This article examines the claim that the recursive operation Merge underlies the generative structure of the natural number system. I argue that this claim rests on a conflation between recursion as a property of syntactic representation and recursion as a property of numerical computation. In syntax, repeated applications of Merge yield hierarchically structured expressions; in arithmetic, the successor function yields successive values. These are not the same kind of operation. Focusing on recent proposals by Mendívil-Giró (2025) and Watanabe (2017), I show that Merge, whether external or internal, does not by itself derive numerical succession, but only structured symbolic objects whose interpretation must be independently determined. I further argue that the principal assumptions needed to sustain a Merge-based theory of natural number (innate numerical generativity, hierarchical structure in the count list, and a primitive lexical item corresponding to 1) lack independent empirical support. I conclude that number generativity is better understood as an emergent property of compositional symbolic structure than as the direct output of a successor-like operation implemented by narrow syntax.
This article examines the claim that the recursive operation Merge underlies the generative structure of the natural number system. I argue that this claim rests on a conflation between recursion as a property of syntactic representation and recursion as a property of numerical computation. In syntax, repeated applications of Merge yield hierarchically structured expressions; in arithmetic, the successor function yields successive values. These are not the same kind of operation. Focusing on recent proposals by Mend & iacute;vil-Gir & oacute; (2025) and Watanabe (2017), I show that Merge, whether external or internal, does not by itself derive numerical succession, but only structured symbolic objects whose interpretation must be independently determined. I further argue that the principal assumptions needed to sustain a Merge-based theory of natural number (innate numerical generativity, hierarchical structure in the count list, and a primitive lexical item corresponding to 1) lack independent empirical support. I conclude that number generativity is better understood as an emergent property of compositional symbolic structure than as the direct output of a successor-like operation implemented by narrow syntax.
A central research question in the history of generative grammar has been whether natural languages fall within the class of context-free languages or occupy a higher position in Chomsky's hierarchy for formal grammars. This research regained attention in recent years because grammatical abilities of non-human species have been investigated to evaluate the (alleged) uniqueness of human language. In this context, identifying the locus of complex grammatical abilities is particularly important. Out of the four papers that proved human language to be context-sensitive, three built their demonstration on clearly syntactic constructions. A fourth paper (Culy, 1982) focused on Bambara and claimed that this language is context-sensitive in its morphology. In this note we systematically investigate whether context-sensitivity indeed lies in the morphology in Bambara, conclude for a positive answer and discuss this finding at the light of the debate between lexicalist and non-lexicalist approaches.
Bernd Bierl eloquently proposes an ambitious synthesis towards biological linguistics. This synthesis aims to establish a new bio-ethological linguistics based on 4E cognition. It considers the foundational ethology of authors like Konrad Lorenz and Nikolaas Tinbergen to be its biological basis. Rather than developing the biolinguistics inspired by Noam Chomsky, which also has its roots in early 20th-century European ethology, the author considers it an inadequate naturalization and suggests subsuming it under a conception of language as behavior, within the context of embodied, embedded, enacted, and extended (4E) cognition. This reply presents some critical observations on this proposal, highlighting the inadequate interpretation of the biolinguistic approach and the potential shortcomings of the proposed research program.
This article proposes a theoretical framework for a Bio-Ethological 4E Linguistics that integrates ethology, biolinguistics, and embodied cognition within a single biological continuum. It argues that the persistent division between linguistic internalism and behavioral biology has produced two incomplete naturalisms-one mental without life, the other biological without mind. By aligning Tinbergen's four questions of ethology (mechanism, ontogeny, function, evolution) with the four dimensions of 4E cognition (embodiment, enaction, embeddedness, extension), the paper reconstructs language as an evolved form of biological sense-making rather than an abstract code. Classical ethology-represented by Tinbergen, Lorenz, Hinde, and Hess-anticipated many principles later formalized in enactive and embodied theories of mind, while contemporary 4E approaches have yet to ground their concepts in the empirical study of behavior. The proposed synthesis restores that missing continuity by treating linguistic interaction as a living process of regulation within ecological and social systems. Language, on this view, functions as an adaptive interface linking individual cognition to collective life. The article concludes with a programmatic agenda for a unified science of communication that spans neurobiological mechanisms, developmental dynamics, social coordination, and cultural evolution. A Bio-Ethological 4E Linguistics thus redefines language as life expressing itself through meaning-an approach that rejoins the study of mind with the study of behavior and situates linguistics within the broader biology of living systems.
This paper draws a deep connection between smuggling (Collins, 2005) and labeling (Collins, 2002; Chomsky, 2013, 2015), showing that the movement of the smuggler in a smuggling derivation can be triggered by the labeling algorithm.
The idea that mathematical ability and language are related in human cognition is an old one. It is commonly assumed in the philosophical tradition, in psychology and in cognitive science, generally implying that knowledge of numbers is indebted to knowledge of language. In this contribution I suggest a more specific model of the relationship between knowledge of numbers and knowledge of language in the light of developments in the neuroscience of numerical understanding and in linguistic theory. The proposed model places the evolutionary development of the syntactic component of the Faculty of Language at the basis of the transition from the innate sense of number that we share with other animals to the sophisticated mathematical ability of our species.
A core component of a successful artificial general intelligence would be the rapid creation and manipulation of grounded compositional abstractions and the demonstration of expertise in the family of recursive hierarchical syntactic objects necessary for the creative use of human language. We evaluated the recently released o3 model (OpenAI; o3-mini-high) from ChatGPT and discovered that while it succeeds on some basic linguistic tests relying on linear, surface statistics (e.g., the Strawberry Test), it fails to generalize basic phrase structure rules; it fails with comparative sentences involving semantically illegal cardinality comparisons ('Escher sentences'); it fails to correctly rate and explain acceptability dynamics; and it fails to distinguish between instructions to generate unacceptable semantic vs. unacceptable syntactic outputs. When tasked with generating simple violations of grammatical rules, it is seemingly incapable of representing multiple parses to evaluate against various possible semantic interpretations. We ran all of these prompts multiple times again through the API and provide basic accuracy scores. In stark contrast to many recent claims that artificial language models are on the verge of replacing the field of linguistics, our results suggest not only that deep learning is hitting a wall with respect to compositionality (Marcus, 2022), but that it is hitting [a [stubbornly [resilient wall]]] that cannot readily be surmounted to reach human-like compositional reasoning simply through more compute.
In this article, we summarise and critically evaluate Ryan Nefdt’s The Philosophy of Theoretical Linguistics: A Contemporary Outlook (2024). In this book, Nefdt brings the tools of philosophy to bear on contemporary linguistics, targeting perennial debates in syntax, semantics, pragmatics, phonology, and the evolution of language. In so doing, Nefdt sketches several tantalising paths for progress on these topics. Although some of Nefdt’s arguments are underdeveloped, this book serves as a worthwhile introduction to the philosophy of linguistics.
Some recent influential work in the Minimalist Program takes the position that Merge, the core language ability to recursively combine two elements together, is free. However, if Merge were completely free, there would be an infinite number of possible derivations for every utterance. Thus, Merge must be constrained in some way. In this paper, I describe a computer model of language that implements a limited form of Merge that is free. I attempt to demonstrate that, within the confines of the language module, Labeling is generally sufficient to constrain Free Merge, and I discuss issues that arise regarding overgeneration of syntactic structures given Free Merge.
Contrary to the prevailing assumption that language is “primarily a tool for communication rather than thought”, I argue that language is, to invoke Oscar Wilde, “quite useless”. Arguing from aesthetic philosophy and the minimalist program for linguistic theory, I conject that language, like art, is not “for” anything—it simply is, conforming to aesthetic rather than utilitarian principles. Of course, like art, language can be a powerful instrument of communication, but its function is not that of expressing thought; it creates thoughts, “primarily” for communicating with oneself, engaging in Popperian critical rationalism, making thoughts (e.g., sentences, constructive proofs) to match Platonic objects (e.g., propositions, classical proofs).
Descartes famously constructed a language test to determine the existence of other minds. The test made critical observations about how humans use language that purportedly distinguishes them from animals and machines. These observations were carried into the generative (and later biolinguistic) enterprise under what Chomsky in his Cartesian Linguistics, terms the “creative aspect of language use” (CALU). CALU refers to the stimulus-free, unbounded, yet appropriate use of language—a tripartite depiction whose function in biolinguistics is to highlight a species-specific form of intellectual freedom. This paper argues that CALU provides a set of facts that have significant downstream effects on explanatory theory-construction. These include the internalist orientation of linguistics, the invocation of a competence-performance distinction, and the postulation of a generative language faculty that makes possible—but does not explain—CALU. It contrasts the biolinguistic approach to CALU with the recent wave of enthusiasm for the use of Transformer-based Large Language Models (LLMs) as tools, models, or theories of human language, arguing that such uses neglect these fundamental insights to their detriment. It argues that, in the absence of replication, identification, or accounting of CALU, LLMs do not match the explanatory depth of the biolinguistic framework, thereby limiting their theoretical usefulness.
A growing number of studies indicate that the hippocampus plays an essential role in language processing as well as episodic memory. However, there is no consensus on how it is engaged in both domains and how it connects the two domains. From a theoretical perspective, this paper delves into the intricate relationship between episodic memory and the narrow syntax of human language. To be more specific, I focus on the functions of the hippocampus in event processing and propose that hippocampal phasic activity supports the cyclic interaction between episodic memory and narrow syntax. Through such cyclic interaction, the event in the episodic memory is assigned a linguistic format that can be communicated, while narrow syntax is provided with an interpretative engine that underlies reference. This hypothesis predicts that when episodic memory is impaired, corresponding abnormalities appear in linguistic reference.
This paper examines Chomsky’s recently proposed and abandoned FormSequence operation and presents a middle-ground implementation of it in a way that conforms to the Strong Minimalist Thesis. Special attention is paid to the role of Hilbert’s epsilon (ϵ) operator in this operation. I argue that while the ϵ-operator can give FormSequence its desired effect, the sequence-choosing mechanism should more adequately be attributed to the cognitive-computational context (mainly the interfaces) instead of Narrow Syntax. In other words, FormSequence is not entirely syntactic in nature but only partly so. I implement its syntactic part as repeated Pair Merge of a coordinator with a number of conjuncts, which yields a partially ordered set as output instead of a sequence. This implementation reconciles FormSequence with the Strong Minimalist Thesis and maintains a purely hierarchical syntactic module of human language. Furthermore, I compare the use of the ϵ-operator in FormSequence and its more established use in formal semantics and eventually promote a domain-general perspective on the fundamental cognitive procedure of sequence formation.
This review first provides a summary of the central ideas in Merge and the Strong Minimalist Thesis and then presents a discussion of the more controversial points. The book offers an introduction to the Minimalist Program. The focus is on Merge, which plays a central role in the Faculty of Language because it is “the primary structure-building device of the syntax” (p. 2). The book clarifies the status of Theta Theory, Search, and Workspace, and provides a novel account of passives and obligatory control.
Recent literature argues that a purportedly long-standing theory—so-called “laryngeal descent theory”—in speech evolution has been refuted (Boë et al., 2019, https://doi.org/10.1126/sciadv.aaw3916). However, an investigation into the relevant source material reveals that the theory described has never been a prominent line of thinking in speech-centric sciences. The confusion arises from a fundamental misunderstanding: the argument that the descent of the larynx and the accompanying changes in the hominin vocal tract expanded the range of possible speech sounds for human ancestors (a theory that enjoys wide interdisciplinary support) is mistakenly interpreted as a belief that all speech was impossible without such changes—a notion that was never widely endorsed in relevant literature. This work aims not to stir controversy but to highlight important historical context in the study of speech evolution.
We identify and analyze three caveats that may arise when analyzing the linguistic abilities of Large Language Models. The problem of unlicensed generalizations refers to the danger of interpreting performance in one task as predictive of the models’ overall capabilities, based on the assumption that because a specific task performance is indicative of certain underlying capabilities in humans, the same association holds for models. The human-like paradox refers to the problem of lacking human comparisons, while at the same time attributing human-like abilities to the models. Last, the problem of double standards refers to the use of tasks and methodologies that either cannot be applied to humans or they are evaluated differently in models vs. humans. While we recognize the impressive linguistic abilities of LLMs, we conclude that specific claims about the models’ human-likeness in the grammatical domain are premature.
