202 Handout #2

PHIL 202: Practical Reason
Winter 2004; David O. Brink
Handout #2: Hume, Instrumentalism, and Internal Reasons

HUME
A. The Simple Argument for Skepticism about Practical Reason

Reason is a faculty that judges of the truth and falsity of ideas (beliefs) [Treatise III.i.1/458].
Actions and desires are not ideas.
Hence, actions and desires cannot properly be said to be conformable or contrary to reason.

B. A More Subtle Form of Skepticism

However, desires (or actions) can be said to be contrary to reason only so far as they depend upon false beliefs about matters of fact or relations of ideas [II.iii.3].
Actions depend on desires.
Some desires are derived from more basic desires and beliefs about how best to satisfy them.
But not all desires can be derived desires; derived desires must ultimately terminate in desires that do not depend upon more ultimate desires [Enquiry appendix I, sect. v].
Hence, derived desires that rest only on true beliefs cannot be contrary to reason.
Hence, ultimate desires cannot be contrary to reason.
Hence, actions that rest on derived desires that rest on true beliefs or on ultimate desires cannot be contrary to reason.
Hence, gross solipsism is not (inherently) contrary to reason [II.iii.3/416].
Hence, gross imprudence is not (inherently) contrary to reason [II.iii.3/416].
It is not the case that desires that rest on true beliefs about instrumental relations are reasonable.
It is not the case that ultimate desires are resonable.
Hence, no actions are reasonable.

Hume articulates (8) and (9) in an infmaous passage

‘Tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger. ‘Tis not contrary to reason for me to chuse my total ruin, to prevent the least uneasiness of an Indian or person wholly unknown to me. ‘Tis as little contrary to reason to prefer even my own acknowledg’d lesser good to my greater, and to have a more ardent affection for the former than the latter [II.iii.3/416].

Note that Hume could argue that solipsism and imprudence are typically, though not inherently, contrary to reason insofar as both are typically incompatible with other ultimate desires we have that presuppose the continued existence of the world and ourselves.

The crucial question in assessing Humean skepticism is why we should accept (1) and Hume's associated claim that

Reason is, and ought only to be, the slave of the passions, and can never pretend to any other office than to serve an obey them [II.iii.3/415].

Hume can appeal to his claim that reason is only a faculty of judging the truth or falsity of ideas (III.i.1/458). But this assumes that there is no such thing as practical reason, only theoretical reason, which is close to defending skepticism by appeal to skepticism. Moreover, as we will see, if there can be theoretical reason that operates on practical content, this may be enough to defeat the claim that reason can only be the slave of the passions.

HUMEANISM
Modern Humeans (e.g. Brandt and Williams) appeal to some Humean claims but draw more constructive conclusions. The modern Humean assumes that practical reason endorses desires or actions that contribute to the satisfaction of one's desires, provided that these desires are not themselves based on false beliefs about instrumental relations. Because ultimate ends are immune to this sort of criticism and all derived desires relate ultimately to ultimate ends, instrumental rationality can be defined in terms of adopting necessary conditions and means conducive to achieving one's ultimate ends.

If Hume is a skeptic about practical reason, then Hume is no Humean. Humeans may reject the asymmetry in (1) and (10). Some Humeans reject (10) without rejecting (11). This is the idea that rationality is purely instrumental. But we might ask how derived ends can inherit normative significance if the ultimate ends from which they are derived have none.

WILLIAMS ON INTERNAL REASONS

Internal reasons are relative to an agent's "subjective motivational set"; whereas, external reasons are not (101-2).
The relevant elements of an agent's subjective motivational set are her desires (broadly construed) (101, 105).
Internalism need not appeal to actual desires if these are based on false beliefs about instrumental means (102-3).
So one can be mistaken about one's internal reasons.
Internal reasons can be relativized to desires an agent would have after suitable deliberation on and from her initial (pre-deliberative) desires (104-5).
Williams is vague about what can count as suitable deliberation (105, 110).
Deliberation cannot involve discovery of facts about value or practical reason, for then the new motivation (desire) would not depend on the old.
Perhaps Williams has in mind content-neutral forms of deliberation, such as full non-normative information, good means-ends reasoning, and vivid appreciation of the antecedents and consequences of one's alternatives.

Those are some main aspects of Williams's version of internalism. But that still doesn't explain why we should endorse internalism. Does Williams do more than subject externalism to the incredulous stare? Williams's official rationale for internalism appeals to links between motivation and possible explanation (106-7).

But if we distinguish between explanatory and justifying reasons, we should reject the idea that justifying reasons are actually explanatory. Williams's own internal reasons are not actually explanatory.
It's not clear that a link between justifying reasons and possible explanation provides any constraint on justifying reasons. For any account of reasons, reasons will explain an agent's behavior insofar as she is rational.

Another strategy is to see Williams as appealing to the sort of resonance or non-alienation requirement at work in his earlier criticism of utilitarianism and other impartial moral theories -- the demands of practical reason must be capable of sustaining an agent's emotional and motivational engagement.

Practical reason must meet the resonance constraint.
Practical reason must be capable of motivating agents.
Motivation involves desire.
New desire can only be produced by content-neutral deliberation from pre-existing ultimate desire.
Hence, practical reason must be relativized to pre-existing ultimate desire.

This argument turns crucially on (3) and (4). (3) is sometimes represented as the Humean theory of motivation. (4) could be respresented as the Humean theory about practical reason, that reason can only be the slave of the passions. One might dispute either (3) or (4).

Some (e.g. Nagel and McDowell) reject (3), offering purely cognitive theories of motivation. (3) is supported by Michael Smith and others who explain intentional action as the product of representational states, such as belief, and practical states, such as desire, which are distinguished by their different directions of fit to the world. Roughly, beliefs are intentional states that aim to conform to the world, whereas desires are intentional states that aim to make the world conform to them.

One might reject (4), claiming that desire can be responsive to reason. If we can reason about the value or appropriateness of desires, then we can form beliefs about how the world ought to be. But given (3)'s conception of belief and desire, such beliefs would normally produce desires to make the world be as it ought to be. If we accept (3) but reject (4), we accept the Humean dictum that action requires desire but reject the Humean dictum that reason can only be the slave of the passions. We can also say that Williams's defense of internal reasons depends on a prior -- but undefended -- commitment to skepticism about the powers of practical reason.

What does this imply about internalism? We can accept internalism when broadly construed as the thesis that practical reason must be capable of motivating agents. We might even accept the thesis that practical reason be capable of producing desires. But if desire can be consequential on recognizing value or the demands of practical reason, then we can and should reject internalism understood as the narrower idea that practical reason must be relativized to antecedent desire. This means that internalism is either ecumenical and plausible or sectarian and problematic.