Skip to main content
Log in

A hard look at moral perception

  • Published:
Philosophical Studies Aims and scope Submit manuscript

Abstract

This paper concerns what I take to be the primary epistemological motivation for defending moral perception. Offering a plausible account of how we gain moral knowledge is one of the central challenges of metaethics. It seems moral perception might help us meet this challenge. The possibility that we know about the instantiation of moral properties in something like the way we know that there is a bus passing in front of us raises the alluring prospect of subsuming moral epistemology under the (relatively) comfortable umbrella of perceptual or, more broadly, empirical knowledge. The good news on this front is that various combinations of metaethical positions and theories of perception arguably have the potential to vindicate moral perception (though I won’t do much to defend this claim here). The bad news, I’ll argue, is that moral perception would be dependent for its epistemic merit on background knowledge of bridge principles linking moral and non-moral properties. Thus, in order to defend a purely perceptual moral epistemology, one would have to argue that knowledge of those principles is likewise perceptual. I further argue it is not.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. As opposed to, for instance, phenomenological motivations. For a recent critique of moral perception along these lines, see Väyrynen (m.s.).

  2. I focus on perception of “thin” moral properties—good, bad, right, wrong, etc. It’s controversial whether perception of thicker moral properties counts as moral perception. See, e.g., Väyrynen (2013). Even if it does, one could use thick moral perception to develop a purely perceptual moral epistemology only if thin moral knowledge is grounded in thick moral knowledge. I highly doubt this, though I cannot argue against it here.

  3. In addition to other works discussed herein, Väyrynen (m.s.) mentions defenses of moral perception from Audi (2013), Blum (1994), Chappell (2008), Cowan (2013), Cullison (2010), Cuneo (2003), DePaul (1993), Greco (2000), McNaughton (1988), and Watkins and Jolley (2002).

  4. The list of low-level properties is borrowed from Siegel (2011). Siegel’s is a list of “non-kind” properties (as opposed to “kind” properties). This is the same distinction—or, at least, non-kind properties and low-level properties are the same, and kind properties are a subset of the high-level properties. The low-/high-level phrasing is borrowed from Werner (forthcoming).

  5. For grammatical simplicity, I will sometimes talk of properties as objects—e.g., say that Norm perceives some behavior, rather than perceiving that something instantiates the property of behaving in some particular way, or that Vera perceives a bus, rather than that she perceives that something instantiates the property of being a bus. This should not be taken as a leap to assuming that we can perceive objects independently of their properties (though I do not deny this, either).

  6. For defense—as well as arguments that this claim is compatible with views it is typically taken to be in tension with, such as naïve realism—see Siegel (2011, Chap. 2).

  7. What is the difference between perception of X and experience as of X? First, I take it perceptions of X need to have a certain kind of “cognitive basicness.” Presumably, both optimists and skeptics about mental state perception would grant that there can be experiences as of anger. Their disagreement would be over whether such experiences ground or are grounded in judgements about anger. Second, for the optimist, a perception of anger is at an intermediate “cognitive level” between an experience as of anger and an anger-judgement. This allows for the possibility of having an experience as of X without perceiving X—say, if I know I’m in a context where there are likely to be a lot of fake Xs. This bit about “cognitive level” is intentionally left vague. It serves only to make certain points more intuitive (I hope); nothing substantive is meant to hang on it.

  8. Which features? Answering this question is difficult. For instance, it can’t just be those features that explain the experience. For, intuitively, it seems that a quotidian experience of water is water-like, but not H2O-like. For the same reason, it can’t be those features that we associate with the object of experience—at least not given that some people know that water is H2O. I suspect the answer has something to do with the features that allow the experiences to meet criteria set by our concepts, but I won’t explore this further here.

  9. This argument mimics the phenomenal contrast arguments Siegel (2011) deploys to defend perception of various complex properties. Werner (forthcoming) deploys such an argument to defend moral perception. Werner’s argument comes up again in Sect. 2.3.

  10. I’m assuming here that Norm might judge Vera to be angry partly on the basis of her behavior. This is roughly in line with the “theory theory” about judgements concerning others’ mental states. In contrast, according to “simulation theory,” Norm might judge that Vera is angry after running through empathetic processing starting with the same input (the bus cutting them off). I set simulation theory aside because it doesn’t seem compatible with mental state perception, which I’m focusing on for illustrative purposes. (For one thing, Vera’s mental state plays no role in the simulation theoretic explanation of Norm’s judgement). For an overview of simulation theory as a reaction to the theory theory, see Gordon (2009).

  11. I take no position here on whether sub-personal processing can count as inferential.

  12. Or, at least, beliefs that themselves have some epistemic merit (e.g., are justified).

  13. Since I completed the paper, a number of people have suggested that this knowledge requirement is too strong. Perhaps Norm is justified in believing that Vera is angry simply because perception is a basic source of justification. And perhaps the explanation for the reliability of that perception is just the fact of some relationship between Vera’s facial expressions and her mental states. What I want to insist on, though, is that in order for Norm’s perception to ground knowledge, whatever explains the reliability of that perception must in some sense be available to him. If the explanans can be relevantly available to Norm without his believing it, then perhaps the knowledge requirement is too strong. But I am confident that shifting from a knowledge requirement to an availability requirement would make little difference to my arguments herein.

    Of course, some—e.g., externalists or coherentists—might reject even the availability requirement. But then Norm’s is just the sort of case that is frequently used to challenge externalism: It seems implausible that an agent can know something when the reliability of his route to that knowledge is deeply mysterious (at least to him). See, e.g., Bonjour’s (1980) case of Norman the Clairvoyant. As for coherentism: If Norm is sensible, he will recognize the need to explain the reliability of his perception. It seems unlikely that his perception-based belief that Vera is angry would cohere with his belief that the reliability of his perception of her anger is deeply mysterious. For application of externalism and coherentism to moral knowledge see, e.g., Shafer-Landau (2005) and Sayre-McCord (1996), respectively.

  14. I first encountered this phrase in Siegel (2011), though I owe my use of it Werner (forthcoming), who uses it to make points similar to those made here about why moral perception alone can’t vindicate a purely perceptual moral epistemology.

  15. Since I’m only interested in epistemically successful cases, I’ll continue to refer to background knowledge, rather than beliefs, though arguably one can have perceptions that involve cognitive penetration even by false beliefs.

  16. This is a case of what we might call augmenting penetration, where background knowledge leads an experience as of Y to ground perception of X. This can be contrasted with what we might call undercutting penetration, where background knowledge prevents an experience as of Y that otherwise would ground perception of X from doing so. For instance, if Norm learned Vera was an android, he might cease seeing her as angry. Except where noted, discussion herein is limited to augmenting perception.

  17. I trust my decision not to have my paper riddled with talk of penetration is self-explanatory. Note that it may also be possible for perceptions to be mediated by non-cognitive attitudes. This might be relevant for those who take certain affective states to themselves be perceptions of moral properties. See, e.g., Oddie (2009) and McBrayer (2010a, b). My arguments apply as well to affective states as to cognitive ones. If Norm’s perception of Vera’s anger is mediated by an affective state, rather than a cognitive one, we would still require a story about how that affective state epistemically vindicates the grounding relation between Norm’s experience as of a scowl and his perception of anger: Norm would still need to “know,” in some sense, that scowls implicate anger.

  18. This is not to say that the perception and the experience must actually be separate. It might be that certain aspects of an overall experiential state with one phenomenal character are grounded in other aspects of that same overall experiential state with a different phenomenal character. I am also not assuming that there is only one unique phenomenal content that counts as being as of permissibility—only that experiences as of permissibility and experiences as of the relevant base properties have different phenomenal contents. More on this in Sect. 2.2.

  19. The exchange begins with Harman (1977). Sturgeon responds in his (1986). This went back-and-forth a bit; next is Harman (1986).

  20. This assumes that the base properties and moral properties aren’t, in fact, the same properties. I return to this issue in Sect. 2.2.

  21. David Slutsky (2001) argues that the test can be improved, by considering cases where the relevant non-moral base properties are causally inefficacious. There are a number of parallels between his argument and my own.

  22. Though again see Slutsky (2001).

  23. One incredible alternative would be that Sam judges the cat-torture to be fake on the basis of failure to perceive wrongness!

  24. I’m simplifying slightly. It is controversial whether the experience caused by the fake cat-torture has the same phenomenal content as that caused by the real. However, given that the fake cat-torture is convincing, we may suppose the contents of the experiences caused by the real and fake cat-tortures are in principle indistinguishable to Sam. The counterfactual test thus suggests that Sam’s moral perception tracks experiences that are indistinguishable from experiences as of cat-torture better than they track the presence of wrongness. The best explanation for this seems to be that his falsidical moral perception is grounded in a non-moral experience that is indistinguishable from an experience as of cat-torture. And the best explanation for this seems to be that his veridical moral perception is grounded in his experience as of cat-torture. For simplicity’s sake, I will continue to speak as though the two have the same phenomenal contents. For those who reject this, “experience as of cat-torture” can be taken to refer both to cat-torture-like experiences and those that are indistinguishable from cat-torture-like experiences. This makes no substantive difference to my arguments.

  25. The idea that this concerns every case is crucial. The point of the counterfactual test is not that if I can falsely perceive X, then perceptions of X are grounded in experiences as of some Y. I accept that there can be false unmediated perceptions, perhaps even of low-level properties.

  26. Though see note 13, above.

  27. This is, of course, a reference to Putnam (1975).

  28. I’m not asserting that this is necessarily the case. I think it may turn out that Vera’s perception here is mediated by knowledge of a relation between certain low-level properties and the presence of busses. For it may be possible to have those low-level experiences but not see them as bus-like. Consider the relation between perception of a duck (or a rabbit) and experiences as of the low-level features of the duck-rabbit. Perhaps all high-level perceptions are like this, and so are all mediated. If that were the case, it would be even easier to reach my skeptical conclusion, so I make the optimist’s position stronger by assuming Vera’s perception might be as suggested here. The general point is that in cases where perceptions are unmediated—and surely, some are—one of two things must be happening: (1) There is some Y such that perceptions of X track experiences as of Y even in the absence of X, but experiences as of X and as of Y have the same phenomenal content (or, at least, indistinguishable—see note 24, above); or (2) There is no such Y, because illusory experiences of X are impossible. (This might be the case if certain mental states are luminous—if we can’t be mistaken which we are in. For an argument against this, see Williamson (2002)).

  29. We need to change Mediation if we are to avoid the conclusion that a perception of X grounded in an experience as of X must be mediated by knowledge of a relation between X and itself. It’s also worth noting that if experiences with the same phenomenal content are identical, only Mediation needs to be amended, for if X-like and Y-like experiences are identical, it is trivial that a perception grounded in one is grounded in the other. Framing things as I do is meant to make it easier to raise certain objections to my arguments.

  30. This is not to say, of course, that there are no puzzles about how experiences as of X generate knowledge about X, given the possibility of illusions, only that these puzzles would pose no special problem for moral perception.

  31. Barring this sort of monism, the situation will be more complicated. For example, suppose causing suffering is wrong, but many other things are, too. In that case, there will be, at best, a partial overlap between experiences as of suffering and experiences as of wrongness. This won’t change the basic issues, though, just make them more complicated.

  32. The inner quotation concerning what it is to lack “affective empathy” comes from Blair (2007, p. 4).

  33. The claim that EEDIs have a theory of mind serves to block the alternative that only Sam perceives certain relevant non-moral features of the cat-torture, such as the cat’s suffering.

  34. Another way to push back here stems from the idea of undercutting penetration (i.e., mediation). See note 16, above. The idea would be that experiences as of suffering and experiences as of wrongness have the same phenomenal content, but certain background beliefs can penetrate the experience to undercut perception of one while retaining perception of the other. This seems possible, but I doubt it can be what’s going on in these cases. Think back to Vera. Suppose she learns that there are a lot of motorcyclists wearing bus-façades on her current route. When she gets cut off, she might fail to perceive a bus. But surely her experience would still be relevantly bus-like. If we asked her if she saw a bus, she’d presumably reply that she seemed to, but knows it very likely wasn’t a bus. In contrast, it seems that Pathos and early Alice don’t even have the relevant moral seemings.

  35. E.g., Moore’s (1903) open question argument, (e.g., Hare’s 1952) missionaries and cannibals, and (e.g., Horgan and Timmons’ 1991) Moral Twin Earth.

  36. I’m inclined to think this just is a Moorean intuition. Can one accept both Moorean intuitions and that moral and non-moral experiences can have the same content? Sort of. Suppose Carla gained her concept ‘wrong’ by ostension. At a young age, when confronted with certain things, she was told: That is wrong. She might thus claim that when confronted with suffering today, she has an experience as of wrongness. Nevertheless, she might still accept the Moorean intuition that, conceptually, being suffering isn’t actually sufficient for being wrong. I’m not sure how psychologically plausible this is. But even if it is, it provides a rather hollow victory for the optimist about unmediated moral perception. For in order to develop a purely perceptual moral epistemology, we need to account for the knowledge of whoever pointed those wrong things out to her. Suppose we trace this back to the origin of the concept. Moorean intuitions seem to block the idea that this origin could itself be ostensive. This is importantly different from certain non-moral cases. For instance, not only might Vera have gained the concept ‘bus’ via ostension; it seems the origin of that concept might have been ostensive, too.

  37. Drawing this conclusion about property identity might require further, non-empirical philosophical knowledge about the nature of property identity, but Cliff could make do with a weaker relation, such as consistent correlation.

  38. One might wish to resist this conclusion. Perhaps Cliff2’s abductive inference relies on something a priori, for instance. Since this would only serve to weaken the case for purely perceptual knowledge of moral bridge principles, I set it aside.

  39. Compare Zangwill (2006).

References

  • Audi, R. (2013). Moral perception. Princeton: Princeton University Press.

    Book  Google Scholar 

  • Blair, R. J. R. (2007). Empathic dysfunction in psychopathic individuals. Empathy in mental illness (pp. 3–16). Cambridge: Cambridge University Press.

    Google Scholar 

  • Blum, L. A. (1994). Moral perception and particularity. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Bonjour, L. (1980). Externalist theories of empirical knowledge. Midwest Studies in Philosophy, 5(1), 53–74.

    Article  Google Scholar 

  • Chappell, T. (2008). Moral perception. Philosophy, 83(4), 421–437.

    Article  Google Scholar 

  • Cowan, R. (2013). Perceptual intuitionism. Philosophy and Phenomenological Research. doi:10.1111/phpr.12023.

  • Cullison, A. (2010). Moral perception. European Journal of Philosophy, 18(2), 159–175. doi:10.1111/j.1468-0378.2009.00343.x.

    Article  Google Scholar 

  • Cuneo, T. (2003). Reidian moral perception. Canadian Journal of Philosophy, 33(2), 229–258.

    Google Scholar 

  • DePaul, M. R. (1993). Balance and refinement: beyond coherence methods of moral inquiry. Colchester: Taylor & Francis.

    Google Scholar 

  • Gordon, R. M. (2009, Fall). Folk psychology as mental simulation. In E.N. Zalta (Ed.), The Stanford encyclopedia of philosophy. http://plato.stanford.edu/archives/fall2009/entries/folkpsych-simulation/.

  • Greco, J. (2000). Putting skeptics in their place: The nature of skeptical arguments and their role in philosophical inquiry. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Hare, R. M. (1952). The language of morals. Oxford: Clarendon.

    Google Scholar 

  • Harman, G. (1977). The nature of morality: An introduction to ethics. Cary, NC: Oxford University Press.

    Google Scholar 

  • Harman, G. (1986). Moral explanations of natural facts—Can moral claims be tested against moral reality? The Southern Journal of Philosophy, 24(S1), 57–68.

    Article  Google Scholar 

  • Horgan, T., & Timmons, M. (1991). New wave moral realism meets moral twin earth. Journal of Philosophical Research, 16, 447–465.

    Article  Google Scholar 

  • McBrayer, J. P. (2010a). A limited defense of moral perception. Philosophical Studies, 149(3), 305–320.

    Article  Google Scholar 

  • McBrayer, J. P. (2010b). Moral perception and the causal objection. Ratio, 23(3), 291–307.

    Article  Google Scholar 

  • McGrath, S. (2004). Moral knowledge by perception. Philosophical Perspectives, 18(1), 209–228.

    Article  Google Scholar 

  • McNaughton, D. (1988). Moral vision: An introduction to ethics. Oxford: Wiley-Blackwell.

    Google Scholar 

  • Moore, G. E. (1903). Principia ethica. Oxford: Oxford University Press.

    Google Scholar 

  • Oddie, G. (2009). Value, reality, and desire. Oxford: Oxford University Press.

    Google Scholar 

  • Putnam, H. (1975). The meaning of ‘meaning’. Minnesota Studies in the Philosophy of Science, 7, 131–193.

    Google Scholar 

  • Sayre-McCord, G. (1996). Coherentist epistemology and moral theory. In W. Sinnott-Armstrong & M. Timmons (Eds.), Moral knowledge? New readings in moral epistemology. Oxford: Oxford University Press.

    Google Scholar 

  • Shafer-Landau, R. (2005). Moral realism: A defence. Oxford: Oxford University Press.

    Google Scholar 

  • Siegel, S. (2011). The contents of visual experience. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Slutsky, D. (2001). Causally inefficacious moral properties. Southern Journal of Philosophy, 39(4), 595–610.

    Article  Google Scholar 

  • Sturgeon, N. (1986). Harman on moral explanations of natural facts. Southern Journal of Philosophy, 24(S1), 69–78.

    Article  Google Scholar 

  • Väyrynen, P. (2013). The lewd, the rude and the nasty: A study of thick concepts in ethics. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Watkins, M., & Jolley, K. D. (2002). Pollyanna realism: Moral perception and moral properties. Australasian Journal of Philosophy, 80(1), 75–85.

    Article  Google Scholar 

  • Werner, P. (forthcoming). Moral perception and the contents of experience. Journal of Moral Philosophy.

  • Williamson, T. (2002). Knowledge and its limits. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Väyrynen, P. unpublished manuscript. Doubts about moral perception. http://www.personal.leeds.ac.uk/~phlpv/papers/moralperception.pdf.

  • Zangwill, N. (2006). Moral epistemology and the because constraint. In J. Dreier (Ed.), Contemporary debates in moral theory (pp. 263–281). Oxford: Wiley-Blackwell.

    Google Scholar 

Download references

Acknowledgments

My thanks to Christian Coons for turning me on to this use of the counterfactual test, to Kelly Trogdon for introducing me to the perception literature that helped me frame it, to Tristram McPherson for raising numerous objections to it, and to all three for subsequent invaluable discussions. Thanks also to Molly Gardner, Graham Oddie, Geoff Sayre-McCord, Walter Sinnott-Armstrong, Jacob Sparks, Christina Van Dyke, Mark van Roojen, Pekka Väyrynen, Preston Werner and participants at the 2013 Rocky Mountain Ethics Congress and the 2014 PPE Workshop at UNC Chapel Hill.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to David Faraci.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Faraci, D. A hard look at moral perception. Philos Stud 172, 2055–2072 (2015). https://doi.org/10.1007/s11098-014-0397-6

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11098-014-0397-6

Keywords

Navigation