Elamur Research⌕
☰
home / papers / 2026 / Frequency modulates structural choice in Turkish suspended affixation: a latent-process account

Frequency modulates structural choice in Turkish suspended affixation: a latent-process account

Summary

The paper examines why derivational suspended affixation (SA) in Turkish is described as highly restricted, yet speakers accept some instances. In SA, one suffix on the last conjunct applies to every coordinated element. The authors propose that SA is structurally ambiguous between two parses:

  • Base generation: the affix attaches to the whole coordination.
  • Ellipsis: the affix sits on each conjunct and is deleted on the non-final ones.

They argue that derivational suffixes are too tightly bound to the root to be deleted, so they require the base-generation parse. Frequent coordinations tend to be stored holistically, which makes that parse more likely.

To test this, they ran an acceptability rating study on four derivational suffixes in SA and full forms, varying how frequent the coordinations were. They analysed the ratings with ordinal regression and a Multinomial Processing Tree (MPT) model, which separates hidden parse choices from the observed responses. Frequency raised the probability of base generation for -sIz and -cI, but not for -lI or -lIk. The authors conclude that frequency influences which structure the parser builds, rather than simply making familiar strings sound better.

Key points

  • Suffixes tested: -sIz ('-less'), -cI ('-maker'), -lI ('-having') and -lIk ('-for'). The model used Bayesian estimation, with 89% credible intervals reported.
  • Base-generation probability at mean frequency: 0.30 for -sIz and 0.38 for -cI (ellipsis dominates), 0.50 for -lI (fully uncertain), and 0.60 for -lIk (base generation modestly dominant).
  • Effect of frequency on base generation:
    • Posterior probability of a positive effect was 0.83 for -sIz and 0.80 for -cI, which the authors call a suggestive positive effect.
    • For -lI it was only 0.12, a suggestive negative trend.
    • For -lIk (0.62) the direction was uncertain.
  • Value of the latent model: for -cI, the MPT detected an influence of frequency on which structure was chosen, an influence that did not appear in the ordinal surface ratings. The authors take this as evidence that a latent-process model is needed.
  • Proposed follow-up: a forced-choice interpretation task. Under base generation the suffix should take joint, wide scope over the coordination, while under ellipsis it should apply to each conjunct separately.

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.scil-main.32

@inproceedings{turk2026frequency,
  title = {{Frequency modulates structural choice in Turkish suspended affixation: a latent-process account}},
  author = {Turk, Utku and Neu, Eva and Bakay, Özge and Dillon, Brian and Jarosz, Gaja},
  booktitle = {Proceedings of the Society for Computation in Linguistics 2026},
  year = {2026},
  doi = {10.18653/v1/2026.scil-main.32},
  url = {https://doi.org/10.18653/v1/2026.scil-main.32}
}

References

46 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. tree modeling. Psychonomic Bulletin & Review, 6:57–86. find
  2. Joan Bresnan. 2007. Is syntactic knowledge probabilistic? Experiments with the English dative alternation. In Sam Featherston and Wolfgang Sternefeld, editors, Roots: Linguistics in Search of Its Evidential Base, pages 77–96. Mouton de Gruyter, Berlin. find
  3. Joan Bresnan, Anna Cueni, Tatiana Nikitina, and R. Harald Baayen. 2004. Predicting the dative alternation. In Gerlof Bouma, Irene Kraemer, and Joost Zwarts, editors, Cognitive Foundations of Interpretation, pages 69–94. Royal Netherlands Academy of Arts and Sciences, Amsterdam. find
  4. Joan Bresnan and Marilyn Ford. 2010. Predicting syntax: Processing dative constructions in American and Australian varieties of English. Language, 86:186– 213. find
  5. George Aaron Broadwell. 2008. Turkish suspended affixation is lexical sharing. In Miriam Butt and Tracy Holloway King, editors, Proceedings of the LFG08 Conference, pages 198–213. CSLI Publications, Stanford. find
  6. Paul-Christian Bürkner. 2017. brms: An R package for Bayesian multilevel models using Stan. Journal of Statistical Software, 80:1–28. find
  7. Bob Carpenter, Andrew Gelman, Matthew D. Hoffman, Daniel Lee, Ben Goodrich, Michael Betancourt, Marcus A. Brubaker, Jiqiang Guo, Peter Li, and Allen Riddell. 2017. Stan: A probabilistic programming language. Journal of Statistical Software, 76:1–32. find
  8. Gareth Carroll and Kathy Conklin. 2020. Is all formulaic language created equal? Unpacking the processing advantage for different types of formulaic sequences. Language and Speech, 63:95–122. find
  9. Miloje Despić. 2017. Suspended morphology in Serbian: Clitics vs. affixes. Glossa: a journal of general linguistics, 2:12. find
  10. Faruk Akkuş. 2016. Suspended affixation with derivational suffixes and lexical integrity. In Proceedings of IMOG 2016, Patras. find
  11. Edgar Erdfelder, Tina-Sarah Auer, Benjamin E. Hilbig, Andreas Aßmann, and Morten Moshagen. 2009. Multinomial processing tree models: A review of the literature. Zeitschrift für Psychologie / Journal of Psychology, 217:108–124. find
  12. Inbal Arnon and Neal Snider. 2010. More than words: Frequency effects for multi-word phrases. Journal of Memory and Language, 62:67–82. find
  13. David Erschler. 2012. Suspended affixation in Ossetic and the structure of the syntax-morphology interface. Acta Linguistica Hungarica, 59:153–175. find
  14. Furkan Atmaca. 2020. Suspended Affixation in Turkish. MA Thesis, Boğaziçi University. find
  15. David Erschler. 2018. Suspended affixation as morpheme ellipsis: Evidence from Ossetic alternative questions. Glossa: a journal of general linguistics, 3:12. find
  16. Furkan Atmaca. 2021. Suspended affixation needs no morphological word: The suffix -(y)Ip. In Proceedings of the Workshop on Turkic and Languages in Contact with Turkic, volume 6, pages 5035–5035. find
  17. Zhiyu Mia Gong. 2021. Postsyntactic lowering and linear relations in Dagur noun phrases. Glossa: a journal of general linguistics, 6:42. find
  18. William H. Batchelder and David M. Riefer. 1999. Theoretical and empirical review of multinomial process Ekaterina Gruzdeva. 1998. Nivkh. Languages of the World/Materials. Lincom Europa, Munich. find
  19. Elina Guseva and Philipp Weisser. 2018. Postsyntactic reordering in the Mari nominal domain: Evidence from suspended affixation. Natural Language & Linguistic Theory, 36:1087–1120. find
  20. Morris Halle and Alec Marantz. 1993. Distributed Morphology and the pieces of inflection. In Kenneth Hale and S. Jay Keyser, editors, The View from Building 20, pages 111–176. MIT Press, Cambridge, MA. find
  21. Morris Halle and Alec Marantz. 1994. Some key features of Distributed Morphology. In MIT Working Papers in Linguistics, pages 275–288. find
  22. Michael J Hautus, Neil A Macmillan, and C Douglas Creelman. 2021. Detection theory: A user’s guide. Routledge. find
  23. Daniel Jurafsky. 1996. A probabilistic model of lexical and syntactic access and disambiguation. Cognitive Science, 20:137–194. find
  24. Barış Kabak. 2007. Turkish suspended affixation. Linguistics, 45:311–347. find
  25. Karl Christoph Klauer. 2010. Hierarchical multinomial processing tree models: A latent-trait approach. Psychometrika, 75:70–98. find
  26. Jaklin Kornfilt. 1996. On some copular clitics in Turkish. In ZAS Papers in Linguistics, volume 6, pages 96–114. Zentrum für Allgemeine Sprachwissenschaft, Berlin. find
  27. Jaklin Kornfilt. 2012. Revisiting “suspended affixation” and other coordinate mysteries. In Laura Brugè, Anna Ferro, Carla Ferraro, Anna Ferro, and Lisa Ferraro, editors, Functional Heads: The Cartography of Syntactic Structures, pages 181–196. Oxford UP, Oxford. find
  28. Roger Ratcliff. 1978. A theory of memory retrieval. Psychological Review, 85:59. find
  29. David M. Riefer and William H. Batchelder. 1988. Multinomial modeling and the measurement of cognitive processes. Psychological Review, 95:318–339. find
  30. Carson T. Schütze and Jon Sprouse. 2013. Judgment data, pages 27–50. Cambridge University Press. find
  31. Tuomas Sivula, Måns Magnusson, Alejandro Andres Matamoros, and Aki Vehtari. 2025. Uncertainty in Bayesian leave-one-out cross-validation based model comparison. Bayesian Analysis. find
  32. Anna Siyanova-Chanturia, Kathy Conklin, and Walter J. B. van Heuven. 2011. Seeing a phrase “time and again” matters: The role of phrasal frequency in the processing of multiword sequences. Journal of Experimental Psychology: Learning, Memory, and Cognition, 37:776–784. find
  33. Anna Vogel Sosa and James MacFarlane. 2002. Evidence for frequency-based constituents in the mental lexicon: Collocations involving the word of . Brain and Language, 83:227–236. find
  34. David A. Swinney and Anne Cutler. 1979. The access and processing of idiomatic expressions. Journal of Verbal Learning and Verbal Behavior, 18:523–534. find
  35. Antoine Tremblay, Bruce Derwing, Gary Libben, and Chris Westbury. 2011. Processing advantages of lexical bundles: Evidence from self-paced reading and sentence recall tasks. Language Learning, 61:569– 613. find
  36. Jochen Trommer. 2008. Hungarian has no portmanteau agreement. In Proceedings of the 26th West Coast Conference on Formal Linguistics, pages 478–486, Somerville, MA. Cascadilla Proceedings Project. find
  37. Emily Morgan and Roger Levy. 2015. Modeling idiosyncratic preferences: How generative knowledge and expression frequency jointly determine language structure. Proceedings of the 37th Annual Conference of the Cognitive Science Society, pages 1649–1654. find
  38. Utku Türk. 2025. Controlling morphosyntactic competition through phonology. In Pavel Caha, Karen De Clercq, and Guido Vanden Wyngaerd, editors, Nanosyntax and the Lexicalization Algorithm. Oxford UP, Oxford. find
  39. Emily Morgan and Roger Levy. 2016. Abstract knowledge versus direct experience in processing of binomial expressions. Cognition, 157:384–402. find
  40. Aki Vehtari, Andrew Gelman, and Jonah Gabry. 2017. Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing, 27:1413–1432. find
  41. Bruno Nicenboim, Daniel Schad, and Shravan Vasishth. 2024. An Introduction to Bayesian Data Analysis for Cognitive Science. Cambridge UP. Online version: https://bruno.nicenboim.me/ bayescogsci/. ↗
  42. Dario Paape and Shravan Vasishth. 2022. Estimating the true cost of garden-pathing: A computational model of latent cognitive processes. Cognitive Science, 46:e13186. find
  43. Steven Pinker and Michael T. Ullman. 2002. The past and future of the past tense. Trends in Cognitive Sciences, 6:456–463. find
  44. Bernhard Wälchli. 2005. Co-Compounds and Natural Coordination. Oxford UP, New York, NY. find
  45. James Hye-Suk Yoon and Wooseung Lee. 2005. Conjunction reduction and its consequences for noun phrase morphosyntax in Korean. In Proceedings of the 24th West Coast Conference on Formal Linguistics, pages 379–387, Somerville, MA. Cascadilla Proceedings Project. find
  46. J. Zehr and F. Schwarz. 2018. PennController for Internet Based Experiments (IBEX). find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/turk-2026-frequency-modulates-structural-choice-turkish/