THE OPEN QUESTION ARGUMENT


R. N. Johnson
Phil. 411
As Darwall, Gibbard and Railton mentioned, one appeal of internalism is that it explains why Moore's "Open Question" argument is so compelling, even if it is universally thought to have failed to expose a fallacy in naturalism. The argument is (very roughly) this:

1. For any moral property M and any non-moral property N, M = N only if (x) "Mx" is synonymous with "Nx". (Putnam calls this the "analytic property identity criterion".)
2. (x) "Mx" is synonymous with "Nx" only if (x) it is not an open question whether, if Nx, then Mx.
3. But (x) it is an open question whether, if Nx, then Mx.
Therefore,

4. It is not the case that, for any moral property M and any non-moral property N, M = N.
Why does this argument fail?

A. We can reject premise 1. The support for it comes from the idea that any speaker competent with two synonymous terms must use the same criteria in applying them. But whether a given term applies to some object depends just on whether it has a given set of properties, and so the presence of these properties are the criteria for applying the term. Hence if term F is synonymous with term G, then their criteria of application are the same, and if their criteria of application are the same, then the objects to which they apply share the same properties. But then if F and G are synonymous, then no competent speaker could be in doubt about whether, if Fx, then Gx. Indeed, could a competent speaker be in doubt about whether, given someone is a bachelor, that he is an unmarried male? Surely not. So the fact that a person could be in doubt about whether, given something is pleasant or in possession of any other natural property, then it is good, must show that good is not identical with any a natural property.

But few accept this reasoning nowadays, for the simple reason that whether two properties are identical is a metaphysical question, not a question concerning the meanings of terms. Moreover, substantive metaphysical questions are taken by most philosophers to be synthetic, and so empirical, questions. Whether two properties are identical is a matter of discovery, not verbal legislation. This is especially the case for natural kind terms ("water", "gold", "hydrogen", etc.). Now clearly the property of being water is identical with the property of being composed of H2O. But "water" does not mean "H2O. Likewise, temperature is identical with mean kinetic energy, but "temperature" does not mean "mean kinetic energy". So someone could be convinced that a given object is, say, a lump of gold, and yet be in doubt that it has the atomic weight of 196.967. Yet the property of being gold and having atomic weight 196.967 are identical properties. So the biconditional in 1 is clearly false.

B. We can reject premise 2. For suppose we were to accept 1. Is it true that competent speakers could not wonder whether, if Nx, then Mx, and yet "Mx" mean the same as "Nx"? Surely not. For one thing, there are "opaque" synonyms. Even "bachelor" falls to this one. For is every unmarried male a bachelor? Or is a bachelor only someone who has never been married? If you are in doubt, you are in the majority of competent english speakers. And examples such as this seem to show that it may be an open question whether two expressions are intersubstitutable for competent speakers of the language, yet those expressions mean the same thing.

So it seems we should conclude that this is not a very good argument against naturalism. But it does seem to put its finger on something important, namely, that moral judgments have a certain authority or magnetism that factual judgments can lack altogether. This is why internalism seems to be so tempting. For internalists explain the "openness" of the question of whether, if Nx, then Mx, as being a function of this magnetic quality of moral judgments. For it seems that, for any non-moral, naturalistic description of something, we can imagine normal persons who do not find it of interest or moving. But a normal person is surely moved by the description of some act as "evil" or "virtuous". So the openness of the above question is explained by the tie of moral judgments to our own responses.

FRANKENA, "OBLIGATION AND MOTIVATION IN RECENT MORAL PHILOSOPHY"

I. In the first half of this article, Frankena presses a single argument against Humeans.They argue:

(1) "I ought to x" entails "I have a reason to x"

(2) But no external judgment about x can entail "I have a motive to x".

(3) Therefore, "I ought to x" is not an external judgment about x.

But this is fallacious, since it requires

(4) "I have a reason to x" entails "I have a motive to x".

(4) states that a justifiying reason for an action entails a motivating reason for it. Until (4) is defended, externalists may accept the claim that "ought" entails reasons, but reject (4), and insist that the connection between reasons and motives is entirely contingent.

II. In the second half, Frankena considers (4). He takes "having a reason" and "having an obligation" to be roughly the same things. Now (4) may have to do with either having, saying one has, or assenting to an obligation. But he thinks that externalism can reject (4), since arguments for (4) all rely on one of these dubious assumptions:

(a) Having a subjective obligation implies or is is identical with objective obligation. (Hence, while subjective obligations may not be possible without motivation, that we really are obliged to do something may well be unknown to us, and so completely untied to any motivation.)

(b) Having an obligation implies or is identical with assenting to or saying one has an obligation. (We may have obligations that we do not assent to, and so no subjective orientation toward that obligation is required.)

(c) That psychological or pragmatic implications or presuppositions cannot explain the motivation connection. (No one denies that it is purposeless to assert that someone is under an obligation if they are incapable of being moved to do it; but that does not support internalism.)

III. Frankena then considers Falk's moral argument against externalism. Falk accepts an internal and an external ought. But the external ought "can't be accepted" by a mature, reflective person aware of his capacity for reflective choice and bent on using it. Since we can't be guided by two different oughts, we ought to be guided by the internal ought.

But, Frankena objects, why not? And should we be guided by our internal ought?

IV. Frankena's ends with what he takes to be an important internalist view. An internalist must prove three things, he thinks:

(i) that having an objective obligation to x logically entails having some motive to x.

(ii) that justifying an objective obligation to x logically entails establishing some motive to x.

(iii) that recognizing an objective obligation to x logically entails having a motive to x.

One way of construing the argument he offers the internalist in favor or these three claims is this:

(1) There is in fact no gap between obligation and motivation.

(2) Externalism implies the possibility of a gap between obligation and motivation.

(3) But, then, externalism must explain the gap. But it can do this only in psychological terms, by positing a desire to do what is obligatory.

(4) But such a psychological claim is very likely false. (And there are other reasons for rejecting this view--See M.Smith's paper in the packet.)

(5) So externalism has no explanation of the absence of a gap.

(6) Internalism does explain it.

(7) So internalism is true.

It is important to see that internalism is the view that there is a logical connection between obligation and motivation, and thus Frankena thinks that it is possible for the internalist to offer the fact that there is such a connection as an explanation of a particular phenomena, viz., the fact that people are for the most part motivated in accordance with their obligations.

The issue between internalists and externalists, Frankena thinks, rests firmly on (a) proving a psychological claim (4) (so the issue between internalism and externalism cannot be settled through conceptual analysis alone); and (b) settling a normative dispute over whether morality ought to be represented as guiding (internalist) or goading (externalist).

FALK "'OUGHT' AND MOTIVATION"

I. Falk first argues that motives are not desires: Pritchard assumes that proving a connection between duty and motive requires proving a connection between duty and desire (to do one's duty). But it does not:

(1) "P has a motive to x" means P is either occurrantly or dispositionally impelled to x.

(2) If "P desires to x" means "P is impelled to x", then desire is a motive in the occurrant sense.

(3) But it does not usually mean this.

(4) So desires are not usually occurant motives.

(5) "P desires to x" usually means "P perceives herself to be impelled to x".

(6) But "P has a motive to x" can be true, and "P perceives herself to be impelled to x" false.

(7) So "P desires to x" doesn't mean "P has a motive to x".

(8) If "P has a motive to x" entails "P desires to x", then P's perception of her being impelled to x impells her to x.

(9) But this confuses the operation of a causal law with our perception of our operation of a causal law.

(10) So "P has a motive to x" does not entail "P desires to x".

(11) If "P is dispositionally impelled to x" entails "P desires to x" , then that P would be impelled to x were she to dwell on it entails that she desires to do it.

(12) But it does not.

(13) So "P is dispositionally impelled to x" does not entail "P desires to x".

One problem with Falk's argument is, I think, that he assumes a phenomenological view of desires. A desire, as he sees it, is a inner feeling of being impelled, a sensation. But if desires were feelings or sensations, then it is hard to see how they could be, as they most certainly are, propositional attitudes; pains, for instance, have no propositional contents. And if desires are not sensations, then Falk's argument does not go through, for then (5) is false. Moreover, Falk does not consider the possibility of teleological, rather than causal, explanations of actions. Hence, even if desires could not enter into causal explanations of behavior, they might enter into teleological explanations. Suppose that motivational explanations are teleological explanations, but not necessarily causal explanations. Then desires could well be motives, if they could figure in teleological explanations. (Whether this is so is something we will take up later in the course).

II. Falk's main concern, however, is in fact motivated by a normative concern, really grounded in the values of the Enlightment. He wants to argue that morality needs no (and ought not have) external sanction. His argument is roughly this:

(1) If moral "oughts" have their origin outside of the agent, then they are completely independent of the psychological constitution, and so motivations, of the agent.

(2) There is a purely formal motivational "I ought x" (i.e., "internal"), viz., that I would be made to will x were I fully rational.

(3) The thought that I would be made to will x were I fully rational can guide my attitudes and behavior (i.e., "What ought I do?" may ask "What would I will to do, were I fully rational?")

(4) So the internal "I ought" is to that exent suitable as a moral "ought".

(5) Now the moral "ought" decisively guides behavior. (A conceptual claim)

(6) But if the moral "ought" were both internal and external, it could not decisively guide behavior (we cannot "serve two masters").

(7) But someone "aware of a capacity of reasoned choice and intent on using it" could not use the external ought alone to decisively guide behavior.

(8) Therefore, we ought not use the external "ought" to decisively guide our behavior.

(9) Therefore, the moral "ought" is an internal "ought".

III. Falk believes that the request for a motive to do what one ought rests on a confusion of the external with the internal "ought". With the external "ought", it is possible to give an answer, but then the connection with motivation is contingent, and will not be satisfying. With the internal "ought", it is not possible to given an answer, since establishing an internal "ought" just is showing that one has a (dispositional) motive. This internal "ought", however, is often "blended" with the external ought out of fear of subjectivist implications, and this gives rise to the dilemma which Pritchard faced (to give an answer without giving an answer to the request for a motive). And doubts about one's having a motive to do what one ought are based on this confusion.