Register here for one or more lectures
Also online: https://meet.google.com/ojw-oshg-fzw
13 November 2023. Jenny Audring, Leiden University.
The importance of being unproductive.
15-17h room T10 Turftorenstraat, Faculty of Arts, University of Groningen
Productivity is a major conundrum in linguistic theory. Even the most fundamental questions are still a matter of debate. Is productivity an actual property of a linguistic construction, say a word formation pattern? Or is it an epiphenomenon of language use? Should productivity be considered the norm, so we need to explain the restrictions that we find? Or is productivity itself the phenomenon that needs explaining?
In this talk I offer a general take on morphological productivity from the perspective of a construction-based approach to morphology (Booij 2010, Jackendoff & Audring 2020). I discuss the necessity to make generous theoretical room for unproductivity when looking at word formation, and I present ways towards an inclusive model that accommodates the regularities as well as the idiosyncrasies that are so abundant in the grammar of words.
Booij, Geert. 2010. Construction Morphology. Oxford: Oxford University Press.
Jackendoff, Ray S. & Jenny Audring. 2020. The Texture of the Lexicon. Oxford: Oxford University Press.
27 November 2023. Kristian Berg, University of Bonn.
Diachronic productivity and corpora.
15-17h room T10 Turftorenstraat, Faculty of Arts, University of Groningen
Diachronic corpora allow us to gauge changes in a pattern’s productivity. In this lecture, I will argue that new words are an intuitively accessible measure of productivity. There are a number of pitfalls with diachronic corpora, however, such as varying corpus sizes over time. These challenges will be discussed, as well as the potential of semantic vectors to track semantic changes of words and patterns over time, and how they relate to productivity notions.
18 December 2024. Tanja Säily, University of Helsinki.
Sociolinguistic variation and change in productivity.
15-17h room A8 Academy Building, Faculty of Arts, University of Groningen
Following Baayen’s (e.g. 1992) work on quantitative aspects of productivity that viewed productivity as a gradient phenomenon, the focus of research into productivity has increasingly been on variation and change. Sociolinguistic variation in productivity was already brought up by Romaine (1983), but it is only in the past twenty years that more empirical and corpus-based research has begun to emerge (e.g. Säily 2014). In this lecture I will go through a few recent case studies illustrating different aspects of sociolinguistic variation and change in productivity, looking into constructions at various levels of linguistic organization in the history of English. We will see that some cases challenge the notion that women tend to lead language change (Labov 2001: 292–293). I will also address the methodological challenges of comparing type-based measures of productivity across social groups and time periods.
References
Baayen, R. H. 1992. Quantitative aspects of morphological productivity. In Geert Booij & Jaap van Marle (eds.), Yearbook of morphology 1991, 109–149. Dordrecht: Kluwer.
Labov, William. 2001. Principles of linguistic change, volume 2: Social factors (Language in Society 29). Malden, Massachusetts: Blackwell.
Romaine, Suzanne. 1983. On the productivity of word formation rules and limits of variability in the lexicon. Australian Journal of Linguistics 3(2): 177–200.
Säily, Tanja. 2014. Sociolinguistic variation in English derivational productivity: Studies and methods in diachronic corpus linguistics (Mémoires de la Société Néophilologique de Helsinki XCIV). Helsinki: Société Néophilologique.
22 January 2024. R. Harald Baayen, Quantitative Linguistics Lab, University of Tuebingen.
New perspectives on morphological productivity.
15-17h room A8 Academy Building, Faculty of Arts, University of Groningen
Word frequencies as collected from language corpora describe usage in language communities. High-frequency words are typically familiar to all speakers. This is not the case for low-frequency words. Individual low-frequency words tend to be well-known to specialists, but are unfamiliar to most other speakers. For instance, the word “harpsichord” is familiar to those who love Renaissance and Baroque keyboard music (like me), but not to speakers with no interest in `old music’.
This simple observation has important consequences for quantitative approaches to productivity. The counts of hapax legomena, which play a central role in the quantitative measures of productivity that I developed in the eighties, are useful for assessing `societal productivity’, but I now think that they are unlikely to properly probe productivity for individual speakers.
In order to approximate productivity from a cognitive perspective, I think it is more useful to make use of the discriminative lexicon model, a computational model of the mental lexicon. The discriminative lexicon model (DLM) posits simple mappings between numeric representations for words’ forms, and numeric representations for word meanings (e.g., embeddings). This model makes it possible to assess productivity both from the perspective of comprehension (how well does the DLM understand novel forms it hasn’t seen during training) and from the perspective of production (how well does the DLM produce novel forms). Importantly, in this approach, differences in productivity clearly emerge not only for derivational morphology, but also for inflectional morphology.
The mappings that the DLM works with to model comprehension and production are learned from usage. It has recently become possible to efficiently learn mappings that are properly sensitive to frequency of use. Typically, these mappings are very precise for the higher-frequency words, but low frequency words, and especially hapax legomena, cannot be learned in this way. As anyone familiar with deep learning will know, a single exposure is not sufficient for learning.
Nevertheless, discriminative learning models provide new insights into morphological productivity by assessing, on the one hand, how well a given model learns the words it is trained on, and on the other hand, how well it can understand and produce novel words that it hasn’t encountered during training.
In my presentation, I will present exploratory studies of the productivity of Estonian case-number inflection, of English plural inflection, and of compounding in Mandarin Chinese.
References
Baayen, R. H. (1992). Quantitative aspects of morphological productivity. In Booij, G. E., and Marle, J. van (Eds), Yearbook of Morphology 1991, Kluwer Academic Publishers, Dordrecht, 109-149.
Baayen, R. H. (1992). On frequency, transparency, and productivity, Yearbook of Morphology, 1992, 181-208.
Chuang, Y.-Y., and Baayen, R. H. (2021). Discriminative learning and the lexicon: NDL and LDL. In Aronoff, M. (Ed.), Oxford Research Encyclopedia of Linguistics.
Chuang, Y. Y., Kang, M., Luo, X. F. and Baayen, R. H. (2023). Vector Space Morphology with Linear Discriminative Learning. In Crepaldi, D. (Ed.) Linguistic morphology in the mind and brain, Routledge.
Heitmeier, M., Chuang, Y.-Y., Axen, S. D., and Baayen, R. H. (2023). Frequency effects in Linear Discriminative Learning. ArXiv, https://arxiv.org/abs/2306.11044.
Heitmeier, M., Chuang, Y-Y., Baayen, R. H. (2021). Modeling morphology with Linear Discriminative Learning: considerations and design choices. Frontiers in Psychology, 12, https://www.frontiersin.org/articles/10.3389/fpsyg.2021.720713.
Heitmeier, M., Chuang, Y.-Y., and Baayen, R. H. (2023). How trial-to-trial learning shapes mappings in the mental lexicon: Modelling lexical decision with linear discriminative learning. Cognitive Psychology, https://www.sciencedirect.com/science/article/pii/S0010028523000567.
Nieder, J., Chuang, Y.-Y., Vijver, Ruben van de, and Baayen, R. H. (2023). A discriminative lexicon approach to word comprehension, production, and processing: Maltese plurals. Language, 99 (2), 1-34.
Shafaei-Bajestan, E., Moradipour-Tari, M., Uhrig, P., and Baayen, R. H. (2022). Semantic properties of English nominal pluralization: Insights from word embeddings. ArXiv, https://arxiv.org/abs/2203.15424v1.
Shafaei-Bajestan, E., Uhrig, P., Baayen, R. H. (2023). Making sense of spoken plurals. The Mental Lexicon, https://www.jbe-platform.com/content/journals/10.1075/ml.22011.sha.
Shen T., and Baayen, R. H. (2023). Productivity and semantic transparency: An exploration of word formation in Mandarin Chinese. The Mental Lexicon, https://www.jbe-platform.com/content/journals/10.1075/ml.22009.she.
Maria Mazzoli, PhD 