40

This is a very interesting passage in all sorts of ways. It is 1970, and Kripke is clearly simply taking it for granted that philosophers appeal to intuitions, with he himself being a prominent example. The only controversy Kripke acknowledges is about the probative value of intuitions, and not their existence or the fact that they are appealed to. Now, Kripke, as we all know, is not a great student of the contemporaneous literature. It’s very unlikely that he would be so confident that contemporary philosophers appeal to intuitions because he had been avidly reading the latest journals and had noticed that talk of intuitions was beginning to take off. He writes as though it’s very old news that philosophers appeal to intuitions, the only question being whether they are right to do so. Second, the intuitions he is talking about are not about linguistic matters but about metaphysical claims: the metaphysical claim that there are necessary versus contingent properties. Third, it is a very explicit endorsement of the probative power of intuitions. But it not only assigns them a role as providing data for our theorizing, providing the first word on the credibility of a view: so much is common to the ideology of many philosophers. What’s fascinating is that he also assigns them the role of having the last word on the ultimate credibility of a view, which is very unusual.1 Wittgenstein provides more direct evidence against Hintikka’s historical claim. There is a famous passage in The Brown Book (which dates from 1934–5) in which Wittgenstein (1958 , p. 143) says: “It is no act of insight, intuition, which makes us use the rule as we do at the particular point of the series. It would be less confusing to call it an act of decision.” This passage was written, more or less as published, in a letter composed in English to Piero Sraffa in 1934. Wittgenstein makes it sound as though invoking intuitions would come very naturally to his interlocutors and needs to be explicitly rejected. I think that sustaining Cappelen’s line about intuitions does ultimately need a historical story about when the unfounded ideology of intuitions took hold and wrongly persuaded philosophers to think of themselves as appealing to intuitions. Otherwise, we may well say to him: Look, even if you are right that this or that original presentation of a classic case doesn’t appeal to intuitions, clearly philosophers in general do appeal to intuitions because they self-consciously act as if they do so and describe themselves as doing so. The only question is how what they are doing should be understood. Let’s set aside, though, this point about needing the debunking historical story. Let’s look at the classic cases. IV Let me start with Putnam’s (1975 ) Twin Earth case, which Cappelen does not explicitly discuss. Putnam, as we all know, is concerned with the Received View of meaning, according to which, • (1) the meaning that a speaker associates with a word is determined by individualistic facts about that speaker,and, • (2) that meaning determines the word’s extension. He seeks to put pressure on this combination of theses by coming up with a thought experiment.2 We are invited to imagine that in another part of our galaxy there is a Twin Earth, a planet as much like Earth as possible, except that the liquid that flows through its lakes, rivers, and faucets, although superficially quite similar to water, is not H2O but XYZ. On Earth, we suppose there to be a user of the word ‘water’, Oscar, who, while not knowing anything about water’s chemical composition, is nevertheless perfectly competent at meaning water by it. On Twin Earth, there is a molecule for molecule duplicate of Oscar’s who also counts as a competent user of ‘water’ by the relevant local standards, although he, too, does not know anything about the chemical composition of the liquid in his environment. Now, Putnam wants to know whether the extension of Oscar’s word ‘water’ is the same or different than Toscar’s. Putnam thinks that, when we reflect on his thought experiment, we will see clearly that, in the scenario he describes, Oscar’s word ‘water’ will have H2O in its extension but not XYZ, and that Toscar’s word ‘water’ will have XYZ in its extension but not H2O. The important claim here is not that XYZ is in the extension of Toscar’s ‘water.’ That much we may take to have been stipulated by Putnam (though it needn’t have been). The important claim is that H2O isn’t in that extension (and vice-versa for Oscar’s word ‘water.’) If he is right about this, the implications are huge, for it immediately refutes the Received View, which predicts that the extensions would be the same. And the Received View is enmeshed with a large number of other claims that we are inclined to believe—claims involving privileged access, psychological explanation, the relation between meaning and reference, and so forth. So the stakes are high. Accepting Putnam’s judgment, it turns out, threatens to upend many central, entrenched philosophical beliefs.His Twin Earth (TE) judgment, therefore, had better be pretty solid. We don’t want to rewrite large swathes of the philosophy of mind, philosophy of language and epistemology, if that judgment is just some thing that Putnam is inclined to say without justification. Clearly, he thinks his judgment is strongly justified and so does each of us when we arrive at the same intuitive judgment about this case. The question is: What is the justification for this judgment about the extensions of Oscar’s and Toscar’s tokens of ‘water’ in this highly contrived and specialized case, far remote from any actual scenario that we might have encountered? There look to be three options: • (a) We can perceptually observe that the TE judgment is true. • (b) We can infer it from background beliefs that we are perceptually justified in holding. • (c) Neither (a) nor (b) is true: our justification for believing the TE judgment comes from some other source. There is no question of (a)’s being true, so we can just set it aside. The contest is between (b) and (c). Now, for (b) to be true there must be (empirically justified) background beliefs from which the TE judgment can be derived. But what would those background beliefs be? They would have to be beliefs about extensions, or beliefs that carried implications about how such extensions are determined. But Putnam’s whole point is that our background theory about such matters—the Received View— predicts the opposite judgment than the one elicited by the thought experiment. It predicts that Oscar’s and Toscar’s tokens of ‘water’ will have the same extensions. That’s why the thought experiment is taken to falsify our entrenched background theory about these matters. So how could Putnam’s judgment be derived from that background theory? At any rate, Putnam does not present the judgment as based on inference from background beliefs. He presents the thought experiment and then, on the basis of the stipulated facts, simply draws the conclusion. On the basis of these considerations, we would have to say that something along the lines of (c) must be correct. Or, at any rate, since this is what matters most to Cappelen’s thesis, that Putnam must be assuming that it is. Putnam is assuming that he is justified in drawing the Twin Earth conclusion—indeed so strongly justified that he can advocate rewriting large swathes of the philosophy of mind and language on its basis—but without this resting either on direct perception or on inference from (perceptually) justified background theory. In Cappelen’s terms, it looks as though Putnam must be assuming that something like feature F2.1 is present in this case. It follows that, by Cappelen’s own standards, the case against the appeal to intuitions has not been secured. V Now, Cappelen will certainly want to resist this line of thought. We can infer what he would say about Twin Earth from what he says about a host of other thought experiments. One of Cappelen’s tactics for dealing with what look to be appeals to intuition—judgments with feature F2.1 in particular—is to say that the author is simply appealing to the idea that these judgments are in the “common ground,” a set of propositions which the author presumes to be shared and so does not feel the need to give them any argument or support. That shouldn’t mislead us into thinking, Cappelen says, that these propositions have a special justificatory source. If these propositions were questioned, the authors would certainly provide arguments for them; the only reason they don’t bother to do so is that they can assume that their readers will share them. We may certainly agree that, in given contexts, some propositions are simply taken for granted in those contexts, without special argument or support. For example, in the context of the Twin Earth thought experiment, the proposition that the Earth did not pop into existence in 1720, or that the chemical composition of water does not change from moment to moment, are simply taken for granted. As I read through the thought experiment, I simply took it for granted that those propositions were being assumed. But I wasn’t similarly taking anything for granted about the non-overlapping extensions of the tokens of ‘water’ or anything else from which that verdict would obviously follow. That verdict was not part of the assumed background. In fact, that I was tempted to make that verdict having read through the scenario and thought about it came as a surprise to me. I tried to resist it, but it kept forcing itself back upon me. It seemed like an unexpected and significant new realization. I despise it even to this day. But in the imagery that Gödel invokes in talking about the axioms of set theory, it keeps forcing itself back upon me. Another tactic of Cappelen’s is to claim that intuitive judgments are nearly never presented without accompanying arguments, arguments that are meant to support the judgments in question. Thus, they are not really intuitive judgments after all, since a necessary mark of such a judgment is that it is not justified by argument. Cappelen illustrates what he intends with Lehrer’s (1990 ) TrueTemp case. TrueTemp, as you may recall, has a device implanted in him that accurately records the ambient temperature and sends a signal to his brain—completely accurate thoughts about the temperature occur to TrueTemp, although he remains ignorant about their source. Cappelen (p. 170) quotes Lehrer who says: Does [TrueTemp] know that it is [104 degrees]? Surely not. He has no idea whether he or his thoughts about the temperature are reliable. … His thought about the temperature is correct information but he does not know this. … Yet the sort of causal, nomological, statistical or counterfactual relationship required by externalism, may all be present. Does he know that the temperature is 104 degrees when the thought occurs to him while strolling in Pima Canyon? He has no idea why the thought occurred to him or that such thoughts are almost always correct. He does not, consequently, know that the temperature is 104 degrees when that thought occurs to him. The correctness of the thought is opaque to him. (p. 187, italics added [by HC]) Cappelen thinks it’s clear that Lehrer is here arguing for his claim that TrueTemp doesn’t know, by appealing to the principle that knowledge requires more than correct information; he’s not presenting the claim that he doesn’t know as an intuitively justified verdict. I believe, on the contrary, that it would make an absurdity of the whole point of constructing the thought experiment to think of Lehrer as arguing for the claim that TrueTemp doesn’t know, by helping himself to the principle that knowledge requires more than correct information. If he already thought of himself as knowing that principle, why would he need to construct an elaborate sci-fi example? When Lehrer says that more than correct information is required for knowledge, he is not providing an argument for his claim that TrueTemp doesn’t know; rather, he is inferring that more than correct information is required for knowledge from the intuitively accessed fact that TrueTemp doesn’t know, even though he’s got correct information. It’s a diagnosis of why True Temp doesn’t know.We need not deny that diagnoses may circle back to provide further support for the intuitive judgment, and perhaps this is the kernel of truth in what Cappelen says. After all, if the principles that explain the intuitively accessed fact are themselves intuitively plausible, then we will have arrived at a local equilibrium that would be satisfying and that would confirm that we are on the right track insofar as these particular issues are concerned. But to admit this is not to deny that the intuitive judgments have a plausibility all their own, which they have independently of justification by independently plausible principles. And we know this because we know that sometimes there is no explanation of the intuitive judgment by independently plausible principle. In Putnam’s case, for example, the intuitive judgment forces us to give up one of two principles that we thought were independently plausible in favor of something that no one thought was independently plausible (meaning ain’t in the head!), but which we feel forced to accept by the thought experiment. So, Cappelen’s ways of defending against (c) are not effective. VI However, another line of resistance to (c) might be thought to be more effective. In my experience, many philosophers are tempted by it—Cappelen himself seems to have come around to it, although it is not what he argues in his book. According to this response, we do after all hold a background theory from which the TE judgment can be derived, despite its being true that the background theory that we are aware of holding—the Received View—could not be what explains our having arrived at it. The thought is that the theory from which we derive the TE conclusion is a tacitly held theory, whereas the Received View is an explicitly held theory that is the result of (potentially problematic) explicit philosophical theorizing. When we react to the TE thought experiment with the TE judgment, we manifest our acceptance of this tacit theory. Where did this tacit theory come from? Here we have two options. According to the first, we acquire such a tacit theory when we acquire some of the relevant concepts— in this case, presumably, the concepts of water and extension, among others. Indeed, acquiring such a concept just is acquiring such a tacitly held theory. And the TE judgment, on this view, is just a manifestation of our grasp of the concepts involved in it. So, here, once again, we see the idea of conceptual competence as underlying intuitive judgments rearing its head. According to the second option, the tacit theory is not constitutive of our grasp of the relevant concepts, but is just some background theory that we have come to learn about concepts and their extensions and that is part of our knowledge base. What would such a tacit theory maintain? We need not get too precise here. The theory would speak in general terms about what concepts are, the different sorts of concepts there are and how their extensions are determined. One point that will be important below, is that whatever exactly such a theory is going to say, it will surely be the case that it says it through necessary propositions. Such a theory will not consist in purely contingent propositions about how things happen to be. Could appeal to the existence of such a tacit theory be used to refute the claim that philosophers appeal to intuitions in arriving at conclusions on the basis of thought experiments? Let us look at this question by initially assuming the first option. It is not open to Cappelen to try to argue in this way because he joins Timothy Williamson (2007 , ch. 4)in denying that the possession of concepts may be thought of in terms of the acceptance of a tacit theory. So this line of thinking is not available to him. But what about a philosopher who doesn’t share Williamson’s skepticism about the tacit theory conception of concept possession; could she appeal to such a view to vindicate Cappelen’s rejection of Centrality? There are several reasons why such an appeal would be ineffective. In thinking about this issue, it is important to remember that what needs explaining here is how a subject’s intuitive judgment about hypothetical scenarios could count as justified. Suppose it’s true, as we are currently speculating, that such an intuitive judgment is either built into, or can be derived from, a theory that the subject has tacitly accepted in coming to have one of her concepts. Is such a fact an adequate explanation of how the subject could be justified in making that judgment? It is no part of our speculation that the subject herself knows that her judgment is part of such a tacit theory. Indeed, our subject may not know that she possesses any tacit theories or that the possession of concepts consists in the acceptance of such theories. All she would know is that she is inclined to make the ‘intuitive’ judgment. However, it is hard to see how the mere fact that some claim is built into the possession of a concept can make it the case, all by itself, that that judgment is justified.3 In addition, there would be the various other problems for such a view outlined at the end of section II . What about the second option, according to which the tacit theory is not part of concept possession but is part of our knowledge base, acquired somehow or other? This option does not mitigate the difficulties raised for the first option, and adds further difficulties of its own. They can be brought out with the following two questions: • (A) In virtue of what are the deliverances of the tacit theory justified on this option? If we are not to think of the tacit theory as concept constituting, we cannot even hope that its propositions would count as justified because they are epistemically analytic, or concept constituting. But if they are not concept constituting, in virtue of what are they justified? Assuming (as we must) that they are justified, surely they are not justified merely perceptually, since they consist, as previously noted, of necessary truths (about concepts and their extensions). So, for all we are entitled to think, they might need to have been justified by intuitions. And that means that, for all that this option does for an opponent of Centrality, it wouldn’t by itself allow him to distance himself from the claim that philosophers appeal to intuitions. The second question is: • (B) Assuming the deliverances of the tacit theory to have some justification acquired somehow, why do we accord the deliverances of the tacit theory a greater justification than those of the explicit one? Why don’t we regard the thought experiment as indicating a standoff? Or why don’t we treat the thought experiment as showing that we hold a false tacit theory, on the grounds that the tacit theory has been shown to conflict with the Received View? Why do we, instead, treat the TE judgment as a counterexample to the Received View? If we discern a conflict between an innately acquired and tacitly held folk physics and an explicitly arrived at physical theory, we certainly don’t immediately privilege the folk physics. On the contrary, we usually regard the folk physical theory as mistaken and we reject it in favor of the explicitly arrived at physical theory. In philosophy, by contrast, we privilege the intuitive judgment over that of the explicitly held philosophical theory. In terms of the option we are currently considering, that is to privilege the tacitly held theory, arrived at we know not how, over an explicitly theorized, amply justified and wellentrenched philosophical theory. In virtue of what is that a sensible practice? It’s hard to see what adequate answer to this question is to be supplied by an advocate of the tacit theory approach. The fact that we often privilege the impressions yielded by thought experiments shows that we regard those impressions as prima facie insights into the nature of modal reality akin to the way in which we regard perceptual impressions as prima facie insights into the nature of actual reality. VII We have been grappling with the question: Do philosophers rely on intuitive judgments in justifying their philosophical views? Well, they certainly rely on thought experiments and on judgments based on them. Should those judgments be thought of as ‘intuitive judgments’? How should this question be decided? I have argued that it should be decided by seeing whether those judgments are treated as (a) highly justified, (b) but not as justified in any standard way via perception or inference from premises perceptually justified, and (c) but only as justified by a certain sort of intellectual seeming experienced during the act of contemplating the hypothetical scenario of the thought experiment. It seems to me eminently clear that philosophers trade in judgments involving all three features very often and in ways that are quite central to their work. Thus, Centrality is true and Cappelen’s case against it fails. Of course, to offer this is not yet to address either various tricky questions about what intuitions are, or various skeptical issues about how intuitions could have the probative powers that they are credited with. It is merely to say that we are justified in believing that philosophers appeal to intuitions because they must be—there is no other viable explanation of their philosophical practice. Since they take themselves to be justified in making certain sorts of judgment on the basis of thought experiments, and since they are in a good position to see that there is nothing else to justify them in making such judgments, a charitable construal of their practice—in line, of course, with all their almost obsessive talk of ‘intuition’—would have them appealing to intuitions.