The Mind's I, Dennet
Man has existential crisis about who he is and what the mind is.
Notes
My notes aren’t great for the first two sections, as I hadn’t started highlighting.
I. A Sense of Self
Borges and I
Borges and I, by Jorge Luis Borges
- Autobiographical essay about the conflict between the self and the persona.
- People think about themselves differently to how they act.
- Which is the true “you”, the one out there in the world or the one in your thoughts?
On Having No Head
On Having No Head, D.E. Harding
- A “charmingly childish and solipsistic view of the human condition”.
- We learn that humans die, we ourselves are humans so one day we will die.
- But the essay describes an experience we get where we feel distinct and somehow different from everyone else.
- The idea of having no head is literal. From the centre of your own experience, you don’t have a head, just a viewport out into the world.
Rediscovering the Mind
Rediscovering the Mind, Harold J. Morowitz
- Biologists used to advocate that the brain was somehow “above” nature (dualism).
- Physicists used to advocate that the universe could somehow be described in purely mechanical terms (reductionism).
- Now the roles have swapped, many famous physicists present elevated views of conciousness whereas many biologists explain it in purely physical terms.
- The discovery of quantum mechanics has played a large role, i.e. “the mysteries of quantum physics and the mysteries of consciousness are somehow one”.
- This contradicts the overall message of the book (i.e. The Mind’s I) that non quantum-mechanical computational models of the mind are possible.
- Many-world interpretation of quantum mechanics.
- The claims about how the many-world interpretation of quantum mechanics is wrong are somewhat hand wavy.
II. Soul Searching
Computing Machinery and Intelligence
Computing Machinery and Intelligence, Alan M. Turing
A classic.
- The first formulation of the imitation game, now referred to as the Turing test. A judge has to decide whether a response coming from an unknown source belongs to a human or belongs to a computer pretending to be a human.
- Meant to make the question of “can machines think” more precise.
Contains the dialogue from this: GPT-3 Fiction, Turing dialogue. What I hadn’t seen before were his “contrary views on the main question”:
- The Theological Objection: God lets man think, so machines can’t think. Turing dismisses this on account of the fact that theological arguments were used for some time to argue against the Copernican revolution).
- The “Heads in the Sand” Objection: Machines thinking would be bad, let’s hope they can’t. Turing says this somewhat tongue in cheek and says that’s it not really an argument.
- The Mathematical Objection: Gödel showed that there are some things that can be seen as true when thinking “outside” a system but cannot be proved within the system. Or, we can see that a machine will halt but you can’t write an algorithm for this. He suggests that we are not as smart as we think, and that there might be things that we cannot prove ourselves but that are still true, but we still think.
- Argument from Conciousness: Tackling “Not until a machine can write a sonnet or compose a concerto because of thoughts and emotions felt, and not by the chance fall of symbols, could we agree that machine equals brain”. Turing says that this is outside the scope of his test and is the reason he’s using the test as a proxy for the question. If a machine could simulate writing a sonnet and explain why it made the choices it did, does that mean it’s thinking?
-
Argument from Various Disabilities: “Machines can do all kinds of things, but they’ll never…”
- Be kind, resourceful, beautiful, friendly
- Have initiative
- Have a sense of humor
- Tell right from wrong
- Make mistakes
- Fall in love
- Enjoy strawberries and cream
- Make someone fall in love with it
- Learn from experience
- Use words properly
- Be the subject of its own thought
- Have as much diversity of behavior as a man
- Do something really new
- …reminds me of the quote from AIMA that’s along the lines of “AI is everything AI cannot do yet”.
- Lady Lovelace’s Objection: The argument that machines can only do what is programmed in them from the start. Turing says that learning is still just programmed in from the start.
- Argument from Continuity in the Nervous System: The nervous system is not a discrete state machine, it’s continuous. Differential analyzers exist, and these are like continuous computers. But even then a discrete state computer could predict the outputs of a differential analyzer to a degree and use it to trick the interrogator.
- The Argument from Informality of Behavior: People’s behavior isn’t formulated as a set of rules, but the behavior of a computer is, so we are not the same. Machines can learn?
- The Argument from Extrasensory Perception: This is where the essay takes a bit of a turn. Turing says that since “telepathy, clairvoyance, precognition” etc. exist, that is proof that we are somehow fundamentally different from machines. I think the consensus these days is that ESP doesn’t exist.