[lit-ideas] Re: Understanding Why Newton Contributed To Human Knowledge With A False Theory

Donal disputes that (and here I'll tamper with the original example by substituting 'p' for 'not-p') 

S. A knows p only if p.

He asks
What is disputed here? 
It is not disputed that 'to know' can be defined in such a way that that 'to
know x' entails 'x (is true)'. But 'to know', and its offshoots - including
the term 'knowledge', can be deployed without this restriction.
If this means that people often use the word 'know,' and 'knowledge,' in various ways, ways which do not depend on any very strict conception of what it is to know something, this is certainly true. 'I know I promised I'd take you to the zoo, but something's come up, and I now I can't.' 'Schrödinger said he had no knowledge of the cat's whereabouts.
But the very idea of knowing is that it is opposed to something, namely, believing, hoping, guessing, surmising, wondering, predicting, and so on. To say that someone knows something is to mark a distinction, a distinction that's been around at least since Plato struggled with it in Meno and Theatetus. The conception of knowledge that underlies S also underlies Aquinas' struggle with the problem of God's knowledge of future contingents. If God knows everything, then what he knows must be true, and if true, for any statement about some future condition you care to mention, thus fated. If God knew everything right from the start, he also knew that 'Adam will eat that damned apple,' was true. The very fact of his knowing it makes it unavoidable.
I mention Plato and Aquinas as examples of philosophers who want to distinguish knowing from something else and, in Aquinas' case, to get around somehow the standard conception of knowing, which is encapsulated in S. The standard conception is not exhausted by S: S simply sets forth a necessary condition for knowing.
S is independent of whether knowledge is justified true belief; it avoids disputes about Edmund Gettier, Eastern mystics, Bertrand Russell, G. E. Moore—that whole crowd. Whatever conception of knowledge one has, it must satisfy S. Just as Tarski's notion that 'P' is true, iff P, leaves it to the truth-seeker to use his or her own favorite way of determining whether P, so S leaves it open as to how 'p' is established. I really should stop, but Donal's post suggests that I should go on for a bit. 

Donal says


The definitional/conceptual question is, I suggest, only worth considering
insofar as it reflects an underlying and substantive dispute as to how we
should view 'knowledge' and 'knowing'.
Let me repeat myself. Any view of knowledge or knowing which is not in accord with S is a different conception of knowledge or knowing entirely. S is not a mere definition. It is at the heart of the distinction philosophers draw between knowing and some other epistemic state. 


First two points concern ordinary usage: (a) ordinary usage cannot be
decisive of how 'to know etc' should be deployed _even if ordinary usage is
unequivocal in how the term is deployed_: this is because ordinary usage may
reflect or conceal a mistaken view of how knowledge works; (b) ordinary usage
is in any case equivocal.
I'm not here appealing to ordinary usage, which is notoriously polychrome, as a Harvard philosopher once sai. 


There are examples we might point to where, as against the assertion 'I know
x', we might use the falsity of x to argue/insist that 'You might have
thought it, believed it etc. - but you did not know it.' But there are many
examples where we deploy 'to know' without the implication as to truth. A
witness in a murder trial testifies 'I know the accused was out of country at
the time of the murder'. When questioned, the basis for their 'knowing' is
that the accused told them they were leaving the country, the witness left
the accused to the airport/saw their plane ticket/ waved to them as they
passed check-in etc. in the hours before the murder. As it turns out, the
accused doubled-back having checked-in and was not on the scheduled plane.
This is at best a case in which the witness would surely say something alone the lines of (I don't want to coach the witness) 'I believed it at the time, but now I see I was wrong.' In any event, this speaks to how it is established that p, and is concerned with evidence and grounds. Claims to knowlege, as here, do not defeat S; in fact, they seem to depend on it. How else could the opposing side establish that the witness didn't really know what he thought he knew? 


But is the witness lying when the witness says 'I know the accused was out of
the country'? Surely not. Is the witness mistaken? Well, that depends on how
'to know' is defined - but surely it is ordinary understanding to read the
claim 'I know x' as here meaning 'I believed, on what seem to me sufficient
grounds to justify a knowledge claim, that x'. The witness may be mistaken
'that x' but, on this view, is neither dishonest nor mistaken in claiming 'I
know x'.
What 'justifies' a knowledge claim (and the Gettier example would work well here) does not establish in any particular case, whether the one who claims to know, knows. Justification here is a slippery notion; I might be justified in saying or doing something insofar as I believe I have really good grounds for saying or doing it and yet, of course, be mistaken. This isn't really news in everyday life or in philosophy. 


If we accept that all knowledge is fallible, and insist that 'I know x'
entails 'x', then all claims of the type 'I know x' are potentially mistaken
- since they are liable to be overturned by showing that 'x' is false. If
this is so then on this definition should we always preface 'I know x' with
'Of course, I may be mistaken but..'? But this seems self-defeating.
I really don't follow this. You've yet to show that 'all knowledge is fallible.' If I have a father, I'm pretty sure I had a grandfather. I know this. It isn't likely to be overturned by anything short of science fictionish objections. If this is too definition-like, then I know that if trout are deprived of oxygen they will die; that square pegs don't fit into round holes; and that no human being can lift a five-ton (tonne?) weight unaided. What might falsify these things, I have no idea, but the mere formula 'all knowledge is fallible' (itself somewhat incoherent suggestion) is about as interesting as global scepticism: we doubt for the same sorts of reasons that we believe, so one wants more, in that case, than a claim, made in a vacuum, that 'nobody really knows anything,' 'nothing is certain,' etc. (I'm trying to draw a parallel between 'all knowledge is fallible' and global scepticism; I don't want to say that Donal's formula was advanced in a vacuum. 


The reality is that all we need to do is clearly distinguish 'knowledge' from
'truth' and accept that all knowledge is conjectural. On this basis it is
supposed truth of 'x' that may be overturned by showing 'x' is false. But the
historical fact that 'x' was believed with what were thought sufficient
reasons, and in that sense 'x was known', is not overturned.
It is conjectural that if I touch the tip of my nose with my forefinger my elbow will be bent? Interesting. Again, I think you've conflated it's being true that sugar is soluble with sugar's being soluble. The former says no more than the latter although it had led philosophers into many dark alleys in the belief that 'is true' somehow helps a sentence along. You need to distinguish, I think, between 

B knows he has a baby brother (in 1951)

and

B knows there's he has a baby brother (in 1949)
Just in case the brother was born in 1950, something that was or could be known at one time wasn't and couldn't be known at another.
Knowledge claims, and knowing is indexed to times and places. Nobody knew at any time that phlogiston was necessary for combustion, although apparently at one time some scientists believed it did and thought they knew it did. And nobody knew at any time that the heavenly bodies were embedded in celestial spheres. But somebody did know, at some time, that the cat was on the mat and the pig in the poke. That we discover that what we once thought we knew isn't, and wasn't the case, the point of claiming that we knew it then, is unclear. 'It was once thought (believed) that q,' is a perfectly good expression. 


A question:- given that Newton's theories are false, must we conclude that
anyone who claimed to 'know' that they were right did not 'know' this? Must
we further conclude that Newton's theories were no contribution to human
knowledge because, by definition, no one ever 'knew' the theories to be true?
If they were a contribution to human knowledge, despite being false, in what
way are they knowledge?
Not all of what you call Newton's 'theories' are false. With slight tinkering, his laws are perfectly adequate for getting robots to Mars, building bridges, and other mundane and spatial tasks. I don't really though want to worry about Newton. 


These questions may lead to others that bring into overall question whether
we should conceive knowledge in terms of justified true belief ['JTB'] and/or
assume that 'to know x' entails 'x is true'.
I would hope they do and that they also bring in others to discuss them. A final note. What you have above seems ill-formed in the following way. In what you say is entailed by knowing x, x should now be in quote marks,' to distinguish it as a statement which is said to be true. But I've already said that to add 'is true' to p is otiose, unless one is trying to assign truth values in propositional logic or answer true-false questions on a quiz. 

Robert Paul

------------------------------------------------------------------
To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html

Other related posts: