If Durkheim's concept of social facts is as I have presented it here – as one which can only properly be understood in terms of a Kantian concept of social phenomena – why did not Durkheim present his own argument in this way and, in short, simply call ‘social facts’ ‘social phenomena’? We have seen already that there is some evidence to suggest that Durkheim uses the terms ‘social fact’ and ‘social phenomenon’ interchangeably in The Rules, almost as though he does not really mind which term we use. Nevertheless it is quite clear that his preferred term for his new concept – the key concept in many ways of his entire sociology – is that of a ‘social fact’ rather than a ‘social phenomenon’. What then is the explanation of this puzzling decision? There seem to have been three main reasons for this.

First, Durkheim seems to have largely inherited this usage. Durkheim's use of the term ‘social fact’ seems to be very much influenced by, and in fact derived from, the writing of the Abbé Henri de Tourville who, together with Edmond Demolins, founded a new journal called La science sociale in 1886. According to Marcel Fournier, ‘Tourville developed a new classification or “nomenclature of social facts” as a guide to fieldwork and the production of future monographs’ (2012, 65), and although there seems to be no actual evidence to support this view, it seems impossible that Durkheim was unaware of the publication of this new social science journal, and was influenced in his thinking on this question by Tourville's work while he was preparing his doctoral dissertation on the subject of the division of labour.