Link Search Menu Expand Document

Against Reflective Equilibrium

Does the conclusion of the loose reconstruction of Greene (2014)’s argument provide grounds to reject Rawls’ method of reflective equilibrium (Rawls, 1999)? This section introduces an argument for the claim that reflective equilibrium will reliably generate incorrect conclusions. (This section also presents a generalisation of the loose reconstruction: it now establishes a conclusion about not-justified-inferentially premises not only concerning particular moral scenarios but also concerning debatable moral principles.)

This recording is also available on stream (no ads; search enabled). Or you can view just the slides (no audio or video). You should not watch the recording this year, it’s all happening live (advice).

If the video isn’t working you could also watch it on youtube. Or you can view just the slides (no audio or video). You should not watch the recording this year, it’s all happening live (advice).

If the slides are not working, or you prefer them full screen, please try this link.

The recording is available on stream and youtube.


Extending the Loose Reconstruction

The loose reconstruction of Greene (2014)’s argument (see Greene contra Ethics (Railgun Remix)) established a conclusion that is restricted to premises about particular moral scenarios. But it is possible to generalise the argument to a broader conclusion by elaborating on step 5. The result is this conclusion:

Not-justified-inferentially premises about particular moral scenarios, and debatable principles, cannot be used in ethical arguments where the aim is knowledge.

With this extension of the argument, we can use it to attempt to show that Rawls (1999)’s proposal about reflective equilibrium should be avoided. This is because it will reliably generate incorrect conclusions.

What Is Reflective Equilibrium?

One standard in ethics is Rawls’ reflective equilibrium idea:

‘one may think of moral theory at first [...] as the attempt to describe our moral capacity [...] what is required is a formulation of a set of principles which, when conjoined to our beliefs and knowledge of the circumstances, would lead us to make these judgments with their supporting reasons were we to apply these principles conscientiously and intelligently’ (Rawls, 1999, p. 41; see Singer (1974) for critical discussion).

Roughly, then, the idea is to start with not-justified-inferentially judgements you are, on reflection, inclined to make. And then to consider which principles might be consistent with these judgements. You may drop some of the judgements you start with depending on how well principles can be made to fit them.

Why Is Reflective Equilibrium Significant?

‘To most moral philosophers who reason about substantive moral issues, it seems that the method of reflective equilibrium, or a process very similar to it, is the best or most fruitful method of moral inquiry. Of the known methods of inquiry, it is the one that seems most likely to lead to justified moral beliefs.’ (McMahan, 2013, p. 111)

(Incidentally, McMahan (2013) is an excellent source for a concise overview of reflective equilibrium, its relations to intuition; there is also a very brief discussion of a challenge from moral psychology. For further evaluations of reflective equilibrium, see Scanlon (2002) and Knight (2023).)

Will Reflective Equilibrium Predictably Lead to Error?

Consider an argument:

  1. The not-justified-inferentially judgements you are inclined to make are an indirect consequence of fast processes (see What Is the Role of Fast Processes In Not-Justified-Inferentially Judgements?).

  2. Reflective Equilibrium is therefore, in effect, a method of identifying principles which characterise how fast processes operate and generalising them. (Roughly doing for ethics what Aristotelians did for physics.)[1]

  3. But the fast processes are fast because they trade away accuracy to gain speed. (All broadly inferential processes face trade-offs between speed and accuracy; see Preview: Ethics vs Physics.) Their function is to provide results that are accurate enough for mundane purposes in a limited but useful range of circumstances.

  4. We know, therefore, that the fast process will predictably be inaccurate in a range of cases. (Even though we cannot yet say much about which cases these are; see Cognitive Miracles: When Are Fast Processes Unreliable?).

  5. So while capable of producing valuable results within limits (much as broadly Aristotelian physics has plenty of applications), we know in advance that reflective equilibrium will reliably generate incorrect conclusions.[2]

Is this argument correct? As there are different varieties of reflective equilibrium (see Knight, 2023), it would be worth checking which, if any, kinds of reflective equilibrium this argument works against. This could lead to an objection against (2) in the above argument.

Another line of objection might be that the dual-process theory of ethical cognition is not well supported by evidence after all (see Evidence for Dual Process Theories). This could lead to an objection against (1) in the above argument.

A much bolder line of objection would be to argue that fast processes need not be inaccurate (see Railton (2014) for an attempt to develop this objection).[3] This could lead to an objection against (3) in the above argument.

Ask a Question

Your question will normally be answered in the question session of the next lecture.

More information about asking questions.


automatic : As we use the term, a process is automatic just if whether or not it occurs is to a significant extent independent of your current task, motivations and intentions. To say that mindreading is automatic is to say that it involves only automatic processes. The term `automatic' has been used in a variety of ways by other authors: see Moors (2014, p. 22) for a one-page overview, Moors & De Houwer (2006) for a detailed theoretical review, or Bargh (1992) for a classic and very readable introduction
cognitively efficient : A process is cognitively efficient to the degree that it does not consume working memory and other scarce cognitive resources.
dual-process theory : Any theory concerning abilities in a particular domain on which those abilities involve two or more processes which are distinct in this sense: the conditions which influence whether one mindreading process occurs differ from the conditions which influence whether another occurs.
fast : A fast process is one that is to to some interesting degree cognitively efficient (and therefore likely also some interesting degree automatic). These processes are also sometimes characterised as able to yield rapid responses.
Since automaticity and cognitive efficiency are matters of degree, it is only strictly correct to identify some processes as faster than others.
The fast-slow distinction has been variously characterised in ways that do not entirely overlap (even individual author have offered differing characterisations at different times; e.g. Kahneman, 2013; Morewedge & Kahneman, 2010; Kahneman & Klein, 2009; Kahneman, 2002): as its advocates stress, it is a rough-and-ready tool rather than an element in a rigorous theory.
not-justified-inferentially : A claim (or premise, or principle) is not-justified-inferentially if it is not justified in virtue of being inferred from some other claim (or premise, or principle).
Claims made on the basis of perception (That jumper is red, say) are typically not-justified-inferentially.
Why not just say ‘noninferentially justified’? Because that can be read as implying that the claim is justified, noninferentially. Whereas ‘not-justified-inferentially’ does not imply this. Any claim which is not justified at all is thereby not-justified-inferentially.
reflective equilibrium : A method that is supposed to provide justification for claims. The idea is to gather considered judgements about particular situations and attempt to identify principles which from which those judgements could be inferred, and then to adjust the judgements and principles so that they cohere. The canonical statement is Rawls (1999) (but Rawls, 1951 is a useful earlier statement). Authoritative secondary sources are Knight (2023) and Scanlon (2002).
unfamiliar problem : An unfamiliar problem (or situation) is one ‘with which we have inadequate evolutionary, cultural, or personal experience’ (Greene, 2014, p. 714).


Bargh, J. A. (1992). The Ecology of Automaticity: Toward Establishing the Conditions Needed to Produce Automatic Processing Effects. The American Journal of Psychology, 105(2), 181–199.
Greene, J. D. (2014). Beyond Point-and-Shoot Morality: Why Cognitive (Neuro)Science Matters for Ethics. Ethics, 124(4), 695–726.
Hogarth, R. M. (2010). Intuition: A Challenge for Psychological Research on Decision Making. Psychological Inquiry, 21(4), 338–353.
Kahneman, D. (2002). Maps of bounded rationality: A perspective on intuitive judgment and choice. In T. Frangsmyr (Ed.), Le prix nobel, ed. T. Frangsmyr, 416–499. (Vol. 8, pp. 351–401). Stockholm, Sweden: Nobel Foundation.
Kahneman, D. (2013). Thinking, fast and slow. New York: Farrar, Straus; Giroux.
Kahneman, D., & Klein, G. (2009). Conditions for intuitive expertise: A failure to disagree. American Psychologist, 64(6), 515–526.
Knight, C. (2023). Reflective Equilibrium. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (Winter 2023). Metaphysics Research Lab, Stanford University. Retrieved from
Kozhevnikov, M., & Hegarty, M. (2001). Impetus beliefs as default heuristics: Dissociation between explicit and implicit knowledge about motion. Psychonomic Bulletin & Review, 8(3), 439–453.
McMahan, J. (2013). Moral Intuition. In The Blackwell Guide to Ethical Theory (pp. 103–120). Wiley.
Moors, A. (2014). Examining the mapping problem in dual process models. In Dual process theories of the social mind (pp. 20–34). Guilford.
Moors, A., & De Houwer, J. (2006). Automaticity: A Theoretical and Conceptual Analysis. Psychological Bulletin, 132(2), 297–326.
Morewedge, C. K., & Kahneman, D. (2010). Associative processes in intuitive judgment. Trends in Cognitive Sciences, 14(10), 435–440.
Railton, P. (2014). The Affective Dog and Its Rational Tale: Intuition and Attunement. Ethics, 124(4), 813–859.
Rawls, J. (1951). Outline of a Decision Procedure for Ethics. The Philosophical Review, 60(2), 177–197.
Rawls, J. (1999). A Theory of Justice (Revised edition). Cambridge, Mass: Harvard University Press.
Scanlon, T. M. (2002). Rawls on justification. In S. Freeman (Ed.), The cambridge companion to rawls (pp. 139–167). Cambridge: Cambridge University Press.
Singer, P. (1974). Sidgwick and Reflective Equilibrium. The Monist, 58(3), 490–517.


  1. In the lecture, I offer a bit more detail on this point (in the form of a dilemma). ↩︎

  2. Unless, that is, it were limited to familiar situations. But this would be hard to do given that we are not in a position to know which situations are unfamiliar (see Cognitive Miracles: When Are Fast Processes Unreliable?). And it would also not be very useful. After all, we have little need for a theory covering only cases that our fast processes already provide us with expertise in dealing with. And if the idea of reflective equilibrium is just to identify principle implicit in responses due to fast processes, it should be subsumed into moral psychology rather than viewed as a method of doing ethics. ↩︎

  3. Note that Railton identifies intuition with the affective system (Railton, 2014, pp. 826–8]). Railton is surely correct that affect is part of a flexible and sophisticated learning system. But evidence that feelings and emotions play at most a limited role in moral judgement (see Moral Intuitions and Emotions: Evaluating the Evidence) indicates that Railton’s identification of intuition and the affective system may not work. ↩︎