By Volkmar Weiss, German Centraly Office for Genealogy, Leipzig
Vol. 35, Mankind Quarterly, 06-01-1995, pp 373.
The Historical Perspective
From the point of view of a European reader, the much discussed book by Herrnstein and Murray consists of two halves, which could have been published independently in two volumes. The first half (part I. -III.) would be a volume of supranational meaning and importance, the second one (part IV.) would be a volume reflecting special social problems of the US from a certain political angle and therefore would be highly controversial. It is a pity, that from the beginning this controversial discussion of this fourth part will overshadow the content of the first three parts, which in the long run will prove to be as timeless and important as Darwin's Origin of the Species. Never before has any author portrayed the relationships between IQ, education, occupation, economy, poverty, unemployment, welfare, crime, and social structure with so much convincing statistical data in such a well written way as done by Herrnstein and Murray. We will restrict our comments to these strictly scholarly parts I., II., and 111.
The two authors observe that (p. 510): "As the century progressed, the historical mix of intellectual abilities at all levels of American society thinned as intelligence rose to the top. The upper end of the cognitive ability distribution has been increasingly channeled into higher education . . . , thence into high-IQ occupations. The upshot is that the scattered brightest of the early twentieth century have congregated, forming a new class." However, already in 1949 (Swedish original edition even in 1944) the sociologist Geiger came to the opposite conclusion that the intelligentsia will never be a "class" characterized by a common purpose and goal and will only remain a stratum. By differing in their relation to power and wealth and striving to increase their influence in such a way, a substantial part of the cognitive elite tends to adhere to a more or less (pseudo)egalitarian ideology and is thriving under such conditions. The development of many countries seems guaranteed at best when the largest number of people honestly believe that they have already reached a maximum of equality, where in reality a maximum of inequality exists. As long as this belief is not disturbed, such countries enjoy social peace and, from this point of view, books as The Bell Curve are only a nuisance, which should be suppressed, even from the point of view of a large portion of the so-called "cognitive elite" (in German already called lntelligenz, in Russian intelligentsia). Several times in history a surplus of cognitive elite has been slaughtered or expelled during the next social revolution, reestablishing in such a way the balanced polymorphism of IQ distribution and underlying gene frequencies. History can even be understood as the struggle between cognitive sub-elites with and without egalitarian ideology.
The channeling of all high-IQ individuals into higher education is not restricted to the United States alone. Parallel shifts can be observed in all industrialized countries, even in the former communist states. For example, in (West) Germany higher education was extended from 3% in 1900 up to about 30% in 1990. In former East Germany this inflation of educational degrees was stopped by governmental intervention by about 12%. Because an IQ of 125 or more is not only a different quantity, but maybe a different quality, i.e. a different genotype (Weiss 1992a, 1994), an extension of higher education beyond 5 or even 10% could even cloud the emergence of the new elite. Only in the United States there is no clouding because we have the unique situation of IQ values available for all the educated population and a ranking of universities (unknown, for example, in Germany). Therefore the book by Herrnstein and Murray could only be written in the States, nowhere else could anybody have obtained statistical data of such quality.
According to both authors (p. 34): "At mid-century America abruptly becomes more efficient in getting the top students to college." In 1964 in East Germany 32% of the highly gifted (IQ above 125) had fathers with higher education, in 1970 already 66% (Weiss 1972). The underlying trend (linear in the sixties) caused the communists to prohibit strongly the publication of all such secret data on the growing educational stratification even within East bloc countries (Weiss 1991). Per definition, the communist intelligentsia in their majority had to be the offspring of workers and peasants. But this has never been true and was becoming less and less true.
Herrnstein and Murray are aware that (p. 26):
Human society has always had some measure of cognitive stratification . . . . This differentiation by cognitive ability did not coalesce into cognitive classes in premodern societies for various reasons . . . . Parents could not [not] pick the brightest of their progeny to inherit the title and the land . . . . Stratification by cognitive ability has been weak and inconsistent until this century because the number of very bright people was so much greater than the specialized jobs for which intelligence is indispensable. A true cognitive elite requires a technological society . . . . A large majority of the smart people in . . . Elizabethan England, and Teddy Roosevelt's America were engaged in ordinary pursuits . . . , living with everyone else . . . . Social and economic stratification was extreme, but cognitive stratification was minor. So it has been from the beginning of history into this century.
And further (p.510):
Among the smart women, a few had professional careers of their own, but most of them kept house, reared children, and were often the organizing forces of their religious and social communities. People from the top of the cognitive ability distribution lived next door to people who were not so smart . . . They socialized with, . . . and married people less bright than themselves as a matter of course.
They here comment that: "This was not an egalitarian utopia that we are trying to recall."
Not a utopia, but a typical American perspective, where nearly every immigrant had to start from the very bottom. From the point of European social history, presumably a crude underestimation of the role the IQ played for stratification already in traditional societies. Maybe, the engineer with the diploma of a university of today has only the same rank as a technician in 1900 and an able smith in 1750.
Saxony is the only large European territory where we can make reliable estimates of the absolute increase of the different social strata in modern time (Weiss 1993). Because the quality of parish registers in Saxony was in some places high from the beginning in 1548, the quality of genealogical research is correspondingly high. In the 18th century parish books, data on status and occupation of a male person are always given, even in the remotest village. In the 16th and 17th centuries our analysis was restricted to parishes were the genealogists have extended the data basis by means of family reconstitution with the use of tax rolls and records of tenure of land. From 500 ancestry and pedigree files we (Weiss 1993) drew five random samples of couples (in most cases married one), each sample comprising about 2200 couples for the following years of marriage: 1548-1649, 16501699, 1700-1749, 1750-1814, 1815-1870. The drawn quotas are representative with respect to main social strata and town/countryside distribution.
The upper stratum of the past, characterized by wealth, education and power, was in many respects a cognitive elite, too. Even more interesting is the occupational group of the "clerks," i.e. relatively poor men with jobs and occupations, where a lot of "cognitive competence" was necessary, such as schoolmaster, clerk, scribe, precentor, forester, or administrator. In Saxony (where a third of all inhabitants lived in towns) this nascent cognitive elite grew from 3% of the total urban population in 1615 to 12% in 1870, with a first leap from about 5% to 10% in the second half of the 18th century; in the rural population from about 2% in 1595 to 5% in 1870. From the beginning, this occupational group was unique in their social mobility. In towns only about a quarter of their fathers belonged to the same occupational group, another quarter were sons of small urban craftsmen or tradesmen, and the remaining half of the sons came from villages, especially in the 19th century, where their fathers were schoolmasters, parsons, peasants or rural craftsmen. As children of the countryside, these smart men entered the town with nothing else than their above average IQ; as children of the town, they inherited no fortune. Therefore, on the marriage market they had nothing to offer and many had to marry wives of the same relatively humble origin or with only a small dowry, i.e. mostly the daughters of craftsmen. But there must have been a lot of assortative marriage with respect to IQ. Only such a marriage strategy can explain the unique upward mobility of this occupational group in the second generation, where about a third of their sons (and daughters by marriage) became members of the wealthy upper stratum (Besitzund Bildungsburgertum) of the cities, either by their own merit and earning or by marriage or in many cases both combined. Because this group of poor "intellectuals" was always growing, a third or more of the sons could remain in such or similar jobs as their fathers and had a second chance of upward mobility in each following generation. The uniqueness of this occupational group is underlined by mentioning, for example, that more than 80% of the urban craftsmen descended from urban craftsmen (as well as more than 80% of peasants from peasants).
Sons and daughters of the upper stratum of the towns, no more than 10% of the total urban population, were never marrying sons and daughters of the low stratum (with the exemption of this mentioned occupational group). Intergenerational social mobility from the bottom to the top or vice versa, mediated by the medium stratum, needed at least two generations. In the countryside even the peasantry was highly stratified into full peasants, smallholders and cotters. The marriage patterns between these three strata were rather consistent and nobody liked to marry downward and to endanger the economical basis of future generations. We tend to agree with Herrnstein and Murray that these patterns of behavior primarily had nothing to do with IQ. But Herzog (1984) showed that in Lampertswalde, again a village in Saxony, during 1700-1799, 4.6 children of full peasants (arithmetical mean per marriage) inherited a total of 612 florins, which translates to 133 fl per child, 5.4 children of millers inherited even 2165 fl., i.e., 401 fl. per child, while 3.3 children of cotters inherited 77 fl. (only 23 fl. per child), which was of no small importance for their chances on the marriage market. Herrnstein and Murray seem to underestimate the possibility of upward and downward mobility in traditional societies by personal achievement. Even in the 16th, 17th and 18th century a small percentage of men were changing their job several times during their lifetime (see Weiss 1993). Even peasants could buy and sell and in such a way, at the end of the Thirty Years War, the peasant Abraham Scheibner (Zschocken/Saxony) was able to place all his seven children into their own property. Surely, he had luck, but, I guess, a high IQ, too.
For the emergence of a new cognitive elite the pattern of assortative mating with respect to IQ plays a decisive role. Assortative mating for IQ means that mated pairs are more similar for IQ than would be expected if they were chosen at random from the population. Herrnstein and Murray (p. 110) state:
When the propensity to mate by cognitive ability is combined with the educational and occupational stratification we have described, the impact on the next generation will be larger than on the previous one, even if the underlying propensity to mate by cognitive ability remains the same . . . . We have been assuming that the propensity to mate by IQ has remained the same. In reality, it has almost certainly increased and will continue to increase.
In spite of this, we should be aware of a counter-hypothesis, stating that with respect to assortative marriage nothing essentially has changed during the last centuries. IQ, wealth and power could always have been correlated in such a way, that marriage patterns mediated on this basis were always resulting indirectly in a correlation of about .50 with respect to IQ of spouses. In other words, to prove the emergence of a new elite needs to prove that the correlation of spouses with respect to IQ is increasing, historically, at present and in the future. Historically this proof will be impossible, at present difficult, in the future an everlasting challenge for research. Assortative marriage has two consequences (Crow and Felsenstein 1968): (1) an increase in the average homozygosity, and (2) an increase in the total population variance. For example, assortative marriage is quite high for deafness, and it might be thought that this is a major factor in increasing the incidence. But it has been estimated that there are at least 35 genes, any one of which can cause deafness when homozygous, and with a small average frequency. Thus, even with strict assortative marriage the incidence of deafness would not be increased by more than 2% or 3%. However, for very high IQ the situation could be quite different and would depend upon whether a major gene locus is very important (Weiss 1992a, 1994) or not. In the case of a major gene locus underlying general intelligence, an increase of assortative marriage with respect to IQ would strengthen the disruptive tendencies within industrialized societies. Both extremes, high IQ and low IQ, would be on the increase, the percentage of the cognitive elite more by an increase caused by assortative marriage than by a growing number of children of their own.
Furthermore, Herrnstein and Murray confirm in a convincing way that the recent distribution of IQ is the result of natural selection in the Darwinian sense and that the future distribution of IQ will be the result of natural selection, too. The rise of the new elite will be impossible without a growing number of jobs for this elite and without the will and the possibility to feed a growing number of children of their own. It is the merit of Herrnstein and Murray to have shown the inter-dependencies between welfare and the proliferation of the dull. In traditional European societies this proliferation was suppressed by a sometimes cruel policy, which nobody wishes to be revived, and Gypsies were even persecuted. Times did change. For example, in 1945 there were about 100,000 Gypsies living in Czechoslovakia, and under communist rule this number grew to about 600,000. In 1980, 1,000 Gypsy women averaged 5.984 children (Kalibova and Pavlik 1988), which contrasted dramatically with the small number of children born to Czech women, now below replacement level. In 1980, 18.7% of all Gypsy children (2.7% of Czech children) were in schools for mentally retarded children, 15.1% were borderline cases of mental retardation (1.0% for Czechs), and only 0.3% were in schools for higher education (7.1% for Czechs).
From these educational statistics we can infer an average IQ of about 85 for Gypsies in former Czechoslovakia. Statistical data from Hungary (Szabo 1991) and Romania are very similar. In the Czech Republic unemployment and crime rate among Gypsies are sixfold to eightfold higher than among the Czech. In order to remedy this situation, communist (sic, communist) Czechoslovakia implemented a program of voluntary sterilization, paying Gypsy women a premium (Fienbork, Miho, and Muller 1992, p. 103). After 1989 Gypsies were among the first to use the new freedom of movement. On July 14, 1993 the daily Leipziger Volkszeitung (p.16) had to write: In the northeastern quarter of Leipzig with its 90,000 inhabitants more than a third of all police actions is dedicated to 350 Gypsy immigrants." Because socially upward mobile Gypsies are assimilated to a certain extent and no longer designate themselves as Gypsies, and on the contrary during each generation a few downward mobile individuals of other populations become integrated into Gypsy tribes since centuries, these statistics relate to some quantitatively unclear mixture of social and racial distinctions. Therefore J.J. Kemeny (cited by Szabo 1991, p.89) has to conclude: "The life-style of the Gypsies is the subculture of the lowest stratum of Hungarian society."
The Bell Curve and Short Memory Capacity
Plomin et al (1994) state: "General cognitive ability (intelligence, often indexed by IQ scores) is one of the most highly heritable behavioral dimensions . . . . General cognitive ability . . . is a quantitative trait with a roughly normal distribution." Herrnstein and Murray's data are based on the Armed Forces Qualification Test (ETCHED). "It consisted the summed raw scores of word knowledge, paragraph comprehension, arithmetic, and mathematics knowledge subtests" (P. 570). This sum should be heavily loaded with crystallized intelligence. "The distribution of AFQT is skewed so that the high scores tend to be more closely bunched than the low scores. To put it roughly, the most intelligent people who take the test have less of an opportunity to get a high score than the least intelligent people have to get a low score . . . . We therefore computed standardized scores corrected for skew" (p. 572) and the ceiling effect, respectively.
Because even a minority of psychometricians (Frank 1985; Lehrl et. al. 1990; Weiss 1992a) is stubbornly claiming that IQ and especially raw scores of subtests of fluid intelligence are not normally distributed, Herrnstein and Murray feel needed to add (p.585):
Empirically, tests . . . , administered to a representative sample of those for whom the test is intended, will yield scores that are spread out in a fashion resembling a normal distribution, or a bell curve. In this sense, tests of mental ability are not designed to produce normally distributed scores; that's what happens . . . . It is also true, however, that tests are usually scored and standardized under the assumption that intelligence is normally distributed, and this has led to allegations that psychometricians have bamboozled people into accepting that intelligence is normally distributed.
An IQ of 140 compared with an IQ of 70 on the surface suggests a double amount or the half of cognitive ability, respectively. However, a look of the raw scores, i.e., of the non-normalized scores, of IQ subtest of fluid intelligence shows that an IQ of 140 means a fourfold amount of cognitive ability compared with an IQ of 70 (see Table 1 and Fig. 1). This relationship holds under the condition that all subtests are elementary cognitive tasks and the superior speed of the most intelligent is not clouded by a ceiling effect.
Since 1959, Frank (see Lehrl, Gallwitz, Blaha and Fischer 1991, for the last updating of Frank's theory and for representative empirical results) is claiming that general cognitive ability is limited by the channel capacity of short-term memory. Frank and his school (Lehrl and Fischer 1990; Weiss 1992b) are arguing that the capacity C of short-term memory (measured in bits of information) is the product of the processing speed S of information flow (in bits/s) and the duration time D (in s) of information in short term memory. (By Lehrl et al. testing of processing speed S was operationalized by reading rates, duration of information D by memory span.)
Hence:
C (bits) = S (bits/s) x D (s)
According to major gene theory of intelligence (Frank 1985), the mean of the IQ-genotype M1M1 is 140 bits, and of the genotype M2M2 70 bits, that means the contribution of a single M1 allele to short-term memory capacity C is about 70 bits, of a M2 allele about 35 bits. For a heterozygote M1M2 hence 70 bits + 35 bits = 105 bits.
Does this difference between the bell curve and the distribution of short term-memory capacity have any practical meaning? For this purpose we have reproduced two figures, from which we can imagine the importance of the fourfold difference in mental speed and memory capacity between high and low IQ individuals. By this, Hernstein and Murray are not contradicted in any way, they are essentially right. But their conclusions as to the importance of IQ ineveryday life and work are an underestimate. Between IQ 70 (-2 standard deviations from the mean) and IQ 140 (+2.66 standard deviations from the mean) is not a twofold, but a fourfold difference in general cognitive ability, with all its socioeconomic consequences.
The Cognitive Elite Breeds True
Many authors assert that regression to the mean levels IQ differences. Reading on p. 26 by Herrnstein and Murray: "Over the course of a few generations, the average intelligence in an aristocratic family fell toward the population average. hastened by marriages that matched bride and groom by lineage, not ability." I was afraid that again the usual misunderstanding of regression was shared and repeated by both authors. However, in the case of the old nobility their statement is true as in all cases where socioeconomic status and IQ are understood as properties of the males only and where wives are chosen according to their beauty, circumference of their breasts, dowry and so on, regression to the mean is an unavoidable consequence.
But now very many smart women have professional careers of their own and membership or non-membership of the cognitive elite is a property not only shared by both husband and wife but also open to scaling of educational degrees and occupations with respect to IQ for both sexes. What in such cases where both married partners are members of the cognitive elite?
For the first time (beside an analog conclusion by Seligman 1992) I am happy to read the only correct conclusion (Herrnstein and Murray, p. 357): "Regression to the mean is a statistical conclusion, not a biological one . . . 'Mean' referred to in 'regression to the mean' is the population's own mean." If we consider the cognitive elite to be a subpopulation of their own, then the regression will be only toward -and even upward - the mean of this elite and never toward an IQ of 100 (Weiss 1982).
In 1980 (p. 114) Jensen stated:
There are several critical thresholds within the total range of IQ, each having important educational and occupational consequences for the individual . . . . The socially and personally most important threshold regions on the IQ scale are those that differentiatewith high probability between persons who because of their level of general mental ability . . . can or cannot succeed in the academic or college graduate preparatory curriculum through high school (about 105), and can or cannot graduate from an accredited four-year college with grades that would qualify for admission to a professional or graduate school (and IQ 115).
In 1970, in former East Germany, a family study, starting with 1329 mathematically gifted with an IQ higher than 130, came not only to very similar thresholds but also from a large body of empirical data (for details see Weiss 1992a) to the conclusion that three overlapping genotypes of IQ (type M2M2, 68% of the total population of Germany, median IQ 94/type M1 M2, 27%, median IQ 112 / highly gifted type M1 M1, 5%, median IQ 130, mean IQ 139) are segregating in a Mendelian manner. Twenty three years later a follow-up (Weiss 1994) of the gifted proved Mendelian segregation among their children (see Table 2) and among their nephews and nieces (see Table 3).
Of course, in social reality, where cognitive ability is embedded into personality and chance plays its role in each biography, a 100% fit between a Mendelian theory of IQ and empirical results cannot be expected. (For example, of the 16 children (the 6.6% in the first row of Table 2) not fitting the theory (they did not pass Abitur), one was physically and mentally disabled, two were from broken marriages, two were midwives, two nurses, five males were already in highly qualified technical jobs (but without Abitur), three stated "students" in the questionnaires without further details, i.e. Abitur not impossible.)
For all gifted individuals M1M1, the chance of their offspring belonging to the cognitive elite is a direct function of the genotype of their respective spouses. Of course, achievement needs education, but individuals are born with fixed, unchangeable levels of cognitive potential. Marriages in which both partners are M1M1 have only M1M1 children (all with Abitur, cutoff IQ with a mean of 112, i.e. very similar to the mean of all US college graduates, of 1990; compare Figure 1 and look at the bell curve from Herrnstein and Murray above). In marriages where the partner of the gifted subject is M1M2 half the children would be M1M2 and consequently a quarter of all children would not pass the Abitur (see Table 2, second row). Because n M1M2-M1M2 marriages with 100 children should segregate theoretically into 25 M1M1 (all with Abitur), 50 M1M2 and 25 M2M2 children (all without Abitur), and half of the M1M2 offspring should attend a high school, the expected value in this marriage group (typically middle class, both partners M1 M2) for obtaining the Abitur should be 50"/0 (see Table 3, third row).
Where polygenic theory predicts a normal distribution of ability among siblings of the same family (Jensen 1980, p. 80), major gene theory (Weiss 1992a) predicts that the IQ distribution of the offspring of homozygotes and their M1 M2-spouses should be quite markedly skewed (compare Table 2, second row). The skewedness should be in the opposite direction for M1M1-M1M2 offspring compared to M1M2-M1M2 offspring (compare Table 3, second and fourth row). M1M2-M1M2 offspring, who are segregating according to Mendelian rules (see above), should have a much greater variance than the offspring of marriages where both partners are homozygotes (either M1M1 or M2M2) and are "breeding true" (compare Table 3, first, third and fifth row), that means only with regression to the mean of the specific type. The major gene effects should be even more pronounced if we compare not normalized raw scores instead of IQ values (we draw attention to Weiss 1992a, figs, 1 and 4). In 1993 in the follow-up sample (Weiss 1994), 97% (n = 357) of the highly gifted males were in professions typically associated with an IQ above 123, compared with 55% (n =77) of the sons, 49% (n =220) of the brothers, 40% (n =346) of the fathers, 18% (n =570) of the male cousins, 22% (n =76) of the nephews, 14% (n =615) of the uncles, 11% (n =2250) of the male cousins of the parents (with no evidence for any excess on the paternal or maternal side and hence no evidence for X-chromosome linked inheritance), 9% (n = 681) of the grandfathers, 5% (n = 1996) of the uncles of parents and 4% (n = 1290) of the great-grandfathers.
In 1903, the famous geneticist Johannsen had already confirmed experimentally that in pure lines there is no regression toward the mean. In the sense of genetics the cognitive elite (in contrast to the old European nobility) is becoming more and more a pure line which breeds true, with all its socioeconomic consequences. This is the message conveyed by Herrnstein and Murray that should outlast the disputes of the day.
Table 1. Non-normalized raw scores in the subtests 3 + 4 + 7 + 8 + 9 of the general cognitive ability test LPS for some selected occupations.
Mathematically sifted tested by Weiss 105
Production engineers 168
Lawyers 162
Psychologists 155
Draughtsmen 146
Clerks 132
Fitters 128
Electricians 116
Hairdressers 105
Bakers 92
Unskilled 60
Data from Horn (1962).
Table 2. Percentage obtaining the Abitur (German high school leaving examination) among the children of gifted MIM1 (from Weiss, 1994).
* These percentages would be expected under the assumption of Mendelian segregation and a cutoff of IQ of 112. Legend for Chart:
A - Marriage combination of proband and spouse B - Children with Abitur; Percentage; obtained C - Children with Abitur; Percentage; expected D - Children without Abitur; Percentage; obtained E - Children without Abitur; Percentage expected
A B C D E
both spouses with IQ 124 and higher- M1M1 x M1M1 93.4 100 6.6 0 n =242 gifted with spouse with IQ 75.5 75 25.5 25 n = 1 84 below 124- M1M1 x M1M2
Table 3. Percentage obtaining the Abitur (German high school leaving examination) among the nephews and nieces of highly gifted M1M1 (from Weiss, 1994).
* These percentages would be expected under the assumption of Mendelian segregation and a cutoff of IQ of 112. Legend for Chart:
A - Sibs of probands and respective spouse B - Children with Abitur; Percentage; obtained C - Children with Abitur; Percentage; expected D - Children without Abitur; Percentage; obtained E - Children without Abitur; Percentage expected
A B C D E
both spouses with IQ 124
and higher- M1M1 x M1M1 91.4 100 8.6 0 n = 70
one spouse with IQ 124
and higher; the other
with IQ below
124- M1M1 x M1M2 71.5 75 28.5 25 n = 130
both spouses with IQ
between 104 and
124 - M1M2 x M1M2 52.3 50 47.7 50 n = 107
one spouse with IQ
below 105-- M1M2 x M2M2 6.9 25 93.1 75 n = 29
both spouses with
IQ below 105 - M2M2 x M2M2 0 0 100 100 n = 12
Americans with and without a college degree as of 1990
Figure 1. Above: the "perfect" bell curve (from Herrnstein and Murray, p. 46), and Below: the bell-shaped overlapping distributions of the three genotypes underlying general cognitive ability. In industrialized societies the percentage of all college graduates fits closely with the sum of the percentage of the genotype M1M2 plus the highly gifted M1M1.
REFERENCES Crow, J. F. and J. Feisenstein 1968 The effect of assortative mating on the genetic constitution of a population. Eugenics Quarterly 15: 85-97.
Fienbork, G., Mihok, B. and S, Muller (eds.) 1992 Die Roma - Hoffen auf ein Leben ohne Angst. Roma aus Osteuropa berichten. Hamburg: Rowohlt.
Frank, H. G. 1985 Is intelligence measurable and is it inherited? Folia Humanistica 23: 671-691.
Geiger, T. 1949 Aufgaben und Stellung der Intelligenz in der Gesellschaft. Stuttgart. Enke.
Hermstein, R. J. and C. Murray 1994 The Bell Curve: Intelligence and Class Structure in American Life. New York: Free Press.
Herzog, J. 1984 Die Entwicklung der Grundherrschaft Lampppertswalde. Phil. Diss., Leipzig: KarI-Marx-Universitat.
Morn, W. 1962 Leistungsprufsystem (LPS). Gottingen: Hogrefe.
Jensen, A. R. 1980 Bias in Mental Testing. London: Methuen.
Johannsen, W. 1903 Erblichkeit in Populationen und in reinen Linien. Jena: G. Fischer.
Kalibova, K. and Paylik, Z. 1988 Cechy demograficzne ludnosci cyganskiej w Czechoslowacji. Studia Demograficzne De Nr. 3/93: 37-55.
Lehrl. S. and B. Fischer 1990 A basic information psychological parameter (BiP) for the reconstruction of concepts of intelligence. European Journal of Personality 4:259-286.
Lehrl, S., Galiwih, A., Blaha, L. and B. Fischer 1991 Geistige Leistungsfahigkeit. Theorie und Messung der biologischen Intelligenz mir dem Kurztest KAI. Ebersberg: Vless.
Plomin, R., McClearn G.E., Smith, D.L, Vignetti,S., Chorney, M.J. Chorney, K, Venditti, C. P, Kasarda, S., Thompson, L. A., Detterman, D.K, Daniels, J., Owen, M. and P. McGuffin. 1994 DNA markers associated with high versus low IQ: The IQ quantitative trait loci (QTL) project, Behavior Genetics 24: 107-118.
Seligman, D. 1992 A Question of Intelligence. New York: Birch Lane Press.
Szabo, G. 1991 Die Roma in Ungarn. Ein Beitrag zur Sozialgeschichte einer Minderheit in Ost-und Mitteleuropa. Frankfurt/M: P. Lang.
Weiss, Volkmar 1972 Ergebnisse zur Genetik der mathematisch-technischen Begabung. Diss. Berlin (East): Humboldt Universitat.
1982 Psychogenetik: Humangenetik in Psychologie und Psychiatrie. Jena: G. Fischer.
1991 It could be Neo-Lysenkoism, if there was ever a break in continuity. The Mankind Quarterly 31:231 253.
1992a Major genes of general intelligence. Personality and Individual Differences 13: 1115-1134.
1992b The relationship between short-term memory capacity and EEG spectral density. Biological Cybernetics 68: 165-172.
1993 Bevolkerung und soziale Mobilitat: Sachsen 1550-1880. Berlin: Akademie-Verlag.
1994 Mathematical giftedness and family relationship. European Journal for High Ability 5: 58-67.
