We’re All Epistemic Deontologists
Abstract: The so-called “deontological conception of epistemic justification” (or “epistemic deontology”) is much discussed in recent epistemological literature. It is not uncommon to hear philosophers claim that they reject the deontological conception of justification. I argue here that this rejection is mistaken and due to an inattention to the variety of kids of views that might be considered epistemic deontologists. Based on a a survey of representative characterizations in the epistemological literature, I distinguish three broad kinds of epistemic deontologies, and then argue that two are implausible and accepted by few epistemologists, but that one is plausible, and accepted by nearly all epistemologists. Thus, I argue that, in an important sense, we are all epistemic deontologists.
Keywords: epistemic deontology, doxastic voluntarism, ethics of belief, belief control, voluntary control, virtue epistemology
1. Introduction
The so called “deontological conception of epistemic justification,” or “epistemic deontology,” is much discussed in recent epistemological literature. Many philosophers have argued that it ought to be rejected. Some argue it should be rejected it because they think it implies “doxastic voluntarism,” the view that, roughly, we have direct voluntary control over what we believe. This objection is based in the idea that if there are things we ought to believe (or not believe) or are obligated to believe (or not believe), then it must be the case that we can believe them (and can refrain from believing them). Critics argue that this latter condition requires abilities to influence what we believe that we do not have, so the deontic conception is mistaken.
The deontic conception is also sometimes said to imply epistemological internalism, the view that epistemic justification depends on internal, or mental, or accessible states alone. Again, this objection might be motivated by a thought along the lines that if there are things we ought to believe, then we must have some access to, or awareness of, what justifies our beliefs, and so some kind of internalism must be true. This thought might be motivated again by a kind of “ought implies can” principle: for an evaluation of what one ought to believe to be true, it must be the case that one can access all of the factors that epistemic justification depends on. Those who reject internalism, often because they think justification depends on, in part, factors that are not cognitively accessible (perhaps because they are non-mental factors), will thereby reject epistemic deontology, if they think it has this consequence.
Another consideration raised against epistemic deontology is that it is also said to imply that some virtue-based understandings of justification are mistaken; those fond of virtue epistemology might then reject it.[1] It might also be thought to objectionably divorce justification from truth since, perhaps, a deontic epistemic evaluation of someone’s beliefs can be appropriate yet their beliefs are not true or likely to be true. Those who accept some version of a truth-conducive conception of justification might thereby reject deontology. Finally, perhaps some might be uneasy with the suggestion that some deontologists make, viz. that there are “epistemic duties.” Perhaps they are uncomfortable with this kind of terminology because it again might suggest doxastic voluntarism, but some think that this makes reasonable belief “too demanding,” much more demanding than it really is.
Many argue that at least some of these implications cast doubt on epistemic deontology: they argue that if it has any of these implications, it is likely false. I will argue that these judgments are premature. While widely criticized, few writers have adequately clarified their understanding of the deontic conception. This clarification is needed to successfully argue against it: if the concept of the “deontological conception” is unclear, then it is not obvious that it has the allegedly false implications that its critics say it does. Thus, when philosophers claim that they accept, or reject, the deontological conception, it is not entirely clear what they have in mind.
Here I address this deficiency. Based on the wide variety of suggestions given by the critics and defenders of epistemic deontology, I develop three general interpretations of what it might be. I argue that some of these interpretations are implausible. Thus, if this is what some philosophers have in mind when they reject epistemic deontology, they are right to do so. However, other conceptions of epistemic deontology are plausible, I will argue, and these conceptions seem acceptable and accepted by all. Furthermore, it would be hard to coherently deny some of these deontic conceptions because to argue against them, it seems that one would have to presuppose the truth of some deontic epistemic evaluations, or presuppose a basic idea motivating many epistemic deontologists, viz. that there are intellectual requirements and that if some belief does not meet them, then it ought not be believed. I will argue that that idea, or variations on it, cannot easily be denied, especially in a plausible manner.
Thus, I argue that we all accept some version, or versions, of epistemic deontology: we are all epistemic deontologists. I show how this result defuses some of the tensions raised above, focusing on the internalism/externalism debate. I conclude that none of the considerations given against a plausible version of epistemic deontology or a deontological conception of justification refute it.
2. The Many Epistemic Deontologies.
As mentioned, there are many understandings of the deontological conception of justification. Below are a large number of representative characterizations of the deontological conception of justification. As we shall see, many different ideas share this label.
Proceeding alphabetically, William Alston (1991, pp. 72-73) characterizes the view this way:
being justified in believing that p consists in some sort of ‘deontological’ status, for example, being free from blame for believing that p or having satisfied one’s intellectual obligations in doing so.
Alston (1988, pp. 258-259) also presents the deontological conception in these ways:
[O]n the deontological conception of the epistemic justification of belief . . to be justified in believing that p at t is for one’s belief that p at t not be in violation of any epistemic principles . . that permit only those beliefs that are sufficiently likely to be true. . . . To say that S is justified in believing that p at time t is to say that the relevant rules or principles do not forbid S’s believing that p at t. In believing that p at t, S is not in contravention of any relevant requirements.
Michael Bergmann (2000, p. 87) discusses a perspective that
epistemic justification is essentially a matter of duty fulfillment.
Lawrence Bonjour (1980, p. 55) says that
the concept of epistemic justification . . . has to do with what one has a duty or obligation to do, from an epistemic or intellectual standpoint.
Anthony Brueckner (1996, p. 527) claims that a deontologist
conceives of epistemic justification as consisting in the fulfillment of epistemic obligation, or duty. . . On a deontological conception of epistemic justification, S is justified in believing that P if and only if S’s belief that P is in conformity to S’s epistemic duty.
Carl Ginet (1985, p. 28), who, according to Alston, is a “model deontologist” who “sets out the conception with admirable directness” (1988, p. 259), presents the view as
One is justified in being confident that p if and only if it is not the case that one ought not be confident that p; one could not be justly reproached for being confident that p.
Alvin Goldman (1999, p. 273) reports that
Epistemic deontologists commonly maintain that being justified in believing a proposition p consists in being (intellectually) required or permitted to believe p; and being unjustified in believing p consists in not being permitted, or being forbidden, to believe p. When a person is unjustified in believing a proposition, it is his duty not to believe it.
John Greco (2001, p. 117) says that
. . the main idea of deontological theories is that justified belief is a function of [or is ‘licensed by’] correct cognitive rules or norms.
Alvin Plantinga (1993) says that
epistemic deontology [is] the view that epistemic duty and obligation are of crucial epistemic importance and that . . being [epistemically] justified is being within our rights, flouting no epistemic duties, doing no more than what is permitted . . [and being] subject to no blame or disapprobation.[2]
And, finally, Matthias Steup (2000, p. 25; 1995, p. 231) reports that
Epistemic deontology is the view that the concept of epistemic justification is deontological: a justified belief is, by definition, an epistemically permissible belief. . . . Our beliefs are justified if, and only if, what we believe is epistemically permissible for us to believe. . . . It is a duty-based view of the nature of epistemic justification.
Epistemic deontology clearly is of interest: many philosophers find it worthy of discussion.[3] Most are interested in criticizing it; only a few explicitly are interested in defending it.
But there are a number of very different suggestions here about what it is in the first place, so many such that, ideally these must be addressed on a case-by-case basis. I aim to categorize these claims in order to identify a clear deontological view to criticize or defend. I will argue that some of these suggestions are plausible, and others are not. Insofar as some of the plausible ones are widely accepted, we are all epistemic deontologists.
3. Deontic Concepts versus Deontic Properties.
Although the topic is sometimes called the deontic conception of justification, I am not immediately concerned with how any philosophers would spell out their concept, or idea, of what justification is. To avoid concerns about concepts being merely “in the head,” and so perhaps highly idiosyncratic, I will focus on the concern of relations between properties, that of bi-conditional entailments between some deontic properties and the property of being justified. So, the fact that many epistemologists would not explain their concept of justification by appealing to any deontological concepts does not show that any epistemic deontologist’s claims about the nature of justification are false. For example, Alston (1988, pp. 257-258) claims that the term “justification” is “most naturally understood” in deontic terms. But for those who do not understand it in these terms, the question remains of what relations (if any) these putative deontic properties have to properties pertaining to properties that constitute epistemic justification.
A first set of suggestions relates the, or a, concept of justifiably believing p with the concepts of being “free from blame” or “disapprobation” for believing that p, and/or not being “justly reproachable” for being confident that p. Alternatively, in light of the suggestion above, claims like these might suggest an essential relation between the putative property of justifiably believing p with these latter blame (and, perhaps, praise)-oriented deontic epistemic properties. I will argue that these deontic conceptions of justification are false. A second set of suggestions relates the concept or property of justifiably believing p with concepts (or properties) of epistemic duties, obligations, requirements and/or permissions. A third kind of suggestion relates justification to epistemic rules, norms, principles or requirements. I will argue that these kinds of claims, these epistemic deontologies, are true and that, for the most part, we all accept them.
For each set of suggestions, we can ask what this relation is between an epistemic property or concept (e.g., the property or concept of justification) and the deontic property or concept in question. Bonjour suggests that the relation is that justification “has to do with” epistemic duties and obligations. Depending on what exactly is meant, this might be a rather weak relation: there is some connection, some relation, between the two kinds of notions or properties. Others make the stronger suggest that justification “consists in” or “is essentially” some other deontic evaluation; thus, some judgment about whether a belief is justification is true just in case another deontic epistemic evaluation is true also: there is a bi-conditional equivalence. This will be the equivalence I will consider. Steup and others suggest this equivalence, and Steup even sometimes appears to suggest something stronger in his claim that epistemic deontology “is a duty-based view of the nature of epistemic justification.” I will present one possible way to understand a stronger claim like this, i.e., as a robust theory of epistemic justification, and argue that it is false.
To evaluate these various claims that justification (or the concept) has some relation to deontic epistemic properties (or concepts), a concrete place is to begin with the variety of possible deontic evaluations. We can ask ourselves if these evaluations are ever true and, if so, whether this in any way seems to depend on our judgments about justification which, we are presuming, are sometimes true. That is, we can consider cases of intuitively justified belief and ask whether a corresponding deontic evaluation would seem to necessarily apply also. This is just to ask if these bi-conditionals are true: we consider a case of (un)justified belief, conjoin it with the bi-conditional, and see if that belief also has the positive (or negative) epistemically deontic status. We could also start with the deontic evaluations to accomplish the same task, in terms of evaluating the bi-conditional.