Intelligence and Fertility
By J. A. FRASER ROBERTS, M.A., M.B., D.Sc., F.R.S.E.
Burden Mental Research Department, Stoke Park Colony, Bristol fertility of the Dull
If the population is grouped in various ways, for example, according to occupation, or geographical situation, it is usually found that the different groups are not Producing at the same rate. This phenomenon is called differential fertility. n this paper the grouping to be considered is based upon general intelligence, as estimated so successfully on mental test scales such as the Binet. It is well known at present in this country, as in many others, that the dull have, on average, more burden than do the bright. The evidence for this is irrefutable; but most of us, We have given the matter thought, will probably have reached the same conclusion simply by observing facts within our personal knowledge. A teacher at a junior elementary school, drawing children from a well-mixed population, is probably as well placed as anyone for appreciating the full extent of the present tendency. General intelligence depends to a considerable extent on heredity. This is reflected in a certain average degree of resemblance between, for example, parents and children, or between brothers and sisters. Of course there are exceptions; and it is the striking exceptions that tend to catch the eye; nevertheless, if considerable numbers are studied, it will be found that bright parents tend, on average, to have children of more than average brightness; similarly for the dull. It is certain that the level of general intelligence is also affected by environmental factors, especially, perhaps, those operating very early in life. But as long as heredity plays a part, and there is no doubt that that part is an important one, the fact that the dull are of higher fertility than the bright must inevitably be that a succeeding generation is inferior on the average in general intelligence to the generation that preceded it.
The Effect of Unequal Fertility
How great is the magnitude of this effect ? Is it comparatively trivial, or is it large ? A short while ago Cattell (1937) on the basis of a large scale experiment estimated the fall in average I.Q. to be about 3 points per generation. It should be carefully noted, however, that the mental test he was employing was one of his own and his I.Q.s show a greater range than do the Binet I.Q.s?so familiar to most people?which can well be accepted as a standard. Allowing for this, i.e. that a given difference in Cattell I.Q. corresponds to a smaller difference in Binet I.Q., the fall per generation in terms of the latter measurement would be between 1^ and 2 points per generation. The studies of my colleagues and myself, carried out on a group of Bath schoolchildren, also permit an estimate to be made. (Some of these observations have been published, 1938?the rest are now being analysed.) Our observations yield a figure in substantial agreement with that of Cattell, indicating a fall in average Binet I.Q. of a little more than 1^ points per generation. Is this a serious fall ? On any view other than one of extremely short range it is a very large and important fall indeed. Its effect, if it were to continue at the present rate for a considerable period, could only be described as catastrophic. At first sight, perhaps, a fall in the average figure for the whole population from 100 to 98 ? 5 might not seem great, but its importance may be better realized by considering what the effect would be on the numbers of highly gifted and very backward persons in the community. Taking an I.Q. of 130 or more to correspond to a very high intelligence, and one of 70 or less to correspond to pronounced dullness and frank mental deficiency, this rate of decline continued for only a single generation would mean that the highly gifted would be fewer by 20 per cent, while the very dull and mentally defective would be more numerous by more than 20 per cent.
Fertility of Mental Defectives
It is sometimes thought that it is sufficient to contrast “defectives ” or very dull persons with “normals”, conceding that the former show a higher average fertility. This is a very inadequate picture. Actually, the very brightest of all produce, on the average, the fewest children; those somewhat less bright somewhat more children ; those of average intelligence produce the average number of children; those who are somewhat below the average somewhat more than the average number; those who are distinctly backward distinctly more than the average number of children. The relation is essentially linear; the degree of infertility of those of above average intelligence is proportional to the amount by which they exceed that average; similarly for the higher fertility of those of below average intelligence. Considering the extremes, it probably would not be far from the truth to say that the dullest Persons in the population have an average of about three times as many children as do the brightest.
One important point should be noted in connection with mental deficiency. High grade mental defectives corresponding to the legal term ” feeble-minded ” are simply extremely dull persons, so dull that care and supervision are often required for their own sake or for that of others. Their reproductive rate is very high. Low-grade defectives, however, are different. They tend to crop up in family groups ?f any level of mentality. For the most part they are themselves incapable of ^production; furthermore, a group of their parents will not usually differ greatly ln average intelligence from the population as a whole; these parents will also display more or less average fertility. The relation between dullness and high fertility, therefore, holds until we reach this very lowest level of all, when it no longer obtains. Idiots and imbeciles, in fact, usually owe their mental state to causes, whether hereditary or non-hereditary, different from those which produce the dullness of the Very backward and the feeble-minded.
It will be seen, therefore, that differential fertility in regard to intelligence is not Slmply a problem related to mental deficiency. It is a tendency common to the whole population, and any analysis of its causes, or estimation of its effects, or any attempts actually to alter it, must take into account this circumstance.
Observations on Brothers and Sisters
So far we have considered differential fertility only in terms of parent and child: that the dull have more children than the bright is in fact the accurate way to describe Vyhat is happening. In practice, however, it is extremely difficult to make observations Slmultaneously on parents and children. In the first place, while a good mental test ls a highly efficient instrument of measurement in the case of a child, it is by no means s? efficient if applied to an adult. A test like the Binet is satisfactory because in any ?rdinary civilized community children are all receiving sufficient education of the ri?ht kind to enable them to answer the test questions correctly if they have attained the appropriate level of mental development. But as soon as they grow up differences 1,1 surroundings, occupation and the like make an increasing difference to the result. ^ school teacher will do better than a blacksmith even if as children their test results Were identical. True underlying, potential mental ability?general intelligence? ecomes very difficult to measure because of confusion arising from different expense. It sounds paradoxical, but is probably quite true, that the effect on the test result of a highly efficient education could hardly be detected, if at all, while that education was taking place, but it would produce an effect, and an increasing one, if the test were given later, in adult life.
A second, and equally important, difficulty is that of obtaining a sample of grown-up people which adequately represents the population. Actually, it is almost impossible. Schoolchildren, however, are available for such psychological observations and with some trouble a good sample can be selected; but any attempt to test the parents would yield an incomplete and certainly biased group. To measure differential fertility in regard to intelligence, therefore, by measuring parents and children, it would be necessary in practice to test the parents as children and then wait until their own children were available for testing: needless to say a troublesome and unpractical proposition.
The great majority of studies are carried out quite differently. Children only are measured and fertility is estimated by the number of sibs (brothers and sisters) that each child possesses. Such an investigation is comparatively easily carried out, and although it yields a measure of differential fertility which is somewhat indirectly calculated, the estimate obtained by means of such studies is perfectly valid. It is found that the dullest children have most sibs and that as intelligence increases so does the number of sibs decrease, the very brightest children having fewest sibs of all. The reason for this is, of course, that dull parents tend to have dull children and many children, therefore the dull child tends to have many brothers and sisters; conversely for bright parents and the above-average children. An estimate of the average fall in I.Q. per generation is quite easily made given one further piece of information: a measure of the extent to which sibs tend to resemble each other in regard to intelligence. This measure, too, can easily be made by testing a population of children which includes many sets of sibs. Given a large group of children of measured intelligence and with known numbers of sibs, we know, first, that if differential fertility were not operating the average I.Q. would be the same in two successive generations and would be equal to that of the group under observation. Next we calculate from the data how many sibs children of each intelligence class possess, and also what the average intelligence of each group of sibs will be. Thus we arrive at the average I.Q. after the lapse of a generation. This is how the figure of one and a half points fall (or a little more) per generation, mentioned earlier in this paper, was obtained.
A Misinterpretation
The fact, however, that investigations are in practice made not on parents and children, which would theoretically be the direct way, but on brothers and sisters instead, has led to some misconception. The observer fully conversant with the whole argument often states his results in terms of his actual observations. That is, instead of saying that dull parents tend to have more than the average number of children he often says simply that dull children tend to have more than the average number of brothers and sisters. This does not matter if he is addressing an audience familiar with the general facts; but it may mislead those who are not. It may actually be thought that what he means is that children are bright because they come from small families.
The possibility of misunderstanding was brought home to me eighteen months ago when I read a paper on intelligence and family size to the Eugenics Society. The audience was perfectly familiar with the general thesis and the paper was concerned not with the fact of differential fertility but with results obtained in a special investigation which threw some light on the nature of the phenomenon. The lecture was reported at some length in the daily press; and the accounts given were substantially accurate. Correspondence and further press comment soon made it clear, however, how much I had contributed to a misunderstanding the possibility of which had hardly occurred to me previously. It may be useful to consider a few of these comments, especially as some of them raise points of much practical importance.
The Daily Sketch (October 20th, 1938) printed a column under the heading ‘ Big Family Slander “, stating that doctors, psychologists, writers, people from the stage, business men, heads of working-class families?all were indignant at the report of the Burden Mental Research Trust which suggested that: ” Bright children were often only children; dull children came from large families.” Mr. J. B. Priestley, interviewed, said: ” I think a child from a large family is far better equipped for modern life than an only child. Early in life he has* had to deal with other people, to face his own problems. As regards intelligence he is quite equal to the only child. The only child may develop a stronger imagination than a child from a large family hut his intellect is not usually any higher.” Mr. Lupino Lane expressed much the same views.
My observations dealt, of course, with intelligence test results, and intelligence m this sense, technically, general intelligence, is not appreciably influenced by such factors as the number of sibs a child possesses. Mr. Priestley was, of course, quite right m stressing the fact that there are important differences between the background of the only child and that of the child from the large family ; but the effects such differences produce are not measured by a mental test of the Binet type. General intelligence may be regarded as basic, native, potential capacity. What use the individual makes, 0r is permitted to make, of his endowments is another matter. Two persons of identical IQ. may serve very differently their own interests and those of society. Nevertheless, it remains true that general intelligence in the technical sense is of basic, fundamental importance and is much the most important measurement that can be made on a child. Because of the operations of differential fertility it is inevitable that at present dull children should tend to come from families of larger average size than do bright ones; that bright children are more often only children than are dull ones. But they are, of course, not bright or dull because of the size of their family. Furthermore, given a certain level of general intelligence there is no doubt that the child from the large family does enjoy advantages which may well permit him to use more successfully the native endowments which neither he nor anyone else can change.
Mr. Sacheverell Sitwell, also interviewed, recalled Napoleon and John Sebastian Bach. ” Read your history,” he said, ” look about you to-day.” To take the Second point first, Mr. Sitwell could easily convince himself that the position to-day ls not what he imagines it to be. Even comparatively few enquiries amongst a well-mixed sample of people could leave no doubt about the magnitude of the present trend. What so often misleads us, unless we are careful, is the tendency to notice the striking and exceptional and not to see the far more numerous but not specially interesting cases. Some people would argue that opposites attract each other; for example, that tall men tend to marry short women, or that blondes tend to marry brunettes. Put to the test of actual observation and measurement it is found that this proposition is not true. On the average, like tends to marry like. The wives of a thousand tall men are taller on the average than the wives of a thousand short men. We may, however, be misled simply because we notice the striking exceptions; a very tall man married to a very short woman impress themselves on the memory; we do not notice a hundred examples in which there is no particular difference. And it must always be remembered that all observations of this kind are concerned with averages. Children of any degree of brightness may come from families of any size, but the average difference is unmistakable. For example, in the group of Bath schoolchildren the brightest 4 per cent, (on the mental test results) included children from families of all sizes. But as many as 25 per cent, were only children, while 16 per cent, had four sibs or more. In the case of the dullest 8 per cent, of children, once again all family sizes were represented; but only 7 per cent, were only children, and 51 per cent, had four or more sibs. The correct deduction to draw from the facts is that it is a great pity that many parents of bright only children did not proceed to contribute large families of children, who would also have been very bright.
Past and Future
Mr. Sitwell’s first point, however, raises a question of much importance. There is no reason to suppose that the present inequality in fertility has existed for very long. His appeal to history would in all probability be upheld. Detailed information is lacking, but there is little doubt that the present marked differences in fertility date from the time of the sharp decline in the birth rate which occurred in so many countries in the latter part of last century. In a word, from that time, when the total fertility of the population declined, the decline was greatest amongst those of highest intelligence. Actually, it is inconceivable that a fall at the present rate in average general intelligence could have been going on for very long.
This immediately raises the point that perhaps differential fertility will soon cease to operate; that fertility will become equalized throughout groups of differing intelligence, and with the same suddenness as it must be presumed to have attained its present disparity. This is quite possible, and there is, in fact, evidence pointing to the conclusion that the rate of decrease of general intelligence is slower now than it was, say, thirty years ago. For this reason some observers refuse to take too tragic a view. Nevertheless, we cannot be certain how soon this change will occur and a fall at the present rate continued for only a century would have very serious results. Possible Causes
A final word may not be out of place about the possible causes of this kind of differential fertility, though in this paper I have been chiefly concerned to explain the facts and to do what I can to remove the erroneous impression which the press account of my lecture apparently helped to foster in the minds of a good many People.
To say that reduced fertility is due to birth control is no answer at all. In one sense this may be largely true; but this is merely the method by which the result is attained; it merely explains how and not why. Questions of prudence, of anxiety about supporting and educating a large family are undoubtedly factors. Thought for the future is a characteristic of the more intelligent, who are deterred from undertaking too great a responsibility. Related to this is the question of age at marriage. Women married at thirty will have, on the average, considerably fewer children than those married at twenty.
It is not impossible that there is a real biological relation between high intelligence and low fertility. It may be that the higher the level of general intelligence, the more numerous are the objects of interest of life, with a consequent weakening sometimes ?f sex and parental instincts.
Another important possibility has been pointed out by Prof. R. A. Fisher. There !s no doubt that at present, in every walk of life, there is an economic advantage in being one of a small family. Above all else this permits parents to give a child opportunities superior to those they themselves have enjoyed. Infertility is in fact a social asset. Now people tend to be promoted in the social scale for many reasons, amongst which are physical strength, good looks and, above all, intelligence. To these we must add relative infertility. The effect is to concentrate these qualities in the same persons. That is, high intelligence and low fertility tend to become associated together not because of a real underlying biological correlation but because both are socially advantageous and thus tend to come together.
Any discussion on causes must, however, be speculative. There is no doubt about the facts and the actual magnitude of the effect can be estimated reasonably accurately. What is required is further investigation into causes, for there is the hope that when the true underlying factors are revealed it may be possible to arrest so unfortunate a tendency by measures under social and economic control.
REFERENCES
CjVITell, R. B., 1937. The Fight for our National Intelligence. P. S. King, London.
Raser Roberts, J. A., Norman, R. M. and Griffiths, Ruth, 1938. ” Studies on a Child Population. III. Intelligence and Family Size.” Ann. Eugen. Camb. 8, pp. 178-215.
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.