Intuition Mongering

I think that readers might be interested in the following short papers:

1. "Intuition Mongering" published in the November issue of The Reasoner
2. "More Intuition Mongering"

Comments and suggestions are more than welcome.

Comments

[Clayton]

Hi Moti,
I know that we've discussed this before. Forgive me if this is territory that we covered back then. In your second paper, you seem to endorse a model of appeals to intuition that's taken from Huemer's work. That might be the model that Huemer works with, but I wonder whether that's the only model available and also whether it's the right model. (I'll set the exegetical issue aside.)

An intuition monger might agree with Bealer that when you have an intuition that p it can seem to you that p. It doesn't seem that the intuition monger is thereby committed to the view that (i) when one appeals to intuition one's belief is based on the fact that it seems to one that p or (ii) when one appeals to intuition one is (implicitly or explicitly) reasoning from the fact/thought/proposition that it seems to one that p to p. Think about how appeals to observation work. If I observe that this raven is black, I suppose that you could say that my belief that this raven is black is formed inferentially (e.g., it's based on the fact/proposition that I observe that this raven is black, I see that this raven is black, or that it seems to me (visually) that this raven is black), but I don't see why you'd have to say that. You could also say that it's based on what I've observed (which is that this raven is black). You might think that at some point some belief is based on the very fact that the belief concerns. Maybe that's how introspection works. If that's the right model for introspection (e.g., my belief that it seems to me that p is based on nothing further than the fact that it seems to me that p), I don't see why it couldn't be the right model for observation (e.g., my belief that this raven is black is based on the observation, which is that this raven is black). And if it's a possible model for observation, why not intuition?

Rather than see 'appeals to intuition' as involving an inference from a premise known via introspection (i.e., that it seems intuitively to me that p), your belief could be based on what you intuit, which is that p. If that's right, there's no reason to think Jackson _infers_ anything about Mary from some further fact about himself.

At one point in the second paper, you seem to suggest that the reason/evidence for a philosophical judgment is a seeming. That looks like a kind of mental state or event. In reconstructing Jackson's reasoning, however, Jackson's belief that (M2) is true (i.e., Mary learns something new upon her release) is supposed to be based on the fact/proposition that it seems to him that Mary learns something new upon release. Isn't there a difference between Jackson's evidence consisting of a fact (or apparent fact) and consisting of a seeming? The fact might be a premise in reasoning, but the mental state/event cannot be. Suppose that you decide to represent Jackson's epistemic situation as follows:

(*) J's belief that Mary learns something upon release is based on a further fact, which is that it seems to J that Mary learns something upon release.

We'd want to know why (*) rather than:

(**) J's belief that Mary learns something upon release isn't based on any further fact. It is evident to him that she learns something and that is his reason.

Is it something special about intuition or is there some general rule about beliefs, facts, reasons, evidence, etc? If it's some general rule about the grounds of judgment, I fear that the general rule will generate some odd results when applied to things like introspective judgment and judgment about validity. (Is my belief that it seems to me that p really going to be based on some further fact to the effect that it seems to me that it seems to me that p? Is my belief that inference from p and ~(p&q) to q is valid really based on the judgment that this inference strikes me as valid?)

If, on the other hand, there's no general rule here, what's special about intuition? If, when it seems introspectively to me that p, my reason for judging that p is just the fact that p, why can't that hold in the case of observation and intuition?

[Moti Mizrahi]

Hi Clayton,

I think we’ve discussed this before. But I appreciate your comments, nonetheless, and am happy to discuss this again.

As I understand it, you’re taking a sort of “direct realist” stance with respect to intuition. When S intuits that p, S has direct access to p, rather than an intellectual seeming that p (from which S infers—either explicitly or implicitly—that p).

I think that there are several problems with this “direct realist” view of intuition:

1. If we’re thinking about intuition by analogy with sense perception, then the “direct realism” model flies in the face of perceptual psychology. Studies on optical illusions (such as the rubber hand) show that perception is belief- and expectation-laden.

2. If intuition gives S direct access to the facts, how do we explain philosophical disagreement? Jackson intuits that Mary learns something new, but Dennett and others intuit that she doesn’t. Do we want to say that Jackson has some special access to the facts that others don’t?

3. If intuition gives S direct access to the facts, how do we explain findings in experimental philosophy, which show that intuitions vary across cultures and are subject to order and framing effects? If people simply intuit what is the case, then why would cultural background make a difference?

4. *It seems just obvious to me that* when S makes an intuitive judgment in response to case C, that intuitive judgment (call it an intellectual seeming) is distinct from facts about C. If intellectual seemings in response to C are not the same as facts about C, then S gets at the facts about C by inference from S’s intellectual seemings in response to C.

That last one was tongue-in-cheek, of course :)

[Clayton]

Hi Moti,

So, I agree that those are worries. How serious are they? They aren't serious enough to exclude from consideration views on which a capacity (introspective, perceptual, etc.) can put one directly in touch with a certain sort of fact. The question as to whether perceptual experience is relational and you can relate you directly to something external seems to me to be open. I think that one can take the hard line and adopt a kind of metaphysical disjunctivist view on these issues and you don't count as an opponent of such a disjunctivist view simply because you say that something seems true to you intuitively.

While you could build an account of intellectual appearances on the model of McDowell's disjunctivist account of appearances, that might not be necessary. Something weaker might suffice. Here's an argument for the sort of view that you'd oppose.
P1. Evidence consists of true propositions or facts, not falsehoods and not mental states/propositional attitudes/mental events.
P2. The set of things that constitutes _your_ evidence will consist entirely of things that are themselves evidence.
C. The set of things that constitutes _your_ evidence will consist of facts.
P3. Nothing will belong to your evidence unless you have the (epistemic) right to treat it as a reason for belief, but if you have the (epistemic) right to treat something as a reason for belief and it is itself a piece of evidence, it belongs to your evidence.
P4. If you know something non-inferentially, you have the (epistemic) right to treat it as a reason for belief and it will itself be a piece of evidence.
P5. The set of things that you know non-inferentially will include some things that you know because it is intuitively evident.
C2. The set of things that constitutes your evidence will include facts that seem to you to be true because they are intuitively evident, not merely facts about what seems or appears to you to be true.

It seems that we disagree about (C2), so our disagreement has to do with (i) the ontology of reasons for belief, (ii) our views about what it takes to have evidence, or (iii) whether there can be non-inferential knowledge on the basis of intuition. I take it that your aim is to argue for a kind of skeptical conclusion, not simply argue from it. If you grant (iii) for the sake of discussion, it seems that our disagreement would really have to do with (i) or (ii). So far as I can tell, the view that I've staked out with respect to (i) and (ii) doesn't commit me to any sort of metaphysical disjunctivism. Instead, it commits me to the ho hum view that when we exercise a set of psychological capacities they can, in favorable conditions, reveal certain facts to us that we can then use to reason from.

So, if I were to argue from the more conservative approach that's neutral on whether direct realism is true, I could grant you something (that I don't currently have a view on) that facts about perceptual psychology undermine direct realism. These facts don't undermine the view that an exercise of our cognitive capacities can put us in direct _epistemic_ contact with a fact when the conditions are favorable.

Addressing your (2) or (3) can be difficult, but I don't think it would be _so_ difficult that we should assume the sort of foundationalist view I've sketched is a non-starter. There are in both cases salient facts that explain why, owing to vast differences in background beliefs, subjects would respond to cases differently. Moreover, if you were truly impressed by the facts about diversity and disagreement, you could simply deny my (P5) and embrace skepticism. Maybe the main point I want to stress is this--what's motivating the seemings/appearances view of intuitive justification that you seem to favor that might be the view that Huemer and others defend is an implicit sort of skepticism. It shouldn't be surprising that if your conception of evidence is formed under skeptical pressures, you'll be able to argue from that conception of evidence to a kind of skeptical conclusion.

[Moti Mizrahi]

Thanks for your response, Clayton.

If by “not serious enough” you mean that my 1-4 don’t amount to conclusive refutations of the direct realist view of intuition you have sketched, then I agree. They were not intended to be such refutations. But they do count as strong evidence against the direct realist view.
In light of my 1-4, I don’t see much going for the direct realist view in terms of positive evidence other than the possibility claim that it could still be true despite 1-4.

Now, I would not characterize my view as a form of skepticism, since I grant that intuitions could count as evidence in philosophical arguments (provided certain conditions are met). What I do not grant is that they count as *non-inferential* evidence. And I also argue that they are rather *weak* evidence as such. For example, suppose that one wants to argue that q, and one thinks that p entails q. The following argument will hardly be a convincing one (not to mention invalid):

It is intuitively evident that p; if p, then q; therefore, q.

Rather, one needs p as a premise in an argument that goes like this:

p; if p, then q; therefore, q.

How does one get p? By inference from ‘it is intuitively evident that p’ or simply ‘it seems that p’. [Note that, *in principle*, one’s interlocutors could accept the inference from ‘it seems that p’ to ‘p’ and grant one ‘p’ as a premise in one’s argument. However, in philosophy, it is usually the case that one’s interlocutors do not grant one the inference from ‘it seems that p’ to ‘p’ because they have different seemings about the same cases, as in the examples I cite in the papers.]

So, as stated, I don’t necessarily disagree with (P5) and (C2). S can know that p is intuitively evident. But knowing that p is intuitively evident is different from knowing that p. To get from ‘it is intuitively evident that p’ to ‘p’ we need inference.

Similarly, for S, F is a fact because it is intuitively evident that F is a fact. But ‘it is intuitively evident that F is a fact’ is different from ‘in fact F’. To get from ‘it is intuitively evident that F is a fact’ to ‘in fact F’ we need inference.

[Clayton]

Hi Moti,

Thanks for your response. I think this is a really interesting topic and I'm glad that you wrote up those two papers because it nicely frames a serious epistemological problem. In introducing a direct realist view into discussion (and then a weaker view that's probably closer to one that I'd defend), I only meant to say that I didn't see what justification you gave for thinking that the people who take intuitions to be evidence would only defend something like this:

Intuitions as evidence (I): Intuitions are evidence in the sense that when one has an intuition one can have as evidence the fact that it seems (intuitively) that p.

This view (IAE (I)) set up an inferential model of intuitive justification that might be right, but I didn't see why talk of what seems intuitive and the like committed me to that view rather than this one:

Intuitions as evidence (II): Intuitions are evidence in the sense that when one has an intuition one can have as evidence the fact that seems intuitively evident. [NB: The evidence is not the fact that p is intuitively evident. The evidence is p, which is intuitively evident. In other words, your evidence is p. It is your evidence because it is intuitively evident.]

It's that second view that I'd probably defend (and it's a view that I don't think requires the direct realist view from above). It's a view that I like to discuss because it seems to be a view nearly universally ignored in discussions of intuitions and it is close to a view that's near and dear to me that I have defended for perceptual justification/reasons/evidence, etc.

To see the differences between the two ways of characterizing the intuition monger's view, consider a case:

Simplest Trolley: A trolley is rolling down an empty track and would come to a stop safely without injuring anyone unless you were to push a man into its path.

According to IAE (I), if you had the intuition that it was wrong to push, the strongest evidence you'd have for a moral judgment about this case and for evaluating various moral theories is the fact that it seems wrong to push. According to IAE (II), that's part of your evidence, but it's also part of your evidence that it's wrong to push. (That it is intuitively evident that it is wrong to push might not itself be part of your evidence. The fact that it is intuitively evident explains why the moral fact is part of your evidence).

The argument I sketched above was in support of the view that facts that seem intuitively evident are part of your evidence and these facts are distinct from facts about what facts seem intuitively evident to you (e.g., in Simplest Trolley, someone who defends IAE (II) might say that your evidence includes something you know non-inferentially: that it is wrong to push). I think this is a view that you want to deny, but I'm not yet sure yet why that would be. We seemed to be on the same page when it came to the ontology of epistemic reasons/evidence, so it seemed that our disagreement would have to be about some other premise in the argument I sketched for IAE (II).

(Also, one minor thing about 'It is intuitively evident that...'. I was taking that to be factive. If I said that it is intuitively evident that it is wrong to push, I took that to entail that it is wrong to push. I can't say that it is intuitively evident that it is wrong to push, but it's not wrong to push. Anyway, the more important point is the fact that p is intuitively evident is the sort of thing that explains why you have p as part of your evidence, but it need not be a further piece of evidence. Better to think of it as an enabling condition.)

[Moti Mizrahi]

Thanks again for this great discussion, Clayton.

I would actually disagree with what you say about ‘it is intuitively evident that p’ if what you mean by ‘it is intuitively evident that p’ is simply what I mean by ‘it seems that p’. That is, it can seem to one that p even if not-p. So, it can seem to one that pushing the man is morally wrong even if it isn’t.

Also, your IAE (II) might merely appear to construe intuitions as non-inferential evidence when in fact it merely construes the inference differently. You say that “the fact that p is intuitively evident is the sort of thing that explains why you have p as part of your evidence, but it need not be a further piece of evidence.” In that case, it could be the case that one engages (even implicitly or non-consciously) in the following reasoning:

1. In response to case C, it seems to me that p is intuitively evident.
2. The best explanation for (1) is that p.
3. Therefore, p.

It can seem to one that p even if not-p because (2) might be false. Perhaps the best explanation for (1) is that one has misunderstood C, is delusional, is considering C under unfavorable conditions, etc.

[jonathan weinberg]

I think Clayton is right that the kind of "enabling condition" epistemology is pretty much completely consistent with existing philosophical practice. Clayton, is the following a correct understanding of your view? The proposition [a is F] can be in my evidence set in virtue of my intuiting that [a is F], without its needing to be the case that it is in my evidence in virtue of being inferred from [I intuit that a is F], antecedently in my evidence set. It may not even occur to me that I am intuiting [a is F], in which case [I intuit that a is F] might not even itself be in my evidence set at the time. (Compare to how all the time we see that, say, [the iphone is on the charger], and thereby have that proposition in our evidence set, without also thinking [I am seeing that the iphone is on the charger], and certainly without inferring the former from the latter.)

That sounds highly plausible to me. I don't think that arguments about the where & when of the trustworthiness of intuitions is _generally_ impacted by the distinction between the enabling model and the inferential model of intuitive evidence. (Just as the enabling model of perceptual evidence doesn't dissolve the problem in real cases of cross-subject perceptual disagreement.) But I suppose it would undercut Moti's attempted analogy to appeals to experts. Though, Moti, I must confess that I don't find that analogy particularly compelling in the first place! Intuition seems to me just as analogous to perception, or to memory, or to getting a reading off of an instrument, or, for that matter, to divine inspiration, as it does to appeals to authority. What they share, is that they are all cases of putative sources of evidence, and hence, all potential argument termination points. What makes intuitive access to evidence more like appeals to authority than any of these other sorts of sources, on your view?

[Moti Mizrahi]

Thanks for your comments, Jonathan.

We should be careful not to confuse intuition with appeals to intuition. An intuition is an intellectual seeming, i.e., when it seems to S that p. Appeals to intuition are arguments that put forth an intuition as support for a conclusion, i.e., it seems that p; therefore, p.

As you know, there are those who claim that philosophers are expert intuiters. If that’s the case, then philosophers’ appeals to intuitions are essentially appeals to expertise. If philosophers’ appeals to intuition are essentially appeals to expertise, then appeals to intuition are strong arguments just in case they meet the necessary conditions for strong appeals to expertise, such as agreement among the relevant experts.

[Clayton]

Hi Jonathan,

The view you sketched (i.e., The proposition [a is F] can be in my evidence set in virtue of my intuiting that [a is F], without its needing to be the case that it is in my evidence in virtue of being inferred from [I intuit that a is F], antecedently in my evidence set. It may not even occur to me that I am intuiting [a is F], in which case [I intuit that a is F] might not even itself be in my evidence set at the time.) was just the sort of view I had in mind. I also agree, fwiw, that this won't have much to do with the arguments about the trustworthiness of intuitions. Why bring it up, then? For one, I think that the view is widely neglected (and since it's one I've spent a few years arguing for, it makes me sad to see the poor little view neglected). For another, I suspect that it's neglect often reflects implicit skeptical commitments/instincts/dispositions, etc. (In class I had a few students blurt out that I couldn't use the premise [here is a hand] in trying to run Moore's proof because I didn't know _that_! This from a group that was trying to defend an internalist line about evidence while claiming not to be skeptics.) And for another, I think it's not at all plausible that in the moral case, say, our moral beliefs are justified by inference from some non-moral fact about our own mental life.

Hi Moti,

You wrote, "I would actually disagree with what you say about ‘it is intuitively evident that p’ if what you mean by ‘it is intuitively evident that p’ is simply what I mean by ‘it seems that p’."

I think we agree about that. My complaint is that some people (e.g., phenomenal conservatives?) aren't as careful in how they take our talk about seemings and appearances in helpfully telling us that we're all really relying on seemings and appearances. If you ask why I endorse some argument and I say that it seemed valid and the premises seemed true, I don't mean to assert that my evidence for thinking that the argument is sound consists of facts about how things seem to me. While I might say that the argument seems valid, I might treat the fact that the argument is valid as a reason for forming some further belief (e.g., for believing that the student who marked it invalid needs to work more on his logic).

[jonathan weinberg]

Moti,
re: appeals to intuition - I think perhaps you need to include a larger set of kinds of moves in your working theory of argumentative discourse. When I appeal to an intuition, I _might_ be drawing an inference from my having so intuited. Or I _might_ be manifesting to you the means by which my justification for P was already obtained. The comparison with perception continues well here: you ask me where the iphone is, I tell you it's on the charger, you say, 'really?', and I say, 'yeah, I saw it there.' I'm not appealing to any inference; I am informing you of my original grounds. On a view like Clayton's, it will make sense for there to be a class of such argumentative acts that nonetheless have no "therefore" in them.

You also write, "As you know, there are those who claim that philosophers are expert intuiters. If that’s the case, then philosophers’ appeals to intuitions are essentially appeals to expertise." But there are at least two big problems here. First, at a minimum it is _controversial_ whether those philosophers are correct about philosophers even _allegedly_ having some sort of intuitional expertise -- I continue to find it suspicious that it occurred to exactly no philosophers in the history of our discipline to claim any such expertise, before x-phi folks started making trouble for them. Second, even if those philosophers are right, and we do have expertise -- full-blown, restrictionist-refuting expertise -- that does not mean that the right way to understand the arguments is on the model of _appeals_ to expertise. Given that no one, anywhere, ever, actually appeals to any such expertise in their first-order philosophizing, that would strongly suggest that the way expertise could come into play is in some different sort of way, perhaps as part of the enabling machinery. (Just as, when I advert to perception as a source of evidence about P, I may need to veridically presuppose that my perceptual faculties are sufficiently reliable about P. But that doesn't mean that that presupposition is figuring as any sort of premise in an argument, under typical circumstances.)

Please note that, on such a picture, one could still potentially make use of dissent among the (putative) expert population to make trouble. But I think it's very important that you avoid framing things so flatly in terms of an inferential appeal to expertise. There are other options here, and it is far from obvious to me that the inferential picture is the most accurate one.

Clayton,
I definitely think that the "enabling" view is interesting and worth exploring for its own epistemological value! Indeed, it's not just interesting, but I think it's likely true, regarding our most ordinary uses of intuition. It seems that we also agree that Moti would be better off framing his arguments in a way that didn't have to pick _this_ particular battle.

[Moti Mizrahi]

Thanks for your reply, Jonathan.

As I said in reply to Clayton’s comments, whether one says that one’s justification is an intuition or not is completely beside the point. For one thing, one can be made aware of one’s justification quite easily in the way you have outlined:

“Why do you believe that p?”
“Because I see it.”
[appeal to sense-perception]

“Why do you believe that p?”
“Because it seems to me that p.”
[appeal to intuition]

“Why do you believe that p?”
“Because expert E says that p.”
[appeal to expertise]

Those who think that philosophers are expert intuiters are committed to the claim that when philosophers appeal to intuition they are essentially appealing to expertise, whether they (or the philosophers) admit that or not.

Please note that I do NOT endorse the claim that philosophers are expert intuiters. Like you, I find it highly suspect. So the problems you mention are serious problems for those who endorse the claim that philosophers are expert intuiters, not for me. I do not endorse that claim. My claim is that *if* one believes that philosophers are expert intuiters, then one should face the consequences of that claim, which are that when philosophers appeal to intuitions, they are essentially making an appeal to expertise. And if philosophers are essentially making appeals to expertise when they appeal to intuitions, then disagreement is a serious problem. The fact of disagreement makes arguments that rely on intuitions (such as Jackson’s argument and Chalmers’ argument) weak arguments. To say that these arguments are weak is NOT to say that intuition can never justify anything. Rather, it is to say that intuition is an *unreliable* source of justification.

[jonathan weinberg]

"For one thing, one can be made aware of one’s justification quite easily in the way you have outlined." Right. That's my point.

Similarly, I wholeheartedly agree with this claim, but not the way that you have plumped for in understanding it: "Those who think that philosophers are expert intuiters are committed to the claim that when philosophers appeal to intuition they are essentially appealing to expertise, whether they (or the philosophers) admit that or not."

My point is: it's a big mistake to construe what's going on here in _inferential_ terms, such that claims about one's expertise, etc. have to be taken as _premises_. That inferentialist picture _might_ be right, or at least right in at least some case. But it's a highly controversial claim, and moreover, it's a claim you just don't need in order to make your argument.

Now, the next step, btw, is to face what I think is the _really_ hard question here: just how much disagreement among (putative) experts is enough to threaten any such (implicit or explicit) commitment to one's expertise? But I suspect that would be a matter for a paper (or at least a blog post) of its own.

[Moti Mizrahi]

Thanks, Jonathan.

Yes, there's still more work to be done to end the intuition mongering.