There are innumerable books on body language. Most postulate that certain poses, certain postures, certain body reactions have specific meanings, regardless of a person’s words. The non-verbal and the verbal are described as separate channels with separate, often different meanings. Although this position is ubiquitous, it is not empirically tenable, and is based on a misunderstanding that has lasted half a century.

You Cannot Not Communicate

Every student of communication knows the five axioms of communication formulated by Watzlawick et al. (1967). These axioms have shaped the way we talk about communication for 50 years, even though Watzlawick et al. saw them as tentative or provisional hypotheses only. In the authors’ view, the five axioms were neither complete nor final and required empirical verification (Bavelas, 1990). Despite their unproven status, the axioms have been accepted as facts. This misunderstanding has been especially prevalent for axiom 1, “You cannot not communicate”.

Axiom 1 consists of two hypotheses:

Part 1 in particular has generated intensive focus on body language, with the following flawed conclusion:

Two Different Logical Arguments

This hype around body language still persists, although the conclusion is based on two logically different statements and is empirically untenable.

Janet Bavelas (1990) herself points out that her two hypotheses make two different logical arguments. “Every behavior is communication” is a universal statement that assumes communicative properties for every behavior, while “one cannot not communicate” simply states that in the presence of other people some behavior must be communicative.

This means that if the first statement is true, the second statement must also be true. If all behavior is communication, then it is also true that one cannot not communicate. However, if the first statement is false and not all behavior is communication, then the second statement can still be true. Even if not all behavior is communication, it can still be the case that one cannot avoid communicating something in the presence of others.

Or the other way round: Even if “you cannot not communicate”, not all behavior has to be communication.

Non-Verbal Behavior and Non-Verbal Communication

Wiener et al. (1972) made the distinction between non-verbal behavior and non-verbal communication as early as 1972. They saw non-verbal communication as a sub-group of non-verbal behavior.

They distinguished between two different forms of non-verbal behavior:

Informative behavior is behavior of one person, interpreted by another one. The interpretation of the behavior depends heavily on the ideas of the interpreter. Since it takes place primarily “in the head” of the person interpreting and not as an interaction between people, it cannot be empirically researched through observation. Informative behavior is not considered to be interactive and is seen as non-verbal behavior, but not as non-verbal communication.

Communicative behavior, on the other hand, is interactive and can be seen as an exchange between people in conversation. Since communicative behavior takes place “between people” — one person intends to communicate something, and the second person relates to the behavior of the first person — it can be observed empirically and is seen as non-verbal communication.

Not All Behavior Is Communicative

The above distinction results in three non-verbal behaviors:

So not all behavior is communicative and certainly not all behavior is meaningful. The first kind of non-verbal behavior (non-informative, non-communicative) isn’t meaningful to either person. The second type (informative behavior) has a meaning for the interpreter. However, informative behavior does not necessarily have to have the same meaning (or indeed, any meaning) for the other person. Only the third type of non-verbal behavior (communicative behavior) is communicative.

Bavelas and Chovil specify that non-verbal communicative behavior takes place only in the presence of other people (to communicate something), but not when the person is alone (Bavelas, 1990). For example, researchers were able to show that people who bite into a salty sandwich show the facial expression of “disgust” when other people are present (Brightman et al., 1975).

It has long been clear in research circles that the 1967 hypothesis that all behavior is communication, is wrong. Yet the idea persists in the general public. In the words of Janet Bavelas (1990): “As for ‘all behavior is communication’, I thought that this error was only in the literature for five years before being corrected by Wiener et. al (1972).”

Can’t You Not Not Communicate?

Does this mean that axiom 1, “One cannot not communicate”, is wrong? No, not necessarily. Even if the derivation of the axiom has been refuted and not all behavior is communication, it can still be the case that certain behavior must be communicative in the presence of others.

Current microanalysis studies show that communicative behaviors (e.g. nodding, gaze, facial gestures, words) occur continuously in human interactions (Bavelas et al., 2017). Even if there is no verbal interaction, it seems impossible not to communicate anything. The presence of other people seems to make it necessary to communicate at least personal availability or unavailability for communication. In situations in which people want to avoid communication, for example, they often quickly look away when eye contact occurs, communicating that they are not available for further interaction. “You cannot not communicate” is thus an hypothesis that can and should be tested empirically.

How Should I Deal With Non-Verbal Behavior That I Cannot Interpret?

Even if many books on body language propagate the idea, not all behavior is communicative and not all behavior is meaningful. Some might say “Still, it makes sense to at least interpret informative or communicative behavior.”

From our point of view, this usually makes little sense. On the one hand, it is often clear what the behavior of the other person shows. If the other person scratches their arm after a mosquito bite or sips their coffee and smiles, no further interpretation is required.

Interpretations are highly dependent on the person making them and have often nothing to do with the other person’s experience. People are different and often react differently. Although books about body language try to tell a different story, they don’t build on scientifically proven findings that show that behavior in general — that is, regardless of individual characteristics and the current interaction — has a meaning. Still, many writers promote this fallacy without a scientific base. An example is the idea that “If someone crosses their arms, it means that the person is rejecting you or your idea”. However, the person may just be cold, or more comfortable with crossed arms, or easing a sore back. If each behavior has a generic, person-independent meaning, then this would mean that all humans generally behave in the same way. Instead, we see every day that this is not the case.

Therefore, instead of interpreting, it is worth asking the other person about their behaviour. The only person who can possibly know if their non-verbal behavior might mean something is the person you’re talking to.