The Philosophy and Psychology of Mind-Reading and Language

Much research in psychology has focused on two capacities often thought to be uniquely human: the capacity for mentalizing and the capacity to use and understand language.

I assume we all know more or less what we mean when we talk about language. The term "mentalizing," however, designates a scientific concept that requires some explanation. Mentalizing is the process of interpreting the actions of oneself and others on the basis of internal states, such as desires, feelings, beliefs, and reasons. By being able to mentalize, we can predict what other people are going to do and thereby navigate our complex social world (e.g., should I ask for a raise?).

I have recently edited a book (alongside Prof. Kevin Ochsner from Columbia University) on the neural bases of mentalizing. In this book, several chapters discuss relevant evidence concerning the potential relation between mentalizing and language. For example, as discussed by Binder (2021), mentalizing and language processing seem to rely on somewhat overlapping brain regions of the so-called default mode network (DMN).

There is some evidence that our mentalizing ability develops as a result of the acquisition of linguistic skills (see de Villiers, 2021, for a discussion). However, other evidence suggests that highly developed mentalizing skills precede language (see Grosse Wiesmann & Southgate, 2021, for a discussion) and even exist in animals (see Krupenye, 2021; but see Arre & Santos, 2021, for divergent views). Moreover, there is reason to believe that acquiring language may rely on mentalizing (e.g., Birch & Bloom, 2002). Such contradictory empirical evidence should lead us to think more carefully about the possible relation between mentalizing and language.

According to philosophers such as George Hebert Mead (and Paul Grice and John Searle, amongst others), the act of speaking inherently involves mental state attribution. Let’s try to explicate the view of Mead by imagining an interview with him.

An imaginary interview with George Herbert Mead

Mead:

When we say, “Could you please pass the salt?” to another person, we are not instinctively barking or blurting out a string of sounds. We are deliberately trying to generate within the listener a picture of the world that coheres with a picture we have inside our heads. To be successful, we need to estimate what are the chances that the other person understands us, and in doing so, we try to “read the mind” of the other person. Speaking may be a bit like playing chess in that it may involve some implicit strategizing regarding the best ways to achieve the desired goal (i.e., a state of belief) in the listener.

This view of speaking as a strategic game (see, for example, Clark & Krych, 2004) might seem odd, but it is actually quite evident when we try to speak with young children, when we try to convey complex ideas, or when we try to talk with someone who has limited understanding of our language:

If you are in a foreign country, and you want the dessert menu, you may start by saying, “Can I see the dessert menu?” When this doesn’t work, you may use concrete words (e.g., Ice cream? Sweet?) and make physical gestures (e.g., reading) to achieve your goal—constantly trying to infer whether the waiter knows what you are talking about.

Michael: OK, there are cases where communication requires mentalizing. However, is it necessarily the case that all communication requires us to be so strategic and mindful of the listener?

Perhaps it’s typically good enough to be quite egocentric and simply assume that the listener understands us completely; in the worst-case scenario, if some of our messages are non-comprehensible to the listener, she will point this out to us. Such a view—according to which my brain starts off with a default “assumption” that my mental states resemble that of a listener—has been echoed by the philosopher Robert Gordon (Gordon, 2021).

Mead: Well, if we assume that the listener already knows what we are saying, why speak at all? If all is known, why expend resources in blabbing on?

It seems to me that when we decide to say something, we are implicitly assuming that the other person lacks some knowledge—that there is a gap between what is in my head and what is in the listener’s head. If I tell you that “I am in a bad mood today,” it means I must have assumed that you do not know that already. Namely, I spoke because I believed that there is a dissimilarity between my mental states and yours.

Michael: Talking isn’t really necessarily about such an exchange of ideas. At least in its most rudimentary sense, it is more like a physical act of pushing someone to do something.

Consider the example of a young baby yelling at his father to bring him a snack. The child might have learned, via simple learning mechanisms, that yelling the word, “Chocolate! chocolate! chocolate!” brings about the desired result. Thus, it may not be the case that the child wishes the parent to change their mental state to that of “knowing that my child wants a chocolate.” The parent may simply be a sort of button the child repeatedly presses until the chocolate is issued. Such a view of language echoes the behaviorists’ (e.g., Skinner) view of language and the mind.

Mead: I agree that language is sometimes used to produce action (in the terminology of Searle, “directives”). However, at a certain point in our lives, it is clear that we definitely begin to say things that are not meant to generate action, but rather generate beliefs: namely, persuade, inform. At a certain age, children begin to express their own beliefs about the world (e.g., Saying “Fireman Sam is brave!”)—also called “assertives” (from “asserting your belief”).

Psychology Essential Reads

The Time Cure: Taking Our Leave

Is a Texas Bitcoin Miner Making People Sick?

If the supposition of any communication is that our beliefs are shared with other people, why is it necessary to express them? Shouldn’t we just keep quiet?

Michael: Perhaps what toddlers are doing is formulating a hypothesis about the world itself (rather than the mind) and putting it out there to the world as an experiment. When my child says, “Fireman Sam is brave!” perhaps he is implicitly asking, “Is fireman Sam brave?” If I say, “Yes, fireman Sam is very brave!” then this is evidence about the structure of the world. In other words, assertives may be less about generating beliefs in other people’s minds and more about finding out answers, like using Google.

Mead: But isn’t it the case that googling something entails implicitly acknowledging my own mental state of ignorance?

Michael: It’s not as if every exploration of the world entails thinking about the fact that we don’t know something. When a rat explores a maze looking for cheese, it is not necessarily thinking to itself: “Geez, I really don’t know where they put the darn cheese.” Nonetheless, it behaves in ways that reduce its levels of ignorance.

Mead: Wait, but isn’t it the case that trusting Mom and Dad (or Google, for that matter) as a source of knowledge about the world entails that I think they “know” something about the world? Namely that somewhere within them, they hide the answers to life’s questions? Isn’t acknowledging such hidden knowledge trapped within Mom and Dad’s head akin to acknowledging the fact that they have mental states?

Michael: I don’t think so. When a scientist runs an experiment, the outcomes of this experiment are hidden from her view until the results are coded and analyzed. Sure, she had to go through a specific procedure (e.g., using statistical software) to unlock the answer, but it is not as if she thought that the statistical software “knows” the answer. In other words, when a child asks me whether fireman Sam is brave, he may be using me as a tool to an answer rather than a knowing subject.

Conclusion

Sorry, I don’t have a clear conclusion. Despite the intuition that any sort of speaking requires us to generate inferences about the mental state of the listener, I am not sure we can flatly rule out a non-mentalistic account of early language use—wherein language can exist before children acquire the capacity to reason about mental states.

It is possible that further philosophical analyses of these issues could help us better understand what sort of evidence we could use in order to answer these fundamental questions about the human mind.

P.S. I wish to thank the philosopher Robert Gordon for his inputs on this piece