What justifies logic?

Jacob McNulty has an article at IAI arguing that the foundations of logic can only be found in metaphysics. (Warning: possible paywall. Alternate link.). He describes a problem called “the logocentric predicament,” that any attempt to justify logic with logic ends up being circular, risking an infinite regress. He notes that the most common response to this historically has been complacency, with Aristotle trying to just dismiss anyone questioning the law of noncontradiction. However subsequent developments in alternate forms of logic apparently make this a difficult stance.

McNulty ends up going through the views of Kant, Fichte, and Hegel. The discussion gets thick with the terminology of continental philosophy. As a result I struggle to understand exactly what he’s selling. But the overall point is that logic can only be justified with metaphysics, by which he means the “ambitious area of philosophy which strives to know God, the soul, the world.”

I actually don’t think this is true, at least not in the sense of requiring pure metaphysics. But it requires doing something logicians may resist, bringing in empirical information. Part of my willingness to do this comes from a comment logician Graham Priest made several years ago in an interview. He was arguing for alternate systems of logic, which he saw as justified because these systems can be seen as basically very basic theories of reality. And reality being complex, can often be viewed and understood through the lens of multiple theories.

One view, which McNulty mentions as problematic, is psychologism, the idea that logic is based on how we think. He argues that the view, “elides the distinction between how people ought to think and how people in fact do.” But this assumes that common thinking isn’t logical. It may not be at one level, the level of social interactions and decision making. But even a computer system, an inherently logical system, will produce results that seem illogical if it has bad or incomplete data, or isn’t working properly. Once we take into account bad or incomplete beliefs, along with various physiological impediments, it’s not hard to see why people often seem illogical too.

Interestingly enough, in the philosophy of logic, psychologism is often regarded as an anti-real stance, as opposed to the realism, which sees logic as existing independent of our minds. But the anti-real view of psychologism, I think, suffers from not continuing the chain of reasoning. Thinking, like digestion, at its most basic level isn’t learned. We just do it innately. Why do we think the way we do? Because it works, providing a survival advantage, which of course is why it evolved. Which means that it has a relation to the environment.

And the stance of psychologism being anti-real likely predates an example we have today, logic machines, like the one you’re using to read this. People argue about whether the mind is physical, but I’ve encountered few people asserting the operations of my phone or laptop aren’t. David Chalmers in his book Reality+ describes computers as causation machines. But I just described them as logic machines. Which is it? My take is that they’re one and the same, which I think gives us a clue to what logic is.

I’ve often referred to computation as distilled causation. Along those lines, I think logic is abstracted causation, or perhaps more fundamentally, abstracted structures and relations that exist in nature.

Of course, a platonist might argue that logic, math, and other abstractions are the more primal reality. But abstract objects in contemporary platonism (as opposed to Plato’s original forms metaphysics) are acausal with no temporospatial extent. If they exist, it’s a different type of “existence” than the patterns in our minds and environment, and don’t seem able to have any effects on those patterns. It’s a view that seems very vulnerable to Occam’s razor.

So logic is based on how we think, and our thinking evolved to resonate with common and repeatable patterns in nature. Logic, in all its various forms, captures these resonances, allowing us to optimize them, bottle them, and put them in our tools, with increasing effectiveness.

Unless of course I’m missing something. Are there problems with my view of logic? Or alternatives that work better? Or is McNulty right that we have to get into the “ambitious area of philosophy which strives to know God, the soul, the world,” to figure this out?

Featured image credit