This paper introduces and defends a new position regarding the question of whether it is possible to have conflicting moral obligations. In doing so, it focuses on what I call a moral pickle. By “moral pickle” I mean a set of actions such that you ought to perform each and cannot perform all. Typically, when people discuss conflicting moral obligations, they focus on the notion of a moral dilemma, which is a type of moral pickle involving two conflicting actions. In other words, a moral dilemma is a pair of actions such that you ought to perform each and cannot perform both. As of yet, there is no debate about the possibility of moral pickles over and above the possibility of moral dilemmas. But as I show, there is good reason to think that moral pickles are possible and moral dilemmas are not.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Here and elsewhere in this paper, in talking about actions one “can” perform or propositions one “can” believe, I mean to be referring to what is logically possible. For example, in the case of Logan, it is logically impossible to both grade each problem correctly and also make mistakes in grading now and again.
As I interpret the phrase “grade in such a way that she makes mistakes now and again”, Logan will only have graded in this way if she actually makes mistakes every now and again. An alternative interpretation is that it merely requires grading in such a way that one is likely to make mistakes. I think my interpretation is the more natural one. Some evidence for my interpretation comes from looking at other phrases with a similar structure. For instance, consider the recommendation: “deliver philosophy papers in such a way that the audience laughs from time to time.” On the most natural interpretation of this phrase, in order to deliver one’s papers in this way, audience members must actually laugh occasionally during one’s philosophy talks. In other words, if I give a number of philosophy talks and no one ever laughs during any of them, I have not delivered my philosophy talks in such a way the audience laughs from time to time. (Just in case the reader is worried that I rest my case for moral pickles on this point about interpretation, I should note that I offer arguments later on in this paper for the claim that Logan ought to make a mistake every now and again, as opposed to merely having the obligation to grade in such a way that she’s likely to make a mistake every now and again.) Thanks to an anonymous referee for pressing me on this.
I am assuming that, for each problem, Logan has a moral obligation to grade it correctly. One reason to think this is because grading it incorrectly would be unfair to the student who submitted it. If you nonetheless doubt that this is a moral obligation, feel free to alter the example to make it more obviously a moral one. One can substitute in any minor, easy-to-perform task, for which it’s possible that someone will make a mistake due to lack of attention. For instance: turning off light switches to save electricity, transcribing credit card numbers for people who have decided to donate to some charity, etc.
There are other interesting distinctions in the neighborhood. For example, one can distinguish between “higher-order moral pickles” and “lower-order moral pickles”. To motivate the distinction, return to the example in which Logan is told she should grade in such a way that she make mistakes now and again. This obligation is (arguably) “higher-order” where a higher-order obligation is an obligation to act in such a way as to bring about other actions. In particular, the action in question is presumably the adoption of some sort of procedure that will in turn lead her to grade problems in a certain way. A higher-order moral pickle, then, is a moral pickle in which some of the obligations are higher-order, whereas a first-order moral pickle is one where none of the actions are.
The obligation to make mistakes is also (arguably) “complex” in that it involves multiple actions. An obligation is “simple” if it involves just one action. (There are presumably multiple ways to individuate what counts as a single action, but on at least some of these, grading in such a way that one makes mistakes from time to time counts as multiple actions, one for each problem one grades.) This yields another distinction: a “complex moral pickle” is one on which some of the actions are complex, and a “simple moral pickle” would be one on which all the actions are simple. Thanks to an anonymous referee for encouraging me to discuss these related distinctions.
Given that attention has not been paid to the distinction between moral dilemmas and moral pickles, there has been some variation in the way “moral dilemma” has been defined in the literature. Often, it is defined as I have defined it (see e.g. Brink 1994, 214–5; Haan 2001, 269; Marcus 1980, 122; Rajczi 2002, 310; Sinnott-Armstrong 1987, 128), but sometimes people use the word “moral dilemma” as I use the word “moral pickle” (see e.g. McConnell 2014). I don’t wish to quibble over which words one uses to pick out which things, as long as these things are properly distinguished. But perhaps I should offer a modest defense of my choice of terminology: in addition to capturing the way that most have used the term “moral dilemma”, it also has etymological advantages, as “di” means two.
So far as I know, no one has offered a version of the preface paradox for moral obligation. That said, recent papers by Goldstein (2016) and Shpall (2016) have focused on similar issues; they both offer a version of the preface paradox for intention. It is perhaps worth briefly noting some ways in which our papers differ. In particular, they focus on whether one can be rationally obligated to hold conflicting intentions, while I am focused on whether one can be morally obligated to perform conflicting sets of actions. This difference in focus yields important differences regarding the upshots of our papers; the key upshot of my paper is that the literature on moral dilemmas has neglected an important but attractive alternative, namely the alternative on which moral dilemmas are impossible but moral pickles are possible.
That many respond in this way is noted in e.g. Sorensen (2011).
In my discussion of the dilemma, I am repeating points made by David Christensen; see Christensen (2004, 33–68).
Thanks to an anonymous referee for raising this.
Perhaps one might respond that my obligation was to ensure that both cats were fed and that this obligation would hold even if the robber broke in (though of course it would be easier to fulfill in that case). But this is subject to the same sort of worry. It was possible for one of my cats to die, in which case there would be no need to feed it. Maybe the obligation was to ensure that both of my cats were fed, so long as they were still alive? But what if one of them suddenly developed some condition that meant if it ate, it would get deathly ill? And so on. Far simpler and more natural just to say that, because none of these unlikely events took place, I had an obligation to feed both cats.
Thanks to an anonymous referee for pressing this worry.
Thanks to an anonymous referee for suggesting that I expand this paragraph and offering some advice on how to do so.
Also known as the weak obligation principle (Horty 2012, 93).
Also known as closure under logical consequence (Horty 2012, 83).
Here’s the argument: suppose there is a moral dilemma, i.e. two actions A and B such that one ought to perform each but cannot perform both. Then because one cannot perform both, performing A entails not performing B, so by Deontic Logic, if A is obligatory, then so is not performing B. But because we’re assuming that performing A is obligatory, it follows that not performing B is obligatory. But we have assumed Deontic Consistency and that performing B is obligatory so it follows that it’s not the case that not performing B is obligatory. But now we have a contradiction; not performing B can’t both be obligatory and not obligatory.
For instance, suppose there are three actions, A, B, and C. Suppose that one ought to perform each, but it’s not the case that one ought to perform any pair, or all three. Suppose that one can perform any two but can’t perform all three. Then both Deontic Consistency and Deontic Logic hold, as can be easily checked.
Brennan, G., & Southwood, N. (2007). Feasibility in action and attitude. Hommage à Wlodek: Philosophical Papers Dedicated to Wlodek Rabinowicz.
Brink, D. (1994). Moral conflict and its structure. The Philosophical Review, 103, 215–247.
Christensen, D. (2004). Putting logic in its place; Formal constraints on rational belief. Oxford: Oxford University Press.
Conee, E. (1982). Against moral dilemmas. The Philosophical Review, 91(1), 87–97.
Estlund, D. (2011). Human nature and the limits (if any) of political philosophy. Philosophy and Public Affairs, 39, 207–237.
Foley, R. (2009). Beliefs, degrees of belief, and the lockean thesis. In F. Huber & C. Schmidt-Petri (Eds.), Degrees of belief. Berlin: Springer.
Goldstein, S. (2016). A preface paradox for intention. Philosophers’ Imprint, 16(14), 1–20.
Haan, J. D. (2001). The definition of moral dilemmas: A logical problem. Ethical Theory and Moral Practice, 4(3), 267–284.
Harman, G. (1986). Change in view. Cambridge: MIT Press.
Horty, J. F. (2012). Reasons as defaults. Oxford: Oxford University Press.
Lawford-Smith, H. (2013). Understanding political feasibility. Journal of Political Philosophy, 21, 243–259.
Makinson, D. C. (1965). The paradox of the preface. Analysis, 25, 205–207.
Marcus, R. B. (1980). Moral dilemmas and consistency. The Journal of Philosophy, 77(3), 121–136.
McConnell, T. (2014). Moral dilemmas. Stanford Encyclopedia of Philosophy.
Rajczi, A. (2002). When can one requirement override another? Philosophical Studies, 108(3), 309–326.
Shpall, S. (2016). The calendar paradox. Philosophical Studies, 173(3), 801–825.
Sinnott-Armstrong, W. (1987). Moral dilemmas and ‘ought and ought not’. Canadian Journal of Philosophy, 17(1), 127–140.
Sorensen, R. (2011). Epistemic paradoxes. Stanford Encyclopedia of Philosophy.
Southwood, N. (2015). Democracy as a modally demanding value. Noûs, 49, 504–521.
Southwood, N. (2016). Does “ought” imply “feasible”? Philosophy and Public Affairs, 44(1), 7–45.
Southwood, N., & Wiens, D. (2016). “Actual” does not imply ”feasible”. Philosophical Studies, 173(11), 3037–3060.
Vihvelin, K. (2004). Free will demystified: A dispositional account. Philosophical Topics, 32, 427–450.
Williams, B. (1965). Ethical consistency. Proceedings of the Aristotelian Society, Supplementary Volumes, 39, 103–124.
Worsnip, A. (2016). Belief, credence, and the preface paradox. Australasian Journal of Philosophy, 94(3), 549–562.
Thanks for helpful comments to Earl Conee, Ryan Hammond, Ting Lau, Caleb Perl, Will Smith, Fritz Warfield, several anonymous referees, and an audience at the 2016 Pacific APA.
Conflict of interest
Author has no financial interest or benefit arising from the direct applications of their research.
About this article
Cite this article
Immerman, D. Moral pickles, moral dilemmas, and the obligation preface paradox. Philos Stud 176, 2087–2101 (2019). https://doi.org/10.1007/s11098-018-1116-5
- Moral pickle
- Moral dilemma
- Preface paradox
- Obligation preface paradox