DOES ETHNIC CONFLICT PAY?
Paul H. Rubin
Department of Economics
Emory University
Atlanta, GA 30322-2240
Voice: 404-727-6365
Fax: 630-604-9609
Email: prubin@emory.edu
http://www.Emory.edu/COLLEGE/ECON/Rubi.htm
March 7, 2000
The author would like to thank three anonymous referees for particularly helpful and
extensive comments.
DOES ETHNIC CONFLICT PAY?: Abstract
It is often argued that ethnic conflict is an extreme example of nepotism, and is
genetically based. This may be so: in the EEA (the evolutionary environment) such
conflict may have been fitness improving, and we may be descended from those who
participated successfully in such conflicts. This would provide us with a “taste” for
xenophobia. But this taste can relatively easily be overcome, as shown by the changes in
behavior in the U.S. in the 50 years since racial segregation was outlawed. Moreover, in
today’s world, such conflict does not provide benefits. There are several reasons for this,
but the most important (and one that is often overlooked, even by evolutionists) is the
possibility of gains from trade in exchanges between ethnic groups. While ethnic
relations in the EEA may have approximated a zero sum game, today a prisoner’s
dilemma is a more appropriate model for interactions, so that there are significant gains
from cooperation. If we want to reduce the amount of conflict in the world, it is probably
better to rely on increasing gains from trade than on increased size of in-groups, since the
latter strategy will reach a natural limit.
DOES ETHNIC CONFLICT PAY?
INTRODUCTION: ETHNIC CONFLICTS
As I write this, there are ethnic conflicts in at least Ireland, the Middle East,
Kashmir, Burundi, and the former Yugoslavia.1 When you read this, many of these
conflicts will still be active, and others may be as well. Conflict between ethnic groups
seems to be a perennial part of the human condition. Because members of the same
ethnic group are more closely related genetically than are non-members, it may be that
ethnic conflict is an evolved human behavior, an extreme form of nepotism. That is, it
may be that by engaging in conflict that is perceived to benefit one’s ethnic group, an
individual is also providing benefits to copies of his genes that reside in members of the
group. If so, this would be an example of kin selection. (van den Berghe, 1981, is an
advocate of this position.) Indeed, politicians induce citizens to join military efforts by
treating individuals as relatives and conflicts as ethnic conflicts (Johnson, 1986.) Even if
all members of the ethnic group are distant relatives with relatively few genes in
common, there may be so many of them that the net effect on one’s genes of some action
that benefits other members of the ethnic group would be beneficial. Then ethnic conflict
would be a remnant of conflicts in the Environment of Evolutionary Adaptedness (EEA),
the period in which our ancestors evolved from an ape like creature to become modern
humans (Barkow, Cosmides and Tooby, 1992). This would be consistent with the
argument that struggles and conflicts between groups have been an important part of
human evolution (Alexander, 1987; Shaw and Wong, 1989; Ghiglieri, 1999).
But even if this ethnic conflict was rational and provided benefits in the EEA, this
does not mean that ethnic conflict actually provides these benefits to individuals today.2
While there are some similarities between the EEA and contemporary conditions
(Crawford, 1998) there are obviously many differences. Therefore, the taste for ethnic
conflict, which might have been useful in the EEA, may now be counterproductive and
may not provide net benefits; see also Goetze, 1998, and Eibl-Eibesfeldt (1998) for
1
I follow van den Berghe (1981) by defining “ethnic” situationally, ranging from small family based
groups (in the EEA) to some large nation-states today.
2
There is no doubt that ethnic conflict has harmful effects on society, where society is defined as all
humans. However, in this paper I am arguing that those engaged in the conflict themselves may lose from
it. In terms used by economists, I argue that this conflict may be privately inefficient; there is no doubt that
it is socially inefficient.
ETHNIC CONFLICT
PAGE 2
similar arguments. This is true whether the benefits are viewed as economic or as fitness
benefits, which are sometimes identical but sometimes not (Rubin, 2000b). Under either
reading, some or most ethnic conflicts in today’s world are likely harmful and
counterproductive to the participants as well as to others. That is, ethnic conflict may not
add to the expected wealth of the participants, and it might not increase their expected
biological fitness and number of genes transferred to the next generation. Moreover,
ethnic conflicts are not inevitable: even if humans have a taste or preference for such
conflict, they can learn that this taste is counterproductive. That is, they can be shown
that the relative price of engaging in this preference is too high and that the conflict will
provide no benefits. In any event, there is some evidence that such tastes are rather weak.
When individuals have relatively more power over government – that is, when
government ids democratic – then conflict of all sorts (including ethnic conflict) is
greatly decreased relative to frequencies under other forms of government (Rummel,
1994).
The level of ethnic conflict can clearly be reduced within a society. The U.S. is
an example: this is a multiethnic society and, while there is some residual ethnic conflict
and some small groups of racists still exist, by and large the level of ethnic conflict is
quite small here. The level of such conflict has been greatly been diminished in a
remarkably short (perhaps 50 year) time period, as dated from the Supreme Court
decision outlawing school segregation.3 Clearly evolution was not involved in this
change; learning and changes in relative social prices were the driving forces. That is,
discrimination once led to rewards and discriminators gained status or other benefits from
their discrimination. Today, those who engage in this behavior lose status, and are
punished in other ways.
Moreover, ethnic groups in the U.S. are much more diverse (much less genetically
related) than groups in those countries where there is significant ethnic conflict. In
Yugoslavia, for example, Serbs, Croats, and Muslims are all Slavs, and the differences
between them are primarily religious, not ethnic or genetic; Cavalli-Sforza et al.
(1994/1996) classify “Yugoslovian” as one genetic group. In the Middle East, Jews and
Arabs are all Semites; Cavalli-Sforza et al. (1994/1996) indicate that the Israelites
3
Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954).
ETHNIC CONFLICT
PAGE 3
(ancestors of modern Jews) were originally one of many Semitic-speaking tribes.
Residents of Bengladesh are of the same group as some Indians and Pakistanis are more
closely related to some Indians than are some other Indians (Cavalli-Sforza et al.,
1994/1996); the difference between these groups is again religion. On the other hand, the
U.S. has large groups of people of African descent, people of Germanic extraction (from
Britain and Scandinavia as well as Germany), Mediterranean peoples (Italians, Greeks),
Semites (Jews and Arabs), Asians (Indians, Chinese, Koreans), Slavs, and members of
many other ethnic groups as well. This diversity is in part due to the greatly increased
mobility of contemporary humans, as stressed by Goetze (1998). If ethnic conflict were
aimed at improving one’s genes chances relative to other genes, the U.S. would be a
much more fertile ground for such conflict than places where it actually occurs because
there are many more unrelated individuals with whom to engage in conflict.
This is an example of a genetic preference responding to costs and benefits and
perhaps to learning. In the U.S. there are benefits from ignoring ethnic differences and
costs of taking them into account, and people learn about and respond to these benefits
and costs. Indeed, we have gone from a society with official state sanctioned
discrimination against blacks to a society with official pollicies such as affirmative action
aimed at benefiting this minority and others. That is, the political process has gone from
active discrimination against minorities to discrimination in their favor.4 Thus, although
there may be evolved preferences, we as humans are able to learn to change behavior if
prices (broadly defined to include all costs and benefits of an action) change. This is an
example of what Sober and Wilson (1998) call “facultative behavior” meaning adaptation
of behavior to changes in the costs and benefits of this behavior. This is the essence of
what is studied by economists. Indeed, I will argue below that there are often benefits
almost everywhere from ignoring the ethnicity of people, and the puzzle is that some
others have not realized these gains.
In what follows I first discuss one crucial difference between the EEA and the
contemporary world: the EEA was likely very close to a zero-sum society (at least with
respect to potential gains from trade between groups), and today there are actual lor
4
This discussion should not be interpreted as an argument in favor of discrimination either for or against a
group. Both may be harmful; see Eibl-Eibesfeldt, 1998.
ETHNIC CONFLICT
PAGE 4
potential large gains from trade. That is, in the EEA ethnic groups competed with little
possible gain from cooperation. Today, ethnic groups may be viewed as playing a variant
of a prisoner’s dilemma game. The strategies are cooperate (that is, engage in trade and
exchange) or fight (engage in ethnic conflict.) I claim that there are gains from trade
today that were not available in the EEA, and also that the gains from conflict are much
reduced relative to the evolutionary past. I analyze the potential costs and benefits of
ethnic conflict. I argue that there are three reasons why ethnic conflict does not pay
today. First, fitness is not constrained by land or other resources. Second, many
countries punish those engaging in ethnic persecution. Third, and most importantly,
ethnic conflict eliminates the possibility of gains from trade, so that any gains from such
conflict are likely to be outweighed by the losses from reduced exchange. This last issue
is important because even students of human evolution may be misled on this point; an
example is provided.
In addition to increasing gains from trade, there is another long term trend in
ethnic conflict. That is the increasing size of entities treating themselves as ethnic
groups. Van den Berghe (1981) indicates that who is included within a particular group
is a variable subject to change, but he does not indicate that the secular trend is to include
more and more subgroups in one ethnic group. For example, Cavalli-Sforza et al.
(1994/1996) indicate that most European countries are heterogeneous and composed of
many subgroups. This trend has been remarked upon by many, and is often viewed as a
desirable moral outcome (e.g. J.Q. Wilson, 1993; Westermarck, 1932, earlier made a
similar point.) This increasing size of ethnic groups is an implication of Alexander’s
(1987) “balance of power” hypothesis: as technology allows increasing density of human
populations, groups become larger in order to more effectively compete with neighboring
groups. To the extent that this increase in size is due to the considerations discussed by
Alexander, hope that it will lead to all humanity viewing itself as one group is unlikely to
be realized because there would be no competitive incentive to coalesce into one group.
On the other hand, increasing gains from trade and reduced gains from conquest may lead
to reduced conflict without bound.
A ZERO-SUM SOCIETY?
ETHNIC CONFLICT
PAGE 5
There is one economic point that will be very important in understanding the
evolved nature of political preferences. Today’s world differs in crucial respects from the
EEA. One major difference is that the EEA was largely a zero sum world, with
competition of the sort that Hirshleifer (1998) calls “absolute competition.” I argue that
gains from trade even within a group were limited, except for sexual division of labor,
which is universal among humans (Brown, 1991). I then argue that there were also
limited gains from trade between groups. I discuss some implications for the likely
nature of evolved human attitudes.
Within Groups
Consider first the possibility of exchange within groups. At one time, our
ancestors hunted and gathered, and the amount of food was approximately given. There
was little if any production of food or other resources. In the EEA it is likely that there
were very few “gains from trade,” except for gender based exchange and intertemporal
exchange aimed at reducing risk. That is, it is likely that resources were relatively fixed
and there was little possibility of value increasing exchange or production, such as we see
all around us today. We evolved in such an environment, and our minds evolved to adapt
to it. The result is that humans in many cases now tend to base decisions on zero sum
thinking when other forms of analysis would be useful.
It appears likely that there was relatively little division of labor evolutionary
times. First, group sizes were relatively small, and as Adam Smith first indicated in
1776, “The division of labor is limited by the extent of the market.” Before there can be
a pin factory, there must be sufficient demand for pins to support it. Among small groups
there was not enough room for substantial specialization and division of labor. If a social
group is small, then a specialist will not be able to work full time at the specialty because
the group will not be able to make use of the output of a full time specialist. The general
point is made by Maynard Smith and Szathm↔ ry (1999, p. 148): “Populations of, at the
most, a few hundred individuals, with little division of labour except, probably, that
between the sexes, have been replaced by societies of many millions, dependent on
extreme division of labour.”
This may explain the sad situation of the Tasmanians. Edgerton (1992) describes
the circumstances of these people, who lived in bands of 40-50 people within a total
ETHNIC CONFLICT
PAGE 6
population of about 4,000. These people had at one time had a relatively advanced
technology developed before the flooding of the land bridge that had connected them to
Australia. Over a 10,000-year period of isolation, they progressively lost most of their
technology. This may simply be because the size of the population and of the small
bands was insufficient to support any significant level of division of labor. Edgerton
describes other societies that lost various forms of technology as a result of isolation or
other sources of population reduction.
Ridley (1996) believes that division of labor and specialization was common in
the EEA. In discussing his vision of the division of labor, Ridley (1996, p. 49) says “One
man made stone tools, another knew how to find game, a third was especially good at
throwing spears, a fourth could be relied upon as a strategist.” But in a group of 50-200
individuals, the likely group size during much of the EEA, there would not be full time
work for most of these specialties. Rather, while people may have had the skills Ridley
mentions, they would have been unlikely to engage in these activities on a full time basis.
That is, while there may have been some limited amount of specialization, it would have
been incomplete. The hunter who knew how to find game would have thrown spears at it
as well and the spear thrower might have engaged in butchering if his spear hit.
Sahlins (1972) argues that the division of labor and specialization was limited in
primitive societies of the sort that are relevant for analysis of the EEA. “… for every man
can and does make the things that men make and every woman the things that women
make.” (Sahlins, p. 9, favorably quoting Marshall, 1961.) While Sahlins work is
controversial, it appears that this particular point has not been debated. For example
Bird-David (1992), in a paper commenting on Sahlins, discusses the simple nature of the
tools used in such societies, with the implication that specialized toolmakers did not exist.
Moreover, exchange would have been difficult, and so again limited. Economists
stress that in non-monetary societies, barter would have been necessary for exchange.
Barter is limited by the requirement for a “double coincidence of wants” – that is, each
party to a potential exchange must have exactly what the other party desires for exchange
to occur. Reciprocal altruism (Trivers, 1971) extends this possibility somewhat, by
allowing intertemporal exchange, but the amount of non-simultaneous exchange is
limited by the possibility of cheating. That is, if one hunter shares his game today, he
ETHNIC CONFLICT
PAGE 7
must be certain that the beneficiary will share his when situations are reversed, but it is
difficult to guarantee this in a world with no contract enforcement. The biological notion
of reciprocal altruism is a mechanism that will alleviate this problem by mechanisms
identified by Trivers, such as moralistic anger in response to cheating. But these
mechanisms will not fully solve this problem. This limited amount of exchange is
consistent with the conclusion of Stiner et al. (1998) that “Low human population
densities during most of the Middle Paleolithic imply that group sizes and social
networks were small, which certainly limited the numeric scope of individual
interactions. Under these conditions the possibilities for evolution of complex sharing
and exchange behavior as ways to counter the effects of unpredictable resource supplies
would have also been quite limited.” This suggests that even the insurance function of
exchange (as discussed by Knauft, 1991) was limited.
Between Groups
If exchange within a group was of limited benefit, then exchange between groups
would also have been of limited value. First, in a world with relatively little production,
resources would of necessity have been limited and amounts could not have been
increased. If one group killed an elephant, then that elephant was not available for
another group to kill. If one tribe gathered in a certain area, then other tribes had to go
elsewhere. Second, exchange between groups would have been even more difficult than
exchange within a group. As mentioned above, barter requires that both parties
simultaneously have something that the other party wants, or that there be some
enforcement mechanism to mandate repayment. Such enforcement mechanisms
(including the moralistic aggression that enforces reciprocal altruism) would have had
much weaker force between group than within groups. Thus, exchange between groups
would have been of limited value in the EEA. Humans evolved in this environment, and
it would have paid for them to evolve xenophobic and nepotistic attitudes.
Over time, group size did expand, and small groups coalesced into larger entities
(e.g., Wiessner, 1998). There were benefits of such larger agglomerations, in terms of
possibilities for exchange (of spouses and goods) and also for more effective conflict
with agglomerated neighboring groups, an implication of Alexander’s (1987) “balance of
power” competition. However, this fusion could not occur until technologies able to
ETHNIC CONFLICT
PAGE 8
support large densities were available. Humans and our ancestors spent much time in the
small bands discussed by Stiner et al. (1998) before coming together into larger bands.
These larger bands then adopted mechanisms borrowed from smaller family sized groups
to generate loyalty within the larger group (Wiessner, 1998; Eibl-Eibesfeldt, 1998).
Moreover, if the amalgamation was due to increased effectiveness in conflict (both
offensive and defensive), then at any given stage of the process (which has been
continuous, from family based groupings to today’s large nations) individuals would have
viewed members of other groups as competitors. Weisnner (1998) does indicate that at
some point (which she identifies as the Early Upper Paleolithic -- about 30,000-40,000
BP) there was widespread homogeneity of decorative art, indicating large open systems.
But before this, there was relatively little exchange; Gamble (1999) indicates that in the
Late Middle Paleolithic most transfers of materials were relatively local.
In a zero sum world of the sort that existed in the EEA, conflict between tribes
could be useful for the victors, both as a group and individually. There is no need of
group selection arguments to explain such conflict. First, if one tribe (band) could defeat
another, then the victorious combatants would have access to the resources commanded
by the second tribe. Second, a common pattern in the EEA was the taking of the females
of defeated enemies. Both of these gains – resources and females – could be translated
into fitness, and so those who engaged in this behavior would have benefited and would
have left more genes in the human gene pool. Thus, in the EEA such conflict could have
been individually desirable with respect to fitness, and we are descended from those who
gained from such conflict.
Once preferences for conflict with our neighbors evolved (and they probably date
back to our ape-like ancestors: chimpanzees engage in something very like war and
genocide; e.g., Wrangham and Peterson, 1996) then they would have been selfreinforcing. If I will attack members of the neighboring tribe on sight, then they will
learn to attack me first. If I can identify them by dress or accent or tattoos then these
should become useful markers for friendship and enmity.5 But this does not mean that
5
Skin color would probably not have been relevant in the EEA. Neighbors with whom one fought would
have been relatively close relations (relative to the entire genetic variation of mankind), and of the same
color. (van den Berghe, 1981, p. 31 makes this point.) This is still true today. The combatants in the
disputes mentioned above (Catholic Irish and Protestant Irish, Jews and Arabs, Indians and Pakistanis, the
ETHNIC CONFLICT
PAGE 9
such conflict is desirable today, even if observed from the viewpoint of the combatants,
or even the victors.
Evolved Attitudes
The result is that we evolved in a world with limited possibilities of gains from
trade or exchange, and limited possibilities of activities that would increase personal or
social wealth. Therefore, our minds are built for understanding a zero sum society. (See
also Rubin (2000a), for some other implications of this analysis.) Economics stresses the
possibilities of gains from trade – it is possible for both parties to come out ahead from
exchange. Indeed, this is what runs a modern economy. But this is a very counterintuitive notion for most people. (This should not be interpreted as arguing that people
cannot understand the notion, but rather that it must be taught.)
Indeed, even when people do engage in mutually beneficial trade, they are not
motivated by this mutual benefit. Rather, each aims at maximizing his or her own
benefits. The fact that for trade to occur, both must benefit is irrelevant for each
individual. To quote Adam Smith again, “It is not from the benevolence of the butcher,
the brewer, or the baker that we expect our dinner, but from their regard to their own
interest.” Thus, there is no reason to expect that an innate mental module to measure
gains from trade has evolved. Rather, we are each selected to try to be sure that we gain
from trade; gains to our trading partner are irrelevant. Moreover, there are mental
mechanisms that work against this recognition of mutual benefit. Even in mutually
beneficial trades, there is also an aspect of competition. Both sides want to engross for
themselves as much as possible of the gains. That is, buyers want the price to be low,
and sellers want it to be high. Therefore, in engaging in trade, an important consideration
is to avoid being victimized. As a result, mental modules aimed at policing transactions
have evolved (Cosmides and Tooby, 1992). These modules focus on the “zero-sum”
aspect of trade (Wright, 1999) – that aspect dealing with the terms of the bargain, rather
than with the gains.
There may be another difference between conflict and trade. To engage in
conflict (at least at the group or tribal level) a conscious policy decision must be made.
various disputants in the former Yugoslavia) are close relatives compared with the entire range of mankind
and cannot be differentiated by skin color.
ETHNIC CONFLICT
PAGE 10
Many persons must be consulted, and pros and cons debated. On the other hand,
individuals automatically engage in mutually advantageous trade. This might mean that
we must think more consciously about gains or losses from conflict, while we can simply
engage in trade without any conscious thought, and especially without any thought of the
mutual benefits.
I do not want to be read as implying that trade is artificial or that people must be
taught to engage in exchange. As an economist, I believe that trade is of fundamental
importance to humans; Adam Smith discussed “the propensity to truck, barter, and
exchange one thing for another.” People engage in mutually profitable exchange when it
is possible, and they do so automatically But for reasons discussed above, I do believe
that people do not have an innate understanding of the benefits of such exchange.6
This lack of understanding may apply particularly to international trade – trade
between countries or, more relevant for the analysis here, between ethnic groups. This is
because in addition to the normal views of trade, international trade also implicates our
xenophobic modules. Krugman (1996) has pointed out the fallacies in many highly
popular arguments against such trade; the relevant point for the analysis here is that the
anti-trade arguments were highly popular and presented in well selling books. Such
arguments are also popular politically. While the U.S. now has relatively small tariffs
and other barriers to international trade, there was a struggle to reduce these levels, and
there is is a continual struggle to maintain these low levels, in part because people do not
fully understand that trade is mutually beneficial.
INEFFICIENCY OF ETHNIC CONFLICT TODAY
There are at least three reasons why ethnic conflict is not likely to be useful for
combatants improving in today’s world. First and most importantly, the modern world is
not zero sum. Rather, there are possible gains from trade in many dimensions. Ethnic
conflict reduces the possibility of these gains, and thus harms the combatants; see also
Hirshleifer, 1998. Second, for most of the world, the issues that may be involved in
ethnic conflict are not the goals of combatants. Third, in many cases the rest of the world
punishes groups for engaging in ethnic combat, so that the net result is not a gain. I
discuss each point.
6
If they did, the teaching of economics would be much simpler.
ETHNIC CONFLICT
PAGE 11
Gains from Trade
The main reason why ethnic conflict does not pay is because of potential gains
from trade that are eliminated by this conflict. These gains come from two sources.
First, if members of ethnic groups have similar skills and abilities, then there are benefits
from simply expanding the size of the market. Second, if members of different ethnic
groups have different skills and abilities for any reason,7 then there are gains from trade.
Polachek (1992) has shown that in the case of countries, increased trade and increased
gains from trade are associated with significantly reduced levels of conflict, and that the
direction of causation is from trade to conflict reduction.
Consider first the size of the market. Remember that “the division of labor is
limited by the extent of the market.” This means that as the number of people in a market
expands, it is possible for each worker to specialize in a narrower area of production. As
this happens, productivity (real wealth) increases. More people in a market also create
the possibility for more specialized consumption opportunities. It is also true that the
existence of more people leads to a greater possibility for technological advance, as
shown by Julian Simon (1981/1996), Michael Kremer (1993), and Charles Jones (1999).8
Of course, additional people can be from the same ethnic group, but if population growth
has slowed or become negative, as is true for much of the developed world, then the only
way to realize these gains is to allow members of different ethnic groups to join the
society.
The second type of gain from trade occurs if the members of different ethnic
groups have different skills and abilities. In that case, there are possibilities of members
of each group specializing in different activities, and then engaging in mutually beneficial
trade. Of course, in this circumstance, those particular members of the dominant group
who compete most closely with members of the smaller group would have incentives to
reduce this trade, perhaps by generating ethnic conflict, but members of the society
overall will benefit from the increased trade.
7
It is not necessary to consider the source of such differences, only their existence.
This is not to say that at some point incomes may fall as population increases. However, there is no
evidence that we have reached such a point. Jones (1999) estimates that the number of new ideas generated
per year has increased by a factor of 108 from 25,000 BC to today as a result of population growth.
Spillovers and better property rights have also caused massive increases in new ideas.
8
ETHNIC CONFLICT
PAGE 12
To make this point more carefully, I will consider Jews as an ethnic group. Jews
have been a minority ethnic group in many societies and have been persecuted in many of
these societies. Moreover, a recent series of books by an evolutionary minded scholar,
Kevin MacDonald 9, has analyzed anti-Semitism from an evolutionary perspective, but
has missed the key point about increased gains from trade. Therefore, examination of
this ethnic group will provide a useful demonstration of the general point. The general
assumption is that Jews will live in a society as a minority, where the ethnic majority is
from another background. Of course, this point would apply to any minority ethnic
group; I merely use Jews as a convenient example.
MacDonald argues that anti-Semitism is caused ultimately by “resource
competition and conflicts of interest” between Jews and gentiles living in he same
society. The key point that is largely missed or ignored is that this “resource
competition” will generally provide economic benefits to gentile society overall, even if
it does harm some segments of gentile society. MacDonald sometimes indicates that
there are benefits from trade with Jews to some gentiles, but only in passing. For
example, p. 9 “… anti-Semitism is expected to be … least common among gentiles who
are actually benefiting from the Jews, such aristocratic gentiles who often profited from
cooperation with them.” There is no mention that Jewish merchants, doctors, teachers,
and moneylenders also provide benefits to their customers, patients, students and debtors.
Even though I do not accept it, I will for the sake of argument grant MacDonald’s
thesis, that Jews are a separate society within gentile society. (This may have been more
appropriate in Midievel Poland than of contemporary society.) However, in his model
there is economic exchange between these two societies. An appropriate model to use in
analyzing this situation is the model of trade between separate countries, the model of
international trade. The results of using this model are unequivocal: members of both
societies, Jew and gentile, gain economically from free exchange. (This use of the model
of international trade to discuss ethnic groups in a society is not original: Gary Becker,
(1956/1971), uses this model in discussing black-white discrimination, and reaches the
same conclusion: both groups benefit from exchange.)
9
This is a three volume study: MacDonald, 1994, 1998a, 1998b. I am here concentrating on 1998a,
Separation and Its Discontents: Toward an Evolutionary Theory of Anti-Semitism because this is the
ETHNIC CONFLICT
PAGE 13
Those parties in direct competition with Jews, such as gentile merchants in
competition with Jewish merchants, might lose from this competition, just as some
American firms and workers lose from international competition with other countries.
But the broader point is that the gains to those who gain from exchange (the customers of
the Jewish merchants, or the clients of the Jewish moneylenders) gain more than the
losers (non-Jewish merchants or moneylenders) lose, so that overall there are net gains to
gentile society from free trade. This is a standard argument in analyzing trade: American
(or, more generally, domestic) consumers gain more from international trade than
competing American (domestic) producers lose. This is the standard economic argument
in favor of free trade, dating back to David Ricardo; for a modern discussion, see
Krugman (1996).
These gains exist even if Jews could succeed in colluding and acting like a cartel,
or a monopoly, which is highly unlikely. (Normally, economists believe that large
groups of sellers, such as all the Jewish merchants in Poland, or even in Warsaw, cannot
successfully collude because of free rider problems of the same sort that some argue may
make group selection impossible.10) For retail trade and most other businesses, the limits
of the ability of a Jewish monopoly to exploit gentile customers are given by the
possibility of other gentiles entering to compete. If Jews are more efficient than gentiles
in a particular industry or trade, then Jewish monopolists will make profits in that
industry, but gentile customers will still benefit from trading because the Jewish
merchants still sell goods for lower prices than would gentile merchants. Jewish
merchants might have been more efficient, perhaps because of efficient links between
countries. However, even for a monopolist, if costs are lower, then the profit maximizing
price will be lower, so that efficient traders would have benefited their customers as well
as themselves. For one example discussing such links for the 11th century Jewish
Maghribi traders, see Greif (1993). Greif indicates that these traders lost their markets
when the Italian merchants became more efficient.
volume that deals with economic relations between Jews and gentiles.
10
I have worked as an economist in numerous antitrust matters, and have personally seen no evidence that
Jews in business compete any less vigorously against other Jews than against gentiles, or than gentiles
compete against gentiles. Indeed, one well known paper argued that discrimination against Jews in medical
school admissions was to avoid the fact that Jews were well known as highly competitive in pricing:
Kessel, 1958.
ETHNIC CONFLICT
PAGE 14
Take the stronger case. Assume that gentiles are forbidden from engaging in
money lending, perhaps by religious prohibitions, so that there is no potential competition
for Jewish moneylenders. (This position is actually false: Barzel, 1992, indicates that the
Jewish moneylenders in England lost their market to the Italians in the 13th century.)
Assume for the sake or argument that all Jewish moneylenders collude and charge the
monopoly price, although this is highly unlikely. Nonetheless, even under these extreme
and highly unlikely conditions, gentile borrowers will still benefit relative to the situation
with no moneylenders. That is, dealing with a monopolist is better than having no
opportunity to trade at all. The “terms of trade” for dealing with a monopolist are worse
than the terms of trade in dealing with a competitive firm, but still some voluntary trade
is better than none.
If my argument is correct, then anti-Semitic societies should have been harmed by
their unwillingness to deal with Jews. In discussing the Inquisition, MacDonald indicates
(p. 15) that “The Inquisition had a very chilling effect on intellectual endeavor in Spain
for centuries.” With respect to Roman Christian anti-Semitism, another example
discussed by MacDonald, there is no discussion of the effects of anti-Semitism.
However, the period following the imposition of restrictions on Jews discussed by
MacDonald is sometimes called the “Dark Ages,” indicating that it may be characterized
by relatively little intellectual advance. With respect to the Nazis, it is interesting to note
that the American nuclear bomb was produced in part with inputs from many Jewish
scientists who left Germany and other European countries in fear of the Nazis. Without
Nazi anti-Semitism some of these scientists might have worked on the German war
effort. Thus, it is at least plausible that anti-Semitism has had the negative effects on the
gentile community that economic theory would suggest.
Jews would also be harmed by separatism, and the harm to Jews would be greater
than the harm to gentiles. This is because there are fewer Jews and so the loss of trade
opportunities with gentiles will harm the Jewish community more than the loss of trade
opportunities with Jews will harm gentiles. Self-sufficiency is more expensive for a
smaller group.
It might appear that economic arguments are irrelevant in an analysis of fitness.
Fitness is often defined in relative terms, and economics deals with absolute incomes.
ETHNIC CONFLICT
PAGE 15
But this issue is not relevant here. The definition of fitness in relative terms is a result of
the mathematics of analyzing a steady state population. Most populations at issue,
Jewish and gentile, were growing over the relevant time period discussed by MacDonald.
Thus, if trade with Jews would have benefited gentiles, then gentile populations could
have grown faster than otherwise. Moreover, there was competition between gentile
populations during this period. Any gentile population that established economic
relations with Jews would have been expected to benefit relative to other non-Jewish
populations.
Conflict between neighbors was an important fact during human evolution;
indeed, some think that it this conflict was one of the major inputs into human evolution
and in particular the evolution of human intelligence (e.g., Alexander, 1987). However,
exchange is also important and there are gains from exchange. It is a common error to
view exchange in competitive terms; discussion of “trade wars” is a staple of the popular
press, and of politics. This is an example of the sort of zero sum thinking discussed
above. The implicit notion is that there is only a certain amount of money to spend, and
that if it is spent with foreigners it is not spent with domestic firms. But one of the most
basic lessons of economics is that this notion is false (e.g., Krugman, 1996).
Nonetheless, it may be a notion that we are selected to believe unless taught otherwise.
Rubin (2000b) discusses in more detail the nature of evolved political preferences.
Thus, ethnic groups today can be viewed as playing a variant of the repeated
prisoner’s dilemma. One strategy is to cooperate and engage in mutually profitable trade.
The other strategy is to attack one’s neighbor and expropriate his wealth. The “folk
theorem” of game theory tells us that cooperative solutions are possible in such games;
Axelrod (1984) has shown that one such solution, “tit for tat” is a successful strategy in
many environments. Goetze (1998) also discusses the benefits of increased interaction
between groups, which he calls functional benefits. Those groups that avoid conflict are
the ones that can choose the cooperative strategy; groups engaging in conflict choose the
attack strategy. There may be some inherited tendency to choose the latter strategy if
tastes have evolved in a zero sum environment. Nonetheless, most potential adversaries
seem to be able to choose the cooperative strategy. Those with an interest in reducing the
level of ethnic conflict might attempt to determine why one strategy rather than the other
ETHNIC CONFLICT
PAGE 16
is chosen. This avenue to conflict reduction is more likely to be successful than an
approach based on increasing the size of in-groups, if the latter are a result of balance of
power competition. This is especially true if there are few benefits from the attack
strategy, as discussed next.
Gains from Conflict
In today’s world, ethnic conflict is not likely to provide net gains, perhaps even to
the victor. Gains must be defined. In biological terms, we think of gains in terms of
fitness. Economists are more likely to think of utility maximization. These may not be
so different; utility functions presumably evolved, and we obtain utility from those things
that were associated with fitness in the evolutionary environment. Wealth is also closely
related to utility, although not identical. In any event, no matter what the maximand, it is
unlikely that ethnic conflict will in general in today’s world achieve any goal.
Consider fitness. The issues that are relevant in ethnic conflict are not related to
fitness for most people. In the EEA, conflicts were over land or territory, and land was a
major input into survival and thus into reproduction. Today, land may be valuable, but it
is only one valuable asset among many. Possession of land is only tenuously related to
the ability to provide food.11 A country such as Japan with little land per capita can
easily feed its people by trading other goods (cars, televisions) for food. A country such
as Russia with huge amounts of land cannot feed itself because its economic and legal
systems are so disorganized that the country cannot function effectively.12 In the U.S.,
agricultural production is only about 1% of GDP.13 Land may still be valuable, and it
may be worth fighting over land on occasion. But land is no different than any other
resource, and in many cases the value of land can be destroyed by the very conflict aimed
at gaining it.
Moreover, biological fitness is not a goal for many people.14 We do not have the
maximum number of children that we could biologically support. In many countries in
11
At least in the developed world. It may be that in some less developed areas, conflict over land would be
fitness improving.
12
It is also subject to harsh weather, but even so agriculture is not very productive.
13
Department of Commerce, Bureau of Economic Analysis, July 1999 Survey Of Current Business, Table
1.7.— Gross Domestic Product by Sector (Web edition.) GDP = $8,511.0 billion; “Farm” Product = $84.3
billion.
14
This is not to say that it was a conscious goal in the past, although for many it may have been: Betzig,
1986.
ETHNIC CONFLICT
PAGE 17
Europe, population growth is actually below replacement levels, and this is not due to
wealth constraints. In China, draconian population policies are aimed at reducing the
number of people in the country. As other countries become wealthier, we may expect
similar patterns. The constraint on the number of children is parental desires or social
policy, not any economic limit. Policies that would have led to increased fitness or
reproductive success in the EEA will not have that effect today. Thus, again, land is
valued for its economic value, not for its ability to grow crops to increase fitness.
Capturing of land would not in general be a wealth maximizing strategy in the
developed world. Much of the value of land comes from minerals or petroleum, which
can be captured through conquest.15 But much comes from the organization of
production. The value of real estate in New York City, Tokyo, and Hong Kong does not
come from the crops that can be grown or because of the minerals under the land.
Rather, they are valuable because of the buildings and, more importantly, because of the
human capital in the workers in the buildings that is agglomerated in the cities. An
invading army conquering one of these cities would not be able to appropriate the wealth
that they create by taking over the land. The buildings without the workers would have
little if any value. Rather, to benefit, a conqueror would be forced to allow the existing
population to continue working. Any benefit to the conqueror would come from taxes, as
in the case of China and Hong Kong. This is very different from massacring the men and
kidnapping and raping the women, the standard form of ethnic conflict in the EEA, and
among chimpanzees.
If humans gain utility directly from attacking those who are ethnically different,
then by definition such attacks would be utility increasing. But it is unlikely that this is
so as a general matter. In certain circumstances, there may be evolved mechanisms that
lead to hostility and conflict. These mechanisms have evolved to adapt to recurrent
situations in the EEA associated with ethnic conflict. They may involve defense from
ethnically different attackers, or they may involve attacks on those who are ethnically
different in circumstances where such attacks would have been likely to succeed. Such
mechanisms are likely to be domain specific (Barkow et al., 1992). But in any case, it
15
The recent Gulf War may have been aimed at eliminating the power of Iraq to obtain a monopoly
position with respect to petroleum.
ETHNIC CONFLICT
PAGE 18
appears obvious that some learning is involved; most of us in the U.S. have learned not to
engage in such attacks, or to anticipate utility from them. While humans can be
indoctrinated to find such aggression desirable (Eibl-Eibesfeldt and Salter, 1998) absent
such indoctrination, we are unlikely to spontaneously engage in this behavior. More
generally, there is evidence that more democratic governments – in which individuals
have more power – are much less likely to engage in either internal or external conflict
than are more authoritative governments (Rummel, 1994). If humans had strong tastes
for such conflict (as opposed to having the ability to be indoctrinated to engage in
conflict) we would expect democracies to disproportionately engage in hostilities. The
fact that they do not is evidence for the weakness of whatever preferences may exist.
Punishment
In many cases, the world punishes those who engage in ethnic conflict. This
punishment is not certain, and does not always occur. But it is probably common enough
so that it reduces the already small return from ethnic persecution or conflict. For one
example, a worldwide boycott of South Africa ultimately caused the ethnically based
apartheid regime to collapse. For another, the Serbian policy of “ethnic cleansing” in
Kosovo was quite costly to the Serbs as a result of U.S. led NATO bombing. Indeed, this
set of Serbian policies in Kosovo and elsewhere in the former Yugoslavia has actually led
to a decline in fertility. In 1998 58,000 babies were born in central Serbia, a 15 percent
decline since 1990. The population growth rate has fallen in the 1990s from a 1.2 percent
annual growth rate to a 1.8 percent decline. (Harden, 1999). It is unlikely that Germany
or Japan, both of which practiced horrid ethnic policies in World War II, gained from
these wars. There were 5.6 million Germans and 2.5 million Japanese killed in this war
(Ghiglieri, 1999, p. 303, note 217) and both countries suffered extensive physical damage
to capital stock from bombs. Both countries are prosperous today, but this prosperity
came from trading and exchanging with ethnically heterogeneous neighbors, not from
exterminating them. Germany, for example, has large numbers of Turkish workers living
in the country. Of course, leaders of countries may still benefit from ethnic conflict, but
the policies are probably not in the genetic or economic interest of the citizens of the
aggressor countries.
ETHNIC CONFLICT
PAGE 19
I do not want to overemphasize this factor; many examples of ethnic conflict do
not lead to any punishment. But the chance of such punishment is one additional cost
that must be considered when contemplating the possibility of ethnic aggression.
SUMMARY
It may be that humans are selected to engage in a certain amount of ethnic
conflict. Such conflict would have been fitness improving for the victors (who are our
ancestors) in the EEA and so we might have inherited tastes for such conflict. But in
today’s world, in many circumstances it is unlikely that benefits from such behaviors
accrue, even to the victors. While we may have played a zero sum game in the EEA,
today a prisoner’s dilemma, with gains from cooperation, is a much more valid
description of the interaction between ethnic groups. In particular, the possibility of
gains from trade indicates that ethnic cooperation is much more likely to increase both
wealth and fitness than is ethnic conflict. There are other reasons why such conflict
might not pay today. Humans can learn this, as shown by the level of inter-ethnic
cooperation in a society such as the U.S., but they must be taught: it may not be a lesson
that humans learn on their own. Indeed, even some students of evolution do not fully
understand the gains from trade or the costs of ethnic conflict. Nonetheless, if we want to
reduce the level of ethnic conflict, reliance on increased gains from trade may be the most
effective method. Those who hope that humans will increasingly identify with all
humans as relevant in-group may be putting more faith in “balance-of-power”
mechanism than is justified. Such mechanisms are limited, in that they operate only
when there is also an out-group with whom to compete. On the other hand, gains from
trade exhibit no such limits.
ETHNIC CONFLICT
PAGE 20
BIBLIOGRAPHY
Alexander, Richard D. (1987), The Biology of Moral Systems, New York: Aldine de
Gruyter.
Axelrod, Robert (1984), The Evolution of Cooperation, New York; Basic Books.
Barkow, Jerome H., Leda Cosmides and John Tooby (1992), editors, The Adapted Mind,
New York: Oxford University Press.
Barzel, Yoram (1992), “Confiscation by the Ruler: The Rise and Fall of Jewish
Lending in the Middle Ages, Journal of Law and Economics v. 35, No. 1,
April, pp. 1-13.
Becker, Gary S. (1956/1971), The Economics of Discrimination 2d ed. Chicago :
University of Chicago Press.
Betzig, Laura L. (1986), Despotism and Differential Reproduction: A Darwinian View of
History, Hawthorne, New York: Aldine de Gruyter.
Bird-David, Nurit (1992), “Beyond ‘The Original Affluent Society’: A Culturist
Reformulation,” Current Anthropology, v. 33, no. 1, February (including
comments and reply.)
Brown, Donald E. (1991), Human Universals, New York: McGraw Hill.
Cavalli-Sforza, L. Luca, Paolo Menozzi and Alberto Piazza (1994/1996), The History
and Geography of Human Genes (Abridged Paperback Edition), Princeton:
Princeton University Press.
Cosmides, Leda and John Tooby (1992) "Cognitive Adaptations for Social Exchange," in
Barkow, Jerome H., Leda Cosmides and John Tooby (1992), editors, The Adapted
Mind, New York: Oxford University Press.
Crawford, Charles (1998), “Environments and Adaptations: Then and Now,” in Charles
Crawford and Dennis L. Krebs, editors, Handbook of Evolutionary Psychology,
Mahwah, New Jersey: Lawrence Erlbaum Associates, 275-302.
Edgerton, Robert B. (1992), Sick Societies: Challenging the Myth of Primitive Harmony,
New York: Free Press.
Eibl-Eibesfeldt, Iran← us (1998), “Us and the Others: The Familial Roots of
Ethnonatinalism,” in Iran← us Eibl-Eibesfeldt and Frank Salter, eds.,
Indoctrinability, Ideology, and Warfare, New York: Berghahn Books.
Eibl-Eibesfeldt, Iran← us and Frank Salter, (1998), eds., Indoctrinability, Ideology, and
Warfare, New York: Berghahn Books.
Gamble, Clive (1999), The Paleolithic Societies of Europe, New York: Cambridge
University Press.
Ghiglieri, Michael P. (1999), The Dark Side of Man: Tracing the Origins of Male
Violence, Reading Massachusetts: Perseus Books.
Goetze, David (1998). "Evolution, Mobility, and Ethnic Group Formation." Politics and
the Life Sciences 17, 1 (March): 59-71.
ETHNIC CONFLICT
PAGE 21
Greif, Avner (1993), “Contract Enforceability and Economic Institutions in Early
Trade: The Maghribi Traders’Coalition,” American Economic Review, V.
83, No. 3, June, pp. 525-548.
Harden, Blaine (1999), “Stresses of Milosevic's Rule Blamed for Decline in Births,” New
York Times, July 5.
Hirshleifer, Jack (1998), “The Bioeconomic Causes of War,” Managerial and Decision
Economics, V. 19, No. 7-8, November. (Special Issue: Management,
Organization and Human Nature. Issue Edited by Livia Markoczy).
Johnson, Gary R. (1986), “Kin Selection, Socialization, and Patriotism: An Integrating
Theory,” Politics and the Life Sciences, V. 4, No. 2, pp. 127-140.
Jones, Charles L. (1999), “Was an Industrial Revolution Inevitable? Economic Growth
Over the Very Long Run,” National Bureau of Economic Research, working
paper 7375.
Kessel, Reuben A. (1958), “Price Discrimination in Medicine,” Journal of Law
and Economics, October, pp. 20-53.
Knauft, Bruce (1991), “Violence and Sociality in Human Evolution,” Current
Anthropology, v. 32, No. 4, August-October, pp. 391-428 (including comments
and reply).
Kremer, Michael (1993), “Population Growth and Technological Change: One Million
B.C to 1990,” The Quarterly Journal of Economics, pp. 681-716.
Krugman, Paul (1996), Pop Internationalism, Cambridge: MIT Press.
MacDonald, Kevin (1994), A People That Shall Dwell Alone: Judaism As a Group
Evolutionary Strategy Westport, Conn.: Praeger, Human Evolution, Behavior, and
Intelligence, Seymour W. Itzkoff, series editor.
MacDonald, Kevin (1998a), Separation and Its Discontents: Toward an Evolutionary
Theory of Anti-Semitism, Westport, Conn.: Praeger 1998, Human Evolution,
Behavior, and Intelligence, Seymour W. Itzkoff, series editor.
MacDonald, Kevin (1998b), The Culture of Critique, Westport, Conn.: Praeger 1998,
Human Evolution, Behavior, and Intelligence, Seymour W. Itzkoff, series editor.
Marshall, Lorna (1961), “Sharing, Talking and Giving: Relief of Social Tensions among
!Kung Bushmen,” Africa, 31: 231-49, quoted in Sahlins, 1972.
Maynard Smith and Szathm↔ ry, E¬ rs (1999), The Origins of Life: From the Birth of
Life to the Origins of Language, New York: Oxford University Press.
Polachek, Solomon William (1992), “Conflict and Trade: An Economics Approach to
Political International Interactions,” in Walter Isard and Charles H. Anderson,
eds., Economics of Arms Reduction and the Peace Process, New York: North
Holland.
Ridley, Matt (1997), The Origins of Virtue: Human Instincts and the Evolution of
Cooperation, New York: Viking Press.
ETHNIC CONFLICT
PAGE 22
Rubin, Paul H. (2000a), “Hierarchy,” Human Nature (In press.).
Rubin, Paul H. (2000b), “On the State of Nature and the Evolution of Political
Preferences,” American Law and Economics Review (In press.)
Rummel, R.J. (1994), Death by Government, New Brunswick: Transaction Publishers.
Sahlins, Marshall (1972), Stone Age Economics, Chicago: Aldine-Atherton.
Salter, Frank (in press), “A Defense and Extension of Pierre Van Den Berghe’s Theory of
Ethnic Nepotism,” in P. James and D. Goetze, eds., Evolutionary Theory and
Ethnic Conflict,
Shaw, R. Paul and Yuwa Wong (1989), Genetic Seeds of Warfare: Evolution,
Nationalism, and Patriotism, Boston: Unwin Hyman.
Simon, Julian (1981/1996), The Ultimate Resource, Princeton: Princeton University
Press.
Sober, Elliott and David Sloan Wilson (1998), Unto Others: The Evolution and
Psychology of Unselfish Behavior, Cambridge, MA: Harvard University Press.
Stiner, Mary C., Natalie D. Munro, Todd A. Surovell, Eitan Tchernov and Ofer BarYosef (1998), “Paleolithic Population Growth Pulses Evidenced by Small Animal
Exploitation,” Science, September 25 (Science Online version.)
Trivers, Robert L., “The Evolution of Reciprocal Altruism,” Quarterly Review of
Biology, v. 46, pp. 35-57.
Van den Berghe, Pierre (1981), The Ethnic Phenomenon, New York: Elsevier.
Westermarck, Edward (1932), Ethical Relativity, New York: Harcourt Brace and Co.
Wiessner, Polly (1998), “Indoctrinability and the Evolution of Socially Defined Kinship,”
in Iran← us Eibl-Eibesfeldt and Frank Salter, eds., Indoctrinability, Ideology, and
Warfare, New York: Berghahn Books.
Wilson, James Q. (1993), The Moral Sense, New York: Free Press.
Wrangham, Richard and Dale Peterson (1996), Demonic Males: Apes and the Origins of
Human Violence, New York: Houghton Mifflin.
Wright, Robert (1999) Nonzero : The Logic of Human Destiny, New York: Pantheon
Books
EMORY UNIVERSITY
DEPARTMENT OF ECONOMICS
Atlanta, GA
DELIVER PACKAGES TO:
324A Rich Building
30322-2240
1602 Mizell Drive
voice: 404.727.6365
fax: 630.604.9609
email: prubin@emory.edu
http://www.emory.edu/COLLEGE/ECON/Rubi.htm
March 7, 2000
Gary R. Johnson, Editor
Politics and the Life Sciences
Lake Superior State University
650 West Easterday Ave.
Sault Ste. Marie, MI 49783-1699
Dear Professor Johnson:
I am resubmitting my revised paper, Does Ethnic Conflict Pay?, for consideration for
Politics and the Life Sciences.
I have extensively revised the paper following the comments of the three referees. I
believe that these comments have greatly improved the paper.
I am looking forward to your decision.
Sincerely,
Paul H. Rubin
Professor