Epistemology and the Regress Problem

Scott Aikin

• Publisher: Routledge
• Year: 2010
• ISBN:
  9781136841897
• Edition: 1
• Topic: Philosophy
• Subtopic: Epistemology in Philosophy

In the last decade, the familiar problem of the regress of reasons has returned to prominent consideration in epistemology. And with the return of the problem, evaluation of the options available for its solution is begun anew. Reason's regress problem, roughly put, is that if one has good reasons to believe something, one must have good reason to hold those reasons are good. And for those reasons, one must have further reasons to hold they are good, and so a regress of reasons looms. In this new study, Aikin presents a full case for infinitism as a response to the problem of the regress of reasons. Infinitism is the view that one must have a non-terminating chain of reasons in order to be justified. The most defensible form of infinitism, he argues, is that of a mixed theory – that is, epistemic infinitism must be consistent with and integrate other solutions to the regress problem.

1
The Regress Problem

In this chapter, I have two objectives. The first is to make a case for an aspirationalist conception of justification. My second objective is to present the epistemic regress problem. On the one hand, I will argue that the problem is pressing for nonaspirationalist conceptions of justification; on the other hand, I will argue that the regress problem is of even greater concern for the aspirationalist program.

1.1 JUSTIFICATION AND ITS DESIDERATA

Having good reasons for our views is a good thing. This seems uncontroversial and perhaps tautological, but let me make the case. First, it seems clear that good reasons are simply valuable to us. Now, that good supporting reasons are valued may not yet show that they are valuable (or better, worthy of that valuing), but showing that we value them for good reasons does. On the assumption that a good reason for a commitment is a reason counting in favor of its truth, believing for good reasons is a means for having true beliefs. At least it is as good a means as we can come up by our own lights. And truth, simply, is what is good in the way of belief. This goodness is both instrumental and intrinsic. Moore’s paradox, that our esteem for truth constrains how we attribute beliefs to ourselves, supports this dual valuing of good supporting reasons. That is, it would be a conflicted thought for you to attribute a belief to yourself that you currently deem false. Moore holds that such a move would be a ‘contradiction in thought,’ because our beliefs, insofar as we believe them, are transparent to us—what it is to believe something is to take that content as true. Moore’s example is the thought “I believe it is raining, but it is not” (1942, 543). Even if it is true (and it could be, as Moore could believe falsely that it’s raining), it would nevertheless be incoherent to think in the first person. Our beliefs are internally related to how we think reality is, and because good reasons are our best tools for reflecting what’s real, it seems clear it would be silly to ignore them while still taking our beliefs as the sorts of things that we reflectively would endorse as true. We may believe these things, but under such a description (perhaps that I believe it is raining, but against overwhelming evidence that it is not), we view our beliefs less as contents we endorse, but as inclinations we have, more akin to symptoms or syndromes we wish we didn’t have. Our integrity as cognitive agents depends on this reflective test we can perform on our beliefs—whether, after having assessed how the belief was formed, how it is currently maintained, and what evidence counts for it and against it, we endorse the belief.

It is not just because we care for the truth of our beliefs that we care about those reasons. It is, second, that with this reflective endorsement of our commitments, we are caretakers of ourselves as thinkers. We are intellectually autonomous. The world is chock-full of crazy and stupid people, and they have the correlate habits of saying crazy and stupid things. Part of what it is to be a cognitively responsible person is to do one’s right best not to develop those habits, to manage one’s beliefs and one’s correlate assertions in a way that eliminates (or at least mitigates) the wildly false, unfounded things we may believe otherwise. We must have skills of cognitive management, and one skill of utmost importance is that of assessing the reasons one has for one’s beliefs, assessing the reasons others give against one’s commitments, and having, on balance, the best of the competing stories. We must sort the contents worthy of our commitment in a way that is reflective of the standing dialectical situation with regards to reasons for and against the issues. That is, in order to be justified, we must not only have a story to tell as to why the things we accept are reasonable from our perspective, but we must also be capable of addressing and answering not just the crazy and stupid, but also the reasonable yet wrong.

Justified commitments are those that, given what else we know, what reasons are on the table, and how the dialectical situation has shaken out, deserve our assent. Justifying reasons are, then, what we look for when we sort the various propositions worthy of our commitment from those that are not. Here is a rough list of beliefs I think I’m in a pretty good position to say I’m justified in believing:

LIST I: 2 + 2 = 4. I have hands. Freedom is valuable. Knowledge is valuable. One should not kill random people. My wife hates my beard. God does not exist.

In fact, I don’t just think the balance of reasons favors these beliefs; I think I am in my rights to say I know them true. But that’s another story as to how justification and knowledge are tied, and I will not be telling that story here.¹ Instead, the point of LIST I is to set out a number of cases of commitment that are the result of some diligent sorting. And the sorting yields another list, which are the propositions inconsistent with LIST I’s members and their dialectical competitors, a list of propositions the balance of reasons counts against:

LIST II: 2 + 2 = 5. I have lobster claws, not hands. Freedom is not valuable, because it’s dangerous. Knowledge is worse. When you get a new samurai sword, you need to test it out, and random people on the road are the best for that. My wife loves my beard. God not only exists, but has plans for me, given that I denied His existence on LIST I.

The point of making LIST II is that the sorting work we perform by attending to justifying reasons has a contrastive² element with the contents of our thoughts, and one of the reasons why that is the case is that justification should be something that can be shared or shown to others who’ve missed the boat. There are folks, some crazy and some quite reasonable, that hold commitments from LIST II, and it would trouble me mightily if I not only couldn’t defend my views from them but also could not say something about why justification doesn’t favor theirs. Part of what it is to be justified in believing that 2 + 2 = 4 and that I have hands, it seems, is not just to understand and have epistemic support for those beliefs, but to be able to answer critical questions about the 2 + 2 = 5 theory and the lobster-claw hypothesis. Those are silly views, to be sure, and what makes them so is the fact that they are easily refuted. Further, people who reject the value of freedom and knowledge often make principled cases (e.g., freedom makes free riders, and knowledge extinguishes wonder), but those cases don’t work. At least in the knowledge case, I’d want to say: does knowing that knowledge extinguishes wonder itself extinguish wonder? I don’t think so … and so we have a counterexample. Regardless, I’m sure an extended argumentative exchange wouldn’t go their way. Same goes for the Samurai who thinks he, in fact, must chop up a random villager to ensure his sword battle worthy. That’s a moral error, and one that, if I’m justified in my belief that one shouldn’t kill random people, I should be able to make the reasons in favor of that explicit. Further, someone may think my wife actually loves my beard. Perhaps this person has inferred, reasonably I might add, that because I look so good with it, my wife must love it. But this person does not know that my wife, despite how distinguished I look with the beard, doesn’t like the fact that it is very bristly and has the tendency to collect mayonnaise. And finally, traditional theism is a serious view held by many serious people, and they’ve got some very impressive arguments. But atheists’ arguments are better, and what it is to be an atheist (or, on the other hand, a certain stripe of theist) is to assess oneself as being on the winning side of the ongoing argumentative exchanges on the issue. Justification, in those settings, is a dialectical accomplishment, and only when one has achieved that does one have the right to have those commitments.

There’s another list, though. LIST I and LIST II capture the contrastive dialectical element of justification, which emphasizes the content one is justified in accepting or rejecting. But there is a second contrastive element to justification—that between being justified and not being justified. And so:

LIST III: The number of stars is even. Yellow is a happy color. There are multiverses. The Good and the Beautiful are conceptually dependent on each other. Cleopatra sneezed three times on her fifth birthday.

These are contents I do not take myself to have justification to believe true or false. The number of stars is an empirical question to which we do not yet (or will ever, as far as I can gather) have any answer. And if the number of stars is infinite, it’s neither odd nor even. I’m only inclined to say yellow is a happy color, but that’s about as good as my reasons get, and there are others with contrary inclinations. I am not even sure about what multiverses are; so, I certainly can’t be justified in any sense that matters to me in believing there are some. Though it sounds plausible that moral and aesthetic questions are separable, they nevertheless seem regularly to inform each other (especially moral judgment directing aesthetic judgment). The point here is that an eye to justification helps sort out three categories: the things we have a right to believe, the things we rightly disbelieve, and those about which we rightly withhold belief. Justification, if it is going to mean something to us, if it is going to be the kind of property of commitments that we value and can be of use to us, needs to be the kind of thing that we look to when we perform these sorting roles with commitment. Justification is an objective of proper cognitive management, dispute resolution, and intellectual self-possession.

Out of this rough story of justification, a few desiderata of a theory of justification worth having emerge. There are patterns in the way we pursue justification that show how justification plays a normative role in our cognitive lives.

Justification is subject-relative. Some people have the justification and others do not. This depends on what evidence people have, how well they’ve thought it through, and so on. So, for example, in the aforementioned beard case we have people who have good reason to sort the beard belief differently, but because of defeating reasons for their commitments, once they talk with me (or my wife), they should change their minds.

Justification is truth-directed. The justification that regulates belief is a matter regarding whether the subject in question has a right to believe a proposition true or false. Justifying reasons count in favor of a belief’s truth, not its preferability or other non-truth-directed concern.

Justification is sharable. We can hold each other to account for things about which we disagree, and we can make explicit what criteria were used in sorting our beliefs. Arguments are the basic model for these sharings.

Justification is dialectical. The justification one has emerges from and can be changed by argumentative exchanges with others. The better one performs in responding to representatives of competing views, the better off one is. The worse one does, the worse one’s justification.

Justification comes in degrees. One way a subject can have more justification is to have more positive evidence. Correlately, one’s justification can be weakened by contrary evidence.

Justification is fallible and correctable. It is possible to be justified in believing something that is in fact false. This is why it is important that subjects have their reasons so that they can update their beliefs as new evidence comes in. Discussion and weighing arguments are the collection of supporting evidence and contrary evidence for the views under consideration.

Justification is required for intellectual integrity. Caring about, checking, revising, improving, and ultimately endorsing one’s beliefs is constitutive of cognitive autonomy. The pursuit of justification is the life of the subject with intellectual integrity. We want to have lives that are livable and endorsable from the inside, and a condition for that livability is that we take ourselves to be believing by our own best lights. Otherwise, our beliefs are not commitments we endorse, but are intellectual symptoms we don’t.

1.2 THE PRINCIPLE OF INFERENTIAL JUSTIFICATION

In the previous section, I outlined a few desiderata for an account of justification worth having. They were axiological criteria for a satisfying story about justification worth the appellation. They are the reasons why we care whether or not we are justified. The basic thought is that if we had a story about justification that shows we have it, but it is not the sort captured by the desiderata, we would naturally think a kind of bait and switch has occurred. Now, sometimes the old switcheroo is appropriate for important concepts. For example, the switcheroo is the right kind of move for a notion like freedom. Metaphysical freedom in a libertarian incompatibilist sense is not only not probable for entities like us, but many of the reasons why we would want it are also reasons to be happy with compatibilism (you keep the idea that the decisions come from you, you keep the thought that punishment and accountability are directed at the entity that performed the act, etc.). The issue with ...

Table of contents

• Routledge Studies in Contemporary Philosophy
• Contents
• Preface
• Introduction
• 1 The Regress Problem
• 2 Infinitism Defended
• 3 Meta-Epistemic Varieties of Epistemic Infinitism
• 4 Foundationalism, Infinitism, and the Given
• 5 Argumentation and Antidogmatism
• Notes
• Bibliography
• Index