In this paper we propose a framework to re-think and widen the theoretical assumptions behind so-called weak signal theory in futures studies. This paper suggests that the present understanding of weak signals views the topic as a question of linear emergence, that is, a sequence of development that perceives the topic evolving from an embryo towards a full-blown future issue or trend. However, this viewpoint does not take into account the context and perspective – or positionality – of the perceiver. This issue has, we argue, wide-ranging effects on the identification and interpretation of weak signals. Thus, as a response to this gap, we outline a relational theory of futures knowledge that aims to consider this missing theoretical baseline and widen the understanding of the nature and formation of weak signals. The theory builds on selected tenets of behavioural economics, psychology, human geography and anthropology. This paper outlines the principles of current weak signal theory and depicts the relational theory. It also provides selected empirical examples to emphasise the points made.