Tuesday, January 31, 2006

impossible omniscience

Absolute knowledge is impossible to obtain. It requires the conjunction of both perspective-dependent and independent knowledge, which is inconsistent. The existence of perspectives in reality generate what might be called “worlds within worlds”, for which no perspective-independent knowledge can be achieved.

The very existence of perspective generates a whole set of facts, the perspective-depending ones, such that any objective knowledge is de facto incomplete. Cases of these knowledge abound the literature: self knowledge like knowing who you are, where you are, when are you reading this, how you feel, etc. The only access to this kind of knowledge is given by means of a perspective.

However, being endowed with a perspective is also a hindrance. For there are also perspective-independent facts for which there is no metaphysic access from any perspective. These also abound: empirical knowledge like knowing that there is charge, that the house is white, that Ann Arbor is North of Mexico City. What is said when we talk about these cases is timelessly true; it is so independently of any perspective or any context. The best access any perspective-dependent subject has to these facts is that of induction.

Now, the problem is this. Supposes there is something like a being X that is omniscient in the objective sense. To be so, such being must lack any perspective, and posses what we nowadays take to be propositional knowledge. By hypothesis X knows up to every detail where in logical space is the actual world. X knows all the propositions that are true of the actual world. However, the price of propositional omniscience is high: in order to achieve such knowledge X must lack a perspective from which to know all that it knows. And so, X is also precluded from knowing all the perspective-dependent facts. Hence, X is not able to know, for instance, what it is, where it is, how it is, and when does it exist. Propositional omniscience implies self-ignorance.

If one would try to endow X with a perspective then the self-ignorance problem would be solved. Needless to say, this would also imply a lack of objectivity in X’s propositional knowledge. Thus, self-knowledge would come but, also, at a very high price: propositional-ignorance.

These considerations are enough to conclude that something like an absolute-omniscience is inconsistent. One and the same knowing entity cannot at the same time posses all propositional knowledge as well as self-knowledge. Either one or the other is available, but not both. Hence, omniscience is only possible at a relative level.

Something interesting concerning the very notion of subject is hidden among these lines. For there is no knowledge without a subject of knowledge, and this in turn forces us to consider a certain perspective. Knowledge cannot be identified with reality. Knowledge necessarily implies the existence of consciousness, and this in turn implies a location although perhaps not a time. At the most, there can be knowledge from a timeless location, but it is still knowledge from a certain location. Thus, all knowledge implies a perspective.

Thinks of Lewis’s gods (see Lewis “Attitudes De Dicto and De se”). They both have propositional omniscience, and they both lack self-knowledge for they do not have access to their particular perspective. One leaves on the top of the highest mountain and throws down manna, the other on top of the coldest mountain and throws down thunderbolts. None of them, however, knows who he is. For all that the manna-god knows he might as well live on top of the coldest mountain.

Two things should be kept in mind. The propositional knowledge of these gods is not empirical knowledge (i.e. knowledge from their mountain-like location). It is, in the clearest way, a priori knowledge. Furthermore, the self-ignorance of these gods is necessary, for their physical instantiation cannot be identified with them. If so were the case, Lewis’s gods would be endowed with a perspective and, thus, with a hindrance precluding them from being propositionally omniscient.

But now that we know that Lewis’s gods are not identical with the entities living on top of the tallest and coldest mountains, we might doubt whether such thing as propositional omniscience is possible. I have claimed that all knowledge is located somewhere in reality, and that being located implies having a perspective. I have also claimed that perspective-dependent knowledge implies propositional ignorance (or at most propositional omniscience with a great lack of certainty: the possibility of a skeptic scenario is metaphysically necessary according to these conjectures). Reality is such that there is no direct access to propositional knowledge, and so, propositional knowledge is possible only by inductive means. It follows that all knowledge is necessarily uncertain, for it must carry some sort of propositional ignorance.

Our first conclusion is that absolute omniscience is inconsistent. Our second conclusion is that propositional omniscience (if by omniscience it is implied a strong degree of certainty as that given by deductive inferences) is impossible. Any subject of knowledge is necessarily endowed with a perspective, and propositional omniscience is incompatible with perspective-dependent knowledge.

It would be interesting to consider whether self-omniscience is possible and, if so, whether this might allow a conscious withdrawal of the self (epoche, identifying one with everything, loosing identity or else) such that the perspective is put into parenthesis. Perhaps this artificial way of achieving perspective-independence may endow us with a greater degree of certainty. I doubt, however, that this could still be tantamount to something like propositional-omniscience for one main reason: if such exercise were possible it would imply the loss of self-location and this, as we have seen, is a necessary condition for knowledge. The loss of self also implies the loss of knowledge. Perhaps that is why many praise that ignorance (absolute) is bliss! I’d rather know something, at least.

Saturday, January 28, 2006

philosophy, the history of one problem

After reading Fodor and Fichte, trying to avoid schizophrenia, I invented a coherent way to interpret all (or pretty much) of the history of philosophy. I'll be simple and state the interpretational claims in a rather easy way:

1) There is one and only one main question: what is the content of a sign?

2) There are at least two subordinated questions: how does something become a sign and have the content that it does? And what is the content that makes such a sign the sign that it does?

3) Answers to this questions give place to all the problems philosophy deals with. The most natural answer is to say that there is a causal relation, but since we can 'actually' mean something beyond what 'actually' exists, this has not been a satisfactory answer.

4) So we get different accounts:

a) Parmenides (very sensibly) eliminates change in the world and so eliminates change in the causal relations and errors in the content of signs. The problem here is not that there is no change, but that there is no time and no freedom (or so some claim).

b) Plato wants both sides of the coin. So he eliminates time among the ideas and assumes time among experience. So we get the content from the ideas, and time, change, and freedom from the empirical world. Problem: this is inconsistent. We cannot fix content from a world (topus uranus) of which with which we have not causal relation. So Plato assume innate knowledge.

c) Aristotle wants both sides of the coin but within the actual world. So he gives us the actual content of representations by means of the 'actual' properties of things and the other non-actual contents by means of the 'potential' properties of things. So we start getting close to possible worlds, but not quite, since everything exists in the actual world.

d) Jump to Descartes (sorry!) and you get that everything is back to the mind, but this time without the topus uranus. So this is some kind of regress for what we are looking for is the world. The world as the content of our representations, at least. So, naturally, what Descartes gives to us is nothing, literally nothing. We are certain that we know, think, believe, or desire absolutely nothing.

e) Then you get the fabulous Hume. The problem is that we have been assuming a very strong connection between our expressions and the world. But there is no such relation: all there is to semantic relations is a mere 'association' between ideas. There are, as it seems, no causal relations between signs and content, mainly because there are no causal relations at all.

f) So Kant wants to save the disaster by claiming that it is, in fact, a mixture of all these ideas, but without the topus uranus. There is in fact a causal relation between our signs and their content. But 'causal relation' is a human (subjective) notion. Thus, what we refer to is the world of experience. Problem: we want to know the world, not just the world of the experience: paradoxical solution, there is such world, but we cannot know it.

g) So we get Hegel and so everything is solved; but no one has decided how everything is solved. Either it is all mental, or it is all physical. What is clear is that everything is solved within the world that Hegel talks about. What is not clear is what world is Hegel talking about. The threat is that he is not talking about the actual world.

h) For brevity's sake let us jump to our contemporary philosophy. After the so called linguistic turn (as if everything was not linguistic since the beginning) we have philosophers of mind, metaphysicians and philosophers of language discussing whether the relation is causal, logical or what. Whether it is possible worlds, biological functions or informational (or computational) states that work around here. We still don't know.

5) What we know is that whatever is an answer to the question of how the content is determined, implies an answer to what is the content. You might as well be sure that if you solve the how (e.g. by solving the mind-world relation, perhaps by means of causal relations), you will therefore get the what.

The problem nonetheless, has been, from the very beginning, the how...

6) If you wonder why, think of the following consequences: if there is no how, then there is no knowledge, no morals, no aesthetics, no theory, and no meaningful expressions whatsoever. And if there is no 'what' then...

What am I talking about?