1. Introduction I thank Ekaterina Abramova (2018) for using my Mirror System Hypothesis (MSH) of ‘how the brain got language’ as the grounding for her thoughtful critique of pantomime. In her abstract, she asserts that ‘the notion of a pantomime [in MSH] presupposes two sophisticated abilities that themselves are left unexplained: symbolization and intentional communication’. She offers ontogenetic ritualization (OR) as ‘an alternative mechanism that can lead to a suitably complex language precursor while avoiding pantomime altogether’ (emphasis added). I will defend the merits of pantomime while showing that OR is better regarded as a complement to pantomime than as a plausible replacement. 2. MSH and the issue of symbolization and intentional communication To respond to Abramova’s critique, and even though her paper offers a good summary, I need to summarize MSH anew. Since the arguments and data in support of what follows are available in my book (Arbib 2012), or the slight update (Arbib 2016), I omit references. 2.1 LCA-m (last common ancestor with macaque) Ability to execute and recognize manual actions (parity). Call repertoire is small, innate. Virtually no ability to learn and use novel vocalization (though some ‘tuning’ of the innate repertoire is possible). 2.2 LCA-c (last common ancestor with chimpanzee; Abramova’s LCA) To the above, add: Simple imitation, attempting to use familiar actions to achieve recognizable goals. A gestural repertoire in which some gestures may be innate but others are learned by, e.g., ontogenetic ritualization (OR). (A correction to Arbib (2012): Figure 8.1 (p. 215) suggests that it takes simple imitation to support OR. This is wrong. Subsequent modelling (Arbib, Ganesh, and Gasser 2014) showed that interaction between two apes could support OR so long as their brains supplemented three capabilities of the macaque brain—(1) a dorsal ‘how’ pathway and a ventral ‘planning’ pathway in the reach-to-grasp system, (2) the capability of mirror neurons to emerge through learning; and (3) opportunistic scheduling of actions—with far better proprioceptive control of arm and hand than that of the macaque. This is an argument for more detailed analysis of brain mechanisms than is usual in discussions of language evolution.) Successful attempts by humans to teach apes the use of lexigrams or some signs (but not the grammar) of a sign language suggest that—if placed in a language-rich environment provided by humans—apes can acquire fragments of a vocabulary but not grammar. The use of gestures shows that intentional communication is already established in LCA-c. The acquisition of human-demonstrated ‘symbols’ shows that LCA-c was symbol-ready, even though LCA-c ‘cultures’ were not symbol rich. 2.3 Hominids prior to homo sapiens The first key addition to primate capability posited post-LCA-c by MSH is complex action recognition and imitation (CAR&IM): complex action recognition (CAR) is the ability to attend both to the goal and subgoals and to some details of the constituent movements of an observed behaviour. Complex imitation (CIM) is the ability to use such recognition to acquire new skills. Pantomime is then posited to be a new skill that rests on CAR&IM—going from (1) a sequence of transitive actions on objects to achieve a praxic goal to (2) using a sequence of related intransitive movements in the absence of the objects to communicate a message to another. Its success then depends on possibly novel brain mechanisms as well as a cultural innovation that supports both the creation of novel pantomimes ‘on the fly,’ and the ability to recognize that a novel behaviour is indeed an attempt to communicate, with interpretation of the message resting on recognition of the behaviour being pantomimed and the context in which the pantomime occurs. Before going further, let us recall Abramova’s assertion that the notion of a pantomime in MSH presupposes symbolization and intentional communication, and these are unexplained. However, since MSH hypothesizes that LCA-c was already capable of limited symbolization and intentional communication, with the range of this increased by immersion in a human symbol-rich environment, there is no need to re-explain these abilities in positing pantomime. Nonetheless, just as MSH found it necessary to distinguish simple from complex imitation, so must we distinguish various complexity levels for symbolization and intentional communication. The claim here is that biological evolution endowed the human brain with a capability for ad hoc pantomime that went beyond that available to LCA-c (but see Russon 2018, for an argument that great ape pantomime may be more powerful than usually suggested). Having said this, the possible transience of an ad hoc performance of pantomime makes it somewhat unsatisfactory as a symbol—but by the posited nature of post- CAR&IM pantomime, it can be recognized by observers beyond the intended recipient of the performance, and can be freely imitated by them. This is the posited extension to the form of CAR&IM directed at objects, and which may require a change in brain-readiness. Pantomime thus escapes the dyadic and asymmetric restriction of OR. However, each pantomime may be highly ambiguous. Being based on the movements made in carrying out some transitive (i.e. object-directed) behaviour it may be taken, as context suggests, to represent either the object at which the behaviour is directed, the overall goal of the behaviour, or some aspect of the behaviour itself, for example. The solution within MSH is, as Abramova notes, ‘the ability to entertain such diverse thoughts and a need to distinguish pantomimes “for action” from pantomimes “for objects” that lead to the practice of introducing small modifications into them, thus fueling the transition to conventionalized protosigns and paving the road to language’. More specifically, according to MSH, pantomime lays the basis for a conventionalization process that does yield indubitable symbols, namely protosigns. Conventionalization (akin to OR; but now shared reciprocally across a group) may transform a set of similar pantomimes used in similar situations into one or more gestures (a protosign) with more constrained meanings. Such gestures then form a system, also called protosign, which initially has little or no grammar but does provide symbols for use in intentional communication. Given this, I agree with Abramova that we should not suppose that concepts for the related action or object are available in advance—the very process of forming new symbols may be engaged in a virtuous circle forming new concepts. Since humans have flexible vocal learning and control, and other primates do not, one must hypothesize that this evolved after LCA-c but (most likely) before homo sapiens. MSH posits that adaptive pressure to add vocal gestures to protosigns favoured individuals with increased control of the vocal cords and that collaterals from manual control of protosign favoured the emergence of mechanisms that supported the emergence of protospeech. This hypothesis is disputed, but lies outside the scope of Abramova’s article. 2.3.1 Homo sapiens MSH holds that early homo sapiens had a language-ready brain but had only protolanguages (a range of ‘protowords’ with little or no grammar) but not languages (with a range of constructions supporting a compositional semantics)—and that it was cultural evolution that led to languages and the communities that could employ them. 3. Ontogenetic ritualization is a complement to pantomime, not a replacement Having disposed of Abramova’s argument that MSH leaves symbolization and intentional communication unexplained, we now assess her claim that ‘ontogenetic ritualization can lead to a suitably complex language precursor while avoiding pantomime altogether’. The catch with OR is that it yields a symbol that is private to a dyad, say A and B, and asymmetric—if A develops gesture Z via OR through interaction with B to elicit behavior Y from B, there is no guarantee that B would perform Z to elicit behaviour Y from A. Abramova claims to show how ‘triadic’ OR can yield a communication system that (1) supports symbolization, (2) is reciprocal, i.e. gestures can be used equally by senders and receivers (presumably, with parity of meaning for sender and receiver) and (3) is intersubjective in that the same signal can be used with multiple partners. She notes that the last two properties capture parity of meaning, but I note that (2) implies (3) if one allows a speaker to address any member of the group under consideration. After some discussion of the relation between tool use and language emergence, Abramova introduces a scenario to support the idea of triadic ontogenetic ritualization (TOR): Let us suppose now, that due to some changes in hominid evolution there is an increased pressure for tool use and joint action more broadly. [And] suppose … that … there arises a need to make or use tools and other objects together, cooperatively, leading to an increase in triadic interactions in the community (i.e. interactions between two individuals around an object, rather than merely between two individuals). (Emphasis added) By analogy to dyadic OR, we could envision the following sequence of events: Individual A performs behaviour X with respect to object T; Individual B reacts consistently with behaviour Y towards the object T; Subsequently B anticipates A’s performance of X, on the basis of its initial step, by performing Y; and Subsequently, A anticipates B’s anticipation and produces the initial step in a ritualized form (waiting for a response) in order to elicit Y towards the object T (e.g. making the joint action more efficient or initiating it). I think Abramova calls this ‘triadic’ because it involves T as well as A and B. However, this sense of ‘triadic’ seems counterproductive as a step toward language, as I shall argue below. Abramova continues: … in a hunting scenario, one could conceive of the following. A is chasing a wild pig (or some other small animal) and trying to hit it by throwing stones at it but the pig keeps running away. B joins the chase and tries to maneuver the pig closer to A's position by making loud noise and waving his arms. Both succeed, kill the pig and a whole tribe feasts on it. The hunting trick gets repeated until A can anticipate that making stone throwing movements in the presence of a pig is enough to request B to execute his maneuvering routine.It seems to me that A’s behaviour, ‘making stone throwing movements in the presence of a pig’, is simply like a ‘pantomime’ but based on long interaction rather than the flexible process posited in MSH. Moreover, in this case, it would be more adaptive for A to ‘pantomime’ arm-waving to signal to B which of his possible actions would be appropriate. Indeed, to have gestures for stone throwing and arm-waving that are specific to a pig-chasing scenario and emerge through a long period of interaction between A and B seems maladaptive. Far better to have pantomimes that can be generated ad hoc by any A to communicate something to any B an appropriate behaviour. In short, pantomime seems like a powerful step to language, whereas TOR does not. Note, too, that a gesture that emerges from Steps (1)–(4) here is neither reciprocal (there is no guarantee that B would make the same signal if roles were reversed) nor, a fortiori, is it intersubjective. In contrast CAR&IM is by its very definition reciprocal and intersubjective in that it was posited to evolve in such a way that any one member of the group can, if motivated to do so, recognize the structure of novel actions of another and learn to imitate it. The way in which ad hoc pantomime is defined as building on this skill means that it inherits these two properties. So, the key question is this: does Abramova offer an evolutionary scenario that takes us from TOR to a system of reciprocal and intersubjective gestures? Unfortunately, her attempt to support the contention that ontogenetic ritualization is ‘an alternative mechanism that can lead to a suitably complex language precursor while avoiding pantomime altogether’ seems unhelpful: As a result [of OR], a single individual is typically only on the producing or a receiving end of a certain gesture. However, if more complex interactions enter the life of the group, part of these interactions could be such that they occur between grown-up individuals who are similar in their background practical knowledge and action possibilities, allowing for their roles to be in principle interchangeable. … After ritualization of such a scenario, the same individual would have an opportunity to both produce a certain gesture and respond to it, creating a reciprocal gesture … . Finally, if such cooperative interactions are important enough for the whole group and scaffolded by the use of the same types of objects, there is no in principle reason why a single individual cannot enter the same interactions with multiple others and therefore [no reason why] the gestures that emerge cannot go beyond a single dyad.To ‘make OR work’, Abramova still requires ritualization by a dyad not only to yield a novel gesture but also for each new person to acquire that gesture, rather than having the more rapid use of pantomime to solve diverse communicative problems as needed. Moreover, a key step in MSH is to get to protosigns. For this, one must go beyond ritualization or pantomime—both of which yield gestures, whether chronic or transient which emerge ‘naturally’ from some behaviour—to gestures which narrow the ambiguity of the original gesture by a process of conventionalization which may yield variants with different applicability which, now, are reciprocal and intersubjective only for members of the protosign community. However, a key point about symbols is that, in general, one can acquire them by imitating their use in appropriate contexts without having been engaged in (or having to be led through) the ritualization or conventionalization processes that initially led members of the community to engage them. In short, one joins that community by processes which (in part) engage imitation, and this mechanism is central to MSH and missing in Abramova’s TOR-centric account. Conflict of interest statement. None declared. References Abramova E. ( 2018) ‘ The Role of Pantomime in Gestural Language Evolution, its Cognitive Bases and an Alternative’, Journal of Language Evolution . Arbib M. A. ( 2012) How the Brain Got Language: The Mirror System Hypothesis . New York/Oxford: OUP. Google Scholar CrossRef Search ADS Arbib M. A. ( 2016) ‘ Towards a Computational Comparative Neuroprimatology: Framing the Language-Ready Brain’, Physics of Life Reviews , 16: 1– 54. Google Scholar CrossRef Search ADS PubMed Arbib M. A., Ganesh V., Gasser B. ( 2014) ‘ Dyadic Brain Modeling, Ontogenetic Ritualization of Gesture in Apes, and the Contributions of Primate Mirror Neuron Systems’, Philosophical Transactions of the Royal SocietyB: Biological Sciences , 369: 20130414. Google Scholar CrossRef Search ADS Russon A. ( 2018) ‘ Pantomime and Imitation in Great Apes: Implications for Reconstructing the Evolution of Language’, Interaction Studies , 19: in press. © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com
Journal of Language Evolution – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.
Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.
It’s easy to organize your research with our built-in tools.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera