I Am a Strange Loop review by Josh

I Am a Strange Loop is Hofstadter's account of how consciousness and personal identity work.

He is a physicalist - meaning that whatever consciousness is he thinks it arises from from the substrate of meat (brains).

At the same time, he rejects the view that consciousness can be understood just by looking at brains or DNA. He repeatedly emphasizes that the level of explanation needs to be calibrated to the phenomena being explained.

It would be silly to attempt to explain a hurricane using just the language of particle physics. Understanding hurricanes requires larger concepts - things wind, rain and pressure systems.

Another analogy: suppose you just created a new pieces of software and a friend asks "how does it work?" You're probably not going to answer by going to the level of binary - 1s and 0s - even though that would technically be correct. If they're an engineer you might talk about technical aspects of how the code works, but even these will be many layers of abstraction above binary. If they're a lay person you're more likely to focus on the user interface level - explaining how it is used to do things.

Hofstadter's point is that it's silly to think too much about the brain or DNA when talking about consciousness. Of course brains and DNA are key building blocks of consciousness, but you can no more understand the phenomenon of consciousness by looking at DNA than you can learn to play Call of Duty by a close examination of the binary code.

Rather, the key building blocks of consciousness are things like concepts, ideas, beliefs, values, tastes and preferences, and the analogy making connections that glue those things together.

My favorite framing of the problem of personal identity comes from the story Funes the Memorious, by Jorge Luis Borges. The eponymous character, has a perfect memory but finds himself unable to draw abstracts. He cannot understand why "a dog at 3:14 (seen from the side) should have the same name as the dog at 3:15 (seen from the front)."

The counter-intuitive answer, proposed by the philosopher Derek Parfit and supported by Hofstadter, is that the dog at 3:15 is not the same as the dog at 3:14. They're very, very similar though so we think of them as the same entity and refer to them by a shared name. Spot, say.

Hofstadter repeats a thought experiment from Parfit: suppose scientists found a way to take a tiny part of Napoleon Bonaparte's brain and transfer it to Parfit. After the transfer he would, perhaps, know a few words of 18th century French or a have a flash of a memory of a battle but would otherwise still be the philosophy professor Derek Parfit. But then suppose the scientists kept repeating the process making him more progressively more Napoleon and less Parfit until eventually they had duplicated Napoleon's brain in Parfit's head.

Traditionalists about personal identity would want to know "when does this person stop being Parfit and become Napoleon?" Parfit's answer (which is also Hofstadter's... though probably not Napoleon's) is that that question is nonsense. There's no answer to whether the Napoleon-Parfit is Napoleon or Parift - it's both and neither.

Hofstadter draws an analogy between Parfit's impact on our understanding of the self and Einstein's on our understanding of physics. Einstein's idea was that the laws of Newtonian physics break down as objects approach the speed of light. That's a very rare edge case, and it makes no difference for things like planets and humans, where Newtonian physics works just fine. Analogously our everyday conception of ourselves and others as indivisible "I"s is pretty much right most of the time, but pushed to the limit in thought experiments it breaks down and the ways in which it breaks down are illuminating and show that our ordinary understanding isn't technically correct.

Both philosophers (Hofstadter quite scathingly) reject the arguments of John Searle, that the substrate of consciousness - the physical stuff out of which it arises (carbon/cells/brains) - matters. Hofstadter discusses the question of whether a self-driving car is conscious. He concludes that it's not, or more precisely that it is conscious to the same degree as an ant, but he doesn't reject the idea out of hand.

What about the title of the book?

A strange loop is a loop with a paradox embedded in it: think Escher's hands drawing each other or his perpetually ascending staircases.

Clearly the mind, for Hofstadter, is a loop: our concepts and interconnections are all ultimately "circular" in the same way that a dictionary is circular (every word is defined in terms of other words or pictures). But what is strange about it?

I think, but I'm not sure, that the purported strangeness arises from the fact that consciousness arises from the physical stuff of brains, but doesn't exist in any particular location of the brain. Similarly, our genes encode how the brain physically develops, but there is no individual gene (or even discr