Social representations research is often undertaken by scholars who seek to ‘give voice’ to knowledge(s) that are held by socially disenfranchised individuals and groups. However, this endeavour poses a number of problems in practice, not least because it assumes that the ‘voices’ voiced by individuals and/or groups in social research will be unambiguous and uniform, and unchanged by the research encounter. Despite the growth of attention to the critical potential of social representations theory, there remains a lack of scholarship on the relationship between the theory's emphasis on the relational, or dialogical, nature of social life (the Ego‐Alter‐Object relation) and the implications of this for critical research. In this article, I argue that the dialogical epistemology from which social representations research must depart incites reflection upon (i) the design of empirical studies, which must equally attend to both ‘Ego’ and ‘Alter’, and (ii) the researcher‐researched relationship, which may itself be best viewed as an Ego‐Alter interaction. I make the case for adopting dialogical epistemology in critical social psychology, and argue that this is essential to undertaking ethical social research. I conclude by suggesting that claims to ‘giving voice’ have little place in critical social psychology in general, and social representations scholarship in particular.