ANTHONY D. YATES

I'm currently an assistant professor at the University of California, Los Angeles (UCLA), with joint appointments in the interdepartmental Program in Indo-European Studies (PIES) and in the Department of Near Eastern Languages and Cultures (NELC). I received my PhD from PIES in 2017, then spent time as a Humboldt Research Fellow in the Lehrstuhl für Historische und Indogermanische Sprachwissenschaft at Ludwig-Maximilians-Universität München before returning to UCLA.

I specialize in Indo-European comparative-historical linguistics and in the philology and linguistics of the Anatolian languages, with particular focus on Hittite. My research is concerned broadly with the synchronic and historical grammar of the Anatolian languages, and with how these inform our understanding of the rest of the Indo-European language family.

Some aspects of my recent research are outlined below; downloadable versions of my publications and conference handouts/posters/slides are available here. Feel free to contact me with any questions here.

[ News ]

10/26/24: My UCLA collaborators and I will be presenting today at the 35th Annual UCLA Indo-European Conference on the Proto-Indo-European nasal infix *–ne– and the development of PIE verbal system. Our slides are available here.
10/17/24: Today I'll be talking (on behalf of myself and UCLA collaborators) at the 55th Annual Meeting of the North East Linguistic Society about the phonology and morphosyntax of the Proto-Indo-European nasal-infix. Why is it an infix? You can find our answer in the slides.
7/1/2024: Today I'm in Athens, GA for the 43rd Annual East Coast Indo-European Conference. I'll be talking about the diachrony of vowel deletion from (pre-)Proto-Indo-European into the attested Indo-European languages. You can check out the slides here.
6/17/2024: I have a new paper coming out in Indogermanische Forschungen, which is concerned primarily with the Anatolian reflexes of Indo-European τομή- and φυγή-type formations and their implications for our understanding of Anatolian and Indo-European morphology and phonology. The manuscript is available here — comments welcome!
3/22/2024: I am in Chicago today for the 234th Annual Meeting of the American Oriental Society, where I'll be presenting joint work with Craig Melchert on the phenomenon of hyperbaton in Hittite. The handout is available here.
12/4/2024: Today I'm in Edinburgh for the Sixth Edinburgh Symposium on Historical Phonology. I'll be delivering a paper this afternoon on vowel deletion in the ancient Indo-European languages. I argue that cyclic vowel deletion should be reconstructed for Proto-Indo-European itself — check out the slides here.

[ Research ]

[ Anatolian ]

A lot of my research deals with the (extinct) Anatolian branch of the Indo-European (IE) language family –– above all, with its major representative, Hittite. I've studied the linguistics and philology of Hittite and the other Anatolian languages (Luwian, Lycian, Palaic, etc.) with Craig Melchert, who supervised my dissertation, and I've since taught these languages at UCLA.
My research focuses on the synchrony and diachrony of the phonological systems of these languages, esp. on the interaction of stress and other phonological processes (e.g., vowel deletion/lengthening/shortening).

[ Lexical accent ]

My recent research has focused on the synchronic analysis and diachronic development of word-prosodic systems –– in particular, those in which stress is morphophonologically determined: morphemes are lexically specified for certain accentual properties, which interact with each other and with phonological principles in the computation of stress. Such lexical accent systems are found in IE languages, both ancient and modern, as well as in various dialects of Japanese, Salishan languages, and Cupeño (Takic, Uto-Aztecan), i.a.
My dissertation examines lexical accent in Cupeño and Hittite, developing novel generative analyses of word-level stress in each. I show that these languages are not just broadly comparable typologically, but in fact share specific principles of stress assignment --- in particular, a general preference for left edge word stress, which emerges when (i) the word lacks accented morphemes ("default" stress) or when (ii) the word has multiple accented morphemes.
Cupeño and Hittite are of special interest because they offer complementary perspectives on the historical development of lexical accent systems. The languages ancestral to Cupeño had fixed (metrical) stress, which is clearly reflected in its closest relative, Cahuilla. I am currently working on tracing the evolution of lexical accent in Cupeño (see the Research page for details).
On the other hand, Hittite inherited a lexical accent system from Proto-Indo-European (PIE) and thus testifies to the types of changes that such systems undergo. Broadly speaking, I am interested in what these changes tell us about morphological processing and the learnability of lexical accent systems (one case is discussed here). However, our ability to diagnose such changes depends crucially on our understanding of PIE word stress --- another major topic of my research.

[ Indo-European word prosody ]

How was word stress determined in the ancient Indo-European languages? What was the relationship between word stress and ablaut (i.e., quantitative and qualitative vowel alternations)? And how did these prosodic systems come to be the way that they are? My work aims to address these questions, in part by synchronic analyses of these languages, and in part by comparative reconstruction of the word-prosodic system of PIE.
One goal of my dissertation was to more systematically integrate the Anatolian evidence into our reconstruction of PIE stress assignment. A major finding was that many of the same principles of stress assignment obtain in Anatolian as in Vedic Sanskrit, which is generally thought to best preserve the PIE system. I am currently working on a monograph that builds on the results of the dissertation (working title: Hittite and Indo-European Word Prosody).
My interest in reconstructing the PIE word-prosodic system is in service of two other aims: (i) explaining how the prosodic systems of the daughter languages arose; (ii) gaining insight into how prosodic systems can change over time. The IE languages offer rich historical data and are thus ideally suited for testing theories of prosodic change — but in order for this data to be useful, it is necessary to understand as precisely as possible how their prosodic systems operated at each diachronic stage.

[ Typology & morphological theory ]

I am interested in linguistic typology (especially morphological and phonological). Typology has long played an important role in Indo-European linguistics and always informs my own research (e.g., here).
One objective that cuts across my research is to make the data of the ancient IE languages more accessible to typologists and theoretical linguists. To this end, I have recently published (with Jesse Lundquist) a critical overview of the reconstructible morphology of PIE in which we attempt to describe traditional IE problems in terms that have wide currency in general linguistics.
Yet my work also attempts to directly bring to bear ancient IE linguistic data on morphological theory. I am currently working (with John Gluckman) on voice alternations in Hittite and their implications for the morphosyntax of voice cross-linguistically (see the Research page for details).

[ Experimental phonology ]

I am interested in how people learn language. A question of particular interest is whether phonological change occurs because certain sound patterns are difficult or even impossible to learn. Experimental methods serve as a natural complement to traditional approaches to historical linguistics in investigating this question, since they allow us to explicitly test hypotheses about language learning and change.
Previously I worked (with UCLA colleagues) on an artificial grammar learning experiment in which we investigated whether learners exhibit “substantive bias,” i.e., a cognitive bias against marked phonological patterns. Specifically, we tested whether a word-final obstruent devoicing alternation, which is well-attested cross-linguistically (e.g. German, Russian), is more easily acquired than an otherwise identical voicing alternation, which is extremely rare (perhaps in Lezgian) or unattested. The results from this study were presented at the 92nd Annual Meeting of the Linguistic Society of America in January 2018; you can download the handout here.