[…] the foundations of a linguistic typology are to be looked for in the way in which each language community proceeds to analyse experience into a number of elements in such a way that the linear order of their succession will not prevent hearers from perceiving the nature of their mutual relations and thus reconstructing the total experience.
- Martinet (1962), a pag.101-102

What contemporary linguists somewhat pompousely call ‘typology’ is not basically different from what a long line of thinkers have tempted to do when they classified languages, not according to their antecedents and genealogy, but with respect to their directly observable characteristics.
- Martinet (1962), a pag.66

When we have to deal with a number of languages which we suspect to have converged, it may prove useful to operate with the sort of concentrated characterization which we call a typology.
- Martinet (1962), a pag.71