Language and Linguistics - Volume 26, Issue 2, 2025

Mandarin ordinal phrases demonstrate an ambiguity between a definite reading and an indefinite quantity reading, first observed by Tsai (2011). However, contra Tsai (2011)’s lexical-ambiguity view, this paper presents a compositionally unified semantic analysis of the ambiguity of Mandarin ordinal phrases. Specifically, the ambiguity of Mandarin ordinals under discussion is derived from the semantic application of type-shifting rules and the pragmatic restriction on the domain of the ordinal morpheme di (which is reminiscent of the in-situ approach to the ambiguity of superlatives), coupled with the semantic requirements of di; therefore, a unified semantics of the ordinal morpheme di is maintained under either reading. Furthermore, this paper suggests two important felicity conditions in general on the use of an ordinal phrase in discourse. Besides theoretical contributions, this paper also presents empirical discoveries showing that some generalizations in Tsai (2011) are not entirely correct. For instance, it is false that all ordinal phrases in Mandarin Chinese must denote a singular entity (contra Tsai 2011). Instead, the semantics of classifiers matters for the denotation of an ordinal phrase: a Mandarin ordinal phrase can denote a plural entity when a group classifier is used. Furthermore, the presence of measure classifiers or certain predicates such as sheng ‘give birth’ is not a reliable diagnostic for teasing apart the two readings; the role of contexts and world knowledge in determining the (un)availability of a given reading is identified and discussed. Finally, this paper illustrates how the proposed analysis explains three puzzling properties of Mandarin ordinal phrases: (a) why the indefinite quantity reading is missing in the subject position; (b) why the indefinite quantity reading is missing when the experiential aspect marker guo occurs; and (c) why an ordinal phrase with canonical sortal classifiers has to denote a singular entity.

This paper focuses on tonal development in Baima, a little-studied Tibetic (Tibeto-Burman) language of China. Based on newly collected data and an updated phonological analysis, it addresses the phonetic and chronological challenges posited by the previous analysis of tonal developments in Baima by Huang & Zhang (1995). As suggested by historical-comparative evidence (correspondences between Baima and Written Tibetan), the three contrastive tonal categories in Baima have likely arisen through an overlay of the phonation difference in consonants (historically breathy vs. non-breathy) and the phonation difference in vowels (tense vs. lax, with pitch and vowel duration as co-articulated cues). I furthermore argue that the development of a tense/lax distinction is likely broadly shared by the Tibetic languages neighboring Baima. Recognition of such a multidimensional contrast (which may variously implicate differences in phonation, pitch, vowel duration, and vowel quality) provides a unified explanation for a number of seemingly unrelated and unusual characteristics of these Tibetic languages (such as rhyme length and onset aspiration as pathways leading to distinctive low register). The clustering of Tibetic languages with a tense/lax distinction in the historically multilingual area at the border of the present-day Northern Sichuan and Southern Gansu is further brought in connection with local non-Tibetic languages, some of which possibly share a structurally similar voice quality contrast that involves sets of correlated phonetic properties associated with phonatory, tonal, and vowel quality.

In Chinese, numerals and classifiers are in principle adjacent to each other, forming a prosodic unit. Apparent violation of Num-Cl adjacency is observed when a zero-level adjective modifies the classifier. Such violation is not observed when the adjectives are phrasal. The current study provides a comprehensive description and analysis of different conditions of Num-Cl Adjacency constraints in Mandarin. Specifically, this paper proposes that the Num-Cl adjacency effect is reducible to the numeral lowering at PF, which is motivated by the morphological dependency of numerals rather than a requirement on the linear relation between numerals and classifiers. This analysis accounts for all observations regarding word order between the numeral, the classifier, and the different types of adjectives in a unified way. In addition, it is shown that the numeral-lowering analysis is also applicable to complex numerals and can solve the problem of some approximant numerals.

In the study of Rma phonology (Eastern Tibeto-Burman < Sino-Tibetan/​Trans-Himalayan), there have been differing perspectives regarding labialized onsets both in synchrony and diachrony. Synchronically, these forms have been treated as CuV or as CwV sequences (Sun 2003). In terms of diachrony, Chang (1967) reconstructs the following set of bilabial-stop initial clusters: *Bt, *Bd, *Bts, *Bts^h, *Bdz, *Bs. Evans (2001; 2006a) reconstructs none of these and proposes that labialized onsets are secondary developments in the dialects that have them.

This paper explores the issue of labialized onsets through the lens of a previously undocumented variety of Rma. It offers a new perspective on the synchronic analysis for these segments as phonologically C^w and contributes to an older discussion about the diachrony of these sounds. The paper finds support for Evans’ (2001) theory that these sounds are secondary developments from *-w- elements while also finding sub-group external evidence that some come from *p- pre-initials.

Apropos of the level of specificity, schematicity is the key indicator of grammaticalization in linguistics; compared to lexical items, the information provided by grammar patterns tends to be more abstract. With recourse to the notion of the quantifiable information content in information theory, the schematicity of a schema can be quantified by comparing the information content provided by the elements occurring in the open slots to that provided by the schema itself. A formula is thereby proposed to measure schematicity. This schematicity measure is able to illustrate the gradience and gradualness of grammaticalization in its applications in synchronic English data and diachronic Chinese data. Compared to previous measures of grammaticalization, there is a notable improvement in reliability and applicability.

This paper means to make a comparative study of the nominals denoting an event in Mandarin and English. I propose that although English has three types of eventive nominals — simple event nominal, Referential (R-) nominal, and Argument Structure (AS-) nominal — Mandarin has only simple event nominals. The category shifting, which is required in the derivation of R-nominals and AS-nominals in English, is an operation unavailable in Mandarin, a result caused by the lack of categorial (c-) functors by assumption. I propose that a Mandarin eventive nominal enters syntax as a category-less root and immediately gets an N category by merging as the complement to a semantic (s-) functor from the extended projection of N. It thus projects no argument structure and undergoes no V-N shift, resulting in its incapability of taking post-nominal arguments. The DPs interpreted as arguments are merged external to the nominal and are thus flexible in terms of their thematic roles. The analysis shows that there is a parametric variation based on the inventory of c-functors between Mandarin and English.