Co-speech hand gestures have largely been assigned a supportive role, ‘enriching’ (Ebert 2024) the primary meaning that is conveyed in speech. As such, the meaning of the gesture itself is often considered too vague to be independently interpretable (McNeill 2005). Instead, one looks to accompanying speech and identifies the primary meaning with which the gesture is ‘affiliated’ (Schegloff 1984). For example, a gesture performed by rotating a hand in small circles, flexing at the wrist, may be interpreted as a “mixing” gesture if it accompanies a spoken description of mixing. I argue that this treatment conflates a gesture’s contextualized function with its inherent meaning (Bavelas 1994), thus inhibiting advancements in our understanding of how gestural meaning actually works.

As an alternative, I take a broadly ‘schematic’ approach, understanding gesture as meaning the action or image it resembles (Cienki 2013; Müller 2017). The contextualized function of a gesture is then determined by how it is incorporated into the utterance (Lascarides & Stone 2009; Parrill & Sweetser 2004). I look specifically at gestures for which a clear affiliate in speech can be identified, and explore how our perception of the gesture’s meaning and function changes when we first conduct a schematic analysis of the gesture itself. This work shifts our notion of gesture as being affiliated with (and dependent on) speech, to being co-expressive with it. As such, building models of gesture meaning in particular become an indispensable part of building multimodal models of linguistic meaning more broadly.