-Knowledge is to be in another (i.e. to be in the subject) as other (i.e. as an object). Simplified, it’s to be in a subject as something other than the subject; to be in the subject as object.
-Any number of epistemological riddles arise from seeing knowledge as being in the subject as subject, that is, to be a modification or mere accident of the subject. “To know something” on such an account means to possess a fact about oneself. This account itself seems incoherent so far as it appeals to the division between a fact and the one possessing it while denying that there is any such division.
– Aristotle spoke everywhere of “the what it is” or “the what it is to be”. The Medievals learned Aristotle by these quirky substantive pronoun sentences, but then coined a word “essence” and made it identical to another word that would have been better left distinct: nature. But “essence” is just shorthand for the pronoun. Our attention should have been drawn to the “is”. The word is the nexus of all affirmations and denials, and therefore of all truths for us, and yet it has a fascinating declension of meanings. I find that whenever my kids ask me “what things are X” I find myself casting about for things that are the X at least in the sense that Aristotle calls said of all (kata panton) but what I’m really looking for are the things said per se. At our most precise, we want even more in addition to this, what Aristotle called kathalou, or “universally”.
Now there is certainly something about things that makes them such that we can have this declension of meanings for “is’. If you want to call this thing “an essence” that gives rise to the per se, then fine. But our initial access to this thing is through our attempt to speak about it and articulate what we know, and this frequently will involve a larger class of data than the word “essence” is appropriate for, even though they are exactly what Aristotle wanted to include among things that were later called “essences”. The division between correlation and causality, for example, is an appeal to the per se, and eventually toward the per se and kathalou. Correlations can never be more than said of all. The desire for a causal connection is thus an aspect of “essence” in the sense that Aristotle is insisting on it, even though the word does not tend to include this in its field of meanings. Again, the long history of trying to figure out what heat is was a continuous attempt to figure out not only what was hot per se (we always knew that fire, the sun, living mammal bodies, etc. were hot per se) but what was hot per se and kathalou. Again, penguins or living things evolve per se, but they are not what evolve per se and kathalou, and a good deal of what one thinks about evolution can be muddled by treating something that is non-kathalou as though it is.
Now there is a good deal of controversy over whether the sciences attain the per se and kathalou simply speaking, or whether all they are interested in is correlating events to quantitative and operational measure-numbers. Dewey, for example, claimed that a scientist telling you “heat is molecular motion” was no different than a power company telling you “heat is $110 for November”. Neither is an attempt to talk about what heat is, but only to relate it to useful units for the purposes of those who are interacting. Dewey is right that this is an aspect of all scientific discourse, and to the extent that it is, it makes no sense to distinguish the per se from the merely said of all. But to the extent that science is actually trying to explain what things are (and it simply goes to far to deny that it does) then the whole point of the quantification is to give a verisimilitude of what is per se and kathalou.
-The human mind seeking clarity and distinction – the scientist – is divided between two poles. On the one hand he has logical structures that specify exactly what would count as distinct (an ultimate division by way of contradiction of any thing from another thing – i.e. a completely made porphyrian tree) on the other hand he can approach this distinction only by way of verisimilitude arising from quantitative and operational phenomena. He has to, in other words, be able to program a machine to be able to discern the differences he is looking for. How else would they be “objective”? (I’m reminded of a Mythbusters episode that debunked the folk-proverb that “you can’t polish a turd” which concluded to the great triumph of placing an [obviously polished] turd under a polish-measuring machine and obtaining a favorable reading.)
So on the one hand we have logical categories that exhibit such strange traits that nothing about them looks mechanical, and on the other hand we only approach these categories when we obtain favorable readouts from machines.
-Artificial intelligence is not a side project or daydream fantasy of scientific method, it is the paradigm of scientific knowledge. Science just is an approach upon artificial intelligence – it just is a convergence upon the objectivity that it only trusts to machines to verify and the subjectivity that is required for knowledge at all. To the extent that artificial intelligence is impossible, science is a convergence on an unattainable limit. But artificial intelligence is impossible; not in the sense of a Turing machine (leave that sense aside for the moment), but in the relevant sense that even if it were attained, the subjectivity of the knower would still remain essentially private and thus non-objective in the scientific sense. Thus if science were to obtain the goal at which it converges, it would not attain it. “Completed science” or “an ideal knower” (where the knower knows by way of physics, chemistry, etc.) are both contradictions. If God or the angels exist, they only know physics by knowing our minds, and not by looking at things.
Just Thomism 