Culture Does Not Optimise

Ya might want to rethink that. They were transitioning from atlatls to bow and arrow. I am largely ignorant about the topic, (except for this pile of spearpoints, arrowheads, scrapers etc. next to my elbow on the table) but I doubt an atlatl point would work very well on an arrow, and a certain amount of fiddling around (otherwise known as experimentation) would have to be done to end up with a useable arrowhead. Not optomized maybe, but better than the clunker on the end of a spear. I think that answers your comment about demonstrating that the later types are in fact better than the earlier ones.
I don't remember where I read the first report of this, ah, Science Daily, I see the link in Afarensis now. Did you read that? You may turn out to be right, but I'll wait til the full article is available. rb

That initial period of tweaking would be too short to show up as a separate phase in the chronology. No hunting-based culture produces unusable arrowheads for any length of time if they ever figure out how to make good-enough ones. And the tweaking would not tend toward optimisation, because lo-tech people can't measure that sort of thing accurately.

Have you read, "Natural selection and cultural rates of change," by Deborah Rogers and Paul Ehrlich? If you have, I would think that you'd acknowledge that certain cultural traits have a functional aspect that can be selected for and are not completely arbitrary.

You criticize the arrowhead model by contrasting it to face painting. But you can't. Those two cultural traits can't effectively be compared, one is functional the other is symbolic. In the Rogers and Ehrlich paper, functionality (traits that affect whether or not the occupants of the canoe will survive or not) changes very little once optimized, whereas symbolism in canoe design (such as aesthetic, spiritual, and decorative) is more variable.

If you don't get it, let me outline another example -- fish hooks and fishing poles. A group of fishermen use different fish hooks and poles to catch fish. Some sport designs that just tie bait to line off a fishing pole that has feathers and trinkets adoring the pole. Some fishermen sport designs that hide a hook within an elaborate bait, tied to a line off a plain fishing pole. Others sport different variations.

The successful design is the one that yield the more fish. The fisherman that walks home with the most fish survives and thrives while the other fishermen, with the failed design decide to become interior designers or bloggers. People don't have to be trained scientist to discern that the hook and pole that yields the most fish will be the one that's selected. Clearly the functional element of the fish hook and pole, the hook, will be selected and optimized, while the symbolic element, the pole will vary.

I really don't get what you mean by there ever existing 'pre-scientific humans' classification? Humans have always been scientific, even before Ibn al-Haytham synthesized the scientific method some 1,000 years ago. Stone tools have always been optimized, as have other functional cultural items!

Kambiz

Ah, yes, I see your point. You're probably right. I'll be interested in seeing what the paper has to say about that.
I'm not sure I can go along with your comments about lo-tech people though. They sussed out some amazingly subtle stuff, and I can't imagine they were any less inquisitive than we are. Some of the Amazonian pharmacological stuff just flat amazes me. The ones involving multiple ingredients and processes, where none of the individual ingredients or intermediate steps results in an active substance. The fact that they got from 'here' to 'there' despite their being lo-tech just causes me to shake my head in wonder.
Even with arrowheads, I'll bet they would have had plenty of time, and the inclination, to experiment. I can imagine a bunch of men and children setting up a target competition, complete with prizes. It wouldn't take long for the winner's design, and modified versions of it, to spread through the group. Just like today, when someone down the river bank reels in a big one. "What kinda lure are you using there, Bubba?" But you're right, that short period of tweaking the design would be unlikely to show up in the record. By the way, I though their illustrative points/arrowheads was a little flaky (pardon me). Anyone with a pile of points could quickly assemble something similar, with no meaning whatsoever. rb

And yet it seems plausible. Without written language, the only way to learn a complex technology is through direct training, usually by familial association. Families with better techniques had an easier time surviving, pushing out other methodologies. Even a slight technological advantage could over long enough time be enough to push out another slightly less optimized way of doing things. Sounds like evolution to me.

Kambiz: I swear I hadn't read your comment before I too went fishing. I wouldn't have bothered, yours was better. rb

I don't see how the study could control for factors like the availability and ease of working of the raw material , cultural preconceptions about what type of point penetrates which type of fur better or size of animal being hunted. Not that I will ever see the study, since it will likely be behind a subscription wall. BTW, the picture of the point with your blog post would be considerably more meaningful with a size scale included.

Cultural trends are optimized in the same way evolution optimizes traits.

"They did not have a good large-scale pragmatic overview of what they were doing, and indeed they would not have acted completely rationally even if they did have access to such an overview." Natural selection doesn't need large scale pragmatic overviews. It just needs success and failure, which can also occur in hunting and many technologically influenced activities.

"If pre-scientific humans had been technological optimisers, then all arrowheads globally would fall into a small number of standardised task-specific types." Natural selection does not only lead to convergent evolution, nor do all similar habitats contain the same species. The tropics of Africa house very different species from the tropics of the Americas, but it not because there was a failure to optimize the species' traits in either group.

"What those ancient archers are likely to have done is make good-enough arrowheads" again the is the exact mechanism of random mutation. Each individual mutation or change does not care if it is 'better' or 'worse', it just cares if it is good enough to pass some arbitrary threshold. For genetic mutations it is the ability to pass on genes to offspring. For technological change it is if the technology provides enough food to feed your family for long enough to pass on the skills to the next generation.

"Today I paint my face red for the ritual dance. Twenty years from now my daughter will paint hers blue. It is deeply significant from a cultural point of view, but completely irrelevant from an evolutionary one." There are many biological traits with direct parallels to this. The feather's of Peacock are a classic example. These can be truly superfluous or have non-unitary goals, such as species (or cultural) identification, where there are an infinite number of "optimal" solutions.

With both biology and technology, random change optimizes when given scare resources. It's true that if either produces non-functional limbs or arrowheads, that those will not stick around very long at all. But consider the following circumstances. If we have two arrow heads, A and B, A allows our average hunter to catch enough to eat 95% of the time, and arrow B allows the same average hunter to substain himself and his family 96% of the time. With even a small difference like that, and staring with an equal amount of people using the two arrow heads at the beginning, the population converges over the course of 200 generations to have 10x as many people using arrowhead B than A.

Did you factor in that arrowheads were not used just to get food, but for tribal warfare? The relative qualities of the arrowheads used had a lot more to do with ultimate survival in that setting than in the food gathering arena.

I think to speak of this in terms of optimization is not warranted, and certainly Lyman et al did not analyze the data in terms of optimization theory. The main point is that a certain pattern namely a burst of variation followed by a subsequent decrease in variation has been observed in their data. The question is what explains that pattern? Since I haven't read the article, that is the best defense I can come up with. I will probably return to it when the article does come out. Like Kambiz, I think you are selling the intellectual abilities of Mesolithic and Neolithic peoples short. Paul Radin's Primitive Man as Philosopher is a good overview of the subject.

I'm with Afarensis. Optimization is a poor choice of term. And really, evolution doesn't optimize either, it's a 'good enough' system too, which is why we have all these backaches and poor eyesight, etc...no designer optimizing us :P

All biological species that exist long enough to get recorded are strongly formed by their environment -- that's what I mean by "optimised". Not so with cultures. Radical changes from one lifeway to another, where whole toolkits get modified beyond recognition, happen without any environmental stimulus.

Archaeological cultures in a certain stable environment will look one way for hundreds of years, and then change into something else that lasts for hundreds of years. It's cultural capriciousness. And that goes both for the choice of flanges and dimensions on arrowheads, and for the design of ceremonial headgear.

Most low tech is not strongly determined by the environment. Arrowheads just need to be pointy. And cultures will hang on to strongly non-adaptive traits (like medicinal blood-letting or the ceremonial eating of dead people's brains) for millennia.

I am convinced that both people in the lo-tech past and people of the present are largely ignorant non-optimisers who mainly do things with no adaptive significance for no good reason. White noise.

you say:
This sounds like complete bollocks to me. If pre-scientific humans had been technological optimisers, then all arrowheads globally would fall into a small number of standardised task-specific types. Culture does not change through time in a manner like that of biology, because cultural selection is largely irrational. Most of what people do is white noise from an evolutionary perspective. Null mutations.

and then:
People bungle forward in a good-enough way, guided by superstitions and half-formed ideas of what the world is like. Some bungle in a non-good-enough way, and starve, or have their culture obliterated by nuclear war or global warming. Archaeology and anthropology is the study of the cultural specifics of this bungling.

most evolution on a molecular level may be neutral, but that does not imply that all of it is, and that selective events are not selective events because most of evolution is neutral. one problem is that evolution on a molecular level has something we can grasp onto concretely so we can quantitize it; discrete sequences of DNA characterized by variation across time and space. culture isn't like that, so asserting that cultural evolution is mostly random-walk or selected through function is subject to distortion on how we classify cultural traits. all that being said, i would tend to agree that most human culture has a strong element of noise, though i think noisy fads and fashions can also exhibit functional saliences in particular contexts.

but in any case, look at what i bolded in the second quote. you do acknowledge functionalism by asserting that some starved because they bungled too much, there were differences in fitness contingent upon cultural variation. there are signals of selection in the noise. the question is just what is noise, and what isn't, it isn't to dismiss non-noise out of hand. modern engineering synthesized with theoretical science is VERY new, most pre-modern technological advance was do to what you might term "bungling," and it was "sub-optimal" in a grand absolute scale. nevertheless, there was persistent and incremental advance in technology over tens of thousands of years before the modern world (as measured by the increase of the malthusian population limit over time). this suggests directionality to the bungling, not a pure random walk.

joe henrich has done some theoretical and empirical work on these parameters re: culture. in small populations the random walk processes do tend to swamp learning and functional advance to the point where isolates lose cultural traits over time; just as small populations lose genetic diversity. in contrast, larger populations tend to advance in their technology slowly but persistently, as generally occurs in terms of adaptation in larger populations genetically.

there's no need to make grand theoretical assertions re: noise vs. non-noise, just focus on the details of the empirical work.

All biological species that exist long enough to get recorded are strongly formed by their environment -- that's what I mean by "optimised".

i don't want to get bogged down in the semantics here; environmentally drive adaptation is important, but you allude above to the ubiquity of null mutations and neutrality in evolution. ergo, you yourself imply that this non-environmental forces are also powerful, that is, parameters endogenous to the system. also, there are debates as to the environmental or random nature of phenotypes. i myself tend to lean toward environmental influences, but there's a debate. IOW, the analogy between biology and culture might be a lot closer than you're implying.

Archaeological cultures in a certain stable environment will look one way for hundreds of years, and then change into something else that lasts for hundreds of years. It's cultural capriciousness. And that goes both for the choice of flanges and dimensions on arrowheads, and for the design of ceremonial headgear.

you could posit a model where there's stabilizing selection. it pays to go along with the group and conform. until that is an exogenous environmental shock throws this equilibrium out of wack and functional selection kicks in. i'm not saying that's how it works, i'm saying that that is a model you could posit where different forces operate at different times and spaces. i'm trying to push back against the dichotomy you seem to want to argue around. i don't think you'd get many defenders of the "optimizing" position you're presenting, because i doubt many people would defend such an extreme position.

i do think that the distinction between arrowheads and head-gear is important. if you don't want to concede the distinction i simply ask that you not eat for 1 week and not wear a hat for 1 week and note that there is a physical difference.

I am convinced that both people in the lo-tech past and people of the present are largely ignorant non-optimisers who mainly do things with no adaptive significance for no good reason. White noise.

again, i'm bolding the critical part: you say "mainly" so you imply that some things are of adaptive significance. i think many people would admit that head gear falls into the noise category re: environment. but arrow heads seem more likely to be part of the minority of traits with definite adaptive salience.

(p.s. you seem to also forget that selective processes are rather noisy themselves)

I don't agree with you here Martin. I think that the same evolutionary processes affect material culture as genes (i.e. natural selection, mutation and drift affects objects ass well as genes). If your material culture contributes to your fitness, it will be under selection in the same way that your genes are. But thats not it. There are even more evolutionary processes that effects culture than affects genes, since cultural transmission can be biased in ways that genetic heritage can't be. E.g if your neighbour allways comes home with more game then you, you will like copy what ever stratagy you neigbour uses. If an infuential person in your comunity does things in a specific way, an easy way to be more influential than you all ready are, is to do thing the same way as the influetial person does. And if everybody else does things in a specific way, you will more than likely do things this way yourself, even if you think that it is a stupid way to act (most people will anyway :-).

Why don't you think that the human species main adaptation (culture) is not under selection as every other adaptation is?

/Micke

I haven't read the original paper, but judging by the comments I have seen, most people ignore the difficult part in evolving better arrowheads: how to make them.

It is pretty easy think about a good shape. It takes less than an idle afternoon to develop an idea of an optimal arrowhead that cuts deeply and stays in the wound. It will take much longer to develop the technology to make that shape, given limited resources. There are only a few kinds of rocks that can be used, and they may not be available anywhere near. More likely the first arrowheads were made of wood or bone.

But once you learn the engineering part, you can experiment with details. At some point you'll notice that putting more effort in the design gives better yield, but there is also an upper limit. Arrows get lost, and investing days to make a good head goes waste. In the end, the Bell curve rulez. It is called optimisation.

There is another wee, offkey (if not OT) matter - simple human eccentric inventiveness.
I am thinking paticularly about fishing - lures etc...

Why don't you think that the human species main adaptation (culture) is not under selection as every other adaptation is?

It's under some selection for adaptive fitness, but not strongly shaped by that selection. The technological changes through the Upper Palaeolithic and Mesolithic, for instance, hardly had any practical consequences, though typologically they were very great. Most of the variation we study is not due to adaptation.

One thing cultural evolutionists seem to be vague about is what unit is under selection. Is it the archaeological culture? The local society? The person? Or even the artefact type?

The technological changes through the Upper Palaeolithic and Mesolithic, for instance, hardly had any practical consequences, though typologically they were very great. Most of the variation we study is not due to adaptation.

I don't know whow you figure that. The introduction of microliths, bow and arrow and other technological adaptations geard at hunting small game would have great practical consequenses. Harvest pressure theory, optimal foraging strategies and diet breadth models have much to offer I think when studying the paleolithic/mesolithic transition.

One thing cultural evolutionists seem to be vague about is what unit is under selection. Is it the archaeological culture? The local society? The person? Or even the artefact type?

I am much an advocate of dual inheritance theory, so I would say that humans are affected by both genes and memes. We inherit both of these, which means that they are copied and thus subject to evolutionary processes. Therefore all of these aspects you mention are worth studying from an evolutionary perspective.

/Micke

Wow, tens of thousands of years passed and they invented the bow and arrow.

I am much an advocate of data loss, ineptitude, unbridled impractical creativity and random staggering in cultural change. If a scholar studies culture from an evolutionary perspective, then she will either have to disregard almost all the data that's really interesting, or commit the panadaptationist fallacy.

If a scholar studies culture from an evolutionary perspective, then she will either have to disregard almost all the data that's really interesting, or commit the panadaptationist fallacy.

Stupid dichotomy. A scholar may want to specifically find and study that which is under selection. This is done with genomes all the time without needing to "disregard" the neutrally evolving parts. And in the Rogers and Ehrlich paper, it appears that they did not need to resort to those sins either.

Martin, I agree with your critics about the functional/symbolic distinction. We do see superior technologies crowding out less effective ones in prehistory. When someone invented a bronze sword with a tang that went all the way through the hilt, so it could both cut and thrust effectively, that technology became standard all across Europe and the old one died out. That wasn't just fashion, because no people ever went back to rivitting their swordblades onto hilts. Where people could make good bows out of readily available materials, they stopped using spearthrowers to hunt. The arbitrary part is between options which are equally adaptive in a particular circumstance (eg. wearing trousers versus wearing a kilt, or all the different ways that a bowl can be decorated).

The researchers have to prove that the later arrowheads were more effective than earlier ones, but their argument is very plausable. Every historical culture that I can think of produced several styles of arrows (with subtypes) to meet several different needs.

Windy, the evolutionary scholar will need to look at nonadaptive stuff only briefly to weed it out of the equation.

Sean, several subtypes of arrows, many of which duplicate the same needs, yet which look different synchronously and diachronously for no practical reason at all.

I am much an advocate of data loss, ineptitude, unbridled impractical creativity and random staggering in cultural change.

A.k.a mutation and drift. ;-)

/Micke

Can someone who's read the paper tell me how Lyman et al. measured an arrow's effectiveness?

Micke, the traits selected for in a drifting population of animals or plants are strongly determined by the environment. With a culture, however, the traits that get picked up and become prominent and long-lived are mainly determined without adaptive consideration.

Just look at Late Bronze Age graves. Full of razors and tweezers, of all the bizarre things they could have chosen. And those things are hallmarks of their culture. People are strange.

Martin: Sure, people can persist in using a technology which isn't as good as it could be for cultural reasons (such as having invested much effort in a technology which is almost as good). But there are many ways of making a good arrowhead, or suit of clothing for a Central European climate in summer, or whatever. You don't have to deny that preindustrial people could judge what worked well and what worked poorly, because each of an infinite variety of slightly different problems has an infinite number of slightly different, roughtly-equally-good, solutions. Where an inferior technology persists after a clearly superior alternative is introduced, its usually counterbalanced by other advantages (European colonists wearing heavy woolen clothes in Africa), isolation from competitors (Japan 1600-1850), or resistance to change (Japan again; or the English-speaking world's obsession with driving everywhere). This has nothing to do with a myth of noble savages and everything with respecting people as good craftsmen even if they didn't know a lot of things we do. The idea that our ancestors were stupid, while we are smart, is a dangerous one.

I have gone back to using a manual reel mower on my lawn rather than a motorized, fossil fuel driven one. It is less efficient but absolutely good enough. I like the fact that it is quieter and doesn't require fuel. Perhaps I will convince some of my neighbors to go this route. Then many centuries from now, some future archaeologist can discover the remains of these mowers and debate whether the residents of Prospect St. were noble savages or died out because we failed to optimize.

Sean, we seem to agree. There are many ways to make a good-enough arrowhead, so adaptation does not explain the shape of any given one.

I'm arguing that our ancestors were stupid and so are we. Just look at all the pseudoscience people buy into today.

Or rather, I'm arguing that people in all centuries devote their intelligence and artistry to the pursuit of non-optimising cultural capriciousness.

"Arrowheads just need to be pointy." Martin

I have shot arrows with heavy huntingheads (steel things with 4 flanges), standard target heads, and rubber blunts (NOT POINTY), all with 28 inch shafts and from 30 pound bows. Even with my lack of skill it is obvious that the different types have different trajectories and a different impact on the object they hit. A rubber blunt would be useless against an elephant, will bruise a human, and can kill a seagull (you should have seen the surprised look on it's face as it fell off the post). If you're after small prey very little is required of the arrow head, what's more important is the fletching. If the feathers are crooked the arrow won't go where you expect it to, and dinner will hop/fly away before you get another chance to hit it.

"I am convinced that both people in the lo-tech past and people of the present are largely ignorant non-optimisers who mainly do things with no adaptive significance for no good reason." Martin

They've got a prefectly good reason - everyone else is doing it. You've got to keep up with the Joneses. ;)

Awesome you accuse O'Brien and Lyman of "just so stories"--that's their main term of abuse for most of the profession.

There isn't a whole hell of a lot of variation in true arrowheads in North America. The range of forms is repeated worldwide (I recently embarrassed myself asking where some "Nodena" elliptical points came from--South Africa, thank you), just like surface treatments on ceramics. I haven't seen their latest book yet, but I think there isn't a lot more variation in the typically early stemmed forms as there is in the typically later triangular forms. What charcateristics will you choose as significant makes a big difference in how much variation you see. Th raw materials used in most of region are fairly similar vaiations on gravel chert and outcrop tabular chert; they do have some variability, mostly in how much heat-treatment helps. Variation in abosolute size, length, width is likewise limited.

I thought the whole point in using style was to find bits of variation that are usefull from a chronological standpoint. he variation observed ethnographically in N Am arrows (whole arrow, not just pointy rock) was generally stylistic variation that marked territory/ethnicity/ownership.

I have never heard any evidence that stemmed points are "less efficient" than triangular ones. If there is any functional difference, the longer ones might be "better", but there are some laonger stemmed forms, esspecially early one (if the Gary var. Issaquena is an arrowhead and not very late, very small dart point).

And what about the blunts (the real "birdpoints"), conical antler arrows/harpoons, gar scales, sharpened, hardened cane, etc.? Those are all vatriations, and not always preserved in the record.

It's a really short chronology anyway--arrows appear in Midsouth/Midwest (O'Brein et al. are from U. Missouri) AD 700 and last into contact period AD 1600-1700. Stems, notches, serrations, whathaveyou do not vary unimodally, so arrowpoints are notoriously unsuited for fine chronology.

I think they have really gone out on a limb with a functional evolutionary argument here, but they do that regularly. They are our chief living proponents of a science-like archaeology.

I thought the whole point in using style was to find bits of variation that are useful from a chronological standpoint.

That's one point in studying style. I've done a lot of work along those lines myself. But Western European archaeologists tend to feel that the main reason to study style is to understand what it meant to people at the time.

A very simple rule could result in optimization through selection: copy the arrowhead recovered from the dead prey. Add a little variation during the copying procedure and, overtime, you have optimization. A large scale systematic comparison and accurate measures of efficacy are not necessary.

Oh how I enjoy this!
Written above;
"The only reason that our forebears' societies survived long-term was their high infant mortality. They wanted to do things that would have wrecked their habitat, but their low population densities prevented them."

Isn't this the key to shred light over the Catholic strategy? No protected sex. And starting off with the very well put sentenses above I'd like to argue that what they really are after is the same as me being a hard rockin' teen. Wreck it all! I'm sorry, I just didn't see it back then. If I had I might've been a bishop by now. ;)

Welcome to Aard