July 1, 2004

Social Intelligence

I've been meaning to write about this for a while, but Carl Zimmer's article on Machiavellian monkeys reminded me of it.

I have had on my in-progress shelf for a few weeks now Robert Sawyer's Far-Seer, a novel about a Galileo-like character in a race, the Quintaglios, evolved from Velociraptor-like creatures on a planet with some unusual geography. (It's a tidally-locked moon falling within the Roche limit whose only continent is on the side facing outward from its primary.)

Sawyer's race are high-intelligence obligate carnivores constrained to live in low-density settlements due to a highly-developed territorial instinct that routinely results in violence.

And yet, and yet, these carnivores, who are nearly obliged to be solitary, are pack hunters. In addition to the fierce territorial instinct, this species also, more or less involuntarily, binds into hunting parties via pheromones.

I won't entirely say this is nonsense. I will say that there is nothing like this on Earth.

It is always risky to draw conclusions about how life must work based on life on Earth. One has a limited ability to generalize from one example. However, nothing suggests that an intelligent species could even remotely operate like this.

Let's take the solitary pack hunters. I can think of only about two groups that behave anything like this: Sharks (especially pelagic ones) and vultures. Both are normally approximately solitary (modulo migrations) and both converge to form large groups at feeding sites. Both can be easily distinguished from the Quintaglios: Vultures are not hunters and sharks are not cooperative. Both groups are far more opportunistic than stalking and neither group is notably intelligent within their grades. All cooperative active hunters on Earth (at least among the vertebrates) live as well as hunt together. There's a likely reason for that: successful cooperation, especially with division of tasks, requires you to be able to anticipate what your packmates are going to do, and that requires familiarity with them. This is especially true as you get more intelligent and thereby have more possible actions to take.

Next, let's tackle the uncontrollable territorial instinct. One of the gifts that intelligence seems to bring is the ability not to have instincts and to override remnant instincts. This is likely to be virtually a definitive characteristic of intelligence, whose nature of being able to assess and anticipate to choose among a broad spectrum of behavior necessarily occupies the same role as instinct in determining behavior. Among humans, our few remaining instincts (probably mostly reproductive or otherwise broadly biological in nature) are extremely general with effects visible only in the aggregate; we certainly have nothing resembling an irresistible urge to attack anyone within our territory or close proximity who's not signaling submission.

It is worth noting that the Quintaglios are well-armed in fang and claw and that squabbles resulting from their instincts are usually injurious, as compared to analogous dominance contests in social animals, which rarely result in serious injuries. But compare to, say, fights for harem control in patriarchal animals like the walrus; these fights are usually serious. The difference is whether or not the fighters will remain in the same pack after the fight. Animals that expect to remain in social contact do not injure each other because they will need to retain each other as allies against external forces.

Let's move on to the low population density due to the territorial instinct. Here's where we come to the biggest objection and probably the strongest statement we can make about intelligence among Earthly life. How did the Quintaglios develop intelligence at all? It is the case on Earth that predators are usually more intelligent than herbivores within the same grade; tuna can outsmart herring, wolves outsmart deer, and lions outsmart kudu. This is illustrated by the old anti-vegetarian slur "How much intelligence does it take to sneak up on a blade of grass?" We can expect that the ancestral non-sapient proto-Quintaglio would be more intelligent than its prey animals. But the only pressure on intelligence offered by predation is that the predator must be more intelligent than its prey. The proto-Quintaglio is constrained by its territorial instinct to remain virtually solitary, probably overlapping only with members of the opposite sex and probably only during mating times (except for, inexplicably, banding together to hunt in packs). They are, in fact, precluded by their territorial instinct from the only other known phenomenon to produce intelligence: sociality.

Social behavior is not sufficient to produce intelligence; one need only look at ants and antelopes to see that. However, it does appear to be a necessary condition for intelligence. Within a grade, the most intelligent animals are social: Ravens and parrots[1] instead of hawks; chimpanzees instead of orangutans; dolphins instead of right whales. Complex social behavior can feed back on itself: if you use your brain to handle social behavior within your group, you're better off with a bigger brain. Intelligence arising from predation increases only in concert with prey intelligence; social intelligence increases based on pressure arising within the species. This appears to be the most productive explanation of intelligence evolution in primates. Primates are across the board both more social and more intelligent than other mammals. They have historically evolved bigger brains faster than any other group. Among living species, the more complex a species' social environment is, the more intelligent it is.

The Quintaglios' prey animals are not notably intelligent, so the predation phenomenon is not responsible for their intelligence, and they are precluded from social reinforcement of increasing intelligence, so there is simply no phenomenon observed on Earth that could explain why their species tended to develop intelligence.

Next time you fight with your in-laws, remember: It is to fight with your in-laws that you developed intelligence in the first place.

[1] Parrots are something of a special case. Parrots are intensely social, but no one would accuse the similarly-social sparrow of intelligence. It is likely that psitticid intelligence was kicked off by another phenomenon beyond pure sociality, with a plausible hypothesis being made that their rain forest environment encourages the development of a mental and temporal map of the territory to keep track of which trees are going to produce fruits and nuts at which times. Such a map is a complex cognitive task requiring relatively large brains. Once they got nudged in this direction, though, sociality would tend to reinforce the trend.

Posted by Greg at July 1, 2004 2:51 PM | TrackBack

Comments
#1 ::: Matt Rossi ::: July 2, 2004 8:55 AM ::: link

Do parrots have color vision? One of the better arguments I've seen for the development of primate intelligence was that it flourished in part in order to allow our forebears to detect fruit at a distance, to process visual cues from a variety of sources and that as the amount of visual input that was available to be processed increased (include facial recognition and the ability to read emotions via facial expression... primates having a very expressive facial structure) the brain increased in sophistication in part to cope with it.

#2 ::: Patrick ::: July 2, 2004 9:59 AM ::: link

That was a very interesting post and I had a few questions/comments:

1. What does the sentence "It's a tidally-locked moon falling within the Roche limit whose only continent is on the side facing outward from its primary" mean? I am only a primitive cave-man lawyer and your technology frightens me.

2. A possible explanation for the emergence of intelligence in this scenario is third-party intervention (a la Uplift), whereby another intelligent species arrives to inflict intelligence upon others. I assume that this is not mentioned or dealt with in the novel at issue but it was a thought.

#3 ::: Greg Morrow ::: July 2, 2004 11:26 AM ::: link

Matt:

So far as I know, parrots have better color vision than primates. Basal reptiles have five kinds of cones (compared to our three); the hypothesis is that basal mammal color vision was suppressed due to being our being primarily nocturnal.

Primates have had relatively good vision from early days, but all you need to handle good vision is a good visual cortex; I wouldn't expect it to drive other kinds of intelligence. Aside from fruit, living in a tree encouraged three-dimensional vision, and once you can do that, being able to see faces clearly comes for free. Color vision comes from diurnality; attributing it to recognizing fruit is crediting the Elbe for the North Sea.

Patrick:

"Tidally-locked" means that a satellite always keep one face to its primary and one face away. The moon is tidally locked to Earth, so we speak of a "dark side" as the face that's always facing away from us. The mechanism that locks the satellite is the same mechanism (gravity having differential force over the width of the satellite, i.e., the near side being attracted more strongly than the far side because the near side is closer) that drives tides, so it's called "tidal".

In general, satellites have slowly changing orbits, e.g., the Moon is slowly moving further away from the Earth. In this case, the Quintaglios' world is moving closer to its primary (presumably a gas giant in the liquid-water zone around its sun).

The Roche limit is how close a satellite can get to its primary before tidal forces break it apart. That is, based on the size of the satellite and the strength of the material it's made of, if it gets too close to its primary, it'll be ripped into shreds and create a belt of debris not unlike the asteroid belt or Saturn's rings. So the Quintaglios are in a crapload of trouble, because in a few thousand years, no more, they're not going to have a home any more.

In the specific case of the Quintaglios' planet, it's mostly a water world, with a single equatorial continent, and that continent happens to be on the side facing away from the primary, so they evolved civilization without ever seeing the planet they orbit. When an explorer sailed far enough out to see the planet, she decided it was God, and, Mecca-like, each Quintaglio once in his life is expected to make a pilgrimage far out to sea to gaze upon the face of God. It's actually a pretty clever piece of mythologizing, especially when you count in the equatorial winds and how they've contributed to the Quintaglios' believing they are on a continental raft floating down a very large river.

#4 ::: Patrick ::: July 2, 2004 11:58 AM ::: link

Well, what do you know? I learned something new today!

Thanks!