STURTEVANT 



problem looks even more difficult. 



One thing we want to know is: 

 What portion of existing mental diver- 

 sities is of genetic origin and what por- 

 tion is of environmental origin? Un- 

 der these conditions the usual scien- 

 tific procedure is to try to hold one 

 variable constant, and then study the 

 effects of variations in the other one. 

 This can in fact be approximated in the 

 problem of human mental differences, 

 through the study of twins. Ordinary 

 fraternal twins arise from the separate 

 fertilization of two eggs and are no 

 more alike genetically than are broth- 

 ers and sisters that are not twins; but 

 the environment to which they are 

 subjected is likely to be more nearly 

 the same. Identical twins arise from a 

 single fertilized egg and are genetic- 

 ally identical. If one studies members 

 of such pairs that were separated in in- 

 fanc\% any observed differences must 

 be nongenetic. 



The difficulty here is in the measure- 

 ment of the properties we are inter- 

 ested in. Such studies of separated iden- 

 tical twins were begun by Muller and 

 have been greatly expanded by New- 

 man, Freeman, and Holzinger. I must 

 confess to a feeling, however, that 

 these studies tell us more about what 

 the psychological tests used are really 

 measuring than they do about the rela- 

 tive effects of heredity and environ- 

 ment. 



There are then inherited differences 

 in the sensory components of human 

 mentality and also in components lead- 

 ing to severe derangements. In the 

 area between these extremes the tech- 

 nical difficulties of getting clear-cut 

 evidence are still unsurmounted. But 

 it seems safe to conclude, from what 

 we know of the genetics of complex 

 characteristics in other organisms, that 

 any property as complex and as vari- 

 able as this must have a large amount 

 of underlying genetic diversity. 



261 



This conclusion applies to individual 

 differences. Analogy with other or- 

 ganisms leads likewise to the conclu- 

 sion that there must also be at least 

 statistical differences between racial 

 groups. This is a rule that has held 

 consistently wherever it has been 

 tested— in many different kinds of ani- 

 mals and plants. 



On general grounds, then, as well as 

 from some direct evidence, one must 

 conclude that there are inherited dif- 

 ferences in mental properties among 

 individuals and, at least statistically, 

 among racial groups. But it is necessary 

 to insist that one must not go beyond 

 this point. Specifically, one must not 

 conclude that a particular observed 

 difference is genetically determined. 

 It is, of course, a platitude to say that 

 no one ever does anything for which 

 he does not have the necessary genes; 

 but one must never forget that there is 

 also a necessary environment. It 

 scarcely needs argument that human 

 behavior is strongly influenced by eco- 

 nomic status, tradition, and training. 

 After all, most of the members of this 

 society are in the business of teaching 

 or, at least, have spent a good deal of 

 time and energy pursuing academic 

 work; we are therefore all of us wit- 

 nesses to the obvious fact that men 

 are teachable— that their behavior can 

 be strongly modified by environmental 

 stimuli. 



This caution about attributing ob- 

 served differences to genetic causes, 

 rather than to environmental ones, ap- 

 plies with special force to comparisons 

 among racial groups, for here the ef- 

 fects of tradition and of public opinion 

 are especially strong. 



Another thing that must be avoided 

 is the view that one race (usually that 

 to which one himself belongs) is "bet- 

 ter" than another. All that can prop- 

 erly be concluded is that they are in- 

 herently different. It follows that 



