I'm writing this review as I go along because the book is long. I read GEB in college and liked it, though I suspected that his idea that consciousness is a kind of self-referential loop might not bear close scrutiny. That's why I picked up this book when I saw it.
However, my hopes have been lowered within the first few pages when Hofstadter tells the reader that some living things have bigger souls/more valuable souls than others. In particular mosquitoes don't have much of a soul that you could speak of. And for people who haven't read the book, 'soul' doesn't mean the Christian or religious soul, only a kind of cognition and introspection that seems to be uniquely developed in humans (so if you think of humans as having the biggest souls, and define a soul as that human quality of thinking and introspection, then you really are talking about a self-referential loop!) (I'm kidding.)
Hofstadter takes it for granted that readers must have some line on the spectrum from humans to rocks where they demarcate those living things with souls we care about from those things that don't have enough of a soul to concern ourselves with. Though the line is kind of arbitrary, it must exist for each person. Even moral vegetarians are killers. After all, plants are living things.
But I actually don't agree with the position which I think is setting up the rest of this book, though I'm sure there will be more details coming. I don't actually see much of difference between inanimate and animate things. I don't see human thought as fundamentally different in nature than other physical phenomena such as the orbit of planets or weather, though it's expressed in a different way. I'm not sure that we're necessarily in safe territory if we say that something like a plant doesn't have consciousness. That's a very biased way of looking at things. Isn't it at least possible that things like plants or worms have a kind of consciousness that is very important to them, but which, being so physically different from them, we can't understand? Might we not lead ourselves into error by assuming that human consciousness is the highest attainment yet along a one-dimensional scale running from 'no soul' to 'maximum soul?'
-----
Starting again because I put it down. I've only gotten as far as I was before, but I had an additional thought, which is that Hofstadter is basically trying in his vegetarianism discussion to evade the moral culpability that one incurs in killing. The vegetarian/vegan solution to this dilemma is to draw a demarcation line between "conscious things" and "unconscious things," and rest on the side of eating unconscious things.
Not only is this a potentially flawed way of thinking, as I said above, it's also an abdication of responsibility for killing. Although there are myriad good reasons to be vegetarian or vegan, such as environmental and health concerns, the more enlightened position is to recognize that in order for one being to live, others must be constantly dying. Except that plants don't usually kill other beings in order to live. I found it strange that Hofstadter is trying to so hard not to admit this point. One must recognize and come to accept one's responsibility in killing to see life correctly. Those who won't kill a deer would kill dozens of chickens or thousands of plants to get the same nutrition that a single deer supplies. Who eats farmed produce kills beasties large and small by the dozen, chemically, mechanically, and by displacement. One can say "I'm resting on the right side of my demarcation line" when one goes vegetarian, but one can only believe it through willful ignorance. Better to draw some other kind of line than a soul-based line, and base it on environmental concerns or cruelty concerns, I think.
---
End of Chapter 6
I don't see why the author is so insistent that we admit that animals don't think like we do. When he's talking about how ideas might be represented by patterns in the brain, I'm on board, But then he keeps mixing in some pretty unconvincing bits about why humans are in a completely different class in symbolic understanding (according to his definition of "symbol" as, basically, an idea in the brain).
You can take it as a general rule that when someone puts a stake in the ground and says "this is where humans are different from animals," that in short order a bunch animals will be found that infringe upon that boundary. But Hofstadter, probably aware of this phenomenon, wants us not to decide whether a mosquito has "interiority," which is debatable, but to admit that some living things don't and therefore justify his "soul" scale.
But I can't. This is an oversimplification. He puts the words of someone who objects into the book, saying "I can't say for sure that a mosquito doesn't have as rich an internal life as I do!" and concludes that such objections are not sincere. Maybe not, but my objection is that this idea of "interiority" is not shown to be the only meaningful expression of consciousness or "souledness." Even beings that act almost as simply as computer programs nevertheless try desperately to feed and mate, often to protect their young, and to survive. Is this not a refutation of the idea that small-souled animals are edible? Even if they only have one single goal, to reproduce as much as possible before dying, be it pursued by reflex, cognition, or even (as in plants) by genetic design, it's that very idea that one violates in swatting the mosquito or eating the chicken or pulling out the weed.
I tried to find a textual justification for this position wherein he states that, in his estimation, "interiority" is the same or largely corollated to soul size. This idea seems to be implicit at this point, but I couldn't find it stated directly. But whether he believes this or not, it's problematic for me, because conflating "interiority" with "soul-size" is basically begging the question of what things have big souls by defining "souls" as essentially the thing that we think humans have the most of, and if he's not making that point, then why all of the animal comparisons and asking the readers to admit that animals don't have souls as big as those of humans? If we took an ant's point of view and tried to define "souls" according to the capacities that make ants unique, humans would come out looking poorly, unable to serve their allotted function in the hive or sacrifice themselves unthinkingly when necessary. So what's the significance?
And my complaint specifically is why we're being asked to concede the point. It doesn't seem to be necessary for his argument, which I think is going to be that the idea of rich "interiority" is essentially a matter of being able to build ideas out of ideas in a self-referential fashion. Does this argument require that we also say that animals without much "interiority" are "small-souled?"
---
Page 210
GEB made such a big impression on me in college, but I think I better not reread it if I want to keep the memories intact. One could make the argument that I'm not qualified to complain about the reasoning going in this book. Fair enough. Hofstadter is a respected academic, and I'm a dingus with a Goodreads account. We both struggled with the "abstraction wall" in higher mathematics, but I won't concede to Hofstadter here, because ultimately it was my slapdash proofwriting style that was my downfall, not the ideas themselves. And I challenge anyone to take as many good ideas as I had and make such bad grades out of them. I must have set some kind of record. How do 5 people working together on a proof all agree on the correct approach, and then 4 get 100% and one gets 50%? The one who came up with the idea in the first place? I'm bitter.
And I'm bringing all that frustration to this book, because if I made half. as many confusing or unsupported claims as Hofstadter does, they would have told me I had a future in the Business school.
But I can't explain it. Obviously Hofstadter is a logically-inclined man of good standing...do his analogies seem so irrelevant to the argument due to a surfeit of simplification? The book is heavily annotated, yet it feels like on most of the claims where I seek citations or explanations, those are the lines where no annotations exist.
But stepping back from the level of simms (the murky argumentation and missing citations and overreliance on a handful of rather limited analogies) to the simmbllic level (heh), I'm finding it hard to understand the significance of the argument, which seems to go something like this: the brain is made up of neurons that act at such a low-level that one can't interpret symbolic meaning from them. However, because they have the ability to persistently represent external events, and to manipulate those "symbols" in such a way that ideas can accrete to other ideas, where a high-level concept like "sitcom" might include "television" which includes "screen" which includes "image", etc, it reaches a threshold where the system is able to conceive of itself in symbolic terms, even though at the symbolic level of thought, there one doesn't observe the individual neurons that physically cause that phnomenon. This self-reference at a more complex level of representation is the eponymous "strange loop."
As in GEB, he does an 80-page summary of Gödel's proof, which is the inspration for strange loops. But I wonder if Hofstadter, enthusiastic as he is over the idea of self-referencing systems, isn't overstating his point. Gödel's genius was, basically to create a mapping from formal symbol logic to whole numbers so that each statement in B. Russell's logical system corresponded to exactly one whole number. In this way, one could embed statements about statements into the system, not just statements about numbers, because statements about numbers could also be statements about numbers that were standing in for other statements. So you could, in effect, say "This statement is unprovable" by making a logical statement about the whole number that uniquely identifies the very statement in question. The details of how to accomplish this are difficult, but the idea is easy. "This statement is unprovable" is true because of how it refers to itself. If you can prove it, then it's false. so obviously you can't prove it and it's true.
Foundation-shattering as that is for mathematics, I don't really see the significance of the analogy to consciousness, except in that Gödel's proof also applies to all systems sufficiently powerful to define exponents within them. The brain is such a system, but it has so many properties so utterly divergent from formal systems that the analogy is not very instructive. For example, human brains can deduce false ideas starting with true premises because of errors of construction. I wonder if Hofstadter will address this. For example "Fat is the word for the flubby stuff around the stomach, and fat is also listed as a macronutrient on nutrition labels, so eating fat adds fat to your stomach." One could earnestly belive this based on the foregoing true statements, but the conclusion is wrong.
And for this reason, it strikes me as less significant that a brain can think of itself, because brains clearly have flexibilities that formal systems don't have. Computers don't have opinions or preferences, for example. The surprise of Gödel's proof, for me, is not so much that it can be interpreted as being self-referential, but what it said about logical systems themselves--that they are incomplete. There are truths that can not be proved.
What came next is also relevant, and I hope I relate this right. For some time, Gödel's proof was considered a kind of tricky corner case, a "gotcha" that could create statements that were unprovable due to their twisty logic, but that non-trivial assertions would have proofs. But Greenblatt? (check on the name) was able to prove that a certain non-trivial theorem was unprovable as well, putting a final bullet in the shiny dome of mathematical completeness. The truth is, nobody knows how many statements are true but unprovable. We just know that such assertions exist. How many famous open questions can never be proved? The response so far has basically been to ignore this inconvenient fact and hope that it's not so inconvenient that people begin to notice.
Which is to say that Gödel could conceivably have found an unprovable statement without trickery that would have been even more shattering. The self-reference was a tool, not really the end in itself. It was just a clever mapping.
So Hofstadter is here to say...that certain patterns of neurons can be mapped to so-called symbols in the brain, and that there is a mapping of neurons that becomes the concept of the brain itself? It makes sense, assuming that this is really how the brain works, but...was this unexpected? And why does he keep bringing animals into it? He also keeps saying things like there is a scale from small soul, with not much of a strange loop, to a big soul, where the self-concept is rich. But again I fail to see the significance. Is it true that beings that don't reflect on themselves are somehow less important than ones that do? After all, when we kill, it's not really the 'I' of the victim that we're hurting-it's those who survive who valued that thing that we should care about, if the injunction is to not cause pain.
I jus really hope at this point that the author doesn't attempt a proof of free will.
---
End of Chapter 16
I realize now what's wrong with this book. Hofstadter doesn't believe his own argument. That's why the arguments are so murky--because they don't quite connect. And he knows they don't quite connect. But he really wishes they did, not only because they're his ideas.
It's clear that he was deeply moved not only by the death of his wife, for which reason he would naturally be inclined to believe, like R. Pirsig, that a person is a "pattern" that in some way survives their death, but by his love even before her death, which was probably sweeping all logic before it.
If it were true that his interest in nested selves predated any emotional perturbation, positive or negative, wouldn't there be more consideration of more germane cases of people who literally switch selves, like multiple personality cases, method actors, or more discussion of authors, as opposed to the couples-first approach? And beyond this, the postscript to chapter 16 should be unnecessary if he knew that his argumentation were solid. Instead of telling us that he had many of the foregoing thoughts prior to his wife's death, he should have said "Go on, try to poke a hole in this argument if you think I'm crazy."
For completeness, I'll do so now: he's trying to have it two ways by refusing to define "soul." He refers to it at times as a kind of metaphor for a real brain pattern, and other times quite plainly in metaphysical sense. He's muddying the issue by on the one hand making an equivalence between a "soul" and "interiority," which is the self-concept, the strange loop, and on the other hand invoking "soul" as the vital essence of an individual, indefinable but definitely present, what gives a being worth, in Hofstadter's eyes. But these senses of the word "soul" are not equivalent. The latter sense is not demonstrated to definitely exist in the first place, nor is it conclusively argued that interiority itself is a meaningful measure from which to judge the "size" of a soul. Ultimately, my first complaint keeps wringing true--the whole premise is flawed. If you don't believe in a metaphysical soul like I don't, then the argument collapses to merely an observation that brains can think about themselves, which is not terribly exciting. And if you do believe in a metaphysical soul, you're being asked to tie it directly to how much a being thinks about itself. But this is not an obvious equivalence, I dare say.
---
Sorry to those people who "liked" this review recently, as it was incomplete. An update that I had tried to submit when I finished the book appears not to have saved.
I wanted to say that in the end, Hofstadter makes his position on several existential questions clear. He does not believe in free will, which makes sense, and he doesn't believe either in mystical, incorporeal souls. So that means that he and I agree on the basics after all, but it makes much of what went before in the book all the more confusing. I am able to grant him that the overgrown ganglion which is the brain can, at a certain point, conceive of an idea or a "symbol" of itself, but since I incorporeal souls don't exist, this is a trivial proposition. The human body, being the only unit associated to each self-aware person, is clearly capable of conceiving of itself, because we all conceive of ourselves.
More intriguing is the idea that the capacity for this kind of abstraction is associated with brain complexity, where certain beings with small brains just don't have the neural power to conceive of such an idea. This, I think, is what he meant to illustrate with his insistence that some animals don't have "souls" or "interiority."
But my original complaint still seems relevant to me. If Hofstadter and I both grant that consciousness is a totally physical phenomenon, like lightning striking or sand dunes, then we have to grant that other beings undergoing complex physical processes might be experiencing other kinds of consciousness no less valid than our own. The question, then, becomes why is the "I" concept uniquely significant? In what way does it imply a being's worth or dispensability? If there were a creature with no "I" concept, whose entire neural network was nothing but pain receptors, and it writhed and screamed at even the slightest touch, would you touch it? Is it a biological machine, only pretending to feel pain? I wouldn't touch it, whether it thinks about itself or not. The "I" concept is just something that seems to be uniquely developed in "higher" animals, so it's a convenient standard to take up as a measure of worth, because we as humans are sure to come out on top. How things would be different if the standard of worth favored a being like the mosquito. It would be our moral duty to let ourselves be sucked dry then, and swatting would be an unforgivable crime.