A pointer to an explanation we could all use from time to time

[johnny9fingers]

bord_du_rasoir  posted this link on politicartoons  which I thought I should share with folk as it is both lucid and elegant. It gives us a clear idea of the difference between a law of nature and a theory.

http://wilstar.com/theories.htm

For the full context of the debate and explanation it's here:

http://community.livejournal.com/politicartoons/2188731.html

Which starts off with a crack at Christine O'Donnell (admittedly a bit like shooting fish in a barrel) and proceeds from there.

Comments

A bit of logic and philosophy of science.

[ankh156.livejournal.com]

All scientific theories and laws are provisional - and thus remain as hypotheses. They become 'corroborated' by the usefulness the scientific communities who use them find them to possess. A theory or law, which should allow the possibility of being falsified, however true it seems, and however many times it appears to have resisted falsification, is still only awaiting the evidence which will falsify it conclusively. Some scientiific theories or laws are found to be so useful that they are affectionately retained as 'rules of thumb' despite having been falsified. The 'problem of induction' demonstrates the assymetry between verification and falsification. No amount of verifying instances which support a theory will verify it ultimately. However, a single clear and disconfirming instance will falsify it for all time.

Cf : David Hume, Emmanuel Kant, Karl Popper, and even Sir Alfred Ayer.

I am in so way supporting the position of Christine O'Donnell in all this. Scientists cling somewhat possessively to their theories and laws. They are human after all. The rules of logic and epistemology appear to be less accommodating.

Re: A bit of logic and philosophy of science.

[johnny9fingers.livejournal.com]

I recognise this: however in terms of "usefulness" I think the distinction is a good one. I mean, I have read Popper, Hume, et al, but when it comes to practical application of the "best possible evidence" notwithstanding falsifiability, Wittgenstein's Poker, and the nature of knowledge and the scientific requirement for workable hypotheses rather than strict logical proof, I think it remarkably elegant.

Re: A bit of logic and philosophy of science.

[johnny9fingers.livejournal.com]

To explain myself better.

On philosophy some time ago I was involved in a debate about something which escapes my memory for now, but it was probably about Descartes meditations.

I ended up blaming the structure of language, because (as I said) 'All language has properties that allow it to pose the question "which came first, the chicken or the egg?" whether meaningful or not'. And it seems to me that even the more formal the language (formal logic, algebra etc) the fault still exists. After all Russell's set paradox exists in the most formal of our philosophically abstract languages.

To a great extent this has left me a brute empiricist.
If one contends, like Ludwig did at one point, that most 'real' questions are not reducible to a priori reasoning, all we are left with is best possible evidence.

I'm prepared to be shown the error of my ways.

Re: A bit of logic and philosophy of science.

[ankh156.livejournal.com]

You can't really hold language to fault because it's capable of referring to, or even describing non-existent entities. Where would human life and imagination be without unicorns, mermaids, or even 'the furthest star' ? Following Kant somewhat, I'm inclined to believe in two basic criteria of 'truthiness'. Correspeondence and coherence. Correspondence is 'a posteriori' and wholly dependent on the evidence of our senses. The other is mathematical or logical coherence and usually ends up indication an equivalence, or even better a tautology. Taken in their pure and extreme forms neither is particularly useful or significant. 'Protocol statements' (as the logical positivists liked to call them) are bare statements of perceived affairs, and are close to meaningless if they are not situated in a collectively recognisable context, or if they are not part for some identifiable programme of observation. Equally, logical statements of equivalence (eg : 2+2=4 - which establishes an equivalence of the terms on each side of the "= sign") don't tend to be too illuminating either. Where most of human discourse, knowledge and deduction take place is between the extremes, and utilising varying admixtures of the two criteria - hence the thorniness of knowledge and truth. Usually we need both prongs of "Humes Fork" (the a prioi and the a posteriori) to have anything understandable, expressible and worth knowing.

Re: A bit of logic and philosophy of science.

[johnny9fingers.livejournal.com]

I don't disagree with your conclusion at all, but if I err, I err on the side of correspondence, because in my perception language tends to decohere: hence (I think) paradox, or meaninglessness.

I have been of the opinion that the observable and measurable set the boundaries of any coherence we can deduce: excepting perhaps in systems which are completely abstract, and which bear little or no relation to the measurable. In the world which is all that is the case, I think the 'why' must agree with 'best possible evidence' rather than making the evidence conform to the 'why'.