate and spontaneous, it is less wrong than when it is done with foresight and planning. One reason why we correctly punish a premeditated murder much more harshly than a crime of passion is that the premeditated murder involves having the intention to commit it for a time, a time over which one has many opportunities to change course. In each case in which an agent's intention changes the morality of the act, a different moral nature is instantiated than what would have been instantiated if the intention had been different.

What motivates the agent also impacts the stringency of the act. When a husband murders a just discovered lover of his wife, this mitigates the severity of the crime. It is not just that the intention has recently been formed. It is that the passion, the motive, overwhelmed the man. The law recognizes the role of motive, for example, in assessing acts of retaliatory violence in response to provocative speech. For example, if Jones says something extremely insulting about Smith's wife when her life is in the balance, it tempers our judgment of Smith's assault of Jones. Circumstantial factors are many. We can see what roles they play in instantiating different kinds. Leveling with a colleague about her poor performance might be honest, necessary for the good of the company, necessary for her to improve as a person in her career, and so it may be right. But doing so during a large company meeting could be cruel and therefore wrong. Spouses making sweet love can be a wholesome uniting of souls. Spouses making sweet love on top of the table in a crowded restaurant is a shocking public indecency and therefore wrong. Sharing private health information about Robinson's condition with a doctor may be obligatory if Robinson's life is at stake. But sharing such information about Robinson in a crowded elevator at work is a breach of confidence and therefore wrong. That we readily identify these acts as an act of cruelty, a public indecency, and as a breach of confidence shows that factors that may seem to be additions to our fundamental ones really indicate other kinds of actions. From all these cases we can see that whether agential factors or circumstantial factors alter the morality of the act, it is because of the different moral kinds their inclusion or exclusion instantiates.

15. Priority Rules

The claims of intuitionism rest on what we really think about morality. We are, in part, describing what we really think, with explanatory purposes in view. The factors involved are not arbitrary. And they are certainly not contrived, as most supreme principles are. By identifying, separating, and analyzing the distinct normative factors that make up the natures of moral acts, we have the basic ingredients for a system of priority rules. As we consider the real normative factors of moral kinds, moral relations, and values, there is no felt need for a sole factor that is invariably used in all cases—such as maximal pleasure or rational agency—as we come to our moral assessments.

What kind of priority rules does my intuitionism offer? As we consider this, we must bear in mind what the theories of our critics currently offer in the way of action guidance. For decades now consequentialists have abandoned the classic utilitarian claim to have a decision procedure (see Bales Reference Bales1971: 263 and Pettit Reference Pettit1993: xvi). Kantians reject the claim that their theory is committed to a decision procedure, and they even deny the claim, despite the clear textual evidence, that Kant holds that the formula of universal law of the categorical imperative can be used as a decision procedure (see Rawls Reference Rawls2000: 155; Nell Reference Nell1975: 73; Kant Reference Kant1993: 8).

Principles of comparative stringency are self-evident priority rules. However, since the grounds of duty are plural and since moral kinds, moral relations, and values can each be decisive, we cannot offer a unique, complete effective set of priority rules. On some occasions the moral relation involved will be decisive. Recognizing that that is my wife will often obviate any need to consider exactly what kind of immoral act doing a might constitute. Other occasions present the values involved as being of paramount moral importance. Recognizing that my friend's mint condition 1962 Shelby Cobra is priceless, informs me that asking whether driving at twice the speed limit now is negligent or reckless is to miss the fundamental moral import of my situation.

Nonetheless, the shape the moral grounds have together gives us a kind of procedure to help us decide what is right. Moral kinds are the integral grounds of our duties. Considering what kind of act we might do will often tell us whether it is permissible or impermissible. To show just how important moral kinds are in deciding, consider the lengths to which people will go to redefine mentally an obviously wrong act as an act of a permissible kind. When there is a large gap in the deontic weights of two moral kinds, as we consider principles of comparative stringency, often that will inform the agent of which duty is more stringent. However, if the moral kinds involved are such that they are roughly equivalent in deontic weight, then the moral relations the agent is involved in can determine what duty is more stringent. Next, if kinds and relations are roughly equivalent in deontic weight, the values involved might be decisive. If after all these factors are considered the two acts have equivalent deontic weight, then either action is the right thing to do.