18 thoughts on “Terry Pratchett — “I’d rather be a rising ape than a fallen angel””
Well, Terry, you are neither an ape nor an angel. You are a human being. There is an essential difference between you and the apes and you and the angels. No Christian or Jew thinks we are “fallen angels”. Nor do you become an angel when you die. What silliness, you are a man. (See Aquinas, Summa Q75.7)
Terry has his finger on the difference. He remarks on how interesting it is that we make (streetlamps, etc) and (though he does not quite say it directly) that we speak. We are makers and speakers. This is a remarkable thing about man. He has put his finger on the right thing, that which makes us essentially different than monkeys. Though he proudly proclaims “I am a monkey”, by doing so he precisely manifests that he is not a monkey by being capable of making such a statement. Man is a rational animal – ‘zoon logikon’ is the expression in Greek. Logos, though, can mean both reasoning but also speaking. Man is the animal who speaks (that is how Heidegger renders that phrase).
As scientists use their words, they never pay attention to the singular miracle that is language. This is one of the main points of Walker Percy’s brilliant book ‘Lost in the Cosmos’ (seriously, I cannot recommend that book enough, one of the most entertaining books I have ever read). What is so interesting about Sagan is that he is so interested in proving that we are not all that unique that he precisely exhibits his own uniqueness (while not realizing it) by speaking and asking those questions!! Scientists, and indeed people in general, live and act in a profound state of self-forgetfulness. But what about this language that we have? What about our ability to make intentional assertions about things? What must be true in order for us to be able to be the kind of speakers we are?
I’m not sure I see your point. Yes, language is a unique trait among humans. As far as we know, we’re the only example of symbolic language use that has arisen on Earth. We’re also Earth’s most prodigious tool -users and -builders. That doesn’t mean we’re not hominids or great apes. Just check a cladogram. (For the record, no, we are not monkeys. We ain’t got tails.)
Even so, our abilities don’t seem to be a “miracle”, in so far as we don’t yet understand exactly how language works in the brain. It’s probably a matter of degree as much as our other skills. Many animals exhibit some measure or other of some particular cognitive skill. Being able to count is likely the basis for logical thought, and elephants, dolphins and chimps can, to a certain extent. Chimps can be taught to use language (ASL, usually) with the relative skill of a preschooler. Yes,they have to be taught, and yes, they max out at the level of a 4 year old human, but the circuitry is there. Chimps in the wild have been shown to make “intentional assertions” (“There’s a cougar in that tree!” or “Hey, I found food over here!”) with the calls and hoots they make. Different chimp tribes in different areas have a different “dialect” of calls. Elephants, dolphins, and certain parrots can show somewhat similar language abilities. For all we know, whalesong might be poetry.
Of course it’s indisputable that humans have a particular synergistic combination of abilities that combined in an unprecedented way, allowing us to eventually build the globe-spanning civilization we have today. We don’t just build tools–we invent technology. We don’t just communicate–we speak and read and write. We walk fully upright (sort of) and have gigantic brains.
No one will deny that this is amazing. But those skills didn’t show up fully-formed, all at once. We’ve been making tools for 2 million years, but they barely changed at all for the first million. The deep roots of language (among the hominid lineage) seem to go back as far as our common ancestor with chimps and gorillas. At the same time, though, it’s unknown whether Neanderthals had the full symbolic language that we do, but I wouldn’t hesitate for a second to call them “human” if that were the case. (The latest research indicates that it is likely that they did have language. It is certain that they never figured out projectile weapons, though.)
My point is this: the fact that we use language doesn’t refute Terry’s description of us as “rising apes.” We are apes. We’re highly derived apes with many specialized skills, but apes we remain.
By the way, I think he was being figurative about the angels thing. Christians and Jews don’t believe we are fallen angels (to paint a broad brush), but many groups do believe we are fallen “from perfection”. Mormons literally do believe we are fallen angels, if I understand what the “Heavenly Host” is. But I guess you took Terry Pratchett “literally” (to possibly misuse the word), because you actually do believe in angels with specific traits. The word “angel” pushes a specific button in your head (symbolic language at work!) that doesn’t precisely match the way I believe he was using it. That’s one of the problems with fuzzy words like “angel”.
I think Kleiner’s point was simply that humans and apes are essentially different. This doesn’t mean that humans didn’t evolve from apes; somewhere along the line, however, humans became something totally distinct from apes. I think it’s clear that this is an essential difference, as Kleiner said. Nobody’s going to mistake an ape for a human. Of course, just because this difference seems obviously essential doesn’t mean we can’t disagree about the degree. I think Kleiner would claim the difference is stark, as big as that that separates a tiger from algae. It’s my feeling that James would compare the difference between humans and apes to the difference between tigers and lions. (Please correct me if I’m wrong.)
It seems clear to me that the difference is stark, at least given my estimation of an ape’s intelligence. Just like everyone else, I’ve never been an ape and so I don’t know how and to what degree they think. Usually, however, I think of animal intelligence under a materialistic model. This means that animals are something like a computer, receiving input and running it through some chips and programs to produce output. I don’t have any idea whether this is actually how an ape’s head works, but if it is, humans and very different.
I remember sitting in one of Huenemann’s classes when he predicted that computers would become conscious as they became more complex. (Please correct me if I misrepresent you.) I have a hard time with this, because it seems so obvious that consciousness is different in kind than unconsciousness. That is, computer’s aren’t conscious because they aren’t going about thinking in the right way, not because they don’t yet have enough connections. Of course, whenever someone says something is “obvious” in a serious discussion, that means they don’t have an argument for it. I’m not sure how to argue that added complexity won’t bring consciousness except by pointing out that extremely complex systems (the internet, for example) don’t seem to be conscious or by asking you to imagine the point at which consciousness will arise out of complexity. What will that be like? How will it happen? At what number (about) will enough connections be made to guarantee consciousness?
On a different note, Pratchett’s claims about the aesthetic benefits of evolution seem sketchy. First, evolution is a boring story. It’s just like all the “begats” he admits skipping in Genesis. (I know the particulars of a species’ evolution can be interesting, but overall, how is this any more exciting than Genesis? Genesis is like the ultimate action movie.) Second, saying “we’ve done pretty good” to evolve to the point where we can build street lamps seems silly. It goes with out saying that streetlamps are awesome, but if we evolved due to the pressures of natural forces, we don’t deserve credit for that (the natural forces do). As Aristotle probably says somewhere in the Nicomachean ethics, we only deserve credit (or blame) for the actions we cause. Although I’m grateful to the natural forces/God/demiurge/whatever that brought my evolution about (I like being a human), I can’t go along with Pratchett and pat myself on the back for it.
“It’s my feeling that James would compare the difference between humans and apes to the difference between tigers and lions.”
Humans are apes. Taxonomically, we are in the hominidae family along with gorillas, chimpanzees, and orangutans. Think hierarchies and subsets of subsets. On a purely genetic level, we are a little further apart from chimps as tigers are from lions–which in fact are very close to each other as they can produce sterile hybrids. (I’m not sure if hybridization is possible between humans and chimps, but I don’t really want to know anyway). But tigers and lions diverged more recently than humans and chimps did, so this is what we’d expect.
Behaviorally, we are worlds apart. All hominids exhibit some capacity for language and culture, but among humans, it’s like we’re running on jet fuel. What caused this? This is an open question and there is lots of ongoing research, and is likely a combination of an increased degree of old traits as well as qualitatively new phenomena. We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it. It could be true that our capacity to create culture and language constitutes a turning point in the history of Earth.
“Usually, however, I think of animal intelligence under a materialistic model. This means that animals are something like a computer, receiving input and running it through some chips and programs to produce output. I don’t have any idea whether this is actually how an ape’s head works, but if it is, humans and very different.”
This is a pretty bad metaphor. Brains–even insect brains–aren’t like any computer we’ve yet built. They’re massively parallel, have all sorts of feedback loops, rewire themselves on the fly, and other messy biology stuff. It can’t be broken into simple input and output. Previous memories, internal states, and chaos (in the mathematical sense) all play a role. And at the very lowest levels, we don’t have any evidence to suggest that our brains are fundamentally different than any other animal. We’ve all the same neurons. I know Kleiner will argue that human intelligence includes non-materialistic factors, and that our minds don’t reside in our brains, and these sorts of things. But that’s not science, and doesn’t contribute anything to our understanding of the brain.
“I’m not sure how to argue that added complexity won’t bring consciousness except by pointing out that extremely complex systems (the internet, for example) don’t seem to be conscious or by asking you to imagine the point at which consciousness will arise out of complexity. What will that be like? How will it happen? At what number (about) will enough connections be made to guarantee consciousness?”
Complexity is a necessary condition for consciousness (whatever that is) but it’s not sufficient in itself. The type of complexity is just as important. The internet is not conscious because it doesn’t have a system-wide organizing component that directs when, where, and how many connections to form–brains do. (Although, for an incredibly loose definition of “conscious”, you might be able to argue that the internet is. It does “know” practically all human knowledge, for example.)
In other words, the software is at least as important as the hardware. It’s very likely that we already have the hardware capability to create artificial intelligences simulating a human brain. We just don’t know how to write that program yet. As for your last few questions in that paragraph, several computer scientists and cognitive neuroscientists have made estimates. If you’re interested I could write up a post on this topic.
“On a different note, Pratchett’s claims about the aesthetic benefits of evolution seem sketchy. First, evolution is a boring story. It’s just like all the “begats” he admits skipping in Genesis. (I know the particulars of a species’ evolution can be interesting, but overall, how is this any more exciting than Genesis? Genesis is like the ultimate action movie.)”
I guess this is a matter of opinion, but Genesis can’t be the ultimate action movie. It doesn’t have semi-trucks crashing into fighter jets, Scottish captains defecting with Russian submarines, or one single Mexican standoff. If you don’t think evolution is interesting (I find it to be one of the most fascinating and beautiful ideas ever thought of) maybe you ought to read more about it. Richard Dawkins’ biology books are really good. I recommend The Blind Watchmaker, Climbing Mount Improbable, and The Ancestors’ Tale. And compared to Genesis, evolution also has the slight bonus of being true.
I have no trouble saying that humans evolved from apes, as JPII said “evolution is more than just a theory”. Frankly, I am pretty agnostic about evolution. I don’t think the story is as ironclad as some would say and I would reject some versions of it, but I think those that deny it outright are a little kooky. I am glad there are really smart people out there trying to work it out. I am agnostic because I think the scientific account of man leaves unanswered (and frankly unasked) most of the existentially interesting question to ask about the human condition. I don’t stay up at night wondering about my genetic ancestry, I do stay up at night wondering about love and hope and loss and joy.
I don’t much care about the taxonimical and zoological distinctions. I understand them to be largely arbitrary. There are very few bright lines to draw. BUT, I think there is a really bright line between man and the apes from which he arose. I think there is clear ontological leap there. Language is the clearest indication of this ontological leap. And this is not just a matter of us not understanding well enough yet how language works in the brain. My first post made a challenge about intentional language and how to explain it. My claim (actually it is Aristotle’s), is that intentional language cannot be explained in totally materialistic terms. We could understand the brain perfectly but still not explain intentional language. In other words, I find the deferred answer from James to be entirely unsatisfying. Yes, scientists are learning more and more about the brain (for which I am glad), but the materialist/evolutionary story just does a terrible job of explaining the remarkable difference between man and the rest of the animal world (language, culture, meaning, etc).
Suggested reading: Ric Machuga’s book: ‘In Defense of the Soul’. It includes chapters on why Aristotelianism is not threatened by either Darwin or artificial intelligence. Basically the argument is no soul, no words. No immaterial form, no meaning. Here is a version of the argument:
1. All relations are either physical or non-physical (i.e., intentional).
2. The relation between a word and its meaning is not a physical relation
3. The person who understands the meaning of a word is active, while the word
itself is passive.
4. That which is capable of action must subsist.
5. Therefore, the agent intellect that understands words must be immaterial and
subsistent.
This is Macuga’s version of Aristotle’s argument (in De Anima) for the immateriality of
the soul. In Q75 of the Summa, Aquinas makes an almost identical argument. This argument might be familiar to some, Huenemann and I have gone to battle over this argument countless times on the usu philosophy blog. Huenemann doesn’t buy it, but (with all due respect to Huenemann) he had really no answer for Machuga when Machuga came here 2 years ago to present his argument.
One quick addition, since SHAFTers are likely to be spooked by any “soul” talk. I include here no religious connotations on the word soul. The Latin for soul is just “anima”. The soul is just that principle that animates. No theological baggage here, this is philosophical hylomorphism, not religion talk.
Just for clarification, my bad computer/animal brain metaphor wasn’t meant to suggest that animal brains are exactly analogous to computers, as if ants were little robots running Windows on tiny chips. I was just trying to point out that I think of brains in terms of electronics (as it seems you do: “parallel,” “feedback loops,” and “rewire”). It’s interesting that we’re always explaining mind in terms of other things rather than explaining other things in terms of mind; as Walker Percy would say, out of all the things in the universe, we know our own selves the least.
I’m more interested in the possibilities of artificial intelligence. I don’t think we’ll ever get there, though we’ll probably write some very good intelligence imitation programs. I’m not sure why I see this as such an impossibility while you seem to think it’s knocking on our door (“We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it). I guess I’m not sure that all mental phenomena is the sort of thing that can be subjected to science, let alone the current paradigms of our neuroscience. I’m very interested in your perspective on this, however, and look forward to your post on artificial intelligence/neuroscience.
I think you’re right; I would probably find evolution fascinating if I got into it. I’m not sure what you mean by “beautiful idea,” other than it explains a lot quite neatly and is good for whapping (fundamentalist) theists over the head. Of course, I can see how some atheists could find Genesis boring. If you don’t think it’s true, you’re only reading it for the literary merit. Since I fancy myself a literary guy, that’s enough for me, but I don’t blame anyone for wanting to skip it.
On one last point, “a system-wide organizing component that directs when, where, and how many connections to form” doesn’t sound like consciousness to me. My unstudied definition of consciousness would have more to do with “knowing you know.” The organizing principle just sounds like self-sufficiency. I do think it’s interesting that you seem to find consciousness in the connections of the brain. One of the most mystifying things about brains (or artificial intelligences) is how a bunch of firing neurons (or 0s and 1s) become thoughts. You seem to find thoughts in the organization of neurons or in the artificial intelligence program; this seems appropriate, as it’s easier to see the immaterial (thoughts) rising from the immaterial (organization or programs). I’m still not sure this chain is free of problems, however.
I just read Kleiner’s post, and although I don’t “find the deferred answer from James to be entirely unsatisfying,” I think Kleiner’s right. Human evolution from and relation to other apes (thanks for the taxonomy tip) doesn’t really matter. Although DNA let’s us know we’re cut from similar stock, our own lived experience lets us know that it doesn’t matter. Not only do we “stay up at night wondering about love and hope and loss and joy,” those purely human experiences and concerns are the major motivations of our human behavior. Neither DNA nor evolutionary theory tell us anything about how to be human.
Although I’m not sure about Machuga’s argument, I can’t see how language could come out of either neurons firing or chips 0 and 1ing. Once again, I look forward to James’ post.
An added point on Source’s first paragraph just above:
The materialist tendency is to explain things in terms of the lower instead of explaining things in terms of the higher. In fact, if there is one difference between pre-modern and modern philosophy (I know, a broad stroke there), I think this is it. As Source points out, Percy really calls readers out on this in his book.
I always find it interesting when people laud evolutionary theory as a “beautiful idea”. It is not that I think ideas can’t be beautiful. But what is beauty? Can you give a materialist account of beauty? I rather doubt it – particularly since beauty seems to have something to do with the organization of parts, and the principle of organization is not itself a material thing. Just another example of materialists using words without thinking about what they are doing and what must be true when they use those words. In fact, they perfomatively disprove their materialism when they say such things.
This is an important thing to emphasize – my target is not evolution (which I don’t find all that threatening and find actually rather interesting). My target is the metaphysics of materialism that the new atheists and SHAFTers always smuggle in (you NEVER hear an argument for it).
“Just for clarification, my bad computer/animal brain metaphor wasn’t meant to suggest that animal brains are exactly analogous to computers, as if ants were little robots running Windows on tiny chips. I was just trying to point out that I think of brains in terms of electronics (as it seems you do: “parallel,” “feedback loops,” and “rewire”).”
It’s funny you mention that, because ants (and other social insects) are probably the best example of biological robots that we have. An individual ant is close to the “given this input and conditions, produce this output” as you described in the context of ape thought. In fact, if there were any sort of “mind” at work here, it would be in the emergent behavior of the ant colony as a whole. The individual ant is a lot more like a neuron. I think the metaphor is pretty bad for ape minds (including ours) and ant colonies, but not for individual ants or neurons. I’m not saying our brains aren’t like computers or aren’t doing computations, they’re just of a different order than the computers we currently have around. (This is all a moot point anyhow if it turns out that consciousness is Turing-complete, although I won’t go into that now.)
“I’m more interested in the possibilities of artificial intelligence. I don’t think we’ll ever get there, though we’ll probably write some very good intelligence imitation programs.”
You might find this interesting. I personally don’t think it much matters if we have a machine with human-level intelligence or a machine that merely simulates human-level intelligence. We can’t really tell the difference, and this is just the problem of other minds applied to machines. As Edsger Dijkstra–one of my favorite computer scientists–said, “The question of whether a computer can think is no more interesting than the question of whether a submarine can swim.” It’s like asking “Yeah, but did Deep Blue really play chess?” It’s a shame, but measuring behavior is all we’ve really got.
“I’m not sure why I see this as such an impossibility while you seem to think it’s knocking on our door (”We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it).”
“Knocking on our door” should be taken with a grain of salt. A few decades (20 – 50 years depending on who’s estimates you look at) for a machine that can pass the Turing test. Probably. The reason I mention brain studies is that I think the first “true AIs” (I say this with a specific meaning in mind, which you probably have an intuitive sense of) will be based off reverse engineering and simulating the operations of a human brain in a mainframe. Figuring out enough in order to do this may lead us to some sort of universal theory of consciousness (as in, systems with properties X, Y, and Z will exhibit behaviors we regard as conscious). I think we already know enough to say that these types of systems will be networks of some kind (ant colonies, brains).
“I think you’re right; I would probably find evolution fascinating if I got into it. I’m not sure what you mean by “beautiful idea,” other than it explains a lot quite neatly and is good for whapping (fundamentalist) theists over the head.”
When I talk about a theory being beautiful, clarity and the succinctness of its explanation are part of it–what mathematicians call “elegance”.
“On one last point, “a system-wide organizing component that directs when, where, and how many connections to form” doesn’t sound like consciousness to me. My unstudied definition of consciousness would have more to do with “knowing you know.””
The organizational component is important, but isn’t close to the whole story. Immediately after their born, human infants have something like ten times as many synapses between the neurons of their brains than adults. As they develop and learn, most of these connections are lost due to a “pruning back” mechanism. It’s like a chaotic bush being cut back into a specific shape. This happens automatically in response to stimulus, beginning as soon as they open their eyes and start trying to process their surroundings. Other animals do this too, but not to the same extent. The amount seems to vary based on how much “preprogramming” a species comes with. Our brains are much more malleable than most, which is part of their success.
This is part of the problem, because “consciousness” doesn’t really have a solid definition. My definition would be something like “theory of mind recursively applied to yourself.” In fact, brain scans show that the same brain regions are activated when we decide our own intentions as when we try to determine someone else’s.
“You seem to find thoughts in the organization of neurons or in the artificial intelligence program; this seems appropriate, as it’s easier to see the immaterial (thoughts) rising from the immaterial (organization or programs).”
Back to the networks, mentioned above. I think–and believe that most neurology and AI researchers would agree–that the pattern of the arrangement and the information it represents is key. The specific material substrate isn’t important, as the same pattern could be represented in meat or in silicon (or whatever else). I suppose that information could be considered immaterial. At the same time though, there must be some type of matter to hold the pattern. My position is essentially this: hardware (material brains) give rise to software (immaterial minds) through the emergent properties of the components that make them up, but you can’t have software without hardware to run it on. To put it in nerdy programmer terms, you can run layers and layers of virtual machines, but there’s got to be metal somewhere.
“Neither DNA nor evolutionary theory tell us anything about how to be human.”
I’m not trying to claim that they do. They may not tell us how to be human, but I do think they tell us an awful lot about what a human is, along with several other sciences. This isn’t worthless, and should not be ignored when we’re trying to answer those other questions.
Claiming that the immaterial mind is merely an “emergent property” of the material components fails to actually explain intentional language (see the argument above, that the immaterial soul must be subsistent, ie not dependent).
Nice to hear James clearly articulate the limits of science. I agree with him – of course genetics and such things tell us a part of the human story. Being is said in many ways, and the scientific way of disclosing human beings is in those terms. It is both interesting and important, but is also insufficient. And we don’t just need other “sciences” to fill out the picture of the human condition. We need philosophy, poetry, religious experience, etc.
“Claiming that the immaterial mind is merely an “emergent property” of the material components fails to actually explain intentional language”
Given our current limited understanding of how the brain produces language (although we can be nearly certain that it does or how it assigns meaning to arbitrary symbols, this is true for now. I don’t see any reason that the above statement should necessarily remain true as we learn more about neurobiology in the coming decades. The emergent behavior of interacting simple rules (4 physical forces and 16 particles in the Standard Model is enough to explain the formation of galaxies, the birth and death of stars, and the evolution of life of Earth. As we are products of that very same stellar and biological evolution, I don’t see any reason why the same type of emergent phenomena shouldn’t be able to produce minds as well as galaxies.
“1. All relations are either physical or non-physical (i.e., intentional).
2. The relation between a word and its meaning is not a physical relation.
3. The person who understands the meaning of a word is active, while the word
itself is passive.
4. That which is capable of action must subsist.
5. Therefore, the agent intellect that understands words must be immaterial and
subsistent. ”
I have to admit I did not understand this at all. It’s using domain-specific terms that I don’t have any handle for. I’m not sure what is meant by “subsist”, “dependent”, or even “immaterial”, “passive” or “active”. Depending on exactly what is meant, Statement 2 doesn’t necessarily have to be true. I can’t say anything about Statement 4; I tried finding a relevant definition, and couldn’t find anything that wasn’t some variation of “To survive on a minimum of resources”. I don’t think that’s what is meant here. All this argument seems to say to me right now is “a ghost explains it”.
My ignorance notwithstanding, it seems that there are other philosophers who don’t see a problem reconciling intentionality with neuroscience and psychology (although I struggled to extract any meaning out of that article section as well). There are also some responses on the Chinese Room article I linked above that seem to be related.
I suppose the crux of it is what is meant by “subsistent, ie not dependent”. I already accept the other conclusion that the mind is immaterial, in the same way that a computer program or the internet or a piece of music is. However, all of these phenomena are tied to (dependent on?) the physical components that produce them (neurons, transistors, single computers or sheet music, instruments, and pressure waves). I guess what I’m saying is that I don’t buy Machuga’s argument (or more precisely your sketch of it), but I can’t answer it in terms you would find satisfying.
It seems patently obvious to me that the mind is dependent on the brain–otherwise, why don’t we see minds untethered to matter, or why don’t we see human-level minds “instantiated in” (attached to, manifested through, whatever) nonhuman animals, regardless of the makeup of their brains? If the agent of word-understanding is not dependent on the physical process of sound waves creating pressure on ear drums, which in turn cause neural signals to be sent to the language networks of the brain which in turn interpret it in some way we don’t yet understand–if this is true, some “channel” between the physical components of the brain and this “agent” must exist that transmits information about which word was heard. (Presumably the agent then responds in some way to send information back to the brain). A causal chain must exist that turns 1) the pattern of air pressure (or the photons of a written word, but this is a slightly different neural process) into 2) an arbitrary symbol, and then 3) attaches meaning to that symbol. These first two (very broad) steps occur in the physical matter of the brain. Could we in theory discover the information channel between the second and third steps? If so, could it not (theoretically) be manipulated or a device engineered in some way as to allow a computer to utilize that channel? After all, the process of symbol-meaning assignment is not dependent on a particular component of the brain, and can be abstracted out to an arbitrary device that simply sends and receives information from the meaning agent.
If the functional properties of the brain are sufficient for assigning meaning to symbols (as I think is the case), then we can understand and replicate that process in a computer of sufficient design. In either case, it should be possible to build machines with true understanding. It’s just that in the first case, we still don’t really understand how it works.
Yes, that argument has some philosophical terminology in that argument. It is hard to avoid. Once soul talk starts up, it is hard to avoid thinking in terms of rigid Platonic/Cartesian dualism (untethered minds, etc). But Aristotle and Aquinas are not dualists. They have a hylomorphic metaphysics. I am ONE thing, not two. I think the soul is “naturally united” to the body (to use Aquinas’ words), it is not the sort of thing you’d find attached to non-human animals. Aquinas’ view would be entirely compatible with our more fully developed understanding of sound, eardrums, the brain, etc etc.
Aristotle knew 2000 years ago that the brain is involved in thought, memory, etc. Those that think this is a recent breakthrough are guilty of chronological snobbery. But just because the brain is necessary for thought and language does not mean it is sufficient. That is such an important point. I continue to be shocked at how often this intro level philosophical distinction is forgotten or ignored by really smart neuroscientists.
If you don’t know what is meant by all of the important terms in the argument (substance, dependent, immaterial, passive, active), then it is rather premautre to say that you “don’t buy Machuga’s argument”. I know you don’t like the conclusion because it offends your ideology, but you can’t judge the argument yet. Seriously, I highly recommend the Machuga book. Perhaps next semester there might be some SHAFTers interested in a reading group where we all discuss it? Being an atheist does not require that you be a materialist, you know. I would be only too pleased to facilitate a group on this, we could meet once a week and discuss a chapter.
Yes, you have some philosophers who are materialists who think they can explain language. But they reduce everything to a behaviorist model, which I think is wrongheaded from the get go (James presumes the behaviorist approach throughout his posts here, and it is only on that reduction of language to a behaviorist account that it makes much sense to speak of apes “learning language”). Bottom line, I think the behaviorists misunderstand what language is, how it works, and have a crappy view of meaning. So when they “explain” intentional language they are not really explaining it.
But these are VERY big questions (language, meaning, intentionality, mind, body, etc etc). Hard to dig into this on a blog. So I’ll restate my offer to read the Machuga book with SHAFTers this spring term. Here is the book:
One more thing to note: these discussions are as old as philosophy. The fancier scientific accounts we have are really important and valuable, but I am of the opinion that they do not change the philosophical questions all that much. In the Phaedo, Plato considers the “harmony objection”, which is basically the view of the mind that James presents here (put in MUCH more primitive terms, though). Have you read the Phaedo, James? You’d likely side more with Simmias and Cebes than Socrates.
Suppose I had built a tiny chip that reproduced in entirety the function of a single neuron, including the ability to form new synapses and power itself from the blood stream. I then pick a random neuron from any point in your brain and replace it with one of these chips. With one neuron out of 100 billion replaced by a chip, for all intents you are unchanged. Keep replacing neurons.
If the immaterial agent that assigns meaning is not dependent on the brain, intentional language will be possible with one neuron replaced, 50% replaced, 99%, or the full brain replaced (this may or may not include the brain stem and spinal column). According to Aristotle as I understand, the formal cause of your brain (the structure of its connections) is unchanged, so you are still fundamentally you. And in your model, there is now a fully artificial brain capable of intentional language through the aid of an immaterial component.
In my model, the outcome is identical. As I think the mind, all its abilities and everything that makes you “you” is a result of the network structure of the synapses in your head, you are still fundamentally you. And since the neural networks responsible for handling symbolic language are still present and have the same structure, you now have a fully artificial brain capable of intentional language.
I would invoke Occam’s Razor at this point, but you seem to hate it.
“Being an atheist does not require that you be a materialist, you know.”
I think patterns are immaterial, but made up of material components. “Soul” could refer to the unique pattern of your brain that is shaped by all your experiences and contains your memories and thoughts and your “you”. In this sense, I think souls exist, but they are obviously not immortal, and while the atoms that make up the pattern will survive indefinitely, the pattern itself–the truly important part–can be destroyed. Does this make me a materialist? I honestly don’t know.
I don’t “hate” ockhams razor. It is a fine principle, though at times too severe. And I think fidelity to it often becomes slavish. I am not offended by plenitude. Anyway, I don’t think it is an absolute principle of intelligibility born from the thigh of Zeus. Too often it is an excuse for excessively severe (and frankly ideological) reductionism. I am NOT a reductionist.
Besides, ockhams razor asks for the theory with the fewest commitments needed to explain the phenonomena … but I don’t think materialism can explain the phenomena. The principle of organization, which has ontological priority since it is the principle of unity, cannot be dependent on the component parts. Quite to the contrary, the lower must be explained in terms of the higher; the way a passible body is must needs be explained by an appeal to form. I don’t think materialism can explain intentional language, and I don’t think a purely material account can explain why/how I am one thing, a unity, that persists through time (that I am the same ME despite changes in my material parts over time). Point is, my objection here is not rooted in a critique of Ockham’s razor, it is rooted in wanting to adequately explain language and myself.
Off to NYC for a conference, so don’t take my silence on any follow-ups personally.
Your thought experiment (“The Brain of Theseus”) was very persuasive. It seems to me that you could replace the brain with tiny brain-chips and still function in much the same way, if the brain-chips acted just like neurons. It was persuasive, however, because I already bought the point you were trying to prove.
It’s similar to the media blitz after someone calls some situation racist/sexist (for example, all the questions of whether Hillary Clinton was the victim of sexism after Obama won the Dem nomination). Some writers will claim there was no sexism, arguing that Hillary was passed over because she rubs people the wrong way (or whatever). Others will argue that she was the victim of sexism. Both sides will use thought experiments as their evidence. “Imagine if she was was a man,” says the sexism-side, “she would never have been treated this way.” We imagine it, and, by Zeus! they’re right. In our imagined world, Hillary as man easily seals the deal. Without sexism, it’s easy for women to succeed.
Of course, this new evidence convincing us that Hillary was the victim of sexism doesn’t come from the world. It comes from our own beliefs. If we believed that feminists were a bunch of PMSing New York FancyPants elites, we would probably find that sexism had nothing to do with Hillary’s failure to get the nomination. Either way, the thought experiment provides us with no new data; it simply reaffirms our beliefs. In rhetorical terms, it’s a reinforcing argument. It works well for those who are already on our side, but it doesn’t affect those who don’t agree.
Of course, just because you used a reinforcing argument doesn’t make your argument wrong; it just means the argument didn’t bring any new evidence to the table. It certainly seems possible to construct an artificial brain by imitating the connections of a real brain. Perhaps future technology will allow neuroscientists to get such a perfect image of a brain that they will be able to construct a mirror image brain out of chips similar to those you describe. That experiment would be a potential paradigm-shifter for all of us.
Perhaps the brain will function just like a human’s. In that case, it would seem that you were right about the organization of connections giving rise to intelligence. If the brain doesn’t work, the upshot will be more ambiguous. Either way, it’s important to note that our reactions to the thought experiment depend on our existing beliefs.
I say all this because we (you and I) might not be right in our thought experiment predictions. If Kleiner is right, it is very possible that the artificial brain would never work. As far as I understand them (and I hope Kleiner can explain this better later), Aristotle’s forms don’t rise out of the organization of given matter. Rather, they give organization to matter. Since this new brain isn’t made of brain stuff, it might not be possible for a human form to work with the brain to create intelligence. This downward flow of organization contradicts our usual idea of upward flow of organization (DNA to cells, cells to organs, etc.), but we can’t say it is wrong simply because it contradicts prior beliefs.
In any case, it’s important to remember that our thinking one way is no justification for our thinking that way. Our ways of thinking are usually the only justifications we have for our ways of thinking. For example, I was once reading Heidegger and thinking, “What a load of bullshit. This attempt to use metaphors to think contradicts every law of good thinking I know.” I realized, however, I had no more base for my ways of thinking than Heidegger had for his (and conversely, no less). Just because I’ve been taught that Ockham’s razor leads to truth doesn’t make it so. Reality might be best explained by a complex theory with many ad hoc adjustments; how can we know? Once we try to step into the noemenal, our best-argued ideas of what makes sense don’t apply. Something’s seeming more probable has nothing to do with it’s truth (or if it does, we don’t know it).
Now, the upshot of all this is that it’s hard to know what neuroscientists will be able to accomplish. The most obvious obstacle to artificial intelligence is the incredible complexity of the human brain, which will probably be overcome in time. Less obvious but more important is our lack of understanding of humanity and speech. If we don’t understand ourselves or our language, how can we build a robot that does? These problems might be solved in time, but I’m not sure they will be. We can (and usually do) look at ourselves as the same as any other thing in the world, but I’m not sure this is justified.
Well, Terry, you are neither an ape nor an angel. You are a human being. There is an essential difference between you and the apes and you and the angels. No Christian or Jew thinks we are “fallen angels”. Nor do you become an angel when you die. What silliness, you are a man. (See Aquinas, Summa Q75.7)
Terry has his finger on the difference. He remarks on how interesting it is that we make (streetlamps, etc) and (though he does not quite say it directly) that we speak. We are makers and speakers. This is a remarkable thing about man. He has put his finger on the right thing, that which makes us essentially different than monkeys. Though he proudly proclaims “I am a monkey”, by doing so he precisely manifests that he is not a monkey by being capable of making such a statement. Man is a rational animal – ‘zoon logikon’ is the expression in Greek. Logos, though, can mean both reasoning but also speaking. Man is the animal who speaks (that is how Heidegger renders that phrase).
As scientists use their words, they never pay attention to the singular miracle that is language. This is one of the main points of Walker Percy’s brilliant book ‘Lost in the Cosmos’ (seriously, I cannot recommend that book enough, one of the most entertaining books I have ever read). What is so interesting about Sagan is that he is so interested in proving that we are not all that unique that he precisely exhibits his own uniqueness (while not realizing it) by speaking and asking those questions!! Scientists, and indeed people in general, live and act in a profound state of self-forgetfulness. But what about this language that we have? What about our ability to make intentional assertions about things? What must be true in order for us to be able to be the kind of speakers we are?
I’m not sure I see your point. Yes, language is a unique trait among humans. As far as we know, we’re the only example of symbolic language use that has arisen on Earth. We’re also Earth’s most prodigious tool -users and -builders. That doesn’t mean we’re not hominids or great apes. Just check a cladogram. (For the record, no, we are not monkeys. We ain’t got tails.)
Even so, our abilities don’t seem to be a “miracle”, in so far as we don’t yet understand exactly how language works in the brain. It’s probably a matter of degree as much as our other skills. Many animals exhibit some measure or other of some particular cognitive skill. Being able to count is likely the basis for logical thought, and elephants, dolphins and chimps can, to a certain extent. Chimps can be taught to use language (ASL, usually) with the relative skill of a preschooler. Yes,they have to be taught, and yes, they max out at the level of a 4 year old human, but the circuitry is there. Chimps in the wild have been shown to make “intentional assertions” (“There’s a cougar in that tree!” or “Hey, I found food over here!”) with the calls and hoots they make. Different chimp tribes in different areas have a different “dialect” of calls. Elephants, dolphins, and certain parrots can show somewhat similar language abilities. For all we know, whalesong might be poetry.
Of course it’s indisputable that humans have a particular synergistic combination of abilities that combined in an unprecedented way, allowing us to eventually build the globe-spanning civilization we have today. We don’t just build tools–we invent technology. We don’t just communicate–we speak and read and write. We walk fully upright (sort of) and have gigantic brains.
No one will deny that this is amazing. But those skills didn’t show up fully-formed, all at once. We’ve been making tools for 2 million years, but they barely changed at all for the first million. The deep roots of language (among the hominid lineage) seem to go back as far as our common ancestor with chimps and gorillas. At the same time, though, it’s unknown whether Neanderthals had the full symbolic language that we do, but I wouldn’t hesitate for a second to call them “human” if that were the case. (The latest research indicates that it is likely that they did have language. It is certain that they never figured out projectile weapons, though.)
My point is this: the fact that we use language doesn’t refute Terry’s description of us as “rising apes.” We are apes. We’re highly derived apes with many specialized skills, but apes we remain.
By the way, I think he was being figurative about the angels thing. Christians and Jews don’t believe we are fallen angels (to paint a broad brush), but many groups do believe we are fallen “from perfection”. Mormons literally do believe we are fallen angels, if I understand what the “Heavenly Host” is. But I guess you took Terry Pratchett “literally” (to possibly misuse the word), because you actually do believe in angels with specific traits. The word “angel” pushes a specific button in your head (symbolic language at work!) that doesn’t precisely match the way I believe he was using it. That’s one of the problems with fuzzy words like “angel”.
I think Kleiner’s point was simply that humans and apes are essentially different. This doesn’t mean that humans didn’t evolve from apes; somewhere along the line, however, humans became something totally distinct from apes. I think it’s clear that this is an essential difference, as Kleiner said. Nobody’s going to mistake an ape for a human. Of course, just because this difference seems obviously essential doesn’t mean we can’t disagree about the degree. I think Kleiner would claim the difference is stark, as big as that that separates a tiger from algae. It’s my feeling that James would compare the difference between humans and apes to the difference between tigers and lions. (Please correct me if I’m wrong.)
It seems clear to me that the difference is stark, at least given my estimation of an ape’s intelligence. Just like everyone else, I’ve never been an ape and so I don’t know how and to what degree they think. Usually, however, I think of animal intelligence under a materialistic model. This means that animals are something like a computer, receiving input and running it through some chips and programs to produce output. I don’t have any idea whether this is actually how an ape’s head works, but if it is, humans and very different.
I remember sitting in one of Huenemann’s classes when he predicted that computers would become conscious as they became more complex. (Please correct me if I misrepresent you.) I have a hard time with this, because it seems so obvious that consciousness is different in kind than unconsciousness. That is, computer’s aren’t conscious because they aren’t going about thinking in the right way, not because they don’t yet have enough connections. Of course, whenever someone says something is “obvious” in a serious discussion, that means they don’t have an argument for it. I’m not sure how to argue that added complexity won’t bring consciousness except by pointing out that extremely complex systems (the internet, for example) don’t seem to be conscious or by asking you to imagine the point at which consciousness will arise out of complexity. What will that be like? How will it happen? At what number (about) will enough connections be made to guarantee consciousness?
On a different note, Pratchett’s claims about the aesthetic benefits of evolution seem sketchy. First, evolution is a boring story. It’s just like all the “begats” he admits skipping in Genesis. (I know the particulars of a species’ evolution can be interesting, but overall, how is this any more exciting than Genesis? Genesis is like the ultimate action movie.) Second, saying “we’ve done pretty good” to evolve to the point where we can build street lamps seems silly. It goes with out saying that streetlamps are awesome, but if we evolved due to the pressures of natural forces, we don’t deserve credit for that (the natural forces do). As Aristotle probably says somewhere in the Nicomachean ethics, we only deserve credit (or blame) for the actions we cause. Although I’m grateful to the natural forces/God/demiurge/whatever that brought my evolution about (I like being a human), I can’t go along with Pratchett and pat myself on the back for it.
“It’s my feeling that James would compare the difference between humans and apes to the difference between tigers and lions.”
Humans are apes. Taxonomically, we are in the hominidae family along with gorillas, chimpanzees, and orangutans. Think hierarchies and subsets of subsets. On a purely genetic level, we are a little further apart from chimps as tigers are from lions–which in fact are very close to each other as they can produce sterile hybrids. (I’m not sure if hybridization is possible between humans and chimps, but I don’t really want to know anyway). But tigers and lions diverged more recently than humans and chimps did, so this is what we’d expect.
Behaviorally, we are worlds apart. All hominids exhibit some capacity for language and culture, but among humans, it’s like we’re running on jet fuel. What caused this? This is an open question and there is lots of ongoing research, and is likely a combination of an increased degree of old traits as well as qualitatively new phenomena. We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it. It could be true that our capacity to create culture and language constitutes a turning point in the history of Earth.
“Usually, however, I think of animal intelligence under a materialistic model. This means that animals are something like a computer, receiving input and running it through some chips and programs to produce output. I don’t have any idea whether this is actually how an ape’s head works, but if it is, humans and very different.”
This is a pretty bad metaphor. Brains–even insect brains–aren’t like any computer we’ve yet built. They’re massively parallel, have all sorts of feedback loops, rewire themselves on the fly, and other messy biology stuff. It can’t be broken into simple input and output. Previous memories, internal states, and chaos (in the mathematical sense) all play a role. And at the very lowest levels, we don’t have any evidence to suggest that our brains are fundamentally different than any other animal. We’ve all the same neurons. I know Kleiner will argue that human intelligence includes non-materialistic factors, and that our minds don’t reside in our brains, and these sorts of things. But that’s not science, and doesn’t contribute anything to our understanding of the brain.
“I’m not sure how to argue that added complexity won’t bring consciousness except by pointing out that extremely complex systems (the internet, for example) don’t seem to be conscious or by asking you to imagine the point at which consciousness will arise out of complexity. What will that be like? How will it happen? At what number (about) will enough connections be made to guarantee consciousness?”
Complexity is a necessary condition for consciousness (whatever that is) but it’s not sufficient in itself. The type of complexity is just as important. The internet is not conscious because it doesn’t have a system-wide organizing component that directs when, where, and how many connections to form–brains do. (Although, for an incredibly loose definition of “conscious”, you might be able to argue that the internet is. It does “know” practically all human knowledge, for example.)
In other words, the software is at least as important as the hardware. It’s very likely that we already have the hardware capability to create artificial intelligences simulating a human brain. We just don’t know how to write that program yet. As for your last few questions in that paragraph, several computer scientists and cognitive neuroscientists have made estimates. If you’re interested I could write up a post on this topic.
“On a different note, Pratchett’s claims about the aesthetic benefits of evolution seem sketchy. First, evolution is a boring story. It’s just like all the “begats” he admits skipping in Genesis. (I know the particulars of a species’ evolution can be interesting, but overall, how is this any more exciting than Genesis? Genesis is like the ultimate action movie.)”
I guess this is a matter of opinion, but Genesis can’t be the ultimate action movie. It doesn’t have semi-trucks crashing into fighter jets, Scottish captains defecting with Russian submarines, or one single Mexican standoff. If you don’t think evolution is interesting (I find it to be one of the most fascinating and beautiful ideas ever thought of) maybe you ought to read more about it. Richard Dawkins’ biology books are really good. I recommend The Blind Watchmaker, Climbing Mount Improbable, and The Ancestors’ Tale. And compared to Genesis, evolution also has the slight bonus of being true.
I have no trouble saying that humans evolved from apes, as JPII said “evolution is more than just a theory”. Frankly, I am pretty agnostic about evolution. I don’t think the story is as ironclad as some would say and I would reject some versions of it, but I think those that deny it outright are a little kooky. I am glad there are really smart people out there trying to work it out. I am agnostic because I think the scientific account of man leaves unanswered (and frankly unasked) most of the existentially interesting question to ask about the human condition. I don’t stay up at night wondering about my genetic ancestry, I do stay up at night wondering about love and hope and loss and joy.
I don’t much care about the taxonimical and zoological distinctions. I understand them to be largely arbitrary. There are very few bright lines to draw. BUT, I think there is a really bright line between man and the apes from which he arose. I think there is clear ontological leap there. Language is the clearest indication of this ontological leap. And this is not just a matter of us not understanding well enough yet how language works in the brain. My first post made a challenge about intentional language and how to explain it. My claim (actually it is Aristotle’s), is that intentional language cannot be explained in totally materialistic terms. We could understand the brain perfectly but still not explain intentional language. In other words, I find the deferred answer from James to be entirely unsatisfying. Yes, scientists are learning more and more about the brain (for which I am glad), but the materialist/evolutionary story just does a terrible job of explaining the remarkable difference between man and the rest of the animal world (language, culture, meaning, etc).
Suggested reading: Ric Machuga’s book: ‘In Defense of the Soul’. It includes chapters on why Aristotelianism is not threatened by either Darwin or artificial intelligence. Basically the argument is no soul, no words. No immaterial form, no meaning. Here is a version of the argument:
1. All relations are either physical or non-physical (i.e., intentional).
2. The relation between a word and its meaning is not a physical relation
3. The person who understands the meaning of a word is active, while the word
itself is passive.
4. That which is capable of action must subsist.
5. Therefore, the agent intellect that understands words must be immaterial and
subsistent.
This is Macuga’s version of Aristotle’s argument (in De Anima) for the immateriality of
the soul. In Q75 of the Summa, Aquinas makes an almost identical argument. This argument might be familiar to some, Huenemann and I have gone to battle over this argument countless times on the usu philosophy blog. Huenemann doesn’t buy it, but (with all due respect to Huenemann) he had really no answer for Machuga when Machuga came here 2 years ago to present his argument.
One quick addition, since SHAFTers are likely to be spooked by any “soul” talk. I include here no religious connotations on the word soul. The Latin for soul is just “anima”. The soul is just that principle that animates. No theological baggage here, this is philosophical hylomorphism, not religion talk.
Happy Christmas!
@ James:
Just for clarification, my bad computer/animal brain metaphor wasn’t meant to suggest that animal brains are exactly analogous to computers, as if ants were little robots running Windows on tiny chips. I was just trying to point out that I think of brains in terms of electronics (as it seems you do: “parallel,” “feedback loops,” and “rewire”). It’s interesting that we’re always explaining mind in terms of other things rather than explaining other things in terms of mind; as Walker Percy would say, out of all the things in the universe, we know our own selves the least.
I’m more interested in the possibilities of artificial intelligence. I don’t think we’ll ever get there, though we’ll probably write some very good intelligence imitation programs. I’m not sure why I see this as such an impossibility while you seem to think it’s knocking on our door (“We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it). I guess I’m not sure that all mental phenomena is the sort of thing that can be subjected to science, let alone the current paradigms of our neuroscience. I’m very interested in your perspective on this, however, and look forward to your post on artificial intelligence/neuroscience.
I think you’re right; I would probably find evolution fascinating if I got into it. I’m not sure what you mean by “beautiful idea,” other than it explains a lot quite neatly and is good for whapping (fundamentalist) theists over the head. Of course, I can see how some atheists could find Genesis boring. If you don’t think it’s true, you’re only reading it for the literary merit. Since I fancy myself a literary guy, that’s enough for me, but I don’t blame anyone for wanting to skip it.
On one last point, “a system-wide organizing component that directs when, where, and how many connections to form” doesn’t sound like consciousness to me. My unstudied definition of consciousness would have more to do with “knowing you know.” The organizing principle just sounds like self-sufficiency. I do think it’s interesting that you seem to find consciousness in the connections of the brain. One of the most mystifying things about brains (or artificial intelligences) is how a bunch of firing neurons (or 0s and 1s) become thoughts. You seem to find thoughts in the organization of neurons or in the artificial intelligence program; this seems appropriate, as it’s easier to see the immaterial (thoughts) rising from the immaterial (organization or programs). I’m still not sure this chain is free of problems, however.
I just read Kleiner’s post, and although I don’t “find the deferred answer from James to be entirely unsatisfying,” I think Kleiner’s right. Human evolution from and relation to other apes (thanks for the taxonomy tip) doesn’t really matter. Although DNA let’s us know we’re cut from similar stock, our own lived experience lets us know that it doesn’t matter. Not only do we “stay up at night wondering about love and hope and loss and joy,” those purely human experiences and concerns are the major motivations of our human behavior. Neither DNA nor evolutionary theory tell us anything about how to be human.
Although I’m not sure about Machuga’s argument, I can’t see how language could come out of either neurons firing or chips 0 and 1ing. Once again, I look forward to James’ post.
An added point on Source’s first paragraph just above:
The materialist tendency is to explain things in terms of the lower instead of explaining things in terms of the higher. In fact, if there is one difference between pre-modern and modern philosophy (I know, a broad stroke there), I think this is it. As Source points out, Percy really calls readers out on this in his book.
I always find it interesting when people laud evolutionary theory as a “beautiful idea”. It is not that I think ideas can’t be beautiful. But what is beauty? Can you give a materialist account of beauty? I rather doubt it – particularly since beauty seems to have something to do with the organization of parts, and the principle of organization is not itself a material thing. Just another example of materialists using words without thinking about what they are doing and what must be true when they use those words. In fact, they perfomatively disprove their materialism when they say such things.
This is an important thing to emphasize – my target is not evolution (which I don’t find all that threatening and find actually rather interesting). My target is the metaphysics of materialism that the new atheists and SHAFTers always smuggle in (you NEVER hear an argument for it).
“Just for clarification, my bad computer/animal brain metaphor wasn’t meant to suggest that animal brains are exactly analogous to computers, as if ants were little robots running Windows on tiny chips. I was just trying to point out that I think of brains in terms of electronics (as it seems you do: “parallel,” “feedback loops,” and “rewire”).”
It’s funny you mention that, because ants (and other social insects) are probably the best example of biological robots that we have. An individual ant is close to the “given this input and conditions, produce this output” as you described in the context of ape thought. In fact, if there were any sort of “mind” at work here, it would be in the emergent behavior of the ant colony as a whole. The individual ant is a lot more like a neuron. I think the metaphor is pretty bad for ape minds (including ours) and ant colonies, but not for individual ants or neurons. I’m not saying our brains aren’t like computers or aren’t doing computations, they’re just of a different order than the computers we currently have around. (This is all a moot point anyhow if it turns out that consciousness is Turing-complete, although I won’t go into that now.)
“I’m more interested in the possibilities of artificial intelligence. I don’t think we’ll ever get there, though we’ll probably write some very good intelligence imitation programs.”
You might find this interesting. I personally don’t think it much matters if we have a machine with human-level intelligence or a machine that merely simulates human-level intelligence. We can’t really tell the difference, and this is just the problem of other minds applied to machines. As Edsger Dijkstra–one of my favorite computer scientists–said, “The question of whether a computer can think is no more interesting than the question of whether a submarine can swim.” It’s like asking “Yeah, but did Deep Blue really play chess?” It’s a shame, but measuring behavior is all we’ve really got.
“I’m not sure why I see this as such an impossibility while you seem to think it’s knocking on our door (”We don’t have a solid understanding of the entire brain or its evolution yet, but people are working on it).”
“Knocking on our door” should be taken with a grain of salt. A few decades (20 – 50 years depending on who’s estimates you look at) for a machine that can pass the Turing test. Probably. The reason I mention brain studies is that I think the first “true AIs” (I say this with a specific meaning in mind, which you probably have an intuitive sense of) will be based off reverse engineering and simulating the operations of a human brain in a mainframe. Figuring out enough in order to do this may lead us to some sort of universal theory of consciousness (as in, systems with properties X, Y, and Z will exhibit behaviors we regard as conscious). I think we already know enough to say that these types of systems will be networks of some kind (ant colonies, brains).
“I think you’re right; I would probably find evolution fascinating if I got into it. I’m not sure what you mean by “beautiful idea,” other than it explains a lot quite neatly and is good for whapping (fundamentalist) theists over the head.”
When I talk about a theory being beautiful, clarity and the succinctness of its explanation are part of it–what mathematicians call “elegance”.
“On one last point, “a system-wide organizing component that directs when, where, and how many connections to form” doesn’t sound like consciousness to me. My unstudied definition of consciousness would have more to do with “knowing you know.””
The organizational component is important, but isn’t close to the whole story. Immediately after their born, human infants have something like ten times as many synapses between the neurons of their brains than adults. As they develop and learn, most of these connections are lost due to a “pruning back” mechanism. It’s like a chaotic bush being cut back into a specific shape. This happens automatically in response to stimulus, beginning as soon as they open their eyes and start trying to process their surroundings. Other animals do this too, but not to the same extent. The amount seems to vary based on how much “preprogramming” a species comes with. Our brains are much more malleable than most, which is part of their success.
This is part of the problem, because “consciousness” doesn’t really have a solid definition. My definition would be something like “theory of mind recursively applied to yourself.” In fact, brain scans show that the same brain regions are activated when we decide our own intentions as when we try to determine someone else’s.
“You seem to find thoughts in the organization of neurons or in the artificial intelligence program; this seems appropriate, as it’s easier to see the immaterial (thoughts) rising from the immaterial (organization or programs).”
Back to the networks, mentioned above. I think–and believe that most neurology and AI researchers would agree–that the pattern of the arrangement and the information it represents is key. The specific material substrate isn’t important, as the same pattern could be represented in meat or in silicon (or whatever else). I suppose that information could be considered immaterial. At the same time though, there must be some type of matter to hold the pattern. My position is essentially this: hardware (material brains) give rise to software (immaterial minds) through the emergent properties of the components that make them up, but you can’t have software without hardware to run it on. To put it in nerdy programmer terms, you can run layers and layers of virtual machines, but there’s got to be metal somewhere.
“Neither DNA nor evolutionary theory tell us anything about how to be human.”
I’m not trying to claim that they do. They may not tell us how to be human, but I do think they tell us an awful lot about what a human is, along with several other sciences. This isn’t worthless, and should not be ignored when we’re trying to answer those other questions.
Claiming that the immaterial mind is merely an “emergent property” of the material components fails to actually explain intentional language (see the argument above, that the immaterial soul must be subsistent, ie not dependent).
Nice to hear James clearly articulate the limits of science. I agree with him – of course genetics and such things tell us a part of the human story. Being is said in many ways, and the scientific way of disclosing human beings is in those terms. It is both interesting and important, but is also insufficient. And we don’t just need other “sciences” to fill out the picture of the human condition. We need philosophy, poetry, religious experience, etc.
“Claiming that the immaterial mind is merely an “emergent property” of the material components fails to actually explain intentional language”
Given our current limited understanding of how the brain produces language (although we can be nearly certain that it does
or how it assigns meaning to arbitrary symbols, this is true for now. I don’t see any reason that the above statement should necessarily remain true as we learn more about neurobiology in the coming decades. The emergent behavior of interacting simple rules (4 physical forces and 16 particles in the Standard Model 
is enough to explain the formation of galaxies, the birth and death of stars, and the evolution of life of Earth. As we are products of that very same stellar and biological evolution, I don’t see any reason why the same type of emergent phenomena shouldn’t be able to produce minds as well as galaxies.
“1. All relations are either physical or non-physical (i.e., intentional).
2. The relation between a word and its meaning is not a physical relation.
3. The person who understands the meaning of a word is active, while the word
itself is passive.
4. That which is capable of action must subsist.
5. Therefore, the agent intellect that understands words must be immaterial and
subsistent. ”
I have to admit I did not understand this at all. It’s using domain-specific terms that I don’t have any handle for. I’m not sure what is meant by “subsist”, “dependent”, or even “immaterial”, “passive” or “active”. Depending on exactly what is meant, Statement 2 doesn’t necessarily have to be true. I can’t say anything about Statement 4; I tried finding a relevant definition, and couldn’t find anything that wasn’t some variation of “To survive on a minimum of resources”. I don’t think that’s what is meant here. All this argument seems to say to me right now is “a ghost explains it”.
My ignorance notwithstanding, it seems that there are other philosophers who don’t see a problem reconciling intentionality with neuroscience and psychology (although I struggled to extract any meaning out of that article section as well). There are also some responses on the Chinese Room article I linked above that seem to be related.
I suppose the crux of it is what is meant by “subsistent, ie not dependent”. I already accept the other conclusion that the mind is immaterial, in the same way that a computer program or the internet or a piece of music is. However, all of these phenomena are tied to (dependent on?) the physical components that produce them (neurons, transistors, single computers or sheet music, instruments, and pressure waves). I guess what I’m saying is that I don’t buy Machuga’s argument (or more precisely your sketch of it), but I can’t answer it in terms you would find satisfying.
It seems patently obvious to me that the mind is dependent on the brain–otherwise, why don’t we see minds untethered to matter, or why don’t we see human-level minds “instantiated in” (attached to, manifested through, whatever) nonhuman animals, regardless of the makeup of their brains? If the agent of word-understanding is not dependent on the physical process of sound waves creating pressure on ear drums, which in turn cause neural signals to be sent to the language networks of the brain which in turn interpret it in some way we don’t yet understand–if this is true, some “channel” between the physical components of the brain and this “agent” must exist that transmits information about which word was heard. (Presumably the agent then responds in some way to send information back to the brain). A causal chain must exist that turns 1) the pattern of air pressure (or the photons of a written word, but this is a slightly different neural process) into 2) an arbitrary symbol, and then 3) attaches meaning to that symbol. These first two (very broad) steps occur in the physical matter of the brain. Could we in theory discover the information channel between the second and third steps? If so, could it not (theoretically) be manipulated or a device engineered in some way as to allow a computer to utilize that channel? After all, the process of symbol-meaning assignment is not dependent on a particular component of the brain, and can be abstracted out to an arbitrary device that simply sends and receives information from the meaning agent.
If the functional properties of the brain are sufficient for assigning meaning to symbols (as I think is the case), then we can understand and replicate that process in a computer of sufficient design. In either case, it should be possible to build machines with true understanding. It’s just that in the first case, we still don’t really understand how it works.
Yes, that argument has some philosophical terminology in that argument. It is hard to avoid. Once soul talk starts up, it is hard to avoid thinking in terms of rigid Platonic/Cartesian dualism (untethered minds, etc). But Aristotle and Aquinas are not dualists. They have a hylomorphic metaphysics. I am ONE thing, not two. I think the soul is “naturally united” to the body (to use Aquinas’ words), it is not the sort of thing you’d find attached to non-human animals. Aquinas’ view would be entirely compatible with our more fully developed understanding of sound, eardrums, the brain, etc etc.
Aristotle knew 2000 years ago that the brain is involved in thought, memory, etc. Those that think this is a recent breakthrough are guilty of chronological snobbery. But just because the brain is necessary for thought and language does not mean it is sufficient. That is such an important point. I continue to be shocked at how often this intro level philosophical distinction is forgotten or ignored by really smart neuroscientists.
If you don’t know what is meant by all of the important terms in the argument (substance, dependent, immaterial, passive, active), then it is rather premautre to say that you “don’t buy Machuga’s argument”. I know you don’t like the conclusion because it offends your ideology, but you can’t judge the argument yet. Seriously, I highly recommend the Machuga book. Perhaps next semester there might be some SHAFTers interested in a reading group where we all discuss it? Being an atheist does not require that you be a materialist, you know. I would be only too pleased to facilitate a group on this, we could meet once a week and discuss a chapter.
Yes, you have some philosophers who are materialists who think they can explain language. But they reduce everything to a behaviorist model, which I think is wrongheaded from the get go (James presumes the behaviorist approach throughout his posts here, and it is only on that reduction of language to a behaviorist account that it makes much sense to speak of apes “learning language”). Bottom line, I think the behaviorists misunderstand what language is, how it works, and have a crappy view of meaning. So when they “explain” intentional language they are not really explaining it.
But these are VERY big questions (language, meaning, intentionality, mind, body, etc etc). Hard to dig into this on a blog. So I’ll restate my offer to read the Machuga book with SHAFTers this spring term. Here is the book:
http://www.amazon.com/Defense-Soul-What-Means-Human/dp/1587430290/ref=sr_1_1?ie=UTF8&s=books&qid=1261687090&sr=8-1
One more thing to note: these discussions are as old as philosophy. The fancier scientific accounts we have are really important and valuable, but I am of the opinion that they do not change the philosophical questions all that much. In the Phaedo, Plato considers the “harmony objection”, which is basically the view of the mind that James presents here (put in MUCH more primitive terms, though). Have you read the Phaedo, James? You’d likely side more with Simmias and Cebes than Socrates.
Or think of a variation of the Ship of Theseus.
Suppose I had built a tiny chip that reproduced in entirety the function of a single neuron, including the ability to form new synapses and power itself from the blood stream. I then pick a random neuron from any point in your brain and replace it with one of these chips. With one neuron out of 100 billion replaced by a chip, for all intents you are unchanged. Keep replacing neurons.
If the immaterial agent that assigns meaning is not dependent on the brain, intentional language will be possible with one neuron replaced, 50% replaced, 99%, or the full brain replaced (this may or may not include the brain stem and spinal column). According to Aristotle as I understand, the formal cause of your brain (the structure of its connections) is unchanged, so you are still fundamentally you. And in your model, there is now a fully artificial brain capable of intentional language through the aid of an immaterial component.
In my model, the outcome is identical. As I think the mind, all its abilities and everything that makes you “you” is a result of the network structure of the synapses in your head, you are still fundamentally you. And since the neural networks responsible for handling symbolic language are still present and have the same structure, you now have a fully artificial brain capable of intentional language.
I would invoke Occam’s Razor at this point, but you seem to hate it.
“Being an atheist does not require that you be a materialist, you know.”
I think patterns are immaterial, but made up of material components. “Soul” could refer to the unique pattern of your brain that is shaped by all your experiences and contains your memories and thoughts and your “you”. In this sense, I think souls exist, but they are obviously not immortal, and while the atoms that make up the pattern will survive indefinitely, the pattern itself–the truly important part–can be destroyed. Does this make me a materialist? I honestly don’t know.
Hope you’re having a happy Christmas Eve with your family, by the way.
I don’t “hate” ockhams razor. It is a fine principle, though at times too severe. And I think fidelity to it often becomes slavish. I am not offended by plenitude. Anyway, I don’t think it is an absolute principle of intelligibility born from the thigh of Zeus. Too often it is an excuse for excessively severe (and frankly ideological) reductionism. I am NOT a reductionist.
Besides, ockhams razor asks for the theory with the fewest commitments needed to explain the phenonomena … but I don’t think materialism can explain the phenomena. The principle of organization, which has ontological priority since it is the principle of unity, cannot be dependent on the component parts. Quite to the contrary, the lower must be explained in terms of the higher; the way a passible body is must needs be explained by an appeal to form. I don’t think materialism can explain intentional language, and I don’t think a purely material account can explain why/how I am one thing, a unity, that persists through time (that I am the same ME despite changes in my material parts over time). Point is, my objection here is not rooted in a critique of Ockham’s razor, it is rooted in wanting to adequately explain language and myself.
Off to NYC for a conference, so don’t take my silence on any follow-ups personally.
@ James:
Your thought experiment (“The Brain of Theseus”) was very persuasive. It seems to me that you could replace the brain with tiny brain-chips and still function in much the same way, if the brain-chips acted just like neurons. It was persuasive, however, because I already bought the point you were trying to prove.
It’s similar to the media blitz after someone calls some situation racist/sexist (for example, all the questions of whether Hillary Clinton was the victim of sexism after Obama won the Dem nomination). Some writers will claim there was no sexism, arguing that Hillary was passed over because she rubs people the wrong way (or whatever). Others will argue that she was the victim of sexism. Both sides will use thought experiments as their evidence. “Imagine if she was was a man,” says the sexism-side, “she would never have been treated this way.” We imagine it, and, by Zeus! they’re right. In our imagined world, Hillary as man easily seals the deal. Without sexism, it’s easy for women to succeed.
Of course, this new evidence convincing us that Hillary was the victim of sexism doesn’t come from the world. It comes from our own beliefs. If we believed that feminists were a bunch of PMSing New York FancyPants elites, we would probably find that sexism had nothing to do with Hillary’s failure to get the nomination. Either way, the thought experiment provides us with no new data; it simply reaffirms our beliefs. In rhetorical terms, it’s a reinforcing argument. It works well for those who are already on our side, but it doesn’t affect those who don’t agree.
Of course, just because you used a reinforcing argument doesn’t make your argument wrong; it just means the argument didn’t bring any new evidence to the table. It certainly seems possible to construct an artificial brain by imitating the connections of a real brain. Perhaps future technology will allow neuroscientists to get such a perfect image of a brain that they will be able to construct a mirror image brain out of chips similar to those you describe. That experiment would be a potential paradigm-shifter for all of us.
Perhaps the brain will function just like a human’s. In that case, it would seem that you were right about the organization of connections giving rise to intelligence. If the brain doesn’t work, the upshot will be more ambiguous. Either way, it’s important to note that our reactions to the thought experiment depend on our existing beliefs.
I say all this because we (you and I) might not be right in our thought experiment predictions. If Kleiner is right, it is very possible that the artificial brain would never work. As far as I understand them (and I hope Kleiner can explain this better later), Aristotle’s forms don’t rise out of the organization of given matter. Rather, they give organization to matter. Since this new brain isn’t made of brain stuff, it might not be possible for a human form to work with the brain to create intelligence. This downward flow of organization contradicts our usual idea of upward flow of organization (DNA to cells, cells to organs, etc.), but we can’t say it is wrong simply because it contradicts prior beliefs.
In any case, it’s important to remember that our thinking one way is no justification for our thinking that way. Our ways of thinking are usually the only justifications we have for our ways of thinking. For example, I was once reading Heidegger and thinking, “What a load of bullshit. This attempt to use metaphors to think contradicts every law of good thinking I know.” I realized, however, I had no more base for my ways of thinking than Heidegger had for his (and conversely, no less). Just because I’ve been taught that Ockham’s razor leads to truth doesn’t make it so. Reality might be best explained by a complex theory with many ad hoc adjustments; how can we know? Once we try to step into the noemenal, our best-argued ideas of what makes sense don’t apply. Something’s seeming more probable has nothing to do with it’s truth (or if it does, we don’t know it).
Now, the upshot of all this is that it’s hard to know what neuroscientists will be able to accomplish. The most obvious obstacle to artificial intelligence is the incredible complexity of the human brain, which will probably be overcome in time. Less obvious but more important is our lack of understanding of humanity and speech. If we don’t understand ourselves or our language, how can we build a robot that does? These problems might be solved in time, but I’m not sure they will be. We can (and usually do) look at ourselves as the same as any other thing in the world, but I’m not sure this is justified.