The introduction to this book describes it as a ‘triumph in modern psychology; the successful development and application of a solid measure of general intelligence.’ It contains 23 articles, divided by subject into six parts, produced by 31 authors 18 of whom are from the USA, 5 from the UK, 3 from Canada and 1 each from Belgium and Denmark. A tribute to Arthur Jensen, a major expert in this field, occupies pp.xiii-xxvi of the introduction and is resumed on a more personal level in chapters 22 and 23.

A large part of the text of 571 pages is given to a description of ‘g’; its nature, biology and its correlation with other specific abilities. The most important questions to be answered by the study are whether g testing provides a satisfactorily accurate measure of an individual’s capacity to 1. grow in ability and to benefit from education and 2. measure how competent the subject will be in employment compared with individuals ranked in intelligence and matched in age. There are differences of opinion about the nature and value of ‘g’. Is it the best measure of all abilities? Are such factors in intelligence tests as ‘picture recognition’ and ‘number calculation’ more accurately measured by g than by specific tests adapted to measure each of the factors to which for instance ‘picture recognition’ and ‘number calculation’ belong? Carroll (chap. 1, p8) investigates this query by applying 29 tests to a large data-set and from multiple comparisons extracting and comparing the correlations which emerge he demonstrates that there is a factor labelled Fluid Reasoning (Gf), p14 that is significantly separate from g thus dislodging the view that Gf is identical with g.

In Part 2, ‘Biology of g’, papers on the anatomy and physiology of intelligence are presented using recent methods of producing measurable images of the brain in whole or selected areas particularly in inaccessible locations using specific apparatus for the purpose and also measuring quantities and effects of biochemical substances involved in brain functions. Positron emission topography exemplifies this with a clear report of work using it.

The limits of using straight-forward tape and calliper measurements of skull or brain for volume can be greatly extended by computer topography which produces photographic transverse sections of the brain itself giving enough evidence to allow correlating its size with a measure of its ‘g’. It is naturally assumed that the vigorous exercise of intelligence requires more power from the brain and this is indeed what is found. Surprisingly however although blood increases to meet the pressure of work in the brain location involved the concentration of glucose there, also a source of power which blood carries, reduces.

‘Reaction time’ is determined by measuring the time interval between a subject’s perceiving a flash of light and responding as rapidly as possible by pressing an appropriate button. Reaction time shows how greatly the field of its use has extended since work on its improvement was first published by Jensen and Munro in 1979. Recently it has been divided to measure inspection time (IT) and reaction time (RT) which have produced promising results but not met queries and answered criticisms sufficiently to produce a universally accepted measure of any of the subjects proposed for it particularly a firm confirmation of a correlation between reaction time and intelligence.

Chapter 8 (p123) ‘The Geography of g’ compares the average g of individual nations worldwide. To do this a representative sample of the population of each nation has been chosen and its individuals’ intelligence tested to produce an average national IQ. This has already been done with results published by Cavalli-Sforza et al (1994), which Professor Lynn describes as ‘The most recent and thorough classification of the world’s peoples by geography and race’ enabling him to produce in the book under review a chapter entitled ‘The Geography of Intelligence’ by adding to the material of these authors data of his own (Lynn 1991) plus data from other authors which he adjusted to update and make comparable. One query arising from this presentation relates to the number of inexact terms by which the race of individuals for testing is classified as for instance nationality, race, geographical-racial group, whites. Cavalli-Sforza et al in the volume referred to above precede their Results Report with a diatribe deploring the use of the term race to select individuals of the human species who show some significant difference (e.g. short-sightedness) to differentiate them racially from other individuals belonging to the human race. Cavalli-Sforza et al give substantial reasons for their view and concern about it goes back to Charles Darwin and Francis Galton. In reporting his results Lynn attempts to avoid the confusion arising from identifying the subjects of his study in this way by naming the country or territory in which they live (Table 8.1-8) and in his Discussion (pp 137-141) using the contrasting colour of their skin etc. for the same purpose thus identifying multiple races among these members of the human race which is a single species (see Lynn, 1991). Anderson in chapter 2 concludes that magnetic resonance technology has given an answer to the question whether brain size correlates with intelligence. It does, r=0.35.

Rushton in chapter 9 reports Jensen’s investigation into Spearman’s hypothesis (1927) that a difference exists between the average g of Negro-Americans and that of White-Americans which is larger and has been found (15-18%) from when IQ testing began. Rushton also reports the results of Jensen’s massive total of data sets comprising 45000 Blacks and 245000 Whites in which the discrepancy in their average g’s persistently shows under several assessments and checks.

There is perhaps a doubt about the validity of the tests used in producing this result because the items constituting the tests are mainly derived from European traditional skills, customs and expectations and not those of the peoples of other nations, environments and communities who are tested. Consider an extreme example, for instance, two sample populations, one black one white, tested in the Sahara desert using a test derived from the traditions held and the skills necessarily employed by the desert’s human inhabitants viz. searching for water, tracking animals, using weapons (e.g. bow and arrow) to procure food, judging distances. Inhabitants of, for instance, European style cities might in a scientifically careful administration of the same test to both sample populations produce a less shiningly superior g than if it were produced by a test formed from and administered in their own environment, for instance altering the progressive matrices to a pattern furnished from a familiar source, e.g. stick figures.

In chapter 18 N. Brody examines the evidence for Jensen’s belief that the difference he has demonstrated between Blacks and Whites in average intelligence scores is in part attributable to genetic differences. Brody disagrees with this conclusion and gives reasons why by first dividing Jensen’s result into the five kinds of study he has used in support of his hypothesis against which Brody places conclusions from his own findings and from those of others. From these he makes precise comparisons between his and Jensen’s conclusions from which he disagrees with Jensen.

‘G’ is shown to be very useful in forecasting the results of training (potential air pilots for instance) and is widely employed for this purpose.

In chapter 14 ‘Education and g’ it is pointed out that in contrast to g-measuring tests, such as Raven’s Matrices, ‘IQ measures are bounded to an age-based framework … but growth of mental abilities with age breaks down as individuals reach adulthood.’ This leads to the suggestion that an assessment made at school entry age (4-5 years) is the least confounded and most reliable measure of g during school life. In retarded subjects g is markedly lower than the average of the general population but a considerably larger proportion of it is transferred to diverse abilities than is the case with that of the general population.

A sociologist and a criminologist take up the question ‘Is intelligence correlated with criminal behaviour?’ and begin with a preliminary caution that ‘it is understandable that people would avoid adding insult to injury by attributing low intelligence to those who have been snared by the criminal justice system, often for doing things that many of the rest of us did without getting caught.’ The authors undertake a world-wide collection of 68 reports from which they found that 60 studies showed a negative correlation and 8 found no significant correlation. Other relationships are derived however and the authors conclude ‘that progress in clarifying the relationship between intelligence and criminal behaviour has been made and the next step is to understand how and why these relationships are as the evidence suggests’. They offer three hypotheses on any or all of which a testing start could be made.

In chapter 15 jobs and life are separated into units of work such as typist, receptionist, auditor for which an average g factor can be calculated and used ‘as a common yardstick for comparing the mental demands of different tasks’. An illustrated example of this is given in a graph-table of 73 occupations (from the Wonderlic Personnel Test) shown in an ascending order of average test scores for each occupation against which one can read off for each occupation its rank position, the intelligence range for satisfactory accomplishment of the required occupation and other pertinent information.

The overall production and arrangement of this book are exemplary. It has separate subject and author indices; references in the text to other literature are conveniently listed at the end of each chapter and there is a complete bibliography of Jensen’s 438 papers and volumes 1955-2000 with 7 in the press. Chapter headings are informative as to subject as are the subtitles for sections. Graphs, tables and lists are clear and well annotated; the photographs of cognoscenti appropriate and pleasing. I think the book binding could have been more robust; in my copy the cover and leaves have begun to part company.

David C. Watt