Causation

Hume N Perception
Aug 25, 2017 · 4 min read

I’ve been listening to a couple of John Searle’s lecture series lately. They are posted online in various places. Searle insists that David Hume is wrong about causation being the constant conjunction of events. I won’t go into details about his opinion about causation, but I am struck with David Hume’s. It seems correct that any explanation of why something (B) happened can be reduced to “Because A happened, and A is necessarily followed B.” Why is it necessary for A to be followed by B? Because it is always followed by B.

Searle complains that this doesn’t explain one-off event pairs. When a passing bus honks its horn and I startle, I don’t have to have that experiment repeated multiple times to know that there is a constant conjunction. The pair of events happened once, that is enough for me. However, I think that we have observed enough similar experiences where loud noises startle us, since we were babies, that now we are almost 100% certain that the startle was caused by the horn. If we had the presence of mind when we were babies to observe ourselves, we wouldn’t be so certain that we were startled by the bus honking its horn. Maybe we might think we were startled because the breeze picked up, or because the person carrying us shrugged, or maybe we always startle at noon on the 27th of August, and we’ve only experienced that date once. We can’t be certain, because there hasn’t been enough of a correlation between the putative cause and effect yet.

Even if causation is only constant conjunction, some explanations that are equally conjoined seem more satisfactory than others. Some seem like realexplanations. Searle talks about gravity causing a piece of chalk to fall to the ground. He says that gravity is a fundamental causal force. He contrasts this with voting for a political party. Why did he vote for Obama? Well, his socioeconomic status — being a middle class professor in the humanities of a well known public institution — virtually guaranteed that he would vote for the Democratic candidate. Does that explain why he voted for Obama? He says it doesn’t, even though there is a nearly constant conjunction of people of his socioeconomic status voting for the Democrat. In this example, event A is ‘having his particular socioeconomic status’, event B is ‘voting for the democrat’. A nearly always proceeds B, but we don’t consider A to be a cause for B. The real cause (according to Searle) was because he thought Obama would be better for the economy.

But let’s break that down. Why did he think Obama would be better for the economy? Without going into models of cognition that explain beliefs and desires, his neurons are wired up in a way that gives rise to that belief. If you took a different brain which also had the wiring which gave rise to that particular belief, the owner of that brain would also have voted for Obama. The wiring between the brain’s of these two people would look very different. They might have more or less neurons, their representations of ‘Obama’ and ‘the economy’ would be stored in different areas of the cortex. But there exists a way of representing the networks of those two brains which would show a kind of isomorphism between the networks. And you could say that anyone with this same set of properties in their neural architecture would also vote for Obama. The neural network, then, caused Searle to vote for Obama.

This seems to be a more satisfactory causal relationship than Searle’s socioeconomic status. Why is that? I suspect it has to do with our experience. We have a lot of experience with gravity. Every single time we have dropped an object heavier than air, it has fallen down. We’ve been dropping things all our lives. We are so certain of this fact, we would stake our lives on it, and we regularly do. And yet we are never 100% certain. It always remains a possibility that gravity will stop pulling things down in the next second. No one can rule it out with literal 100% confidence. But we are much more certain of gravity than we are of certain demographic’s voting patterns. Not only do we not have as much experience with it, it also has a more probabilistic nature, so we consider it less of a cause and more of a factor. The neural explanation of Searle’s behavior may seem more satisfactory than the demographic explanation because we are more familiar with physical, mechanical systems than with analysis of populations and political parties. But what if the situation were reversed, and from the age of 1 month we were much more concerned about political abstractions than learning the business of interacting with our physical environment. Maybe then the demographic cause would have more explanatory power. “Professors always vote Democrat, without exception, since time immemorial,” the 8 year old version of you would say.

When we explain things, we seek to put them in terms we are more familiar with. Good speakers do this all the time. You can give your audience the illusion of understanding just by comparing the thing you’re trying to explain with a thing that is common place, even if the two things have nothing in common. And real explanation consists of convincing yourself that B was inevitable because it always follows A, I’ve seen it or something like it innumerable times myself.

The universe contains patterns. Not all events happen randomly. Our brain evolved to detect these patterns. When we find them, we can use them to make predictions. For most of our lives we are confronted with things we don’t understand, but we try to. And when we have enough experience, or we find a similarity between the new event and our experience, then ‘click’, we find we understand what caused what. Some may say that extrapolating patterns from the past and present into the future is delusional. But my extensive experience tells me otherwise.

)
Hume N Perception

Written by

perception, reasoning, induction, neural networks, artificial intelligence, categories, concepts, analogies, unsupervised learning, pattern completion

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade