Whamit!

Speaker: Lina In?iurait?-Noreikien? (Vilnius University)
Title: Derivational Transparency and Opacity in Borrowed Nouns
Time: Thursday, October 8th, 5pm - 6pm
Location: 32-D769

Abstract: Derivational relations in borrowed nouns have long attracted attention in Lithuanian
word-formation studies. However, relatively little research has systematically classified borrowed nouns according to their degree of derivational transparency. I examine borrowed nouns in contemporary Lithuanian from the perspective of derivational morphology. They can be divided
into two major categories: derivationally transparent and derivationally opaque. Derivationally
transparent nouns participate in a derivational opposition with independently attested base words and exhibit clear formal and semantic relations. By contrast, derivationally opaque nouns lack synchronically identifiable base words in Lithuanian and therefore do not participate in a
derivational opposition. Derivationally opaque nouns comprise two distinct subtypes: nouns with
bound stems that occur only in paired formations and nouns containing foreign formants that occur in isolation without corresponding pairs. Finally, I examine the semantic categorization of derivationally transparent and derivationally opaque borrowed nouns.

Time: Thursday, 12:30pm - 2pm

Location: 32-D461

Talk #1:

Speaker: Zhouyi Sun (MIT), Vlad Orlov (MIT), Oddur Snorrason (Queen Mary University of London)

Title: DD support at LF: On the last-resort nature of Distributed Deletion



Abstract: In this talk, we show that the interpretation of a discontinuously realized nominal (apparent left branch extraction; LBE) depends on the presence and attachment height of a parasitic-gap-containing (PG) adjunct licensed by it.

Building on Bondarenko and Davis (2023) who show than Russian LBE licenses parasitic gaps, we observe that in the absence of a relevant PG-licensing requirement, LBE obligatorily yields a split-scope interpretation with respect to an intervening scope-taking element (1a). Overt pied-piping of the corresponding nominal, by contrast, permits both high and low interpretations of the nominal restrictor (1b). Strikingly, this asymmetry disappears when a PG-containing adjunct attaches above the intervening operator: in this configuration, LBE and overt pied-piping have the same scope possibilities.

(1)

a.  [Skol?ko statej]1 Petja potreboval, ?toby ty otpravil t1
     How.many papers Peter required that you sent
     OK — papers require
     OK — require papers
b. Skol?ko3 Petja potreboval, ?toby ty otpravil [t3 statej]1
     How.many Peter required that you sent papers
     * — papers require
     OK — require papers

We argue that this interaction is problematic for adopting any single uniform analysis of LBE, whether in terms of direct subextraction, or distributed deletion. We propose instead that LBE has two competing derivations. Direct subextraction is preferred whenever possible, and blocks distributed deletion. Distributed deletion is available only as a last resort when required by an independent LF condition, such as PG licensing. We also show that this last-resort strategy is constrained by Economy Conditions, in that it moves only when necessary and only to the height necessary to satisfy the relevant LF requirement.

We further show that the same interaction is found in Icelandic and other languages and discuss additional LF conditions that force the distributed-deletion derivation. More generally, we propose that the last-resort nature of distributed deletion can be motivated by LF-requirements in addition to the previously established PF-constraints (e.g. Franks 1998, Bos?kovic? 2001).

Talk #2:

Speaker: Zhouyi Sun (MIT)

Title: Tigrinya gradable verbs: When verbalizers comparativize

Abstract: In Distributed Morphology, words are built syntactically from acategorial roots and categorizing functional heads such as v, a, and n (Halle & Marantz 1993, et seq.). In addition to determining syntactic category, categorizers may contribute semantic content. For verbalizers, previous work has focused primarily on meanings associated with argument structure and event structure (e.g. Arad 2005, Harley 2009). The range of possible meanings verbalizers can contribute remains an open question.

This work takes comparative meaning as a case study. A potential connection between verbalization and comparative meaning has been suggested by degree achievements (e.g. worsen; Bobaljik 2012: 171–172, 205–207), English out-comparatives (e.g. outweigh; Ahn 2022), and Mandarin transitive comparatives (schematically, ‘John tall Mary 5cm’; Grano & Kennedy 2012). Yet in these cases, it remains unclear whether the relevant comparative operator is contributed by a verbalizer. I suggest Tigrinya (Ethio-Semitic) offers a useful case for investigating this possibility: its gradable roots exhibit a systematic, category-conditioned positive-comparative opposition. Their adjectival forms have positive interpretations, as in b?lla? ‘smart’, whereas their verbal forms have comparative interpretations, as in t?-b?ll?? ‘she is smarter’. I show that the relevant forms are morphosyntactically verbal and semantically comparative, and argue that the comparative operator is contributed by the verbalizer. The analysis thus identifies comparison as a possible semantic contribution of verbal categorization.

For our Phonology Circle this Monday (10/5), we will be hearing two AMP practice talks from Hani Na’eem (MIT) and Chelsea Tang (MIT).

Time: Monday, October 5 5pm-6:30pm

Location: 32-D831

Speaker #1: Hani Na’eem (MIT)
Title: The nature of emphasis spread in Jordanian Arabic
Abstract: 

In this practice for an AMP poster presentation, I will give a summary of my work on the phenomenon of emphasis spread (ES) in Jordanian Arabic (JA). ES is a process whereby emphatics, coronals with a secondary constriction near the uvula, lower the second formant (F2) of sounds in their context, most saliently backing low vowels which are otherwise front. Previous work on ES has documented a directional asymmetry of the effect: while vowels to the left of the emphatics have their F2 uniformly lowered regardless of distance and the consonantal context, the effect on following vowels gradually fades out with distance and, in some descriptions, is blocked by high segments. The gradience of the rightward effect led to varying accounts with respect to the domain of spreading and/or the nature of the mechanism that gives rise to the effect. Some analyses (Davis 1995; Herzallah 1990; Jaber et al. 2019; McCarthy 1997; Younes 1982, 1993) posit a phonological account of rightward [rtr] harmony with various restrictions on its domain of application; others (Zawaydeh 1999; Zawaydeh & de Jong 2002) argue that ES is a ‘phonetic’ rather than a ‘phonological’ process.

Based on acoustic data from a production experiment, I argue that (i) emphatics spread [rtr] non-iteratively to two vowels to the right of the trigger, (ii) that the effect on the second vowel to the right does not reduce to a purely phonetic/coarticulatory mechanism, and (iii) that the observed gradience/smaller acoustic effect on this distant vowel is due to anticipatory coarticulation with the following context, which is outside the direct purview of the ES process. Adopting Walker’s (2011) formalism, I analyze both leftward and rightward ES through licensing constraints that seek to maximally license [rtr] in the surrounding context. While the constraint responsible for leftward spreading is highly ranked, the one responsible for rightward spreading is dominated by a CrispEdge constraint that delimits the spreading domain which I propose – as a first approximation – is sensitive to foot structure. The resulting analysis captures the directional asymmetry in ES while allowing the gradient phonetic realization of rightward spreading to emerge from the interaction between a categorical feature-spreading process and independent coarticulatory mechanisms.

Speaker #2: Chelsea Tang (MIT)
Title: From Denasalization to Backcopying in Nyandarua G?k?y?: the Emergence of Base-Reduplicant Identity from Sound Change
Abstract:

This presentation examines how a historical change in the consonant inventory of Nyandarua G?k?y? gives rise to a new pattern of consonant backcopying. Earlier descriptions of G?k?y? document a consonant system with prenasalized voiced stops but no corresponding plain voiced stops, while Nyandarua and other northern and southern varieties have undergone the loss of prenasalization, resulting in only plain voiced stops (Ngure 2005). The two varieties also differ in reduplication. Both Standard and Nyandarua G?k?y? show templatic backcopying, where the reduplicant can affect the shape or vowel quality of the base, but consonant backcopying is found only in Nyandarua.

I argue that the emergence of consonant backcopying follows from the historical change in Nyandarua Gi?ku?yu? consonant inventory. Diachronic work in Optimality Theory models sound change through changes in constraint ranking (Jacobs 1995; Holt 2003, 2015; Lleó 2003), and many analyses show that a single instance of constraint reranking can have consequences for multiple (sometimes apparently unrelated) phonological processes (Holt 2003; Miglio & Morén 2003; Padgett 2003; Smith 2023; among others). Building on these accounts, I analyze the loss of prenasalized stops as a reranking of *NC and IO-Id[nas], such that *NC comes to dominate IO-Id[nas] in Nyandarua, while varieties that retain prenasalized stops have the IO-Id[nas] >> *NC grammar. In reduplication, templatic backcopying in both varieties establishes that BR-faith is active. However, the effect of BR-faith on consonants is masked by a higher-ranked IO-Id[nas] that preserves prenasalized stops in Standard G?k?y?. By contrast, in Nyandarua, the reranking of *NC and IO-Id[nas] responsible for denasalization also places IO-Id[nas] below BR-Id[nas], allowing consonant backcopying to surface. The historical loss of prenasalized stops therefore not only changes the consonant inventory, but also gives rise to consonant backcopying in Nyandarua G?k?y?.

We are delighted to announce that the paper “Scale-sensitivity of comparatives and measure phrase interpretations in Gitksan” by 5th-year student Yurika Aonuki has just been published in Semantics of Natural Languages. Congratulations, Yurika!

Here is the abstract:

This paper demonstrates that in Gitksan, measure phrases (MPs) occurring with relative gradable adjectives (GAs) like ’wii ’nakw ‘tall’ receive differential interpretations even in the absence of an overt comparative operator or an overt standard phrase, a pattern which has been only reported in Japanese (e.g., Snyder et al. 1995) and Korean (Oda 2008; contra Sawada and Grano 2011). In contrast, MPs occurring with minimum-standard GAs like k?’ak? ‘open’ receive absolute interpretations, as in many languages including Japanese (Sawada and Grano 2011). However, Gitksan differs from Japanese in two crucial ways. First, the distinction between relative and minimum-standard GAs is observed even without MPs: relative GAs can easily receive comparative interpretations on their own, while minimum-standard GAs cannot. Second, Gitksan has a morphologically free, overt morpheme k’aa, which can give rise to comparative readings with any GA and hence makes the possibility of a covert comparative operator less plausible. On a closer investigation, the comparative readings of relative GAs without k’aa show mixed behaviours of what have traditionally been called ‘explicit’ and ‘implicit’ comparatives (Kennedy 2007). I propose that these comparative readings arise from inherently context-sensitive denotations of relative GAs, building on Oda’s (2008) analysis of Japanese. The latter half of the paper examines the semantics of k’aa and analyzes it as a superlative morpheme, and highlights the role of consideration of alternatives to the subject in comparative readings observed without k’aa. 

The paper is open access: https://doi.org/10.16995/snl.29102