Phillip G. Rogers

Syntactic architecture of the lexicon

Contrary to traditional assumptions in generative linguistic theory, recent work has demonstrated that the lexicon contains fine-grained, probabilistic information about the syntactic contexts in which a word participates. In this project, we explore some of the implications of this finding cross-linguistically. Are words clustered or dispersed in this syntactic space? Are there systematic relationships between syntactic distributions and other features of the lexicon such as wordforms and meanings? Might syntactic distributions play a role in grammatical phenomena such as case systems and grammatical gender assignment? These questions serve as a convenient locus for testing psycholinguistic models with opposing predictions, and the results provide new insights into how language is structured for efficient use.

Relevant publications:

Rogers, Phillip & Stefan Th. Gries. (2022). Grammatical gender disambiguates syntactically similar nouns. Entropy 24(4), p. 520. https://doi.org/10.3390/e24040520

Lester, Nicholas A., Sandra Auderset, & Phillip Rogers. (2018). Case inflection and the functional indeterminacy of nouns: A cross-linguistic analysis. Proceedings of the 40th Annual Conference of the Cognitive Science Society. https://escholarship.org/uc/item/4q41w313

Documentation and description of the Bitur and Abom languages

The Bitur language is spoken by less than a thousand people in Western Province, Papua New Guinea. It is a member of the Lower Fly family, the least-known language group in Southern New Guinea. The nearby Abom language is a Trans-New Guinea isolate with just three native speakers. I am working with these communities to document and describe their languages. These projects are driven by community goals, and the results inform Papuan language history and typology.

Relevant publications:

Rogers, Phillip. (In revision). The birth of a noun class: Gender actualization in Bitur, a language of Southern New Guinea.

Rogers, Phillip. (2021). The phonetics of Bitur. In Kate L. Lindsey & Dineke Schokkin (eds.), Language Documentation & Conservation Special Publication No. 24: Phonetic fieldwork in Southern New Guinea, pp 108-119. Honolulu: University of Hawai'i Press. http://hdl.handle.net/10125/24996

Rogers, Phillip. (2018). Documentation and description of Bitur, a Lower Fly language of Papua New Guinea, and preliminary investigation of the moribund Abom language. London: SOAS, Endangered Languages Archive. http://hdl.handle.net/2196/86fb67aa-a44c-4f7c-b8d8-723db240b821

Cross-linguistic tendencies in the phonological length of affixes

with Tim Zingler

This research seeks to fill an empirical gap in the literature on the phonological length of affixes cross-linguistically. Using a genetically and geographically balanced sample of languages, we test claims such as whether affixes tend to be monosyllabic and whether prefixes differ from suffixes in length. Our findings are interpreted in the context of the various factors that conspire to motivate affix length, including economy, perceptibility, phonotactic constraints, and processing considerations.

Relevant publications:

Zingler, Tim & Phillip Rogers. (2024). The interaction of affix size, type and shape: a cross-linguistic study. Linguistic Typology. https://doi.org/10.1515/lingty-2024-0004

Cross-clausal associations in complex sentence constructions

with Jesús Olguín Martínez

In this research, we explore cross-clausal associations among morphosyntactic features of complex sentences in a large and diverse sample of languages. Using statistical methods such as hierarchical configural frequency analysis, we identify types and antitypes—preferred and dispreferred associations among these features. Our results speak to the different ways languages negotiate competing communicative pressures of economy and discrimination.

Relevant publications:

Olguín Martínez, Jesús & Phillip Rogers. (2024). A cross-linguistic analysis of cross-clausal associations: Counterfactual conditionals. STUF - Language Typology and Universals, 77(4), pp. 467-514. https://doi.org/10.1515/stuf-2024-2014

Modeling language phylogeny and feature dynamics in real geographic space

Despite recent innovative efforts to understand the distribution of languages and language features in time and space, many questions remain unanswered. We introduce a model for simulating languages and their features over time in a realistic geographic environment. This model operates over a number of discrete timesteps at which languages can speciate, migrate, or die. The settings for these parameters are calibrated by comparing the model output to real languages and families. Languages can also be assigned values of a feature, which can change at each timestep randomly or under the influence of surrounding languages. The resulting model is both flexible and realistic, and it can be employed to answer a wide range of questions concerning language phylogeny and feature dynamics.

Relevant publications:

Kapur, Rhea & Phillip Rogers. (2021). Modeling language evolution and feature dynamics in a realistic geographic environment. Proceedings of the 28th International Conference on Computational Linguistics, pp 788-798. https://doi.org/10.18653/v1/2020.coling-main.68

Typological hierarchies and the learnability of gender and number syncretisms

This experimental research seeks to shed light on the motivations behind two well-known implicational universals of typology: the number and animacy hierarchies. In one study, participants are tasked with learning an artificial language featuring syncretisms in grammatical gender and number that either do or do not violate these hierarchies. In another study, we introduce variation to see if participants regularize toward or away from cross-linguistically dispreferred syncretisms. The results speak to differences between the two hierarchies, as well as the role of constraints on learnability as a possible explanation for cross-linguistic patterns.

Illustrating the prototype structures of parts of speech

Radical Construction Grammar proposes that parts of speech can be explained as prototypes that emerge from the use of broad semantic classes of words—objects, properties, and actions—in basic propositional act functions of discourse—reference, modification, and predication. Equipped with conceptual targets covering the range of prototypicality for each part of speech, we identify lexemes corresponding to these concepts in eleven diverse languages. Typological markedness criteria are used to identify asymmetries in how these lexemes are formally encoded relative to each other across the three propositional act functions. A multidimensional scaling analysis of these assymetries illustrates the prototype structure of parts of speech and sheds light on the semantic primitives that motivate these structures.

Relevant publications:

Rogers, Phillip. (2015). Illustrating the prototype structures of parts of speech: A multidimensional scaling analysis. MA Thesis, University of New Mexico. https://digitalrepository.unm.edu/ling_etds/28