This paper develops a framework for analysing paralanguage, initially inspired by systemic functional linguistic (hereafter SFL) research on early child language development. A distinction is drawn between non-semiotic behaviour (somasis) and meaning (semiosis), and within semiosis between language and paralanguage (using the term paralanguage to refer to semiosis dependent on language and realised through both sound quality and body language, the latter including facial expression, gesture, posture and movement). Within paralanguage a distinction is drawn between sonovergent resources in sync with or in tune with the prosodic phonology of spoken language, and semovergent resources supporting the ideational, interpersonal and textual meaning resources of spoken language's content plane. The paper closes with a brief discussion of the intermodal relations among language, paralanguage and other modalities of communication. Embodied meaning In this paper we introduce a social semiotic framework for analysing paralanguage. Our approach was inspired by Chris Cléirigh's contributions 1 to New South Wales Youth Justice Conferencing research consolidated in Zappavigna and Martin (2018) and Martin and Zappavigna (2018). 2 Cléirigh's work drew on Matthiessen's synopses (Matthiessen 2004; Matthiessen 2007; Matthiessen 2009) of Systemic Functional Linguistic (hereafter SFL) research on early child language development. Following Matthiessen (2009) we use the term paralanguage to refer to gestural resources arranged along what McNeill (1992) christened as 'Kendon's Continuum' (gesticulation, pantomime and emblems), along with the vocalisations outlined in van Leeuwen (1999) not usually included in linguistic descriptions of the segmental and prosodic phonology of spoken language (timbre, tempo, tension, pitch range etc.). In this paper however we will consider only gestural systems. There are of course many ways to classify gestural resources. Kendon 2004 (Chapter 6) provides a thorough historical survey. The most useful vantage point from which to compare classifications is Kendon's Continuum. The introductory chapters in McNeill (McNeill 2000a, 2000b; McNeill 2012) include clear presentations of the model outlined in Fig. 1 below (taken from Sekine et al. 2013). We will cross-reference our work to this model as we present our framework, setting aside the sign languages of deaf communities (ASL, BSL, Auslan, LSF etc.) since these are languages in their own right (themselves involving paralanguage; Johnston 2018).