Bio-QWERTY?? sci.bio.evolution,sci.econ,alt.politics,economics cpwUhUh@Spamrahul.net wrote: [snip] >thought sprang from reading an article in the New Yorker while in a waiting >room that mentioned Microsoft and a paper written by Brian Arthur about how >a company can 'lock in' a market. It struck a chord with me and it also >struck me that there was a similarity to a biological phenomenon I had read >about in a book by Stephen J. Gould, "Wonderful Life" which talked about >the Cambrian explosion and how certain phyla of life had mysteriously >become extinct while others, which in that time, seemed no more likely to >flourish, such as the progenitors of the insects, had in fact done so. >There was speculation that some temporary advantage at a critical moment >had become permanent and I thought aha, there was some quality of systems, >whether biological or economic, at work. Someone else was bound to have >noticed and I wanted to ask who. andrew@pimlott.student.harvard.edu (Andrew Pimlott): Steven Gould is a bright guy, and AFAIK his speculation that the Cambrian explosion may have been a path-dependant "lottery" is well respected by biologists (I took a course from him last spring :) ). Yet you seem to believe that path dependance has been discredited in the economic sphere: > Well, reading about QWERTY and Microsoft in here has given me >pause. Apparently there's not so much evidence to support the >contention of what's being called 'path dependence' as I thought, People like to say this, but lets compare this situation with that of Gould's theory. Is there any direct evidence that the selection of the organisms that prospered in the Cambrian was by-and-large a random process, not correlated with the fitness of the organisms? Of course not!--prehistory only happened once, and though we have a decent record of it, the most we can say is that we can't find any strong trends in the characteristics of the organisms that succeeded. So we conclude that it is likely that evolution is a highly contincent process. The "problem" is that neither evolution nor economics is a primarily experimental science (economics probably less so than evolution). We don't get a chance to repeat events with slightly different initial conditions to see whether they come out differently. Nor do we get to observe systems with the same precision as in physics. So, it would be almost impossible to prove that a specific outcome arose by path dependance; but it's very possible to look at a set of outcomes and say that it's very likely that some were path dependant. The people who led you to believe that path dependance is a failed economic theory have tricked you into taking a very narrow view of the evidence. By a broader view (which allows for common-sense interpretation, by the way), path dependance is much more plausible. Hi, I have some thoughts on this. I agree that since life evolved only once (ON EARTH!) there is no control experiment. But someday we will have a field of "comparative evolution" or comparative paleontology that compares species and evolution on different planets around the universe, that have similar conditions. Until then there is just "common sense", life on earth, economics, and computer simulations. My guess is that "path dependence", or what I think of as the "local minimum" problem, is less of a factor in economics than it is in evolution. The Local Minimum I call it "local minimum" because think of a large table top with concave dips: some are shallow and a few are deep. Drops of water or mercury are randomly put on the table. They "want" to minimize their potential energy, so they flow down hill. Some follow a slope into a shallow dip: they then reach a relatively high minimum. To get into a deeper minimum they must flow "uphill" to get to a slope that will take them to a deeper minimum. Chance puts some drop on the slope to a deep or shallow minimum. A drop in such a shallow dip, when comparing notes with a friend who made it into a deeper hole, can see that he did not have enough foresight to skip the immediate gain and go for better long range goal. "It seemed like a good idea at the time" he laments. In biology here on earth, it seems clear to me that initial conditions play a critical role. Australia, and some Pacific islands were dominated by species that were less will adapted (even to their local conditions) than species that lived out- side. The marsupials of Australia were able to dominate only because they were in competition with other marsupials. They were in a shallow minimum. In economics, the same idea applies, but for most situations, it is much easier to jump from a higher minimum to a lower one. (It is a more "quantum mechanical" table?). This is perhaps because while a marsupial cannot decide to convert to a rodent or primate, you can change your QWERTY keyboard to a DSK, or sell your Betamax and get a VHS. People can communicate and decide to change if the invested effort is judged to be worth the gain. I can even picture a sort of Darwinian test of "path dependence". Newsgroup posters who believe in it will change their keyboards from QWERTY to the Dvorak DSK, and if they can then type 20% (or 40%?) faster as claimed, they will come to dominate the the discussions, and Path Dependence will become the accepted conventional wisdom :-) One exception I see is the situation where YOU cannot make the change unless "everyone" also does too. Like the DSK before computers. Or (my favorite): you want to ride a bike or take a bus/train/monorail/mass transit to work, but they build only to accommodate cars. You are locked into a high cost system (and stuck in traffic much of your life) and cannot move to the low cost better system unless you can convince "society" to change. I have put several articles on this topic in a QWERTY file located on my web page book review secrion next to PEDDLING PROSPERITY (where I first encounted this topic). ,,,,,,, _______________ooo___(_O O_)___ooo_______________ (_) REPLY: From: fivelongs@acsworld.net (Mitch Long Family) To: "jim blair (by way of larclark@acsworld.net (Larry Clark - Lawrence D. Clark)) " Subject: Re: Bio-QWERTY? Interesting conversation. Amazing how a linear concept like path dependence is applied laterally to economics. Thank goodness Blair brings up the point of individual choice in economics. AND THIS: mturton@ms.showtower.com.tw (Michael Turton): >... if there is no path dependence, how is it that different cultures >have different technologies, problem-solving strategies and different >rates of technical acquisition? Hi, This in an interesting point, but just which "different technologies" do you mean? I was thinking almost the exact opposite: that different cultures all over the world have developed along such similar lines that it makes me question the assumption that they were not in contact. Swords, bow and arrow, domestic animals, even building pyramids in Egypt and the New World. And the fact that all civilizations started fermenting plant sugar to make alcoholic drinks as soon as they had agriculture. The details were a little different, but to me they seem too similar to be accidental. Especially in weapons and war. There, you can't do it any old way. With the resources available, you must do it the best way. But of course some have advanced faster than others, both because of available resources and because the incentives for innovation are very different. . ,,,,,,, _______________ooo___(_O O_)___ooo_______________ (_) jim blair (jeblair@facstaff.wisc.edu) Madison Wisconsin USA. This message was brought to you using biodegradable binary bits, and 100% recycled bandwidth. Subject: Re: QWERTY, Gould, economics & nonsense Date: 26 Feb 1998 23:35:55 GMT From: rvien@see.sig.com (Robert Vienneau) Organization: Dreamscape Online Newsgroups: sci.econ, comp.os.linux.advocacy Some economists who have considered connections between economics and evolution: o Thorstein Veblen ("Why is Economics Not an Evolutionary Science") o Alfred Marshall o F. Hayek o G. Hodgson (who, I believe, is active in the European Association for Evolutionary Political Economy ) o Some contributors to a book edited by Philip Mirowski called _Natural Images in Economic Thought: Nature Read in Tooth and Claw_ There's a lot of interesting work on how history matters in economics that is not necessarily explicitly connected to biological metaphors, for example, o Paul Davidson's interpretation of Keynes' GT as being at home in a non-ergodic environment, unlike the classical special case assumption of ergodicity o Richard Goodwin's investigation of chaotic and non-linear dynamical systems o Those building on Schumpeter to understand innovations and firms. Path-dependence relates to more than individual products. Think of how city layouts are altered by decisions about the use of cars or light rail and subways. Or how whole related industries of gas stations, materials, and part manufactors, etc. are affected by a late 19th century/early 20th century initial dominance of gasoline-powered cars over electric. These issues have been discussed in the theory of economic development for a long time, e.g. by Hirschman. You might also be interested in work coming out of the Sante Fe Institute. For example, Stuart Kaufmann's _At Home in the Universe_ is a popularization of theoretical biology (which reminds me of the theory of computing automata). I don't claim to understand any of this stuff in any depth. Economics is just a hobby for me. -- Robert Vienneau