Dynamic Linguistics (DL) was developed as an offshoot of late generative semantics in the seventies by Bailey (1973), as an attempt to bridge the advances in the understanding of language production and the incorporation of variation in the transformational machinery. One of its most important contributions is the idea of lect, as “a completely non-committal term for any bundling together of linguistic phenomena” (Bailey, 1973, p. 13). Mental grammars would therefore be polylectal and “whatever the level of abstraction represented by a grammar may be, it should contain underlying representations and rules which will generate all the systematic variation in the data at the systematic phonetic level of every lect abstracted from” (Bailey, 1973, p. 13). In addition, following Schuchardt (1885), Bailey expressed the need of incorporating internal reconstruction and the comparative method when a descriptivist writes a polylectal grammar (Bailey, 1973, p. 33), as well as the concept of ‘creolisation’ as the axis of creation of new systems in the historical sense.1 Unfortunately, the incorporation of these in the model was never achieved, with the exception of Seuren and Wekker´s Principle of Semantic transparency in language genesis (1986). A decade later, Seuren (1982) published a model known as Internally Variable Competence that incorporates sociolinguistic variation within Semantic Syntax,2 developed in parallel to that of Bailey’s. Seuren also advocated the importance of formalisation when analysing linguistic phenomena in flux. One of his main contributions was the incorporation of what I call “otherness” into the grammatical machinery of language, intended as a radical dropping of the idealisation of the speaker/listener (Chomsky, 1965, p. 3) as a member of a completely homogenous community. In this talk I will show how such an encompassing model of competence. i.e DL, updated and informed by recent developments in third wave sociolinguistics, particularly the notion of non-polyglossic multilingualism (Lüpke, 2017), as well as lingueme (Croft, 2000) or item-based approaches (Enfield, 2014), can help us to understand the development of small language families or isolates and contact languages in South America and Australia. I will focus on (i) Shawi, a Kawapanan language spoken in Peruvian Northwestern Amazonia, and (ii) Kukatja-Yingkutja, a Western Desert language spoken in the Kimberley region, in Western Australia. The last few minutes of the talk will be devoted to future avenues in DL, as well as ongoing studies, involving languages such as Cocama, Chiquitano, Mochica, and Wolof.