[Paleopsych] CPE: Douglas Glen Whitman: Hayek contra Pangloss on Evolutionary Systems

[Premise Checker]

Douglas Glen Whitman: Hayek contra Pangloss on Evolutionary Systems
Constitutional Political Economy, 9, 45-66 (1998)

[This article is an excellent discussion of Panglossism in evolutionary 
(and functionalist, for that matter) thinking. as well as a good summary 
of the controversy surrounding Hayek's thought. But I learn more the 
fallacies Hayek didn't commit than what he actually argued. How is 
evolutionary thinking important?

[Various comments of mine in brackets below.]

Department of Economics, New York University, New York, N.Y. 10018

Abstract. Some analysts have criticized Friedrich Hayek's theory of 
cultural evolution for implying that the rules, customs, norms, and 
institutions that emerge from the evolutionary process are necessarily 
efficient or desirable in all cases. This charge is unfounded. The present 
article defends Hayek versus his critics in two ways: First, it restates 
Hayek's own objections to the idea that cultural evolution produces 
optimal outcomes. Second, it shows, through an analogy with biological 
evolution, that Hayek's theory need not imply any such conclusion. 
Contrary to a widely held misconception, biological evolution does not 
produce organisms that are perfectly adapted to their habitats; insofar as 
cultural evolution shares common features with biological evolution, 
cultural evolution may be expected to display similar types of 
suboptimality or mal-adaptation. Insights from the theory of biological 
evolution also help to illuminate some areas of controversy with regard to 
Hayek's theory of cultural evolution, including: Hayek's advocacy of 
gradual change; the question of what selective forces drive the process of 
cultural evolution; and the alleged conflict between group selectionism 
and methodological individualism.

JEL classification: B25, B31, K40

1. Introduction

Nearly all of the political and economic doctrines of Friedrich Hayek have 
drawn heated criticism from one quarter or another, but few have attracted 
so much critique and rebuke, from authors of diverse persuasions, as his 
theory of cultural evolution. The idea that the morals, customs, habits, 
conventions, and even laws of modern civilization may owe their origin to 
a lengthy process of variation, competition, and selection has a long--and 
sometimes unsavory--history in intellectual thought, and Hayek was by no 
means its first exponent. He may, however, be credited with reviving the 
concept as a serious tool for social theory and normative judgment in the 
latter half of this century, and his most evolutionarily oriented works, 
the Fatal Conceit (1988) and Law, Legislation and Liberty (1973, 1976, 
1979), have served as a lightning rod for renewed discussion of the merits 
and flaws of evolutionary theory in the social sciences.

Among the most frequently repeated charges lodged against Hayek's theory 
of cultural evolution is that Hayek, like the Social Darwinists, has 
committed the Panglossian fallacy: he has suggested or implied that social 
evolution must necessarily produce the best of all possible worlds, a 
world in which "whatever is, is desirable," or (to put the economists' 
spin on it) "whatever is, is efficient."2 John Gray (1989: 98), for 
instance, claims that "Hayek frequently affirms that the sheer persistence 
of a tradition or a form of life suggests that it must possess some 
general utility." Martin De Vlieghere (1994: 293) characterizes Hayek as 
contending that "only those cultural attainments can survive and spread 
that are beneficial. So, the very longevity of an institution proves its 
value: : : ." According to Stefan Voigt (1992: 465, n.20), Hayek commits 
the naturalistic fallacy in his support of evolved institutions: "The 
currently existing institutions (the 'is') have emerged because they have 
been more viable than other institutions, from which Hayek concludes that 
they ought to exist."

In the economists' camp, Joseph Stiglitz (1994: 275) argues that "those 
who appeal to the evolutionary process [e.g., Hayek and Armen Alchian] 
also claim too much: There is no reason to believe that evolutionary 
processes have any optimality properties:::," and he goes on to say, "It 
seems nonsensical to suggest that we should simply accept the natural 
outcome of the evolutionary process." James Buchanan, an author usually 
friendly to Hayekian themes, nonetheless perceives Hayek as being 
adamantly opposed to all reform of evolved institutions. "We may share 
much of Hayek's skepticism about social and institutional reform, however, 
without elevating the evolutionary process to an ideal role," says 
Buchanan (1975: 194, ch. 10, n.1). "Reform may, indeed, be difficult, but 
this is no argument that its alternative is ideal." Sociologist Bjorn 
Hallerod (1992: 34) is notably less sympathetic. He argues that "Hayek 
ends up in a situation where every existing form of society is a good 
society or otherwise it would not exist," which means that Hayek must find 
even Nazism acceptable.

The critiques have been severe and sometimes overstated, but they are in 
substance correct: evolutionary systems cannot be characterized as 
unambiguously efficient or desirable (however these terms might be 
defined) in their effects. Where Hayek's critics err is in directing these 
criticisms at Hayek. Hayek's theory can be faulted in a variety of ways, 
but Panglossianism is not one of them.

My intention in this article is two-fold: first, to restate for the record 
Hayek's rejection of the idea that cultural evolution necessarily produces 
optimal results; and second, to elaborate some of the reasons why his 
theory need not imply any such thing. I will conclude by explaining how a 
better understanding of suboptimality in evolutionary systems can 
illuminate some areas of controversy that have arisen with regard to 
Hayek's theory.

The second goal will be pursued via an extended analogy with biological 
evolution. This approach may require some justification. It is my 
impression that many opponents of cultural evolution theories assume 
Panglossian implications because of a conscious or unconscious analogy 
with biological evolution, which is widely--and incorrectly--perceived as 
a process that produces optimal fitness in organisms relative to their 
habitats. Gray (1989: 98), for example, states as an objection to Hayek's 
theory, "we have nothing in society akin to the mechanism of natural 
selection of genetic accidents in Darwinian theory which guarantees the 
survival of useful social practices," as though he believes biological 
natural selection does make such a guarantee. My response, then, proceeds 
by showing that, even if the analogy between biological and cultural 
evolution is close (and the analogy does seem closer to me than many 
analysts would like to admit), biological evolution does not and cannot 
produce optimal results in all cases. Insofar as cultural evolution shares 
common features with biological evolution, it, too, will be subject to 
inefficiency.

Although Hayek often tries to distance himself from the analogy with 
biological evolution, he apparently does so not mainly because he doubts 
the analogy's validity, but because he wishes to eschew the errors of the 
Social Darwinists. Hayek repeatedly emphasizes that Darwin's theory of 
biological evolution was inspired by the evolutionary thinking of the 
moral and social theorists who preceded him (particularly David Hume, Adam 
Smith, and the other Scottish moral philosophers).3 After Darwin, Hayek 
(1979: 154) laments, "those 'social Darwinists' who had needed Darwin to 
learn what was an older tradition in their own subjects, had somewhat 
spoiled the case [for cultural evolution] by concentrating on the 
selection of congenitally more fit individuals," rather than on the 
selection of rules and practices adopted by groups. Hayek hastens to point 
out the differences between cultural and biological evolution that make it 
the case that rules and practices are far more significant than 
individuals in the process of cultural evolution. Specifically, he notes 
that "cultural evolution simulates Lamarckism"; that cultural traits can 
be acquired "from an indefinite number of 'ancestors,"' not merely from 
one's parents; that learning as a mode of transmission makes cultural 
evolution occur more quickly than biological evolution; and that cultural 
evolution is more likely to be subject to group selection (1988: 25). 
Nonetheless, Hayek recognizes that while their specific mechanisms differ, 
all forms of evolution share common features.

Although the "literal use" of Darwinian theory leads to "grave 
distortions" when focused upon individuals rather than rules, "the basic 
conception of evolution is still the same in both fields," he says (1979: 
23). Biological and cultural evolution "both rely on the same principle of 
selection: survival and reproductive advantage. Variation, adaptation and 
competition are essentially the same kind of process, however different 
their particular mechanisms, particularly those pertaining to 
propagation."4 Again, to the extent that cultural and biological evolution 
are united by kindred processes, they can be expected to exhibit similar 
characteristics, including their capacity to produce efficient and 
less-than-efficient outcomes.

In much of this article, I will be purposely vague about the definition of 
efficiency. Even within economics, efficiency has been defined in a 
variety of ways, from strict Pareto efficiency to wealth maximization. The 
standards by which the efficiency of rules and institutions are judged 
sometimes differ from the standards employed to judge efficient activity 
within given rules and institutions; for example, when Hayek speaks of the 
efficiency of rules, he usually seems to have in mind the degree to which 
rules promote the utilization of knowledge and the coordination of plans. 
Biologists typically employ the concept of "reproductive fitness," by 
which they mean the capacity of traits to increase the probability of an 
organism surviving long enough to reproduce as effectively as possible 
subject to environmental constraints. In general, all such concepts of 
efficiency are related to the idea, broadly conceived, of "doing the best 
you can given certain constraints," and fortunately, the point I wish to 
make does not require any greater specificity. I will contend that, 
whatever specific definition of efficiency may be adopted, an evolutionary 
system could not be expected to achieve it in all cases, although some 
brands of efficiency may be more easily approached than others.

2. F. A. Hayek: No Panglossian

By all indications, Hayek was fully aware of the "all's for the best" 
charges that might be leveled against his theory. He was particularly 
concerned with the tendency of some social theorists to reject all 
evolutionary theories of culture out of hand because of the errors of 
Social Darwinism. His disclaimer is therefore worth quoting at length:

Bertrand Russell provides a good example in his claim that "if 
evolutionary ethics were sound, we ought to be entirely indifferent to 
what the course of evolution might be, since whatever it is is thereby 
proved to be the best": : : . This objection, which A. G. N. Flew : : : 
regards as "decisive," rests on a simple misunderstanding. I have no 
intention to commit what is often called the genetic or naturalistic 
fallacy. I do not claim that the results of group selection of traditions 
are necessarily "good"-- any more than I claim that other things that have 
long survived in the course of evolution, such as cockroaches, have moral 
value (Hayek 1988: 27).

Nor does he claim that the products of cultural evolution should be immune 
to criticism or change; again, it is best to quote Hayek directly:

It would be wrong to conclude, strictly from such evolutionary premises, 
that whatever rules have evolved are always or necessarily conducive to 
the survival and increase of the populations following them. : : : 
Recognizing that rules generally tend to be selected, via competition, on 
the basis of their human survival-value certainly does not protect those 
rules from critical scrutiny (Ibid.: 20).

Notably, Hayek believes that the cultural selection process selects for 
survival and reproduction of groups (a questionable hypothesis that will 
be considered later), yet even by that criterion of efficiency, the 
resulting rules cannot be assumed to be efficient. It would be 
particularly odd, then, for those rules to be efficient according to some 
other standard, such as neoclassical economic efficiency or classical 
liberal value judgments.

The above quotations appear in Hayek's latest work, but they do not 
represent retrenchments in the face of criticism of Hayek's previous 
works; the same message appears repeatedly in his earlier works. In the 
Constitution of Liberty, for instance, we find Hayek admitting,

These considerations, of course, do not prove that all sets of moral 
beliefs which have grown up in a society will be beneficial. Just as a 
group may owe its rise to the morals which its members obey, : : : so may 
a group or nation destroy itself by the moral beliefs to which it adheres 
(Hayek 1960: 67).

Of course, this statement could be interpreted as merely a view of 
selectionism-in-progress, in that "bad" moral views are characterized as 
leading inevitably to their own demise. The point, however, is that Hayek 
does not perceive the process as finished: at any point in time, including 
the present day, we may find undesirable rules and customs that have not 
been weeded out by selective forces, at least not yet. Hayek never eschews 
the modification and reform of rules; he simply points out that any such 
revision of particular rules must necessarily take place in the context of 
a complex of other rules that are taken as given for the time being: "This 
givenness of the value framework implies that, in our efforts to improve 
them, we must take for granted much that we do not understand" (Ibid.: 
63).

In Law, Legislation and Liberty, Hayek again emphasizes the need for 
reform of established rules--this time in the context of a narrower 
evolutionary system, the common law.

The fact that law that has evolved in this way has certain desirable 
properties does not prove that it will always be good law or even that 
some of its rules may not be very bad. It therefore does not mean that we 
can altogether dispense with legislation (1973: 88).

Indeed, Hayek (Ibid.: 89) even admits the possibility that general 
principles of justice (embodied in the remainder of the body of law) may " 
require the revision not only of single rules but of whole sections of the 
established system of case law."

These are not the statements of a Panglossian. But neither do they suffice 
to shield Hayek's theory from the charge that it implies that whatever 
exists is the best of all possible worlds; Hayek's objections 
notwithstanding, his theory may have implications beyond his words. The 
question is, does an evolutionary theory unavoidably lead to Panglossian 
conclusions? In answering this question, we can gain insights by taking a 
closer look at the well-developed evolutionary theory of another field: 
biology.

3. The Flaws of the Adaptationist Paradigm

Evolutionary biologists have, unfortunately, contributed in part to the 
misconception that evolutionary systems must yield optimal results. 
Particularly in the early days of biological evolutionary theory, 
biologists could be found using Spencer's phrase "survival of the 
fittest," and that phrase has proved more than a little misleading. 
Biologists of the "panadaptationist" stripe have perpetuated the idea that 
all traits of all extant organisms may be construed as optimizing those 
organisms' fitness relative to the environment. Even modern biologists 
occasionally slip into this way of thinking; consider the following 
passage from biologist Ledyard Stebbins:

["Fitness" must refer to something in the world and not just as whatever 
survives, in order to avoid charges of circularity and tautology. This 
much we know from Mary B. Williams on group selection and from Mario Bunge 
in general. But I do not know how biologists do describe fitness.]

... all modern species and races of organisms have existed as successful 
populations, well adjusted to their environment, for thousands or millions 
of generations. We would expect, therefore, that all mutations that might 
improve the organism's reproductive fitness to its particular environment 
would have occurred at least once during this long period. If so, they 
would have been incorporated by natural selection in the gene pool 
(Stebbins 1977: 58).

>From statements like this one, it would be easy--but wrong--to draw 
Panglossian conclusions. Although extreme adaptationism reigned for a 
while in the biological literature, most biologists (including Stebbins) 
now reject pan-adaptationism (Vromen 1995: 95f.).

Two highly problematic assumptions are required to justify evolutionary 
theories of the pan-adaptationist variety. The first is the "t goes to 
infinity" assumption: evolutionary processes are presumed to have reached 
the ultimate result thatwould obtain if the processes continued for an 
infinite period of time. The paradigmatic example is the anecdote about 
100 monkeys (actually, just one would do) pounding on typewriters for an 
unlimited amount of time: sooner or later, one of the monkeys will type 
out the entirety of Gone With the Wind. If the t-goes-to-infinity 
assumption is taken seriously, the logic is inexorable; every combination 
of letters (or gene/trait combinations, or cultural taboos) will 
eventually appear. Everything that can happen will happen, so an 
appropriate selection mechanism will presumably capture the best of all 
possible worlds. In real-world processes, however, infinite time is never 
the case, at least not from the perspective of an analyst observing the 
products of evolution at any given point in time. From our perspective, 
evolution is an ongoing process, and we should not be surprised to find 
incomplete--and suboptimal-- adaptation. The assumption of infinite time 
bypasses considerations of process altogether.

[I love this "t goes to infinity." Happens all the time in economics. And 
it gets invoked by those who reject multi-selection evolution in biology. 
I have hounded those who use verbal arguments (which are all the worse 
when supplemented with equations: I say this as a math majopr with a 
vested interest in my speciality) against the possibility of group 
selection by drilling down the arguments till I find some mention of "the 
long run," like "in the long run no altruistic genes can survive. But by 
the same token, no groups can survive without altruists. The whole thing 
is like dividing by zero.]

Indeed, it is tempting to argue that, once infinite time is presumed, the 
optimal result is implicit in the initial conditions, in much the same way 
that the solution to a system of equations is implicit in the equations 
themselves. I will resist that temptation, however, because a second 
assumption is necessary for that conclusion: a stable, exogenous 
environment against which selection takes place. The environment is, of 
course, the standard on the basis of which adaptation (and optimality) is 
usually measured; in most theories, the environment is actually the 
selective mechanism. When the environment is stable and exogenous, the 
adaptive "target" remains fixed, and infinite time assures the process 
will eventually achieve it. But with amoving target, even infinite time 
cannot force a conclusion of optimal adaptation. As J. Maynard Smith 
(1994: 97) has pointed out, "Optimization is based on the assumption that 
the population is adapted to the contemporary environment, whereas 
evolution is a process of continuous change. Species lag behind a changing 
environment." In other words, one cannot assume perfect tracking of 
environmental changes by changes in the genome of resident species. Even 
if time were infinite, the protean nature of the environment would 
restrict the relevant adaptation time for any organism to the interval 
between changes in its environs.

The endogeneity of the environment complicates the matter even further, by 
raising the possibility that the definition of a "good" mutation may 
depend crucially upon prior mutations. The appearance of a new, desirable 
trait in a species causes changes in the environment, and those changes 
alter the selective pressures impinging on the species--possibly rendering 
other prevailing traits non-adaptive. One puzzling consequence of such a 
pathdependent process is that fitness may not be transitive: trait B might 
supersede trait A, and C supersede B, and then A supersede C (Wesson 1991: 
141). If changes in the traits of an organism can shape the environment as 
well as be shaped by it, the very idea of optimal adaptation gets murky 
because it is unclear that a steady-state relationship between organism 
and environment will always occur.

[The next paragraph is good.]

The twin assumptions of infinite time and stable environment underpin the 
usual case for optimality in evolutionary systems. When they are relaxed, 
we can understand a variety of actual phenomena in such systems as being 
non-adaptive or mal-adaptive, rather than dream up ad hoc justifications 
of how such phenomena might be optimal (as modern biologists have 
unfortunately tried to do in many cases5). This is true even when 
"adaptiveness" or "efficiency" has been defined specifically in terms of 
the environment that acts as a selective mechanism upon traits and 
organisms. That is, even if we specifically tailor our definition of 
efficiency to fit the direction of the evolutionary forces at work, we 
still cannot realistically expect perfectly efficient outcomes. A 
fortiori, we should not expect an evolutionary system to yield efficient 
outcomes with respect to some other brand of efficiency defined 
independently of the selective forces at work (except, perhaps, purely by 
coincidence).6

In what follows, I will explain some of the most widely recognized types 
of less-thanperfect adaptation in biological evolution. In addition to 
mentioning specific cases of such suboptimalities in biology, I will also 
provide some examples of how similar suboptimalities might occur in 
cultural evolution. Where possible, I draw my examples from Hayek himself. 
These examples should, however, be taken with a grain of salt: they are 
intended as suggestive, not definitive. A convincing case for why any one 
of the examples given indeed constitutes an example of suboptimal 
adaptation would probably require an article of its own.

3.1. Errors of Omission, Errors of Commission

Naturalists regularly encounter organisms with traits that defy attempts 
at explanation in terms of adaptation to prevailing environmental 
conditions. Often the best explanation for such traits comes from an 
examination of the organisms' phylogenetic histories (even though 
optimalitywould imply that current conditions alone should provide 
sufficient explanation). Apparently, selective forces are not always 
strong enough to remove all unnecessary or harmful traits from a genome in 
a finite period of time. The best examples are the so-called "vestigial 
structures" that appear in numerous species, including human beings. 
Vestigial structures in humans include the vermiform appendix (may have 
been a gizzard in our ancestors), ear muscles (needed for directional 
hearing), and caudal vertebrae (used to be a tail).7 None of these 
features provides any apparent selective advantage any longer, and 
appendices often require removal when they pose a positive danger to human 
life; they are actually mal-adaptive. These traits constitute errors of 
omission: they are features that selective forces have failed to 
eliminate.

It is not terribly difficult to imagine possible analogs in cultural 
evolution. Although Hayek often fails in his works to explain why the 
processes he describes may not always yield optimal results, he seems to 
have recognized the persistence of no-longer-adaptive traits as one 
possible reason. In Law, Legislation and Liberty, Hayek notes that mankind 
maintains multiple layers of rules, "according as [sic] traditions have 
been preserved from the successive stages through which cultural evolution 
has passed. The consequence is that modern man is torn by conflicts which 
torment him and force him into ever-accelerating further changes" (1979: 
159). Hayek harks back to the conflict between new and old rules in the 
Fatal Conceit (1988: 18f.) when he attributes the collectivist desire to 
implement altruism society-wide to a misapplication of the morals of the 
small group (which evolved very early in humanity's cultural history) to 
the extended order called civilization (whose rules developed later, and 
often in conflict with the prior set of rules).

Biology also provides various cases in which traits that would clearly be 
beneficial are conspicuously absent. Smith cites the sula bassana gannet, 
which lays only one egg at a time, even though itwould be capable of 
raising (and the environment capable of sustaining) two young at a time. A 
related gannet in very similar conditions does, in fact, lay two eggs at a 
time (Smith 1994: 98). Why, then, doesn't the sula bassana? Two answers 
seem plausible: first, that the environment has changed recently in a more 
favorable direction and the gannet's genome has not caught up yet; or 
second, such a mutation may have appeared one or more times but been 
eliminated by accident (say, because the one chick with the mutation 
happened to fall out of the nest and die before reproducing). The second 
scenario would constitute an error of commission, a case of selective 
forces accidentally eliminating a desirable trait. In either scenario, the 
fact remains that evolution has not placed all adaptive traits in the 
current genome.

Again, it is not difficult to imagine analogs in cultural evolution. Of 
course, many suggestions of "beneficial traits we haven't adopted" may be 
nothing more than the wishful thinking of social reformers or cultural 
entrepreneurs, but this observation does not mean that truly beneficial 
but unused or untried cultural traits cannot exist. Hayek admits this 
possibility with a particular example: "The institutions of property, as 
they exist at present, are hardly perfect; indeed, we can hardly yet say 
in what such perfection might consist. Cultural and moral evolution do 
require further steps if the institution of property is in fact to be as 
beneficial as it can be" (1988: 35). Some might argue that property rules 
and other customs and conventions are perfect as they are, but a belief in 
the idea of cultural evolution certainly would not warrant such a 
conclusion. There is every reason to believe that cultural evolution can 
produce errors of omission and commission just as does biological 
evolution.

3.2. Linkages and Pleiotropism

[This is a nice section, esp. for those who have forgotten their biology.]

Students of biological evolution have long been familiar with the fact 
that traits often travel together in packs, even when there is no apparent 
adaptive advantage to the traits appearing together. This may occur, for 
instance, when two or more genes are located very close to each other on a 
chromosome so that it is unlikely that they will be separated during 
crossing-over (the process in sexual cell division whereby chromosomes 
exchange sections, thus creating a greater variety of gene combinations). 
It may also occur in organisms in which crossing-over does not take place, 
such as in male fruit flies and some bacteria; in cases like these, the 
entire chromosome is the smallest unit of selection. In such situations, 
it becomes possible for non-adaptive or mal-adaptive traits to tag along 
with traits of high adaptive value, a phenomenon P. W. Hendrick calls 
"genetic hitchhiking" (Dodson and Dodson 1985: 212).

Linkages between traits may also occur when a single regulator gene (a 
gene that activates or otherwise regulates the activities of other genes) 
turns a number of genes "on" or "off" as a group. If some of those genes 
confer substantial advantages, the unfortunate effects of other genes in 
the group may be outweighed. Some biologists suspect that a small number 
of mutations in regulator genes may have yielded the vast phenotypic 
differences that separated human beings from their ape-like ancestors; as 
Robert G. Wesson (1991: 272) puts it, "Hairlessness, tender skin, and 
exceptional intelligence seem all to be parts of an evolutionary package, 
elements of which are evidently unadaptive." Similar linkages may occur 
because of a pleiotropic gene, a single gene that causes multiple effects. 
An example of a pleiotropic gene is the gene for sickle cell anemia, 
which, in addition to its well-known harmful effects, provides some degree 
of protection against malaria (Stebbins 1977: 126).

It might be argued that some linkages are unavoidable, and it is therefore 
optimal for an organism to have linked traits so long as the good 
outweighs the bad (since optimal means only the best of all possible 
worlds). This is probably true of pleiotropic genes, and possibly true for 
regulator-complexes. Linkage by proximity, on the other hand, is clearly a 
matter of historical accident. The relevant question is, do these traits 
need to be connected? Are there no other formations or combinations of 
genes that could separate good from bad effects? If the answer is no, then 
the existing situation must be considered suboptimal.

Cultural analogs leap to mind. It is clear enough that many ideas and 
practices travel in groups, even though they could theoretically be 
separated. Religions, for example, are complex structures that comprise 
multiple beliefs and mores. One might expect a religion to persist if it 
provided sufficient selective advantages to outweigh any disadvantages 
involved. On this subject, Hayek argues:

Customs whose beneficial effects were unperceivable by those practising 
them were likely to be preserved long enough to increase their selective 
advantage only when supported by some other strong beliefs; and some 
powerful or magic faiths were readily available to perform this role 
(1988: 138).

Fantastic beliefs about the nature of theworld might, therefore, piggyback 
on beneficial religious practices. Such beliefs could be disadvantageous 
because they impede the acquisition of more accurate and scientific models 
of nature, yet survive because they facilitate useful modes of behavior. 
(What constitutes a selective advantage or disadvantage in cultural 
evolution is, of course, an open question--one that will be partially 
addressed later.)

Another case of linkage in cultural evolution might arise from the fact 
that the growth of government power could have both beneficial and harmful 
consequences. As Hayek observes,

Those [governments] that gave greater independence and security to 
individuals engaged in trading benefited from the increased information 
and larger population that resulted. Yet, when governments became aware 
how dependent their people had become on the importation of certain 
essential foodstuffs and materials, they themselves endeavoured to secure 
these supplies in one way or another (1988: 44).

[Mancur Olson's stationary bandit theory fits the facts much better.]

Consequently, security of trade routes and abuse of power have tended to 
travel together, although whether they can ever be separated is an open 
question.

3.3. Evolution by Chance and Evolutionary Trends

Evolutionary change can also take place simply by chance, particularly in 
small, isolated populations. In a small population, the death of a single 
individual can have large repercussions in terms of gene frequencies. Over 
several generations, these random effects can drive out genes and reduce 
the variability of the population's genome (Sober 1994: 486). Random 
genetic changes can also accumulate over time with almost no effect, until 
a marginal mutation, such as the emergence or disappearance of a regulator 
gene, causes substantial changes to take place all at once. Most 
importantly, chance selection explains why adaptive mutations could appear 
yet fail to spread. Stephen J. Gould and Richard Lewontin (1994: 82) 
observe that "new mutations have a small chance of being incorporated into 
a population, even when selectively favored. Genetic drift causes the 
immediate loss of most new mutations after their introduction." In short, 
chance can provide the basis for the activation of complexes of linked 
genes and magnify the incidence of errors of commission.

[How big of a mole hill are these chance effects?]

A well-known, though trivial, cultural example of this phenomenon is the 
shrinking of the pool of surnames within small villages in New England, 
Wales, and elsewhere (Stebbins 1977: 127f.). When a new settlement was 
established by a small number of founders, the chance death of a single 
person could substantially reduce the frequency of the victim's surname in 
the population, even though the surname itself had no selective impact. 
Whether there exist non-trivial examples of chance selection in cultural 
processes depends in part on the level at which selection takes place. If 
group selection (as opposed to individual selection) is an actual 
phenomenon--as Hayek believed it to be--then some form of cultural drift 
should become more likely as groups become larger in size and fewer in 
number. To the extent that the entire world may be considered a single 
community, the relevant population has only a single member, and drift 
could therefore be quite dramatic. (The debate about levels of selection 
is a live issue in biological evolution as well as in cultural evolution 
that will be discussed more fully later.)

Biologists have also observed that selective processes can sometimes lead 
to the persistence and enlargement of trends; that is, a kind of 
evolutionary multiplier effect may cause the same mutation to occur again 
and again. Suppose gene A creates cellular conditions under which mutation 
B is likely to occur. Then an organism with gene A will tend to have 
progeny that carry both gene A and mutation B. If B has a selective 
advantage, then the progeny will be likely to survive and create progeny 
of their own that carry gene A and mutation B twice. And the next 
generation may have gene A and mutation B thrice. The phenomenon occurs 
because the same forces that favor a trait (mutation B) must also favor 
the genetic conditions that make the trait likely to occur in the first 
place (gene A). Possible examples include the multiplication of legs on 
the millipede and the growth of the brain in humans (Wesson 1991: 194). A 
possible example of trend persistence in cultural evolution, which seems 
in keeping with Hayek's previously cited suggestions about the abuse of 
government power, is that the forces which favor groups that solve certain 
coordination or public good problems may also favor the growth of 
institutions or attitudes that allow for these social solutions to be 
reached. (The institutions or attitudes that allow the solutions to be 
reached are analogous to "gene A"; the solutions themselves are analogous 
to "mutation B".) Selective forces may therefore reinforce cultural 
attitudes that favor an increase in social control, even though only 
specific forms of social control yield a selective advantage.

3.4. Multiple Adaptive Peaks

Finally, biologists have also recognized the possibility that an organism 
may follow multiple routes in its adaptation to an environment. It is by 
no means certain that all routes must lead to the same end point; there 
may be different end points that represent the highest adaptability of an 
organism along the different paths. Such end points are referred to as 
"multiple adaptive peaks" (Gould and Lewontin 1994: 84). Which path is 
"chosen" may depend crucially on the order in which mutations occur. A 
beneficial mutation may arise early on in the phylogenetic history of a 
species and be incorporated into its genome. Then, subsequent mutations' 
"fitness" will depend on how well they fit with the organism's new genome. 
Thus, an early mutation may place an organism on a path to one adaptive 
peak rather than another as a result of historical accident.

A number of economists have observed the evident connection between the 
idea of multiple adaptive peaks and the game-theoretic concept of a 
coordination game. (Many of these economists owe much to J. Maynard Smith, 
who pioneered the use of game-theoretic tools in biology.) Viktor Vanberg 
(1986: 93) has used game theory to offer a sharp critique of Hayek's 
theory of cultural evolution, noting that "once a coordination rule is 
established in a group, it cannot be assumed that a shift to a more 
beneficial rule can, in general, be brought about by a spontaneous, 
invisible-hand process." To put that in biological terms, switching from 
one adaptive peak to another is an extremely unlikely phenomenon, even if 
one peak is demonstrably superior to the other. If a species reaches an 
adaptive peak that is not sufficient to preserve the species in the 
relevant environment, it seems more likely that it will go extinct than 
switch to a different evolutionary path. If, on the other hand, a 
suboptimal peak is sufficient for species survival, then the species could 
persist indefinitely in a less than optimal state.

Despite Vanberg's criticism, Hayek seems to have been aware of the 
possibility of multiple adaptive peaks; indeed, Hayek's cultural 
relativism (which may seem inexplicable to those who interpret Hayek as a 
Social Darwinist) is intimately related to the concept. Hayek does not 
deny the fact that some cultures have developed in completely different 
directions from that of Western civilization and yet somehow managed to 
survive:

There are, undoubtedly, many forms of tribal or closed societies which 
rest on very different systems of rules. All that we are here maintaining 
is that we know only of one kind of such systems of rules, undoubtedly 
still very imperfect and capable of much improvement, which makes the kind 
of open or "humanistic" society possible where each individual counts as 
an individual and not only as a member of a particular group, and where 
therefore universal rules of conduct can exist which are equally 
applicable to all responsible human beings (1976: 27).

Hayek deliberately argues, therefore, from the context of the adaptive 
route taken by Western civilization, and he argues for internal 
improvement within that system. Hayek also recognizes the possibility 
that, even within a given tradition, path dependency may result in 
suboptimal consequences for particular subsets of that tradition. In the 
common law, for example, he points out that "The development of case-law 
is in some respects a sort of one-way street: when it has already moved a 
considerable distance in one direction, it often cannot retrace its steps 
when some implications of earlier decisions are seen to be clearly 
undesirable" (1973: 88). In situations like these, we find Hayek once 
again arguing for the occasional corrective reform, which would be 
unnecessary in a perfectly self-correcting (or instantaneously optimal) 
evolutionary system.8

4. Broader Implications for the Theory of Cultural Evolution

Biological evolution does not provide any justification for the belief 
that evolutionary processes necessarily lead to optimal results. But 
neither does it support the opposite conclusion, that evolutionary systems 
exhibit no desirable or efficient qualities whatsoever. Outrageously 
mal-adaptive traits have a high likelihood of being weeded out of the gene 
pool, and the organisms we observe in the natural world have clearly 
inherited remarkably sophisticated and effective structures and behaviors 
that allow them to survive and reproduce. The adaptiveness of at least a 
large number of traits observed in existing organisms has never been in 
question; what is in question is whether such traits represent the best 
solutions possible in all cases, and whether every single trait must serve 
some adaptive purpose. As Wesson has observed, "It is only necessary, 
however, that any particular characteristic be sufficiently functional to 
permit the species to survive. If there is an optimal shape of leaf for 
certain conditions of light and humidity, or webs for snaring flies, and 
so forth, most species are far from it" (1991: 154). The challenge for 
biologists, then, is to discern which traits took hold for truly adaptive 
reasons, which traits emerged for other reasons, and how such emergence 
took place.

The challenge for the evolutionary social scientist, I claim, is much the 
same. If evolutionary theory told us that all existing laws, customs, 
conventions, and mores were optimal adaptations to the conditions of human 
life, there would be little left to do but look around and describe what 
is already known to be best. But since evolutionary theory does not 
justify that conclusion, the social scientist's task is more difficult: he 
must attempt to identify which cultural norms possess truly adaptive 
qualities, which cultural norms emerged and persisted for non-adaptive 
reasons (and which may even have mal-adaptive effects), and how these 
norms came into being.

Evolutionary theory can, therefore, provide a sound basis for both 
advocacy of reform (when structures appear mal-adaptive or detrimental in 
some way) and for the defense of tradition (when the traditions seem to 
produce desirable results on net, or when they may be indispensable to the 
ongoing system as a whole). (Of course, to engage in such internal 
criticism, one would have to approve, normatively, of whatever standard of 
"efficiency" is implicit in the selective forces at work--which Hayek 
appears to do.) That we should keep "good" traditions and change "bad" 
ones might seem truistic, but some of Hayek's critics have accused him of 
inconsistency in so arguing. De Vlieghere (1994: 294), for instance, calls 
Hayek's advocacy of piecemeal reform "lip-service" because it is "in 
contradiction with his Darwinian theory" which devalues the contributions 
of reason. Similarly, Barbara M. Rowland (1987: 54) says that Hayek 
"inconsistently" draws the conclusion "that people can learn from studying 
the valuable role evolved institutions have played in advanced societies" 
so that reforms will fit smoothly into the evolved order. But as we have 
seen, Hayek's reformist and traditionalist tendencies present no 
contradiction; they are perfectly consistent if viewed from an appropriate 
evolutionary perspective.

These statements should not be taken to imply, however, that Hayek's 
theory of cultural evolution has no flaws or drawbacks. Indeed, the 
criticisms and doubts about Hayek's theory are too numerous to state 
here.9 I will instead show how the biological metaphor and a recognition 
of the possibility of suboptimality in evolutionary systems can help us to 
address some unresolved issues associated with Hayek's approach.

4.1. Gradualism

One conclusion that Hayek has drawn from his evolutionary analysis is that 
gradual or piecemeal change ought to be preferred to radical or wholesale 
change. At first, this conclusion appears to fit in with the evolutionary 
paradigm nicely. In the biological sphere, Stebbins states, "If an 
organism is well adjusted to its environment, slight changes in its 
genetic makeup may alter it better to modifications of that environment, 
but drastic alterations of one or a few characteristics are almost certain 
to make it function more poorly under any environment" (1977: 60). In 
Hayek's view, ill-advised reformers who wish to jettison rules or 
conventions whose functions are not immediately clear or whose systemic 
implications are not understood may seriously threaten the stability of an 
interdependent system. Hayek therefore advises that all reforms be judged 
within the context of a complex of other rules taken for the time being as 
given.

While all these points are well taken, Hayek's plea for gradualism cannot 
be taken as a universal rule--at least, not on the basis of evolutionary 
arguments alone. The potential existence of multiple adaptive peaks 
indicates that a system very different in all respects from the status quo 
could, conceivably, have more desirable qualities. In order for such a 
peak to be reached, radical changes might be required. For Hayek to argue 
against such wholesale reform, he must (and does) muster other arguments 
that he has elaborated elsewhere. For instance, in his case against 
socialism, Hayek might like to say that evolutionary considerations alone 
should be sufficient to relegate socialism to the dustbin of bad ideas. 
And evolutionary arguments do carry him part of the way to that 
conclusion, inasmuch as they lead the analyst to consider the functional 
properties of institutions such as several property and security of 
contract. But it is not inconceivable, prima facie, that the status quo 
represents but one of many adaptive paths. In order to make the case 
against socialism, Hayek must also rely on a variety of other tools such 
as economic theory to demonstrate that socialism could not, in fact, 
achieve the results its proponents suggest.10

Hayek also grounds his argument for gradualism on a strong epistemological 
challenge: people whose civilization has evolved along one path may simply 
lack the knowledge necessary to identify viable alternatives that differ 
substantially from the status quo. That other adaptive paths are 
conceivable does not imply that ignorant human beings can see what they 
are and implement them. To theorize that evolution could have led to a 
different and possibly superior outcome is fine, but to say precisely what 
that outcome would have been is an act of the imagination, and trying to 
realize an imaginary outcome in the real world is to engage, not in 
evolutionary theory, but in rational constructivist design. When proposed 
changes differ only marginally from the status quo, the imagination can 
(perhaps) be relied upon for some valid judgments; but in the case of 
massive system-wide changes, the demands placed on human knowledge are far 
higher.11 Itwas exactly that sort of hubristic endeavor to which Hayek 
applied the term, "the fatal conceit."

4.2. The Dual Selective Mechanism of Cultural Evolution

There is a great deal of confusion, in both Hayek's work and the 
literature on cultural evolution in general, about the exact means by 
which selection takes place in cultural evolution. It is often unclear 
whether the emergence of cultural norms is a matter of individual and 
collective choice or purely a product of impersonal environmental factors. 
An understanding of how evolutionary systems may fail to yield optimality 
can shed light on this matter.

An evolutionary system consists of two fundamental features: units of 
selection, and a selective mechanism. The selective mechanism consists of 
those forces in the system which allow for the differential survival and 
reproduction of the units of selection. Units of selection are structures 
or entities that have the capacity to replicate themselves (that is, to 
reproduce) under certain conditions. In biology, the most fundamental 
units of selection are genes (out of which higher level structures such as 
organisms and species are formed). But if genes are the most basic units 
of selection in biology, what are the corresponding units of selection in 
cultural evolution? The smallest units are, in fact, cultural traits or 
features with the capacity to be adopted, consciously or unconsciously, by 
human beings. Richard Dawkins dubbed these entities "memes." In his words,

Examples of memes are tunes, ideas, catch phrases, clothes fashions, ways 
of making pots or of building arches. Just as genes propagate themselves 
in the gene pool by leaping from body to body via sperm and eggs, so memes 
propagate themselves in the meme pool by leaping from brain to brain via a 
process which, in the broad sense, can be called imitation (Dawkins 1976: 
206).

[This is most definitely not a definition. Russell (and after him 
Wittgenstein) also thought he had a "unit" of meaning. No characterization 
of such units have been found have achieved any consensus. This does not 
render the notion useless. Uselessness comes in degrees.]

If memes are indeed the smallest such units, then the psychology and 
preferences of individuals may constitute significant selective forces, 
inasmuch as these factors determine which memes can successfully "infect" 
human minds. "We do not have to look for conventional biological survival 
values of traits like religion, music, and ritual dancing," Dawkins 
argues, "though these may also be present" (Ibid.: 214).

[Dawkins would probably agree that a simple graph is a meme. The first 
graph was drawn by Nicole Oresme in 1340, not by a Greek or Roman.]

The simplest way to grasp this point is to conceive of cultural evolution 
as a massive hitchhiker trait. The mutations that created the ability of 
the human brain to imitate, learn, and evaluate obviously had substantial 
adaptive qualities for the human species, and for that reason they tended 
to be selected. But such a brain is capable of far more than enhancing an 
organism's survival and reproduction; this sort of brain can also desire, 
imagine, and create. The complex of mutations that created the human mind 
set in motion myriad effects, only a fraction of which necessarily 
possessed biologically adaptive qualities; the rest just came along for 
the ride. Biologist Philip Kitcher observes, "All that natural selection 
may have done is to equip us with the capacity for various social 
arrangements and the capacity to understand and to formulate ethical 
rules" (1994: 440). In so doing, natural selection created the conditions 
for another kind of evolution, cultural evolution, that is only 
peripherally related to biological factors. The entire process of cultural 
evolution may be accurately characterized as a playing out of the full 
implications of a particular genetic configuration--the human brain--that 
emerged from the process of biological evolution. Consequently, human 
culture may be regarded as responding to a dual selective mechanism. On 
the one hand, the reproductive capacities of units of cultural 
transmission (memes) are subject to a selective process in terms of their 
plausibility, attractiveness, utility, and ease of imitation--as 
determined by the human minds that consciously or unconsciously adopt 
them. These standards may or may not have anything to do with the memes' 
capacity to help or hinder the survival and reproduction of human beings 
in their environments. On the other hand, which cultural traits human 
beings adopt will often have indirect impacts on human survival and 
reproduction, and natural selection of an environmental variety will 
necessarily come into play if the impacts are sufficiently positive or 
negative. I will refer to selection of the former variety as 
"psychological selection," and to selection of the latter variety as 
"environmental selection." I should note that what I am calling 
environmental selection is the sole selective mechanism at work in 
biological evolution, while both forms of selection are at work in 
cultural evolution.12

Like any evolutionary process, cultural evolution does not exhibit a 
strictly linear chain of causality. The feedback generated by selective 
forces (in this case, psychological and environmental selection) means 
that the reason a trait comes into being may differ from the reason a 
trait persists. Cultural traits come into being because humans are 
equipped with brains capable of imagining and conceiving of different 
rules, practices, and ideas. But of the many cultural traits that may come 
into being, only some will survive both psychological and environmental 
selection. (Similarly, in biological evolution many traits can come into 
being via mutation and recombination, but only some will survive the 
process of selection.)

The two selective mechanisms of cultural evolution need not always work in 
the same direction. Sometimes they will reinforce each other; other times 
they may conflict. It is even possible that some cultural traits may run 
contrary to the apparent demands of environmental selection, because of 
the overwhelming influence of psychological factors. Cavalli-Sforza and 
Feldman provide a fascinating example: the decline of birth rates among 
European women after the onset of industrialization. If child-bearing were 
a purely genetic disposition passed from mother to daughter, a disposition 
to limit one's child-bearing would tend to die out in an environment 
wherein raising more children were possible. Those mothers with a 
disposition to bear more children would pass that disposition to more 
future mothers, while those with a disposition to bear fewer children 
would pass that disposition to fewer future mothers. The fact that the 
European birth rate diminished indicates that some form of cultural 
transmission of dispositions had to have been at work; only then could 
horizontal (intra-generational) and oblique (from one generation to 
non-offspring members of the next) transmission of dispositions have taken 
place. That is, the ability of European women to learn a new disposition 
rather than inherit an old one made a drop in the birth rate possible.13

But how could this environmentally mal-adaptive meme have survived the 
process of environmental selection? The answer lies in recognizing the 
dual nature of cultural selection. Evidently, the disposition to limit 
one's pregnancies became psychologically (and financially) appealing to 
women after industrialization took place in Europe; as a result, the 
disposition to have more children suffered from a magnified "death" rate, 
since large numbers of women were abandoning it. Successful memes must 
survive at this psychological level of selection before the environmental 
level of selection can even become operative. On the environmental level, 
the disposition to restrict one's pregnancies might have been expected to 
lead to its own demise if its net effect had been to reduce the number of 
the disposition's adherents in each generation. But the European 
population was still, on the whole, rising because of improvements in 
sanitation, food provision, and other factors. As a result, a meme 
yielding lower birth rates was enabled to survive despite the selective 
advantage of higher birthrates in the context of environmental 
selection.14

Perceiving cultural evolution as responding to a dual selective mechanism 
allows the idea of spontaneous order, a concept to which Hayek devoted a 
considerable amount of attention, to be more fully integrated into the 
theory of cultural evolution. A spontaneous order such as a market-based 
economic system does not respond to or serve the specific, unitary ends of 
a society; rather, it serves the multiplicitous and largely unknown ends 
of all the individuals whose transactions create the order. This sort of 
order is an "abstract order of the whole which does not aim at the 
achievement of known particular results but is preserved as a means for 
assisting the pursuit of a great variety of individual purposes" (Hayek 
1976: 5). It is not at all clear why an order that serves individuals' 
multifarious purposes should survive in an evolutionary system in which 
survival and reproduction of groups is the only criterion for natural 
selection (as Hayek sometimes implied). One of the advantages of a 
spontaneous order is its capacity to mobilize information that is 
dispersed among many individuals in the order. Among the pieces of 
information transmitted (or summarized) by this sort of order, in addition 
to information about technologies and resource supplies, are the 
subjective tastes and preferences of the participating individuals. The 
need for such information would be inexplicable if group survival and 
reproduction were the only selective forces at work; the tastes of 
individuals would be precisely irrelevant. It is only because there is 
another type of selection involved--the satisfaction of the psychological 
demands of human minds--that information about tastes and preferences 
might be relevant to an adaptive process.

4.3. Group Selection and Methodological Individualism

In explaining his theory of cultural evolution, Hayek embraces the concept 
of group selection: the idea that cultural traits and behavioral features 
are naturally selected on the basis of advantages and disadvantages they 
create for the groups of people who practice them. A number of authors 
have found Hayek's group selectionism troubling, and Vanberg (1986) argues 
that group selection conflicts with Hayek's professed methodological 
individualism. Since the idea of group selection is "theoretically vague, 
inconsistent with the basic thrust of Hayek's individualistic approach, 
and faulty judged on its own grounds," Vanberg (1986: 97) contends that 
group selection ought to be jettisoned to save methodological 
individualism. Geoffrey Hodgson (1991) agrees with Vanberg that there is a 
conflict between the two doctrines, but recommends instead that 
methodological individualism should be abandoned (or at least modified) in 
order to keep group selection. Some of the insights from the foregoing 
discussion of the dual selective mechanism of cultural evolution may help 
to resolve the Vanberg-Hodgson debate.

Vanberg defines methodological individualism as "the guiding principle 
that aggregate social phenomena can be and should be explained in terms of 
individual actions, their interrelations, and their--largely 
unintended--combined effects" (Vanberg 1986: 80). Group selection 
conflicts with methodological individualism, Vanberg argues, because it 
attempts to explain cultural norms in terms of the functional roles they 
play for groups rather than their emergence through individuals' behavior. 
He proceeds to argue that group selection is a troublesome and flawed 
concept even in biology, because it is unclear how "altruistic" behavior 
patterns that benefit groups could possibly survive in the presence of 
selective pressures that favor "selfish" behavior by individuals. Vanberg 
says that it seems to be the "dominant opinion among biologists" that the 
conditions necessary for true group selection "rarely exist in nature" 
(Ibid.: 69). Interestingly, Vanberg also maintains that methodological 
individualism was a factor in the development of the theory of biological 
evolution because it supported a shift "from the species as the 
theoretical unit to the individual organism as the central unit of 
analysis" (Ibid.: 80).

[I have been beating a drum for group selection in biology for a decade, 
even before Eliot Sober and David Sloan Wilson's _Unto Others_ came out. 
Wilson (alone) went on to apply the group selection idea to human religion 
and to revive functionalism in sociology, in _Darwin's Cathedral. 
Functionalism had gone by the wayside in sociology as much as group 
selection went in biology. It was Stephen Sanderson's _The Evolution of 
Human Sociality: A Darwinian Conflict Perspective_ (a terrific book and 
the first one to import sociobiology into sociology), or rather his 
dismissal of functionalism that turned me into a functionalist! He argued 
that functionalism implies the existence of equilibrating mechanisms, 
which would restore the social system back to where all the parts worked 
together. Well, there *are* these mechanisms. Sanderson was criticizing 
*pan*-functionalism. In the process, he convinced me that there must be 
something to a less thoroughing version of it. And _Darwin's Cathedral_ 
reached the same result.]

Hodgson argues cogently that Vanberg has misconstrued the biological 
literature on the debate over units of selection. The biological 
"reductionists" on whom Vanberg relies for support do not contend that the 
individual organism is the most basic unit of selection in the 
evolutionary process. Reductionists like Richard Dawkins contend, on the 
contrary, that selective forces ultimately operate on the smallest units 
of selection, genes. The misleadingly labeled group selectionists, on the 
other hand, argue that natural selection operates on higher level 
structures as well. Genes come in complex groups called individual 
organisms, organisms come in groups called populations, populations comes 
in groups called species, and so on; and all of these structures, the 
group selectionists believe, may be subject to weeding and culling by 
evolutionary forces. "In other words," Hodgson says, "selection operates 
simultaneously on different types of unit, depending on the time-scale and 
the type of selection process" (1991: 69). The real debate in biology, 
then, is not selection of individuals versus selection of groups, but 
selection of genes versus selection at multiple levels of a hierarchy. To 
the extent that Vanberg relies on the support of biological reductionism 
to support methodological individualism, his argument collapses because 
there is no particular reason to focus on individuals. "Simple reduction 
to the individual level is unacceptable because the same arguments 
concerning reduction from groups to individuals apply equally to reduction 
from individual to gene. To avoid this double standard, one must either 
accept multiple levels of selection, or reduce everything to the lowest 
level [i.e., genes] in the manner of Dawkins : : : and Williams" (Ibid.: 
71).

Although Vanberg's use of reductionist argumentation is vulnerable to 
Hodgson's critique, the case for methodological individualism is stronger. 
In most of his analysis, Vanberg implicitly portrays individuals as units 
of selection in the evolutionary process. If this were the theoretical 
basis for methodological individualism, then methodological individualism 
would indeed be threatened by Hodgson's clarification of the 
levels-of-selection debate. But the earlier discussion of the dual 
selective mechanism of cultural evolution suggests that the individual 
human is not merely a unit of selection; the individual human is actually 
part of the selective mechanism that influences the survival and 
reproduction of cultural traits (or memes). And it is this fact, I will 
argue, that is crucial in the case for methodological individualism. In 
addition, it dissolves the alleged conflict between methodological 
individualism and group selection, allowing the concepts to co-exist in 
the same theory.

For the social scientist interested in the process of cultural evolution, 
the relevant explananda are the cultural norms (including beliefs, rules, 
behavioral regularities, and institutions) that emerge from that process. 
In order to understand why some memes have survived and prospered while 
others have grown rare or disappeared, he must direct his attention to the 
selective forces that have imposed differential death rates on various 
cultural practices and beliefs. That means asking, first and foremost, how 
and why some practices and beliefs were adopted in the first place by 
human beings and others were not. In other words, it is necessary to 
enquire into the effects of psychological selection, the first prong of 
the dual selective mechanism. Then, the analyst must explore the systemic 
effects that would result from the adoption of certain norms. Such effects 
might include changes in the constraints that influenced individuals' 
adoption of those norms in the first place, in which case another round of 
psychological selection could occur, and the same process could be 
iterated indefinitely. The systemic effects of norms might also include 
changes in the capacity of individuals and groups to serve their 
physiological needs, resulting in population growth or population loss; 
that is, the second, environmental, prong of the dual selective mechanism 
could come into play.

It might appear that allowing for two selective mechanisms, instead of 
just psychological selection, represents a break from methodological 
individualism. But there is no contradiction here: the tenets of 
methodological individualism do not require that social phenomena be 
explained without reference to the constraints that impinge on 
individuals' actions. If environmental constraints affect the survival of 
individuals (and the groups composed of them) in such a way that the norms 
they practice and the things they believe have a reduced probability of 
being absorbed by other individuals (either outsiders or subsequent 
generations), then the environmental prong of the dual selective mechanism 
is consistent with a methodological approach that explains social outcomes 
in terms of the actions, choices, and behaviors of individuals.

[Just what is inconsistent with "methodological individualism"? I am now 
thoroughly confused.]

Notably, in this account individuals are not units of selection upon which 
selective forces operate, except insofar as an individual may be perceived 
as a conglomeration of multiple memes and genes. What is essential for a 
methodologically individualist account of the evolution of cultural 
outcomes is that individuals constitute a filter (i.e., a selective 
mechanism) through which memes must pass before they can begin to have 
systemic effects. Vanberg is correct to chastise Hayek for giving too 
little attention to this filtering process in his later work: Hayek 
regularly refers to the unexpected prosperity of groups that "happened to 
change them [cultural rules] in a way that rendered them increasingly 
adaptive" (Hayek 1988: 20) while giving little detail about how the 
individuals in those groups might "happen" to adopt such changes. Vanberg 
is also correct, therefore, to draw attention to the question of how, for 
instance, groups of individuals might happen upon appropriate rules for 
escaping Prisoners' Dilemma-type situations. It is also clear, however, 
that groups that did--somehow--find solutions to that kind of dilemma 
(e.g., tit-for-tat or "grudger" strategies) would create advantages for 
their members over the members of other groups that did not discover 
similar solutions. In other words, if a set of beneficial social rules can 
survive the gauntlet of psychological selection, then groups of 
individuals who adopt those rules will be favored by environmental 
selection. It is worth pointing out that the psychological gauntlet may 
not be as difficult to clear as Vanberg suggests, since individuals may be 
guided as much by an instinct to imitate as by rational optimization. 
(Hayek contends that that kind of rationality is a product, not a 
predecessor, of cultural evolution (Ibid.: 21).) Of course, any strategy 
that survived psychological selection would still have to be capable of 
surviving environmental selection as well. (That is, it would have to be 
an "evolutionarily stable strategy," to borrow J. M. Smith's terminology.)

[Most people give little or no thought to what they do. They are no 
obsessive Premise Checkers.]

Finally, I should be explicit about howthis discussion relates to the 
issue of group selection. Without necessarily agreeing with the group 
selectionist hypothesis, it is easy enough to see that group selection is 
at least not incompatible with methodological individualism, once it is 
recognized that methodological individualism does not depend upon 
individual organisms being the (sole) unit of selection. With the 
methodological issue out of the way, the debate between Vanberg and 
Hodgson largely disappears. Like the biologists from whom they draw 
support for their respective positions, Vanberg and Hodgson apparently 
agree that group selection is a conceivable phenomenon; they merely 
disagree about its empirical relevance in the world. Opinion on this 
matter seems to have converged on the position stated by Sober:

Group selection acts on a set of groups if, and only if, there is a force 
impinging on those groups which makes it the case that for each group, 
there is some property of the group which determines one component of the 
fitness of every member of the group.15

There remains a debate as to how often these conditions hold, in both 
biological and cultural evolution. But Vanberg's coordination games and 
Prisoners' Dilemmas present fine examples of how these conditions could, 
at least in principle, apply to certain realworld situations faced by 
human beings. There seems to be some basis, therefore, for Hayek's focus 
on group selection in his evolutionary theory.

5. Concluding Remarks

The critics of theories of cultural evolution have often chided cultural 
evolutionists for their alleged belief that "whatever is, is desirable." 
Although some theorists of cultural evolution (like the Social Darwinists) 
have in fact reached such conclusions, Friedrich Hayek was not one of 
them. Repeated statements by Hayek indicate that he did not regard 
cultural evolution as a perfect process.

["Social Darwinism" is a social construction of 20th century historians. 
None called themselves that at the time. Likewise, there are no 
self-styled "neo-Nazis." Sometimes, though, an enemy's label gets worn as 
a badge of pride. Methodism is the best-known example.]

Nor does an evolutionary approach justify or imply such a conclusion. The 
well-developed field of biological evolution provides innumerable examples 
of how an evolutionary process may fail to produce perfectly adapted 
organisms. The assumptions of infinite time and constant environments 
could sustain the idea of perfect adaptation, but these assumptions are 
untenable. In a real-world evolutionary system, whether of the biological 
or cultural variety, one should therefore not be surprised to find errors 
of omission and commission, "hitchhiker" traits, chance selection, trend 
persistence, and path dependence.

Indeed, such "suboptimal" phenomena in the phylogenetic history of mankind 
may be responsible for the very existence of cultural evolution. Trend 
persistence and chance, as well as adaptive selection, led to the 
formation of a powerful human brain capable of imitation, learning, and 
cognitive thought. That brain produced multiple effects, only some of 
which could be considered adaptive on a purely biological level. The other 
traits merely tagged along, and among those traits was the capacity for 
desires and preferences--often for things with no discernable adaptive 
value whatsoever, such as fine art and literature. The very persistence of 
cultural traits that are non-adaptive (or even mal-adaptive, in the sense 
of counteracting the demands of environmental selection) constitutes a 
fantastic error of omission; human beings are constantly engaged in a 
multitude of costly, energy-consuming activities that add nothing to the 
reproductive fitness of the species. The species can remain in existence 
because the biological advantages of having powerful brains--such as 
providing food, shelter, and clothing--are sufficient to justify the 
biological burdens of having those brains.

Those burdens include the vast majority of what we call "culture" (and few 
people would consider them burdens in a pejorative sense of the word). The 
process of cultural evolution may usefully be treated as responding to two 
masters. One is environmental selection, meaning the process by which 
certain cultural traits may lead to the demise or proliferation of those 
who hold them because they inhibit the production of food, cause the 
population to shrink, etc. The other is psychological selection, meaning 
the process by which some cultural traits dwindle and others spread 
because of their appeal, utility, plausibility, and capacity for imitation 
by human minds. Both sets of selective forces are, of course, highly 
imperfect; both are subject to all of the adaptive limitations imposed by 
finite time, trait linkage, path dependence, and so on.

When a two-fold selection criterion is fully and explicitly incorporated 
into Hayek's theory of cultural evolution, the theory can more easily be 
squared with Hayek's theory of spontaneous order. The idea of a dual 
selective mechanism also provides a ready defense against the charge that 
his theory conflicts with the principles of methodological individualism. 
Stripped of all Panglossian implications, real or imagined by critics, 
Hayek's theory of cultural evolution may provide a powerful tool for the 
analyst searching for a critical theory of social development and the 
growth of institutions.

[This strikes me as exactly right, namely that combining group selection 
with spontaneous order will result in a terrific way of looking at social 
institutions. Don't forget there are other constraints, even those imposed 
by physics. There's a fine little book, C.J. Pennycuick, _Newton Rules 
Biology: A physical approach to biological problems_ (Oxford UP, 1992) 
that should probably be read by all.]

Notes

1. The author wishes to thank Roger Koppl, Mario Rizzo, an anonymous 
referee, and participants at the Austrian Economics Colloquium at New York 
University for their useful comments and suggestions.

2. In Voltaire's novel Candide, the eminent Dr. Pangloss maintained that 
we live in the best of all possible worlds. "It is proved," he said, "that 
things cannot be other than they are, for since everything is made for a 
purpose, it follows that everything is made for the best purpose." 
(Voltaire 1947 [1759]: 20).

3. Hayek (1960: 59); Hayek (1973: 23); Hayek (1979: 154); Hayek (1988: 
23f.).

4. Hayek (1988: 26). Indeed, Hayek argues that the same principles are 
applicable to the study of all complex orders: "We understand now that all 
enduring structures above the level of the simplest atoms, and up to the 
brain and society, are the results of, and can be explained only in terms 
of, processes of selective evolution: : : " Hayek (1979: 158).

5. See Gould and Lewontin (1994: 78f.).

6. Examples of types of "efficiency" defined independently of the 
selective forces at work might include conformity to an aesthetic 
standard, or consistency with an ideological viewpoint such as classical 
liberalism.

7. Dodson and Dodson (1985: 213).

8. The fact that detrimental path dependence is possible in an 
evolutionary system does not necessarily mean it is common. Some of the 
most famous examples of detrimental path dependence in economics, such as 
the alleged inferiority of the QWERTY keyboard, have turned out to be 
unfounded. See Liebowitz and Margolis (1990).

9. See Kley (1994) for examples.

10. This does not mean that a socialist system could not be created in the 
first place, only that it could not work the way its proponents suggest it 
would. As noted earlier (in section 3.1), an evolutionary system is 
capable of a form of retrogression when no-longer-adaptive traits have 
been superseded but not weeded out. Hayek attributes the collectivist 
impulse behind socialist schemes to a misapplication of small group morals 
to the extended order that evolved later.

11. The Eastern European economies that are attempting to transform 
themselves into market economies after the socialist experiment may face 
similar problems of trying to implement a "jump" from one path to another. 
They have the advantage, however, of knowing from observation of existing 
market economies that a market economy is at least possible (an advantage 
not shared by the socialists early in this century, who tried to engineer 
a jump to a purely hypothetical socialist economy).

12. My distinction between psychological and environmental selection 
parallels Cvalli-Sforza and Feldman's distinction between "cultural" and " 
Darwinian" selection, which they define as follows: ": : : cultural 
selection refers to the acquisition of a cultural trait, while Darwinian 
selection refers to the actual test by survival and fertility of the 
advantages of having or not having the trait" (Cavalli-Sforza and Feldman 
1981: 16). I have chosen not to adopt their terminology because their use 
of the word "cultural" might be misleading. I use the word "cultural" to 
refer to all traits that are not transmitted genetically, and I use " 
psychological" and "environmental" to refer to the selective forces that 
impinge on cultural traits.

13. See Sober (1994: 482-4).

14. It has been suggested to me that simple cost-benefit analysis would be 
sufficient to explain the drop in European birth rates. But this 
explanation begs the question: the whole issue is which costs and benefits 
may be considered. The environmental (i.e., strictly biological) costs and 
benefits clearly pointed toward more child-bearing (since better 
sanitation, food, etc., made children easier and cheaper to sustain). A 
lower birth rate could only have arisen from "cost-benefit analysis," 
then, if some psychological costs and benefits could also come into play.

15. Quoted in Hodgson (1991): 70.

References

[The anthology by Elliot Sober is comprehensive, often too specialized for 
me, and cheap considering its thickness. These are articles that should, 
as Mr. Vining used to say, be read "bolt upright in a hard chair.]

Bradie, M. (1994) "Epistemology from an Evolutionary Point of View." In: 
Sober, E. (ed.) Conceptual Issues in Evolutionary Biology, 2nd ed., 
Cambridge, Mass.: The MIT Press.

Buchanan, J. M. (1975) The Limits of Liberty: Between Anarchy and 
Leviathan. Chicago, Ill.: University of Chicago Press.

Cavalli-Sforza, L., and Feldman, M. (1981) Cultural Transmission and 
Evolution: A Quantitative Approach. Princeton, N.J.: Princeton University 
Press.

Dawkins, R. (1976) The Selfish Gene. New York: Oxford University Press.

De Vlieghere, M. (1994) "A Reappraisal of Friedrich A. Hayek's Cultural 
Evolutionism." Economics and Philosophy 10(2): 285-304.

Dillon, L. S. (1978) Evolution: Concepts and Consequences, 2nd ed. St. 
Louis: The C.V. Mosby Company.

Dodson, E. O., and Dodson, P. (1985) Evolution: Process and Product. 
Boston: Prindle, Weber & Schmidt.

Eldredge, N. (1985) Unfinished Synthesis: Biological Hierarchies and 
Modern Evolutionary Thought. New York: Oxford University Press.

Gould, S. J., and Lewontin, R. C. (1994) "The Spandrels of San Marco and 
the Panglossian Paradigm: A Critique of the Adaptationist Programme." In: 
Sober, E. (ed.) Conceptual Issues in Evolutionary Biology.

Gray, J. (1989) Liberalisms. London: Routledge.

Hallerod, B. (1992) "Friedrich August von Hayek--Apostle for Freedom?" 
Sociologisk Forskning 29(3): 12-34.

Hayek, F. A. (1960) The Constitution of Liberty. Chicago, Ill.: University 
of Chicago Press.

Hayek, F. A. (1988) The Fatal Conceit: The Errors of Socialism. Chicago, 
Ill.: University of Chicago Press.

Hayek, F. A. (1973) Law, Legislation and Liberty, vol. 1: Rules and Order. 
Chicago, Ill.: University of Chicago Press.

Hayek, F. A. (1976) Law, Legislation and Liberty, vol. 2: The Mirage of 
Social Justice. Chicago, Ill.: University of Chicago Press.

Hayek, F. A. (1979) Law, Legislation and Liberty, vol. 3: The Political 
Order of a Free People. Chicago, Ill.: University of Chicago Press.

Heath, E. (1992) "Rules, Function, and the Invisible Hand: An 
Interpretation of Hayek's Social Theory." Philosophy of the Social 
Sciences 22(1): 28-45.

Hodgson, G. M. (1991) "Hayek's Theory of Cultural Evolution: An Evaluation 
in the Light ofVanberg's Critique." Economics and Philosophy 7: 67-82.

Kitcher, P. (1994) "Four Ways of 'Biologizing' Ethics." In: Sober, E. 
(ed.) Conceptual Issues in Evolutionary Biology.

Kley, R. (1994) Hayek's Social and Political Thought. Oxford: Clarendon 
Press.

Liebowitz, S. J., and Margolis, S. E. (1991) "The Fable of the Keys." 
Journal of Law and Economics 33 (April): 1-25.

Rowland, B. M. (1987) Ordered Liberty and the Constitutional Framework: 
The Political Thought of Friedrich A. Hayek. New York: Greenwood Press.

Smith, J. M. (1994) "Optimization Theory in Evolution." In: Sober, E. 
(ed.) Conceptual Issues in Evolutionary Biology.

Sober, E. (ed.) (1994) Conceptual Issues in Evolutionary Biology, 2nd ed., 
Cambridge, Mass.: The MIT Press.

Sober, E. (1994) "Models of Cultural Evolution." In: Sober, E. (ed.) 
Conceptual Issues in Evolutionary Biology.

Stebbins, G. L. (1977) Processes of Organic Evolution. Englewood Cliffs, 
N.J.: Prentice-Hall, Inc.

Stiglitz, J. (1991) Whither Socialism? Cambridge, Mass.: The MIT Press.

Tomlinson, J. (1990) Hayek and the Market. London: Pluto Press.

Vanberg, V. (1986) "Spontaneous Market Order and Social Rules: A Critical 
Examination of F. A. Hayek's Theory of Cultural Evolution." Economics and 
Philosophy 2: 75-100.

Voigt, S. "On the Internal Consistency of Hayek's Evolutionary Oriented 
Constitutional Economics--Some General Remarks." Journal des Economistes 
et des Etudes Humaines 3(4): 461-76.

Voltaire (F.-M. Arouet) [1947 (1759)] Candide. Translated by John Butt. 
Penguin Books.

Vromen, J. J. (1995) Economic Evolution: An Enquiry into the Foundations 
of New Institutional Economics. London: Routledge.

Wesson, R. G. (1991) Beyond Natural Selection. Cambridge, Mass.: The MIT 
Press.