Abstract

A precise definition of ‘gesture’ is not essential in order to observe that the word carries a form of human embodied expression. Thus, when extending this word to artifacts like robots, its application raises some questions in terms of ambiguity. Why do we use words typically reserved for the living when referring to artefacts? Usually, the state-of-the-art dives into this matter via a socio-cognitive perspective. It is about investigating how robots impact human perception by observing how the human brain tends to attribute intentions to moving and shape-alike objects. This approach presents the agentive language as an effect and describes how we come to talk about robots as wanting to go right or left, making decisions, being intelligent, autonomous, etc.In this paper, I propose a reversal of this approach, examining the connection between the perception of movement and natural language from a linguistic perspective; I discuss what – within language itself – can eventually explain the specific way in which humans translate their perception of movement into words.Firstly, I review some of the general principles that describe the influence of language on cognition with the ambition to highlight the relevance of adopting a linguistic approach to the issue. Then, I support the idea that movement raises a problem of ineffability (i.e. that motion resists linguistic coding). I show how this problem constraints the programming of moving machines and I argue that this linguistic matter has an impact on our common way of thinking and talking about the artefacts that we call ‘robots’.