Post by zarahan on Jun 12, 2010 17:30:11 GMT -5
J.P. RUSHTON AND HBD RACIAL "EVOLUTIONARY' CLAIMS DEBUNKED
Africanist archaeology and ancient IQ: racial science and cultural evolution in the twenty-first century
Scott MacEachern
World Archaeology. (2006). Vol. 38(1): 72-92 Race, Racism and Archaeology
----
Abstract
Over the last two decades, a number of psychometric researchers have claimed that very substantial
differences in intelligence exist among modern human racial groups, as these groups are traditionally
defined. According to these researchers, African populations suffer severe cognitive deficits when
compared to other modern humans. Philippe Rushton, particularly, places these claimed mental
deficits in an evolutionary context, advancing environmental explanations for such deficits and
asserting that such cognitive differences existed prehistorically as well. Such substantial cognitive
differences should be evident in human behavioural patterns, and thus in the archaeological record.
Archaeological data can thus be used to test these claims about human evolutionary development
and modern human cognitive difference. Examination of the archaeological record does not support
the claims made by these researchers. This suggests that regional differences in IQ test score results
should not be ascribed to variations in human evolutionary development.
---------------------
EXCERPT
Behaviour, selection and African environments
There are three basic elements in Rushton’s theorization of human difference. First, he
amasses a large amount of material on physical, behavioural and social differences
between the groups that form the popularly accepted triptych of human races: ‘Negroids’,
‘Caucasoids’ and ‘Mongoloids’ (Rushton 2000: 17–183; see also Rushton and Bogaert
1987: 265–8). The scope of these topics is very wide: it includes material on brain size, IQ
(and related) test scores, dental development, speed of sexual maturation, age of first
intercourse, life span, number of sexual fantasies, penis size, number of multiple births,
permissive attitudes toward sex, aggressiveness, law-abidingness, ‘mental durability’,
AIDS rates, cultural achievements and much more. Most striking, and most relevant for
this paper, is Rushton’s (2000: 133–7; see also Rushton and Skuy 2000) acceptance of
Richard Lynn’s (1991b, 2003; Lynn and Vanhanen 2002: 197–225) claim that the average
IQ test score of African populations is approximately IQ¼70. In the psychiatric literature,
this correlates to a state of borderline mental retardation (American Psychiatric
Association 1994), implying that the great majority of African people living today suffer
some degree of cognitive deficit, ranging from mild to very severe. Rushton ranks the three
races on the basis of these assembled characteristics (Rushton 2000: 5, 119, 148, 152, 162,
166, 168, 171, 214).
He has stated on many occasions that this is a straightforward scientific description of
humanity, but it is clear that any taxonomy describing Africans, for example, as less
intelligent, more promiscuous, less altruistic, more aggressive, less law abiding,
investing less effort in child-rearing and less culturally developed than Europeans
or Asians – as Rushton’s taxonomy does – is also a ranking of fundamental human worth.
Rushton’s accumulation of these data and the uses that he makes of them have been
very stringently critiqued (see, for example, Brace 1996; Czernovsky 1995; Graves 2002b;
Lieberman 2001; Peters 1995), and that critique need not be repeated here. However, it
should be noted that the sources, characteristics and quality of his data are diverse, and
often of extremely low reliability. Rushton’s use of late Victorian travel pornography
(‘French Army Surgeon’ 1896) as a central reference on human sexual characteristics and
behaviours is perhaps the most egregious use of poor-quality data, but it is very far from
being the only one. His data on racial differences in intelligence test scores is taken from
wildly disparate sources ranging over the last century (Rushton 2000: 38–9, 135–7),
including sources using techniques that even advocates of IQ testing have dismissed as
‘primitive’ and unreliable (Jensen 1988). Many of his demographic data – on longevity and
reproductive rates, for example – have been subject to large-scale fluctuations even within
populations and in relatively short periods of time, making it unlikely that such
comparative measures have any evolutionary underpinnings.
The second element in Rushton’s theorization of racial differentiation involves
accounting for the biological basis of the data accumulated. He hypothesizes that the
‘Mongoloid’, ‘Caucasoid’ and ‘Negroid’ racial groups have been subjected to different
natural selection pressures over evolutionary timescales, and that because of this these
racial groups have evolved different life-history strategies. This hypothesis makes use of
the r-/K-selection schema, developed by MacArthur and Wilson (1967) as a method of
modeling density-dependent natural selection. The consequences of these different
reproductive strategies are, according to Rushton, to be found in the data on physical,
behavioural and social differences between races amassed in earlier sections of the book. In
the formulation used by Rushton, r-selected species, adapted to unstable, rapidly
fluctuating environments, evolve reproduction strategies with prolific production of
offspring and relatively low parental investment in individual offspring, while K-selected
species, adapted to more stable environments, produce relatively lower numbers of
offspring but invest more care in each. Rushton (1987, 2000: 199–216; see also Rushton
and Bogaert 1987) then claims that ‘Negroid’ populations, in particular, are more
r-selected than are ‘Caucasoid’ populations, which are in turn more r-selected than
‘Mongoloid’ populations. In essence, Africans are held to invest bodily resources more
heavily in sex and impulsiveness/aggression, while Europeans and Asians are supposed to
invest such resources more heavily in intelligence, altruism and restraint. While Rushton
does claim that ‘Negroids’ are only relatively r-selected, the data that he amasses give the
impression of very substantial differences between these racial groups, an impression
eagerly seized by racists around the world.
It might be objected that a concentration on ‘Negroid’ characteristics, vis-a`-vis those
of the other major races, imputes an unfair racializing subtext to Rushton’s work, because
he situates the three major races along an r- to K-selected continuum: ‘Negroids’ –
‘Caucasoids’ – ‘Mongoloids’. However, in practice (see below), ‘Caucasoids’ and
‘Mongoloids’ are lumped together as temperate-/cold-climate races against the tropically
adapted ‘Negroids’ throughout this work (e.g. Rushton 2000: 199, 262). In other writings
(Rushton and Horowitz 1999), Rushton speculates upon temperamental differences that
might disadvantage ‘Orientals’ in scientific and cultural achievements vis-a`-vis ‘Whites’.
In the theories of modern racial scientists, the higher IQ scores of ‘Mongoloids’ cannot be
held to imply significant intellectual superiority over ‘Caucasoids’.
This element of Rushton’s work has also been strenuously critiqued (Anderson 1991;
Graves 2002a, 2002b; Silverman 1990; Weizman et al. 1990). There are very serious doubts
about the utility of the r-/K-selection model as an explanation for behavioural differences
among and especially within animal species, especially when the concept is generalized far
beyond its origins (Boyce 1984; Stearns 1992). There exist many cases in which the
predictions of this theory do not hold (Graves 2002a: 66–7), and a number of characteristics
of Rushton’s three racial groups – most notably body size – do not fit the model.
The linkages between r-/K-selection and the characteristics that Rushton associates with
those different forms of selection are quite unclear (Weizman et al. 1990: 4–5). Finally,
Rushton has never satisfactorily established the environmental circumstances in which his
different racial groups were exposed to these different selection pressures. It is difficult to
avoid the conclusion that Rushton turned to an r-/K-selection model because of a fairly
impressionistic resemblance between that model and his image of ‘racial’ variability, and
that he then tuned the model to fit his assumptions.
It is at this point – when we begin to speak about the ancestors of modern humans as
evolving populations in particular environments – that palaeoanthropology and archaeology
become important to Rushton’s account. He claims that ‘Negroid’ populations
differ from ‘Caucasoids’ and ‘Mongoloids’ because the former evolved only in tropical
Africa, while the latter groups moved out of Africa into colder climates (Rushton 2000:
199, 217–33, 262–4). There are two components to this explanation: tropical African
environments are less stable and less predictable than are temperate and cold environments
(thus favouring r-selected strategies of prolific reproduction and low parental
investment in offspring), and in addition temperate and cold climates posed cognitive
demands upon ‘Caucasoid’ and ‘Mongoloid’ populations not experienced by those
peoples living in balmier climes. It should be emphasized that only the first of these
components involves r-/K-selection theory; the second derives in large part from speculations
about the evolution of intelligence published by Richard Lynn (1987, 1991a).
(Lynn (2003: 141), in fact, seems to be claiming that modern Africans are cognitively
unchanged descendants of the hominids that occupied Africa 250,000 years ago.) This
bipartite explanation is supplemented by archaeological data that Rushton claims support
his thesis.
Less attention has been paid to this last component of Rushton’s model than to other
elements of his work (but see Graves 2002a: 71–4; Lieberman 2001: 79), in part because
there are so many other obvious targets for criticism in his work and in part because most
of this critique has come from psychologists and biological anthropologists. It remains,
however, central to Rushton’s accounting of difference between his three racial groups. He
is concerned with producing an integrated explanation of the differences between those
groups, and to do this using r-/K-selection theory he must account for the origins of these
differences in an evolutionary context. This requires engagement with archaeological and
palaeoanthropological data. There is also a long tradition of appeals to environmental
differences in racialist descriptions of African peoples. In almost all such cases, Africans
are held to be cognitively disadvantaged, either because African environments are so
benevolent that they provide a reduced cognitive challenge to African populations (cf.
Herder 1968: 297; Kant 1997 [1775]: 46), or because those environments are so hostile to
humanity that they inhibit the intellectual development of those populations (cf. Cornelius
de Pauw in Duchet 1969: 123). Rushton’s explanation uniquely combines elements of both
of these models.
His approach shares another characteristic with many eighteenth-century accounts of
human racial variation: it treats continental land-masses as undifferentiated geographical
units, each characterized by a particular set of prevailing environmental conditions. Thus,
African environments are collapsed to subtropical savannas, which are prone to
unpredictable droughts and ‘devastating viral, bacterial, and parasitic diseases’ (Rushton
2000: 231), while Europe and Asia are similarly characterized as cold (but predictable) arid
tundra and glacial landscapes.We are never told why all African savannas are supposed to
be prone to unpredictable droughts or the reason for believing that tropical and
subtropical plant foods are going to be available year-round (Rushton 2000: 228). We are
not told why dispersed tropical foragers should be prone to devastating diseases that are
now associated with higher-density, agricultural lifeways (see, for example, Armelagos
et al. 1991; Barnes 2005; Tishkoff et al. 2001). We are never told why Rushton believes that
‘Mongoloids’ evolved in Siberia (Rushton 2000: 229), rather than further south in Asia.
We are never told why environmental predictability is characteristic of temperate and
Arctic climates, while unpredictability reigns in Africa. (Apparently, the only sort of
seasonality that Rushton recognizes is that between summers and winters; the equally
predictable and often very challenging cycling between wet and dry seasons in many areas
of Africa is unknown to him.) Examples could be multiplied. Rushton simply asserts that
racial evolution took place as he says, without providing evidence.
His archaeological and historical evidence is a similar mishmash of unsubstantiated
assertions and obsolete ideas. At various points, Rushton claims: that natural brush
fires would have been unknown in Eurasia during glacial periods, and because of this
human production of fire would have been more challenging there than in Africa; that
clothing and shelter were unnecessary for prehistoric African populations; that Middle
Stone Age Africans could ‘barely’ be considered big game hunters because they lacked
bows and arrows; that the greater numbers of known Cro-Magnon sites in Europe
indicate that those people were more successful than those same MSA Africans; that
‘Mongoloid’ populations entered the New World between 40,000 and 24,000 years ago
(as part of a classic evolutionary just-so story that purports to explain the lower IQ test
scores of Native American peoples); and that Africans and south-east Asians never
developed agriculture (Rushton 2000: 224–33). He further (Rushton 2000: 142) makes
use of John Baker’s (1974) racist and unsystematic list of twenty-one ‘criteria for
civilization’ – which begins to evaluate cultural advance by scoring the amount of clothing
that people wear and ends with criteria like ‘some appreciation of the fine arts’ – in
order to dismiss African and American cultural achievements. Again, examples can be
multiplied.
In the early 1990s, as today, exposure to an introductory university course in world
prehistory would equip an undergraduate to dismiss such a farrago of elementary errors
and distortions. Rushton’s account of human cultural development is one that
systematically valorizes cultural developments in Europe and (to a lesser extent) Asia,
while denigrating such developments by peoples of Africa, the Americas and Australia,
with no serious attention paid to the quality of his data. His sources for these claims
derive only to a minor degree from archaeologists and palaeoanthropologists, but come
rather from Richard Lynn and Edward Miller, respectively a psychologist and an
economist, who share his views on race and the racial inferiority of Africans. It would
be more surprising to see such claims promulgated in what is supposed to be a
legitimate study of human behaviour if they were not so common in other parts of the
book as well.
Race, reason and reality
Race, Evolution and Behavior makes use of an obsolescent, Eurocentric model of human
cultural advance, one that assumes a ‘creative explosion’ particularly in Europe at
approximately 35,000 years ago, simultaneous with the appearance of modern humans in
that area and well before equivalent developments anywhere else on the globe (Rushton
2000: 225). There is no evidence that Rushton comprehends recent archaeological research
in different areas of the world or the effects that differences in research intensity can have
upon our understanding of cultural developments world-wide. However, his book has one
very significant strength: the great claims made within it are testable archaeologically.
Such testing does not merely involve disproving the archaeological claims to be found in
his text. That is a trivial exercise, one that could be carried out with an up-to-date
introductory textbook. Instead, we can use archaeological data – with which, after all,
archaeologists claim to be able to speak authoritatively about the human past – in order to
assemble an account of African history that can then be compared with the expectations
generated by Rushton’s model.
One particular fact makes such testing possible: the magnitude of behavioural, intellectual
and social differences claimed by Rushton to exist between the three racial groups is
very great indeed. There is no great subtlety in the picture of human racial variation that
he paints. This is probably most striking for IQ test scores, as noted above. If African IQ
test scores indicate an average IQ¼70, as indeed they appear to (Lynn 1991b; Lynn and
Vanhanen 2002), and if IQ test scores are an accurate reflection of the general intelligence
of individuals and populations, then we would expect African populations to be
characterized by various degrees of cognitive deficit, with borderline mental retardation
compared to other human populations as the representative intellectual state on the
continent. We would further expect that such significant inferiority in the average mental
functioning of an entire continent’s population would have substantial social and cultural
consequences, and that those consequences would be expressed in the material traces of
the populations involved.
Rushton himself assumes that to be true. He frequently explains modern cases of
poverty, conflict, social disruption and disease in Africa as the consequences of human
evolutionary history on the continent, just as the late-twentieth century economic success
of Japan and the Asian Tigers is supposed to be due to the evolutionary history of
‘Mongoloids’. Similarly high levels of difference are evident in the other measures he uses.
Rushton makes use of r-/K-selection models that were, after all, originally developed to
compare the reproductive strategies of different species, and his definition of ‘race’ appears
to be more or less equivalent to ‘subspecies’ (Rushton 2000: 305). He is not especially
forthcoming on what cultural characteristics we might expect of a population with an
average IQ of 70. In an editorial, he notes that an IQ of 70 equates to a mental age of 11,
and says that 11-years-olds can, with supervision, drive cars, work (as child labourers) in
factories and (as child soldiers) go to war (Rushton 2004). However, presumably they
cannot design those cars or factories, or plan the wars that they find themselves involved
in. Gottfredson (2003) claims that people with IQ less than 100 are incapable of carrying
out any sort of sophisticated managerial task, including acting as merchants or bureaucrats;
this implies that such people capable of such tasks would be quite rare in African
societies generally (comprising less than 3 per cent of the population) and essentially
absent in some countries.
Some proponents of IQ tests, and of racial differences in intelligence, have noticed that
the results of IQ testing in Africa (and in other areas of the world) actually pose a
substantial challenge to the validity of those tests. The idea that the average intelligence of
Africans is severely decreased relative to that of people in other parts of the world simply
does not accord with the experience of people who have worked on or even visited that
continent. It is as if Rushton, Lynn and their colleagues were claiming that all Africans
were actually only four feet tall. If such a claim is made, and one is asked to choose
between doubting the evidence of one’s own eyes in Africa and doubting the calibration of
the ruler used in measurement, most people will doubt the ruler. The problem is even
greater than this, in fact, because testing for some African countries gives average IQ
scores of well under 70 (for example, Congo-Zaire¼65; Equatorial Guinea¼59;
Ethiopia¼63; Sierra Leone¼64; Zimbabwe¼66) (Lynn and Vanhanen 2002: 202–4,
217, 225).
Psychometricians have proposed a number of ingenious explanations in order to get
around this problem. Arthur Jensen (1998: 367–9), the dean of this research, claims that
African-American individuals with low IQ scores tend to be more socially adept than
Euroamericans with the same IQ scores (because retardation in the latter cases is more
likely to derive from organic damage), so leading observers to overestimate the intellectual
capabilities of the African-Americans in question. This has then been generalized to the
African case: Africans are supposed to be cognitively disabled, but this fact is not apparent
because of their social and verbal skills. This seems extremely unlikely: any substantial
contact would presumably pierce that facade of verbal skill, and the fact that many such
interactions take place in languages and cultural settings not native to one or both
participants would seem to indicate that verbal/social skills cannot be determinative. In
addition, this sort of post hoc explanation undercuts the relationship between IQ test
scores and the general intelligence factor (g) that is supposed to be reflected in those scores:
IQ test scores appear to mean something very different in terms of human functionality in
Africa than in other parts of the world. In any case, consideration of archaeological data
allows us to compare cultural achievements in Africa with other areas of the world,
without the distortion supposedly inflicted by verbal/social skills and over evolutionarily
significant time spans.
African history
To a degree, it seems grotesque that one must, at the beginning of the twentyfirst
century, marshal evidence for African potential for cultural progress through time.
Rushton’s model of race, behaviour and intelligence is a profoundly archaic intellectual
construct, one that has more affinities with Victorian assumptions of the inferiority of the
lesser breeds than with anything more recent. At the same time, Africanist archaeology
itself for a long time shared many of those assumptions, particularly expressed by an
unwillingness to accept evidence for indigenous African cultural advance (see the articles
in Robertshaw 1990). Thus, the roots of various kinds of sophisticated behaviour –
agriculture, iron-working and state formation are the best known – had to be sought
beyond the continent. Over the decades in which Rushton and his colleagues were
refurbishing old theories of African intellectual and cultural inferiority, archaeologists
working on the continent were developing very different models of the continent’s
history. These models do not require that African culture history exactly mirror historical
developments in other areas of the world in some unilinear evolutionary progression
(Fuglestad 1992; Neale 1986; Stahl 2005b), but they do indicate that African history is
entirely comparable to that of other regions and other continents in terms of the human
capabilities that it evokes.
Rushton (2000: 217–34) notes the palaeoanthropological debates about modern human
origins, and accepts some form of an ‘Out of Africa’ model in his book. Such a model
provides him with a well-defined, simple origin narrative for modern humans, as well as a
mechanism for moving them beyond Africa to the cognitive challenges of the non-African
world. He does not consider the evolutionary advantages that might allow modern
humans to expand beyond the African continent into areas in many cases already occupied
by other hominids. He does not explore in any detail the relationship between biological
and behavioural evolution in moderns, saying only that blade technology appeared in
Africa at 100,000 BP (itself an erroneous claim) but hastily adding that Africans at that
time were not much advanced beyond their forebears (Rushton 2000: 225). As noted
above, he accepts that significant behavioural advances among modern humans occurred
especially in Europe, and to a lesser extent in Asia, around 35,000 years ago.
Archaeological research over the last two decades provides a very different picture.
The linkage between such developments and occupation of Europe has been systematically
broken down, as evidence for advanced behaviours among humans has been found in
earlier sites and in sites well beyond that continent. Many of these data have been derived
from African contexts (see Brooks and McBrearty 2000; d’Errico et al. 2003; Henshilwood
and Marean 2003; Marean and Assefa 2005). Thus, we have a series of sophisticated bone
harpoon points, presumably used in composite weapons, at Katanda in Zaire and dating
to more than 75,000 years ago (Brooks et al. 1995; Yellen 1998); evidence for symbolic
behaviour and advanced tool production – engraved bone and ochre, perforated shell
beads, worked bone tools – at Blombos Cave in South Africa at c. 75,000 BP (d’Errico et al.
2001, 2005; Henshilwood et al. 2001, 2002); advanced composite tool technologies, long
distance acquisition of raw materials and probable symbolic behaviour in southern
African Howieson’s Poort assemblages, dated to c. 75,000–65,000 BP (Ambrose 2002;
Deacon and Wurz 1996; Wurz 1999); and bead production at Enkapune ya Muto in
Kenya at c. 41,000 BP (Ambrose 1998). It is likely that much of this behaviour has
significantly earlier roots; for a more extensive discussion, see Brooks and McBrearty
(2000) and d’Errico et al. (2003). Even figurative art, a phenomenon very frequently
associated with late Pleistocene archaeological occurrences in Europe, occurs at Apollo 11
cave in Namibia, where it is dated to 25,000–27,000 BP – that is, contemporary with early
European rock art (Vogelsang 1998).
It would be wrong to assume that the appearance of technologically sophisticated
artefacts is necessarily coterminous with the appearance of complex behavioural,
symbolic or conceptual systems, in Africa or elsewhere (Wadley 2001). Such systems may
not manifest themselves in persistent material culture (or in any realms of material
culture at all), and the cultural meanings of particular technological systems will quite
probably vary drastically across space and time. The definition and detection of behavioural
modernity among hominids is an extremely complex topic (Henshilwood and
Marean 2003), and not one that can be treated in detail in this paper. However, the
examples above very strongly indicate that Africa played a central and continuing role in
the appearance of such behavioural modernity, a role at least as important as that played
by temperate Eurasia.
Africa’s role as the birthplace of humanity is widely accepted today, by the general
public as well as by archaeologists and palaeoanthropologists. However, this acceptance is
often accompanied by the assumption that the continent has been a cultural backwater in
more recent times. Thus, even an avowedly non-racialist account of human history
(Diamond 1998: 98–101, 186–7) finds it necessary to provide an explanation of African
cultural lag, especially in agricultural development. Diamond provides a rather deterministic
explanation of plant and animal domestication based to a large extent upon the
geographical orientation of the different continents. An east-to-west Eurasian transect of
approximately 8900 kilometres, between Brussels and Shanghai, is paralleled by a similar
transect across the African Sudanian and Sahelian zones, between Dakar and Djibouti, of
about 6600km. Diamond’s model locates the early success and continuing influence of
Near Eastern domesticates in the ease with which they could spread across the long reach
of Eurasia, but does not explain why similar success did not attend Sudanian-Sahelian
agricultural systems. (It seems unlikely that a 2300km difference in east–west distances
actually produced the different Holocene continental histories that Diamond thinks exist.)
Diamond tends to underestimate the diversity and sophistication of African agricultural
systems, despite a much deeper understanding of African history than that possessed by
Philippe Rushton.
Discussions of domestication processes in Africa suffer from the fragmentary nature of
archaeological data – especially in the tropical forests of Central Africa – but the earliest
firm evidence for sub-Saharan African plant domesticates dates to just after 4000 BP in
both Africa and India (Neumann 2003). This implies African domestication of millet,
sorghum and cowpea during the fifth millennium BP. This is certainly later than was the
case in many other areas of the world. On the other hand, the variety of indigenous
African plant domesticates is very striking indeed, comparable to that from earlier centres
of domestication in the Near East and probably exceeding the diversity of plant
domesticates in East Asia and the Americas (Harlan et al. 1976), and there is no evidence
that the inspiration in their development came from beyond the continent. This parallels
the situation in New Guinea, another tropical area often assumed to be a cultural
backwater but that now appears to be a centre of domestication in its region (Denham
et al. 2004).
Economies based in large part upon animal domesticates, especially cattle and caprines,
seem to be significantly older in the Sahara and sub-Saharan West and Central
Africa, dating to the eighth millennium BP and afterward (Gifford-Gonzalez 2005: 200;
Marshall and Hildebrand 2002), and there is accumulating evidence that Saharan
populations played a significant role in cattle domestication in the early-/mid-Holocene
(Bradley et al. 1996; Grigson 1991; MacHugh et al. 1997). A stable and eminently successful
pastoralist adaptation, based upon animal domesticates, the exploitation of wild
plant and animal resources and eventually domesticated cereals, and capable of supporting
populations of significant size and complexity, can hardly be dismissed as unsophisticated.
African experiences with domestication seem entirely comparable to those of other
areas in the world – and rather innovative compared with, say, the agricultural record of
Europe.
The ability independently to develop state-level societies was another capability
traditionally denied to Africans by European authors, who tended to look for inspiration
beyond that continent and especially in the Mediterranean Basin and western Asia
(Delafosse 1912, I: 207; Desplagnes 1906: 544–6; Murdock 1959; Palmer 1936). Probably
the most developed example of this attitude was Charles Seligman’s (1957: 10, 43)
‘Hamitic hypothesis’, which traced virtually every cultural advance in sub-Saharan Africa
to light-skinned immigrants from the north and north east, or to later contacts with
Semitic populations.
Again, research across the continent over the last three decades decisively disproves this
point of view. The literature on this topic is expanding rapidly: there is, however, no doubt
that complex polities in the Nile Valley (O’Connor 1993; Welsby 1998), in West Africa
(Gronenborn 2001; Holl 1985; MacEachern 2005; McIntosh 1991, 1999; McIntosh and
McIntosh 1984), in North-east Africa (Curtis 2004; Fattovich 2000; Munro-Hay 1993), in
East and Central Africa (de Maret 1999; Kusimba 1999; Schoenbrun 1999; Sutton 1993)
and in south-eastern Africa (Huffman 1996; Pikirayi 2000; Sinclair et al. 1993) were indeed
African, developing according to their own internal logics. The social and political
hierarchies, the external relations and the economic and trading systems of these states
were entirely comparable with those of similar polities on other continents, and were
frequently recognized as such by European visitors before corrosively racist views of
Africans had time to develop (cf. Brooks 1993; Northrup 2002). They did not appear in
isolation – indeed, neither did states in other parts of the world, including Europe – and,
again, they were not mirror-images of states in those other regions (cf. McIntosh 1999).
The culture history of the continent is one of change and development comparable to that
of Europe and Asia, one where particular cultural systems – the development of external
symbolic systems, agriculture or states, for example – occur in particular areas, which in
turn affect neighbouring regions in different ways. This paper provides only an extremely
cursory survey of those data, on a limited number of topics, but more broad-based
examination (cf. Stahl 2005a) would provide the same results. Such results provide no
basis for the differentiation of homogeneous continental blocs of humanity, still less for
the ranking of those blocs one against the other.
Interpretations
We are thus presented with a problem. The picture provided by African archaeological
data is entirely incommensurate with claims by Rushton and his colleagues that African
populations suffer severe cognitive deficits or other behavioural disadvantages when
compared with human populations from Europe and Asia. There is no evidence in those
data that Africans as a continental population suffer from the degree of mental retardation
that would be indicated by an IQ of 70, or from any degree of mental retardation at all.
Both of these data sets are internally consistent: IQ test scores for African populations do
indeed yield an average IQ of roughly 70, while the archaeological (and historical)
evidence indicates that Africans have the same cognitive and cultural abilities as people
living in other regions of the world, over evolutionary time spans and today. How may we
reconcile these results?
A number of possibilities present themselves. In the first place, one might claim that the
most intelligent people in African societies (perhaps the 3 per cent of the population with
IQ scores greater than 100, as indicated by Gottfredson (2003)) have acted as a tiny
‘cognitive elite’, themselves almost entirely responsible for African cultural advances. The
existence of such an elite would lead to an overestimate of the cultural and behavioural
similarities between Africa and other continental regions. This seems unlikely on a number
of levels. It implies drastically different intellectual arrangements in African societies than
in societies in other areas of the world, and there are no archaeological, ethnographic or
other data indicating that such differences exist. It would also divorce IQ test scores from
social and cultural consequences to an extent resisted by psychometricians on both
theoretical and practical grounds. It should be noted that neither Rushton (2000) nor
Lynn and Vanhanen (2002) accept this explanation, because they claim that there is a good
correlation between IQ, behaviour and cultural indicators on both a continental and a
national level.
In the second place, we must remember that the evidentiary basis for many of these IQ
tests is extremely weak, and in some cases the data are presented quite selectively. Some of
the tests, like the Army Beta administered in the 1920s to South Africans, are known to be
severely culture-bound. Claims made by Lynn (1991b; Lynn and Vanhanen 2002) and
Rushton (2000) about the intelligence of different groups of South Africans also ignore the
very significant debates about mental testing in South Africa during the twentieth century
(Dubow 1995: 197–245), and the fact that much of this debate involved the IQ test scores
of (usually Afrikaans-speaking) ‘poor whites’. A sense of the quality of reporting of these
tests comes from Lynn’s (1991b; see also Lynn and Vanhanen 2002: 219) description of
tests administered by Owen (1989) on different ‘racial’ groups in South Africa as ‘[t]he best
single study of the Negroid intelligence’. Owen himself (1989: 60, 62–8) indicated
significant problems with these tests, many involving language difficulties experienced by
the African test-takers, and did not assign IQ scores for the results. Similar kinds of
problems, where authors’ cautions about test circumstances are ignored by Lynn and
Vanhanen, exist in a number of the other African cases. At the same time, it is unlikely
that straightforward bias could explain all of the test score results in question.
A third possibility also presents itself, one derived from another realm of debate about
IQ test results. The African case is not the only circumstance in which unaccounted-for
differences exist in IQ test scores across cultural boundaries. James Flynn (1984, 1987,
1998, 1999; see also Dickens and Flynn 2001) has documented a steady rise in IQ test
scores from the late nineteenth century onward in countries where longitudinal data exist,
an increase now widely known as the Flynn Effect. (Longitudinal data are unfortunately
available almost exclusively from Western countries.) Depending upon the test, this rise
varies from less than 10 points to as much as 20 points per generation, with the greatest
increase in ‘culture-reduced’ tests like the Raven’s Progressive Matrices. This effect is both
reliable and predictable: it also indicates significant gains in intelligence over a period too
short for any evolutionary effects to come in to play. Explanations for the Flynn Effect
vary and none appears completely satisfactory, but there seems little doubt that it is due to
some combination of environmental and cultural factors at play in Western societies,
factors that remain significant even on tests where cultural influences are supposed to be
largely excluded.
Taken at face value, the Flynn Effect implies that the average North American adult
living about a century ago would have had an IQ of approximately 75 in modern terms, a
value closely comparable to that derived for twentieth-century African populations by
Rushton, Lynn and their associates. Mid-nineteenth-century North Americans would
have been even more deficient mentally. This is a nonsensical result, and is widely accepted
as such; no North American believes that our great-grandparents were mentally deficient.
As two enthusiasts for racial comparison in intelligence testing, Richard Herrnstein and
Charles Murray, said a decade ago in The Bell Curve:
"Whether one looks at the worlds of science, literature, politics, or the arts, one does not
get the impression that the top of the IQ distribution is filled with more subtle,
insightful or powerful intellects than it was in our grandparents’ day. . . . No one is
suggesting, for example, that the IQ of the average American in 1776 was 30 or that it
will be 150 a century from now.+ -(Herrnstein and Murray 1994: 308–9)
In this case, Herrnstein and Murray are absolutely right. Moreover, the evaluative criteria
used by Herrnstein and Murray to judge this claim involve the examination of cultural
accomplishment, as does the present paper. This is the only way to test such claims for
societies before the invention of intelligence tests. The fact that similar nonsensical results
concerning Africans are widely promulgated in the psychometric literature may indicate
simple ignorance of African societies or a more pernicious readiness to place Africans
below other human populations in a ranking of inherent human worth.
The Flynn Effect is a measure of IQ test performance across a substantial cultural
divide, with biology held more or less constant. This divide exists in time: it looks from the
early twenty-first century back toward the late nineteenth century. The IQ testing that has
taken place in Africa since the 1920s has taken place across a comparable cultural divide,
one from economically and politically dominant Western societies and test designers to
African societies and individuals in almost all cases at significant economic, social, cultural
and/or political disadvantage. In this case, the cultural divide exists both in time (because
the tests have been administered since the 1920s) and in space. The parallels between time
and race have been remarked upon by Flynn (1999: 14–15; see also Dickens and Flynn
2001). It may well be the case that depressed African IQ test score results are best
explained by a combination of obvious test bias and the subtle and additive environmental
differences that produce variation in even ‘culture-reduced’ tests like the Raven’s
Progressive Matrices, in a geographic analogy to the Flynn Effect. The archaeological
data would seem decisively to contradict the claim that these differences stem from
reduced cognitive potentials in African populations.
Conclusions
This paper presents the juxtaposition of two data sets, one archaeological and one
behavioural and psychometric. It is now a commonplace in the comparative psychometric
literature to claim that low IQ test scores among African populations indicate severely
diminished average intelligence among those populations. Rushton (2000) places these
claims in a behavioural and evolutionary context, one paralleled by similar explanations
applied to poor and relatively powerless populations in other parts of the world and
supplemented by data of other kinds. Rushton’s model posits quite major behavioural
differences among the different continental populations, and especially between tropical
African populations and the inhabitants of temperate and Arctic Eurasia. The magnitude
of these differences is such that they should be detectable archaeologically, and indeed
Rushton presents archaeological evidence that he believes bolsters his case. His archaeological
interpretations are for the most part obsolete and/or erroneous.
However, Rushton is probably correct in claiming that such a magnitude of racial
differences should be demonstrable archaeologically. Archaeological data provide an
independent test of his hypothesis, one not subject to the obscuring effects associated with
modern mental testing and interpretation. Examination of archaeological data on the culture
history of African populations, and comparison of those data with data from other parts of
the world, yields no evidence for the behavioural and cognitive disparities claimed by
Rushton. African cultural history is entirely comparable with that of other regions of the
world, not in terms of lockstep evolutionary schema but rather in the evident sophistication
with which African populations have met the challenges of their physical and social
environments through time. To interpret the conflict between these two data sets, it may be
useful to examine possible confounding factors in the behavioural and psychometric data.
The behavioural data are quite variable and often of poor quality, but it is striking to note
that the field of intelligence testing is grappling with a phenomenon analogous to continental
differences in IQ test scores – the Flynn Effect. In both cases, testing across cultural
boundaries yields results that systematically disadvantage populations culturally removed
from our own, results that on their face defy logic. It is now up to intelligence researchers to
identify the confounding effects in their tests, and let archaeologists and other researchers get
back to looking at the Africa that actually exists today, and existed in the past."
knol.google.com/k/mainstream-academic-research/human-biodiversity-racial-evolutionary/3q8x30897t2cs/28#view
Africanist archaeology and ancient IQ: racial science and cultural evolution in the twenty-first century
Scott MacEachern
World Archaeology. (2006). Vol. 38(1): 72-92 Race, Racism and Archaeology
----
Abstract
Over the last two decades, a number of psychometric researchers have claimed that very substantial
differences in intelligence exist among modern human racial groups, as these groups are traditionally
defined. According to these researchers, African populations suffer severe cognitive deficits when
compared to other modern humans. Philippe Rushton, particularly, places these claimed mental
deficits in an evolutionary context, advancing environmental explanations for such deficits and
asserting that such cognitive differences existed prehistorically as well. Such substantial cognitive
differences should be evident in human behavioural patterns, and thus in the archaeological record.
Archaeological data can thus be used to test these claims about human evolutionary development
and modern human cognitive difference. Examination of the archaeological record does not support
the claims made by these researchers. This suggests that regional differences in IQ test score results
should not be ascribed to variations in human evolutionary development.
---------------------
EXCERPT
Behaviour, selection and African environments
There are three basic elements in Rushton’s theorization of human difference. First, he
amasses a large amount of material on physical, behavioural and social differences
between the groups that form the popularly accepted triptych of human races: ‘Negroids’,
‘Caucasoids’ and ‘Mongoloids’ (Rushton 2000: 17–183; see also Rushton and Bogaert
1987: 265–8). The scope of these topics is very wide: it includes material on brain size, IQ
(and related) test scores, dental development, speed of sexual maturation, age of first
intercourse, life span, number of sexual fantasies, penis size, number of multiple births,
permissive attitudes toward sex, aggressiveness, law-abidingness, ‘mental durability’,
AIDS rates, cultural achievements and much more. Most striking, and most relevant for
this paper, is Rushton’s (2000: 133–7; see also Rushton and Skuy 2000) acceptance of
Richard Lynn’s (1991b, 2003; Lynn and Vanhanen 2002: 197–225) claim that the average
IQ test score of African populations is approximately IQ¼70. In the psychiatric literature,
this correlates to a state of borderline mental retardation (American Psychiatric
Association 1994), implying that the great majority of African people living today suffer
some degree of cognitive deficit, ranging from mild to very severe. Rushton ranks the three
races on the basis of these assembled characteristics (Rushton 2000: 5, 119, 148, 152, 162,
166, 168, 171, 214).
He has stated on many occasions that this is a straightforward scientific description of
humanity, but it is clear that any taxonomy describing Africans, for example, as less
intelligent, more promiscuous, less altruistic, more aggressive, less law abiding,
investing less effort in child-rearing and less culturally developed than Europeans
or Asians – as Rushton’s taxonomy does – is also a ranking of fundamental human worth.
Rushton’s accumulation of these data and the uses that he makes of them have been
very stringently critiqued (see, for example, Brace 1996; Czernovsky 1995; Graves 2002b;
Lieberman 2001; Peters 1995), and that critique need not be repeated here. However, it
should be noted that the sources, characteristics and quality of his data are diverse, and
often of extremely low reliability. Rushton’s use of late Victorian travel pornography
(‘French Army Surgeon’ 1896) as a central reference on human sexual characteristics and
behaviours is perhaps the most egregious use of poor-quality data, but it is very far from
being the only one. His data on racial differences in intelligence test scores is taken from
wildly disparate sources ranging over the last century (Rushton 2000: 38–9, 135–7),
including sources using techniques that even advocates of IQ testing have dismissed as
‘primitive’ and unreliable (Jensen 1988). Many of his demographic data – on longevity and
reproductive rates, for example – have been subject to large-scale fluctuations even within
populations and in relatively short periods of time, making it unlikely that such
comparative measures have any evolutionary underpinnings.
The second element in Rushton’s theorization of racial differentiation involves
accounting for the biological basis of the data accumulated. He hypothesizes that the
‘Mongoloid’, ‘Caucasoid’ and ‘Negroid’ racial groups have been subjected to different
natural selection pressures over evolutionary timescales, and that because of this these
racial groups have evolved different life-history strategies. This hypothesis makes use of
the r-/K-selection schema, developed by MacArthur and Wilson (1967) as a method of
modeling density-dependent natural selection. The consequences of these different
reproductive strategies are, according to Rushton, to be found in the data on physical,
behavioural and social differences between races amassed in earlier sections of the book. In
the formulation used by Rushton, r-selected species, adapted to unstable, rapidly
fluctuating environments, evolve reproduction strategies with prolific production of
offspring and relatively low parental investment in individual offspring, while K-selected
species, adapted to more stable environments, produce relatively lower numbers of
offspring but invest more care in each. Rushton (1987, 2000: 199–216; see also Rushton
and Bogaert 1987) then claims that ‘Negroid’ populations, in particular, are more
r-selected than are ‘Caucasoid’ populations, which are in turn more r-selected than
‘Mongoloid’ populations. In essence, Africans are held to invest bodily resources more
heavily in sex and impulsiveness/aggression, while Europeans and Asians are supposed to
invest such resources more heavily in intelligence, altruism and restraint. While Rushton
does claim that ‘Negroids’ are only relatively r-selected, the data that he amasses give the
impression of very substantial differences between these racial groups, an impression
eagerly seized by racists around the world.
It might be objected that a concentration on ‘Negroid’ characteristics, vis-a`-vis those
of the other major races, imputes an unfair racializing subtext to Rushton’s work, because
he situates the three major races along an r- to K-selected continuum: ‘Negroids’ –
‘Caucasoids’ – ‘Mongoloids’. However, in practice (see below), ‘Caucasoids’ and
‘Mongoloids’ are lumped together as temperate-/cold-climate races against the tropically
adapted ‘Negroids’ throughout this work (e.g. Rushton 2000: 199, 262). In other writings
(Rushton and Horowitz 1999), Rushton speculates upon temperamental differences that
might disadvantage ‘Orientals’ in scientific and cultural achievements vis-a`-vis ‘Whites’.
In the theories of modern racial scientists, the higher IQ scores of ‘Mongoloids’ cannot be
held to imply significant intellectual superiority over ‘Caucasoids’.
This element of Rushton’s work has also been strenuously critiqued (Anderson 1991;
Graves 2002a, 2002b; Silverman 1990; Weizman et al. 1990). There are very serious doubts
about the utility of the r-/K-selection model as an explanation for behavioural differences
among and especially within animal species, especially when the concept is generalized far
beyond its origins (Boyce 1984; Stearns 1992). There exist many cases in which the
predictions of this theory do not hold (Graves 2002a: 66–7), and a number of characteristics
of Rushton’s three racial groups – most notably body size – do not fit the model.
The linkages between r-/K-selection and the characteristics that Rushton associates with
those different forms of selection are quite unclear (Weizman et al. 1990: 4–5). Finally,
Rushton has never satisfactorily established the environmental circumstances in which his
different racial groups were exposed to these different selection pressures. It is difficult to
avoid the conclusion that Rushton turned to an r-/K-selection model because of a fairly
impressionistic resemblance between that model and his image of ‘racial’ variability, and
that he then tuned the model to fit his assumptions.
It is at this point – when we begin to speak about the ancestors of modern humans as
evolving populations in particular environments – that palaeoanthropology and archaeology
become important to Rushton’s account. He claims that ‘Negroid’ populations
differ from ‘Caucasoids’ and ‘Mongoloids’ because the former evolved only in tropical
Africa, while the latter groups moved out of Africa into colder climates (Rushton 2000:
199, 217–33, 262–4). There are two components to this explanation: tropical African
environments are less stable and less predictable than are temperate and cold environments
(thus favouring r-selected strategies of prolific reproduction and low parental
investment in offspring), and in addition temperate and cold climates posed cognitive
demands upon ‘Caucasoid’ and ‘Mongoloid’ populations not experienced by those
peoples living in balmier climes. It should be emphasized that only the first of these
components involves r-/K-selection theory; the second derives in large part from speculations
about the evolution of intelligence published by Richard Lynn (1987, 1991a).
(Lynn (2003: 141), in fact, seems to be claiming that modern Africans are cognitively
unchanged descendants of the hominids that occupied Africa 250,000 years ago.) This
bipartite explanation is supplemented by archaeological data that Rushton claims support
his thesis.
Less attention has been paid to this last component of Rushton’s model than to other
elements of his work (but see Graves 2002a: 71–4; Lieberman 2001: 79), in part because
there are so many other obvious targets for criticism in his work and in part because most
of this critique has come from psychologists and biological anthropologists. It remains,
however, central to Rushton’s accounting of difference between his three racial groups. He
is concerned with producing an integrated explanation of the differences between those
groups, and to do this using r-/K-selection theory he must account for the origins of these
differences in an evolutionary context. This requires engagement with archaeological and
palaeoanthropological data. There is also a long tradition of appeals to environmental
differences in racialist descriptions of African peoples. In almost all such cases, Africans
are held to be cognitively disadvantaged, either because African environments are so
benevolent that they provide a reduced cognitive challenge to African populations (cf.
Herder 1968: 297; Kant 1997 [1775]: 46), or because those environments are so hostile to
humanity that they inhibit the intellectual development of those populations (cf. Cornelius
de Pauw in Duchet 1969: 123). Rushton’s explanation uniquely combines elements of both
of these models.
His approach shares another characteristic with many eighteenth-century accounts of
human racial variation: it treats continental land-masses as undifferentiated geographical
units, each characterized by a particular set of prevailing environmental conditions. Thus,
African environments are collapsed to subtropical savannas, which are prone to
unpredictable droughts and ‘devastating viral, bacterial, and parasitic diseases’ (Rushton
2000: 231), while Europe and Asia are similarly characterized as cold (but predictable) arid
tundra and glacial landscapes.We are never told why all African savannas are supposed to
be prone to unpredictable droughts or the reason for believing that tropical and
subtropical plant foods are going to be available year-round (Rushton 2000: 228). We are
not told why dispersed tropical foragers should be prone to devastating diseases that are
now associated with higher-density, agricultural lifeways (see, for example, Armelagos
et al. 1991; Barnes 2005; Tishkoff et al. 2001). We are never told why Rushton believes that
‘Mongoloids’ evolved in Siberia (Rushton 2000: 229), rather than further south in Asia.
We are never told why environmental predictability is characteristic of temperate and
Arctic climates, while unpredictability reigns in Africa. (Apparently, the only sort of
seasonality that Rushton recognizes is that between summers and winters; the equally
predictable and often very challenging cycling between wet and dry seasons in many areas
of Africa is unknown to him.) Examples could be multiplied. Rushton simply asserts that
racial evolution took place as he says, without providing evidence.
His archaeological and historical evidence is a similar mishmash of unsubstantiated
assertions and obsolete ideas. At various points, Rushton claims: that natural brush
fires would have been unknown in Eurasia during glacial periods, and because of this
human production of fire would have been more challenging there than in Africa; that
clothing and shelter were unnecessary for prehistoric African populations; that Middle
Stone Age Africans could ‘barely’ be considered big game hunters because they lacked
bows and arrows; that the greater numbers of known Cro-Magnon sites in Europe
indicate that those people were more successful than those same MSA Africans; that
‘Mongoloid’ populations entered the New World between 40,000 and 24,000 years ago
(as part of a classic evolutionary just-so story that purports to explain the lower IQ test
scores of Native American peoples); and that Africans and south-east Asians never
developed agriculture (Rushton 2000: 224–33). He further (Rushton 2000: 142) makes
use of John Baker’s (1974) racist and unsystematic list of twenty-one ‘criteria for
civilization’ – which begins to evaluate cultural advance by scoring the amount of clothing
that people wear and ends with criteria like ‘some appreciation of the fine arts’ – in
order to dismiss African and American cultural achievements. Again, examples can be
multiplied.
In the early 1990s, as today, exposure to an introductory university course in world
prehistory would equip an undergraduate to dismiss such a farrago of elementary errors
and distortions. Rushton’s account of human cultural development is one that
systematically valorizes cultural developments in Europe and (to a lesser extent) Asia,
while denigrating such developments by peoples of Africa, the Americas and Australia,
with no serious attention paid to the quality of his data. His sources for these claims
derive only to a minor degree from archaeologists and palaeoanthropologists, but come
rather from Richard Lynn and Edward Miller, respectively a psychologist and an
economist, who share his views on race and the racial inferiority of Africans. It would
be more surprising to see such claims promulgated in what is supposed to be a
legitimate study of human behaviour if they were not so common in other parts of the
book as well.
Race, reason and reality
Race, Evolution and Behavior makes use of an obsolescent, Eurocentric model of human
cultural advance, one that assumes a ‘creative explosion’ particularly in Europe at
approximately 35,000 years ago, simultaneous with the appearance of modern humans in
that area and well before equivalent developments anywhere else on the globe (Rushton
2000: 225). There is no evidence that Rushton comprehends recent archaeological research
in different areas of the world or the effects that differences in research intensity can have
upon our understanding of cultural developments world-wide. However, his book has one
very significant strength: the great claims made within it are testable archaeologically.
Such testing does not merely involve disproving the archaeological claims to be found in
his text. That is a trivial exercise, one that could be carried out with an up-to-date
introductory textbook. Instead, we can use archaeological data – with which, after all,
archaeologists claim to be able to speak authoritatively about the human past – in order to
assemble an account of African history that can then be compared with the expectations
generated by Rushton’s model.
One particular fact makes such testing possible: the magnitude of behavioural, intellectual
and social differences claimed by Rushton to exist between the three racial groups is
very great indeed. There is no great subtlety in the picture of human racial variation that
he paints. This is probably most striking for IQ test scores, as noted above. If African IQ
test scores indicate an average IQ¼70, as indeed they appear to (Lynn 1991b; Lynn and
Vanhanen 2002), and if IQ test scores are an accurate reflection of the general intelligence
of individuals and populations, then we would expect African populations to be
characterized by various degrees of cognitive deficit, with borderline mental retardation
compared to other human populations as the representative intellectual state on the
continent. We would further expect that such significant inferiority in the average mental
functioning of an entire continent’s population would have substantial social and cultural
consequences, and that those consequences would be expressed in the material traces of
the populations involved.
Rushton himself assumes that to be true. He frequently explains modern cases of
poverty, conflict, social disruption and disease in Africa as the consequences of human
evolutionary history on the continent, just as the late-twentieth century economic success
of Japan and the Asian Tigers is supposed to be due to the evolutionary history of
‘Mongoloids’. Similarly high levels of difference are evident in the other measures he uses.
Rushton makes use of r-/K-selection models that were, after all, originally developed to
compare the reproductive strategies of different species, and his definition of ‘race’ appears
to be more or less equivalent to ‘subspecies’ (Rushton 2000: 305). He is not especially
forthcoming on what cultural characteristics we might expect of a population with an
average IQ of 70. In an editorial, he notes that an IQ of 70 equates to a mental age of 11,
and says that 11-years-olds can, with supervision, drive cars, work (as child labourers) in
factories and (as child soldiers) go to war (Rushton 2004). However, presumably they
cannot design those cars or factories, or plan the wars that they find themselves involved
in. Gottfredson (2003) claims that people with IQ less than 100 are incapable of carrying
out any sort of sophisticated managerial task, including acting as merchants or bureaucrats;
this implies that such people capable of such tasks would be quite rare in African
societies generally (comprising less than 3 per cent of the population) and essentially
absent in some countries.
Some proponents of IQ tests, and of racial differences in intelligence, have noticed that
the results of IQ testing in Africa (and in other areas of the world) actually pose a
substantial challenge to the validity of those tests. The idea that the average intelligence of
Africans is severely decreased relative to that of people in other parts of the world simply
does not accord with the experience of people who have worked on or even visited that
continent. It is as if Rushton, Lynn and their colleagues were claiming that all Africans
were actually only four feet tall. If such a claim is made, and one is asked to choose
between doubting the evidence of one’s own eyes in Africa and doubting the calibration of
the ruler used in measurement, most people will doubt the ruler. The problem is even
greater than this, in fact, because testing for some African countries gives average IQ
scores of well under 70 (for example, Congo-Zaire¼65; Equatorial Guinea¼59;
Ethiopia¼63; Sierra Leone¼64; Zimbabwe¼66) (Lynn and Vanhanen 2002: 202–4,
217, 225).
Psychometricians have proposed a number of ingenious explanations in order to get
around this problem. Arthur Jensen (1998: 367–9), the dean of this research, claims that
African-American individuals with low IQ scores tend to be more socially adept than
Euroamericans with the same IQ scores (because retardation in the latter cases is more
likely to derive from organic damage), so leading observers to overestimate the intellectual
capabilities of the African-Americans in question. This has then been generalized to the
African case: Africans are supposed to be cognitively disabled, but this fact is not apparent
because of their social and verbal skills. This seems extremely unlikely: any substantial
contact would presumably pierce that facade of verbal skill, and the fact that many such
interactions take place in languages and cultural settings not native to one or both
participants would seem to indicate that verbal/social skills cannot be determinative. In
addition, this sort of post hoc explanation undercuts the relationship between IQ test
scores and the general intelligence factor (g) that is supposed to be reflected in those scores:
IQ test scores appear to mean something very different in terms of human functionality in
Africa than in other parts of the world. In any case, consideration of archaeological data
allows us to compare cultural achievements in Africa with other areas of the world,
without the distortion supposedly inflicted by verbal/social skills and over evolutionarily
significant time spans.
African history
To a degree, it seems grotesque that one must, at the beginning of the twentyfirst
century, marshal evidence for African potential for cultural progress through time.
Rushton’s model of race, behaviour and intelligence is a profoundly archaic intellectual
construct, one that has more affinities with Victorian assumptions of the inferiority of the
lesser breeds than with anything more recent. At the same time, Africanist archaeology
itself for a long time shared many of those assumptions, particularly expressed by an
unwillingness to accept evidence for indigenous African cultural advance (see the articles
in Robertshaw 1990). Thus, the roots of various kinds of sophisticated behaviour –
agriculture, iron-working and state formation are the best known – had to be sought
beyond the continent. Over the decades in which Rushton and his colleagues were
refurbishing old theories of African intellectual and cultural inferiority, archaeologists
working on the continent were developing very different models of the continent’s
history. These models do not require that African culture history exactly mirror historical
developments in other areas of the world in some unilinear evolutionary progression
(Fuglestad 1992; Neale 1986; Stahl 2005b), but they do indicate that African history is
entirely comparable to that of other regions and other continents in terms of the human
capabilities that it evokes.
Rushton (2000: 217–34) notes the palaeoanthropological debates about modern human
origins, and accepts some form of an ‘Out of Africa’ model in his book. Such a model
provides him with a well-defined, simple origin narrative for modern humans, as well as a
mechanism for moving them beyond Africa to the cognitive challenges of the non-African
world. He does not consider the evolutionary advantages that might allow modern
humans to expand beyond the African continent into areas in many cases already occupied
by other hominids. He does not explore in any detail the relationship between biological
and behavioural evolution in moderns, saying only that blade technology appeared in
Africa at 100,000 BP (itself an erroneous claim) but hastily adding that Africans at that
time were not much advanced beyond their forebears (Rushton 2000: 225). As noted
above, he accepts that significant behavioural advances among modern humans occurred
especially in Europe, and to a lesser extent in Asia, around 35,000 years ago.
Archaeological research over the last two decades provides a very different picture.
The linkage between such developments and occupation of Europe has been systematically
broken down, as evidence for advanced behaviours among humans has been found in
earlier sites and in sites well beyond that continent. Many of these data have been derived
from African contexts (see Brooks and McBrearty 2000; d’Errico et al. 2003; Henshilwood
and Marean 2003; Marean and Assefa 2005). Thus, we have a series of sophisticated bone
harpoon points, presumably used in composite weapons, at Katanda in Zaire and dating
to more than 75,000 years ago (Brooks et al. 1995; Yellen 1998); evidence for symbolic
behaviour and advanced tool production – engraved bone and ochre, perforated shell
beads, worked bone tools – at Blombos Cave in South Africa at c. 75,000 BP (d’Errico et al.
2001, 2005; Henshilwood et al. 2001, 2002); advanced composite tool technologies, long
distance acquisition of raw materials and probable symbolic behaviour in southern
African Howieson’s Poort assemblages, dated to c. 75,000–65,000 BP (Ambrose 2002;
Deacon and Wurz 1996; Wurz 1999); and bead production at Enkapune ya Muto in
Kenya at c. 41,000 BP (Ambrose 1998). It is likely that much of this behaviour has
significantly earlier roots; for a more extensive discussion, see Brooks and McBrearty
(2000) and d’Errico et al. (2003). Even figurative art, a phenomenon very frequently
associated with late Pleistocene archaeological occurrences in Europe, occurs at Apollo 11
cave in Namibia, where it is dated to 25,000–27,000 BP – that is, contemporary with early
European rock art (Vogelsang 1998).
It would be wrong to assume that the appearance of technologically sophisticated
artefacts is necessarily coterminous with the appearance of complex behavioural,
symbolic or conceptual systems, in Africa or elsewhere (Wadley 2001). Such systems may
not manifest themselves in persistent material culture (or in any realms of material
culture at all), and the cultural meanings of particular technological systems will quite
probably vary drastically across space and time. The definition and detection of behavioural
modernity among hominids is an extremely complex topic (Henshilwood and
Marean 2003), and not one that can be treated in detail in this paper. However, the
examples above very strongly indicate that Africa played a central and continuing role in
the appearance of such behavioural modernity, a role at least as important as that played
by temperate Eurasia.
Africa’s role as the birthplace of humanity is widely accepted today, by the general
public as well as by archaeologists and palaeoanthropologists. However, this acceptance is
often accompanied by the assumption that the continent has been a cultural backwater in
more recent times. Thus, even an avowedly non-racialist account of human history
(Diamond 1998: 98–101, 186–7) finds it necessary to provide an explanation of African
cultural lag, especially in agricultural development. Diamond provides a rather deterministic
explanation of plant and animal domestication based to a large extent upon the
geographical orientation of the different continents. An east-to-west Eurasian transect of
approximately 8900 kilometres, between Brussels and Shanghai, is paralleled by a similar
transect across the African Sudanian and Sahelian zones, between Dakar and Djibouti, of
about 6600km. Diamond’s model locates the early success and continuing influence of
Near Eastern domesticates in the ease with which they could spread across the long reach
of Eurasia, but does not explain why similar success did not attend Sudanian-Sahelian
agricultural systems. (It seems unlikely that a 2300km difference in east–west distances
actually produced the different Holocene continental histories that Diamond thinks exist.)
Diamond tends to underestimate the diversity and sophistication of African agricultural
systems, despite a much deeper understanding of African history than that possessed by
Philippe Rushton.
Discussions of domestication processes in Africa suffer from the fragmentary nature of
archaeological data – especially in the tropical forests of Central Africa – but the earliest
firm evidence for sub-Saharan African plant domesticates dates to just after 4000 BP in
both Africa and India (Neumann 2003). This implies African domestication of millet,
sorghum and cowpea during the fifth millennium BP. This is certainly later than was the
case in many other areas of the world. On the other hand, the variety of indigenous
African plant domesticates is very striking indeed, comparable to that from earlier centres
of domestication in the Near East and probably exceeding the diversity of plant
domesticates in East Asia and the Americas (Harlan et al. 1976), and there is no evidence
that the inspiration in their development came from beyond the continent. This parallels
the situation in New Guinea, another tropical area often assumed to be a cultural
backwater but that now appears to be a centre of domestication in its region (Denham
et al. 2004).
Economies based in large part upon animal domesticates, especially cattle and caprines,
seem to be significantly older in the Sahara and sub-Saharan West and Central
Africa, dating to the eighth millennium BP and afterward (Gifford-Gonzalez 2005: 200;
Marshall and Hildebrand 2002), and there is accumulating evidence that Saharan
populations played a significant role in cattle domestication in the early-/mid-Holocene
(Bradley et al. 1996; Grigson 1991; MacHugh et al. 1997). A stable and eminently successful
pastoralist adaptation, based upon animal domesticates, the exploitation of wild
plant and animal resources and eventually domesticated cereals, and capable of supporting
populations of significant size and complexity, can hardly be dismissed as unsophisticated.
African experiences with domestication seem entirely comparable to those of other
areas in the world – and rather innovative compared with, say, the agricultural record of
Europe.
The ability independently to develop state-level societies was another capability
traditionally denied to Africans by European authors, who tended to look for inspiration
beyond that continent and especially in the Mediterranean Basin and western Asia
(Delafosse 1912, I: 207; Desplagnes 1906: 544–6; Murdock 1959; Palmer 1936). Probably
the most developed example of this attitude was Charles Seligman’s (1957: 10, 43)
‘Hamitic hypothesis’, which traced virtually every cultural advance in sub-Saharan Africa
to light-skinned immigrants from the north and north east, or to later contacts with
Semitic populations.
Again, research across the continent over the last three decades decisively disproves this
point of view. The literature on this topic is expanding rapidly: there is, however, no doubt
that complex polities in the Nile Valley (O’Connor 1993; Welsby 1998), in West Africa
(Gronenborn 2001; Holl 1985; MacEachern 2005; McIntosh 1991, 1999; McIntosh and
McIntosh 1984), in North-east Africa (Curtis 2004; Fattovich 2000; Munro-Hay 1993), in
East and Central Africa (de Maret 1999; Kusimba 1999; Schoenbrun 1999; Sutton 1993)
and in south-eastern Africa (Huffman 1996; Pikirayi 2000; Sinclair et al. 1993) were indeed
African, developing according to their own internal logics. The social and political
hierarchies, the external relations and the economic and trading systems of these states
were entirely comparable with those of similar polities on other continents, and were
frequently recognized as such by European visitors before corrosively racist views of
Africans had time to develop (cf. Brooks 1993; Northrup 2002). They did not appear in
isolation – indeed, neither did states in other parts of the world, including Europe – and,
again, they were not mirror-images of states in those other regions (cf. McIntosh 1999).
The culture history of the continent is one of change and development comparable to that
of Europe and Asia, one where particular cultural systems – the development of external
symbolic systems, agriculture or states, for example – occur in particular areas, which in
turn affect neighbouring regions in different ways. This paper provides only an extremely
cursory survey of those data, on a limited number of topics, but more broad-based
examination (cf. Stahl 2005a) would provide the same results. Such results provide no
basis for the differentiation of homogeneous continental blocs of humanity, still less for
the ranking of those blocs one against the other.
Interpretations
We are thus presented with a problem. The picture provided by African archaeological
data is entirely incommensurate with claims by Rushton and his colleagues that African
populations suffer severe cognitive deficits or other behavioural disadvantages when
compared with human populations from Europe and Asia. There is no evidence in those
data that Africans as a continental population suffer from the degree of mental retardation
that would be indicated by an IQ of 70, or from any degree of mental retardation at all.
Both of these data sets are internally consistent: IQ test scores for African populations do
indeed yield an average IQ of roughly 70, while the archaeological (and historical)
evidence indicates that Africans have the same cognitive and cultural abilities as people
living in other regions of the world, over evolutionary time spans and today. How may we
reconcile these results?
A number of possibilities present themselves. In the first place, one might claim that the
most intelligent people in African societies (perhaps the 3 per cent of the population with
IQ scores greater than 100, as indicated by Gottfredson (2003)) have acted as a tiny
‘cognitive elite’, themselves almost entirely responsible for African cultural advances. The
existence of such an elite would lead to an overestimate of the cultural and behavioural
similarities between Africa and other continental regions. This seems unlikely on a number
of levels. It implies drastically different intellectual arrangements in African societies than
in societies in other areas of the world, and there are no archaeological, ethnographic or
other data indicating that such differences exist. It would also divorce IQ test scores from
social and cultural consequences to an extent resisted by psychometricians on both
theoretical and practical grounds. It should be noted that neither Rushton (2000) nor
Lynn and Vanhanen (2002) accept this explanation, because they claim that there is a good
correlation between IQ, behaviour and cultural indicators on both a continental and a
national level.
In the second place, we must remember that the evidentiary basis for many of these IQ
tests is extremely weak, and in some cases the data are presented quite selectively. Some of
the tests, like the Army Beta administered in the 1920s to South Africans, are known to be
severely culture-bound. Claims made by Lynn (1991b; Lynn and Vanhanen 2002) and
Rushton (2000) about the intelligence of different groups of South Africans also ignore the
very significant debates about mental testing in South Africa during the twentieth century
(Dubow 1995: 197–245), and the fact that much of this debate involved the IQ test scores
of (usually Afrikaans-speaking) ‘poor whites’. A sense of the quality of reporting of these
tests comes from Lynn’s (1991b; see also Lynn and Vanhanen 2002: 219) description of
tests administered by Owen (1989) on different ‘racial’ groups in South Africa as ‘[t]he best
single study of the Negroid intelligence’. Owen himself (1989: 60, 62–8) indicated
significant problems with these tests, many involving language difficulties experienced by
the African test-takers, and did not assign IQ scores for the results. Similar kinds of
problems, where authors’ cautions about test circumstances are ignored by Lynn and
Vanhanen, exist in a number of the other African cases. At the same time, it is unlikely
that straightforward bias could explain all of the test score results in question.
A third possibility also presents itself, one derived from another realm of debate about
IQ test results. The African case is not the only circumstance in which unaccounted-for
differences exist in IQ test scores across cultural boundaries. James Flynn (1984, 1987,
1998, 1999; see also Dickens and Flynn 2001) has documented a steady rise in IQ test
scores from the late nineteenth century onward in countries where longitudinal data exist,
an increase now widely known as the Flynn Effect. (Longitudinal data are unfortunately
available almost exclusively from Western countries.) Depending upon the test, this rise
varies from less than 10 points to as much as 20 points per generation, with the greatest
increase in ‘culture-reduced’ tests like the Raven’s Progressive Matrices. This effect is both
reliable and predictable: it also indicates significant gains in intelligence over a period too
short for any evolutionary effects to come in to play. Explanations for the Flynn Effect
vary and none appears completely satisfactory, but there seems little doubt that it is due to
some combination of environmental and cultural factors at play in Western societies,
factors that remain significant even on tests where cultural influences are supposed to be
largely excluded.
Taken at face value, the Flynn Effect implies that the average North American adult
living about a century ago would have had an IQ of approximately 75 in modern terms, a
value closely comparable to that derived for twentieth-century African populations by
Rushton, Lynn and their associates. Mid-nineteenth-century North Americans would
have been even more deficient mentally. This is a nonsensical result, and is widely accepted
as such; no North American believes that our great-grandparents were mentally deficient.
As two enthusiasts for racial comparison in intelligence testing, Richard Herrnstein and
Charles Murray, said a decade ago in The Bell Curve:
"Whether one looks at the worlds of science, literature, politics, or the arts, one does not
get the impression that the top of the IQ distribution is filled with more subtle,
insightful or powerful intellects than it was in our grandparents’ day. . . . No one is
suggesting, for example, that the IQ of the average American in 1776 was 30 or that it
will be 150 a century from now.+ -(Herrnstein and Murray 1994: 308–9)
In this case, Herrnstein and Murray are absolutely right. Moreover, the evaluative criteria
used by Herrnstein and Murray to judge this claim involve the examination of cultural
accomplishment, as does the present paper. This is the only way to test such claims for
societies before the invention of intelligence tests. The fact that similar nonsensical results
concerning Africans are widely promulgated in the psychometric literature may indicate
simple ignorance of African societies or a more pernicious readiness to place Africans
below other human populations in a ranking of inherent human worth.
The Flynn Effect is a measure of IQ test performance across a substantial cultural
divide, with biology held more or less constant. This divide exists in time: it looks from the
early twenty-first century back toward the late nineteenth century. The IQ testing that has
taken place in Africa since the 1920s has taken place across a comparable cultural divide,
one from economically and politically dominant Western societies and test designers to
African societies and individuals in almost all cases at significant economic, social, cultural
and/or political disadvantage. In this case, the cultural divide exists both in time (because
the tests have been administered since the 1920s) and in space. The parallels between time
and race have been remarked upon by Flynn (1999: 14–15; see also Dickens and Flynn
2001). It may well be the case that depressed African IQ test score results are best
explained by a combination of obvious test bias and the subtle and additive environmental
differences that produce variation in even ‘culture-reduced’ tests like the Raven’s
Progressive Matrices, in a geographic analogy to the Flynn Effect. The archaeological
data would seem decisively to contradict the claim that these differences stem from
reduced cognitive potentials in African populations.
Conclusions
This paper presents the juxtaposition of two data sets, one archaeological and one
behavioural and psychometric. It is now a commonplace in the comparative psychometric
literature to claim that low IQ test scores among African populations indicate severely
diminished average intelligence among those populations. Rushton (2000) places these
claims in a behavioural and evolutionary context, one paralleled by similar explanations
applied to poor and relatively powerless populations in other parts of the world and
supplemented by data of other kinds. Rushton’s model posits quite major behavioural
differences among the different continental populations, and especially between tropical
African populations and the inhabitants of temperate and Arctic Eurasia. The magnitude
of these differences is such that they should be detectable archaeologically, and indeed
Rushton presents archaeological evidence that he believes bolsters his case. His archaeological
interpretations are for the most part obsolete and/or erroneous.
However, Rushton is probably correct in claiming that such a magnitude of racial
differences should be demonstrable archaeologically. Archaeological data provide an
independent test of his hypothesis, one not subject to the obscuring effects associated with
modern mental testing and interpretation. Examination of archaeological data on the culture
history of African populations, and comparison of those data with data from other parts of
the world, yields no evidence for the behavioural and cognitive disparities claimed by
Rushton. African cultural history is entirely comparable with that of other regions of the
world, not in terms of lockstep evolutionary schema but rather in the evident sophistication
with which African populations have met the challenges of their physical and social
environments through time. To interpret the conflict between these two data sets, it may be
useful to examine possible confounding factors in the behavioural and psychometric data.
The behavioural data are quite variable and often of poor quality, but it is striking to note
that the field of intelligence testing is grappling with a phenomenon analogous to continental
differences in IQ test scores – the Flynn Effect. In both cases, testing across cultural
boundaries yields results that systematically disadvantage populations culturally removed
from our own, results that on their face defy logic. It is now up to intelligence researchers to
identify the confounding effects in their tests, and let archaeologists and other researchers get
back to looking at the Africa that actually exists today, and existed in the past."
knol.google.com/k/mainstream-academic-research/human-biodiversity-racial-evolutionary/3q8x30897t2cs/28#view