Deduction, subjectivity & culture are all real enough

People and philosophers tend to get bogged down in binary oppositions like objective vs subjective, nature vs culture and inductive vs deductive knowledge. Continental philosophers tend to complicate such binaries, trying to show how both terms are inadequate. Analytic philosophers tend to talk themselves into the ground trying to prove or disprove the existence of one of the terms.

But I think the unusually practical philosopher Daniel Dennett goes about these things the right way and his method is quite generally applicable.

Take the inductive/deductive debate in the philosophy of science and epistemology. Is all learning inductive (based on experience) or can we learn by performing deductive reasoning without reference to experience? This argument is old as fuck ¹⁾See for example Kant’s analytic vs synthetic knowledge and the related binary of a priori knowledge vs a posteriori knowledge. .

Instead of trying to boldly claim that deductive knowledge is impossible (roughly empiricism or positivism) or that deduction is the surest way to true knowledge (roughly idealism and rationalism) Dennett would note the evolutionary history of these things.²⁾Dennett’s approach is underneath a lot of his writing but is best illustrated in his books Darwin’s Dangerous Idea (1995) and Freedom Evolves (2003). The former traces the evolution of species, reasons and meaning; the latter looks at the evolition of selves, agency and free will.  There must have been, in the long history of the world, inductive knowledge before there was anything claiming to be deductive knowledge. Simple organisms embody inductive knowledge, i.e. knowledge about the environment accidentally gained by earlier generations through experience and encoded in their genome. Before that, experience wasn’t even used and really there was just some randomness and chance that allowed some self-replicating molecules to accrue. Before that there was just energy and before that there may not have been anything at all.

If we have deductive knowledge today that can be usefully separated out from inductive knowledge, it has at least evolved out of inductive knowledge. We seem to be able to deduce statements about hitherto unthought of mathematics (as can a computer). This seems like a good candidate for deductive knowledge, but obviously it has to ride on the shoulders of a whole heap of hard-won inductive knowledge. You have to have a brain or a processor to even begin the exercise and these things are too complex to spring up by chance, so some heritage of consolidated inductive knowledge is necessary.

We can also now conjecture about things we’ve never experienced. Is this deduction? Well, some hardcore inductivists might say that even the insight that future things exceed experience, is itself a lesson learned from experience. Thus we know from previous disasters, say, that unprecedented black swan events can come along that go beyond anything witnessed before. Is this pure unadulterated deduction or just souped-up induction? Uncle Dennett would probably say, “Don’t look for a bright dividing line at twilight even if there is a useful difference between night and day.” In other words once you get to these high level investigations of knowledge, this latter deductive type of knowledge seems like a pretty different species to induction. But if you try and tease it apart and find the liminal case, it will be too blurry to adjudicate.

So I reckon we have developed knowledge that we might as well call deductive, but that it evolved out of the older inductive kind. I think that induction was a necessary condition for the first epoch of knowledge creation but is now an insufficient condition for further knowledge creation.

And of course in the future we will probably develop even more elaborated ways of coming up with knowledge that are further away from homespun induction (simulations, learner algorithms, ways of leveraging stochastic processes, etc.), and these will in turn lead to more elaborate ways of learning, but they will all still have emanated from the foundation of induction. But call them something different, because for practical purposes they are.

This same dynamic applies to culture as well: it came after and out of nature. If you want to be a stickler then even all of human culture is really still part of nature because we are a natural species and part of the biosphere. But there’s a useful distinction nowadays between the products of human culture and everything else in the biosphere. Environmentalists and ecologists tie themselves in knots over this. I feel a Dennett-style analysis works pretty well. Again it involves accepting that there will not be a bright dividing line. (Were early Homo sapiens natural or cultural? Are traditional peoples living today, but cut off from modernity, natural or cultural? etc.)

And finally, ditto for objective/subjective, or indeed object/subject. The subject (a conscious person) is made up of objects, chiefly cells. There was a time when there were only objects and no subjects. Gradually subjects evolved from objects. Similarly we can say there are subjective claims about the world or subjective experiences, but these too presuppose objective events that gave rise to them. If you look hard enough there will be tricky cases that blur the distinction depending on how strict your definitions for each term are. (Can chimps have subjective experiences? Lizards? Bacteria? Viruses? Carbon atoms?)

The evolution of species is the best lesson for this general kind of reasoning. It reminds us that things happen gradually over time and that differences or distinctions between things (in the broadest sense) arise as complexity increases, but had to start from somewhere — from much simpler forms — and if you go back far enough the distinctions disappear. Such is the case for essentially all phenomena in the universe. Because complexity really does exist today, and yet really didn’t exist before, there was a history of its evolution in which it’s impossible to draw clear dividing lines.