Critically Assess the View that Moral Properties are Best Understood as like Secondary Qualities such as Colours and Tastes. 

The analogy with secondary qualities is ultimately not the best means of understanding moral properties, because secondary qualities are not themselves very well understood, and because the analogy between moral properties and secondary qualities is interpreted in different ways.  The moral anti-realist employs the analogy to explain how subjective moral qualities gain an apparent objectivity.  The moral realist employs the analogy to explain how objective moral properties gain that essential element of motivational subjectiveness.  Hence how the analogy will be interpreted depends on one's preconceptions about moral properties.  It is not, therefore, the best avenue to an understanding of those moral properties.

In this essay, I will examine the distinction between primary and secondary qualities and then explore the deployments of the analogy by anti-realists and realists.  I will thereby demonstrate that an understanding of moral properties must precede an interpretation of the analogy, rather than vice versa.

Primary versus Secondary Qualities

The distinction between primary and secondary qualities originated with Locke.  According to Locke, shape, extension, solidity, motion, and number are primary qualities.  While colour, odour, taste, texture, and sound are secondary qualities.  Locke reasoned that primary qualities are an essential part of the object of which they are predicated -- hence they are considered objective.  Secondary qualities, by comparison, have an essential element of subjectiveness about them.

"The ideas of primary qualities of bodies are resemblances of them, and their patterns do really exist in the bodies themselves, but the ideas produced in us by the secondary qualities have no resemblance of them at all. There is nothing like our ideas [of secondary qualities] existing in the bodies themselves."
                                                                                                (Locke, 1969, Pg 69)

In a different possible world, where nothing like us existed, these secondary qualities would not exist, and could not be predicated of anything in that world (Kripkean trans-world rigidity aside).

"these tastes, odours, colours, etc., so far as their objective existence is concerned, are nothing but mere names for something which resides exclusively in our sensitive body, so that if the perceiving creature were removed, all of those qualities would be annihilated and abolished from existence.
                                                                        (Galileo, 1623. Pg 28).

But Locke's approach conflates two senses of meaning for the quality predicates.  There are two senses under which we can understand something to be "red".  There is the perceptual sense in which something appears "red" to us -- let's call that sense "red-s" (for subjective).  The property of being "red" can be conceived as applying only to those things which we perceive as "red-s" (given, of course "normal" circumstances, and a "normal" human visual apparatus).  But because this sense of being "red" is an entirely subjective experience, there is no assurance that the way in which you see things "red-s-ly" is the way in which I see things "red-s-ly".  Learning the meaning of the word "red" is just learning to call the same things "red".  But we cannot have any more than an empirical assurance that the set of things that I would call "red" coincides with the set of things that you would call "red".

On the other hand, there are some physical characteristics of surfaces (or translucent volumes) that are such that under "normal" viewing conditions, and in the presence of a "normal" visual apparatus, they will be seen to be "red-s" by an arbitrary observer.  Let's call surfaces (or volumes) that meet that physical description "red-o" (for objective).  Now we can conceive of the property of being "red" as applying only to those things that are "red-o" - regardless of how or even whether they might appear to us.  "Red-o" is thus a Lockean primary quality because it can be defined independently of any relation to the way we are.  When Locke would say that the "red" that I perceive is in me and not in the thing I am viewing, I can suggest that the "red-s" that I perceive is just the way that you would expect that a bio-chemical sensing device like the ones with which I am equipped would see a "red-o" surface/volume -- I don't have a "subjective colour sense", I have an "objective surface/volume properties detection sense".

This separation of secondary qualities into an "-s" and an "-o" sense can be done for any of the predicables that are collected under the "secondary quality" label.  It is also possible to take each of the primary qualities, and come up with the same two senses.  Think of the primary quality of motion and sitting on a train in a station.  Which train is moving -- the one you are on or the one next to you?  So there is "motion-s" that applies to the train that appears to be moving.  And there is "motion-o" that applies to the train that is moving in the sense independently of us.

The same approach can be taken with any of the moral properties.  We can, for example, understand "courage-s" as the value that we "see" (or judge or appreciate) in actions, circumstances, and people.  And we can understand "courage-o" as the value that exits in those things independently of how we might "see" them.  Of course, some philosophers will argue that there is no such thing as "moral-property-o" -- but I will side-step that argument for the moment.  I should also note that I do not intend to rule out the possibility of such a predicate being relational.  It is possible, for example, that the "good-o" of something might be its moral value to someone for some purpose.

The Anti-Realist Analogy

Many philosophers think that moral properties are inherently motivational -- that it is part of our concept of moral properties that they be motivational to some degree.  On this view, judging that "X is Good" is at least a prima facie reason for me to act on that judgment.  Our motivations are captured in our desires.  We can capture this link between values and desires in a bi-conditional --

(M) Something is good if and only if we desire it (in the proper conditions).(1)

Values and desires are "made for each other" as Wiggins puts it.

"Philosophy has dwelt nearly exclusively on differences between 'good' and 'red' or 'yellow'. I have long marvelled at this. For there resides in the combined objectivity and anthropocentricity of colour a striking analogy to illuminate not only the externality that human beings attribute to the properties by whose ascriptions they evaluate things, people, and actions, but also the way in which the quality by which the thing qualifies as good, and the desire for the thing are equals—are, 'made for one another' so to speak."
                                                                                    (Wiggins, 1998, Pg 107)

The anti-realist regards moral properties as inherently subjective -- relegating the objective phenomenology of moral language to a consequential role.  The parallel with colours that the anti-realist relies upon is that colours are understood to be inherently visible.  As Strawson puts it "colours are visibilia or they are nothing".(2)  This thought can be captured in another bi-conditional --

(C) Something is red if and only if it looks red to us (in the proper conditions).(1)

Neither bi-conditional is taken as empirical.  Both are presented as a priori understandings of our concepts of value and colour. 

Yet, despite these subjective understandings of colour and value, we employ the concepts as properties of the world.  At the pre-philosophical level, we believe that (and behave as if) colours are objective properties - out there in the world.  As Lewis says -

"[deny these facts] and the most credible explanation of your denial is that you are in the grip of some philosophical (or scientific) error"(3)

At the same pre-philosophical level, we treat moral properties the same way. As Mackie puts it --

"It [the apparent objectivity of values] has also a firm basis in ordinary thought, and even in the meanings of moral terms."(4)

The anti-realist employs the analogy between moral properties and colours to explain this apparent objectivity.  For colours, the primary is the subjective experience of something looking red, the consequent is that the thing is red.  By analogy, then, the primary is the desire, the consequent is that it is good.  The analogy with secondary qualities argues that moral properties are not properties independent of our sensibilities.  Moral properties are what they are because of our sensibilities.  In another possible world, where nothing like us exists, there would be no moral properties.  The subjective sense of the predicate determines the scope of reference of the objective sense of the predicate. 

Of course, what the anti-realist is ignoring is the temporal direction of the causal link between a thing being "red-o" and it being "red-s".  (Assuming, of course, that they admit that there is such a thing as causation at work here.)  According to the physics of the situation, we perceive something as "red" only because of the interaction of photons with the surfaces/volumes of the thing and the bio-chemistry of our sensory apparatus.  Physics argues that it is "red-s" because it is "red-o" - not the other way around.  Hence, we can argue that something exhibits a "moral-property-s" (or is desirable) only because of the sociological, psychological, biological, bio-chemical, physical facts about the thing -- its having a "moral-property-o" (or is good).  The anti-realist interpretation of the analogy thus works only under the prior supposition of an anti-realist metaphysics  that denies there is a "red-o" or an element of causation linking "red-o" to "red-s". 

The Realist Analogy

The realist employment of the analogy, by comparison, starts with the acknowledgement that the subjectivism of the anti-realist is contrary to how we commonly think about morality.  The anti-realist maintains that the objective phenomenology of our normal way of thinking about morality is grossly in error.  The realist denies this.

I cannot see how to refute the arguments for the subjectivity of ethical values, but I find myself incapable of believing that all that is wrong with wanton cruelty is that I don't like it
                                                                                    (Russell, 1960, Pg: 146-7).

The anti-realist assumes that moral motivation is internal and not part of the objective world.  Hence the anti-realist assumes that the objective phenomenology of thinking about things moral must be mistaken.  The analogy of the projection of color-sensations is used to explain the error.

The realist, by comparison, maintains that motivation is external rather than internal.  Hence, to the realist, there is no error to explain.  Moral properties are objective - "out there" (albeit perhaps as relational entities).  What the realist uses the analogy to explain is how objective moral properties acquire their essential element of motivational subjectiveness.

On a causal understanding of colour, a "red-o" object causes "red-s" subjective experiences (under the usual "normal" conditions).  Colours are objective properties because the surface/volume properties of things are out there in the environment to be discovered.  Things that are "red-o" affect the human-type perceivers by generating "red-s" perceptions in them.  As McDowell says --

"Secondary quality experience presents itself as perceptual awareness of properties genuinely possessed by the objects that confront one. And there is no general obstacle to taking that appearance at face value.... [A]n experience of something as red can count as a case of being presented with a property that is there anyway, there independently of the experience itself.
                                                                                    (McDowell, 1998, Pg 202)

The realist interpretation of the analogy applies the causal account of colours to moral properties.  They are discernability-involving, yet objectively "out there" in the environment.  Attitudes and motivations are the causal consequences of the objective moral properties of things.  Contrary to Mackie's argument from queerness, moral values are no more "queer" than any other discernable property. 

The realist's understanding of the analogy has the added bonus of highlighting an additional difficulty for the anti-realist's understanding of the "moral-property-s" to "moral-property-o" linkage.  The anti-realist maintains that something has a "moral-property-s" that we "see" (or judge or appreciate).  And that it is the existence of the "moral-property-s" that dictates the content of the concept of the "moral-property-o" (just the way that the "red-s" subjective experience dictates what we accept as "red-o").  But what the anti-realist does not address is just how it is that we "see" the "moral-property-s".  Intuitionists aside, we do not have a "moral sense".  So the only way to conceive the "see" in question here, is that we perceive (in the usual way) some non-moral properties of the thing that meets some (humanly created) criteria for accepting this thing as having the appropriate moral properties.  But this is just the realist's casual understanding of what happens.  So where is the anti-realist left to retreat to?


As you can see from the foregoing discussion, just how one chooses to understand the primary-secondary quality distinction, and just what prior suppositions about moral properties are brought to the table, determines how one interprets the analogy between secondary qualities and moral properties.  Given the extent of prior suppositions on both scores, the analogy cannot be the best avenue to an understanding of moral properties.


Notes & References

(1) Benbaji, Hagit; "'Red' and 'Good'" in Occasional Papers of the Institute for Advanced Study, School of Social Science, Paper Number 30, Oct 2007. URL=<>

(2) Strawson, Peter; "Reply to Evans" in Philosophical Subjects: Essays Presented to P.F.Strawson, Z. Van Straatten (ed.), Oxford University Press, New York, New York, 1998. Pg 87-137.

(3) Lewis, David; "Naming the Colours' in Australasian Journal of Philosophy, Vol 75, pg 325.

(4) Mackie, J.L.; Ethics: Inventing Right and Wrong, Penguin Books, London, England. 1990. ISBN 0-14-013558-8. Pg 31.


Bennett, Jonathan; "Substance, Reality, and Primary Qualites" in American Philosophical Quarterly, Vol 2, No 1 (Jan 1965), pp 1-17. URL=<>

Boyd, Richard N.; "How to be a Moral Realist" in Essays on Moral Realism, Geoffrey Sayre-McCord (ed.), Cornell University Press, Ithica, New York, 1988. ISBN 0-8014-9541-5.

Cuneo, Terence; "Are Moral Qualities Response-dependent?" in NOUS, Vol 35, No 4 (2001), Pg 569-591

Dancy, Jonathan; "On Moral Properties" in Mind, New Series, Vol 90, No 359 (Jul, 1981), Pp 367-385, URL=<>

Galileo Galilei; "The Assayer (1623)" in Discoveries and Opinions of Galileo, as translated by Stillman Drake, Knopf Doubleday Publishing Group, New York, New York. 1957. ISBN 0385092393.

Hume, David; A Treatise of Human Nature, Selby-Bigge (ed.), Oxford, England. 1967.

Locke, John; An Essay concerning Human Understanding, Pringle-Pattison, ed., Oxford 1969

McDowell, John; "Values and Secondary Qualities" in Mind, Value & Reality, Harvard University Press, Cambridge, Mass., 1998. ISBN 0-674-00713-1.

Millgram, Elijah; "Moral Values and Secondary Qualities" in American Philosophical Quarterly, Vol 36, No 3 (Jul, 1999), URL=<>

Moore, Michael S.; "Moral Reality Revisited" in Michigan Law Review, Vol 90, No 8 (Aug 1992), pp 2424-2533. URL=<>.

Platts, Mark; "Moral Reality" in Essays on Moral Realism, Geoffrey Sayre-McCord (ed.), Cornell University Press, Ithica, New York, 1988. ISBN 0-8014-9541-5.

Russell, Bertrand; "Notes on Philosophy" in Philosophy, Vol 75 (Jan 1960), Pg 146-147.

Sayre-McCord, Geoffrey; "Moral Theory and Explanatory Impotence" in Essays on Moral Realism, Geoffrey Sayre-McCord (ed.), Cornell University Press, Ithica, New York, 1988. ISBN 0-8014-9541-5.

Sturgeon, Nicholas L.; "Moral Explanations" in Essays on Moral Realism, Geoffrey Sayre-McCord (ed.), Cornell University Press, Ithica, New York, 1988. ISBN 0-8014-9541-5.

Thompson, Brad J.; "Moral Value, Response-Dependence, and Rigid Designation" in Canadian Journal of Philosophy, Volume 36, Number 1 (Mar 2006), pp. 71-94, URL=<>

Wiggins, D.; "Truth, Invention and the Meaning of Life" in Needs, Values, Truth (3rd Edition), Clarendon Press, Oxford, England. 1998. ISBN 978-0-19-823719-8.

Wilson, Catherine; "Perception, Evolution and the Moral Properties Debate" in Proceedings of the Aristotelian Society, 132nd Session, Dec 2010.

Zangwill, Nick; "Against Moral Response-Dependence" in Erkenntnis, Vol 55, No 2 (2001), pp 271-276, URL=<>

[Up] [Home] [Next]