Linguistic convergence is universal: just like charity, it begins at home and, just like charity, it extends to the whole of mankind; it takes place among those who feel they belong to the same language and social group and believe they speak alike with all respects, as between a newcomer and the former resident in the suburban district; but it takes place just as well among Russian and Norwegian fishermen who happen to settle in the same neighbourhood along the Arctic shores, and results in the development of a new form of speech. Besides, convergence will inevitably breed divergence: the new-comer who adapts his speech to that prevailing in his new surroundings will thereby deviate from what had been his set of linguistic habits so far, and all the quicker if the original linguistic differences between the two parties did not humper immediate oral communication.
- Martinet (1962), a pag.104-105

When we come across two languages which present striking resemblances and which, for some good reason, we do not believe to be genetically related, we are inclined to assume a process of convergence determined either by protracted contacts between two comminities or by some common substratum […] It is, however, frequently assumed that convergence is likely to breed purely outward resemblances, the ones that result from direct imitation of sounds, or from borrowing of lexical elements or loose grammatical items, which, since they are not welded to radicals, are easiest to isolate and to transfer from one language to another. There is some truth this, although I am of the opinion that there is no limitation to the extent to which to languages can converge. Convergence will show in trimmings before it manifests itself in fundamentals. Therefore, if similarity is found exclusively in the structural cores of the two languages, we may be induced, when genetic relationship is ruled out, to reject as an explanation the sort of convergence that results from the contact.
- Martinet (1962), a pag.71-72