It has become fashionable in recent years to bemoan the manner in which the social sciences lag behind the physical sciences. Few, however, of those who deplore the slowness of scientific advance in man's knowledge of himself, have any concrete suggestions to make towards increasing the speed of that advance, nor are they clear as to whether the lag is necessary and inevitable, or accidental and removable. The common opinion appears to be that the causes of the lag are entirely, or at least mainly, historical, residing in the fact that while the physical sciences started earlier and have by now advanced further, the social sciences began later, and are still, therefore, lagging behind. This explanation is at least implied in the frequent urging by editors and politicians that the social sciences should hurry to catch up with the physical sciences, or alternatively, that the physical sciences should slow down and wait for the social sciences to overtake them. Viewing the matter thus, as a simple foot-race between physical science and social science, with the latter handicapped only by a slow start, is not altogether desirable, since it appears to ignore certain difficulties involved in a scientific approach to human behaviour, which, while not entirely absent in physicial science, are nevertheless easier to cope with when dealing with non-human material than when dealing with human subject-matter. It is with these difficulties that the present paper will be concerned.

The nature of these obstacles will perhaps be made a little clearer if we begin with a preliminary consideration of what a scientific sociology entails. There is no mystery about this. The programme for the new science was set forth clearly and in detail by the man who invented the word sociology, Auguste Comte.