Today we began studying Sosa’s Epistemology in Professor Kvanvig’s Epistemology seminar. We finished Zagzebski’s Virtues of the Mind last week, and as I made my way through it I noticed some potential problems regarding her views on the acquisition of virtues. In the book Zagzebski outlines a motivation-based virtue theory, which includes an account of intellectual virtues for the purpose of an epistemic application. In what follows I will first provide some groundwork, outlining the internal component of a virtue in terms of a dispositional motivation, the external component of a virtue, where virtue is defined as a success term based on a reliability constraint, and the standards for virtue possession. After the ground has been cleared, I will propose a potential regress which faces her account as antecedently developed.
Virtue is defined by Zagzebski in terms of one’s motivations, where a motivation is a “persistant tendency to be moved by a motive of a certain kind.”1 A motive is, in other words, a low-level, action directing emotion. The particular virtues are based on these dispositional motivations, with each virtue taking a particular motivation as its source. This motivation-based account of virtue has both general and specific consequences. That is to say, the motivational component of a virtue will provide impetus for the agent to develop certain skills needed to carry out one’s goals, with the particular consequences of driving one to find out certain non-moral facts about particular circumstances, dependent upon the ends required by the motivational component of the virtue. These internal conditions lead directly into the external aspect of a virtue, where virtue is defined in terms of success.2 The success component of a virtue is independent of the motivation component, and it provides that reliable success in attaining the ends of the motivational component is a necessary condition for virtue possession. Reliability is given in terms of truth conduciveness, but this link may be either direct or indirect. For directly truth conducive virtues I will asume that the reliability component is cashed out in terms of reliably generating true beliefs. Indirect truth conduciveness, however, is given a somewhat vague definition, in that Zagzebski “suggest[s] that we may legitimately call a trait or procedure truth conducive if it is a necessary condition for advancing knowledge in some area even though it generates very few true beliefs and even if a high percentage of the beliefs formed as the result of this trait or procedure are false.”3
Zagzebski adds to this account of intellectual virtues that they are acquired “through the imitation of virtuous persons and practice in acting virtuously.”4 This launches an immediate regress:
(1) Person A acquires virtue V through imitation of person B.
(2) Person B acquires virtue V through imitation of person C.
(3) Person C … D.
(4) Ad Infinitum.
This is not the end of the matter, however. The previous regress may be sharpened through appropriate supplementation to more accurately account for the way in which a virtue is acquired:
(1) Person A assesses virtue V in person B, as B exhibits behavior X in appropriate situations (a1, a2, …, an), and therefore person A acquires virtue V through imitation of person B.
(2) Person B assesses virtue V in person C, as C exhibits behavior X in appropriate situations (b1, b2, …, bn), and therefore person B acquires virtue V through imitation of person C.
(3) Person C … D.
(4) Ad Infinitum.
Now, with this more accurate sketch of the acquisition regress involved, the question may be posed as to how one assesses behavior in appropriate situations. This behavior is to be considered successfull if it is to exhibit the virtue involved, and the success must be reliable. Reliability, however, is a vague concept. Given this, and the previous distinction between directly and indirectly reliable virtues, it would seem that any given directly reliable virtue may be reduced to an indirectly reliable virtue provided a slight modification of the previous regress:
(1) Person A assesses virtue V in person B, as B exhibits behavior X in appropriate situations (a1, a2, …, an) with degree of success P1, and therefore person A acquires virtue V through imitation of person B.
(2) Person B assesses virtue V in person C, as C exhibits behavior X in appropriate situations (b1, b2, …, bn) with degree of success P2=(P1-1), and therefore person B acquires virtue V through imitation of person C.
(3) Person C assesses virtue V in person D, as D exhibits behavior X in appropriate situations (c1, c2, …, cn) with degree of success P3=(P2-1), and therefore person B acquires virtue V through imitation of person D.
(4) Ad Infinitum.
Here we have it. Given any directly reliable virtue, with an arbitrarily high degree of success, as the route is traveled through iterations of virtue acquisition success approaches 0.
- Zagzebski (1996), p. 132 ↩
- Cf. Zagzebski (1996), p. 100 and p. 134 ↩
- Zagzebski (1996), p. 182 ↩
- Zagzebski (1996) p. 157-158 ↩
