top of page

PROJECTS

My research aims to bring creative solutions to long-standing analytical and empirical problems in phonology, morphology, and the interface between the two, with the goal of improving our theoretical modeling of those domains. My work falls into three main projects, all of which work in concert to build a coherent grammatical architecture of the phonology-morphology interface. 

​

​​​Project #1: Thorny Problems in Reduplication

​​

​Reduplication is a word-building process that involves the copying of a circumscribed portion of the root or stem, thus leading to distinct sound sequences in words with different sounds or different shapes. This is distinct from typical affixation, which builds a complex word by adding a consistent sound sequence, regardless of the properties of the word it is attaching to. 

​

Reduplication has played an important role in the development of phonological theory, because it involves so many different moving parts, all interacting with each other in complex and interesting ways. For this very reason, many aspects of reduplicative theory remain unsettled, and there are multiple competing theories of reduplication on the market today. This project seeks to better understand the true predictions of these competing models -- which often are at odds with what the proponents of those models have claimed -- and to develop new analyses of important empirical patterns which can help us properly evaluate these revised predictions.

​

The Austronesian languages -- a language family spoken widely throughout East and Southeast Asia, the Pacific, and even Madagascar -- have played a major role in the literature on reduplication, as they happen to contain some of the most crucial kinds of interactions for shedding light on the machinery of reduplication. As such, much of my work on the matter has focused on Austronesian languages, such as Malay, Tawala, Ponapean, and Hawaiian. But my work has also drawn on languages from other parts of the globe, including Indo-European languages, aboriginal Australian languages, and indigenous languages of North America.

​

Throughout this work, I argue for a model of reduplication known as Base-Reduplicant Correspondence Theory (BRCT). BRCT asserts that the two parts of the reduplicated word -- the base and the reduplicant -- are built at the same time, and with reference to one another. Similarity between the two parts is enforced via a "correspondence relation", a formal connection between the individual components of the two parts which is evaluated through specialized faithfulness constraints. This theory stands in contrast with all other models of reduplication, which introduce some manner of serial derivation, wherein the base is computed first and fixed in place, and then the reduplicant is built with reference to it (or, in at least one case, without reference to it).

 

In several strands of my work, I show that certain claimed predictions of these alternative models, which are said to prefer them to BRCT, do not hold up to scrutiny, and instead do not distinguish the respective models at all. 

​​

​

Within this work, as well as in other strands, I show that certain empirical patterns robustly argue in favor of BRCT, as alternative models fail to generate them entirely. First and foremost among these is the interaction between reduplication and nasal spreading in Malay. This pattern has long been identified as favorable to BRCT, but has drawn skepticism due to a lack of instrumental documentation. In fieldwork led by my co-author Jian-Leat Siah, we have documented the veracity of this pattern, and reaffirmed that it stands as a strong argument in favor of BRCT. 

​

​

In related work with Juliet Stanton, I argue that an equivalent correspondence-based theory is the right way to model "copy epenthesis", a phonological pattern that inserts a copy vowel to repair a bad sound sequence rather than a default vowel, which is the more common type cross-linguistically. Based on evidence from copying of properties above the level of the segment and the feature -- properties like stress and length -- we show that traditional accounts based on autosegmental spreading are not sufficient.

​​​

Project 1

​​​Project #2: Word-Building at the Phonology-Morphology Interface

​​

​Reduplication is not the only word-building process that consists of something other than simple affixation. While much less common than simple affixation, or even reduplication, word-building processes like root-and-pattern morphology, templatic morphology, infixation, and mobile affixation are crucial to our understanding of grammatical architecture.

 

In this project, I have developed a novel framework for explaining how these processes work, which also applies seamlessly to simple affixation and reduplication. I call this the "Mirror Alignment Principle", or the MAP for short. This approach uses "alignment" constraints -- constraints that pull the respective pieces of a complex word (called "morphemes") towards one edge of the word or the other -- to explain the order in which those pieces appear. But the prioritization of the alignment constraints (their "ranking") is not random; rather, it is determined by the syntax on a dynamic basis.

​

Using alignment constraints tied to the syntax successfully implements the well-known "Mirror Principle" generalization. We know that languages vary with respect to whether they express particular ideas by means of strings of morphemes within a complex word (morphology) or by means of strings of words within a complex phrase (syntax). The Mirror Principle states that the ordering properties of morphemes within words directly mirror the ordering properties of the equivalent words within a phrase. The MAP derives this generalization, while simultaneously allowing phonological constraints to monkey with the outputs in interesting ways. The atypical word-building processes mentioned above are precisely the result of this monkeying around. 

​

Here are some of my research outputs that apply the MAP and related frameworks to these different empirical processes. (Some are listed multiple times since they cover multiple patterns.)

​

​Root-and-Pattern Morphology in Arabic

​

  • Zukoff, Sam. 2024. Morpheme Ordering Happens in the Phonology. Job talk, UCLA Department of Linguistics, January 29, 2024. [Slides]

  • Zukoff, Sam. 2023. The Mirror Alignment Principle: Morpheme Ordering at the Morphosyntax-Phonology Interface. Natural Language & Linguistic Theory 41(1): 399-458. doi:10.1007/s11049-022-09537-2. 

  • Zukoff, Sam. 2021. Deriving Arabic Verbal "Templates" without Templates. Proceedings of the LSA 6(1): 144–158. doi:10.3765/plsa.v6i1.4955. [PDF]

​

​Templatic Morphology in Bantu

​

  • Zukoff, Sam. 2023. The Mirror Alignment Principle: Morpheme Ordering at the Morphosyntax-Phonology Interface. Natural Language & Linguistic Theory 41(1): 399-458. doi:10.1007/s11049-022-09537-2. 

  • Zukoff, Sam. 2023. The Mirror Principle and Pseudo-cyclicity in Bantu Templatic Morphology. Talk presented at UCLA Phonology Seminar, May 31, 2023. [PDF]

​

​Infixation in Indo-European and Arabic

​

  • Baki, Klaus, Anthony D. Yates, & Sam Zukoff. 2025. A Phonology–Morphosyntax Interface explanation of the “Nasal Infix” in (Proto-)Indo-European. Proceedings of NELS 55, edited by Duygu Demiray, Roger Cheng-yen Liu, and Nir Segal, vol. 1:39–52.  [PDF]​

  • Baki, Klaus, Anthony D. Yates, & Sam Zukoff. 2024. Indo-European Nasal Infixation and the Mirror Alignment Principle. Paper presented at WeCIEC 35, UCLA, October 25-26, 2024. [PDF]​

  • Baki, Klaus, Anthony D. Yates, & Sam Zukoff. 2024. A Phonology-Morphosyntax Interface Explanation of the “Nasal Infix” in (Proto-)Indo-European. Paper presented at NELS 55, Yale University, October 17-18, 2024. [PDF]

  • Zukoff, Sam. 2023. The Mirror Alignment Principle: Morpheme Ordering at the Morphosyntax-Phonology Interface. Natural Language & Linguistic Theory 41(1): 399-458. doi:10.1007/s11049-022-09537-2. 

​

​Mobile Affixation in Huave and Moro

​

  • Zukoff, Sam. 2024. Morpheme Ordering Happens in the Phonology. Job talk, UCLA Department of Linguistics, January 29, 2024. [Slides]

  • Zukoff, Sam. 2021. A Parallel Approach to Mobile Affixation in Huave. Supplemental Proceedings of AMP 2020, edited by Ryan Bennett, Richard Bibbs, Mykel Loren Brinkerhoff, Max J. Kaplan, Stephanie Rich, Nicholas Van Handel, and Maya Wax Cavallaro. Washington, DC: LSA. doi:10.3765/amp.v9i0.4910.

  • Zukoff, Sam. 2021. The Role of Alignment in Moro Affix Mobility: A Friendly Amendment to Jenks & Rose (2015). Ms., USC, October 19, 2021. lingbuzz/006265.

Project 2

​​​Project #3: Reduplication in the Ancient Indo-European Languages

​

​My first major research project, which began with my undergraduate thesis and led all the way to my dissertation, was on the analysis of reduplication in the ancient Indo-European (IE) languages, focusing on Ancient Greek, Hittite, Gothic, and Sanskrit, plus a number of other IE languages. It centered on a particular reduplication pattern which is inherited into many of the IE daughter languages, to one extent or another, namely the Consonant-Vowel (CV) prefixal reduplication pattern of the Proto-Indo-European (PIE) perfect tense. This pattern is noteworthy because, across the language family, its properties are remarkably similar, yet diverge in interesting ways. These divergences fall into two main types: what happens when the root begins with a "bad" sequence of consonants (a bad "cluster"), and which clusters count as bad.

​

The details of these patterns across the languages have been known since the advent of Indo-European studies in the 19th century. And yet, there had been very few attempts to explain these fine details in a theoretically-rigorous and empirically-adequate way -- and, if you ask me at least, none which had succeeded in doing so. My proposal centers on a novel constraint penalizing consonant repetitions in particular phonetic environments. The variation in which are the bad clusters results from cross-linguistic differences in the specification of the required phonetic properties in the environment, related specifically to consonant contrast perceptibility. The variation in what do you do with bad clusters results from language-specific rankings of a small number of relevant constraints -- you perform the operation that violates the lowest-ranked constraint. 

​

I used the analysis of the synchronic systems of the attested languages, and the pre-languages accessible through shallow reconstruction, to argue for a reconstruction of the reduplicative system of PIE that is at odds with the traditional reconstruction. The traditional reconstruction proposes a system equivalent to Gothic (cluster-copying for bad cluster roots), with reductions into the daughter languages. I, on the other hand, equipped with synchronic analysis as a tool for historical reconstruction, proposed to reconstruct a system with single-consonant copying to PIE, akin to what is observable in Old Irish, and in demonstrably archaic forms in Ancient Greek, Latin, and Old Persian.

​

Consult my dissertation for the most extensive treatment of these topics:

​

​

Here are other relevant papers:

​​​​​

  • Zukoff, Sam. 2025. Reconstructing the Phonology of Proto-Indo-European Reduplication. In Satoko Hisatsugi, Junichi Ozono, and Matilde Serangeli (eds.), The Reduplication in the Indo-European Languages. Studien zur historisch-vergleichenden Sprachwissenschaft 22. Hamburg: Baar Verlag, 27–46. [Pre-publication version]​​​​​​

  • Zukoff, Sam. 2024. Sonority Hierarchy (Ancient Greek). Encyclopedia of Greek Language and Linguistics Online, edited by Giorgios K. Giannakis. Brill. doi: 10.1163/2666-6421_GLLO_COM_056949.​​

Project 3
bottom of page