Live data from Hacker News

The Dream of Enlightenment by Anthony Gottlieb review

theguardian.com

51–52 of 52 posts

Re: The Dream of Enlightenment by Anthony Gottlieb review

#51

Earlier quoted context omitted.

"After all, scientifically speaking, everything beyond our sensory nerve-endings is latent and is inferred." This is not a scientific claim, it is a philosophical one, and a dubious one at that ( http://plato.stanford.edu/entries/sense-data/ ).

You don't seem to understand. I wasn't making a philosophy-of-mind claim about philosophical sense-data. I was making a claim, a very basic claim, in neuroscience. In some theories, the nerves don't even pass unaltered signals up to the brain! Everything beyond the sensory nerve-ending itself may involve some amount of inference, inferences performed in the spinal cord and sensorimotor cortex themselves long before a…

You may not think you're talking about sense data, but you might as well be.

Science doesn't operate from the first person perspective. The data of science is not neural firings, it is public observations and measurements.

The sense in which a subatomic particle is inferred from its effects is completely different from the sense in which the brain "infers" things based on neural signals. The former is part of theorizing, the second is a bit of information processing that's prior to theorizing of any kind. The former is a fact about how science works. The latter is a scientific discovery which does not (directly[0]) affect how we do science.

[0] I say directly because there are contexts where we have to account for bias, optical illusions, and whatnot. It's the same sense in which astronomy is not directly about telescopes, though astronomers spend a lot of time worrying about telescopes.

Re: The Dream of Enlightenment by Anthony Gottlieb review

#52

Earlier quoted context omitted.

You don't seem to understand. I wasn't making a philosophy-of-mind claim about philosophical sense-data. I was making a claim, a very basic claim, in neuroscience. In some theories, the nerves don't even pass unaltered signals up to the brain! Everything beyond the sensory nerve-ending itself may involve some amount of inference, inferences performed in the spinal cord and sensorimotor cortex themselves long before a…

You may not think you're talking about sense data, but you might as well be. Science doesn't operate from the first person perspective. The data of science is not neural firings, it is public observations and measurements. The sense in which a subatomic particle is inferred from its effects is completely different from the sense in which the brain "infers" things based on neural signals. The former is part of theoriz…

> You may not think you're talking about sense data, but you might as well be.

Ah, isn't the imperialism of philosophers so likeable? Hey scientists, you're not talking about the stuff in your experiments, you're talking about this other thing we made up! Because we say so!

>The sense in which a subatomic particle is inferred from its effects is completely different from the sense in which the brain "infers" things based on neural signals. The former is part of theorizing, the second is a bit of information processing that's prior to theorizing of any kind. The former is a fact about how science works. The latter is a scientific discovery which does not (directly[0]) affect how we do science.

Actually, according to our most up-to-date paradigm, they're both instances of hierarchical Bayes inference, one of which is implemented in hardware, and the other of which is using said hardware as a primitive base to build higher-precision inferences.

For someone trying to stick up for philosophy, you should read more of it, since I'm getting many of the details from Andy Clark's recent book.

Post reply on HN