Human language evolution: a view from theoretical linguistics on how syntax and the lexicon first came into being

Abstract

Human language is a multi-componential function comprising several sub-functions each of which may have evolved in other species independently of language. Among them, two sub-functions, or modules, have been claimed to be truly unique to the humans, namely hierarchical syntax (known as "Merge" in linguistics) and the "lexicon." This kind of species-specificity stands as a hindrance to our natural understanding of human language evolution. Here we challenge this issue and advance our hypotheses on how human syntax and lexicon may have evolved from pre-existing cognitive capacities in our ancestors and other species including but not limited to nonhuman primates. Specifically, we argue that Merge evolved from motor action planning, and that the human lexicon with the distinction between lexical and functional categories evolved from its predecessors found in animal cognition through a process we call "disintegration." We build our arguments on recent developments in generative grammar but crucially depart from some of its core ideas by borrowing insights from other relevant disciplines. Most importantly, we maintain that every sub-function of human language keeps evolutionary continuity with other species' cognitive capacities and reject a saltational emergence of language in favor of its gradual evolution. By doing so, we aim to offer a firm theoretical background on which a promising scenario of language evolution can be constructed.

Keywords: Disintegration; Functional categories; Language evolution; Lexicon; Merge; Motor control origin.

Fig. 1

Basic architecture of the human language faculty. The lexicon provides inputs to the syntactic computational system (syntax), which combines these lexical items into a hierarchical phrase structure. This structure is then transferred to the two interpretive systems, the conceptual-intentional (CI) system for semantic interpretation and the sensorimotor (SM) system for phonological interpretation, including signs and other forms of surface realizations, via the two interfaces which roughly correspond to Logical Form (LF) and Phonetic Form (PF) in earlier versions of generative grammar. The syntax-CI connection is adaptive for internalization that takes place within an individual (thought, inference, planning, etc.), while the syntax-SM connection is for externalization (communication with other individuals)

Fig. 2

Motor control origin of Merge. Linguistic Merge evolved from action Merge (action grammar). Starting from the Pairing strategy of action Merge, motor action and linguistic syntax evolved into more and more complex forms. Proto-Merge characterizes animal communication systems (metaphorically, animal “language”), where at most only two (sets of) signals are combined (Miyagawa and Clarke 2019), as in the pyow-hack sequence of putty-nosed monkeys (Schlenker et al. 2016a) and the ABC-D call of the Japanese tit (Suzuki et al. 2018). Protolanguage had at best Pot-Merge, given the plausible assumption that it had only linear syntax (Jackendoff and Wittenberg 2016) and hierarchical syntax was yet to come. With the advent of Sub-Merge, exapted from Subassembly strategy of action Merge, human language with all its structural properties emerged

Fig. 3

From action Merge evolved a domain-general combinatorial operation (generic Merge), and from the latter derive a variety of domain-specific operations including linguistic Merge. The formation of generic Merge corresponds to what Hauser (2009) refers to as “the release of recursion from its motor prison to other domains of thought.” A possibility is that only this generic Merge is our innate capacity, and domain-specific operations develop through ontogeny, much in line with Marcus’ (2006) descent-with-modification version of modularity. If the minimalist thinking is correct that UG contains only Merge and nothing else, it will follow that there is no UG

Fig. 4

The structure of grammar in Distributed Morphology (adapted from Fujita ; cf. Embick and Noyer ; Marantz 1997). The elements of the narrow lexicon are used by syntax (Merge) to generate hierarchical structure. On the morphophonological side, this structure is mapped to the SM system via the SM interface (PF) based on the information of the vocabulary. On the conceptual-semantic side, it is mapped to the CI system via the CI interface (LF) by referring to the encyclopedia

Fig. 5

Layers of grammatical development (adapted from Heine and Kuteva ; see also Heine and Kuteva 2012). AGR stands for agreement marker, ASP for (verbal) aspect, CAS for case marker, CPL for complementizer, DEF for marker of definiteness (definite article), NEG for negation marker, PAS for passive marker, REL for relative clause marker, SBR for subordinating marker of adverbial clauses, TNS for tense marker. The dotted arrow indicates an indirect relation