If there’s permanent dissensus on an issue, we’re not willing to trust those philosophers who assert with full confidence that they have knowledge (they have justified the truth of their substantive philosophical theses with knock-down arguments), and explain the dissensus by saying that their interlocutors are unable to see the compelling force of their arguments. In Chapter 4 of my book, I sarcastically call these thinkers “I’m the only one” philosophers and depict them as figures who are unable to put themselves into others’ perspective, are wholly insensitive to the epistemic attraction of others’ pre-philosophical convictions, resistant to the convincing force of others’ arguments, and suffer from epistemic blindness.

Jennifer Lowell reconstructs the line of thought in this chapter precisely (and largely concurs with it), but she isn’t completely satisfied with my solution. As she puts it, “[t]hough this chapter presents a compelling assessment of and farewell to the epistemic unjust ‘I’m the only one’ view, I wonder if this chapter’s dismissal of what Tőzsér rightfully identifies as the wrong approach to philosophy’s failure is also a bit guilty of a failure to acknowledge possibilities for responding to them.” So, let me take another shot to the problem in a wider context, picking up the thread at the introductory line of thought in Chapter 2 of my book.

*   *   *

Let’s suppose that Tom and Steve disagree with each other. Tom thinks that p is true, while Steve thinks that not-p is true. There may be two things in the background of this disagreement. Either it is that while one of them has knowledge, the other is mistaken, or it is that neither Tom nor Steve has knowledge (at most one of them has true belief). Generally speaking, we can explain all disagreements with two things. Either it is the case that one of the participants of a debate knows something while others don’t know it, or it is the case that none of the participants know the thing under debate (at most one of them has true belief). In a word, we can safely say that if there is permanent dissensus in a question, then either all interlocutors of the debate save one or all of them without exception have beliefs about the subject of the debate that have been shaped by factors that don’t track the truth.

Now, if it doesn’t follow from the fact of given dissensus that nobody has knowledge, then the question arises: why do we strongly tend to think that none of the sides to the debate has knowledge? Why are we reluctant to accept the possibility of individual (or “small-community”) philosophical knowledge—why do we think that philosophical knowledge can only be consensual (or collective)?

We can conceive three scenarios in which S knows some substantive philosophical truths. In the first one, S has a compelling argument for the truth of p—an argument whose conclusive (truth-conducive) nature is questioned by no rational person. An argument which takes away our cognitive freedom for believing otherwise than what the conclusion of the argument asserts. The second case is when S’s argument for the truth of p doesn’t have compelling force, but S’s belief p is in a proper relationship with the truth at issue—p is caused, in an appropriate way, by a fact or state of affairs that makes p true. That is, S’s belief p is produced by a reliable (truth-conducive) belief-producing process, but—and this is what distinguishes this case from the previous one—from her subjective perspective, S doesn’t have access to every factor which is responsible for the reliability (truth-tracking feature) of her argument. In the third case, S’s argument for the truth of p is not a knock-down one, but beyond/above her arguments, she has some further private evidence for the truth of her belief. This evidence is private because it is ineffable (S cannot build it into her argumentation), and yet it reliably indicates to S that her reasoning tracks the truth. This third scenario is akin to the first one and differs from the second in that from her subjective perspective, S has access to every element of her argumentation—her justification is internalist, although she’s unable to make every element of it public. And it is akin to the second one and differs from the first in that the argument itself doesn’t have compelling force—the credibility of the conclusion depends on something that is not part of the argument itself.

As for the first case: if S has knock-down arguments for the truth of p, but there’s permanent dissensus on the issue among philosophers, then it means (it cannot mean anything else) that those who disagree with S don’t understand S’s arguments—they’re unable to see their conclusive nature. As for the second case: if S knows that p in an externalist sense, but there’s permanent dissensus on the issue among philosophers, then it means (it cannot mean anything else) that those who disagree with S are unable to recognize S’s epistemic excellence—the fact that S is a person with a “blessed” belief-producing process. And as for the third case: if S knows p in the sense that she has some further reliable private evidence for the truth of p over and above her arguments, but there’s permanent dissensus on the issue among philosophers, then it means (it cannot mean anything else) that those who disagree with S are unable to recognize S’s exceptional insight force—the fact that S is a person who has come into possession of such experience which show her the truth.

To sum up, although we can conceive three epistemic scenarios in which a philosopher has substantive knowledge, we don’t put our trust in her in light of the permanent dissensus. Nor do we trust those who are convinced that they have gotten into the possession of arguments with compelling force. Nor those who see themselves as epistemically “favored”—the Holy Spirit has restored their minds, restituting their innate but corrupted sensus divinitatis (e.g. Plantinga 2015); their moral greatness leads to their epistemic greatness (while the flaccid and non-autonomous spirits are empiricists, the free and autonomous ones are idealists, see Fichte 1797/1994) or they’re just the lucky winners of the epistemic lottery. And, we don’t trust those, either, who vindicate themselves of philosophical knowledge by citing their private evidence—be it some phenomenologically vivid (visionary or ecstatic) experience or a quiet signal (inner voice or inner compass), which “assures” them of the rock-solid fundamentals of each premise and step in their argument. The question that arises again: why don’t we trust any of them?

*   *   *

I think this is what most of us have on our mind:

It is true indeed that it does not follow from any permanent philosophical dissensus that nobody has knowledge—every dissensus can be just as well explained by saying that S knows something that the others don’t know as with saying that nobody knows anything. It is also true that we cannot rule out any of the above three epistemic scenarios—for how on earth could we justify this decision of ours beyond all doubt? At the same time, we still have more reason to interpret the fact of permanent dissensus to mean that there’s no individual knowledge than to interpret it to mean that while some know some things, others are mistaken. In a word, we have more reason to distrust those who claim to know substantive philosophical propositions than to trust them.

Let’s take a look at the first scenario. Why do we consider the fact of permanent philosophical dissensus as a sign that philosophers haven’t come up with knock-down arguments? It is because if philosophers have not been able to come up with philosophical arguments in which no points could be disputed by other philosophers (and they haven’t), then we have every reason to suspect that they have not produced any compelling philosophical arguments—similarly, if the community of mathematicians judges a mathematical proof not to be conclusive, then we have every reason to suspect that it actually isn’t. In other words, insofar as there were compelling philosophical arguments, then we would recognize their compelling force—similarly to the community of mathematicians recognizing the conclusive force of conclusive mathematical proofs. Or, as Peter van Inwagen puts it: “[i]f any reasonably well-known philosophical argument for a substantive conclusion had the power to convert an unbiased ideal audience to its conclusion (given that it was presented to the audience under ideal conditions), then, to a high probability, assent to the conclusion of that argument would be more widespread among philosophers than assent to any substantive philosophical thesis actually is” (2006, 52–53).

Let’s now turn to the second and the third scenarios. Why do we consider the fact of permanent philosophical dissensus as indicating that there are no philosophers who are epistemically favored and have special insight force? The reasoning is similar: if there were such philosophers, then we’d recognize them. We’d accept them as epistemic authorities and would give credence to their word—we’d believe them that things stand the way they say they stand. (Just like we’d believe God or angels whatever they say). Of course, it isn’t easy to precisely specify the conditions under which we’d recognize Mary or Esther as epistemic authorities—what matters is that we’d recognize their epistemically privileged position. Just as we recognize someone as a good chicken sexer (as he regularly asserts of male chickens to be male and female to be female), we’d recognize (even if it’d take much more complex weighing) those philosophers who regularly hold true propositions to be true (due to their epistemically favored status or special insight force) and regularly hold false propositions to be false.

I think that considerations like the above are on our mind when we want to give reasons for why we don’t believe in individual philosophical knowledge. Our distrust rests on our trust that we’re able to recognize compelling arguments (we can distinguish them from non-compelling ones) and we’re able to recognize true epistemic authorities (we can distinguish them from false prophets).

*   *   *

However, we may have some residual bad feeling about the foregoing. For, after all, what if we’re wrong, and our belief in this ability of ours is false? What if there are knock-down philosophical arguments and philosophers who are epistemically favored or have special insight force, and it’s just that we blindly pass them by? As these doubts may arise in us from time to time, we need to say something about them—preferably something reassuring.

Here’s what I can say by way of reassurance. Let’s suppose that we don’t have the ability to recognize arguments with compelling force (whose conclusions we must believe) and epistemic authorities (whose words we must heed). If this is how things stand (we don’t have this ability), then our unease and doubt are meaningless: we’re not to blame for not having recognized the ones with compelling force from among the lots of arguments, and the epistemic authorities from among the lots of philosophers. Our unease and doubt are meaningful only if we have this ability—if we’d recognize knock-down arguments and epistemic authorities (provided there were any).

In another approach, we’re entitled to hold the belief in this ability of us because the call for giving up, suspending or just tempering our belief in this ability has normative weight (regardless of uttered by others or a doubtful inner voice) only if we have the ability that we’re called to give up. Namely, only if we’re able to recognize an argument as one whose conclusion we must believe, and a person as one whose words we must give credence to. In the opposite case, if we’re unable to do so, we are free to believe anything.

Thus, there are two possible cases. Either we’re unable to distinguish between knock-down and non-knock-down arguments as well as between true epistemic authorities and false epistemic prophets—in this case, we’re not open to criticism by others or by ourselves if we continue to hold our belief that we’re able to do so. Or we’re able to distinguish between the above mentioned—and that’s why we’re entitled to believe that we do the right thing when we maintain distrust in those arguments and assertions whose conclusiveness and authority are surrounded by permanent disagreement. In a word, there can be no epistemic situations in which our belief in this ability of us could be shaken—our doubt has epistemic-normative weight only if the object of doubt is true. If you wish, we are entitled to hold this belief under all circumstances—this belief of ours is a hinge belief that is given to us gratis.

References

• Fichte, J.G. (1797/1994). “First Introduction to the Wissenschaftslehre.” Introductions to the Wissenschaftslehre and Other Writings. Indianapolis: Hackett Publishing.
• Plantinga, A. (2015). Knowledge and Christian Belief. Grand Rapids: Eerdmans Publishing.
• van Inwagen, P. (1996). “It Is Wrong, Everywhere, Always and for Anyone, to Believe Anything upon Insufficient Evidence.” In Faith, Freedom and Rationality, edited by J. Jordan and D. Howard-Snyder, 137–54. Savage, MD: Rowman & Littlefield.

I’d like to react to two points in László Bernáth’s review. The one is that according to him, the pessimistic picture I paint of philosophy (it’s a failed epistemic enterprise—its truth-seeking and justificatory tools are inadequate and unsuitable for establishing substantive philosophical truths) rests on a merely inductive basis. As he writes, “this induction heavily relies on the presupposition that the tools of philosophy will be the same in the future as it were in the past,” and somewhat rebukingly adds that I don’t “give a detailed picture on what the tools of philosophy are and why they should be the same forever.” The other is that according to him, I don’t explain properly “why we should care so much about the insufficiency of philosophy [including my intellectual breakdown].” By his account, “this ’should’ can have a normative weight only if there is a robust ethics of belief in the background,” and he goes on saying, also not to praise, that “the book has close to nothing to say about [this background].”

*   *   *

I think that Bernáth’s first criticism is only partly rightful. On the one hand, I clearly distance myself from induction (pp. 203–6). On the other hand, even if only as a hint, I say something about the nature of philosophical theorizing and through it about the reason for the failure of philosophical knowledge (pp. 72–5). But then again, I don’t go into this issue in detail.

There are several explanations of the reason for permanent philosophical dissensus (PD) and the unresolvedness of philosophical problems—let’s see a list of the most important ones.

(1) Philosophical problems are meaningless—they are actually the products of an appearance-creating mechanism (for example, the misunderstanding of language).

(2) There are no objective philosophical truths—things are not in any way independent of our mental and linguistic representations.

(3) It is only science that can solve theoretical problems—philosophy, however, is not a science.

(4) Philosophical problems are extremely complex—no one has succeeded in reassuringly clarifying the fundamental conceptual connections, and those scientific results that are necessary for solving philosophical problems have not yet been achieved.

(5) We are closed off from the solutions of philosophical problems—our epistemic equipment is unsuitable for creating those concepts that are necessary for solving philosophical problems.

(6) Every philosopher’s beliefs (even those of the most prominent ones) are shaped by bias factors—these factors are unreliable and do not track the truth (for example, the philosophers’ temperament and moral characters, their social and cultural environments etc.)

We can categorize these explanations in more than one way. Firstly, while according to (1) and (2), there’s nothing that could be known but not actually known by us, according to (3), (4), (5) and (6), there are plenty of things that are not known by us but could be known. This is because while according to (1) and (2), philosophical problems are not genuine, and so nor are philosophical disagreements (the debate is about meaningless sentences in the first case, and about sentences with no objective truth value in the second case), according to (3), (4), (5) and (6), philosophical problems are meaningful, the sentences expressing them have objective truth values, and so the dissensus about them is genuine. In what follows, I won’t discuss (1) and (2).

Secondly, while according to (3) and (4), there are no obstacles in principle to solving philosophical problems and so we can be optimists, according to (5) and (6), there are obstacles in principle to solving philosophical problems, and so we cannot be optimists. Those accepting (3) may argue that the sciences themselves, or perhaps the kind of philosophy that is based only on the best scientific theories (on solid scientific results) will be able to answer philosophical questions in the future—including the big “ones.” Those accepting (4) may say that we should wait until the beginning of the “Epistemic End of Days” (MacBride 2014, 231)—we’ll be safe to pursue philosophical theorizing once we clarify the relevant conceptual relations and/or the sciences deliver those results that are key to the solutions of philosophical problems. Now, as I cannot commit myself neither to (3) nor to (4) on account of my pessimism (I cannot say that we haven’t yet solved the problems of philosophy but we have all the reasons for hope), the ones that remain are (5) and (6).

Thirdly, although both (5) and (6) see the reason for the unresolvedness of philosophical problems to lie in our deficient epistemic equipment, it is important to see the difference between them. According to (5), philosophical problems are unresolved because the constitution of our minds is adapted to the Stone Age environment, so our cognitive equipment doesn’t enable us to solve philosophical problems—our misfortune stems from the fact that “[w]e can envisage questions that require conceptual and theoretical resources that exceed the contingent limits of [ours]” (McGinn 1993, 8). I don’t sympathize with this proposal, partly because if our philosophical abilities were mere spin-offs of the process of natural selection, then the question arises as to “why the same does not apply to the ability to do abstract mathematics or highly theoretical science” (Chalmers 2015, 28), and partly (mainly) because this explanation says nothing about the nature of PD. The unresolvedness of philosophical problems means that we don’t know which of the competing theories is true—yet, in the spirit of (5) it is difficult to account for the permanent fronts or fault lines concerning philosophical problems.

Thus we’re left with (6) as the most promising explanatory strategy. However, I’m not completely satisfied with this proposal, either. Indeed, personal (and at once distorting) factors must have a role in the explanation (after all, if not these, then what else could explain PD?)—but this is just one side of the coin, not the complete picture.

My explanation rests on two pillars. One is the nature of philosophical problems, and the other is the personal characteristics of philosophical theorizing. According to this, each philosophical problem is given to us in the form of propositions that are epistemically attractive (seemingly true when considered in themselves), but jointly constitute a set of inconsistent propositions (or at least they can be reconstructed as such without loss of content), and the factors playing the ultimate and crucial role in shaping our philosophical beliefs (and so in our choice of the proposition we want to give up from the set) are unreliable and don’t track the truth.

Let’s take the first pillar and suppose that Esther and Sophie hold opposed positions on the mind-body problem. While Esther is a physicalist (an adherent of the type-identity theory), Sophie is anti-physicalist (being a substance dualist). Nevertheless, they agree on certain things. They agree that the mind-body problem emerges in consequence of the joint inconsistency of certain propositions—moreover, they also agree in identifying the same propositions. Let’s suppose that both of them think that the mind-body problem arises as a consequence of the following inconsistent propositions.

(1) Conscious experiences are not physical events.

(2) Conscious experiences can cause physical events.

(3) Every physical event has a sufficient physical cause.

(4) Human actions (as physical events) are not overdetermined by conscious experiences and physical events.

Now, if they agree on all of the above, then they must also agree that the disagreement between them is to be accounted for by the fact that the propositions they hold to be false of this inconsistent quadruple are different. While both hold (2) and (4) to be true, Esther holds (1) but Sophie holds (3) to be false.

The reason why they cannot convince each other of which of these inconsistent propositions must be held true or false is that each of these propositions is epistemically attractive. Each of them has “juice”—there are considerations in favor of each of them. Esther has to admit that (1) is epistemically attractive—that the gap-intuition is “vigorous” and “tenacious” and can even be “pumped” further and further (think of the variety of increasingly sophisticated conceivability arguments). And Sophie has to admit that (3) is epistemically attractive—that there’s good reason to believe that the most successful epistemic enterprise of mankind is physics, and “[t]he success to date of current physics in finding sufficient physical causes for physical effects […] provides inductive evidence that all physical events […] have sufficient physical causes” (Melnyk 2003, 288). In a word, both of them have to admit that they precisely understand why the proposition they deny is so attractive to the other one. Moreover, I think they also have to admit that they themselves are not completely immune to the epistemic attractiveness of the proposition they deny—were it not inconsistent with the rest, they, too, would tend to hold it to be true. That is, no matter how they decide and which proposition they hold to be false, they do so against their already existing epistemic affection.

Now, as I see it, one can extend the above characterization to every philosophical problem. In the final analysis, every problem of philosophy is a set of inconsistent propositions—or else it wouldn’t be a problem to begin with. And every philosophical problem is a set of inconsistent propositions such that each element of this set is epistemically attractive—or else the dissensus wouldn’t be permanent.

As for the second pillar of the explanation: why is it that Esther and Sophie give up different propositions? At first blush, we have an obvious explanation: because the rest of their beliefs are different. If Esther is an atheist, a strong determinist on free will and has perdurantist views about the persistence of physical objects, then it is no wonder that she is also a physicalist—this is how her worldview can be coherent. And, if Sophie is a theist, an adherent of agent causation concerning free will and a presentist about the persistence of physical objects, then it is no wonder that she is also an anti-physicalist—this is how the big picture fits together for her.

However, this is not the ultimate explanation. The ultimate explanation is that the factors that “price in” Esther’s and Sophie’s epistemic affections (i.e. the greater or lesser extent to which they are epistemically attracted to concrete propositions) and shape or determine their full-fledged philosophical beliefs are such that shouldn’t do so—as they don’t track the truth and have nothing to do with it.

In my opinion, this is the case with all of us concerning every philosophical problem. During philosophical theorizing, we usually “make decisions” about which of our epistemic affections we ought to trust, and which we oughtn’t. And, in our better moments, while self-reflectively monitoring ourselves, we’re able to glimpse down to a certain depth—to realize those distorting mechanisms that are responsible for our epistemic character and for our having precisely those substantive philosophical beliefs that we actually have rather than that we don’t have.

To sum up, the ultimate failure of philosophical theorizing can be traced to deficiencies in our epistemic equipment. We are creatures with epistemic affection for jointly inconsistent propositions (this is a part of conditio humana)—creatures who (at least in the case of philosophical problems) are unable to reliably filter the deceptive ones from among their epistemic affections.

*   *   *

Bernáth’s second objection is valid. I admit that my discussion is not sufficiently deep of why we should care about the fact of philosophical disagreements—why we should account for the epistemic status of our philosophical beliefs.

Well, to be able to assert that “we should,” I certainly have to commit myself to three substantive theses (which belong to the ethics of belief). The first is that there are values—at least in the low-cal sense that it is more valuable to know something than to err or just hit on the truth; justified beliefs are more valuable than unjustified ones. The second commitment is that our epistemic equipment is not altogether bad—no such ultra-skeptic scenario obtains in which we cannot at all trust our ability to distinguish between our truth-conducive and our not truth-conducive cognitive processes. The third one is that there are epistemic duties—we must do all we can in order to see clearly what we may, what we have to and what we must not believe.

So, it is completely certain that I must commit myself to these three theses. Because I don’t have any arguments with compelling force for their truth (not even stronger arguments than the ones for their negations), at most I can show why I’m entitled to stick with them.

I borrow the core idea from hinge-epistemology, which says that there are beliefs which don’t need justification due to their special contents—they’re outside of the circle of those beliefs that we have to justify. We’re entitled to hold these beliefs under all circumstances—for if we didn’t have these beliefs, then it would be impossible for us to attain any epistemic goal. According to hinge-epistemologists, one such hinge belief is that there are other minds and there’s an external world. Now, I think that the aforementioned three beliefs are hinge beliefs, too.

Why? (1) The belief that “justified beliefs are more valuable than unjustified ones” is a hinge belief (and here’s why I’m entitled to hold it) because the only case in which someone might rightfully and strongly criticize this belief (saying “Give it up!” or “Suspend it!” is when this belief is true—in the opposite case, this criticism and the call to discard it have no normative weight, since every belief is equally valueless. (2) The belief that “at least sometimes we’re able to distinguish good arguments from bad ones and the veridical instances of seeming-to-be-true from the delusive ones” is a hinge belief (and here’s why I’m entitled to hold it) because the only case in which someone might challenge this belief (expecting me to give it up or suspend it) is when this belief is true—in the opposite case, I, as the addressee of the objection, am not in the kind of situation in which I can properly decide whether the interlocutor’s argument is good or bad. (3) The belief that “there are epistemic duties” is a hinge belief (and here’s why I’m entitled to stick with it under any circumstances) because the only case in which someone might call me to give it up is when this belief is true—in the opposite case, if there are no epistemic duties, I don’t have to believe or not believe anything at all.

In a word, a belief b is a hinge belief if the truth of b is necessary for epistemically (in a normative sense) criticizing the holding of b. The above considerations help us see that it can be held by anyone under any circumstances. There are no such epistemic situations in which someone could come up with a criticism with the effect that we’d have to give it up or suspend it—we can hold it gratis.

It’s an interesting question as to what further beliefs may claim the status of hinge beliefs. According to Bernáth, the belief in free will is a hinge belief (Bernáth and Paár 2024). As for myself, I rest content with the above two plus three ones.

References

• Bernáth, L. and Paár, T. (2024). “Responsibility First: How to Resist Agnosticism about Moral Responsibility.” Dialectica (forthcoming)
• Chalmers, D. (2015). “Why Isn’t There More Progress in Philosophy?” Philosophy 90 (1): 3–31.
• McBride, F. (2014). “Analytic Philosophy and its Synoptic Commission: Toward the Epistemic End of Days.” Royal Institute of Philosophy 74: 221–36.
• McGinn, C. (1993). The Limits of Inquiry. Oxford: Blackwell.
• Melnyik, A. (2003). A Physicalist Manifesto. Cambridge: Cambridge University Press.

Before reconstructing Eli Minor’s objection and presenting my response to it, let me outline the context very briefly (see Chapter 5).

If a philosopher doesn’t believe that knock-down philosophical arguments are possible, she must settle for a more modest epistemic goal. Namely, to develop theories that are in harmony or equilibrium with her pre-philosophical convictions, and to maintain the resulting theories intellectually—to defend them against various objections and to clean them of the arising internal contradictions. But in the absence of knock-down arguments, doubt can overwhelm one from time to time: “Can I firmly trust in the truth of my philosophical views merely on the basis of the fact that they are in equilibrium with my pre-philosophical convictions and no compelling arguments can be brought up against them?”—“Can I take epistemic responsibility for the truth of my philosophical beliefs if my pre-philosophical convictions on which these beliefs are based are unjustified, and the most I can say for them is that I cannot abandon them without damaging my personal integrity and/or my cognitive household?” And this doubt can be followed by what I call “epistemic schizophrenia”: “On the one hand, I clearly see that a merely egocentric justification of my views doesn’t entitle me to seriously and sincerely believe in their truth, and on the other hand, I cannot hold my beliefs to be irrational and suspend them—I keep sticking to them.” There are philosophers who aren’t overcome by doubt and epistemic schizophrenia, because although they don’t think that their arguments are compelling, they do think that they have some further private evidence for the truth of their philosophical beliefs—and referring to IT, they attribute privileged status (extra weight) to their own views. However, I don’t think that citing this private evidence can actually help them—its reliability cannot be justified in any way.

Minor takes aim at the latter point. According to him, “[w]hen [I] criticize the reliability of first-person ‘inner voice’ as lacking any ‘right criteria’ for proper truth-tracking […], [I] assume a conversationalist view of mind and rationality which takes communicability of reasons to be the hallmark of epistemic property.” That is, I presume that “any suitable criteria for discriminating good knowing from bad knowing will turn to out to be an essentially public affairs, a matter of conversation” (italics in original), and that “[c]onversation […] is the ultimate tribunal for what shall come (or not) to pass for properly justified true belief”—ultimately “this is why ‘private justification’ rings as an oxymoron” for me. In a word, my “criticism of ‘phenomenological evidence’ as ‘inner monologue’ goes through only if we assume that social justification is the only epistemically legitime kind of justification.”

However, if we break with this “roughly conversationalist view of mind and rationality” and interpret the phenomenology of experiencing private evidence not as an “inner monologue” or “inner conscience” but, as is more customary, as “what it is like to be a such and so,” then according to Minor, my “criticism of private evidence as phenomenological evidence might be problematic because certain people and groups of people do occupy epistemically privileged positions in relation to some others” (italics in the original). As he puts it: “[i]n virtue of being located in the social space in a certain way, one thereby has certain experiences which confer legitimate knowledge and evidence to that knower which, in principle, inaccessible to those who do not occupy that same position” (italics in original). And, finally, he concludes that “there may be evidences beyond conversation and these are ones that phenomenological thinking seeks to vindicate,” and “[p]henomenological thinking properly understood might provide us with a richer, more expansive conception of rationality and evidence beyond what passes muster in the normative space of giving and asking for reasons” (italics mine).

As far as I can see, I disagree with Minor on almost everything—I think his criticism misses the mark, and his phenomenology-based optimism is unjustified.

*   *   *

First, I don’t think that Minor gains anything by interpreting the phenomenology of private evidence not as an “inner conscience” but as “what it is like to be a such and so.” It makes no difference that he phrases the question not as “Should we listen to the ‘inner voice’ that assures us that there is a reliable cognitive process leading to our beliefs?” but rather “Should we give credence to our ‘phenomenal experience’ in which the cognitive process that leads to our beliefs seems reliable?” The question is the same anyway: “Should we trust these private evidences or should we not?” In other words, regardless of whether we call it an “inner voice/conscience” or a “seeming-to-be-true,” we have to decide whether or not we can trust the veridical nature of the experience in question.

Secondly, I didn’t accidentally begin my answer by describing the context. According to it, S doesn’t refer to private evidence at the beginning of philosophical theorizing, but after he has already developed a theory and has gone beyond its intellectual maintenance—he has packed all the evidence available to him (even those with phenomenological roots!) into his arguments for his theory and against the objections. After all this, S cites some further evidence (which is uncommunicable with others and thus cannot be incorporated into his arguments) in an attempt to prevent doubts and convince himself that his theory (equilibrium) is nevertheless privileged. So, I am not arguing against the reliability of phenomenological insights (and appearances) in general, but it is in relation to this particular seeming-to-be-true that I claim that we cannot trust it, because there is no criterion for correctness or veridicality in this case.

*   *   *

My main concerns are with Minor’s optimistic vision—especially the element that gives phenomenology a privileged role: “[t]he singular insight of phenomenological thinking is that first-person experience is philosophically indispensable and irreducible because we have reason to grant conceptual priority to experiences over reasons” (italics mine).

Take the problem of perception. Our veridical perceptual experience has two, usually highlighted, characteristics, both of which are derived from phenomenological reflection. The one is that, during perceptual experiences, we’re only aware of the object(s) that exist independently of our actual perceptual experience. For example, when we perceive a red tomato, the only object perceived is the red tomato whose existence isn’t dependent on our actual perceptual experience. The other is that during our perceptual experiences, the perceived objects are presented to us in a robust way—the nature of our perceptual experience is (partly) determined by the nature of the object(s) being perceived. For example, when we perceive a red tomato, then our experience is (partly) determined by the way the red tomato actually is (see Crane 2005).

Thus, during veridical perception, the perceived entities are presented to us and exist independently of our current perceptual experience—the phenomenal character of our perceptual experiences is shaped (at least partially) by the experience-independent objects we perceive. These two phenomenological characteristics form the distinctive feature of perceptual experiences—they make our perceptual experiences different from other types of our conscious experience.

Now, if we want to commit ourselves to a theory of perception that takes both phenomenological characteristics of perceptual experience at face value, we face the following well-known difficulty. Hallucinations are possible (we may have an experience that is indistinguishable from a veridical perception of an object from a first-person perspective, but in which there’s nothing we perceive)—and insofar as our veridical perceptual experience and our hallucination are subjectively indistinguishable to us, they’re conscious experiences of the same type. Here’s a reconstruction of the argument from hallucination:

(1) The thesis of the possibility of hallucinations: S may have experiences that are indistinguishable from the veridical perceptual experience of an experience-independent object O from a first-person perspective, but where S doesn’t perceive O (O isn’t presented to S as an experience-independent object).

(2) Common kind thesis: if S’s hallucinatory experience is indistinguishable from S’s veridical perceptual experience from a first-person perspective, then S’s hallucination and the corresponding veridical perception are of the same type of conscious experience.

Therefore:

(3) S’s veridical perceptual experience of O ≠ the presentation of O to S as an experience-independent object.

This means that if the argument from hallucination works, then no theory of perception can be given that takes both phenomenological characteristics of perceptual experience at face value. Either, we have to deny that during veridical perception, we are aware of experience-independent objects, as proponents of the sense-data theory do—saying that during our perceptual experiences we are aware of mental/mind-dependent entities. Or, we have to deny that the phenomenal character of our veridical perceptual experience is (partly) determined by the nature of the object(s) of experience being perceived by us, as proponents of the intentional theory of perception do—saying that the contents of our perceptual experiences (like those of beliefs and thoughts) merely represent the experience-independent world.

Of course, if one wants to take both features of the phenomenology of perceptual experience at face value at all costs (wouldn’t like to withdraw one’s confidence of neither), one can do so. Either [contra (1)] one must claim that hallucinations subjectively indistinguishable from veridical perceptual experiences are not actually possible, or [contra (2)] one must claim that they are possible, but that they are different types of conscious experiences, despite their subjective indistinguishability.

Since the denial of (1) doesn’t seem promising, those who reject the conclusion of the argument from hallucination deny (2). This is the strategy of the disjunctive theory of perception: “[d]isjunctivism about perceptual appearances […] is a theory that seeks to preserve a naïve realist conception of veridical perception [one that is in full harmony with the phenomenology of perception] in the light of the challenge from the argument from hallucination” (Martin 2006, 215).

However, the commitment to disjunctivism comes at a price. This is because if we claim that veridical perceptual experiences and those hallucinations that are subjectively indistinguishable from them are different types of conscious experience, then we must also reluctantly claim that from the first-person perspective [from the phenomenological point of view] we do not have access to the factors that determine the type of conscious experience in question. In other words, as disjunctivists, we must claim that the type of our conscious experience is not determined by factors that we can access from the first-person perspective [from the phenomenological point of view]. In other words, we need to radically curtail our self-awareness—we need to go against our deep-seated belief that we have privileged access to our phenomenal content.

Furthermore, we must also say something about the nature of hallucinations. We must come up with an interpretation that allows us to take the phenomenology of veridical perceptual experience at face value, and to do so despite the fact that hallucinations and veridical perceptual experiences are indistinguishable from each other from a subjective perspective. And that doesn’t come free either. We have to commit ourselves “to saying that, at least when it comes to a mental characterization of the hallucinatory experience, nothing more can be said than the relational and epistemological claim that it is indiscriminable from the perception” (Martin 2004, 72)—we have “to accept that although the hallucinatory situation lacks the property in question, one cannot know that the property is absent simply by reflection on this situation and hence that it at least seems to be present” (Martin 2004, 66). In a word, we have to say that “hallucinations lack phenomenal character”; to accept “eliminativis[m] about hallucinatory phenomenal character” (Fish 2009, 93)—we have to bite the bullet saying that although there isn’t anything it is like to hallucinate, a hallucination (taking place in complete darkness) may have the causal patterns as the corresponding veridical experience: it can produce “the same beliefs or judgement that a veridical perception of that kind would have produced” (Fish 2009, 94).

*   *   *

Well, by appealing to the problem of perception, I wished to illustrate that phenomenology—pace Minor—is not privileged (and even much less protected). Our phenomenological insights (or rather, our “seemings-to-be-true”) are legitimized (in Kant’s words) before the “Tribunal of Reason” just as our non-phenomenological insights are. In philosophical theorizing, we decide whether or not to give our trust to this and this “seeming-to-be-true.” However, the latter doesn’t require any further phenomenological investigation (let alone the hunt for private evidence), but rather, a cost-benefit analysis by reason.

References

• Crane, T. (2005). “What is the Problem of Perception?” Synthesis Philosophica 20 (2): 234–67.
• Fish, W. (2009). Perception, Hallucination and Illusion. Oxford: Oxford University Press.
• Martin, M. (2006). “On Being Alienated.” In Perceptual Experience, edited by T. Gendler and J.H. Hawthorne, 215–72. Oxford: Oxford University Press.
• Martin, M. (2004). “The Limits of Self-Awareness.” Philosophical Studies 120 (1-3): 367–425.

Ágnes Katona’s objection concerns the relationship between science and philosophy. In her opinion, my characterization of philosophical theorizing (namely, that the factors leading to our considered philosophical beliefs don’t track the truth; our philosophical theorizing is based on intuitions and pre-philosophical convictions that have a good chance to be biased; our personal and epistemic character has an inescapable role in the formation of our philosophical views etc.) is also true of scientific theorizing—“these kinds of ‘bias’ aren’t unique to philosophy.” As she sees it, “the situation is analogous to choosing between empirical equivalent theories in science.” According to her, “it’s natural to include extra-theoretical criteria like parsimony, elegance or fit with our other theories in evaluating competing interpretations, and our judgment about those criteria are probably influenced by our taste and epistemic character.”

Now, if this is so, I must face the following dilemma. Either I trust in the success of scientific theorizing despite the fact that some epistemically “suspicious” elements have a role in it, too—but in this case, I don’t have any reason to doubt the success of the kind of philosophical theorizing that is exclusively based on scientific theories. Or, I don’t trust the latter—but in this case, I have no reason to trust in the success of scientific theorizing either, as it is influenced by such “suspicious” factors to the same extent. By choosing the first fork of the dilemma, I’ll have to admit that the kind of philosophy that is based on scientific theories can be successful: “the legitimation of the sciences extends to philosophy in these cases.” Choosing the second fork means that “both cases are equally problematic and demand that scientists, too, stop believing in the truth of those theories where they appeal to intuitions.”

First of all, I’d like to address Katona’s objection within a somewhat broader conceptual framework, and then to show that the above dilemma is misleading—the real fault line runs elsewhere.

*   *   *

Firstly, as for the philosophical interpretation of scientific theories: the most fundamental question is whether we accept their scientific realist (SR) or anti-realist (SAR) interpretation. SR can be characterized with three theses. (1) The world has a definite and mind-independent structure. (2) Scientific statements about both observable and unobservable entities and mechanisms are to be interpreted literally—they’re definitely true or false. (3) The current mature, predictively successful, well-confirmed, etc. scientific theories and models are true of the world, including its observable and unobservable (fundamental) domains—at least some of them, and at least approximately true: they won’t be modified, only refined over time; scientific progress converges on the final true account. SAR has several versions, but all of them have in common that they reject the epistemological thesis of SR. According to them, the success of sciences is limited to their being able to make exact predictions and retrodictions, systematize observational reports, manipulate certain complex phenomena, and give causal explanations of certain observable phenomena—which amounts to no more than recognizing certain regularities or patterns of contrafactual dependence among the groups of certain observable phenomena.

To sum up, while according to SR, the sciences have epistemic authority, they can tell what the world is really like (the world really is the way our best scientific theories describe it), according to SAR the success of the sciences is actually not an epistemic success, they cannot tell what the world is really like (our best scientific theories tell nothing about what kind of fundamental things exist and act on each other).

Secondly, as for the underdetermination of scientific theories by data: this is the basis for one of the main objections to SR. The argument runs like this. (1) The best scientific theories (might, must) have empirical equivalents: they’re equally well supported by and consistent with any possible body of empirical evidence, but differ with respect to their postulated unobservable (fundamental) entities and mechanisms. (2) Empirical evidence is all the evidence we can have. Therefore (3) no evidence will ever permit us to decide between rival scientific theories—there’s no evidence for deciding which of the empirically equivalent theories is true.

According to those who accept SAR, the argument from contrastive underdetermination is spot-on—we must reconcile ourselves with the sciences not being able to do more than come up with empirically adequate theories. According to those who accept SR, the argument misses the mark because it doesn’t follow from the fact that T1 and T2 are empirically equivalent (equally well confirmed by empirical evidence) that there can be no further evidence for the truth of one of them. At this point, the theoretical virtues that Katona refers to come into play: parsimony, simplicity, fruitfulness, etc. By weighing these further pieces of evidence, we can and must commit ourselves to the truth of some of the rival empirically equivalent theories. If T1 is more comprehensive and yet more parsimonious, more fruitful and more elegant than T2, meaning that it performs better in terms of the theoretical virtues, then—even if the two are equally well confirmed empirically—we must accept T1 as true. We must conclude that T1 is true—at least probably and approximately.

Thirdly, as for the inference to the best explanation (IBE): we can entertain several well-known scruples against its reliability. (1) While choosing a theory, we have reason to believe in the truth of the theory that we reached by IBE only if the true theory has already been present in the first place among the rival theories available to us. In the opposite case, as van Fraassen writes, we may end up with “the best of a bad lot” (1989, 143). (2) Here’s an open question: why should we trust that the theoretical virtues forming the basis or principles of IBE are connected to the truth? Why should we believe that the various theoretical virtues will lead us to the true theory, rather than to believing that “[v]alues of this sort […] provide reasons for using a theory, or contemplating it, whether or not we think it true, and cannot rationally guide our epistemic attitudes and decisions” (van Fraassen 1980, 87)? (3) Another open question: what should we do in cases when empirically equivalent theories have conflicting theoretical virtues? What should we do if T1 fits better with our background knowledge but less fruitful than T2? Or, if T1 brings together a larger body of data under one explanation but is less elegant and beautiful than T2? In such cases, we inevitably have to rank the various theoretical virtues, but it’s unclear on what basis we should do that—how we could properly justify any epistemic ranking of theoretical virtues (should it contain anything). (4) Focusing specifically on the most often mentioned theoretical virtue, ontological parsimony: why should we give credence to the epistemic principle that the more parsimonious of empirically equivalent theories is the true one? If the unobservable and fundamental domain of our world happens to have a “baroque” ontology, then we’re wide of the mark in choosing the most parsimonious one from among empirically equivalent theories and starting to believe in its truth—what we do isn’t “I to the best E,” but in fact “I to the worst E.”

According to those who accept SAR, these scruples are legitimate—there’s no way to justify the reliability of IBE. On the contrary, according to those who accept SR, these scruples can be dispelled—it can be justified (with sufficient force) that theoretical virtues like simplicity, beauty or parsimony (can) lead us to the truth in choosing a theory.

So, we’ve reached a point where we have two consistent combinations (equilibria). One is that: “SR is true and IBE is reliable.” The other is that: “SAR is true and IBE is unreliable.” Thus, either I have to claim that IBE is reliable despite all emerging epistemic suspicions, and so commit myself to SR, in which case I’ll have no reason to doubt the success of that kind of philosophical theorizing that is exclusively based on scientific theories. Or, I have to claim that IBE is unreliable and so commit myself to SAR, claiming that “scientists too [have to] stop believing in the truth of [their] theories.”

*   *   *

If the question is whether I believe in the potential epistemic success of the kind of philosophy that allies itself with the sciences and is exclusively based on scientific results in theorizing, then my answer is an emphatic NO. But the reason why I’m a pessimist is not that I’m committed to the second fork of the dilemma posed to me by Katona (the SAR and non-IBE combination), thinking that scientists can at most come up with empirically adequate theories. My pessimism doesn’t stem from my putting in my two cents about SAR, saying that it has stronger underpinnings than SR. It is about something else—the real fault line runs elsewhere.

The crucial point concerns what we should think about the relationship between scientific and philosophical theorizing. On the one hand, we may think that the two form a continuum; there’s no essential difference between them; there are no scientific problems with philosophical ones over and above them; there are no special philosophical truths, which we could only get to know using philosophical methods that differ from the tools of science—there’s no such thing as a distinctively a priori philosophical project of theorizing. In a word, we may commit ourselves to metaphilosophical or methodological naturalism (MN). On the other hand, we may think that there’s an essential difference between scientific and philosophical theorizing; there are some synthetic propositions that can only be justified a priori—via rational insight, intellectual grasp; phenomenological Wesenschau, the use of the “faculty of reason.” In a word, we can commit ourselves to the falsity of MN; saying that philosophical problems or questions are more fundamental than scientific ones.

Now, one may say that there can be a kind of philosophy that is epistemically successful (is able to establish substantive philosophical truths), is allied with the sciences, and is based exclusively on the best scientific theories only if one commits oneself to the truth of both SR and MN. However, I think that this optimistic combination is untenable. The reason why I think it to be untenable is not that I think that SR is false. The reason why I think it to be untenable is that I think MN to be false. I think that one can a priori recognize that—pace esp. Psillos (1999), ch. 4—the reliability of IBE and thus indirectly the truth of SR cannot be justified a posteriori, that is, within the framework of MN. That is, one can a priori recognize the truth of the conditional: “If SR/IBE is justifiable at all, then SR/IBE is justifiable only a priori.”

Here’s how the pieces of the pessimistic picture fit together. The question whether “Is SR true or false?” is a par excellence philosophical question, and as such, it’s more fundamental than any scientific question in an epistemic (and in a related conceptual) sense. We don’t know the answer to this question. Since we don’t know the answer to this question, our best scientific theories are merely “If … then …” type truths. All we are justified in saying is that if SR is true, then (but only then) our final physical theories can be our final metaphysical theories—or at least they can give us a clue to the solution of metaphysical problems.

I see the real fault line as follows. Either we believe that SR can be properly justified, but in this case, we must also believe in the epistemic success of the kind of philosophical theorizing that is independent of scientific theorizing. Or, we do not believe in the epistemic success of philosophical theorizing that is independent of scientific theorizing, but in this case, we don’t have any reason to consider SR to be properly justified. I, for one, prefer the second option, and the reason why I do so is that I still think that the factors leading to our considered philosophical beliefs don’t track the truth; our philosophical theorizing is based on intuitions and pre-philosophical convictions that have a good chance of being biased; our personal and epistemic character has an inescapable role in the formation of our philosophical views.

References

• Psillos, S. (1999). Scientific Realism: How Science Tracks Truth. London: Routledge.
• van Fraassen, B. (1980). The Scientific Image. Oxford: Oxford University Press.
• van Fraassen, B. (1989). Laws and Symmetry. Oxford: Clarendon Press.