Elamur Research⌕
☰
home / papers / 2026 / Metric Grammars: A usage-based grammatical formalism that supports generation, parsing and morphological innovation

Metric Grammars: A usage-based grammatical formalism that supports generation, parsing and morphological innovation

Summary

The paper (full title: "Stretched Tree Metric Grammars…") addresses what the authors call the Life ↔ Grammar question: how individual acts of language use can reshape a grammatical system. Discrete formalisms such as CFGs, HPSG, LFG, Minimalism or Construction Grammar assign structures to categories with no intermediate distances, so they struggle to capture gradual change. The authors propose Stretched Tree Metric Grammars (STMGs), which place grammar in a continuous metric space. A backpropagation-trained neural network encodes syntax and semantics in a single space. Grammaticality is a graded "harmony" value, and expressions above an observability threshold are predicted to occur. Every usage event slightly reshapes this harmony landscape, so forms just below the threshold can be pushed into acceptability. In a toy English simulation, the model generates and parses simple sentences. Adding a growing share of a novel denominal verb to training raised its harmony to the level of attested forms. Comparable cross-category novelties, such as denominal determiners or modals, stayed below that level. The work offers a formal, more interpretable cousin of LLMs for modelling usage-driven innovation.

Key points

  • STMGs are one subtype of "Metric Grammars," a class of dynamical systems defined over complete metric spaces whose job is to recognize or generate sequences of symbols. LLMs also belong to this broad class.
  • Syntactic differences correspond to large distances in encoding space, while semantic differences are smaller shifts within those regions. Many accumulated small changes can therefore add up to a syntactic shift.
  • In the experiment, the proportion of novel denominal-verb forms in training rose from 0% to 8%. Results average 5 runs, with 10 test items per condition.
  • The authors suggest the innovation succeeds partly by analogy to an existing denominal verb ("dog") in the training grammar. They note this explanation is still speculative.
  • Acknowledged limitations:
    • The simulation is a toy and has not yet been scaled up.
    • Introducing the novel verb seemed to lower the harmony of existing constructions.
    • Robustness has not been tested.
  • Initial parameterization can come from a PCFG or from any labelled context-free parsed corpus.
  • Proposed future work:
    • Testing predictions against human judgments.
    • Modelling grammaticalization.
    • Checking whether SLMs show similar innovation tendencies.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.conll-main.46

@inproceedings{tabor2026metric,
  title = {{Metric Grammars: A usage-based grammatical formalism that supports generation, parsing and morphological innovation}},
  author = {Tabor, Whitney and Lee, Hyosun},
  booktitle = {Proceedings of the 30th Conference on Computational Natural Language Learning},
  year = {2026},
  doi = {10.18653/v1/2026.conll-main.46},
  url = {https://doi.org/10.18653/v1/2026.conll-main.46}
}

References

58 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Nicolas Asher, Tim Van de Cruys, Antoine Bride, and and Márta Abrusán. 2017. Integrating Type Theory and Distributional Semantics: A Case Study on Adjective–Noun Compositions. Computational Linguistics, 42(4):703–725. find
  2. Michael Barnsley. 1988. Fractals Everywhere. Academic Press, Boston. find
  3. Douglas Bates, Martin Mächler, Ben Bolker, and Steve Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67(1):1– 48. find our card
  4. Katrien Beuls and Remi Van Trijp. 2016. Computational construction grammar and constructional change. Belgian Journal of Linguistics, 30:1–13. find
  5. Ina Bornkessel-Schlesewsky, Dietmar Roehm, Robert Mailhammer, and Matthias Schlesewsky. 2020. Language Processing as a Precursor to Language Change: Evidence From Icelandic. Frontiers in Psychology, 10. find
  6. Lieselotte Brems. 2003. Measure Noun constructions: An instance of semantically-driven grammaticalization. International Journal of Corpus Linguistics, 8(2):283–312. find
  7. Bastian Bunzeck and Sina Zarrieß. 2024. Fifty shapes of BLiMP: syntactic learning curves in language models are not uniform, but sometimes unruly. In Proceedings of the 2024 CLASP Conference on Multimodality and Interaction in Language Learning, pages 39–55. Association for Computational Linguistics… find
  8. Joan Bybee, Revere Perkins, and William Pagliuca. 1994. The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World. University of Chicago Press, Chicago, IL. find
  9. Joan L. Bybee. 1985. Morphology: A Study of the Relation between Meaning and From. John Benjamins, Amsterdam. find
  10. Pyeong Whan Cho, Matthew Goldrick, and Paul Smolensky. 2020. Parallel parsing in a gradient symbolic computation parser. find
  11. Noam Chomsky. 1986. Knowledge of language : its nature, origin, and use. New York : Praeger. find
  12. Charles J. Fillmore, Paul Kay, and Mary Catherine O’Connor. 1988. Regularity and Idiomaticity in Grammatical Constructions: The Case of Let Alone. Language, 64(3):501–538. find
  13. Adele E. Goldberg. 2006. Constructions at Work. Oxford University Press, Oxford. find
  14. Carolyn Gottfurcht. 2008. Denominal verb formation in English. PhD, Northwestern University. find
  15. Stevan Harnad. 1990. The symbol grounding problem. Physica D: Nonlinear Phenomena, 42(1):335–346. find
  16. Bernd Heine, Ulrike Claudi, and Friederike Hun-nemeyer. 1991. Grammaticalization: A Conceptual Framework. University of Chicago Press, Chicago. find
  17. John E. Hopcroft and Jeffrey D. Ullman. 1979. Introduction to Automata Theory, Languages, and Computation. Addison-Wesley, Menlo Park, California. find
  18. Paul J. Hopper. 1991. On some principles of grammaticization. In Elizabeth Closs Traugott and Bernd Heine, editors, Approaches to Grammaticalization, v. 1, pages 17–36. John Benjamins. find
  19. Paul J. Hopper and Elizabeth Closs Traugott. 2003. Grammaticalization, 2nd Edition. Cambridge University Press, Cambridge, England. find
  20. Frank Keller. 2000. Gradience in grammar: Experimental and computational aspects of degrees of grammaticality. find
  21. Gerard Kempen and T. Vosse. 1989. Incremental syntactic tree formation in human sentence processing: a cognitive architecture based on activation decay and simulated annealing. Connection Science, 1:273– 290. find
  22. Alexandra Kuznetsova, Per B. Brockhoff, and Rune H. B. Christensen. 2017. lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software, 82(13):1–26. find
  23. G Lakoff. 1977. Linguistic Gestalts. Gestalts, 13:236–287. find
  24. Linguistic Ronal W. Langacker. 1988. A Usage-Based Model. In Brygida Rudzka-Ostyn, editor, Topics in Cognitive Linguistics, pages 127–164. John Benjamins. find
  25. Noam Chomsky. 1995. The minimalist program. MIT Press, Cambridge, MA. find
  26. B. Coecke, M. Sadrzadeh, and S. Clark. 2010. Mathematical foundations for a compositional distributional model of meaning. Linguistic Analysis, 36:345–384. find
  27. Ronald Langacker. 1977. Syntactic Reanalysis. In Charles N. Li, editor, Mechanisms of Syntactic Change, pages 57–139. University of Texas Press. find
  28. Hendrik De Smet. 2012. The Course of Actualization. Language, 88(3):601–633. ERIC Number: EJ986712. find
  29. Ronald W. Langacker. 1987. Foundations of Cognitive Grammar, v. 1. Stanford University Press, Stanford, California. find
  30. Jeffrey L. Elman. 1991. Distributed Representations, Simple Recurrent Networks, and Grammatical Structure. Machine Learning, 7:195–225. find
  31. Russell V. Lenth. 2025. emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.11.1. find
  32. Frantisek Lichtenberk. 1991. On the gradualness of grammaticalization. In Elizabeth Closs Traugott and Bernd Heine, editors, Approaches to Grammaticalization, v. 1, pages 37–80. John Benjamins. find
  33. Kanishka Misra and Kyle Mahowald. 2024. Language Models Learn Rare Phenomena from Less Rare Phenomena: The Case of the Missing AANNs. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 913–929, Miami, Florida, USA. Association for Computational Linguisti… find
  34. Richard Montague. 1970. English as a formal language. In B. Visentini et al, editor, Linguaggi nella e nella Tecnica, pages 189–224. Edizioni di Comunità. find
  35. Michael Spivey. 2007. The Continuity of Mind. Oxford University Press, New York. find
  36. Luke Steels, editor. 2011. Design Patterns in Fluid Construction Grammar. Constructional Approaches to Language. John Benjamins Publishing Company. find
  37. Whitney Tabor. 2000. Fractal encoding of context-free grammars in connectionist networks. Expert Systems: The International Journal of Knowledge Engineering and Neural Networks, 17(1):41–56. find
  38. Whitney Tabor. 2021. On the Relationship between Syntactic and Semantic Encoding in Metric Space Language Models. Journal of Cognitive Science, 22(2):35–67. find
  39. Cris Moore. 1998. Dynamical Recognizers: Real-time Language Recognition by Analog Computers. Theoretical Computer Science, 201:99–136. find
  40. Whitney Tabor, Bruno Galantucci, and Daniel Richardson. 2004. Effects of merely local syntactic coherence on sentence processing. Journal of Memory and Language, 50:355–370. find
  41. Jean Petitot. 2011. Cognitive Morphodynamics: Dynamical Morphological Models of Constituency in Perception and Syntax. Peter Lang AG, Internationaler Verlag der Wissenschaften, Bern, Suisse. find
  42. Whitney Tabor and Sean Hutchins. 2004. Evidence for Self-Organized Sentence Processing: Digging In Effects. Journal of Experimental Psychology: Learning, Memory, and Cognition, 30(2):431–450. find
  43. Carl Pollard and Ivan A. Sag. 1994. Head-Driven Phrase Structure Grammar. The University of Chicago Press, Chicago. find
  44. Whitney Tabor and Hyosun Lee. 2024. Metric Grammars. Poster presented at the 2024 Conference of the Cognitive Science Society. find
  45. Ian Roberts. 1993. A Formal Account of Grammaticalisation in the History of Romance Futures. Folia Linguistica Historica, XIII/1–2:219–58. find
  46. Elizabeth Closs Traugott and Graeme Trousdale. 2016. Constructionalization and Constructional Changes. Oxford Studies in Diachronic and Historical Linguistics. Oxford University Press, Oxford, New York. find
  47. David E. Rumelhart, James L. McClelland, and the PDP Research Group. 1986. Parallel Distriubted Processing. Explorations in the Microstructure of Cognition. Volume 1. MIT Press, Cambridge, Massachusetts. find
  48. Joanna R aczaszek-Leonardi and Terrence W. Deacon. 2018. Ungrounding symbols in language development: implications for modeling emergent symbolic communication in artificial systems. In 2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob), p… find
  49. Ferdinand de Saussure. 1916. Course in General Linguistics. Columbia University Press. Pages: 336 Pages. find
  50. Hans-Jörg Schmid. 2020. The Dynamics of the Linguistic System: Usage, Conventionalization, and Entrenchment. Oxford University Press, Oxford, New York. find
  51. Garrett Smith and Whitney Tabor. 2018. Toward a Theory of Timing Effects in Self Organized Sentence Processing. In Proceedings of the 16th international conference on cognitive modeling, pages 138–143. find
  52. Paul Smolensky and Geraldine Legendre. 2006. The Harmonic Mind: from neural computation to optimality-theoretic grammar. MIT Press, Cambridge, MA. find
  53. Paul Van Eecke and Katrien Beuls. 2018. Exploring the Creative Potential of Computational Construction Grammar. Zeitschrift für Anglistik und Amerikanistik, 66(3):341–355. find
  54. Frank van der Velde and Marc de Kamps. 2006. Neural Blackboard Architectures of Combinatorial Structures in Cognition. Behavioral and Brain Sciences, 29:37–108. find
  55. Lara Verheyen, Jonas Doumen, Paul Van Eecke, and Katrien Beuls. 2025. You Shall Know a Construction by the Company it Keeps: Computational Construction Grammar with Embeddings. In Proceedings of the Second International Workshop on Construction Grammars and NLP, pages 75–83. find
  56. Sandra Villata and Whitney Tabor. 2022. A self-organized sentence processing theory of gradience: the case of islands. Cognition, page 104943. find
  57. Theo Vosse and Gerard Kempen. 2000. Syntactic structure assembly in human parsing: a computational model based on competitive inhibition and a lexicalist grammar. Cognition, 75:105–143. find
  58. Uriel Weinreich, William Labov, and Marvin Herzog. 1968. Empirical Foundations for a Theory of Language Change. In W. P. Lehmann and Yakov Malkiel, editors, Directions for Historical Linguistics, pages 95–188. University of Texas Press. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/tabor-2026-metric-grammars-usage-based-grammatical/