Skip to main content Skip to navigation

Origins of Syntax Conference, December 12th-13th 2023

The origin of syntactic structure has been a topic of contention in research on the evolution of language. In recent years, proponents of Chomskyan and non-Chomskyan views have disagreed about whether syntactic structure is learned or unlearned; the product of gradualistic or saltational evolution processes; and about whether syntactic abilities are a product of natural selection for better communication, or for other cognitive processes. While both sides have agreed that syntax is likely to be uniquely human, comparative psychologists have also produced new evidence of combinatorial capacities in the communication of non-human species. However, it remains controversial whether this evidence meets the criteria for hierarchically structured syntax that have been proposed by linguists and philosophers; and consequently whether animal combinations count as evolutionary precursors of human syntax.

In this interdisciplinary conference, we bring together philosophers, comparative psychologists, and cognitive scientists from a range of disciplines to discuss their recent work on the ontogenetic and phylogenetic origins of syntax, in order to make progress in our understanding of these fundamental issues.

The Origins of Syntax conference is organised by the UKRI-funded Communicative Mind research group. It will take place at Radcliffe Conference Centre at the University of Warwick on December 12th and 13th December 2023. In person attendance is free, although you are requested to register in advance because capacity is limited. To register, please contact giulia.palazzolo.1 [at]

Online attendance will also be possible.

Confirmed speakers:

Nick Chater (University of Warwick)

Cas Coopmans (Max Planck Institute for Psycholinguistics)

Cathy Crockford (ISC Marc Jeannerod)

Olga Feher (University of Warwick)

Richard Moore (University of Warwick)

Nirmalangshu Mukherji (Delhi University)

Ross Pain (University of Bristol)

Giulia Palazzolo (University of Warwick)

Ronald Planer (University of Wollongong)

Ljiljana Progovac (Wayne State University)

Simon W. Townsend (University of Warwick and University of Zurich)



12th December


13th December


Coffee on arrival


Coffee on arrival


Palazzolo & Moore








Morning beverage


Morning beverage














Afternoon beverage


Afternoon beverage









Lunch is served at the Radcliffe restaurant and is provided for all attending participants.


Chimpanzees show the capacity to communicate about concomitant daily life events

Cathy Crockford et al.

A universal of human language is its versatility in communicating combined information about everyday events. Versatility is achieved through using a diversity of information combination and modification mechanisms. Versatile combinatorial systems of communication can be selected for if (a) several vocal units are flexibly combined into numerous and long vocal sequences and (b) vocal sequences convey information about numerous and diverse daily life events. If (a) and (b) are in place, then we expect to find (c) a diversity of information combinatorial and modification mechanisms. Versatile combinatorial systems combining (a), (b) and (c) have not yet been found in non-human animals. Chimpanzees are a candidate for such a system since they fulfil (a).

Here, we test the potential for (b) and (c) in chimpanzees. We analyzed 9391 vocal utterances across the repertoire of 103 wild chimpanzees, Tai Forest, Ivory Coast, and the events occurring during emission. In support of (b), chimpanzees used vocal sequences across a range of daily life events and twice as often when experiencing concomitant than single events. We also found a positive correlation between the diversity of utterances and the diversity of events. In support of (c), when focusing on two-unit utterances (bigrams) and the events during which the bigrams were uttered, we found patterns consistent with several information modification mechanisms found in other animal species: new information creation, combination of information and ordering effects. Previously, usually only one such mechanism has been found per species.

Our results show the potential for chimpanzee utterances to convey combined information about numerous daily life events. This capacity is likely achieved by utilizing a diversity of information modification mechanisms. Whilst we did not assess the construction of meaning through hierarchical ordering, a requirement of syntax, the chimpanzee vocal system may demonstrate a step from which generalized combinatoriality could have evolved.

Grammar through spontaneous order

Nick Chater

How do the algebraic regularities in natural language, described by generative grammar, emerge? One traditional viewpoint has been that these are encoded of a species-specific and innately specified universal grammar, which has somehow come to be part of the human biological endowment. From this point of view, the strange mix of regularities, subregularities and downright exceptions observed across languages and levels in linguistic analysis are somewhat puzzling. An alternative perspective is that language begins through attempts to solve immediate communicative problems between specific people on specific occasions; but each new communicative exchange draws on precedents from past exchanges, and sets precedents for future exchanges. Over time, specific linguistic patterns will become entrenched, and layered upon each other, to create a complex spontaneously ordered system, analogous to case law. From this point of view, the grammatical patterns in language are always various, partial and subject to exceptions. This work is joint work with Morten Christiansen.

Origin of core syntax: Some methodological issues

Nirmalangshu Mukherji

I will approach the issue of origin of syntax of human languages from a somewhat wider

perspective and narrow down to the notion of core syntax. By ‘syntax’ I broadly mean the formal properties of any cognitive domain whose ‘expressions’ may be understood in terms of structure descriptions. Each syntactic system consists of two broad components: some atomic units and a generative device. Except near the end, I will be largely concerned with the generative device of language. Suppose we wish to find out if all domains, including language, share the basic generative device. Call it ‘Core Syntax’ (CS). We study the syntax of language to extract, if at all, some idea of CS. Once we locate CS, we are in a position to ask, how CS originated. It is obvious that simpler the conception of linguistic syntax, the sharper the quest for CS and its origin.

Survival of the wittiest (not friendliest)

Ljiljana Progovac

Research on language evolution has largely neglected the artistic dimension of language, including eloquence and wittiness, and yet the fitness in humans has been found to be correlated with linguistic prowess, and human mate choice even today is often influenced by displays of cognitive abilities through the creatives use of language. My argument is that selection for quick-wittedness (“using words in a clever and funny way”), specific to language and unique to humans, needs to be added to the complex picture of human evolution, relevant from the earliest stages of language. Wittiness is that kind of trait which allows competition (by ‘outwitting’ others) while at the same time favoring “friendliness” in the sense that it provides an excellent platform for replacing physical aggression with verbal behavior. There are several previous findings, both theoretical and experimental, that have paved the way toward the view of human evolution as the “survival of the wittiest,” offering better explanatory power than the “survival of the friendliest”.

Copying hierarchical structure: The human-ape difference maker as regards both language and cumulative culture?

Ron Planer

Humans and other great apes differ with respect to their communicative capacities. Nowhere is this clearer than in the domain of syntax. Likewise, humans and other great apes differ with respect to their cultural learning capacities. Nowhere is this clearer than in the domain of know-how transmission. While it is possible that these two differences are unrelated, this talk pursues the idea that they are a dual manifestation of a common underlying cause. More specifically, I explain how both might be the result of a differential ability for copying hierarchical structure. I call this possibility the “hierarchical copying hypothesis.” This hypothesis is attractive both for its simplicity and for its unificationist character. I summarize the evidence that currently exists for this hypothesis, and then suggest some specific lines of future research that will allow us to better evaluate the hypothesis.

Hierarchical cognition and the evolution of syntax

Ross Pain

Abstract: There are a diverse range of theories attempting to explain how the capacities underwriting syntax evolved. One important issue concerns the role of hierarchical cognition. Some accept that syntactical processing requires hierarchical cognition, but argue that the later cannot be given a Darwinian explanation. They hence favour saltationist accounts (e.g. Berwick & Chomsky, 2016). Others reject saltationism, but accept that the evolution of hierarchical cognition cannot be Darwinian. They hence provide sequential accounts of syntactical processing (e.g. Frank et al., 2012). Others attempt to give a Darwinian account of the evolution of hierarchical cognition in the primate lineage (Planer & Sterelny, 2021). In this paper, I sketch a different view. The account is based on Karl Lashley’s influential 1951 paper (Lashley, 1951), and draws on the claim that hierarchical processing is required to explain even the simplest actions, and is hence a ubiquitous feature of cognition. Adopting this view changes the way we think about the evolution of syntax. The key question becomes: why did a phylogenetically widespread trait become so sophisticated in our lineage? I argue this is a more tractable challenge than that facing rival views, which must either: (i) explain the evolution of hierarchical cognition in our lineage alone; or (ii) show that the production of syntax does not require hierarchical processing.

The role of interactive learning in the evolution of syntax in birdsong and language

Olga Feher

Juvenile songbirds learn species-specific vocalisation from adult individuals during a limited sensitive period in early life. Birdsong is also hierarchically structured, culturally transmitted and exhibits social and geographical variation. These characteristics, along with other important parallels to language, make birdsong a popular animal model for human linguistic behaviour. Birdsong models are especially useful for studying how innate cognitive and social-cultural mechanisms combine to shape vocal communication systems. Songbirds and humans both show innate biases for certain orders of constituent elements. These favoured patterns emerge in learners who are trained on songs or languages that do not initially exhibit them. Learning plays an essential role in the evolution of any linguistic feature, because only those that are learnable can survive multi-generational transmission. However, although birdsong and language are socially learned, interaction during learning is a relatively under-researched area. I will discuss experimental evidence in both songbirds and humans for the important role of interactive learning in the emergence of species-typical patterns in component ordering.

What (dis)similarities between language and action tell us about the origins of syntax

Cas Coopmans

A frequently expressed view in cognitive science is that language and action rely on cognitive and neural resources that are partially shared. An important argument for this idea of cross-domain convergence is that the two systems are structurally analogous: similar to the hierarchical structure underlying sentences, action sequences are thought to be governed by hierarchically organized action plans. Despite supporting evidence from different cognitive science, neuropsychology, and neuroscience, a formal characterization of the structural properties of actions is lacking. I will attempt to fill this gap and evaluate whether language and action are indeed structurally analogous. Depending on the outcome of this formal comparison, the results have important implications for the origins of syntax, which might have precursors in the ability to represent and perform complex actions. I will end by exploring whether language and action additionally exhibit functional similarities.

Syntax evolution: Insights from chimpanzee vocal communication

Simon W. Townsend

The capacity for syntax (i.e. the ability to combine meaning-bearing units together into larger meaningful structures) is a key defining feature of the human species. The evolutionary roots of syntax (how old is syntax?, where did it come from?) are, however, less clear. One powerful way to begin to answer this question is to investigate the communicative abilities of our closest-living relatives, the primates to work out what aspects of syntax are shared with our primate cousins and what are potentially derived and thus unique to humans. In this talk I will review recent observational and experimental work we have conducted investigating the presence of proto-forms of syntax in the communication system of wild chimpanzees with whom we shared a common ancestor approximately 6 million years ago. Together this work is beginning to suggest that the cognitive building blocks underlying syntax might be much older and likely evolved prior to the emergence of language itself.


Dr Richard MooreLink opens in a new window

Dr Giulia PalazzoloLink opens in a new window


To register email giulia.palazzolo.1 [at] Please specify whether you would like to attend remotely or in person.