Rethinking Commonsense Psychology

In everyday life, we spend much of our time interacting and coordinating with other people. In order to do so, we must be able to understand them and there is considerable interest amongst philosophers, psychologists and others in the question of what interpersonal understanding involves.
Most recent approaches start off with the assumption that we can perceive behavior but not the internal mental states that cause it. Thus, in order to understand, explain and predict what others will do, we need to somehow work out what their internal mental states are. 1t is generally maintained that we achieve this through a grasp of human psychology.
This cannot be a scientific psychology, familiar to only small Proportions of the population, as almost all of us use it in our day to day lives. Hence it is claimed that we understand each other by employing a 'commonsense' or 'folk' psychology, an everyday knowledge of mental states. There is a remarkable degree of consensus concerning what this 'folk psychology' most centrally consists of. Most of those working on the topic maintain that it is a matter of inferring what beliefs and desires a person has, which allows one to anticipate their future actions and explain their past actions. Debates about folk psychology are not concerned so much with what we do but with how we do it. Some think that it is a body of conceptual knowledge or 'theory', like a scientific theory. Others think it is a practical ability, rather than a body of knowledge, more like riding a bike than understanding Newtonian mechanics.
This practical ability is said to consist in 'simulating' others, by putting oneself in their situation or psychological predicament, working out what one would do in such circumstances and predicting that they will do something similar. There are also 'hybrid' approaches, which suggest that folk psychology is enabled by a mixture of conceptual knowledge and practical abilities. All these accounts tend to share a conception of what folk psychology is: An ability to attribute internal mental states, it is often claimed, is what facilitates almost all of human social life.
My aim in Rethinking Commonsense Psychology is to challenge this view. I claim that we often interact with others without assigning internal mental states and that we can do so because we share all sorts of social norms with them. For example, 'one drives on the left' and 'one pays for things in shops'.
It is usually possible to predict what someone will do on the basis that 'it's what people do in that situation' or 'it's what they ought to do'. An appreciation of shared norms usually does most or even all of the work of interpersonal interpretation and it is not at all clear why it should require an understanding of beliefs and desires.
I also challenge the assumption that mental states are internal, hidden entities and suggest that we can, to some extent, 'see' other people's emotions, gestures and expressions; their mental states are partially embodied in their behaviors. In order to defend this claim, I appeal to evidence from neuroscience and to the work of various philosophers known as 'phenomenologists', who have provided rich and insightful descriptions of many aspects of experience. I go on to argue that interacting with other people is central to our ability to understand them and that characterising interpersonal understanding in terms of attributing mental states fails to accommodate this.
Furthermore, the terms 'belief' and 'desire' might initially appear to have a clear meaning but turn out to be extremely vague. I bring all these points together to argue that the orthodox conception of 'folk psychology' is inadequate in various important respects. I also note that it is misleading to call it 'commonsense'. In fact, it is no more commonsensical than the account of shared norms, interactions between people, and perception of mental states that I offer in its place.