[wsy]

> just about every culture distinguishes between empirical and logical truths

Actually, having this distinction is quite young even in our own western culture: depending what exactly you mean, ~300 years (Kants distinction between "Synthetisches Urteil" vs. "Analytisches Urteil"), or 150 years(Gottlob Frege's 'invention' of formal logic). And modern philosophy of science is already doubting it again (e.g., W.V. Quine).

[dvt]

Nope. Descartes, Leibniz and Spinoza were rationalists. Hume, Berkeley, Locke were empiricists. This distinction wasn't formalized until relatively recently, but the intuition is very old.

In antiquity, Pythagoras and Plato fall in the former camp; Serapion of Alexandria and Philinus of Cos fall in the latter.

[wsy]

I have no expertise in ancient philosophy, so can't comment on these philosophers.

Is 'cogito ergo sum' a logical or an empirical truth? For Descartes it was a logical truth (undoubtably true). However, from the point of view of Hume, it must be viewed as empirical truth, because it is based on an empirical observation. So if even philosophers of that time couldn't agree on the distinction between logical and empirical truth, I think I can safely conclude that this distinction was not at all an established part of the Western culture at that point of time.

Rationalists vs. Empiricists is what we call these groups of philosophers now, it is not that they grouped themselves in two schools of 'logical truth' vs. 'empirical truth'. In fact, Hume most probably believed that his own philosophical insights can be derived just by thinking, so they must be logically true. And still, this is quite different from what we today call logically true (i.e., formally provable).

[dvt]

First off, Hume rejected Descartes' cogito argument. Second of all, in Sceptical Doubts concerning the Operations of the Understanding[1], Hume makes a very rudimentary distinction between "Matters of Fact" (what we'd now call logical truths) and "Relations of Ideas" (what we'd now call empirical truths). Again, these classifications were intuited long before Hume/Descartes, but they exemplify them pretty well.

[1] http://www.livingphilosophy.org.uk/philosophy/David_Hume/Sce...

[wsy]

You claim that "the now well-established distinction between a and b was intuited long ago", but in the discussion refer to various a'/b' distinctions that separate the whole into quite different parts than today's a/b distinction.

I concede that expecting predecessors to split the whole exactly along today's a/b border would be asking too much. On the other hand, there is no objective measure that would allow us to test if a'/b' is similar enough to a/b to support your claim. So if you think it is similar enough, and I think it isn't, it seems to be more a matter of opinion than a matter of knowledge.

And my argument on "cogito ergo sum" was not about if Hume would consider it as true or false, but if he would consider it as logical or empirical statement. So your remark about that is true but irrelevant here.