[hist-analytic] Chapter 1 of Aune's An Empiricist Theory of Knowledge

Baynesr at comcast.net Baynesr at comcast.net
Wed Oct 7 14:29:40 EDT 2009

I've been dealing with a lot of idiots, lately, which makes it more difficult
to deal with Aune. Still, here is my best effort, at this point, all things

Russell held, along with others, that there is a difference between knowledge by achquaintance and knowledge by description. Aune notes that philosophers used to think that "knowledge that" had its source in "knowledge of." There is a very complex story to be told, here, one that I cannot tell in full. Still, the "plot" should be outlined. 

For Russell knowldege by acquaintance is not "knowledge of" in Aune's sense. In Aune's sense we have knowledge of such things as buildings, streets and parks. However, for the early positivists acquaintance was not knowledge of objects, it was knowledge of sense-data. Knowledge of buildings etc. was "knowledge that" certain sensa (for short) belong together in a construct. Thus, what Bruce is calling "knowledge of" would be characterized as "knowledge of," thus the distinction which was actually operant is just a bit obscured. Recall, the point of acquaintance was to supply the empirical basis of knowledge in the sense that the propositions we know must reduce to predicates with which we have acquaintance in the form of sense data. Later, the positivists would divide over what the objects of acquaintance are, but this move took place concurrently with a rejection of Kant, whose views were being addressed by people like Russell. So my first concern is that knowledge required for meaning must be distinguished from knowledge of truth if sense is to be made of the historical antecedents that concern Bruce. Bruce is right about the historical impact of Wittgenstein's private language argument, despite the fact that many rejected the argument. I reject it as well, but that is another story. Aune's own position is, however, interesting and worth examining, but let's get clearer on what it is. 

I would add as an aside that the notion of a sense-datum is not the idea of a sensation, but more like the idea of a Kantian intuition. The relation of sensation to intuition is Kant is an interesting story few think worth the telling these days, but I think it is fundamental to philosophy and in particular philosophy of science. More on that later, perhaps. 

Bruce says (p. 7) that his is a "dual account": "...one in which a concept of knowing for certain is distinguished from a minimal concept that does not require rational certainty." This will make certainty a big issue. This is a huge topic! We will pursue within the confines of his discussion but it is a vast topic that when traced historically will be found to lead to the very entrance of any good discussion of what motivated belief in analyticity and a number of other theses. The idea of 'certainty' as, merely, psychological must be made distinct from the philosophically relevant sense. Bruce's Humeanism may militate against making this a clear distinction. I'm not sure. The next point which I think is absolutel fundamental is one with which I agree. 

Bruce says that when it comes to deciding between two of the most important positions, contextualism and invariantists, there is no fact of the matter. Now here I agree completely. Take the idea of being a 'paper weight'; there is no fact of the matter whether something is a paper weight. One looks at use. Similarly, there may be no real definition of 'knowledge'. What knowledge is may depend on use, but this is seem pretty close to contextualism. Contextualism is the view that 'S knows that p' is subject to stronger or weaker standards, depending on context. A quick word from me on contextualism. 

I am always suspect of 'context'; it has been used to justify just about every crazy idea that has ever come down the pike. We need a much clearer idea of exactly what it is etc. But Aune's point is well taken; whether we accept this position or its antithesis is not a matter of fact. (p. 9); but if not fact, then what? Pragmatic considerations? If so, then Aune is an epistemological pragmatist or so it would seem. How far does this go, and when is it merely convenient and when does it obscure a difficult problem a solution to which we flee in the seach in favor of the "easy" pragmatic answer of 'context'. But there is the fact of the matter that the original problem seems undecidable, but if it is, then how important can the question be? Not sure. Next there is a discussion of D. Lewis's paper. 

I'm not sure, but as far as I can tell, Lewis simply takes Kripke's notion of epistemic counterparts, the ones he discusses in his paper related to issues of the necessary a priori, and he (Lewis) says that knowledge of p is when my evidence rules out any such counterparts inconsistent with the p. On the face of it, this seems to be palpably ridiculous when applied to scientific knowledge; it is to forget the main lessons of Duhem and others. I can't go into this here and now, but there are a number of considerations in this regard. I'll mention one. Knowledge by my lights is knowledge of facts, but on the proferred definition we don't need ANY facts of the matter; all we need is some reliable mechanism for ruling out certain epistemic alternatives. But is the remaining alternative one that corresponds to a fact of the matter? And if there are many then, surely, we aren't entitled to speak of facts of the matter. So for Lewis there can be knowledge not only without belief but without facts of the matter. Like some other of Lewis's views, this is "cute" but not convincing, assuming I understand him right. What is wrong with being "cute"? An answer may be forthcoming in looking at the torment that Gettier unleashed on epistemologists. 

The strongest point Bruce makes in this chapter on Gettier is this: 

"…the truth of the proposition embodying the information ust not be ‘accidental’ so far as that evidence is concerned." (p. 27) 

Bruce cites a couple of recent papers by Steup and Heller, but I think the most careful examination of such a proposal is to be found in a much older work, a work that in my opinion is striking in its care and precision. Here I am referring to Brian Skyrm’s marvelous (but long) paper "The Explication of ‘X Knows that p’." (JP June, 1967, pp. 373-389) 

Skyrms, it will be recalled, remarked that "evidential warrant and belief must be connected," and 

"By parity of reasoning, we deny that there is knowledge when the evidentially warranted belief flows from one disjunct and the truth from the other" (p. 100) (here my page references are from _Knowing: Essays in the Analysis of Knowing_ edited by M. Roth and Leon Galis, Random House. 1970). Now as I recall (I haven’t read this paper in about 25 years), Skyrms adds at some point that there are cases where a causal connection is *required* of the evidential connection. I let this pass. 

I, also, like Aune’s use of "truth maker." The issue of facts brings in a lot of ontological baggage. Moore, circa 1914, suggested facts as truth makers. This has stuck, more or less, despite the popularity of coherence in some pragmatic circles. Indeed, there are philosophers – recall the "sling-shot" arguments – who, actually deny facts. This is a symbol of troubled times in my opinion. There are facts. I cannot know something to be true unless I know it to be a fact. There is more to the dispute, including the ontological status of "connections." But here we are "doing" epistemology." I would raise one question, and I do not have an opinion on the answer: "Is there a sense of "ought" that is not ethical etc which is such that I can say ‘I ought to believe this but I know it is not true’. Behind my question is a concern as to how distant, or – to use Aune’s expression – "indirect" evidence can be in relation to truth maker? 

A couple of small points. Bruce’s example (p. 32) of the Gettier example of the clock which is right once a day comes from Russell. I believe from the Problems of Philosophy. So Russell had an idea along Gettier’s lines but didn’t put it to use as far back as about 1910. 

Certainly, one of the best things in the chapter is Aune’s distinction bweteen two sorts of knowledge. I won’t rehearse the contents. But there is a loose and strict sense of knowledge that must be distinguished. One minor correction. Bruce mentions that Chisholm introduced the difference between a loose and strict sense of terms in his 1976 _Person and Object_. Actually, Chisholm introduces the idea, first, in his paper "The Loose and Popular and the Strict and Philosophical Senses of Identity" in Care and Grimm’s 1969 anthology _Perception and Personal Identity_. I might add that the paper by Hintikka in this volume "On the Logic of Perception" is one of Hintikka’s best works. One other detail of little importance is that he cites Moore’s paper ‘Certainty’ correctly as 1959 (although it was written much earlier) but in the bibliography it comes out 1969. And, before I forget, he references "Klempke" in the bibliography when it ought to be "Klemke." No big deal. Now a concluding remark with a little more substance. 

At one point, (p. 34), Bruce says that with respect to the loose sense of ‘know’ I may be said to ‘know’ p without believing p. Now I’m inclined to have my doubts about this. While I agree on the loose and strict distinction, I’m not sure where the looseness ends and the strictness, really, begins. There are it seems two sense of ‘believe’. Suppose I see that my wallet is missing and I see it hanging out of Mojita’s pocket (‘Mojita’ is in fact the name of a dearly departed friend). It has a unique design which I carved into it and a distinctive color and I had only shortly before handed it to her to prove I’m broke. I say ‘Mojita I do believe you have my wallet’. This is one use of ‘believe’. Now another. I am at an air show. I see a plane. I see a plane coming in our direction. I say ‘I believe that is a DC3’. Now I often have confused DC3s and DC9s at a distance, otherwise I would have said ‘Hey, here comes my favorite plane, a DC3!’ But in this case I’m not sure. In the first case I have the sense of ‘believe’ that I most often associate with knowledge; in the second case, I have what I would call a "hedging" sense of ‘believe’. Now the difference between loose and strict sense of ‘know’ may very well be tied to this. After all, the evidence we have is evidence for believing not knowing. I do not say ‘Why do you know that’ I say ‘Why do you believe that’. This question may arise when I say ‘I know such and such’ not ‘I believe such and such’. 

Anyway, this is my first impression of the first chapter of Bruce’s book. Were I to give it the real justice it deserves I would spend a week thinking about counterexamples to his definitions ‘certain’ knowledge and ‘imperfect’ knowledge. I haven’t discussed his discussion of Kripke. While I think there may be some value in the theory of rigid designation (surely there is), I’m unconvinced by Kripke’s attack on Wittgenstein and its implications for Kant. I’ll devote a separate post to this unfortunately widely accepted gambit. Whoops! I see that is in the next chapter. 


Steve Bayne 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rbjones.com/pipermail/hist-analytic_rbjones.com/attachments/20091007/810d1729/attachment-0002.html>

More information about the hist-analytic mailing list