Floyd's Blog

Children of Memory (Tchaikovsky 2023)

This book brings together some very big ideas, such as (I have to be a bit cryptic to avoid spoilers) how we can recognize human-level intelligence and how would we recognize fundamental physical reality versus an immersive simulation—or do these really matter? The author has a background in zoology and psychology, and it shows. I think you have to read the two earlier books in this series (1st, Children of Time, 2018, then 2nd, Children of Ruin, 2019) in order not to be hopelessly lost about a lot of central character and setting details. (However, the author does try to briefly recap these to set the stage at the beginning.) Midway, I was a bit torn about how I would ultimately rate this book. I felt like it got a bit on the silly fantasy side, and I prefer hard science fiction, but it went for a really deep dive from the characters’ point of view (at a certain point, you feel confused (by the author’s design) reading this—I’ll leave it for you to see what I mean) and brought it back together. I am recommending this book along with the two previous ones in the series.

One quote to keep in mind while reading Children of Memory is, “Any sufficiently advanced technology is indistinguishable from magic” (Clarke 1962). This might be especially true if you didn’t realize you were interacting with an advanced technology or mistakenly thought that you were the more advanced technology. Also, how do we know that we are not in a simulation? This type of idea has a long history from (at least) Plato’s Allegory of the Cave to the 1999 movie The Matrix (it also reminds me of a more obscure science fiction novel, True Names by Cory Doctorow, 2008)—and does it really matter? If our experiences behave as if they are derived from fundamental reality, then they are what we have to work with, and we treat them as such, at least until we can find inconsistencies and break through to a deeper understanding in some way. After all, aren’t the perceptions created by our brain an immersive simulation? We know we are subject to biases and illusions, with our brain often filling in the details of what we expect to observe (our visual blind spot being a literal example but also in how we process language, for example), and we only sense a small slice of the wide potential scale of sensory inputs (like the narrow band of visible light in a mix of only three colors or our audible frequency range and completely miss details outside of this or things like magnetic fields which some organisms appear to be able to detect). We hallucinate daily (dreams and otherwise). How far on this scale can we go before we would define our perceptions as not objective reality? Also, how real or alive is an organism if it is simulated or not—does this really make a difference?

This brings to mind René Descartes’s 1637, “I think therefore I am” (cogito, ergo sum). It is referred to as Descartes’s first principle of philosophy. Even if everything around us is an illusion to one degree or another, we know that something, including ourselves, exists (we know there is not nothing, which is possibly the most fundamental mystery of reality, that something does exist, which spins off into ideas related to the anthropic principle) and can act from there. Some more Descartes is also very relevant to this book, “Animals are like robots: they cannot reason or feel pain” (attributed to René Descartes by Proctor et al., (2013); it perhaps results from a paraphrasing of Part V of Descartes’ 1637 Discourse on the Method; interestingly the discussion in Part V touches on what some might consider early ideas of the Turing test to determine intelligence). Many of us don’t really believe this today, that animals lack any form of genuine intelligence, but how do we determine if animals have degrees of human-like intelligence?

Operant conditioning can be used to teach animals to perform some amazing tricks (and there are also limitations in what animals can be taught). However, we can be tricked by our empathy and biases into thinking they understand more than they do. A famous example is Clever Hans, the counting horse. The trainer unknowingly gave non-verbal clues to the horse as to when to stop counting, which made the horse look like it had the ability to do arithmetic.

It is very novel and surprising to us when birds vocalize human language. We are predisposed to think that this represents an understanding of the language when it is simply a learned response to a stimulus, literally parroting.

“Button dogs” are a recent phenomena that are also problematic in how humans interpret the responses.

There has been a lot of hype about AI, not that it is not a potentially very useful tool, but how much of this is the computational equivalent of us falsely perceiving intelligence in non-humans, as we have done with animals, and falsely applying our empathy to something that is simply a complex response to a stimulus. One relevant idea here is the “Chinese Room Argument”. An input written in Chinese characters could come into a slot on one side of a room. A person (or AI) could follow complex instructions contained in books or computer drives in the room to generate a corresponding output in Chinese characters and drop it out of another slot. If the rules are good enough, it may seem like an appropriate response in Chinese. However, the person or AI in the room may not understand a single word of Chinese and is just simulating an apparent ability to understand Chinese. How would we tell the difference?

However, if we go further down this track before long, we start running into some soul-searching and difficult questions. Flipping this around, how much of what humans do when they communicate is simply learned responses to a stimulus? Does 90% of our daily communication not take a lot of deep thought and is simply responding in a way that we are expected to and trained to respond in? How do we really measure how much other humans understand what the words we use in a language actually mean? (And what do we really mean when we say this?)

So, here is where we get to Alex and Irene Pepperberg. Tchaikovsky added two minor characters in Children of Memory that are an obvious (especially in the context they appear in in the book) reference to them, Alex Tomasova and Renee Pepper. If that is not enough, the dedication at the front of the book is “For Irene Pepperberg and Alex”. Irene Pepperberg worked with Alex, who was an African gray parrot. Alex appeared to be an extremely intelligent gray parrot and could do some amazing things, and they were surrounded by quite a bit of controversy in the animal behavior world. After the history of interactions like Clever Hans, there is, rightly so, a great deal of caution in interpreting animal-human language interactions. Alex could talk and reply to statements from Irene Pepperberg. This included answering questions about objects such as their size, shape, number, color, and material they were made out of. Alex could also ask for different types of food and other things, such as to take a break from being tested. The instant modern reaction is to assume this is simply a learned response to a stimulus with no real understanding of how to use language. However, Alex did some things that had not been seen before. One of these was that Alex asked an unprompted untrained question about himself—this had never happened before in the history of animal-human-language research and may seem simple on the surface but is very profound the more you think about it. Another startling thing is that Alex made up new words and combinations of words. When presented with an apple, a fruit Alex had not seen before, he called it a “banerry”, which may be a portmanteau of banana and cherry that Alex was familiar with. Alex described a cake as “yummy bread”, not knowing a word for cake. Again, this seems so simple to us that it is easy to overlook that an animal combined an adjective and noun to, apparently appropriately, describe something new. This is hard to explain as simply a memorized, trained response without some ability to apply the underlying mechanisms of language and the meanings of words.

A couple of family notes: My wife would probably kill me if I failed to mention that she met Irene Pepperberg once at a talk she gave years ago. Also, I was talking to my son about some ideas related to the ones in this post, and the conversation led to him asking me about the possibility of genetically modifying birds with the human sequence of FOXP2 (a gene that affects language in humans, mimicry in birds, and echolocation in bats among other things).

This starts to sound religious (and gets into some more science fiction plot ideas), but it is interesting to speculate, from a scientific biological perspective, about rather than viewing animals as lower on a scale starting from us, to turn the perspective around and ask if there is a higher level of consciousness-awareness-intelligence possible above humans and what that would be. Would an alien species see us as basically an animal level of intelligence? We react to stimuli, have instincts, have memory and communication, ask questions, are creative, appreciate art, have empathy, can record ideas and data, can build amazingly complex tools, and think about how to interpret data with math and statistics to overcome our biases and challenges in how we think, etc., but is there something we are missing? (I don’t just mean individual organism-based intelligence but also our collective behaviors (economic examples abound with what we can accomplish as a group and are largely unaware of on an individual level), which could include multiple species and AI with human collective behavior intelligence.) Maybe it is something simply on a scale like just more focus and being able to hold more ideas in mind to work with simultaneously and make connections—or is there a dramatically different way to think that we are blind to?

Just because we can’t think of an example doesn’t mean it can’t exist, especially in this case where we are thinking about the limits to our thinking. This is a fundamentally hard thing to think and communicate about because we don’t really have the words and thoughts for it, something like a human-transcendental (in the sense of surpassing the ordinary, not spiritual) intelligence. Is it a mistake to think of this as a higher level versus just a different way of doing intelligence that we are blind to (human-alternative intelligence)? Are there other species that have fundamentally different components of intelligence that we are missing but could begin to recognize and define, at least by its absence in ourselves? (In math, we are able to identify that unprovable things exist and that there are limits to formal knowledge (like Gödel’s incompleteness theorems)—could we do this in some way with intelligence and identify our own gaps and missing dimensions of intelligence, even if we don’t understand what they actually are?) Could AI be designed to evolve and develop new forms of intelligence and communicate them to us? Or do we have too much of something, like (from our perspective) animals having too much instinct that overrides, acts as too much of a crutch, or takes the place of other thoughts, such as peer pressure, biases, creativity, or language (how do we measure the volume of a container if it is not making any sound?—a joke, but I am trying to point toward an overreliance on linguistic rules and assumptions influencing our thought, kind of like the idea of the Sapir–Whorf hypothesis of linguistic relativity or a linguistic version of functional fixity where our assumptions and categories we place things into limit our use of them), that we need to tone down to focus on other areas of thought and knowledge (this might sound like I am referring to the Flower Sermon of Zen Buddhism). Language seems like our defining feature to enable human intelligence. It has given us a tremendous boost in our abilities (an understatement), and is hard to imagine what humans could be like without language, but could an overreliance on language be holding us back somehow (as animals seem to over-rely on instincts from our perspective, not that instincts or language are not useful), inhibiting or biasing a clearer view of thought in some way in some situations—with the irony that I am using language to write this thought? Perhaps working with math and equations is a good example of an alternative way of thinking; what else is possible? It sounds sacrilegious to consider the idea that language might be cryptically inhibiting intelligence in some way, which makes it even more interesting to me. This is all speculative, so I will stop here. For it to progress, there needs to be some way to rigorously define our intelligence’s “blind spots” and “blind directions” beyond storytelling and wishful thinking (again, even if we don’t understand them knowing that they might be there is a place to start). However, I sense that this is something that I will wonder about for a while. (Now that I am writing this, perhaps looking for commonalities in common fallacies and biases that we, as individuals, are susceptible to and collective human behavior is susceptible to is a good place for me to further look into.)

Or, are the tools we are using and our way of thinking the maximum level of intelligence possible? This would seem to be even more surprising than the alternative, but that doesn’t mean that it is wrong.

… So, these are the directions that reading this book (and the previous books in the series) led my thoughts.

Update: An article about parrots making video calls to each other: https://news.northeastern.edu/2023/04/21/parrots-talking-video-calls/

Links

Wordlist

  • Acerbic—direct and critical.
  • Acquisitive—eager to possess things.
  • Agog—eager. I knew the definition of this word, but I think it is such a funny-sounding work that I wanted to include it.
  • Anatomizing—analyze in detail.
  • Argot—a particular jargon.
  • Askance—suspicion and disapproval.
  • Atavistic—reverting to an older state.
  • Cenotaph—a monument to someone that is not buried there.
  • Comme il faut—proper behavior.
  • Contritely—remorseful.
  • Conurbations—multiple towns or cities merging into a larger urban area.
  • Crenulations—a small notch or a scalloped irregular wavy appearance.
  • Cuprous—of or containing copper or specifically copper with a valence of 1 (copper(I), Cu+).
  • Demuring—to object to something. Don’t confuse this with demure, shy and reserved. Is this almost a contranym? They seem to both be related to the idea of delaying something but in different ways.
  • Détente—achieving an easing of strained relations.
  • Devolving—to break up, degenerate, or delegate power.
  • Doggerel—badly written or irregular verse, sometimes comic.
  • Dour—stern and gloomy.
  • Dysphoria—unease.
  • Earnest—serious and sincere. I knew this, but something was bugging me about it, so I wanted to look it up to double-check.
  • Equanimity—calm and composed in difficult situations.
  • Ersatz—an inferior substitute.
  • Gadding about—wander around for entertainment.
  • Gallimaufry—things jumbled together like a mishmash, hodgepodge, or grab bag.
  • Garrulous—talks excessively and rambles about unimportant details.
  • Genially—pleasant, cheerful, comfortable.
  • Gnomically—terse hard to understand speech that may seem wise but is hard to interpret.
  • Graft—I was thinking about the similarity between the crime meaning, bribery or stealing funds you are entrusted with, and the plant meaning, which is loosely growing off of another plant.
  • Hoary—gray-white and/or old.
  • Imago—a fully developed organism at its last stage of metamorphosis or an idealized image of someone. Imago means image in Latin.
  • Indefatigability—tireless.
  • Inimitable—so unique and high quality that it is not able to be imitated or copied.
  • Interlocutor—someone you have a conversation with. In the context of the use in this book it seems to indicate a translator.
  • Irascible—easily angered.
  • Keens—wailing in grief. Don’t confuse this with keen (sharp, eager, intelligent).
  • Lambency—glowing, shiny, or appearing as if it is reflecting light.
  • Languorously—happily relaxing or lazy.
  • Limning—This one seems to have a few meanings. To draw or outline something. To depict something in art or writing. To shine a light on or backlight something (so it is outlined in light). Or just to paint an image.
  • Louring—to look angry.
  • Matryoshka doll—Russian nesting dolls. These are typically in a long series from larger to smaller that fit inside each other. https://en.wikipedia.org/wiki/Matryoshka_doll
  • Motes—a small speck of something.
  • Panoply—a complete collection, such as a suit of armor with all of its parts.
  • Pareidolia—seeing a pattern when there is not actually one present (like seeing faces in wood grain patterns). This might be a very useful term. More than once lately I have found myself trying to describe or point out this effect.
  • Paroxysm—a sudden occurrence of something.
  • Penurious—very poor or stingy.
  • Plaintive—a sad sound.
  • Pore over—to carefully study something. For some reason, this one caught me, so I noted it. I am used to pore meaning a small hole. It almost feels like this should be spelled pour-over (like pouring your attention out over a subject), but this is incorrect (to the best of my knowledge). It uses the other meaning of pore.
  • Potted history—a short summary that just outlines the main events.
  • Presaging—an omen of a coming event.
  • Pugnacity—quick to argue or fight.
  • Quizzically—doing something in a way that shows you are questioning it.
  • Quotidian—an ordinary thing or something that occurs daily.
  • Revenants—a creature, person, or ghost returning after death.
  • Rheumy—refers to watery eyes.
  • Riven—torn apart.
  • Sanguine—optimistic.
  • Sensorium—all of the combined senses.
  • Sobriquet—nickname.
  • Susurration—a whispering or murmuring sound.
  • Teratogens—compounds that cause fetal abnormalities.
  • Unkindness—mistreating someone and a collective noun for a group of ravens. This is a “term of venery” that developed in English in the Late Middle Ages as a part of developing unique hunting terms for herds of animals.
  • Verisimilitude—appears to be authentic.

Quotes (There are a lot of references to other literature in this book. I am sure I missed a ton of them, but a few that I caught are included below.)

  • “You will fail, and when you do, you must do everything you can to fail as little as possible.” p. 18
  • “What if, buried down there, was the machina from which that deus might arise?” p. 29, a reference to the deus ex machina plot device.
  • “Things fall apart, though, and entropy is the landlord whose rent always gets paid.” p. 67. This might be, in part, a reference to Yeats’s (1919) The Second Coming and to the second law of thermodynamics, in the same sentence.
  • “We profess an inordinate fondness for beetles. The bluish-green ones, in particular, are delicious.” p. 72. This is a reference to J. B. S. Haldane, an evolutionary biologist that is also a very quotable person. In the original quote, he was referring to God having an inordinate fondness for beetles because there were so many species of them. I am not aware that he ever tried to eat them or found their flavor enjoyable.
  • “Everyone’s face says they don’t know her, in that moment. That she doesn’t belong. If anything, the other refugees’ faces say it louder, because if it’s her, it’s not them. Because the best way to make yourself one of us is to find someone more them than you are” p. 139; “Angry faces, and faces making sure they look angry so nobody looks angrily at them.” p. 372, an example of how things can go wrong in collective human behavior.
  • “Sometimes the friends we meet along the way were inside us all along…” p. 433, cf. p. 178
  • “To strive, to seek, to find, and not to yield […]”
    “What was that? […]”
    “I believe it’s literature […]”
    p. 202, a reference to Tennyson’s (1842) Ulysses. It might be making a slight joke about this line being over quoted.
  • “[…] a great wheeling congregation of birds. Not an unkindness […] but a kindness of ravens.” p. 292. This plays with the weird English language use of unkindness as a collective noun for a group of ravens.
  • “When shall we three meet again, hmm?
    Traditionally it’s when the hurly burly has been concluded, […]”
    p. 339, a reference to the three witches in Shakespeare’s 17th-century Macbeth.

Posted

in

by

Floyd

Tags:

Comments

One response to “Children of Memory (Tchaikovsky 2023)”

  1. A Fire Upon the Deep (Vinge 1992) – Floyd’s Blog

    […] intelligences. Coincidentally, it touches on some points that I explored in a previous post about Children of Memory regarding levels of intelligence, different ways to be intelligent, and if an individual is real or […]

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *