On the murky dissociation between expression and communication

Abstract

The authors present an ambitious attempt to outline the gradual evolution of the cognitive foundations of ostensive communication. We focus on three problematic aspects of the distinction between expression and communication: ambiguity in the distinction's central principle of “complementary mechanisms,” inconsistencies in the application of the distinction across taxa, and the dismissal of mentalizing in nonhuman primates.

The authors propose a fundamental distinction between expression and communication: They refer to “expression” as behavior whose function is to induce an intended reaction in the other and reserve “communication” for behavior whose function is to generate an intended reaction by means of stimulating evolved complementary mechanisms of interpretation in the other. This distinction ascribes the status of communication to the waggle dance in honeybees because the behavior of the signaler triggers an evolved complementary interpretive mechanism that induces the receiver's response (to visit a particular location). Crucially, the authors dismiss nonhuman primates' intentional use of gestures as not communication because although they induce responses in the recipient, and the authors concede that informative intentions may drive the signaler's actions, primates may lack the capacity to represent the mental states their signals create in others. In their view, “a gorilla thumping his chest might associate this with the behavioral effect of conspecifics backing away, but not with the effect of them being frightened.”

This distinction between communication and expression creates three interpretative problems. First, contrary to the authors' proposal, the gorilla chest-beating display may qualify as communication because it (and other intentional gestures by nonhuman primates) may indeed trigger an evolved mechanism of interpretation in the recipient. Chest-beating is a species-specific signal typically used in agonistic contexts (although juveniles also use it in play) to intimidate others by conveying strength. In this sense, chest-beating is not so different from other species-specific signals, such as red deer roaring, commonly used in male–male contests. It is thought that such signals evolved precisely because opponents can interpret them as a reliable indicator of the signaler's strength. Thus, production and interpretation of gorilla chest-beating and other nonhuman primates' intentional gestures are likely to have evolved (or developed during ontogeny) as complementary systems.

Second, if nonhuman primates' intentional use of gestures does not qualify as communication on the grounds that signalers do not mentalize their effects on others, it is unclear why the honeybee dance constitutes communication. Bee signalers and recipients have evolved a fixed but complementary system of signal decoding, where mentalizing is unlikely to play a role. Therefore, the assertion that bees engage in true communication seems to hinge on the complementary nature of their system. However, the authors do not present a precise explanation of the factors which make a mechanism “complementary” that can be mapped onto behavior occurring at various levels of cognitive engagement. The authors argue that nonhuman primates' intentional gestures are expressive but not communicative, because although the signal can generate a desired behavioral reaction, it does so by “trigger[ing] a mixed set of mechanisms that may not be complementary in the relevant way.” This distinction seems to imply that for mechanisms to be “complementary,” they must be symmetrical in their level of cognitive engagement. In other words, the honeybee's dance uses fixed signal encoding and decoding, thus the signaler and the recipient engage with the communication in a cognitively similar manner. In contrast, the gorilla's chest-beating display may consist of informative intention from the producer, while the recipient may engage only with a behavioral reaction to their own emotional response of fright: One engages at a more complex level of cognition than the other. In our view, however, for informative gestures (like chest-beating) to be selected and maintained in the evolved repertoire of the species, they must successfully induce their intended reaction in the other often enough to remain sufficiently beneficial for fitness and survival. This makes it more likely that species-specific display gestures achieve their effect by inducing specialized functional mechanisms, whether they are cognitively symmetrical or not, that account for their sufficient degree of success and stability rather than inducing an unspecified “mixed set of mechanisms.”

Third, nonhuman primates' intentional use of gestures may qualify as communication after all, even when mentalizing is required. We are uncertain as to how the authors substantiate the assertion that the displaying gorilla does not knowingly intend to induce a frightened state, when there is relevant evidence that apes are aware of some causal psychological mechanisms mediating their intended effect on their conspecific. Significantly, before presenting a species-typical gesture that displays relevant information about the context of their subsequent behavior – for example, an “arm-raise” gesture to inform a conspecific partner that the ensuing hitting behavior is meant in play – apes first check and make sure that their addressee is positioned so that they have perceptual access to the behavioral display (Tomasello, Call, Nagell, Olguin, & Carpenter, 1994). In fact, signalers facilitate perceptual access by engaging in tactile gestures (e.g., throwing objects, poking, touching of distal body parts), or by moving into the line of sight of an individual to ensure that they see and attend to a visual gesture (Liebal, Call, & Tomasello, 2004). This audience-sensitive, selective use of visual gestures indicates that apes can monitor the other's perceptual orientation and intentionally modify their own signaling behavior in response. This competence establishes the preconditions for mentalization of informative intention; apes engage with the relevant psychological mechanisms of the other that are necessary for and involved in perceiving the gesturing ape's informative behavior. Furthermore, apes' coordinated use of “attention-getters” together with the subsequent display of species-specific gestures that encode relevant information has been posited as convincing evidence of non-verbal intentional communication in apes (Tomasello & Call, 2019; Warren & Call, 2022).

It appears to us that the relevant and intriguing question for evolutionary cognitive science is to further explore and characterize the underlying cognitive mechanisms and functional adaptations that serve apes' capacity for intentional communication. In particular, we should investigate whether and to what degree this rudimentary communicative system can be considered a proto-form or evolutionary precursor of ostensive communication proper as it emerged in humans. The authors attempt to identify how the relevant cognitive capacities that serve ostensive communication may have evolved in a gradual manner. We feel, however, that the theoretical distinction and definitions proposed to differentiate between expression and communication fail to serve this purpose and lead to more confusion than clarity in how these definitions can be applied to nuanced behavior across taxa.