El Experiencialismo de Comesaña: Sobre Being Rational and Being Right de Juan Comesaña

Comesaña’s Experientialism: About Being Rational and Being Right by Juan Comesaña*

Esta obra está bajo una Licencia Creative Commons Atribución-NoComercial 4.0 Internacional.

Juan Comesaña’s Being Rational and Being Right is a brilliant work that displays the virtues of analytic philosophy at its best. Written in concise prose and using formal tools to formulate and analyze the author’s claims and theses, it unfolds as an array of interwoven arguments for a view he calls “Experientialism,” which provides an answer to an ancient philosophical question: What is the role of experience in rationalizing our beliefs?

Although this is not a book for beginners, since many of Comesaña’s arguments are dense and complicated, and require plenty of prior background to be correctly followed and understood, I assure the reader that the effort that it demands will be highly rewarded. One’s understanding of the themes it touches on are substantially enriched thanks to Comesaña’s rigorous and carefully crafted arguments.

Given the complexity and theoretical richness of this book, there are many issues that deserve careful discussion, and in this brief note I will not be able to do justice to that richness. To make the most of the space available, I will proceed as follows. I will first describe the fundamental features of Experientialism. Then, I will briefly go over how the chapters of the book unfold. Finally, I will describe and discuss one problematic issue concerning how Comesaña’s views interact with the liberalism/conservatism debate in the theory of justification.

The identity of Experientialism emerges from Experientialism’s opposition to its rivals: Psychologism and Factualism. These views offer different conceptions of how experience provides us with evidence for our beliefs: for Psychologism, an experience with the content that p is itself evidence for the belief that p (pp. 118-119);¹ for Factualism, the experience that p does not work like evidence for p but rather the experience provides p as evidence, and the experience plays this provider role if and only if it makes p known to us (p. 119). Experientialism agrees with Factualism in that the experience that p is not evidence for p, and that it provides p as evidence instead. But Experientialism disagrees with Factualism in that the experience can play this provider role not only when the experience makes p known, but also when the experience makes it justified for us to believe p (pp. 117, 195), where “justified” is understood as non-factive.

In cases where the experience with the content p makes it rational for S to believe that p, Psychologism categorizes the case as one where the rationalization is achieved through some evidence, i.e. the experience. In contrast, Factualism and Experientialism classify those cases as ones where the rationalization does not involve any evidence; the experience justifies one in believing p without being evidence for it. This illustrates the point that “not all justification is justification by evidence” (pp. 4-5). Although both views classify the cases as cases of “basic justification,” i.e. justification that does not arise from the belief being based on evidence, only Experientialism allows that an experience delivers such non-inferential justification for p, even when that experience fails to make p known to the agent.

So, Experientialism connects evidence and rationalization or justification² thus: what makes a proposition p gain membership into the evidential corpus of S is that an experience with the content p makes it justified for S to believe p.

We can summarize the fundamental features of Experientialism as follows:

1. The empirical evidence S has at a time t is constituted by the propositions that S’s experience basically justifies for S at t.

2. An experience can justify p for S even if p is false, so there can be false evidence.

3. vs Psychologism: An experience justifies p for S without being evidence for p.

4. vs Factualism: An experience can make p become part of S’s evidence without making p known to S.

One of the central arguments for Experientialism connects practical and theoretical rationality. This argument uses as a premise what Comesaña calls “Fumerton’s principle,” which says that rational action presupposes rational belief (pp. 3, 71), and the principle that, in deciding what to do, an agent should consider only those options that are compatible with her evidence (pp. 92-93, 123). Combining these two premises with Experientialism, it follows that false belief can rationalize action because, according to Experientialism, false beliefs can be part of one’s evidence. The corollary that false belief can rationalize action enables Comesaña to deliver simple explanations of cases where Psychologism and Factualism sanction as irrational the actions of an agent who, intuitively, is rational in what she does (e.g. pp. 2, 87-88). So, Experientialism is superior to its rivals to the extent that it is extensionally more adequate.

However, Experientialism’s central claim that there can be false evidence leads Comesaña to argue against the combination of Traditional Decision Theory and Knowledge-First Epistemology, understood as a theory of evidence, for these theories combined entail that falsities cannot be part of one’s evidence and, hence, cannot rationalize action. Chapter 2 explains the mathematics of subjective Bayesianism (a theory of rational credence that combines Probabilism and Conditionalization) and of Traditional Decision Theory (which presupposes Probabilism). Chapter 3 criticizes subjective Bayesianism in favor of Objective Bayesianism, as the latter is deemed to be “closer to the truth” (p. 43). However, in chapter 4, Comesaña goes on to attack a version of Objective Bayesianism that emerges from those aspects of Knowledge-First Epistemology that provide a theory of rational credence; in particular, it attacks the equation K=E by making powerful arguments to the effect that not all knowledge is evidence and not all evidence is knowledge. Yet this does not mean that he ends up rejecting Objective Bayesianism tout court, since he believes that his Experientialism constitutes a better version of it (pp. 69, 116).

One response by knowledge firsters to cases where intuitively false belief can rationalize action (which militate against them and in favor of Experientialism), invokes a distinction between justifications and excuses: the persistent intuition that false beliefs can have a sort of positive epistemic status is explained by saying that, although they cannot be justified, they can be excusable. Chapter 5 contains a careful and insightful critique of two prominent instances of this “excuses maneuver.” The chapter also contains a discussion of another maneuver made by other knowledge firsters that concedes more than the excuses maneuver in that it grants that false beliefs can be justified, but it insists that the sense of justification in question is explainable in terms of knowledge. Chapter 6 moves on to highlight the merits of Experientialism as a theory of basic empirical evidence by contrasting them with the difficulties faced by Psychologism and Factualism; the chapter also contains an illuminating discussion of how the three theories fare in dealing with different types of defeaters. Chapter 7 discusses the question of what normative requirements hold for an agent when her initial beliefs are irrational; isn’t there a sense in which the subsequent attitudes she goes on to form on the basis of those starting points ought to “cohere” with them, notwithstanding their irrationality? The discussion connects this issue in a complex and novel way with the problem of contrary-to-duty obligations. This chapter is perhaps the book’s most challenging. Chapter 8 develops Comesaña’s solution to the “problem of easy knowledge.” Throughout its history, the way of presenting this problem has changed and the reader will find that Comesaña’s presentation of it is quite different from the way it is usually presented by other authors. But these differences are superficial, since the principles at the root of the problem remain the same. Comesaña’s solution to the “problem of easy rationality,” as he re-labels it, is among the boldest in the market as it involves denying that a certain form of non-deductive justification is possible. Chapter 9 has a kind of retrospective flavor and constitutes a fit closing for the book. Comesaña compares the Experientialism of Being Rational and Being Right with Reliabilism and Evidentialism, and also contrasts it with an earlier incarnation of his own views that he used to dub “Evidentialist Reliabilism.”

In the remainder of this critical note, I want to briefly discuss an issue I find problematic in the foundations of Experientialism: How exactly does experience justify belief?

To abbreviate, I will use ‘EXP’ for ‘an experience with the content p’.

As we’ve seen, Experientialism says that what makes p part of S’s evidence is that EXP justifies p for S, without being evidence for p. Comesaña is also clear that this justification is provided by EXP, even if it is not used by S as the basis of the corresponding belief; that is, the justification in question is ex ante or propositional, not ex post or doxastic:

[An] important feature of my view is the fact that it requires basic rational belief as a condition on evidence-possession […] the subject’s evidence is constituted by those propositions he is ex ante basically justified in believing at the time (p. 117. Order of sentences reversed)

But Comesaña does not elaborate on what makes EXP a provider of p as evidence. Other mental states can have p as their content without providing p as evidence, and without making p justified for one. So, what makes the mental state of having EXP a provider of p as evidence? What makes it the source of justification for p? It doesn’t seem to me that Comesaña is very interested in answering this question, which is puzzling, considering that there are well known theories (some of which could be seen as competitors of Comesaña’s) that invest copious theorizing in articulating answers to precisely those questions.³

Rather than providing an explanation of what makes EXP a source of justification, given that EXP itself is not evidence, Comesaña is interested in stressing (and putting to work) the fact that the justification provided by EXP is basic, in the sense this term is understood in his theory:

[…] the notion of a basic reason (or a bit of evidence) functions as somewhat of a technical term within Experientialism, defined mainly by its functional roles in the theory, the roles of being a proposition that a subject is justified in believing but not on the basis of being justified in believing other propositions […] (p. 131, my emphasis)

This explanation of basic justification is just a reminder of the point that the cases in question are not cases of justification by evidence, i.e. of justification achieved by basing a belief on other propositions one is justified in believing; EXP justifies p for S without being evidence for p.

Comesaña’s silence on what makes EXP a source of justification for p, together with his insistence that the justification EXP provides is basic in the above sense, may encourage the thought that EXP on its own has the power to justify p for S (whatever it is about EXP that endows it with this power). Even the final statement of his view encourages this idea:

A credence x in p by S is justified if and only if:

Either:

1. x=1 and p is a logical truth or S’s experiences provide him with p; or

2a. S’s experiences provide him with E;

2b. S’s credence x in p is based on E;

2c. Pr (p|E) = x

2d. There is no more inclusive body of evidence E’ had by S such that Pr (p|E’) ≠ x. (p. 209, my emphasis)

Conditions 2a and 2b concern inferential cases where the subject already holds some propositions as Evidence provided by his experiences and, based on those, forms further credences. The cases that concern us in the present discussion are the basic, non-inferential ones and these are referred to in the second disjunct of condition 1, with the phrase I’ve italicized: S’s experiences provide him with p. Nothing else in addition to the experience itself is mentioned as a condition for S’s experiences to “provide the subject with an initial corpus of propositions as evidence” (p. 208). So, it sounds as if experience on its own justifies p for S.

However, it turns out that Comesaña does not think that EXP on its own can justify p for S, for he remarks that the justificatory power of EXP exhibits one form of epistemic dependence. The question then arises as to whether such dependence does not create trouble for the basicness of the justification in question.

The epistemic dependence I’m alluding to consists in the fact that the justification that EXP can give S for p depends on S’s having justification for believing other propositions. This dependence comes up in Comesaña’s rejection of “transmission principles” for justification. Like other epistemologists, he thinks that some cases that were historically wielded as counterexamples to closure are instead counterexamples to transmission:

[…] epistemic dependence relations do not always mirror entailment relations. Indeed, I take it that the right moral to draw from alleged counterexamples to closure such as Dretske’s painted mule case […] is not that Single-Premise Closure is false, but rather that transmission principles are false. That the animal in the pen is a zebra entails that it is not a painted mule, but our justification for believing that it is a zebra depends on our justification for believing that it is not a painted mule, not the other way around (p. 187, my emphasis).

In what follows, I will abbreviate “The belief that the animal in the pen is a zebra” as “The belief that ZEBRA,” and “The belief that the animal in the pen is not a painted mule” as “The belief that not-MULE.”

Comesaña does not explain what he means by saying that our justification for believing that ZEBRA “depends on” our justification for believing that not-MULE, but a natural and minimal reading of this phrase is that, contrary to what is suggested by the official statements of Experientialism, the experience of a zebra on its own doesn’t do all the job of giving justification for the belief that ZEBRA; it needs the concurrence of justification for other propositions. This type of dependence, brought about in discussions of closure and transmission principles, is very reminiscent of the debate between “conservatism” and “liberalism” in the theory of justification. Conservatives endorse that kind of dependence because they think that an experience can justify a corresponding belief only if the subject has independent justification for certain background presuppositions; liberals reject the dependence and hold that an experience on its own can justify a corresponding belief, with no need for independent justification for background presuppositions.⁴ Since Comesaña’s dependence is in the spirit of conservatism, in what follows I will refer to it as such.

Now, the case in which one visually identifies an animal at the zoo would seem to be a paradigmatic case of non-inferential or basic justification, if so, Experientialism says that the visual experience of the zebra gives S basic justification to believe that ZEBRA and hence the proposition ZEBRA becomes part of S’s evidence. Is Comesaña’s conservatism compatible with this story about how ZEBRA becomes part of one’s evidence? Conservatism is not an immediate threat to the basicness of the justification for ZEBRA, because the epistemic dependence it postulates is not among justified beliefs but among justifications to believe. It doesn’t say that the justified belief that ZEBRA depends for its justification on being based on the justified belief that not-MULE, but merely that the justification S has for ZEBRA depends on S having justification for not-MULE. So, strictly speaking, the epistemic dependence postulated by conservatism is consistent with the justification for ZEBRA being basic, in Comesaña’s sense, which only excludes “evidential justification,” i.e. the circumstance in which the justification of the belief depends on its being based on other justified beliefs.

Comesaña’s notion of basic justification rules out the possibility that the justification of a belief depends on its being based on other justified beliefs, but it seems to be consistent with depending on S having independent justification for believing other propositions. So, his conservatism in rejecting transmission principles seems to be consistent with his Experientialism.

Nevertheless, if conservatism is an unacknowledged part of Comesaña´s complete views, the claim that the justification provided by experience is basic, because experience justifies “without itself being evidence” (p. 130), needs at least some sort of caveat because it turns out that even in such cases of basic justification evidence still plays an indispensable positive role. Experience does not work as evidence in justifying beliefs, but in order to play that non-evidential role it nevertheless requires the aid of evidence in the background. For example, in the zebra case, the independent justification for believing that not-MULE could consist of several propositions that are part of S’s evidential corpus because they are justified through the testimony of the Zookeeper or through the results of a DNA test. A subject with a sufficiently impoverished experience, which has not provided her with the needed independent evidence to justify not-MULE for her, will be such that her experience of the zebra will not be able to provide her with ZEBRA as evidence, because her experience will not justify ZEBRA for her, because she will lack the requisite justification to believe not-MULE, upon which her justification for ZEBRA is said to depend.

It then turns out that S’s experiential justification for ZEBRA does depend on some independent evidence she must already have (granting that the dependence does not consist in the fact that her belief that ZEBRA needs to be based on that evidence). Maybe this is true, but it sounds misleading to insist that the experiential justification subject to such conditions is basic justification, that it is the justification whereby “experiences provide the subject with an initial corpus of propositions as evidence” (p. 208, my emphasis). Instead, it looks like that initial corpus of evidence (or at least a good relevant chunk of it) must already be there, for experience to play its justifying role.

Another aspect of Comesaña’s attitude toward conservative ideas that I find puzzling is that he does not seem to apply them uniformly. For example, when describing how Experientialism handles some cases that are central examples throughout the book, he completely omits any reference to the sort of conservatism that he himself relies on when handling the zebra case in the context of his rejection of transmission principles. Consider the case of the apparent candy:

Bad Lucas: Tomás would like to eat some candy. Lucas offers him a marble that looks just like candy, and Tomás reaches for it and puts it in his mouth […] (p. 87)

Let’s abbreviate the proposition “Lucas is offering Tomás a candy” as “CANDY” and the proposition “Lucas is offering Tomás a marble that looks like candy” as “MARBLE.”

Comesaña argues that one of the virtues of his Experientialism is that it can vindicate the commonsense verdict that “Tomás acts and believes rationally in Bad Lucas” (p. 94), while Psychologism and Factualism cannot (pp. 2, 87).

According to Psychologism, Tomás relevant evidence consists of his experience with the content CANDY. Since Tomás’ candy experience is compatible with the possibility that MARBLE, and with a myriad of hypotheses where it appears to him as if CANDY but it is not true that CANDY, then, given that in deciding what to do he ought to consider the options compatible with his evidence, Tomás should consider all those hypotheses. But, since he doesn’t, he is irrational. From the perspective of Factualism, Tomás’s relevant evidence also turns out to be insufficient to rationalize what he does. Tomás’s relevant evidence is too meager: Since he doesn’t know that CANDY (because CANDY is false), CANDY is not part of his evidence; since he doesn’t know that not-CANDY (because he doesn’t believe it), not-CANDY is not part of his evidence either. So, he should consider the option that MARBLE because his evidence doesn’t rule it out. Since he does not consider such an option, he is irrational.

In contrast, according to Experientialism, CANDY is part of Tomás’s evidence because his experience basically justifies him in believing that CANDY. So, Tomás’s evidence is incompatible with the option that MARBLE, hence he need not consider it in deciding what to do and his action is therefore rational.

In explaining how Experientialism handles this case, Comesaña never mentions any conservative idea. He only emphasizes that the only condition required for Tomas’s experience to play its epistemic role is the absence of defeaters:

Tomás was rational […] In his situation, where nothing indicated that he couldn’t take his experience at face value, Tomás’s evidence was the content of his experience: that Lucas was offering him candy (p. 2, my emphasis)

[…] I appeal to the notion of an experience providing a proposition as a reason […] this will happen whenever the justification that the subject has for believing the content of the experience is not defeated by other beliefs the subject is justified in having (p. 209, my emphasis)

His experience provides Tomás with the proposition CANDY as evidence because it makes that proposition justified for him, and the experience delivers this ex ante justification on condition only that Tomás lacks defeaters. Comesaña is insisting here that the condition for the experience to deliver justification is merely that Tomás lacks justification for MARBLE but, if he applied conservatism to the case, he should add a stronger condition: that Tomás should have justification for not-MARBLE.

Note that if conservatism is applied to the candy case, as Comesaña applies it to the zebra case, then, for Experientialism to retain its explanatory advantage over Factualism and Psychologism, Tomás must have independent justification for not-MARBLE. For if he didn’t, his experience would not justify him in believing that CANDY, the proposition CANDY would not be part of his evidence, so he would have to consider that MARBLE in deciding what to do (because MARBLE would not be excluded by his evidence) and his action would be irrational. In this scenario, just as is the case with its rivals, Experientialism is not able to validate the commonsense verdict that Tomás´s action was rational.

In dealing with the candy case, Comesaña does not go into the important consequences that conservatism would ultimately have on it. This suggests that he simply does not regard conservatism as part of the correct view about the epistemic structure of the case. But why should we treat differently the case of Tomás’s belief that CANDY from María’s belief that ZEBRA? If Comesaña allows conservativism concerning the latter, why doesn’t he allow it in the former? Such disparity appears unwarranted. Since both are cases of casual visual experience, prima facie the conditions for experience to deliver justification should be of the same type in both. Comesaña owes us an explanation of why he treats those cases differently on a point which has such important consequences for how Experientialism deals with them.⁵

redalyc-journal-id: 3400