Brains in Vats, Cats and Dr. Deming

You need only scan the daily news to see how deeply mired we are in the frankensteinian “brains in vats” nonsense. For example, in today’s NYT we learn that Google’s Scientistic scientists have:

“…created one of the largest neural networks for machine learning by connecting 16,000 computer processors, which they turned loose on the Internet to learn on its own.

Presented with 10 million digital images found in YouTube videos, what did Google’s brain do? What millions of humans do with YouTube: looked for cats.”

and, they say…

“…the software-based neural network created by the researchers appeared to closely mirror theories developed by biologists that suggest individual neurons are trained inside the brain to detect significant objects.”

The network of computers said to be learning of its own accord, how to identify cats, is a perfect example of the “brain in a vat” thinking that is taking us down the wrong road.

Are we to suppose that a disembodied assemblage of CPUs forming a so-called “neural network”, without a single neuron to be found anywhere, decided of its own accord to look for cats?

Who was it that thought cats interesting enough to warrant searching for two-dimensional photographic representations of cats posted on the Internet? Was it the computer network or the programmers who thought online pictures of cats interesting?

The intention lies with whole and fully embodied human beings who act by wiring computers together to create a virtual “brain in a vat” and then imagine that pictures of cats on the internet have much to do with cats at all.

Dr. W. E. Deming would remind us that finding cats, say for the PURPOSE of counting them, is not as simple as it sounds.

Think about it! What is a cat? Where do cats begin and end? Is a picture of a cat a cat? How do I know when I have found a cat? When I see a cat I may not have found a cat. It might be a stuffed animal, a photo, a drawing or even a dog in a cat-skin cape. There’s a lot more that goes into being a cat than a picture on the Internet. Deciding what to count as cats and non-cats is going to take a bit of work. But first we will need to decide exactly why it is we want to identify cats because cat-ness depends upon our relation and interest in creatures WE call “cats”.

Our minded experience of cats requires that we understand our aims. We need to know first, why cats of are of interest at all. Is it that we think them cute and cuddly or that we want to eat them or that we fear being eaten by them? I find cats especially interesting because I am allergic to them. I know I have found a cat well before seeing it and that a picture of a cat on the internet is not for me at least, a cat.

And once we know why we find cats worthy of our attention, which is after all a matter of our relationship to them as embodied creatures who act in the world, there’s the matter of coming to an agreement among ourselves about all that makes up our idea of cat-ness: sights, sounds, smells, scratches, feelings, and in my case, itchy eyes.

Computer chips, no matter how many interconnections, can only be extensions of our intentions in which for example, we take an interest in pictures of cats. And if we add to our electronic version of a brain in a vat, more and more sensors that measure sounds and smells and itchiness, then still our digitally electronic virtual brains in vats can only reflect OUR INTENTIONS with respect to cats or anything else in the world.

Google would like to create a machine that advances THEIR INTENTION to make more money, and a machine that can return to a person searching the Internet, pictures of cats, will presumably make more money. But if we confuse Google’s intentions by imagining intentional machines that are one-in-the same in principle, with our own capacity to experience and know the world…!

Whoa Nellie!

PS – The observations above apply to pictures of cats as well as cats. For example…

PICTURES OF CATS