We must, however, at certain points in anthropological inquiry have recourse to the conjectural method. Its value lies in the fact that it states, and states clearly, the issue which is before us, and it is always possible to take up the conjectural position and endeavour to ascertain whether the neglected facts of human history which it expresses can be recovered. Its danger lies in the neglect of certain anthropological principles which can only be noted from definite examples, and the significance of which can only be discovered by the handling of definite examples. I will refer to one or two of the principles which I have in mind. Thus, it is necessary to distinguish between what is a practice and what is a rule. A practice precedes a rule. A practice incidental to one stage of society must not be confused with a rule, similar to the practice, obtaining in a different stage of society. Again, it must be borne in mind that identity of practice is no certain evidence of parallel stages of culture, and already it has been pointed out that identical practices do not always come from the same causes. Thirdly, it has to be borne in mind that primitive peoples specialise in certain directions to an extreme extent, and correspondingly cause neglect in other directions. The normal, therefore, has to give way to the special, and it is the degree of specialisation and the degree of neglect which are measuring factors of progress; in other words, it is the conscious adoption of certain rules of life with which we alone have to do. These principles are apt to be wholly neglected, and, indeed, the last-mentioned element in the evolution of human society does not enter into the calculations of analytical anthropologists. They provide for the normal according to scientific ideas of what the normal is. They either neglect or openly reject what cannot be called abnormal, because it appears everywhere, but which they are inclined to treat as abnormal because it does not fit into their accepted lines of development. That which I have ventured to term specialisation and neglect is a great and important feature in anthropology. It obtains everywhere in more or less degree, and accounts for some of the apparently unaccountable facts in savage society, where we are frequently encountered by a comparatively high degree of culture associated with a cruel and debasing system of rites and practices which belong to the lowest savagery. Dr. Haddon has usefully suggested the term "differential evolution" for this phenomenon in the culture history of man,[310] and as I find myself in entire agreement with this distinguished anthropologist as to the facts[311] which call for a special terminology, I gladly adopt his valuable suggestion.