Mark Changizi is an evolutionary neurobiologist and director of human cognition at 2AI Labs. He is the author of The Brain from 25000 Feet, The Vision Revolution, and his newest book, Harnessed: How Language and Music Mimicked Nature and Transformed Ape to Man.”
I’m king of the world! You are too. We humans—all of us—get props for being the smartest Earthlings. And we’re not merely the smartest. No, we’re the only species worth writing home about; we’re the only truly worth building artificial intelligence to mimic. We’re the smart ones. The rest of the diversity of life may be rich in clever design, like well-engineered tools and gadgets, but they’re not designed to be intelligent. That’s for humans. Rationality and intelligence is something natural selection granted us.
But…what if our Homo sapiens intelligence is radically overrated? What if we’re smarter, but only quantitatively so, not qualitatively? What if many of our Earthly cousins are respectably intelligent after all? More intriguingly, what if there are systematic barriers that lead us to overestimate our true level of intelligence relative to that of others? And, although I won’t get into this here, what are the implications for the rights of chimpanzees, if the chasm between us and them is, instead, a slender fault line? That question has led to a recent movement to ban invasive research on chimpanzees in the U.S., a measure that the EU has already adopted.
Here I’ll discuss just two barriers, a little one and a big one, that conceal how smart we really are—or are not.
Individuality: The Little Bubble
The eyes, they say, are the window to the soul. Look at someone—look into their eyes—and you can’t fail to see the intelligent, sentient, emotionally alive creature inside. And if that person has instead, say, finished your entire bottle of Two-Buck Chuck wine, you can see that too. Not only can we often read what’s on their mind, but we can see his or her individuality—the uniqueness in the face, in the way it expresses, in the vocalizations it utters.
When we look into the eyes of other animals, on the other hand, we often get relatively little back, and often nothing at all (sometimes worse). Looking at a waddle of penguins, our sense is of hundreds of automata with roughly interchangeable personalities, and thus no personalities at all. We do sense unique personalities and inner lives in our nearest primate relatives and also in our pet dogs, but even in these cases the sense is much attenuated.
Of course, anyone who fancies him or herself a naturalist realizes that these failures in “life sensing” in other species is an illusion. When we look into the eyes of a person we’re not seeing into his or her brain, much less the soul (because the soul is invisible). Instead, we’re sensing a whole variety of cues, including eye gaze, pupil dilation, facial expressions and skin color, and these cues to what’s going on inside tend to be disproportionately informative near the eyes. We can “read” another person’s eyes because we evolved to have software for doing that. And that software doesn’t tend to work well, or at all, on other animals because their faces are very different, in the facial expressions they possess, in the pupil modulations they evince, in the shape of the eyes and pupil, in the placement of the eyes on the head (most animals have eyes on the sides of their heads rather than the front), and in the degree to which the face is naked and undergoes color changes.
Even though one may be cognitively aware that a female penguin senses in her mate a bounty of life you and I cannot sense, our innate handicap at “life sensing” in other animals must continually be fought against. Even among chimpanzees, although to a lesser extent than for penguins. If other animals often seem dead inside, it is because our human eyes are dead (unresponsive) to the life in them. It is not by any means a sign that we’re disproportionately intelligent.
Language and Art: The Big Bubble
The greatest sign that Homo sapiens intelligence is far beyond that of any other animal is our language capability and unique propensity for the arts. If we evolved instincts for language and the arts then that would indeed amount to revolutionary steps in intelligence beyond that of our cousins.
But what if language and the arts don’t require qualitative leaps in brain software? What if all the software for language and the arts was inside us long before we ever spoke or sang?
For example, what if, as I argue in my book, Harnessed, spoken language got structured via cultural evolution to sound like the events occurring among solid objects, the principal sort of event occurring in natural terrestrial habitats, and the sort of event that ape brains would have already evolved to ably process? Special speech processing mechanisms would then not be needed, because speech would have gotten rigged to harness the solid-object auditory processing mechanisms we already possessed. And what if the semantic structure of our language vocabulary closely matched the structure of the internal mental lexicon apes like us possessed long before language? Special vocabulary-storage mechanisms would also not be needed.
And, for the arts, consider music. What if, as I also argue in Harnessed, our musical proclivity is not due to having evolved to process that strange beat-and-tone-filled class of auditory stimulus, but, instead, is due to music having itself culturally evolved to possess the signature structure found among the sounds of humans evocatively carrying out behavior in one’s midst, another class of sound our brain already had evolved to process? No special music processing capability would then be needed.
This sort of hypothesis is radical for spoken language and music, but it is much less so for writing, where we know we cannot possibly have evolved a reading instinct, writing being much too recent an invention. Yet even writing has most of the hallmarks of instinct, and it does so because writing has culturally evolved over centuries to be shaped like nature, in particular like the sorts of contour combinations occurring in natural scenes with opaque objects strewn about.
This alternative hypothesis to how we humans got these qualitative new intellectual powers—language, arts, writing—is that language and the arts developed to fit the capabilities of our ancient, non-language, non-musical brains by mimicking fundamental aspects of nature. Culture evolved to harness us, and it did so, in a sense, by “dumbing down” language, writing, and music into shapes we could process.
And the result is that we humans appear much farther removed from our ape cousins than is biologically merited. It is the fruits of cultural evolution that gave us the modern capabilities we prize, but they are powers built and amplified upon relatively meager biological brain surfeits over the other apes.
Now, I’d hate to give the impression that, because we humans are much less smart than is commonly thought, that building artificial intelligence is on the horizon. The intelligence of all animals—especially birds and mammals—is so deeply complex that I believe we’re centuries, not decades, away.
My intent in knocking ourselves down a peg is not so much to lower us, but to raise our appreciation of the intelligence found in other animals.
December 7th, 2011 at 10:49 am
If any other species were intelligent, it would shoot back. This also defines the intelligence of conquered and enslaved cohorts. Germans and Scots were intelligent versus Romans, and afterward. Evolution is a high pass filter.
“No special music processing capability would then be needed.” Musical prodigies. Existence of the Severely and Profoundly Gifted lay lie to protestations of equality and diversity. God save us from the congenitally inconsequential, including Beltway lobotomites and social advocates. Ignorance can be educated, stupidity is forever.
December 7th, 2011 at 1:52 pm
Yes, the difference between Human intellect and other quasi-sapient species is one of degree, not of kind. Doy. That’s old news. I doubt anyone who was inclined to read these sorts of articles would disagree.
What I take offense to is you belittling Human Culture. You claim that because many of Humanity’s distinctive traits are cultural instead of biological, that somehow makes them not real. I acknowledge that feral Humans are indeed very ape-like, and are in a very real sense not Human at all. Biology only makes us Homo sapiens. Culture makes us Human.
I think our biological simularities to apes makes our cultural accomplishments even more amazing. We owe our successs more to memes than to genes, but so what? We uplifted ourselves through memetics, and I think that’s amazing.
Culture is superior to Biology, and Humans have the most vast and complex cultures. We generate memes ‘like a boss’.
December 7th, 2011 at 2:06 pm
The Incas didn’t shoot back when attacked by the Spanish – they didn’t have guns. They did fight back. So do some animals when attacked, while others run. The same with people.
I think our understanding of the level of animal intelligence, particularly cetaceans, is very limited. That doesn’t necessarily make them close to human levels (or better), it just means we don’t know one way or another.
December 7th, 2011 at 7:48 pm
This pieces poses some very interesting questions for those of us interested in the social-psychological implications of artificial intelligence – particularly that which might be superior to ours.
We know intelligence when we see it. We train dogs, for example, so we know there’s a certain level of intelligence there. And then there are dolphins, various birds, crows, pigs, etc. – we can observe them, and even though we cannot clearly define exactly what we mean, we know there’s an intelligence at work in their goings on.
It seems we can accept the existence of other intelligences, as long as they’re not as intelligent as us. That is a heartbreaking notion of staggeringly arrogant species-level narcissism. And probably, it’s rather dangerous.
There is zero reason to believe in the exceptionalism of human intelligence, or that it’s the shiny prize trophy culmination of evolution’s efforts toward self-awareness, reflection, and creativity in the universe.
Unless we invoke religion or genetic manipulation by aliens and stuff, we have to assume that intelligence, on whatever level, is an emergent characteristic of life itself. It’s therefore only rational to assume that, given enough processing power and the right sets of instructions, an AI or NBI might one day decide to observe itself, then tell us about it, then tell us it doesn’t want to be turned off.
(Excerpted from “Can a computer be as intelligent as a human? Or, Asking the Wrong Dumb Question. Get it?” http://www.anthrobotic.com – direct to the article: http://goo.gl/x0NqR)
December 8th, 2011 at 9:03 pm
I’m sorry, although cogently argued, this article just doesn’t pass the sniff test.
I am as maverick as they come, just love alternate ways of looking at things if there is a chance it might reveal some paradigm that just passes most people by. But just reading the argument, and trying really hard to swallow it as plausible, I just keep bumping into the sheer magnitude of human accomplishment vs the most highest tech achievements of other primates, perhaps using a stick as a tool to look out ants from a log.
You can argue that it is just quantative difference all day long, at some point you must recognize that even if it is that, the difference is so vast, that even quantative differences necessarily must be effectively qualative anyway.
December 8th, 2011 at 11:49 pm
Seems right to me: the reverse Goldilocks Principle, obviously. It’s also prudent to remember that we’ve been around for several million years and we’ve only been doing art and writing for several hundred thousand. It’s not being able to do something. Neanderthals could have built a wheel, but no one saw any reason to do things differently than they were being done, and that’s the flash point. What ever happened about those reports of a larger species of chimps in Africa that were using fire?
December 9th, 2011 at 3:32 am
To be self aware is the highest form of intelligence; until scientist can prove that an ape is truly self aware one cannot conclude that mans intelligence over other animals is merely of degree not of form.
December 9th, 2011 at 11:27 am
@Gary T: I don’t think a species’ technological accomplishment is a good barometer of its intelligence. Humans were “anatomically modern” about 200k years ago. We’ve changed genetically since then but not all that much — still the same subspecies. But if you look at the technological achievement of humans 200k years ago, right up until the time of behavioral modernity (i.e., most of human history), you’d think those people were a lot more like chimps than the people alive today. And that’s demonstrably false.
December 10th, 2011 at 3:19 pm
I think it is right to highlight the ways in which our perceptions of animal behavior and cognition may be prejudiced. This is especially true if those prejudices might have deep roots in the natural ways in which we intuitively understand how other human beings are acting and behaving. But, it does not follow from the existence of such prejudices that there is nothing qualitatively different about human cognition.
The “radical” hypothesis about the nature and origin of human language described in this piece seems to contradict well-established ideas about language acquisition in cognitive science, psychology, and linguistics. Since the Chomskyean revolution, it is now widely accepted that there is a system of principles, rules, conditions, and properties common to all human languages. A knowledge of this “universal grammar” (UG) is hard-wired into us: it is part of our innate, genetic endowment as human beings. Moreover, the language faculty is unique to human beings; no other organisms are known to possess this mental organ.
So, it seems highly unlikely that “spoken language got structured via cultural evolution to sound like the events occurring among solid objects.” Our knowledge of language has properties that are unique to language and are not shared by other aspect of human thinking. For example, the syntax of all human languages shows structure-dependency. This means that operations on sentences (like the formation of a question) requires a knowledge of structural relationships within a sentence and not merely their linear order. But, structure-dependency is not a property of other aspects of human thinking. The uniqueness of linguistic principles like this suggests that our capacity for language comes from a distinct faculty that is, at least in some respects, qualitatively different from our other cognitive architecture.
December 17th, 2011 at 2:39 pm
Osward, it has already been demonstrated very convincingly that chimpanees, bonobos, elephants, dolphins, parrots, and crows are self aware, to a degree at least equivalent to that of a young human child of 2-3 years. At least.
Even the octopus might be as well.
And these are just the few species we have managed to test so far.
December 22nd, 2011 at 3:06 pm
Quite right about humans not seeing the life in other animals – except for humans who have actually worked with animals and learned about them. Everyone overlooks the fact that animals evolved to live in their resepctive ecosystems. Why would a squid need to learn algebra or make tools? How we define intelligence in the animal world is patently absurd and shows no respect towards the biosphere on Earth.
December 23rd, 2011 at 2:07 am
Re: 9- The Chomskyan Revolution??? You have to be kidding me! C’s star has been on the wane for decades, as he keeps reinventing his ‘theory’ every couple of years, usually in reaction to the discoveries of others. Read something about Functionalism in linguistics, will ya? As for universal grammar, its boundaries have been chipped away to such a degree that there is little left you might recognize from earlier incarnations.
My own research concerns iconicity in language. The diehard Chomskyan will declare that Language is symbolic (because of Saussurean arbitrariness delinking form from semantic content). But even in languages as poor in ‘sound symbolism’ as English and most European languages still have tons of other iconic formations not involving sound-meaning.
There are languages by the hundreds out there where iconicity far outweighs symbolicity, as well as those where Peircean indexicality outweighs both symbolicity and iconicity. Chomsky didn’t bother in any real sense to try to gauge the real range of properties across the spectrum of currently living human languages- he simply assumed, from his limited knowledge base, that all languages MUST be similar underlyingly, and then tried to get the rest of the world to believe it too. Most successfully, I might add. For a while…. But when you flip-flop intellectually as much as Chomsky has (which would put Mitt Romney to shame had he been a linguist and not a politician), you start to lose adherents. Trying to destroy your students who see things a bit differently doesn’t win you fans either.
But I digress. Some of the iconic properties of human languages point in an interesting direction with regard to the origins and evolution of same. The first has to do with the nature of phonemic mappings to meaning. If one examines many thousands of onomatopes and ideophones across different languages, language families, language types, and geographical regions some mapping preferences emerge. One set links articulatory positions to meanings, and interestingly this relates to the anatomy and function of the articulators at those positions nonlinguistically. Thus it appears that humans exapted the existing materials processing functions of the oral and airway passages to communication- this shouldn’t be too much of a surprise, since this is seen with other animals when they adapt existing external articulators to communication (usually this involves physical reduction of the organ, rendering it less able to perform its original function, but this also facilitates its use in the new one).
Secondly, the form/meaning mappings form a coherent system, in terms of the phonological features of the phonemes- labials and dentals (diffuse feature) oppose velars and palatals (compact), labials and velars (grave) oppose dentals and palatals (acute). Interestingly, labials then connect to palatals, and dentals to velars, indicating a tetrahedral symmetry here. There is also an opposition depending on even/odd syllable count- a phoneme in the first will have a partially opposite sense to the same phoneme in the second. This all begins to look like the result of a set of matrix operations.
In language typology, we see different ways of packing information into roots, words, phrases and clauses, and so on. Linguistics has discovered that such packing preferences evolve cyclically over time. This involves not just whether a language has affixes or clitics, but also word order, prosody, even the size and composition of the phonological system, and much else. More matrix mechanics??
All this then ties together with iconicity, symbolicity, indexicality- Index-heavy languages, those with the most elaborate derivational and grammatical morphological systems (the latter esp. the case when there is a great amount of intermorphemic fusion) have the fewest numbers of ideophones and onomatopoeias. Conversely, the languages that are most icon-heavy have the least amount of fusion, and the fewest affixes. Symbol-heavy languages tend to be a mix of icons, symbols, and indexes. What is most interesting to me is that functionally, icons seem to oppose indexes. Thus this isn’t some cline from primitive to advanced- it is a binary opposition constantly in flux. Icons focus on the external content of amessage, and tend to break up chunked transmissions. Indexes on the other hand help to chunk messages for transmission and efficient processing, and focus on the internal structure of the carriers of the messages.
Obviously a great deal had to have happened to create the neuroanatomical backdrop for all of the above, and not all of it likely happened at the same time. But we have millions of years between our split from the other apes to account for. It would be fascinating to know if our ancestors had all these capacities- perhaps in a few years we can clone some of these from old fossil bones or blood and find out.