Definitions. Let the initial "objects" of moral judgment be moral problems. I don't want to say practical problems insofar as instrumental reasoning can be seen as solving practical problems by identifying actions as those solutions, yet "merely" instrumental reasoning doesn't cover moral reasoning through and through. I also hesitate to say normative problems, although the gist of the idea I have in mind is along the lines of Christine Korsgaard's cartographic disanalogy between concept-application, as theoretical reasoning, and practical reasoning (c.f. Onora O'Neill, and even H. A. Prichard, on why moral judgment might not best be construed as applying moral concepts to already-given particulars so much as constructing those concepts "over" particulars). At any rate, allow the possibility of a Moore/Ross faculty of deontic intuition, not as justifying our beliefs directly in specific moral answers so much as in justifying our gloss of specific practical problems as the moral ones.

Then list the following relations between actions and such problems:

  1. An action that solves the problem.
  2. An action that doesn't solve the problem, but doesn't make it harder to solve either, but which is not irrelevant to the problem (it is, let us suppose, such as merely sustains the problem on some current difficulty level).
  3. An action that aggravates the problem (makes it harder or perhaps even impossible to solve).
  4. An action that has no relevance to the problem.

We will define a right action as one that just is a case of (1), a wrong action as a case of (3), a differently permitted action as a case of (2), and an indifferently permitted action as a case of (4).

The occurrent problem. There does not seem to be room, in this scheme, for talk of "doing the right thing in the wrong way." Yet Kant opened the door to such talk with his doctrine of the motive of duty: albeit in the Groundwork/second Critique, he does not avow that fulfilling a duty from a lesser motive is a bad thing, yet by the time he goes over the levels of radical evil in Religion within the Limits of Reason Alone, he seems to countenance such talk rather clearly. This is bad for my thesis, seeing as my thesis is so well-inspired, here, by the neo-Kantian corpus (at least modulo Korsgaard, O'Neill, and even Susan Neiman).

The Aristotelian clause. What hath Athens to do with Königsberg? I'm having an intuition that phronetic logic (not so-called) in Aristotle's philosophy, might be caught up with an erotetic definition of deontic operations as per (1) through (4), but so I wonder how to situate the possibility or impossibility of "doing the right thing in the wrong way" vs. Aristotle's ethical theories. There is a similarity between the idea of a "golden mean" and the possibility in question; but there is a dissimilarity, too, in that again, it is hard to credit talk of "too much of a good thing" when the kind of goodness at issue is the goodness of right actions, supposing these to have some rightness that is not just a function from how they support or produce more static or objectual value. Perhaps the erotetic construction, in action-theoretic space, of the deontic operations, is out of line with both Kant and Aristotle. Does that tell against the construction, or against these elder luminaries? If the latter, nevertheless, I am sensitive enough to the parameters of John Rawls' method of reflective equilibrium to desire an explanation as to why Aristotle and Kant got this question wrong.

A Kantian way out? A biting-the-bullet idea that comes to my mind is: Kant could be comfortable with (1) through (4), and have it, then, that no action sufficiently satisfies (1) unless it comports directly with the motive of duty. That is, to actually, completely (we must not say "properly") solve a moral problem, one must perform the relevant action from the motive of duty as well. This motive, as itself enacted, is part of the greater action-theoretic solution-context. Such a Kantian "way out" is tempting to me, but not compelling enough quite yet, because not really clear enough to me, yet, either.

  • 1
    Famous psychologist Alfred Adler of the Viennese school of psychotherapy proposed people striving to do right/good things ultimately originates from the superiority complex while striving to avoid the wrong way ultimately originates from the inferiority complex. And some psychologists such as Adler believe both complexes exist in a single individual and forms certain dynamics, so under this POV it's possible to do the right thing in the wrong way if inferiority complex happens to be not working during the functioning of the superiority complex... Jul 24, 2022 at 22:48
  • @DoubleKnot It does seem to be one of those "folk moral theory" propositions, as even Aristotle and Kant buy into it for their own reasons. But Rawls suggests that, when we jettison elements of folk moral theory (or any theory we are predisposed to prefer), we are better off explaining why the prior is being "abandoned" rather than just fanatically proclaim our own new/exotic standpoint. Perhaps tracing the at-issue talk to malformed psychological dynamics would be a step in said direction? Jul 25, 2022 at 1:07
  • Indeed it's possibly the right direction to trace said psychological dynamics malformation and that usually leads to radical eliminativism claiming there's no such dynamics ontologically and perhaps neither epistemically w.r.t right/wrong as hoped by deontic logics. However, your material conditional 'if one does good then one does it in wrong way' is still a true claim due to vacuous truth under Russell's definite description analysis of the non-referring "good"... Jul 25, 2022 at 2:14
  • My interpretation of the categories of "rightness" and "goodness" is as encoding specifically ordinal degrees, or cardinal amounts, of a "thin" normative/deontic function's outputs. So the right/good priority problem mentioned by Rawls corresponds to the question, "Are all cardinals well-ordered?" and hence, albeit not absolutely, to the question of the choice axiom (or axiom metascheme, as encompasses countable, dependent, etc. versions of choice). Jul 25, 2022 at 3:09
  • 1
    Do you mean "achieving a good end using evil means"?
    – Geremia
    Aug 13, 2022 at 21:06

2 Answers 2


Yes, there is and T S Eliott stated it The last temptation is the greatest treason:

To do the right deed for the wrong reason.

It involves prudence in what you do and charity in your attitude in doing it.

This is all illustrated well by Slavery and its 3 positions

  1. Douglas: Whatever people want is okay with me. if they don't want slavery that is okay. Now that is surely a bad motive behind a good result

  2. Abolitionists: I will have no compromise with slavery a la John Brown (Note this is known to have increased slavery)

  3. Slavery is inherently wrong but it must be righted in the right way and the minds of people must be changed. That is Lincoln

  • Maybe the best thing would have been to pass a law to make slavery completely legal for anyone, on anyone, at any time. It would have taken about 1 day for the correct solution to be enacted after that. "Give the people what (they say) they want!" But, that would have been... wrong.
    – Scott Rowe
    Apr 2, 2023 at 21:26

Your list of four possible relations between actions and problems assumes a consequentialist reasoning- you are categorising the actions purely according to the nature of their outcome. On that basis, achieving the right outcome might be morally justified. Kant, surely, would not accept that, but instead would adopt a deontological view, considering motivation. Suppose a person is about to be killed by a passing car, and I snatch them out of the path of the vehicle just in time. That is a right action according to your scheme. But suppose it emerged a week later that I was a hitman, and I saved the person only because they were leading me to two other people whom I killed. For some, that would rather take the shine off my action of saving the first person from the car- especially if I decided later to shoot the first person too.

  • Consequentialism is not a well-defined term (as John Rawls notes in A Theory of Justice, ignoring consequences is crazy, but Rawls is canonically not a consequentialist). The deeper distinction is between theories that prioritize the concept of objects-having-value over the concept of actions-being-right, or vice versa, and so far the problem-solving picture is neutral between those orders-of-priority. And as I said, incorporating the motivation factor into the equation might offer a "Kantian way out." Mar 3, 2023 at 6:57
  • I should also add that Korsgaard and O'Neill, from whom I get the problem-solving picture (Susan Neiman too IIRC), which is corroborated in Kant (in the Doctrine of Virtue) when he talks about the "erotetic" character of the analysis there, are not canonically consequentialist, and Korsgaard, I think, pursues the avoidance of conflating action with production at some length. Worse, another analyst, Allen Wood, thinks Kant was a teleological (good-before-right) ethicist after all; at any rate, there is a more often now a better term, "aggregationism," for "consequentialism." Mar 3, 2023 at 7:38
  • 1
    I'm not sure why you need a 'way out'. Actions can have unintended and unpredictable consequences, so it will always be possible for an action classed as good to have a later bad consequence and vice versa. Mar 3, 2023 at 8:59
  • 1
    Also, your four point classification assumes that a problem can be unambiguously 'solved'. Many types of moral problem do not have a binary solved-unsolved outcome. Often they involve trade-offs between competing considerations, with a wide range of potential factors to be weighed and balanced. Mar 3, 2023 at 9:01
  • Aren't unsolved/unsolvable problems covered by the merely-permitted function? Like, if an action neither solves nor worsens a problem, it's neutral, so it's either differently or indifferently permitted (the SEP article on deontic logic covers this sort of obscure distinction). Mar 3, 2023 at 15:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .