Lucky Virtue and Neighbor's Knickers: Two Problems with Goldman's Veritistic Unitarianism

In his essay “The Unity of the Epistemic Virtues,” Alvin Goldman defends a species of virtue unitarianism which he calls veritism—a consequentialist, truth-linked account of what, exactly, shall qualify as an epistemic virtue. Goldman defines veritism as:

V1: A process, trait, or action is an epistemic virtue to the extent that it tends to produce, generate, or promote (roughly) true belief.

Veritism is a species of virtue Unitarianism not in traditional senses which suggest the virtues are either reducible to one, or inseparable, but rather, because there is a thematic unity (i.e. all epistemic virtues produce, generate or promote true belief). This position which Goldman advances is to be contrasted with virtue pluralism as well as with any other Unitarian account that stipulates a non-veritistic thematic unity.

Goldman takes labor in his paper to defend veritism against objections from a variety of breeds of opposition, and for that reason, he has at least made the case that veritism has some excellent explanatory virtues. I think, though, that there are two problems that stand out as potentially damaging for any endorsement of veritism.

(1) The problem of “Lucky Virtue”

In section four of his paper, Goldman defends veritism against an objection that, for simplicity, I’m characterizing as:

1. If veritism is to be a Unitarian approach, then it must posit one (not two) epistemic values.
2. Veritism is committed to positing two epistemic values, true belief and error avoidance.
3. Therefore, veritism is not a Unitarian approach.

P1 is correct, and so Goldman seeks to disrepute P2 by showing how maximizing true belief and avoiding error can be blanketed under one veritic value. Goldman strategy here takes the form of his proposing a veritistic value scale. Given a proposition p, Goldman postis that the highest degree in belief in p would be 1.0 (subjective certainty), the lowest degree of belief would be 0, and .5 would represent “maximum subjective uncertainty” toward p. Goldman writes: “I propose that the highest degree of belief in a true proposition counts as the highest degree of “veritistic value” (witih respect to the question at hand, e.g., whether p or not-p is the case). In general, a higher degree of belief in a truth counts as more veritistically valuable than a lower degree of belief in that truth” (Fairweather and Zagzebski, 36).

As I see it, by positing this veritistic value scale, Goldman is essentially dodging the objection he intended to at the expense of opening the door to a separate and I think equally troubling objection. I will present this objection as “The case of Lucky Virtue”

The Case of Lucky Virtue

Lucky and Careful are investigating separate barnyards, with the goal of identifying cows. Lucky got a fortunate draw, for in the barnyard in which he is inquiring, there are five fat cows, and five giraffes, and those animals exhaust the barnyard. Lazy sits in a lawnchair, takes a quick survey and immediately develops subjective certainty (degree of belief 1) that there are five cows in the barnyard. “For surely,” he thinks, “cows have not got 30 foot necks.”
Careful, on the other hand, is investigating a barnyard with five skinny cows, five fat goats, five oxen, five black-painted fattened sheep, five tusk-free rhinos and five cow facades. Rather than to take a quick survey as Lucky had done, Careful takes out his notebook and tenaciously surveys all of the animals. His pre-barnyard-investigating knowledge of cows was equivalent to Lucky (mediocre at best), but the ecumenical obfuscation of his barnyard makes his epistemic position such that, after his three hour survey, he does not have subjective certainty in his belief that there are five cows. (His belief, though correct, is only held with .9 confidence).

Out intuition should be, I think, that Careful is more intellectually virtuous than Lucky. However, if we apply Goldman’s veritistic value scale to the situation, then Lucky will have acquired more “veritistic value” given that he has a higher degree of certainty in his true belief. Because Careful actually displayed much more intellectual virtue in his inquiring, though, the verdict Goldman would give seems counterintuitive.

I’ll move now to the second potentially problematic objection to Goldman’s verisitsic Unitarianism.

The Case of the Neighbor’s Knickers

After counting cows, Careful (now discouraged for having lost—he thinks unfairly—veritistic value competition to Lucky) sits on his front porch in disgust. Two neighbor children, Tommy and Bobby approach him. Tommy tells him that he peeked at the neighbor girl’s knickers and found that they were green. Bobby disagrees and says that the saw them and that they are yellow. Careful thinks to himself, “Well, I certainly like to produce, generate or (roughly) promote true belief, and so I’ll climb the ladder and peep into the neighbor’s bedroom to determine which is correct.” Careful climbs the latter, catches a glimpse of the neighbor’s knickers (they are red, by the way), and climbs down the ladder grinning.

Shame on Careful! But wait: should we slap his wrist or honor him as a beacon of intellectual virtue? This is a tricky question for Goldman’s account. First off, consider what process Careful is using to arrive at the true belief: “The neighbor’s knickers (contrary to opinion) are red!” Perhaps the process of careful investigation? Tenacious inquiring? Verifying testimony? No matter how the pie is cut, Careful’s disposition to investigate the color of the knickers reflects a disposition that would “promote, generate, or (roughly) produce true belief” which secures his disposition a spot on Goldman’s list which fall under the unified theme of veritistic. In section eight of his paper, in his discussion of welfare Unitarianism, Goldman makes it clear that he recognizes that problems of knicker-peeping sort will occur—i.e. situations in which what morality seems to demand will set limits on the extent to which epistemic inquiry ought to be carried. He says:

Epistemological or scientific value sometimes conflicts with moral value, and when they conflict, epistemological value must give way. There is a moral “side constraint” on scientific research, which is that the conduct of such research should not violate human rights or injure people.” (45)

This might appear to be a solution to the problem, but I think it is an overly hasty one. Ought epistemological values always yield to moral values? Those who identify intellectual virtues as a subset of the moral virtues (i.e. Zagzebski 1996) don’t seem to run into this sort of problem, but because Goldman does, I think he must also meet the task of defining clear conditions under which the exercise of veritistic virtues should yield way to the exercise of moral virtues that yield contrary courses of action. Merely offering the default position that intellectual virtues ought always yield to moral virtues seems somewhat careless (especially, for example, when the consequences are great.) For example, if it was not knickers Careful was attemping to peep into through the window, but rather, some word printed on the neighbor’s wall that, if he uttered it on the streets, a bloody war would end, it is not so clear that the intellectual virtue that would dispose him to peep should give way to the moral virtue of (say) not violating rights.
Another problem I see is that, even if Goldman were to outline careful “trumping” conditions, it seems that veritistic virtues would then be a sort of “conditional” virtues (i.e. investigate carefully only if you aren’t violating x, y, and z) that they would be quite far removed from the Aristotelian conception of virtue as being such that it leads us to act naturally in line with the virtue. Even though it is true that Goldman certainly doesn’t limit what qualifies as virtues as proper character traits, his position of veritisim nonetheless allows for them to make the list by meeting his veritistic condition. And so, I fail to see how an adequate picture of the virtues that are “deep and enduring character traits” (i.e. such as Careful’s trait of careful inquiry) could be adequately explained on his view, given that his view would require a careful conditionalizing of them—a conditionalizing to the extent that they would not longer be dispositions to act. Virtues of character, then, would be reduced to maxims which the agent must consider carefully (i.e. I’ll act intellectually virtuously only if x, y, and z) which seems to be quite a stretch from whatever we ought to call a “virtue theory.”