Causal loop diagrams between structure and behaviour: A critical analysis of the relationship between polarity, behaviour and events

Causal loop diagrams (CLDs) are a qualitative diagramming language for representing feedback‐driven systems. One of their conceptual cornerstones—polarity—has been critiqued for some shortcomings. However, their use has not diminished, and polarity, as well as the relationship between behaviour and events continues to be used in a pragmatic way. A definition of how events relate to behaviour is proposed for a world of continuous behaviour. Then, several previously unconsidered limitations of CLDs are discussed, which prove to be closer to the event‐related thinking and therefore provide little help in avoiding faulty thinking about the behavioural consequences of causal structure. Also, the reasons underlying previously discussed limitations of CLDs become clear. Polarity is shown to be an interface between behaviour and structure. Inexperienced users need external help to use polarity in going from behaviour to structure and back. Such help could come from incorporating behaviour‐over‐time (BOT) graphs into the diagrams. Several areas of further research are identified.