And Cornford and Lévy Bruhl make the same mistake in regard to 'collective representations' as in regard to the group self-consciousness--namely, they credit the savage with theories for the explanation of the beliefs implicitly involved in the 'collective representations,' for example, the theory of mystic participation, which is said to replace for the savage the civilised man's theory of mechanical causation. But, when we regard any material object as holy, or sacred, or as of peculiar value, because it was given us by a departed friend, or belonged to and perhaps was once worn by a beloved person, our behaviour towards it is not determined by any theory of participation; if, for example, we touch it tenderly and with reverent care, that is the direct expression of our feeling. We even behave as if we held the theory of participation, to the extent of believing that the dead or distant person will suffer pain if we ill-use or neglect the object which is associated with him in our minds, but without actually holding that belief; and still more without elaborating a theory of the nature of the process by which our action will produce such an effect. It is only a late and highly sophisticated reflection upon behaviour of this kind which leads to theories for the justification of such behaviour. It is not true, then, that we are logical individuals, while savages are wholly prelogical in virtue of the dominance among them of the collective mental life. The truth rather is that, wherever emotion qualifies our intellectual operations, it renders them other than purely and strictly logical; and the savage or the civilised man departs more widely from the strictly logical conduct of his intellect, in proportion as his conceptions of things are absorbed without critical reflection and analysis and are coloured with the traditional sentiments of his community. The average savage, being more deeply immersed in his group, suffers these effects more strongly than the average civilised man. Yet the interval in this respect between the modern man of scientific culture and the average citizen of our modern states is far greater than that between the latter and the savage.
If one had to name the principal difference between the conditions of life of the typical savage and those of the average civilised man, one would, I think, have to point to the lack in civilised life of those conditions which so inevitably develop the group consciousness of the savage. The family circle supplies to the young child something of these conditions, but in a very imperfect degree only. At an early age this influence is much weakened by general intercourse. As the individual approaches maturity, he finds himself at liberty to cut himself off completely from all his natural setting, to transplant himself to any part of the world, and to share in the life of any civilised community. He can earn his livelihood anywhere, and he knows nothing of communal responsibility.