Interlocution, perception, and memory

Philosophical Studies

Volumn 86, Issue 1, 1997, Pages 21-47

  Burge, Tyler ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0002990744     PISSN: 00318116     EISSN: 15730883     Source Type: Journal    
DOI: 10.1023/A:1004261628340     Document Type: Article

References (14)

1. David Christensen and Hilary Kornblith, "Testimony, Memory and the Limits of {lie A Priori" Philosophical Studies 86 (1997), pp. 1-20 (this issue). Their paper responds to my "Content Preservation" The Philosophical Review 102 (1993), pp. 457-488.
2. Thus, given the views I outlined three paragraphs back, I maintain that in interlocution one can on occasion be a priori entitled to rely upon one's intellectual, seeming understanding of (and even a priori entitled to believe) propositions which can only be known empirically. The default entitlement to believe propositions one receives in interlocution presumes a more primary epistemic warrant somewhere in the chain of interlocutors. An a priori entitlement carried by the recipient of communication frequently relies upon necessarily empirical primary warrants that others have. I presume that a primary epistemic warrant for belief in the general proposition about zebras can only be empirical. Inasmuch as knowledge gained through interlocution depends on there being primary epistemic warrants that others have, the recipient's knowledge will be empirical even though his own individual entitlement to belief may be a priori - in the sense of "a priori" denned in "Content Preservation", op. cit. Some of the matters just glossed are discussed in more detail in that article.
3. One of the ways in which the elaborated characterization of intellectual understanding is a first approximation is that it does not elaborate what it means to understand assertive mode intellectually. Getting this right is complex, and not something I will attempt here. The initial explication of "intellectual understanding" will have to suffice in that case, for the present. The assertive mode of an assertion is not part of its content, so the approximate characterization in terms of application of elements of the content is not directly relevant. But the main difficulty is that understanding assertive mode, though commonly what I want to count "intellectual" in the rough intuitive sense I began with, is not entirely conceptual. (Cf. note 12 for an analogous source of complexification in the account of intellectual understanding, even of understanding of intentional content.) So my account of intellectual understanding in terms of conceptual understanding is an oversimplification from the beginning. The claim that understanding assertive mode is commonly intellectual thus invites further elaboration. The situation may appear to be even worse for me in that the claim is surely one that Christensen and Kornblith would dispute, on any elaboration of it that would serve my purposes. So clarification is needed at this central point. (The approximation that I have given in the text will suffice for the content of empirical generalizations, such as zebras are larger than poppies. My claim that understanding in those cases can be intellectual, in the sense given in the text, should be uncontroversial. The controversy in that case centers on whether we can be a priori entitled to rely upon seeming understanding of what an utterance's content is, granted that the understanding is intellectual.) I believe, however, that most of what I will say in favor of the a priority of our entitlement to rely upon our seeming understanding of particular putative assertions is independent of exactly how I characterize intellectual understanding of assertive mode. For I think that the most salient epistemic issues arise about understanding expressive events, or uttered presentations of content as opposed to abstract content, regardless of whether the mode or the content of the uttered expressions is at issue. I will, however, have more to say about understanding assertive mode.
4. John Biro, in "Testimony and Apriori Knowledge" Philosophical Issues 6 (1995), argues that particular applications of the general entitlement to rely upon one's understanding cannot be a priori because one cannot know a priori that there are no (possibly empirical) counterconsiderations to one's general entitlement. This argument rests on a misunderstanding of my Acceptance Principle in "Content Preservation" op. cit. The Acceptance Principle is: A person is [a priori] entitled to accept a proposition that is [taken to be] presented as true and that is [seemingly] intelligible to him, unless there are stronger reasons not to do so. (I have entered the bracketed clarifications. All are explicit in my original explanations of the principle, though I failed to make them explicit in the statement of the principle itself. None of them is relevant to Biro's argument.) The force of "unless there are stronger reasons not to do so" is to indicate that the person's entitlement is prima facie. The principle says that the entitlement holds unless there are stronger reasons (available to the person) that override it. It does not say that the person must know there are no stronger reasons; the individual may lack the concept of a stronger reason, and in any case need not rule out the existence of defeaters in advance. It is enough for the individual's being warranted that there are no defeaters; defeaters of the entitlement must be available to him. As Glenn Branch pointed out to me, an individual with the right conceptual abilities might know through non-empirical self-knowledge that there are no relevant counterconsiderations. On the other hand, it might be that some candidate counterconsiderations could be known not to be stronger only through empirical considerations. But the main point is that the individual does not have to know that there are no stronger reasons counting against the a priori entitlement in order for the a priori prima facie entitlement to entitle him to belief. It is enough that there be none that are available to him.
5. Although Christensen and Kornblith acknowledge this distinction between genetic and epistemic roles for a cognitive capacity, they often state the dispute in ways that ignore it. For example, they write, "The root intuition we are trying to capture [in deciding whether something is a priori] involves a distinction between learning about the world through our senses and learning about the world through thought" (p. 9). On my view, and most rationalist views, one's learning about the world through the senses, as a genetic matter, need not compromise the a priority of a justification, as long as the senses play a merely enabling role rather than a justificatory one. Similarly, they write, apparently in criticism of my view, "Prepositional content does not pass directly from one mind to another; rather the passage is mediated by perception of an utterance ..." (p. 12). As a description of the causal process there is of course no disagreement here. My metaphor of passage of prepositional content from one mind to another concerned the account of justification, and explicitly acknowledged the essential role of perception in mediating communication.
6. Cf. "Content Preservation", op. cit. p. 480, note 19. There are also, of course, differences in what is required for knowledge. Although one has a prima facie entitlement in interlocution to belief based on seeming understanding of a putative assertion (presuming that one's seeming understanding involves a genuine content that one is thinking), one has knowledge only if one does understand a genuine assertion. In the mathematical case, one can have knowledge as well as justification regardless of whether one has understood an assertion, or indeed any communication, as long as one genuinely understands a relevantly simple logical or mathematical truth.
7. The argument for the a priori connections between seeming understanding of putative assertion and prima facie rationality of the source, and between such rationality and the truth of the putative assertion, is given in "Content Preservation", op. cit. There is much to be said about this argument. But Christensen and Kornblith do not discuss it. They concentrate on my claim that we have a priori prima facie entitlement to rely upon seeming understanding of an assertion. So I will not defend the larger argument here.
8. It does not entail it, of course, because one might claim that the entitlement to rely on the understanding derives from other previous perceptual entitlements. I do not think that this is so. I do not think that any reasonable inductive or perceptual inference can explain our entitlement to rely upon our seeming understanding of strangers in unusual contexts. And I do not think that perceptual experience that precedes acquiring understanding plays an indispensable role in justifying applications of our intellectual faculties. But the point in the text is not meant to settle the issue about the nature of our entitlement to rely upon understanding. It is just to explore the relation between understanding and perception, showing that our entitlement to rely upon understanding can be independent of particular perceptual entitlements in the context ofthat understanding.
9. "Content Preservation", op. cit. p. 479.
10. As empirical scientific instruments approach the sophistication of androids in the way they express and process information, they become more nearly like interlocutors. Whether there is a sharp line between a rational source and a sophisticated empirical instrument (with lots of the functions or reason) is a deep question in the philosophy of mind.
11. This is, I think, one important difference between computers that we rely on to solve mathematical problems and most scientific instruments. I argue elsewhere in "Computer Proof, Apriori Knowledge, and Other Minds", forthcoming - that we can obtain a priori mathematical knowledge through computer aided proofs. It is not just that the computers rely on rational processes, and that we can be entitled to understand this by understanding their output. It is that the computers do not put us into de re empirical relations to a subject matter. Our understanding of their output is purely intellectual.
12. I think that all de re relations to times or particulars in time involve nonconceptual elements in the mode of reference. Cf. my "Belief De Re" The Journal of Philosophy 74 (1977), pp. 338-363. But I do not think that understanding contents that involve such de re elements always involves empirical warrants, or perceptual application. I believe that certain de re contents (for example, those involving / and certain uses of now and perhaps here) can be understood without being backed by a warrant whose epistemic force derives (even partly) from sense experience. One can understand such de re contents through mastering a conceptual rule, and applying the rule with such understanding in a context, where the application is not guided by the senses, only by thought. In effect, one allows the context to provide a re given the intellectual application of the rule. The application need not be guided by perception. Such understanding seems to me to be intellectual in a way that would not preclude an a priori justification or entitlement. So it appears that applicational elements in thought can be intellectual, without themselves being wholly conceptual. I think that understanding instances of the cogito is intellectual but not purely conceptual. The intentional content of such instances involves de re non-conceptual elements of application, but the understanding of the instances is not normally guided by sense experience. There are analogs in understanding de re elements (e.g. tensed elements) in thoughts about physical objects. These matters obviously complicate the relation between de re contents and the nature of one's epistemic entitlement to rely on one's understanding of such contents.
13. I will return to "appropriate way". Preservative memory does not add to one's justification or entitlement. But there may be conditions on preservation of justification or entitlement that go beyond merely preservation of the belief. I am sceptical about such conditions' looming large, but I leave room for their possibility.
14. There is perhaps an even more serious mistake in the discussion. It was part of my characterization of preservative memory that it does not confer or enhance warrant (neither a priori nor empirical warrant). It preserves beliefs and, normally, their warrant. We are entitled to rely on it. But it is not itself an element in an individual's justification: it provides no justification, and adds nothing to the force of justification or entitlement. They appear to be asking whether the warrant provided by preservative memory is genuine, overridden, or dependent on other matters. This would be a misunderstanding, unless "provided" simply means "preserved". Later (p. 18) they speak of the "justificatory power of memory", again suggesting this basic misunderstanding, insofar as preservative memory is at issue.