As you may have figured out from this OP, I am basically a half-wit when it comes to topics like these, but am interested. Also, in the light of day, I think it's "Alex" the parrot and Irene was the parrot trainer.
On the subject of octopus intelligence.
Animals that are very very smart that don't get the dolphin/chimp headlines: elephants, killer whales, and crows. Everyone should look at some of the new caledonia crow footage. Killer Whales have an amazing level of culture passed down, for example they have strong cultural prejudices about what food to eat (if you were raised on seals you won't eat fish), and mothers teach their kids to overcome the beaching instinct so that the can hunt on the shore.
What the hell is up with the beaching instinct? What is that?
I was writing a big long thing about this and then realized that I completely don't have time today at all.
The short answer is that it's probably possible to talk seriously about this, but there's a lot of disagreement on how. Some people think that language underlies human general intelligence -- see here -- but a lot of people don't. The question of what general intelligence is and if it can be effectively measured is also very much open. One candidate for measuring it is by measuring this kind of thing, which points to something going on somewhere in the brain that happens serially (as opposed to most perceptual and motor and decision processes, which are at least somewhat paralellized). For some of the people who believe in the primacy of linguistic processes, that's sort of tantalizing, because spoken language is a serial process. But it could easily be something else entirely (this guy, who has done a lot of the work on the PRP, would probably argue fairly strenuously that language was not the key faculty).
As far as the parrot (who was named Alex; Irene is the woman who trained him), he undeniably learned some really impressive things. But whether those count as "language", or what "language" really means, are subjects of great ongoing debate.
The question of how much you can anthropomorphize animal social cognition (especially canine social cognition) is a whole other bag of worms.
Anyhow, the Tomasello book is well-written and accessible, and a pretty good overview of the animal cognition landscape (with some pretty strong biases).
One of the features of human intelligence historically thought to have been distinguishing is theory of mind, something that most children develop in more or less fully-fledged form by about 5 years. People with autistic spectrum disorders are typically impaired in this area.
Most research into animals has shown (or at least proved) only limited theory of mind. Dogs, for instance are good at some aspects of theory of mind (understanding human pointing/gaze, for instance) which chimpanzees can't/don't do. This isn't too surprising from an evolutionary perspective, despite chimps being genetically closer, given that for 7,000 years dogs have been selected for their ability to understand the wishes of their owners. Evidence is mixed on other aspects such as acting upon another's true or false beliefs, but there is at least some empirical support that chimps can and do do this, even in the wild. A fair bit of chimpanzee power politics seems to be based upon what a given member of the troop knows or doesn't know.
What is "theory of mind"? Metacognition? Realizing that other individuals have separate experiences and that they're talking from their perspective, as opposed to keeping everything omniscient?
For some of the people who believe in the primacy of linguistic processes, that's sort of tantalizing, because spoken language is a serial process.
It's not always strictly serial, though. There's plenty of research to demonstrate that in production we anticipate words/sounds to come and in comprehension we revise our initial interpretations based on new information.
Fish-eating orcas (which live in small regions in large complex groups and communicate a lot via sound because fish can't hear well) will *pretend to be seal eating Orcas* (who travel much larger areas and hunt relatively silently) to scare seals away from fish. That probably doesn't really need a theory of the mind, but it is pretty awesome.
What is "theory of mind"? Metacognition?
Basically, yes, though it is obviously a complex subject. In short, understanding that other people/animals have their own minds and their own knowledge, emotions, goals etc.
Most research into animals has shown (or at least proved) only limited theory of mind.
I think negative results from these experiments don't tell us anything (that is, it's always possible that the experiment isn't testing a situation that would evoke the right response). If she wants something I've denied her, one of my cats will wait until my attention wanders, which strikes me as a very useful predatory behavior and one requiring a theory of mind.
Well, limited 'theory of mind' in the context of reading other animals' behaviour and making assumptions about their state of mind must be fairly common - threat displays etc.
mothers teach their kids to overcome the beaching instinct so that the can hunt on the shore.
I've seen video of dolphins doing this, and it's very clever. I worry that it's evidence that their normal food sources are disappearing, but that's probably just my paranoia.
Every now and then I start to suspect you each have your own minds.
Has it been shown that other minds exist, or that the mind esists at all? I thought that that was controversial. Shouldn't animals which seem so sure that minds exist be thought of as unintelligent, or at least as unphilosophical?
8: Absolutely, but it makes more sense to say that spoken language has to have a significant serial component than it does to say that about, say, object tracking, or episodic memory.
11, 12: I think predicting another animal's behavior on the basis of demeanor (threat displays, inattention) doesn't require theory of mind -- you need to be reacting to what you know about the other animal's state of knowledge, not to what you know about its propensity to behave in a certain way.
This is right in rfts's wheelhouse, so I'll ask her to take a crack at the questions when she gets back.
12: you can get a lot of that behaviorally without resort to theory of mind per se; you could have a learned reaction (when that thing over there gets bigger and makes loud noises in that way that it does, run) without having any real sense that the other animal has thoughts and emotions and goals and so on.
I don't have any insight into the definition of intelligence, but here are a few facts and thoughts.
Many birds and octopus have incredibly powerful spatial memories, so are excellent navigators. Octopus are also very good at manipulating small objects to see if there's anything inside.
Smart birds (finches, parrots, ravens) and smart mammals are all pretty long-lived-- decades. Octopuses do not have long lives. This seems significant to me, because having time to benefit from an investment in learning seems like a necessary condition for real intelligence. So spatial ability is handy, provides leverage for abstraction, but is IMO not the main point.
Humans and chimps are 15-50x more different from each other genetically than are a pair of humans. That's a lot of difference. Human-chimp paper
Language and an ability to sustain abstraction seem very closely tied together, and to amplify social intelligence. Understanding the inevitability of death seems like a big species-independent threshold of intelligence. I like Gary Marcus' books and articles for reading about how the human brain works.
It's a good question, I think that looking at the content of birdsong and cetacean communication is worthwhile. The mathematical methods for analyzing information content suggest that they are much poorer than human language. I suspect that we are the first species to have passed some threshold, that like eyes, intelligence has evolved many times independently.
Given the problems people have defining human intelligence (see IQ, Steven Pinker, etc. passim), and artificial intelligence, is this question even susceptible to inquiry?
The smart-ass part of me thinks, whenever people are describing the brilliance of chimps/hummingbirds/whatever, "But really. We have X-boxes. And trips to the freaking moon. And sky-scrapers."
22: But we still haven't learned to live in harmony with the Earth! Perhaps it is we who are the "dumb animals"!
Man. Nobody really knows anything about protein folding, but everybody understands the nature of intelligence! Who knew it was such an easy problem.
Douglas Adams:
It is an important and popular fact that things are not always what they seem. For instance, on the planet Earth, man had always assumed that he was more intelligent than dolphins because he had achieved so much -- the wheel, New York, wars and so on -- whilst all the dolphins had ever done was muck about in the water having a good time. But conversely, the dolphins had always believed that they were far more intelligent than man -- for precisely the same reasons.
Or, alternately, it's such a hard problem that the failure of some of the initial approaches to studying it means that it cannot be studied at all.
No way. The other animals aren't smart enough to kill the earth as well as we do.
Vancouver is famous for being the home of Wesley Allan Dodd. They're probably happy with being misidentified.
No mind, no world, no buddha
So there, Emerson
And the stupid commie walked us in the freezing rain, and then gave us a treat for it. Who's smarter?
Birds have language. I watched a documentary about turkeys last week where this guy figured out 20 different turkey noises that correspond to different meanings including "That bird in the sky won't hurt us, but be mindful" and "Oh shit, there's a rattlesnake right there!"
Crows are very smart. They can distinguish between human faces and tell other crows which humans to watch out for, they use tools and they use their beaks and wings in the same way that humans use their hands.
Which species has the most words for "snow"?
Quick note on Alex:
While Alex's achievements probably don't count as "having language" in any important sense of the term, they are far from being bullshit. This was not a "clever Hans" situation, where the results are just a trick that don't reflect animal intelligence at all.
Alex was able to identify many complicated properties of things he saw and describe them using complicated noun phrases. If presented with five blue cubes, he could say "five blue cubes." If you presented him with five blue cubes and three yellow pyramids, and asked him how many blue cubes there were, he would say "five."
My take on this is that demonstrates the incredible visual intelligence of parrots. Alex had a detailed grasp of what was in his visual field, and was able to associate all of these visual details with phrases.
This does not amount to the Chomskyite holy of holies, a knowledge of recursive grammar. But it ain't bullshit. He was smart the way parrots are supposed to be smart: visually.
Given the problems people have defining human intelligence (see IQ, Steven Pinker, etc. passim), and artificial intelligence, is this question even susceptible to inquiry?
So far as I know, no one's saying anything like 'The average octopus has an IQ of 78'. (Okay, maybe someone says that. That would be stupid.) But studying what kinds of things different animals are capable of and what kinds of mental capacities those capabilities would seem to indicate seems quite within the realm of science.
How far did other people read in 33 before they realized this was not referring to the Yorkshire Ranter?
Sure, they may not have X-Boxes, but pidgeons have been observed playing a crude version of Pong, so really it's just a difference of processer speed and memory.
I used to have an African Grey when I was little and he would specify what he wanted. He would ask for scratches and name the type of treat he wanted (grape, peanut). He also had a great sense of humor and would mimic my mother's voice to call for me and make me run downstairs. I've heard anecdotes of parrots calling for the dog and then scolding him.
I've heard anecdotes of parrots calling for the dog and then scolding him.
From the dogs?! That is some verbal ability.
The question of what general intelligence is and if it can be effectively measured is also very much open. One candidate for measuring it is by measuring this kind of thing, which points to something going on somewhere in the brain that happens serially (as opposed to most perceptual and motor and decision processes, which are at least somewhat paralellized). For some of the people who believe in the primacy of linguistic processes, that's sort of tantalizing, because spoken language is a serial process.
That's really interesting -- if I'm understanding the idea correctly at all, is a semi-accurate oversimplification that the ability to quickly, but serially, process a chain of information may be the key, as opposed to quickly perform a function in parallel. And the serially-processed activity may (or may not be) language; language might be the key thing that we've evolved to process serially, and therefore our brains are good serial processors, or it might be that our brains are good serial processors and therefore we are able to use language.*
*N.b., I only have the vaguest notion of the difference between serial and parallel processing.
He also had a great sense of humor and would mimic my mother's voice to call for me and make me run downstairs.
They seem to be good at that.
39: I can't believe he gave up the parrot! That parrot was totally watching his back.
||
Life imitates The Onion again.
|>
9 is very cool, though it doesn't say much for the seals that they're fooled.
Isn't there something about each of an octopus's arms being controlled by its own brainlet? Or something? (rfts showed me this a couple of years ago and I have been laughing about it ever since.)
GY, how does the anticipation-and-revision process you mention work? I'm picturing it as being sort of like Dasher. Is it at all well understood?
Humans and chimps are 15-50x more different from each other genetically than are a pair of humans. That's a lot of difference.
I feel like this fact does not get a lot of play in the popular press.
38: That's about the size of it. The hypothesis is that there's some kind of cognitive faculty that humans have, but that no animals have, which allows us to do this kind of symbolic manipulation of the inputs and outputs of other cognitive faculties (memory, vision, proprioception, executive function, whatever), all of which are shared by non-human primates (at least). This function is assumed to be essential for language, and correlates well with experiments that measure the psychological refractory period, which itself correlates strongly with other things (IQ tests and the like) that theoretically measure general intelligence. If you buy all that, the question then becomes whether we developed language first, and got all the rest of the symbolic intelligence as kind of a gimme when we figured out how to do the symbolic manipulation necessary to express concepts more complicated than "AAAH, EAGLE", or whether we developed the symbolic intelligence first and that left us well-placed to develop language.
Of course, all of the hypotheses you have to believe even to get to that question are pretty controversial themselves; lots of people would argue that symbolic manipulation isn't actually very important to practical general intelligence, that it's not actually very important to spoken language, that practical general intelligence isn't a cognitive function that's terribly necessary to most things that allegedly require "smarts", that the correlation of the PRP and other measures of general intelligence isn't meaningful, or some combination of all of those.
Ii> Most research into animals has shown (or at least proved) only limited theory of mind.
I read somewhere that chimps can deliberately lie to each other -- specifically, give an alarm call in order to get another chimp to drop their bannana, and then steal it. Seems to be a theory of mind. Actually, some rudimentary theory of mind seems so useful that it would be surprising if animals didn't have it. Some rough ability to predict predator behavior would definitely be selected for.
Along the lines of group selection, it's interesting to think about distributed intelligence. Leaf-cutter ants evolved to create agriculture (growing and harvesting crops) hundreds of thousands of years before humans did. But they would make a lousy guest on a talk show.
Unfortunately, I don't have much time to participate right now, but listen to President Not Really an Expert!
Birds have language. I watched a documentary about turkeys last week where this guy figured out 20 different turkey noises that correspond to different meanings including "That bird in the sky won't hurt us, but be mindful" and "Oh shit, there's a rattlesnake right there!"
That's not language.
There is, of course, plenty of debate over this stuff, but to get a starting idea of what kinds of things might make language language, you can take a look at Charles Hockett's proposed design features of language.
A chimp would definitely have done a better job with the italics html then I did there.
Understanding the inevitability of death seems like a big species-independent threshold of intelligenceteen brooding.
(Hockett's work is old and much picked apart, but it gives a sense of the kinds of things you should be thinking about when you think "is this a language?")
Do I recall correctly that the human vocal apparatus is very unusual among primates, in that we can just make a whole lot more subtly distinct sounds than other primates can (which is why the effort is in teaching chimps to sign, not talk, because they physically can't talk)? I have no idea what this suggests evolution-wise about which developed first, the larynx or the mental capacity to use language, but it seems like the sort of issue that would be interesting to look into.
But they would make a lousy guest on a talk show.
In fairness, so would most farmers.
The other animals aren't smart enough to kill the earth as well as we do.
Animals, feh. Check out what the plants can get up to. Who needs smart when you've got brute numbers? Although very carefully working out how to dig them up and burn them is a peculiarly human thing to do.
Us parrots wouldn't be caught dead doing that!
50: People were really excited by that one for a while but as far as I know (which isn't terribly far; this is second-hand based on dismissive comments by people who should really know what they're talking about, but who also have a vested interest in it not being true) it hasn't panned out as a plausible causal mechanism.
52.link: Wow. I had never heard of that.
Descartes* assumed that humans have souls and free will, whereas animals are automatons ruled by instinct or whatever. This was not based on observation of either humans or animals, but was partly a philosophical assumption and partly a way of keeping the inquisition off his back. Since then his expedient compromise has caused endless trouble, primarily by formalizing the mind-body problem in a way that made it hard to solve, and to a lesser degree by promoting unrealistic assumptions about the stupidity of animals.
As time went on, various non-human species showed evidence of learning ability, and tool use, and cultural transmission, and something like language, all of which had once been regarded as unique to humans, and because of the original assumptionthese discoveries always caused more fuss than necessary. So animals are smarter than we assumed, but there still is a gap.
*I'm sure the story was more complicated. But the point is that the assumption of animal stupidity became orthodox, and that it did so without ever having been tested.
53: Sure, I can see it being perfectly plausible in the other direction too -- you get the symbolic intelligence from somewhere, and then there's suddenly a huge selection pressure in favor of the apes who can make subtler vocal distinctions, because they win all the arguments on Paleolithic blogs.
I do like the idea of an episode of Richard and Judy with the guests, an octopus, a chimp, and a colony of leaf-cutter ants.
42."Or something" is awesome. I want a subscription, as well as to the companion journal Something something New Yorker Something.
The newest individual animal getting attention for his 'language' skills is a Boarder Collie who knows the names of his many toys and can retrieve the appropriate toy when shown it in a picture. This doesn't sound super smart but recognizing a flat, two dimensional image and translating that to the real object is a form of intelligence (that maps well to what we humans consider intelligence - I'm sure that dogs think this is dumb).
I don't know how she's thought of in the animal behaviour field but Patricia McConnell write some interesting things on dog intelligence and emotion. (I don't know how to do links...http://www.theotherendoftheleash.com/yup-dogs-can-be-disgusted and http://www.theotherendoftheleash.com/could-you-learn-1022-new-nouns)
Sorry! http://www.theotherendoftheleash.com/yup-dogs-can-be-disgusted
There was a recent episode of Philosophy Bites on animal minds which dealt in detail with chimps' understanding of other minds. The guest said that chimps could definitely know when another chimp was ignorant of a fact, but we aren't sure whether they are actually able to attribute a false belief to another chimp, which is a bit more complicated.
Jack Russell Terriers are supposed to be smart too, but boy are they a pain in the ass. There are pros and cons on this animal intelligence thing. Just imagine if turkeys were intelligent.
But studying what kinds of things different animals are capable of and what kinds of mental capacities those capabilities would seem to indicate seems quite within the realm of science.
Sure, as long as you call it that. Getting all anthropomorphic about it and comparing those capabilities to a human attribute which isn't even understood to the extent that it has an agreed definition, less so.
Anthromorphic thinking is pretty double-edged, since people like the Churchlands don't want you to think anthromorphically about people either,
Do we even have any quantifiable way to talk seriously about this, or is the question itself nonsensical?
Last I knew, people are mostly trying to figure out whether there's a not nonsensical way to talk about it, much less quantify it -- indeed, what "it" is. The whole endeavor will necessarily struggle with anthropomorphization issues, e.g. language for humans is like this, something language-like for birds is like this, although we don't speak bird, of course, and we're defining pretty much everything relative to ourselves.
33: My take on this is that demonstrates the incredible visual intelligence of parrots. Alex had a detailed grasp of what was in his visual field, and was able to associate all of these visual details with phrases.
It's difficult to get away from the fact that (mere) sensory/perceptual differences between humans and various other species make for different kinds of intelligence. My favorite along these lines: a dog's sense of smell is to a human's sense of smell as the surface of the (human) brain is to the surface of an egg. We are not intelligent in the olfactory realm.
These are mostly just random thoughts. President NRAEBKSP has the real way of it.
Also, did you know that there is one invertebrate that has rights in British law? Octopus vulgaris, of course. Weirdly, it was our deeply horrible refugee-bashing, music-hating Tory home secretary Michael Howard who is responsible for this.
67. What rights are these? Can it set up an Academy for its spawn independent of the Local Education Authority?
Presumably the relevant distinction is between communication and language. Communication can be quite sophisticated without being equivalent to language. Not having read the aforelinked Hockett thing, I don't know whether it's a continuum or a true threshold.
I should handwave towards AI at this point, I believe.
65: I've met people who it would have been a real mistake to anthropromorphize. But I don't work in Big Law anymore.
69. They said the same thing about a bunch of positions endorsed by Barry Goldwater which are now mainstream.
Are Gingrich and Trump more or less intelligent than a New Caledonian Crow? Show your working.
My under-informed impression is that it's within the realm of possibility that Elephants understand that they're going to die. At any rate, Elephant response to death is somewhat ritualized
they use their beaks and wings in the same way that humans use their hands
Beaks maybe. But I don't believe it's possible for crows to use wings for masturbation.
a [...] is to [...] as [...] is to [...]
Parsimon is banned!
73: Experiment time! We'll need to phrase it carefully for the IRB.
Which species have the best mental whateverness?
77: it's between philosophers and lawyers, I believe.
I was thinking of trying to convince elephants that they were going to die. We need a corrupt mahout, a chainsaw, and a really strong rope.
The used to cull elephant herds by killing everyone over some age, now they know that you have to cull whole families because otherwise the kids grow up traumatized and maladjusted. For example, "Young elephants who saw their mothers being killed often wake up screaming."
The way I understand it, a lot of the traditional assumptions about which qualities or abilities truly set Man apart from The Beasts -- or at the very least differentiate higher primates from other animals -- have been seriously compromised over the past decade and a half:
- Tool use is a no go as a benchmark (crows and octopi use tools, let alone chimps);
- Cultures and socialization don't work, as these are now recognizable in a number of species;
- The presence of self-consciousness and metacognition (the "theory of mind") in a wide range of species is still controversial, but the trend toward its collapse as a salient of human or primate uniqueness is pretty clear;
- The ability to weigh options and make "rational" decisions was originally supposed to be a human quality but is present to some degree or other in a much wider range of species (there are hints that the collective cognition of ant colonies actually performs consistently better in this area than humans do);
- The contention that animals can't feel pain or suffering -- treated as credible by some philosophers as recently as the late Eighties -- is much more widely recognized now as a vestigial prejudice, and while it's not possible to quantitavely measure "suffering" as yet, it's widely evident that animals are subject to and react to psychological as was all physical distress;
- And the two granddaddies of all distinctions -- Language and War -- are on their way out, too: whale songs have hierachical syntax (theorized as far back as the Seventies but it can now be backed up with computer analysis), and theories about animal warfare are proliferating.
Maybe the emerging tendency to describe such-and-such animal as being "as smart as a 3-5 year old human" is related to the ongoing collapse of the ability to talk about humans as being unique in their "sapience"? It could be, I don't know, and in any case I don't know how meaningful it is. That we still have ongoing debates over whether IQ is a measure of intelligence makes it clear that our ability to scientifically assess even human intellect is still pretty primitive; until we've sorted that aspect of things out, it seems to me that comparing animal to human intelligence will be mostly meaningless. But the overall trend in biology seems to me to have been toward talking less about where animals fit on a great chain of being with humans at its apex, and more on how they're fitted for their particular niches and environments. I get the sense that researchers in animal cognition more and more have a tendency to look at it this way.
"Young elephants who saw their mothers being killed often wake up screaming."
Sometimes they engage in even more bizarre behaviors, like wearing tailored suits and learning arithmetic.
My Hmong students did not accept the Babar books. They carefully explained to us that elephants can't talk. They knew their elephants.
46: I understand that warning about the rattlesnake is too simplistic to be language, but stating that the bird is not an imminent danger but should still be watched seems like language to me. Based on the Hockett's design, it seems like the concept is missing traditional transmission, a duality of patterning and productivity. But this list looks like it was designed to distinguish humans from everyone else because we're superior and no one can be as wonderful as humans.
83: What the Babar books didn't specify was that all the elephant talk was subsonic.
84: Hockett's list of features looks in some ways ludicrously overspecific and frequently redundant, but has in common with other definitions of language the presence of the ability to speak abstractly and the presence of hierarchical syntax to organize statements (which he confusingly calls "duality of patterning"). These are the big factors that are generally supposed, I think, to distinguish language from food or warning calls or songs (although these kind of calls can still have fairly sophisticated degrees of what Hockett calls "semanticity").
Duality of patterning isn't syntactical structure; it that languages have phonemes which make words which make sentences.
81: But the overall trend in biology seems to me to have been toward talking less about where animals fit on a great chain of being with humans at its apex, and more on how they're fitted for their particular niches and environments. I get the sense that researchers in animal cognition more and more have a tendency to look at it this way.
I think this is right -- or hope so.
75: Parsimon is banned!
Nah, that was not a bannable-style analogy. The point of the surface-of-the-brain-to-suface-of-an-egg thing was to say not that a dog's sense of smell is just more better than a human's, but a whole 'nother realm entirely. I think I picked the analogy up from Vicki Hearne, animal trainer cum philosopher. Among her specializations was the training of scent dogs for rescue missions and police work and so on. While she was not herself a dog, she knew whereof she spoke, and I believe her when she says that we have no idea what a dog's scent intelligence must be.
Anyway, as Castock and numerous others upthread have observed, the question of how intelligent other animals are compared to humans is all but meaningless. At best it comes to an attempt to define human intelligence (and language use, etc.) -- there's nothing wrong with that project, but we wouldn't want to be confused about what we're trying to define.
The way I understand it, a lot of the traditional assumptions about which qualities or abilities truly set Man apart from The Beasts -- or at the very least differentiate higher primates from other animals -- have been seriously compromised over the past decade and a half
Cooking food? Being the being for whom its being is a matter of concern? Watching reality TV?
According to the Hockett link provided in 46, he's looking for features distinctive of human language and human language alone. It will be no surprise if no other language-like thing possessed by other species fails to meet the specifications. Saying of some other species' method of communication that "it's not language" doesn't get very far, then.
Being the being for whom its being is a matter of concern? Watching reality TV?
These are just different expressions of the same thing, right?
the question of how intelligent other animals are compared to humans is all but meaningless. At best it comes to an attempt to define human intelligence (and language use, etc.) -- there's nothing wrong with that project, but we wouldn't want to be confused about what we're trying to define.
The "all but meaningless" seems both wrong to me and pretty clearly contradicted by some of the research mentioned upthread. Sure, you have to define "intelligence" in a certain way -- ability to process language, some concept of general intelligence, whatever -- and, sure, dogs are better smellers than humans or whatever. Still, that there's a different kind of cognitive ability that humans have, including but not limited to language, that no other animals have seems pretty clear, and I don't see why we can't call that "intelligence" or do what we can to measure it and figure out how and in what quantity we differ in that regard from other animals.
Being the being for whom its being is a matter of concern?
I remember way back when I first studied Heidegger -- I was like 20? -- I asked my tutor slightly plaintively, "So is Dasein just human being? Or what?" He refused to answer. Annoying. (Also, I always translated it in my head as "for whom its being is an issue for it," which seems more poetic somehow.)
that there's a different kind of cognitive ability that humans have, including but not limited to language, that no other animals have seems pretty clear,
No, it is not clear. What seems increasingly clear is that humans do not have any particular "kind" of cognitive ability that no other animal possesses. What humans have is a cognitive and biological toolkit some ingredients of which are very sophisticated in comparison with many or most other animals' capacities in the same areas (although not necessarily as much more sophisticated as we'd like to think).
I don't know much, but I know 95 is definitely false, or at least really, really confused.
Has anyone quoted that Wittgenstein bit about the speaking lion yet? If not, well, that, and I think it likely that the smarter the animal the more it applies. Cf. recent speculations regarding whales' metaphysics.
41: Enthusiasm for apprenticeships reminds me of (i) Johnny Deformed Tremaine and (ii) the (probably apocryphal but too good to check) family story that the first Flippanter arrived in the New World an indentured servant and promptly skipped out on the deal. Mumble heritage of problems with authority mumble.
(95 is really just a follow-up to 81. Is there something in 81 that appears "false" or "really confused"?)
there's a different kind of cognitive ability that humans have, including but not limited to language, that no other animals have seems pretty clear
Not clear to me either. The differences appear to be in distribution and amounts, not in kind. And anecdotally, there seem to be vast differences in "smarts" in animals of the same species. I'm dealing with one very smart cat and one who relies on her looks. On the veldt....
103: "Almost all of it" what? False, and if so how? Confused, and if so how?
83: That Thai movie about the guy who uses martial arts to rescue a baby elephant from poachers is pretty awesome. He jumps across a room to knee a dude in the face!
Ability to process language? Which isn't to say that you're wrong, it's measuring precisely the degree of difference that prompted my original question.
As I've said many times, I am basically a blundering idiot in this area, but it strikes me that all three of the following propositions are pretty silly: (a) humans are completely cognitively distinct from animals, in some deep Cartesian sense, making human intelligence incommensurable with that of animals; (b) it is meaningless to measure "intelligence," since all animals can do different things well and who's to say what's intelligent and what's not; and (c)nonhuman animals have a capacity for language or something meaningfully similar to human language.
107 to 95. And I guess "ability to process language" should just be "ability to use language."
It's hard to know what "kind" means. Sure tapirs have a prehensile nose, but I think it strains credulity to say that Elephants trunks aren't unique and differentiating. Trunks are a lot of what makes elephants elephants. Similarly, language does play a large role in making humans humans, even if there are a few other species with vastly vastly inferior "versions" of the same "kind."
108: No, that was his first one. I think the elephant one is The Protector, at least in the U.S.
107.1: Language was the closest thing to a bulwark for a uniquely human form of cognition, but that was dependent on whale song being categorized as non-linguistic (where a true language is defined as the organization of phonemes into syntactic statements). But whale song does in fact have syntax, so I don't think even language holds up as our pure distinction from the animals anymore.
I don't think even language holds up as our pure distinction from the animals anymore.
What about iTunes?
81, 95: Seems to me that you're missing something when you can't find a difference in kind between the 'culture' of some chimp banging on a coconut and the species that went to the moon and invented reality TV. At a certain point it becomes a qualitative difference.
Go ahead, call me a speciest. I don't care.
Go ahead, call me a speciest. I don't care.
Classist.
117: Any species that invents the Kardashians and puts them on TV *needs* extinction.
The Kardashians weren't invented, they were discovered.
From the link in 115: the authors do not claim that humpback whale songs meet the linguistic rigor necessary for a true language
117: Seems to me that you're missing something when you can't find a difference in kind
Definitional confusion. The existence of cultures is a reference to a specific kind of phenomenon: whether a species' adaptations to its environment are purely instinctive or are mediated by some form of localized group dynamic, the kind of complex of traditions and contingent individual and community preferences that might be called a "culture.'' This was at one time supposed to be a uniquely human phenomenon, and then (as primatology advanced) a phenomenon found only in higher primates, and now has broadened further. This obviously has nothing to do with denying that humans went to the moon and invented reality TV, dubious an achievement as the latter might be.
120: Oh? I thought they were the output from a secret artificial stupidity project.
If people want to find a difference in kind between humans and other species, and I'm sure we can, eventually, or at least we'll keep trying, that's fine. Calling that a difference in intelligence tout court is a different matter: it's important to be clear that what's being defined is human intelligence.
But I've said this about five times now.
Whalesong has hierarchical structure. Until someone understands the whalesong, whether or not there's syntax is unknown.
Here's the paper:
http://asadl.org/jasa/resource/1/jasman/v119/i3/p1849_s1
A problem with 117 is that many human cultures have very primitive technology. The species with the most sophisticated communication (some birds, parrots and ravens that I know of, and whales) cannot make even simple tools.
121: The article's author is of course using "true language" to mean "human language" in circular fashion, as I suspect you're also doing. When the authors are actually quoted, what they say is that whale song is not an exact equivalent of human language, not that it is not "true language."
But I've said this about five times now.
How many times would a whale have had to say it?
124. Congratulations on your persistence. I gave up after three.
it's important to be clear that what's being defined is human intelligence.
This seems true but meaningless. Sure, a fly is better at flying than I am. So what? What seems open to measurement is the degree of cognitive capacity that allows us to do things, like language and abstract thought, that we seem to do well but other animals can't do at all or only extremely poorly in relative terms.
Incidentally, I notice that talking about animal cognition seems to bring out a familiar sort of anxiety in people. It's familiar from the history of evolutionary theory and Darwinism. When it first made it into the popular press that evolution placed humans on a continuum with other animals, rather than setting humanity above the animials as absolutely unique, there was born a fallacious but insistent meme -- still with us today in the form of Creationism and Intelligent Design -- that "Darwinism" was preniciously trying to claim there was no difference between people and "monkeys." Most, indeed I think almost all, Unfoggedtarians would recognize the ridiculousness of this meme now. And yet, when animal cognition comes up...
129: Halford, yeah, we can engage in human style language-use and abstract thought, and other species don't, so much. If all you want to assess is how human-like other species are, go for it.
as I suspect you're also doing
I'm not, no.
131 -- well, that seems like the question we're answering if the question is about animal intelligence, unless you want to just say that whatever an animal has evolved to do functionally well is its "intelligence."
132: Okay, then I'm not sure why you didn't know better than to cite the second-hand reference to "true language" at me as though it were supposed to be meaningful, instead of referring to the substance of what Suzuki et al actually say.
133: unless you want to just say that whatever an animal has evolved to do functionally well is its "intelligence."
I'm sorely tempted to say that. But I actually have to be off for a while, and I've been getting a bit irritated anyway.
well, that seems like the question we're answering if the question is about animal intelligence, unless you want to just say that whatever an animal has evolved to do functionally well is its "intelligence."
So your definition of "intelligence", against which animal intelligence must be measured, is human intelligence? Why? But there is no agreement on how to define human intelligence, so you are asking to define animal intelligence by analogy to an undefined standard. how does that work?
133: unless you want to just say that whatever an animal has evolved to do functionally well is its "intelligence."
You would presumably want to talk about an animal's cognitive tools in relation to the process of its evolution, and the problems those tools evolved to solve. Right? This is done with humans -- for example there are common theories relating our well-developed talents for patterns recognition and our poorly-developed talents for probability assessment to our remote ancestors' hunter-scavenger-gatherer pasts on The Veldt -- so why wouldn't you want to do it with other animals?
how does that work?
Well, that was the point of the question. And there do seem to be answers that are more than just philosophers throwing their hands up in the air, even if definitions of intelligence are contested, as indicated by 5 et al. They could be wrong for all I know.
134: sorry, no, I meant that I wasn't being circular in my definition.
||
Hey, Bitch is going to be blogging at Crooked Timber. (Under that odd new moniker she goes by, Tedra.)
|>
You would presumably want to talk about an animal's cognitive tools in relation to the process of its evolution, and the problems those tools evolved to solve. Right? This is done with humans -- for example there are common theories relating our well-developed talents for patterns recognition and our poorly-developed talents for probability assessment to our remote ancestors' hunter-scavenger-gatherer pasts on The Veldt
Of course, when "this" is "done with humans," the results are often trash. It's very hard to cash out either the relationships between behaviors and capacities or the relationships between either one and "problems" they evolved "to solve." The relationship between selection pressures and the traits they select for isn't really one of problems and solutions to those problems, no matter how tempting it is to try to cast it in those terms.
(I'm afraid my participation in this thread is going to be characterized by a scattered, widely spaced sequence of shallow, drive-by comments. Sorry.)
140: Happy as I am to see B blogging again, the combination of her writing and the kind of cranks CT tends to attract makes me leery.
I wonder if they've brought her in partially as a lightning rod/hockey enforcer. I spent some time bickering over at CT in the last two days on a post about the iPhone/Siri abortion-search fuckup -- the usual CT clowns were being jerks about it, and Hargittai doesn't generally (and didn't this time) get into it with commenters. Tedra might have been brought in partially to raise the tendency of the bloggers addressing gender issues to kick ass and take names; Belle can't be the only pugnacious woman on the site, not with their commentariat.
141: Of course, when "this" is "done with humans," the results are often trash.
A fair point.
5: Some people think that language underlies human general intelligence -- see here -- but a lot of people don't.
I'm confused by this as a description of the argument of The Cultural Origins of Human Cognition. I remember it arguing that theory of mind and it's permission of certain kinds of information transmission -- not necessarily linguistic -- that separates humans from animals. Specifically, information transmission that understands X did Y with the intention of achieving goal Z, and he thinks that language is an outgrowth of this capacity. Possible I misremember or misunderstand something though.
6: Most research into animals has shown (or at least proved) only limited theory of mind.
This may be true (not sure), but I don't think it's true of the state of the literature on great apes in particular. Tomasello himself has abandoned the "theory of mind is what's uniquely human" argument because it's not well-supported (so TCOoHC is unlikely to be a great review of the current consensus). Given that that means reversing himself on the subject of a very well-regarded book he wrote, I think this is an indication that there's overwhelming evidence for great ape theory of mind at this point. He'd moved to a "humans are more cooperative than chimps" argument in the last paper I read. No? Yes? Anyone very well-informed say different?
I certainly recall seeing the "humans are more cooperative than chimps" argument all over the place lately.
146.1: No, that's right, I think. I was being a bit handwavey there (in part because it's a cool book). I was really thinking of the work of Liz Spelke, who talks about things in terms of Tomasello's work but turns it to her own purposes, somewhat. (And the ideas of Spelke's that I was talking about are already her handwavier ones.)
140:Just as I'm getting banned. Damn.
Also, 120 really made me laugh, and it seems no one has yet given it love. Let me be the first!
He'd moved to a "humans are more cooperative than chimps" argument in the last paper I read.
Tomasello has indeed revised his position, but it's a revision of the original claim rather than a reversal. The original claim was not exactly that "theory of mind" was uniquely human and the underpinning of our symbolic competence (and, incidentally, I want to point out that there is a lot of disagreement about whether or not the ability to understand other people as having mental states that differ from our own is properly understood in terms of a "theory of mind"). Instead he argued that the central thing was the ability to understand oneself as part of a joint attentional scene, in which two or more people are mutually aware that they are intentionally sharing attention to some external thing, often in the course of pursuing some shared goal involving that external thing. He also argued that full-blown joint attention involved the ability to recognize other people as intentional agents like ourselves.
Since then he has shifted to arguing that shared attention is not as special or crucial as he originally thought, and that the way humans understand, share, and coordinate intentions might be where the action really is. The cooperation line of investigation follows out from there.
So, yes, the current consensus on great ape social cognition is that they've definitely got some. But it really is not the case that "there's overwhelming evidence for great ape theory of mind at this point," in part because the whole notion of "theory of mind" is a contentious one.
Specifically, information transmission that understands X did Y with the intention of achieving goal Z, and he thinks that language is an outgrowth of this capacity. Possible I misremember or misunderstand something though.
And yes, this is exactly correct.
The cooperation line of investigation follows out from there.
By this I mean that the cooperation stuff is his current research program aimed at trying to pin down what parts of intention understanding and coordination seem to be truly unique to humans. This current line does definitely involve moving away from a bunch of his earlier claims (and with good reason), but it's not a complete sea change.
All these interesting threads and I have to spend the next five hours in planes and airports. Not that this distinguishes me from birds, of course.
Thanks Tia!
The description of Tomasello in 151 makes me wonder if anyone's written about him and Davidson together.
whether or not the ability to understand other people as having mental states that differ from our own is properly understood in terms of a "theory of mind"
rfts, how would you define "theory of mind"? I would have said that the first part of phrase defined the term. (Although if pressed I would elaborate that there are lots of subcomponents. I went to talk by this guy once and he enumerated five, but I can't remember what they were. It was in the context of a very interesting argument/line of research purporting that which subcomponents were achieved by children earliest during development depended on cultural values about which aspects of understanding of persons were most important.)
Anybody know what the current state of play is on the Piraha (a Brazalian tribe) and Daniel Everett's contention that their grammar lacks embedding or some such recursive feature of language once thought to be universal among humans?
I vaguely remember seeing a claim that it had been debunked, but I don't remember it either well enough to find it or to evaluate how solid it was.
Wtf is this Christmas-themed version of Toto's "Africa"? Why is Logan Airport subjecting me to it? Will I miss my connecting flight due to this delay? I can only imagine that whales have erudite conversations on similar matters as they glide through the ocean depths.
Here's a link to a paper claiming to have debunked it, which I haven't read.
155: Right, well, a couple of things -- one is that I'm reluctant to use "theory of mind" as the general term covering something you might alternately call "concept of mind" or "social cognition," because it brings along a lot of theoretical baggage about what that understanding of other minds consists of. There's the notion that it is some kind of coherent "theory" of other minds, where the most explicit, fully articulated, conscious representations of other minds is the capacity in question.
There's the the related notion that whatever it is is a modular capacity that you can simply "have" or "lack". As you already suggested with your discussion of subcomponents, the thing often referred to as "theory of mind" doesn't seem necessarily to be a single entity. Different kinds of social-cognitive competences seem to come online at different points in development, or be more easily accessible under some circumstances than not others. (For example, chimps do a lot better at implicit false-belief tasks when both food and hierarchical dominance relations are immediately involved in the task.)
So, sure, it's very common for people to use "theory of mind" to mean "the ability* to understand other people as having mental states that differ from our own," but then the waters get pretty muddied when you start to talk about whether all great apes have "theory of mind," because the phrasing threatens to conflate all kinds of activities, from the most contingent and implicit to the most explicit and fully articulated, that seem to take into account any aspect of another agent's perspective.
*and obviously, from what I've said here, I should be more punctilious about saying anything like "the ability," either.
There's the notion that it is some kind of coherent "theory" of other minds, where the most explicit, fully articulated, conscious representations of other minds is the capacity in question.
And, of course, whether people actually use anything similar to this on a day-to-day basis is its own question, with some people arguing fairly strongly that we do (and that this function is linked to episodic memory, which would mean that animals who lacked this would have a very different understanding of their own past and future than humans do), where other people argue that most of the time people are just using heuristics about how somebody's likely to act, based on past experience.
Fuckola. Any posters want to anonymize that for me?
OT: Garrett Morris is still working. Now I feel young.
Maybe the emerging tendency to describe such-and-such animal as being "as smart as a 3-5 year old human" is related to the ongoing collapse of the ability to talk about humans as being unique in their "sapience"? It could be, I don't know, and in any case I don't know how meaningful it is.
Having had 3 dogs and 2 kids, I fully endorse that trope. For the record. No question that, at some point between 2 and 4, the kids and the dog swapped slots in the hierarchy.
On an unrelated note, when AB went to fetch Kai from his nap this evening, he was asleep under his rug. Snug as the proverbial bug. Weirdo.
I'm curious if this discussion could be linked up with previous remarks here about the Flynn Effect and Cosma's review of Flynn's recent book.
As I understand it, Flynn's explanation of the Flynn effect is not that people are getting smarter, but that they are moving from a concrete style of thought to an abstract one favored by industrialized, Western societies. This change can be describe alternately as "cultural imperialism" or "liberation from the concrete" depending on whether you are a fan of abstract thinking.
I think the honorific kind of "intelligence" we question whether animals have is related to the kind of "intelligence" we try to measure with IQ tests in a number of ways. Politically speaking, both were set up to exclude. More importantly, both wind up tracking a kind of abstraction. Not just problem solving ability, but a kind of problem solving that translates more easily across domains and subject matters.
Now I know that to compare differences in cognition among humans to differences between humans and animals is to invite charges of racism. This is because any comparison between humans and animals is taken as an insult to humans and not a complement to animals. If I say benighted humans of the year 1900, with their average IQ of 70, thought in a more ape like fashion, everyone will assume I'm on a slippery slope to calling black people monkeys who should go back to Africa.
That's not where I'm coming from at all (although M/tch can go ahead and call me a racist.) Mostly I just want to know more about what this "abstract thinking" is that people seem to be pushing toward here. I am particularly wondering what is lost when we become a planet of abstract thinkers.
166: If the PRP hypothesis is correct (that is, that the PRP corresponds to some kind of serial function that marshals other kinds of thinking into a symbolic framework), it would correspond fairly exactly to the kind of abstract style of thought that is useful for IQ tests, I think.
I am particularly wondering what is lost when we become a planet of abstract thinkers.
Poetry. The wisdom of the East.
Not sure I buy this description, sounds like something self-serving for (yeccch) scientists. That "derivative" is more "abstract" than "daffodil." Is this picture-theory of language or something? I don't know how to measure "abstraction."
Damn good question, though.
No question that, at some point between 2 and 4, the kids and the dog swapped slots in the hierarchy.
Social promotion.
This tumblr is the best thing I've seen on the internet all week, present company (mumble, mumble).
the kind of abstract style of thought that is useful for IQ tests, I think.
Question: when I took an IQ test, in 2nd grade, it was all pattern recognition - red and white triangles and tangrams and such. Then I got older and found out that some IQ tests include shit like vocabulary and other things that are obviously based on teaching, not aptitude. Does this still happen? Does this have any relevance to the broader IQ debate? Or is ability to identify patterns and solve simple, abstract problems exactly as fraught as knowing that "ennui" comes from the French word for boredom?
I get that any cognitive skill will be enhanced/degraded according to stimulus/opportunity/SWPL, but it seems to me that the 2 kinds of IQ test are measuring completely different things.
167: If the PRB hypothesis is just about the serial ordering of mental functions, does that mean that this New Caledonian crow is exhibiting The Big Important Mental Capacity That Makes Us Human, albeit to a lesser degree?
I was disappointed to learn that none of the crows I see belong to the exalted New Caledonian species, which is actually limited to some south Pacific islands controlled by the French. (Are they a product of some island biogeographical effect?) A part of me wants to release a bunch of them in North America to see if they can take over.
The same part of me wants Kanzi to go to the Congo and teach the other bonobos how to use fire and make stone tools, just to give them more of a fighting chance.
171: Sounds like you took the Raven which is assumed to be less culturally loaded because it isn't linguistic, although that is crap if you think about it for a second. Scores on the Raven do correlate with other IQ tests.
I think the effect is reversed from what you'd think. The highly abstracted tests are where the Flynn effect shows up like gangbusters, and that has to be about education and environment somehow.
And to get anecdotal, if you've ever played with something that's like those kinds of abstract IQ test questions, you know how learnable they are. Any puzzle like that is hard until you get used to it and get fluent with it.
less culturally loaded because it isn't linguistic, although that is crap if you think about it for a second
I don't know about "crap". Vocabulary is roughly 200% cultural, whereas pattern recognition, while culturally bound, has independent existence.
Put it this way: give me a vocab-based IQ test in any non-Euro language, and I score at goldfish levels. I'm sure that a smart Japanese kid - or a Javanese - could do a tangram puzzle* even if she hasn't grown up with that sort of game.
I might add that, at some point, ability to succeed on even a culturally-bound IQ test is going to map onto ability to succeed in general. It's not as if all kids with shitty circumstances score identically on IQ tests. I guess this is a banned analogy, but it's a bit like adjusting batting stats for ballpark factors: every hitter is at a disadvantage at the Padres home field, but better hitters still do better than worse ones.
* that is, Take these shaped pieces and assemble them to match the shape in the picture. And yes, it's testing for a particular intelligence (visual/spatial), but not an especially cultural one. Geometric ornament is super-duper not exclusive to the West, or the UMC, or SWPL.
whale songs have hierachical syntax
What human languages have is not just syntax but syntax in the service of semantics - we can use the syntactic structures available in our language to construct sentences that express propositions. That's the usual story, anyway. I don't imagine that there is any evidence yet that whale-songs have propositional content to which their syntactic structure is relevant.
175: I guess. I honestly don't think anyone ever gave me red & white triangles to play with as a kid. I had a plastic Pacer and wooden blocks with letters and ballpoint scribbles from my sister.
And again: no amount of cognition will tell you what "hyperborean" means, unless you've been exposed to lots of Greek-derived words. Geometric patterns are comprehensible (Kai still doesn't get numbers, but he gets shapes) and extrapolate-able.
To exactly match human language, an animal language would have to deploy syntax in the service of semantics as a part of having sex over the phone.
But this just brings out the fact that human language ability, rather than being a single unique thing, is a combination of abilities that exist in animals to various degrees. Vervet monkeys have some semantics. Humpback whales have syntax. So far as we know though, no other animals come close to us in the sex-over-the-phone arena.
I'll cop to this: I may be misunderstanding what the vocal-based IQ tests are like. My assumptions are (roughly) in terms of SAT-like tests, which are as culturally bound as a test on which side the knife and fork go on.
no other animals come close to us in the sex-over-the-phone arena.
Presumably because most of them can lick themselves.
If the PRB hypothesis is just about the serial ordering of mental functions
It's not. If I gave that impression I was oversimplifying.
176:
Vocabulary is roughly 200% cultural, whereas pattern recognition, while culturally bound, has independent existence.
As far as non-linguistic tests being less culturally contingent: they often are, but you could easily imagine creating a purely visual test (say a set of Rebuses) that relied on understanding of English grammar. Similarly, you could imagine that Raven's matrices (or whatever), despite being purely visual, reflect the biases of their creators. Also, tests like Raven's are piloted and normed against populations of a certain culture; they're going to select a set of problems for Raven's such that a population of literate, relatively educated westerners produce a good distribution on it. It makes sense, given that, that it would correlate well with more culturally contingent (verbal ability-based) measures of intelligence.
at some point, ability to succeed on even a culturally-bound IQ test is going to map onto ability to succeed in general
Exactly this idea was behind the development of the idea of "g", which is basically an attempt to take all of the psychological tests you might be able to give a person and figure out how much of that test is due to "ballpark factors" (which is to say cognitive functions that aren't general intelligence) and which are just due to general intelligence (g). This approach has a number of problems which are somewhat statistically involved, and have me really wishing Cosma would show up and talk about them, because I'm not sure I could do them justice.
Vervet monkeys have some semantics. Humpback whales have syntax.
Yes, but it's the combination that's connected to the human ability to express arbitrarily many novel propositions, which, you have to admit, is a pretty rad ability to have when you're having sex over the phone.
Then I got older and found out that some IQ tests include shit like vocabulary and other things that are obviously based on teaching, not aptitude.
I remember doing an IQ test in middle school in which there was a section where you had to put several picture cards in order of "the story." I could tell by the reactions of the test giver that I kept doing it "wrong," and I had a pretty good idea what the "right" order would be, but I stubbornly liked my narratives better.
182.2: makes plenty of sense, esp. wrt how the test is normed. No matter what the facts of the test, if you define a good score as the kind of score a smart UMC American gets, then that's what you'll reward. As 184 kind of reflects.
182.3: interesting, but it seems like a situation where every time you add an equation, you also add another variable, such that you can never solve for "g". It's not like you can subtract the spatial biases of the Raven from the linguistic biases of whatever in order to isolate "g".
37: From the dogs?! That is some verbal ability.
"I guess I shouldn't have said DiMaggio."
166: Remember that even those people sunk furthest in the deep idiocy of rural life are cultural and linguistic animals, in ways which (so far as we can tell) no other species comes close to being. I see that people are already plugging Tomasello, so I will just add that I am very much looking forward to this talk.
On the subject of animal cognition, I strongly recommend Shettleworth's Cognition, Evolution, and Behavior, but it's a big fat textbook.
182: For "somewhat statistically involved", read "ought to be fatal". (186.2's "adding an unknown with each equation" problem is real and has been known since at least 1928, see note 2 at that link.)
184, 186.1: I gather that when I take a test like that, one of the answer-pictures is supposed to feel right, the way an analogy-completion problem does; but none of them ever do. I could only solve them by trying to guess which answer the test-maker wanted, which gets into cultural ideas about symmetry and so forth.
151: Since then he has shifted to arguing that shared attention is not as special or crucial as he originally thought, and that the way humans understand, share, and coordinate intentions might be where the action really is . . . it really is not the case that "there's overwhelming evidence for great ape theory of mind at this point," in part because the whole notion of "theory of mind" is a contentious one.
To certain laymanpersons, this sort of thing really is reminiscent of a scientific (or perhaps scientistic) variant of the "God of the gaps" phenomenon, in that it smacks of an attempt to force the evidence fit a foreordained conclusion rather than an attempt to follow the evidence wherever it leads. If the parameters of "theory of mind," previously acceptable as a rough working terminology for things like self-awareness and metacognition, proves to include other species than the great apes, then of course it must be the case that the uniqueness of the primates must be sought elsewhere, and then elsewhere again, and then elsewhere again.
If the parameters of "theory of mind," previously acceptable as a rough working terminology for things like self-awareness and metacognition
That's not at all a good characterization of what theory of mind is supposed to consist of, even among the strongest proponents of a unitary Theory of Mind module. Theories about "Theory of Mind" are talking about something that's a descendent of the notion of a "folk psychology" -- that is, the ideas you have about how other people's minds work.
Some things complicating/confusing the issue:
- in practice, do the ways you take other people's thoughts and actions into account actually tend to involve recourse to your explicit model of how their minds works?
- to what extent is our ability to understand that other people have mental states like our own, but often different from our own, actually a single ability?
- to what extent *do* we understand that?
- maybe other things we do, which might not immediately seem like they're all about "theory of mind", rely on the same stuff that lets us take other people's thoughts and plans into account!
- does that mean that "having theory of mind" is what lets us do that stuff?
- or maybe it means that "having theory of mind" and doing various other things, like symbolic reference, both depend on some third thing.
- et cetera.
190: that is, the ideas you have about how other people's minds work
I thought this was what "metacognition" meant. Am I wrong?
(And I don't doubt that there are things complicating the issue, and do appreciate your itemization of some possibilities. I'm just struck by the way in which some outlooks seem determined to arrive at something like their original assumption regardless of the lay of that territory.)
There's various ways to actually test aspects of having a theory of the mind. You do stuff like you have a dog watching another dog hide a bone, then watch someone else move that bone, then have the first dog come back and either 1) go to the original hiding spot to dig or 2) go to the new hiding spot to dig, and you see if the observing animal expresses surprise/confusion or just looks longer in one of those cases. You can do these tests with babies and they switch at some age to knowing that other people don't know everything that they themselves know. Presumably people are actually doing work on this stuff, but it's not really a deep mystery.
191: Metacognition usually refers to cognition about one's own thoughts. "Theory of mind" is supposed to be something very different.
I thought this was what "metacognition" meant. Am I wrong?
Well, you would have to be able to engage in metacognition to have an explicit folk psychology. But the folk psychology itself would be a set of specific notions of how other people's minds work. Metacognition is thinking about or referring to thoughts/thinking. It's certainly related to the concept of "theory of mind," and that term is, as I mentioned earlier, often and sloppily used to refer to a bunch of different things.
But there are lots of problems (some of which I gesture towards in my messy list above) with the narrative that we used to have a good healthy "rough working terminology for things like self-awareness and metacognition" and that we're only pulling it apart now because it turns out that other creatures are self aware or can form explicit representations of thoughts.
Indeed, one thing that's going on with cross-species research--and also with early childhood research--is that people are designing more studies that illuminate behaviors that don't seem to reflect anything that rises to the level of old, unitary criteria like "self-awareness" or "metacognition," strictly understood, but do seem to show some kind of interesting capacity on that continuum. In other words, a lot of this gap finding arises from recognizing the need to make space for the fact that non-human, non-adult, or non-neurotypical cognition doesn't just "have" or "not have" the magic of theory of mind.
And at the same there is (and maybe you just don't buy that there is--but you're going to have to work really hard to convince me that there isn't, because no study I've read actually demonstrates otherwise) a very large gap between neurotypical adult human behavior and what animals, even other primates, seem to do. Which is intriguing!
196: And at the same [time] there is . . . a very large gap between neurotypical adult human behavior and what animals, even other primates, seem to do.
Of course I buy that there is. I'm just more and more convinced -- in my half-arsed leyman's fashion, of course -- that this is an emergent result from the varyingly developed elements of a toolkit and isn't located in some singular cognitive Thing that humans do and which other animals absolutely do not do.
because it turns out that other creatures are self aware or can form explicit representations of thoughts
I'm afraid my rhetorical structure in 190 got a little obscure. The third paragraph is supposed to make it clear that it doesn't, so far, turn out anything so exciting.
God, I write the most convoluted fucking sentences. Syntax!
I am very much looking forward to this talk.
Ooh, let's make it a date!
I'm not kidding, btw.
Cosma's g post is such an enduring classic. (Mind you it also falls into a category of blog posts that mean I will never read such-and-such a book because I now know that despite its reputation it's hopelessly, maddeningly wrong.)
200: That would make me very happy.
I'm going to put that Tomasello book on my x-mas list.
By the way, I wanted to thank everyone who advised us to get a new shower head after I complained about the shower in my house. No advice from this blog has improved my life more since you guys pointed me to Janelle Monáe.
If the parameters of "theory of mind," previously acceptable as a rough working terminology for things like self-awareness and metacognition
Some of the mirror neuron people hypothesize that self-consciousness is basically the theory of mind applied back to the agent him/herself. Michael Graziano's "God Soul Mind Brain" is a great introduction to this idea.
Is there any experimental evidence for mirror neurons in humans yet?
...because the last time I looked in on this, it wasn't looking good.
206: It seems to be a matter of some ongoing controversy. (I haven't read either paper yet, and am struggling with a strong and almost certainly unfair instinct to dismiss Caramazza and believe Friston a priori.)
I read somewhere that chimps can deliberately lie to each other -- specifically, give an alarm call in order to get another chimp to drop their bannana, and then steal it. Seems to be a theory of mind.
Yeah, this was some of the positive evidence I had in mind. There's also evidence of deception when it comes to fertility displays in some ape species. Designing studies that entirely rule out a behaviourist explanation can be tough though, especially as wild and captive behaviour can vary significantly.
Indeed, one thing that's going on with cross-species research--and also with early childhood research--is that people are designing more studies that illuminate behaviors that don't seem to reflect anything that rises to the level of old, unitary criteria like "self-awareness" or "metacognition," strictly understood, but do seem to show some kind of interesting capacity on that continuum. In other words, a lot of this gap finding arises from recognizing the need to make space for the fact that non-human, non-adult, or non-neurotypical cognition doesn't just "have" or "not have" the magic of theory of mind.
Having introduced theory of mind to the thread and then abandoned it for 200 posts, I should point out that I'm not arguing for a unitary/binary "theory of mind" capacity in animals or humans. I was simply highlighting it as one non-linguistic thing that was historically used to distinguish animal intelligence from humans and that recent research is showing to be a complicated picture.
Great thread, BTW.
||
This seems like an apt thread to mention that I have officially survived discussing my research in public. I mean, semi-public (it was just the larger lab; about 25 people or so), but they ask some tough questions, geez. I now feel prepared to... possibly present the same material in front of a slightly larger audience someday.
|>
209: Let me be the first to offer you congratulations!
Let me be the first to suggest Fresh Salt offer you congratudations.
Let me be the last to offer you congratulations. After this comment, no one should offer Sifu any congratulations.
202: On my calendar. And it's a Friday (comes between Thursday and Saturday), so we should go get blitzed somewhere afterwards.
215: Deal. If only there were a good bar somewhere nearby...
Are the two portions of the evening a fixed package?
217: I see no reason why they should be. I presume you want to hear about the evolution of human intelligence without setting yourself up for drunken, brain-damaging debauchery?
A close reading of Moby's commenting history could lead to no other possible conclusion.
drunken, brain-damaging debauchery
Does this mean that Cosma will be reduced to having only 1000% of the brain of a normal human?
Damn, I wish I could be there. For the talk, I mean.
If my brain gets damaged, I know where the best MRI in town is.
I bet he has the address sewn into all his underwear.
209: Fun, wasn't it? Congratulations!
225: it... kind of was. I recorded the whole thing, as I suspected the altered state that went along with the logorrhea would leave me unable to remember any questions. I tried to listen to it later this afternoon, and oh my god I hated myself. Bullshitting little dickweed.
226: If it's any consolation, that's why I can't listen to any recordings of myself lecturing or giving talks.
Bullshitting little dickweed.
That's also what I sound like to myself.
223: Walk across Oakland and your watch will get pulled the right way.
227, 228: no, I mean, I think I'm fairly aware that this means I did at least an okay job. The sort of vaguely folksy "let me explain this complicated subject in which I have a great deal of expertise" tone works gangbusters on me when somebody else deploys it in a talk. I just know how little I actually know, which makes it appalling.
230: You're only discovering impostor syndrome now?
231: I'm sort of a late bloomer.
209: Yay! The ones where they ask tough questions actually tend to be the best, in my experience. Talks where people sit there silently lead me to start thinking "oh no! they're all really bored! I'm being really boring!" and then I start stumbling.
Having just had a conversation this afternoon about how looking at Cosma's CV yesterday had me feeling like an utter fraud, I am astonished by 231. Oddly comforted, though.
Congratulations on the talk, Sifu, and especially on having found a lab that asks good hard questions in lab meeting. That's indispensable, IME.
Re Tomasello talk and the proposed uniquely human propositional-conceptual-reflective cognition-related drinking activities. I'm pretending that I will take off work early and attend, although I've been so busy being a real boy at work that I've been unable to read much of this or any other thread.
Well, I have the opposite of imposter syndrome.
235: Surely you'll have had enough of being a good boy by late January.
