At this point in my discussion, I need to make fully explicit something that I have only briefly sketched above, which is that there are three essentially different logico-semantic types of Kantian moral principles, and that these types should be carefully sorted into a three-levelled hierarchy of principles running from the highest level to the lowest level, as follows:

LEVEL 1: Absolutely Universal and Objective Moral Meta-Principles, that is, the four analytically equivalent formulations of the Categorical Imperative, knowable by essentially reliable self-evident moral intuition.

LEVEL 2: Fairly Universal First-Order Substantive Ceteris Paribus Objective Moral Principles, knowable by fairly reliable cognitive and practical construction.

LEVEL 3: Moral Duties and Moral Licenses, that is, the obligatory or merely permissible complete act-intentions or fully meaningful maxims, knowable by moral judgment.

In broadly Kantian nonideal dignitarian moral theory, moral principles inherently guide the intentional acts of human real persons. So this is what I’ll call The Hierarchy Interpretation of the broadly Kantian nonideal dignitarian theory of moral principles, not only in the sense that the principles themselves must be hierarchically sorted in this way, but also in the sense that their psychological realization in the wills of intentional agents carries the same basic hierarchical structure.

In the three-levelled hierarchy of Kantian moral principles, principles at lower levels or types presuppose all the principles at the higher levels or types. Furthermore, the principles occurring at each level are all logico-semantically and normatively equivalent with all the other principles occurring at that level.

More specifically, the various formulations of the absolutely universal and objective Categorical Imperative are all analytically equivalent with one another across logically and conceptually possible worlds at LEVEL 1.

In turn, the so-called perfect and imperfect “duties,” the fairly universal first-order substantive ceteris paribus objective moral principles, are all necessarily extensionally equivalent with one another across relevant, really possible act-worlds at LEVEL 2.

And finally, the moral duties and moral licenses are all biconditionally equivalent with one another across the actual world at LEVEL 3.

The modal scope of each of the three levels is also distinct from the others.

The absolutely universal and objective moral meta-principles at LEVEL 1, namely, the four analytically equivalent formulations of the Categorical Imperatives, hold for all logically and conceptually possible worlds, or for all and only finite, embodied rational beings, that is, beings possessing an innately-specified capacity, or faculty, of pure practical reason, along with a faculty of desire and a sensibility. My own view is that necessarily, all finite, embodied rational beings are also minded animals, and therefore living organisms,[i] but this thesis is not absolutely required for the main point I want to make here. The main point is that the domain of finite, embodied rational beings comprehended by the moral meta-principles at LEVEL 1 is not restricted to human beings, but instead includes any finite, embodied rational beings whatsoever, whether they are human, or artificially-constructed biological systems (as are the “replicants” in Philip K. Dick’s famous science-fiction novel, Do Androids Dream of Electric Sheep, and in its correspondingly famous film version, directed by Ridley Scott, Blade Runner), or alien.

By sharp contrast, the fairly universal first-order substantive ceteris paribus objective moral principles at LEVEL 2 hold for all and only finite, embodied rational human beings or real human persons, that is, rational beings who are also essentially embodied as living organisms, or animals, in essentially the way we are embodied, and are thereby naturally disposed to pursue and produce happiness.

And finally the moral duties and moral licenses at LEVEL 3, as principles of volition or rational choice, hold for all and only the actual, context-dependent intentional acts of rational human beings or real human persons.

Here’s a compact diagram of the structuralist system of broadly Kantian nonideal dignitarian moral principles I’ve just been describing:

Now according to broadly Kantian nonideal dignitarian moral theory, as a specifically nonideal dignitarian moral theory, the three-levelled hierarchy of Kantian moral principles is governed by two distinct but closely-related level-theoretic structural constraints. Here’s the first structural constraint:

1. The No-Global-Violation Constraint. In order for a choice or act to be permissible or obligatory in any actual act-context, there cannot be any violation of moral principles of the highest type and at the highest level in the hierarchy of principles. That is, there cannot be any violation of any of the analytically equivalent formulations of the Categorical Imperative at LEVEL l, even if there are violations of first-order substantive ceteris paribus objective moral principles at LEVEL 2.

The No-Global-Violation Constraint strictly forbids violations or inconsistencies of any absolute moral principles at LEVEL 1, which would be global or context-invariant violations or inconsistencies of moral principles, even though the Constraint also permits, in some act-contexts, local or context-sensitive violations or inconsistencies of first-order substantive ceteris paribus objective moral principles at LEVEL 2. And here is the second structural constraint:

2. The Excluded Middle Constraint. If an agent has a moral duty in an actual act-context, then there is always one and only one moral duty for the agent in that act context — no matter how difficult it is for the agent herself to discern it — and acting on any other conflicting first-order substantive ceteris paribus objective moral principle applying to the agent in that context, is morally impermissible in that context.

It should be obvious how The Excluded Middle Constraint relates to the sharp distinction between moral principles and moral duties. Even though several different and possibly even really conflicting first-order substantive ceteris paribus objective moral principles might apply to a given intentional agent in a given actual act-context, nevertheless she only ever has one moral duty in that context, if indeed she has any moral duty at all in that context. The Excluded Middle Constraint is a level-theoretic constraint that guarantees an empirical moral realism within the non-platonistic, non-naturalistic (that is, not reductively or scientifically naturalistic) framework of Kantian constructivism in ethics. And it is obviously also intimately related to the fact that The Formula of Universal Law is a formally presupposed, absolutely universal and objective minimal moral meta-principle that imposes global moral consistency on the hierarchical system of moral principles in a thoroughly nonideal natural and social world.

I’m now in a position to begin to face up to the problem of moral dilemmas on behalf of broadly Kantian nonideal dignitarian moral theory According to the standard interpretation of the Kantian theory of moral dilemmas, the Kantian moralist accepts that there are apparent or prima facie moral dilemmas, but then denies that there are genuine or real moral dilemmas. The standard interpretation is adequately supported by the (unfortunately, very few) relevant texts in which Kant explicitly discusses moral dilemmas. But the glaring philosophical problem with this standard Kantian theory of moral dilemmas is that it just does not seem to comport with the actual facts about our moral lives on the ground in a thoroughly nonideal natural and social world. That is, everyday moral experience, moral commonsense, and above all, existential insight, all self-evidently tell us that there are real moral dilemmas, or at the very least, that moral dilemmas are really possible.

So, let’s suppose instead that we accept either the actuality or the real possibility of moral dilemmas, as some existentially insightful ethicists — for example, Kierkegaard, Sartre, Bernard Williams, and Martha Nussbaum[ii] — would have us do. If so, then the Kantian system of moral principles would apparently lead to contradictions or dilemmas, and “morality totters.” The so-called supreme moral principle, the Categorical Imperative, would then be bogus, and pure practical reason would then be nothing but an evolution-generated cognitive self-defense mechanism for deceiving ourselves about the awful truth — a morally chaotic world. Tragically, it would be even worse than Hobbes’s imaginary state of nature, in which human life, although nasty and brutish, was at least mercifully short. In the world of the bogus and self-contradictory Categorical Imperative, it would be the end of the world as we morally know it, forever.

I call this The Moral Doomsday Scenario. It’s clear and distinctly true that neither the actuality nor the real possibility of The Moral Doomsday Scenario can be permitted by any acceptable version of Kantian moral theory. Therefore, in order to avoid both the actuality and the real possibility of moral dilemmas, any acceptable version of Kantian morality must be able to prove that there cannot be such things as real moral dilemmas. This in turn leads to a meta-dilemma:[iii] Either we accept the standard interpretation of the Kantian theory of moral dilemmas, in which case Kantian moral theory ends up not being in conformity with everyday moral experience, moral commonsense, and existential insight, by denying the actuality and real possibility of moral dilemmas; or else we accept the claims of moral experience, moral commonsense, and existential insight, in which case we leave Kantian morality open to the actuality or real possibility of The Moral Doomsday Scenario.

Of course, Kantian moral theory is not the only moral theory that has to face up to this sort of meta-dilemma.[iv] More generally, for any normative ethical theory grounded on universal principles, either one rejects the existence of moral dilemmas, in which case one’s meta-ethics ends up not being in conformity with everyday moral experience, moral commonsense, and existential insight, by denying the actuality and real possibility of moral dilemmas; or else one accepts the claims of moral experience, moral commonsense, and existential insight, in which case one leaves one’s meta-ethics open to the actuality or real possibility of a relevant analogue of The Moral Doomsday Scenario.

Now, what should we do? To be sure, every contemporary Kantian moral theorist has their own take on the problem of moral dilemmas.[v] As a defender of broadly Kantian nonideal dignitarian moral theory, however, I think that it would be a very good thing indeed if there were a fully distinct third candidate interpretation of the Kantian theory of moral dilemmas, over and above the standard interpretation and also the interpretation mandated by everyday moral experience, moral commonsense, and existential insight, that’s also importantly different from those of other contemporary Kantian moral theorists. This would be a fully distinct third interpretation that is relevantly supported by the Kant texts just like the other interpretations, and independently defensible, but also resolves the meta-dilemma.

What might such a fully distinct and philosophically adequate third interpretation look like? In view of what I’ve already argued, one point that should immediately strike us is the obvious structural analogy between moral contradictions or dilemmas in a Kantian structuralist system of moral principles in a thoroughly nonideal natural and social world on the one hand, and alethic contradictions or paradoxes (antinomies, hypercontradictions) in classical or non-classical logical systems on the other. This in turn provides us with a philosophical working clue: What if we thought about moral dilemmas in terms of the broadly Kantian nonideal dignitarian hierarchical structuralist system of Kantian moral principles, in parallel with thinking about logical contradictions or paradoxes in non-classical logical systems? And in particular, what if we thought about a formal analogy between that hierarchical structuralist system of Kantian moral principles on the one hand, and dialetheic paraconsistent logical systems on the other?

Here’s the formal analogy I have in mind. A dialetheic paraconsistent logic explicitly allows for local, logically restricted, or non-Explosive contradictions, while also explicitly ruling out global, logically unrestricted, or Explosive contradictions that lead to logical anarchy or chaos. Moreover, and now also thinking in terms of logic, semantics, and linguistic pragmatics, we can think of local, restricted, or non-Explosive contradictions as being context-sensitive and systematically variable, and of global, unrestricted, or Explosive contradictions as being context–insensitive and systematically invariant. Correspondingly then, we can hold that a properly-designed broadly Kantian nonideal dignitarian structuralist system of Kantian moral principles will explicitly allow for real local, restricted, context-sensitive and systematically variable moral dilemmas, while also explicitly ruling out global, logically unrestricted, context-insensitive and systematically invariant moral dilemmas, which would Explosively entail moral anarchy or chaos — that is, moral contradictions or dilemmas literally all over the place, moral Doomsday. The actuality and real possibility of local moral dilemmas would conform to everyday moral experience, moral common sense, and existential insight; and ruling out global moral dilemmas would also prevent the possibility of The Moral Doomsday Scenario. So we would then have a third interpretation of the Kantian theory of moral dilemmas that prevents the meta-dilemma, and correspondingly we would also have a richer interpretation of the broadly Kantian nonideal dignitarian structuralist system of Kantian moral principles.

Now, granting me also The Hierarchy Interpretation of the broadly Kantian nonideal dignitarian structuralist system of Kantian moral principles as a working hypothesis, I can then develop it in full conjunction with the paraconsistent logic analogy, and thereby spell out — while also, as I mentioned earlier, riffing on Emerson’s famous remark about the hobgoblin of small minds — what I call The No-Foolish-Consistency Interpretation of our broadly Kantian nonideal dignitarian hierarchical structuralist system of Kantian moral principles, in five steps, as follows.

First, although least importantly, the relevant Kantian texts adequately support The No-Foolish-Consistency Interpretation:

Act only on that maxim by which you can at the same time will that it should become a universal law. (GMM 4: 421, boldfacing added)

A conflict of duties would be a relation between them in which one of them would cancel the other (wholly or in part)…. But since duty and obligation are concepts that express the objective practical necessity of certain actions and two rules opposed to each other cannot be necessary at the same time, if it is a duty to act in accordance with one rule, to act in accordance with the opposite rule is not a duty but even contrary to duty; so a collision of duties and obligations is inconceivable. However, a subject may have, in a rule he prescribes to himself, two grounds of obligation, one or the other of which is not sufficient to put him under obligation, so that one of them is not a duty. (MM 6: 224, boldfacing added)

On the appropriate reading of those texts, (i) the phrase “should become a universal law” means that no moral principle can be self-contradictory in a global, logically unrestricted, context-invariant way, which makes the Formula of Universal Law, when understood independently of its real-world application to particular actual act-contexts, the moral equivalent of Minimal Non-Contradiction, (ii) the phrase “a collision of duties and obligations is inconceivable” means that there cannot be global, unrestricted, context-invariant moral dilemmas, because this would entail that some moral principles are globally, unrestrictedly, and Explosively self-contradictory, (iii) the phrase “ground of obligation” means a first-order substantive ceteris paribus objective moral principle, and (iv) the phrase “a subject may have, in a rule he prescribes to himself, two grounds of obligation” means that there can still be real local, logically restricted, context-sensitive moral dilemmas between first-order substantive ceteris paribus objective moral principles.

Second, and more importantly, in order to support The No-Foolish-Consistency Interpretation in the ways I’ve just indicated, we need only to formulate, within the broadly Kantian nonideal dignitarian hierarchical structuralist framework of moral principles, a proper analogue of a special paraconsistent no-Explosion axiom, that is, a moral meta-principle which explicitly allows for real local, logically restricted, context-sensitive moral dilemmas in a thoroughly nonideal natural and social world, while at the same time explicitly constraining their logico-moral powers so that logico-moral chaos cannot Explosively result from them. Correspondingly, my proposal is that we adopt the following broadly Kantian nonideal dignitarian moral meta-principle, which, as I mentioned above, I call The Lesser Evil Principle:

Given a real local moral dilemma between first-order substantive ceteris paribus objective moral principles, you ought to choose the first-order substantive ceteris paribus objective moral principle which in that actual act-context is the lesser of the several evils, in the sense that acting on it keeps rational faith with the Categorical Imperative to the greatest possible extent.[vi]

By virtue of The Lesser Evil Principle, any real violations of or real inconsistencies between first-order substantive ceteris paribus objective moral principles (that is, principles at LEVEL 2 in the hierarchy), are automatically resolved by the moral agent’s being required, in any given actual act-context, to choose and act on the first-order substantive ceteris paribus objective moral principle that is the lesser of several evils in that context, in the sense that this is the moral principle which in that actual context keeps rational faith with the Categorical Imperative to the greatest possible extent.

What, more precisely, do I mean by a moral principle’s “keeping rational faith with the Categorical Imperative to the greatest possible extent” in a given actual act-context? One thing I mean is that a moral principle MP1 keeps rational faith with the Categorical Imperative to the greatest possible extent in a given act-context if and only if

(i) in that act-context MP1 belongs to a holistic conceptual network which necessarily analytically includes all of the four necessarily equivalent formulations of the Categorical Imperative, as instantiated or implemented in that act-context — hence MP1is either analytically entailed by or analytically entails the Categorical Imperative in that act-context — and

(ii) any other relevant moral principle MP2 in that act-context is either

(iia) merely analytically consistent with, or else (iib) analytically rejected by that holistic network.

And another thing I mean is that a moral principle MP1 keeps rational faith with the Categorical Imperative to the greatest possible extent in a given act-context if and only if

(i) in that act-context MP1 adequately expresses the Categorical Imperative, and

(ii) any other relevant moral principle MP2 in that act-context is either

(iia) merely consistent with the adequate expression of the Categorical Imperative, or else (iib) fails adequately to express the Categorical Imperative.

I’m taking these two formulations of the notion of rational-faith-keeping-to-the-greatest-possible-extent, in terms of either analytic entailment or adequate expression, to be necessarily equivalent.

Third, and also importantly, we should explicitly note that insofar as The Lesser Evil Principle and The Excluded Middle Constraint together select a certain first-order substantive ceteris paribus objective moral principle to be the agent’s duty in an actual act-context, all sorts of ineluctably actualist or real-world factors will contribute to determining precisely which moral principle is to be chosen, including human desires and causal act-consequences of acts, amongst which will be private utility and public utility. But it would be a serious non sequitur to conclude from this, that the selected principle which is my duty in that context must be an ethically egoistic or act-consequentialist principle, or that it has been selected for ethically egoistic or act-consequentialist reasons. More generally, it’s crucial not to confuse the phenomenon of context-sensitivity and systematic variability in broadly Kantian nonideal dignitarian moral theory with either ethical egoism or act consequentialism. And it’s equally crucial not to fall into an instrumentalist fallacy of thinking that just because every intentional act in this thoroughly nonideal natural and social world begins with human desires and causal act-consequences, therefore its guiding principle must be dependent on human desires and act-consequences. It should be obvious that there is a close parallel here with Kant’s famous warning against falling into the Empiricist fallacy of thinking that just because every cognition begins with experience, therefore its content or truth must be derived from experience (CPR B1).[vii]

Fourth, less importantly, The Lesser Evil Principle also captures a plausible reading of this relevant Kantian text:

When two such grounds [of obligation] conflict with each other, practical philosophy says, not that the stronger obligation takes precedence … but that the stronger ground of obligation prevails. (MM 6: 224, boldfacing added)

On the appropriate reading of the text, (i) the phrase “when two such grounds [of obligation] conflict with each other” means morally unlucky real-world situations in which there is a local, logically restricted, context-sensitive moral dilemma between first-order substantive ceteris paribus objective moral principles, and (ii) the phrase “the stronger ground of obligation prevails” means that you ought to choose the first-order principle which in that context is the lesser of the several evils, in the sense that acting on it most keeps rational faith with the Categorical Imperative.

Fifth, finally, and most importantly, The Lesser Evil Principle together with The No-Global-Violation Constraint and The Excluded Middle Constraint, jointly guarantee that in any act-context in which the agent has a moral duty, there will be one and only one first-order substantive ceteris paribus objective moral principle that is her moral duty in that context. This is because The Principle together with the two Constraints collectively rule out any global, logically unrestricted, context-invariant violation or inconsistency of moral principles in the overall structuralist system of principles, even in cases of real local, logically restricted, context-sensitive violations or inconsistencies of first-order substantive ceteris paribus objective moral principles, that is, even in cases of real local moral contradictions or dilemmas.

One crucial consequence of The No-Foolish-Consistency Interpretation of the broadly Kantian nonideal dignitarian structuralist system of Kantian moral principles is that The Lesser Evil Principle reinstates, in a non-classical global version, the classical Kantian “ought implies can” principle, which would otherwise be directly threatened by the existence of real local moral dilemmas. If The Lesser Evil Principle does not hold, then if in some actual act-context I am obligated to choose or do A (for example, lie to a murderer because otherwise he will murder an innocent person) and also in that context I am obligated to choose or do not-A (that is, not lie to the murderer because it is always wrong to lie), then morally I cannot choose or do either A or not-A in that act-context — and ought does not imply can. But if The Lesser Evil Principle does hold, together with The No-Global-Violation Constraint and The Excluded Middle Constraint, then I am obligated to choose or do either A or not-A in that act-context, depending on which is contextually selected to be my duty by the Principle together with the Constraints, and then it follows that either morally I can do A (because in that context I am obligated to do A) or morally I can do not-A (because I am obligated to do not-A), but not both — and ought implies can again.[viii}

It’s also crucial to notice that the process of discovering just which of the conflicting first-order substantive ceteris paribus objective moral principles is the lesser of the several evils in that actual act-context is an issue for moral judgment, but not an issue for the non-classical skinny logic and the non-classical fat semantics of Kantian moral principles. The fact that it may be extremely or even almost impossibly difficult for a moral agent to discover just which principle is the lesser of the several evils in that actual act-context — for example, the poignant case in which you are lucidly asked by your suffering sibling, parent, or life-partner to assist in his or her suicide, as per the real-world case of Stephan and Edith Körner — is entirely irrelevant to the application of The Lesser Evil Principle to that act-context. The Lesser Evil Principle rules out the possibility of any global moral dilemma, and it thereby rules out The Moral Doomsday Scenario. But it does not in any way underestimate or undermine the force and significance either of real local moral dilemmas, or of the very real epistemic difficulties of moral judgment in these actual act-contexts, or indeed of the very real human pathos and tragedy of such situations.

As a consequence of this last point, we rightly feel deep moral pity for Sophie in Sophie’s Choice when she must choose which of her two children will be killed by the Nazis; and we also rightly feel a deep moral terror when we imagine her awful situation. We thereby fully acknowledge the reality of local moral dilemmas, and also fully acknowledge the corresponding problem of moral judgment in local moral dilemma situations. But it would be no moral comfort whatsoever to Sophie or anyone else if the Moral Doomsday Scenario were true, if the Categorical Imperative turned out to be bogus, if moral chaos were lurking like a post-ethical Godzilla behind the thin façadeof everyday etiquette and prudential conduct, and if human morality and rationality went into the abyss. There is a world of difference between introducing a tragic sense of life[ix] into Kantian ethics on the one hand, and being a moral terrorist like Peter Verkhovensky in Dostoevsky’s The Devils on the other hand. Fully acknowledging the existence and implications of real local moral dilemmas is a deep and ultimately life-affirming and morality-affirming existentialist insight. But asserting or allowing for the existence and implications of real global moral dilemmas is, emphatically on the contrary, tantamount to moral nihilism.[x]

In any case, we can now see that The No-Foolish-Consistency Interpretation of the Kantian theory of moral principles is specifically designed to impose on the total structuralist hierarchy of broadly Kantian nonideal dignitarian moral principles precisely the same sort of invulnerability to global, logically unrestricted, context-invariant moral dilemmas that Alfred Tarski’s meta-linguistic solution to the semantic paradoxes (and in particular the Liar sentence) imposes on natural languages and logical systems.[xi] Tarski’s idea was to stipulate that the logical and semantic predicates, especially including the truth-predicates, for a given language, always belong to its meta-language. Then it is impossible to form a Liar sentence in any language. It’s only if the hierarchy of languages and meta-languages is collapsed into a single undifferentiated “flatland” language, that paradoxes can occur. In a precisely analogous way, in the context of the broadly Kantian nonideal dignitarian structuralist system of moral principles, it’s only if all moral principles were treated as if they logically, semantically, and normatively belonged to the same level, that global moral dilemmas could occur. Just to give it a convenient name, I’ll call this rationally disastrous fallacy of collapsing the levels of the hierarchy of principles, whether in logic or morality, the flatlander fallacy.

In the mid-1970s, Saul Kripke noticed that, due to the irreducible presence of indexical terms in natural languages, instances of The Liar Paradox can occur contingently, and also that natural languages can consistently contain their own truth-predicates.[xii] Ruth Barcan Marcus then also noticed that moral dilemmas can have essentially the same logical and semantic structure as contingent occurrences of The Liar, even when sets of first-order moral principles are consistent.[xiii] Correspondingly, according to The No-Foolish-Consistency Interpretation, real context-sensitive and systematically variable conflicts of first-order substantive ceteris paribus objective principles are allowed to occur at the second level of the hierarchy of moral principles. This adequately captures the standpoint of everyday moral experience, moral commonsense, and existentialist insight. But at the same time, we are rationally enabled to live with these real local moral contradictions or dilemmas in the thoroughly nonideal natural and social world, precisely because The Lesser Evil Principle holds, moral Explosion is thereby prevented, and The Moral Doomsday Scenario is thereby ruled out.

Taken together, then, The No-Global-Violation Constraint, The Excluded Middle Constraint, and The Lesser Evil Principle collectively guarantee the intact good will of any rational “human, all too human” agent who cognitively constructs, practically constructs, and then wholeheartedly volitionally implements the broadly Kantian nonideal dignitarian hierarchical structuralist system of moral principles in our thoroughly nonideal world. Here we must always remember that, according to broadly Kantian nonideal dignitarian moral theory, as a version of Kantian constructivism, a system of moral principles is neither an abstract object in moral platonic heaven nor a shifting bundle of merely natural facts. Instead, this Kantian constructivist system of moral principles is nothing more and nothing less than a categorically normative immanent structure in a rational human minded animal’s or real human person’s actual conscious and self-conscious will; and this immanent structure has both irreducible psychological reality and also robust causal efficacy.

NOTES

[i] See R. Hanna and M. Maiese, Embodied Minds in Action (Oxford: Oxford Univ. Press, 2009).

[ii] See, e.g., B. Williams, “Ethical Consistency,” in B. Williams, Problems of the Self (Cambridge: Cambridge Univ. Press, 1973), pp. 166–186; B. Williams, “Moral Luck” and “Conflicts of Values,” both in B. Williams, Moral Luck (Cambridge: Cambridge Univ. Press, 1981), pp. 20–39 and 71–82; and M. Nussbaum, The Fragility of Goodness (Cambridge: Cambridge Univ. Press, 1986).

[iii] See, e.g.,T. Hill, “Moral Dilemmas, Gaps, and Residues,” in T. Hill, Human Welfare and Moral Worth: Kantian Perspectives (Oxford: Clarendon/Oxford University Press, 2002), essay 12, pp. 362–402; A. Donagan, “Consistency in Rationalist Moral Systems,” Journal of Philosophy 81 (1984): 291–309; and A. Donagan, “Moral Dilemmas, Genuine and Spurious: A Comparative Anatomy,” Ethics 104 (1993): 7–21.

[iv] See, e.g., G. Sayre-McCord, “A Moral Argument Against Moral Dilemmas,” available online at URL= <https://philosophy.unc.edu/wp-content/uploads/sites/122/2013/10/A-Moral-Argument-Against-Moral-Dilemmas.pdf>.

[v] See, e.g., C. Korsgaard, “The Right to Lie: Kant on Dealing with Evil,” in C. Korsgaard, Creating the Kingdom of Ends (Cambridge: Cambridge Univ. Press, 1996), pp. 133–158, Hill, “Moral Dilemmas, Gaps, and Residues”; and S. Shiffrin, Speech Matters: On Lying, Morality, and the Law (Princeton, NJ: Princeton Univ. Press, 2016), esp. ch. 1.

[vi] For a similar idea, see T. Schapiro, “Kantian Rigorism and Mitigating Circumstances,” Ethics 117 (2006): 32–57.

[vii] See also R. Hanna, Cognition, Content, and the A Priori: A Study in the Philosophy of Mind and Knowledge (THE RATIONAL HUMAN CONDITION, Vol. 5) (New York: Nova Science, 2018), section 4.5.

[viii] See also, e.g., D. Jacquette, “Moral Dilemmas, Disjunctive Obligations, and Kant’s Principle That ‘Ought’ implies ‘Can’,” Synthese 88 (1991): 43–55.

[ix] See, e.g., M. Unamuno, The Tragic Sense of Life, available online at URL = <http://www.gutenberg.org/files/14636/14636-h/14636-h.htm>.

[x] On and against moral nihilism, see also A. Camus, The Rebel, trans. A. Bower(New York: Vintage, 1956).

[xi] See A. Tarski, “The Semantic Conception of Truth and the Foundations of Semantics,” Philosophy and Phenomenological Research 4 (1943–44): 341–375.

[xii] S. Kripke, “Outline of a Theory of Truth,” Journal of Philosophy 72 (1975): 690–715.

[xiii] R.B. Marcus, “Moral Dilemmas and Consistency,” Journal of Philosophy 77 (1980): 121–136.