This last view, though it has had a certain vogue among linguists, is worth nothing as a psychological explanation, for it is quite clear that the word is only a starting-point or a vehicle for the process of thought, which is the sole agent of the metamorphosis. If the _nomina_ become _numina_, it is by a disease of imagination or thought, rather than of language.[190] ----- Footnote 190: For a discussion of this point see Goblet d’Alviella, _L’idée de Dieu_, pp. 60 _et seq._ ----- As for the principal thesis, the alleged primitive notion of the Infinite, which is the source of henotheism, it is a metaphysical hypothesis of extreme improbability. Primitive man, enclosed within hard conditions of life, is practical and positive rather than a dreamer; he does not naturally tend towards the Beyond. But a better reason than this, and an entirely psychological one, is that he is incapable of attaining to even a medium degree of abstraction and generalisation. How could a savage, who cannot count up to four, form any idea whatever of the Infinite? Evidently, this notion of the Illimitable is far beyond him. There is only one way of imparting a certain psychological verisimilitude to Max Müller’s view, viz., to strip it of its intellectualist character, and admit as its origin a feeling rather than a notion, a craving, a tendency, rather than a cognition. Of these two factors, which make up all religious belief, one intellectual, the other emotional, which has the priority? Did the notion produce the feeling, or the feeling excite the notion? Such is the problem which lies at the heart of all debates on the origin of religious manifestations. Some place it in the region of instinct: so Renan when he compares religion in humanity to the nest-building instinct of the bird. Others maintain that every feeling presupposes an object. “At first sight this latter theory seems to have logic on its side. It is clear that, in order to love or fear any being, one must have conceived the notion of his existence. Yet, however indispensable it may be to assume an intellectual operation at the beginning of religion, we must recognise that the feelings set in motion by this operation must have long preceded the most ancient formulas of primitive theology.”[191] For my own part, I am inclined to accept the priority of feeling, though unable to supply any arguments based on fact; the period of origins being also that of conjecture. ----- Footnote 191: Goblet d’Alviella, _op. cit._, p. 50. -----