Wittgenstein noted that human beings adopt facial, postural and vocal behaviors that are congruent with the displays they perceive, and that these displays often represent mimicry. Social perception, understood by Wittgenstein as the ability to directly read off emotion and thought from a person’s behavior, has a direct effect on social behavior. As a result, we often do what we see others doing. How can a person see that the child wants to touch the dog, but doesn’t dare? Wittgenstein’s answer is that an observer ‘may also mimic a human being who would like to touch something, but doesn’t dare’ (Wittgenstein, 1980a, §1066). This suggests that mimicry is essential for understanding what the perceived behavior means. If a person could not do this, then he would fail to understand something important about what is expressed. The perception-behavior link Wittgenstein argues for is the human tendency to act in the same way as we see others act.

Wittgenstein considers as facial mimicry the congruent facial reactions to the emotional displays of others. Specifically, mimicry is understood as an expressive component. As facial mimicry is the direct consequence of perception, we do not need additional private mechanisms to engage in it. No introspection is required, nor a conscious decision. We just do it. For example, the difference between a friendly smile, a wry, sarcastic, ironic, or cruel smile may be no more than a minute difference (a thousandth of an inch) in the orientation of the facial features, but we perceive it, and not by measuring the difference. Moreover, we mimic such expressions without looking in the mirror, and without monitoring or checking the precise disposition of our features (Wittgenstein, 1958, §285). We see friendliness in a face; we do not infer it from such-and-such a facial configuration we know accompanies our own feelings: ‘We describe a face immediately as sad, radiant, bored, even when we are unable to give any other description of the features’ (Wittgenstein, 1980b, §570). The same can be said of facial mimicry: ‘One may also say: ‘He made this face’ or ‘His face altered like this‘, imitating it – and again one can’t describe it in any other way’ (Wittgenstein, 1980a, §920). There is something primitive about the ability to transpose what we perceive in the facial expression of someone else, to one’s own facial expression and activity. Indeed, one-month-old infants have been shown to smile, stick out their tongues, and open their mouths when they see someone else doing the same (Meltzoff & Moore, 1977).

However, facial mimicry may not always occur. An unsightly and unusual facial expression that leads to puzzlement or prejudice may not stimulate congruent facial expressions in an observer. Depending on the nature of the decoding task, facial reactions to emotional facial expressions may be either affective or cognitive. If we were to meet people who randomly mixed up posed and spontaneous emotional facial expressions, it is conceivable that ‘we should not know where we were within them’ (Wittgenstein, 1992, §201). And, if we had to decide whether an emotional facial expression was posed or spontaneous, no facial mimicry may accompany a cognitive judgment such as ‘I have no idea what is going on inside of them’. By contrast, Wittgenstein suggests that we may find facial mimicry when observers are asked to make affective judgments concerning emotional facial expressions.

In other cases, counter-mimicry may result. Wittgenstein imagines a face incapable of making gradual and subtle movements, and which had a limited number of facial expressions. When it altered, it would snap straight from one expression to another. He asks: ‘Would this fixed smile really be a smile? I might not be able to react as I do to a smile. Maybe it would not make me smile myself’ (Wittgenstein, 1980b, §614). This suggests that facial mimicry is not obligatory, and can be inhibited. Wittgenstein may have supported a conception of facial mimicry as ‘default social behavior’: we naturally mimic, as long as other considerations do not have a reason to intervene. For example, ‘a person who had seen only one facial expression couldn’t have the concept ‘facial expression’. ‘Facial expression’ exists only within a play of features’ (Wittgenstein, 1992, §766). This means that a person who had only seen ‘happy’ faces could not mimic them as happy. 

In sum, Wittgenstein acknowledges that observers mimic facial displays. However, the possibility of counter-mimicry he mentions may suggest that this reaction is somewhat less automatic and reflex-like than otherwise thought.