Transcendental Arguments.

Transcendental arguments are a favorite among Gangadeanians. Anderson once told me in conversation that he thought analytic philosophers failed to pay enough attention to them. Roughly a transcendental argument functions in the following way. Take something that is true, and then consider what else must be true in order for that original thing to be true. In other words, you pick a premise, a starting point that you believe is true, and then sort of figure out what it entails---i.e., what else must be true for the starting point to be true. So Gangadean thinks that meaning is possible. He then considers what else must be true for there to be the possibility of meaning. It's obvious to him that clarity at a basic level must obtain in order for meaning to be possible. Thus, he concludes there must be clarity at the basic level.

One can resist a transcendental argument in at least 2 ways. First, you can outright deny (or at least call into question) the starting premise or phenomena that is taken for granted. In this case it would be to deny that meaning is possible. That sounds pretty bad, though (at least the outright denial). Do you really want to deny that meaning is even possible? Isn't it seemingly self-undercutting to do so. Aren't you assuming meaning in denying meaning? I feel like that is mostly right although Gangadean again has failed to tell us what he means by 'meaning.' Maybe he takes it as a primitive (which again must introduce intuitions--after all, one needs to ask how do we come to grasp primitives, correctly?). Moreover, I think I know what 'meaning' means. But I'm not sure. Like I use the word all the time, and I suspect at least many of my usages are appropriate or correct, but it also seems to have various senses. Like if I say, "life has no meaning" by 'meaning' I intend to convey something like "objective or subjective purpose." But semanticists also talk about the meaning of sentences, which they identify at least partially with "truth conditions" (i.e, the conditions under which the sentence is true). Or perhaps we can look at ordinary language-- like often when I ask someone what a word or phrase means, I'm asking for a definition of sorts---I'm asking someone to identify the word or phrase with synonyms that I am more familiar with in some suitable sense. So when Gangadean takes for granted that meaning is possible, as his starting point, what is he actually saying? I think he's hoping that you have some sort of intuitive grasp of the word/concept and that you won't press any further. But it turns out that 'meaning' is a tricky concept in and of itself. Still, you won't get anywhere by questioning the possibility of meaning with a Gangadeanian, so this might not be the most promising push-back point. So here's a second way to resist a transcendental argument--- show that the preconditions that the arguer has presented are not actually entailed by the starting point. Remember transcendental arguments have at least two main parts. A starting point and then the purported preconditions to that starting point. I think that for some of Gangadean's transcendental arguments, the right move is to actually deny the starting point. But perhaps for the one about the possibility of meaning, it's more savvy to deny the preconditions.

In effect, my last post was along these lines. At least if we think about the conversation highlighted in that post, as an informal version of a transcendental argument (or at least I think a transcendental argument was in the background for my interlocutor). Here's what I have in mind. My interlocutor was thinking that it was common ground that intelligible discourse, assertions, thinking, and the like are possible. That's the starting point. After all, I was talking and thinking and making assertions, as was he. So we must both accept that such activity (qua intelligible activity) is possible. Note this doesn't necessarily mean that we accept the starting point in the same exact way---after all he's a Ganagdeanian and so he accepts it with absolutely certainty, whereas, I may not. Next, his move was to try and show me that the necessary precondition for the possibility of such activity was that there is clarity at the basic level. And this is just what I called into question in my last post. It is what my interlocutor thinks is entailed by the starting point. Intelligible thought, talk, assertion and the like necessarily requires the precondition of clarity at the basic level. But he failed to actually show this is true---instead he simply assumes it.

In a similar way, when Gangadean insists that clarity at the basic level or clarity of basic distinctions is requisite (or entailed by) the possibility of meaning. So if you don't find it amenable to resist the starting premise that (meaning is possible) you can always question the entailment. After all has he given you any reason to believe that meaning is possible only if there is clarity at the basic level?

In passing, I said above that there are other transcendental arguments where I suspect the right move is to actually deny the starting point or premise that is assumed rather than the preconditions (I'd imagine there will be cases where you ought to deny both). I have in mind the case of skepticism. Let me just say a few words about it.

Gangadean thinks that the skeptic is being self-referentially absurd. The skeptic as Gangadean defines it, is one that denies that knowledge is possible. But remember for Gangadean knowing that P = not possibly being wrong about P (what I sometimes refer to as epistemic certainty). Now he (rightly) shows that a skeptic (understood in these terms) is denying with absolute certainty that absolute certainty is not possible. That's self undercutting. So skepticism (as the view that it is absolutely certain that absolute certainty is not possible) is shown to be false. And on some occasions Gangadeanians will use this conclusion as a starting point for a transcendental argument. What must be the case if skepticism is false (i.e., knowledge is possible?) Clarity at the basic level for instance, might be suggested as necessary for the possibility of knowledge.

But there's a lot of equivocating under the surface that Gangadeanians exploit (I suspect they do so unawares rather than anything sinister). Again it all comes back to Gangadean's theory of knowledge. He thinks that knowledge requires absolute certainty--where you know something only if you couldn't possibly wrong about it. Skepticism is in turn defined as the denial that such knowledge is possible. After showing the self-referential nature of the skeptic (as he defines her) he will then consider what must be possible for knowledge (but remember he means absolute certainty when he refers to knowledge). So then he goes into a theory about the preconditions for the possibility of absolute certainty while calling it knowledge. Of course, you might follow with him because you are not aware about how loaded the notion of knowledge is (you should ask yourself at this point whether you agree with him that knowledge requires absolute certainty). So Gangadean says, if knowledge is possible, then this requires clarity at the most basic level (where clarity again is something like having absolute certainty) you start to buy into what he calls the necessary preconditions.  And if you're not careful you think, well then I must have absolute certainty about very basic things.

But what if knowledge doesn't require absolute certainty? That is, what if we resist or at least call into question the starting point?  In fact, the vast majority of philosophers think knowledge doesn't (they are called fallibilists about knowledge). At the very least, it isn't anything like a settled matter even though Gangadean treats it as such. So you can question/doubt whether knowledge requires certainty (without outright rejecting it). Thus we have reason to not affirm Gangadean's starting point that knowledge (as he define the word) is possible. You are not doing anything self-undermining in doing so. You're not saying that you know with absolute certainty that absolute certainty is not possible. You are merely saying that it's not obvious that absolute certainty is possible. By doing so you block the transcendental argument from getting off the ground since the starting point hasn't earned it's place at the table as such. This means there is no rational reason to accept the purported preconditions that are said to be entailed by the starting point since they rest crucially on the starting point qua starting point.