It is proposed that the term "dialect" be held in reserve for the time being and that, for purposes of structural analysis as set forth here, it be replaced by "variety". In deference to the non-structural sense of "dialect" as a type of speech which may itself be heterogeneous, some linguists have broken down the object of description even further to the "idiolect" level.
- Weinreich (1954), Pag. 389

n.17 Some people are not averse to calling modern standardized languages "Indo-European dialects", or speaking of "literary dialects". Dialectology in the sense proposed in this paper need not restrict itself to the folk level, but such usage is one more reason why the term "dialect" ought to be held in abeyance.
- Weinreich (1954), Pag. 397

One problem […] is to determine what structural and non-structural features of language have [...] helped to break up the folk-language continuum into the non-technical units of "dialects" [...] etc.
- Weinreich (1954), Pag. 398-399

This combined research might get to the heart of the question of diasystems as empirical realities rather than as mere constructs. One of its by-products might be the formulation of a technical concept of "dialect" as a variety or diasystem with certain explicit defining features.
- Weinreich (1954), Pag. 399

To designate the object of the description which is in fact a subdivision of the aggregate of systems which laymen call a single language, the term "dialect" is often used. But if "dialect" is defined as the speech of a community, a region, a social class, etc., the concept does not seem to fit into narrowly structural linguistics because it is endowed with spatial or temporal attributes which do not properly belong to a linguistic system as such.
- Weinreich (1954), Pag. 389