S. Joel Garver
The Power of "Truth":
Foucault and Doxastic Practices
"On Truth and Lie in an Extra-moral Sense"
Let us review the ground we have covered thus far. It seems that we regularly hold people responsible for what they believe or fail to believe, praising or blaming them when it seems appropriate. We found that while responsibility for a state of affairs implies some form of control over that state of affairs, it need not presuppose basic or immediate or direct control over it. In fact, we hold people responsible for many states of affairs over which they have no direct control.
The control necessary for responsibility need only be a form of long term control or a control theorized in terms of influence. As such it does not require "act foresight" of the consequences of actions and practices. Since we do not appear to have direct control over belief, our responsibility for belief, therefore, must be accounted for in terms of how we come to have and how we engage in certain fundamental patterns of doxastic practice or habits.
Accordingly, there are many cases that are clearly ones of epistemic responsibility, but this does not imply that praise and blame attach primarily to individual beliefs and their propositional contents. Rather, it implies that beliefs are praiseworthy or blameworthy in respect to how they were formed--which doxastic practices were engaged in and in what manner they were engaged. Cases of epistemic blame can, therefore, be distinguished from those cases in which allowable practices engaged in properly are engaged in an ethically blameworthy situation. Thus the student who sneaks a peek at the exam his professor has left on the desk gains knowledge in an ethically blameworthy fashion, though the knowledge itself is not epistemically impeachable. That is to say, there is a distinction between general ethical norms (including those that define when a practice is properly engaged in, merely ethically speaking) and epistemic norms (including those that define how a practice is to be engaged in properly for the purpose of epistemic success).
Epistemic obligations were related to justification in the following ways. A belief's having or not having justification (i.e., it's being in a state of justification) is not the same as a belief being such that it is reasonably taken to be (prima facie) justified. Nor is it the same as being able to justify a belief by arguing that it is reasonably taken to be prima facie justified. That is to say, reasonable ascriptions and practical defenses of justification do not precisely track actual cases of justification, though they are sufficient, pragmatically speaking.
Thus, while a deontological account of a belief's being in a state of justification cannot be given successfully, it may be possible to give a deontological (or prescriptive) and action-guiding account regarding how to form beliefs that are reasonably taken to be (prima facie) justified, how to make practical ascriptions of (prima facie) justification, and what kinds of belief are defensible. The basic epistemic obligation, then, is to form beliefs only on the basis of well-established doxastic practices that are properly engaged.
Which practices count as well-established is a matter of some debate. At the very least, the following characteristics can be among those that are (jointly) sufficient for a practice being well-established: it is engaged in by significant portions of a population; it does not produce outputs that are (massively and persistently) inconsistent either internally or in relation to outputs of other practices; it provides significant self-support; it is well-structured; and so on. Being well-established is clearly a matter of degree.
In respect to the practices discussed in the last chapter together with the material from Millar, these points imply that a well-established practice will make use of a coherent conceptual framework (this is also, in part, what leads to consistent outputs). Furthermore, it is primarily inconsistencies between the outputs of several practices indexed to the same conceptual framework that are epistemically important. It is perilous, therefore, to compare the outputs of a practice within one conceptual framework to those of another practice (or even the same or analogous practice) within another framework. Such outputs must be placed in their own conceptual contexts and the extent of overlap between frameworks is epistemically important to evaluation.
It may also be the case that, since conceptual frameworks are pragmatically structured and action guiding, that teleological considerations must be taken into account in evaluating practices--both in terms of their presuppositional function and actual effects. That is to say, it may be the case that the ends presupposed by a certain practice must be epistemically legitimate ends for a practice to count as well-established. Furthermore, the effects of pursuing those ends might have to be epistemically legitimate effects. What this exactly means will occupy us below, especially in Chapter 6. For now it is important to note that such an approach may necessarily import normative assumptions--whether teleology is evaluated in terms of the will of God or in terms of Nietzschian considerations of its life-affirming value or in terms of some other norm. In other words, more properly ethical considerations may well enter into epistemic evaluations of doxastic well-establishment.
Furthermore, since doxastic practices are pragmatically aimed, epistemically important inconsistencies within a practice or between practices within a conceptual framework cannot be limited to mere logical inconsistencies between propositional doxastic outputs. Rather, there can be inconsistencies between those outputs of a practice (e.g., the belief that individual freedom and rights are of great importance and value) and the practical effects that the same practice produces or enables to be produced (e.g., a proliferation of disciplinary technologies that are believed to contribute to freedom, but really enslave; cf. Foucault 1979; Bartky 1990:63-82).
These sorts of inconsistencies may be concealed from most of those who engage the doxastic and non-doxastic practices that give rise to them and that concealment may be a function of the practices themselves. Conditions regarding the accessibility of grounds may, therefore, also be important in these situations for the purpose of evaluating belief.
Also note that the scenarios of these last paragraphs while not, I think, unusual in any way, also do not characterize most of our rudimentary beliefs about the world from sense perception and memory (apart from those held by subjects we classify as suffering from psychoses and the like). These scenarios are, however, important for many of the beliefs that are most dear to us--regarding ethics, value, aesthetics, personal fulfillment and satisfaction, politics, art, literature, our neighbors' welfare, religion, worship, conversation, economics, family, sexuality and gender, race, and the like.
Finally, there is the matter of what it means for a practice to be engaged in properly. Proper engagement is largely a socially defined matter that includes a number of social factors: whether the subject is engaging a practice to the correct epistemic ends, whether the subject is in an epistemically legitimate position or has the right sort of epistemic authorization to engage the practice, and so on. Though this area is socially regulated it will also draw out epistemically important characteristics of the believing subject: her emotional state, her personal identification with or rejection of social forms, her desires and goals, and so on. Deviance from some ethical norms in these areas may have significant epistemic fallout and thus it is proper in those cases to classify such norms as not merely ethical, but also as epistemic norms.
Towards the end of clarifying the issues outlined in the last several paragraphs, we will begin by turning to the work of Michel Foucault and his theorization of power/knowledge, desire, and so on. Foucault gives a lot of attention to conceptual frameworks (roughly, what he terms "discursive formations"), social epistemic norms, social epistemic teleology, and the like. Thus he will be of great assistance in discussing and penetrating many of the issues raised. This chapter will introduce the basic outlines of Foucault's theories in regard to what are more external, clearly epistemic, and conceptual aspects of belief-formation. The following chapter will focus more on what is internal to the subject, on teleology, and on the ethical.