Methods of quantifying distance between sound sequences are known as phonological distance measures. Despite the wide application across subfields, phonological distance has been calculated mainly with features related to consonants and vowels. This research report establishes new measurements of phonological distance that incorporate lexical tone through experimental approaches and modeling, using Hong Kong Cantonese as a case study. Results show correspondences between the experimental data and predictions from information-theoretic measures, including entropy measures and functional load, suggesting that lexical components which play a more crucial role in phonological distance judgments are lexically less predictable as well. Implications for phonological distance measures are discussed.

This article investigates how dependencies are constructed in prenominal relative clauses of Mandarin Chinese by comparing the comprehension of two types of relative clauses: Possessive Relative Clauses (PRCs), where the head noun is associated with a dependent noun phrase in the embedded clause, and Adjunct Relative Clauses (ARCs), where the head noun takes the whole embedded clause as its complement. The results of a naturalness-rating experiment and two self-paced reading experiments showed distinctive reading patterns of PRCs and ARCs. The comprehension of a PRC is sensitive to the grammatical position of the dependent noun in the prenominal clause: retrieval of a dependent noun at the subject position is less costly than that of a dependent noun at VP-internal nonsubject positions. The comprehension of an ARC reflects the structural frequency of the whole prenominal clause: more-canonical structures like SVO sentences were read faster than less frequent structures such as disposal and passive sentences. These results support the importance of structural locality and subject prominence for constructing gap-filler dependencies in prenominal relative clauses.

This article examines the syntax of extra be constructions, common in nonprescriptive English and often considered a curiosity, such as: The problem is, is that she hates apples. It has been claimed that there are many different types of extra be constructions, with the two main types being double be and single be, but this article argues that these distinctions are largely superficial. The article reviews previous accounts, presents the complex data, and categorizes most cases of extra be into one unified syntactic construction, the SHARED SHELL-NOUN CONSTRUCTION. It is argued that such constructions are syntactically fairly ordinary biclausal specificational copular sentences, consisting of a setup clause and a resolution clause, which share an argument. A second construction is also proposed for one subset of examples, the LINKING FOCUS BE CONSTRUCTION, where be lexicalizes a left-peripheral focus head.

What causes Indonesian to lenite word-final /k/, American English to lenite word-final /t/, and Spanish to lenite word-final /s/? This article shows that all three observed lenition patterns can be motivated using a single principle: languages preferentially lenite segments that provide relatively low informativity compared to the amount of informativity those segments carry in other languages. In a comparison of a diverse sample of seven languages from the LDC CALLHOME and CALLFRIEND corpora, Indonesian /k/, American English /t/, and Spanish /s/ are found to have the lowest informativity, predicting that they would be more likely to be affected by sound change processes. In a subsequent regression-based corpus study, low informativity predicted the propensity of word-final lenition of all obstruents in American English after phonetic and phonological factors were controlled for. This article therefore provides a partial solution to the famous actuation problem (Weinreich et al. 1968) with respect to the actuation of lenition processes.

An article published in Language (Sproat 2014a) questions our findings on the Indus script and Pictish symbols published in the journals Science (Rao et al. 2009a), PNAS (Rao et al. 2009b), IEEE Computer (Rao 2010), and the Proceedings of the Royal Society (Lee et al. 2010a,b). Sproat's article does not accurately present our methods and findings, and its conclusions are based on what appears to be a misunderstanding of our proposed approach. For example, the article's results on entropic measures seem to favor, rather than contradict, the inductive hypothesis that the Indus script may represent writing. The article selects results to draw a particular set of conclusions and convey a specific viewpoint. In light of these issues, we stand by our original findings.

Nouns in Nuer (Western Nilotic) have been presented as an extreme example of inflectional complexity, where a ‘chaotic’ distribution of suffixes combines with dozens of different stem modifications to yield dozens of inflection classes (Frank 1999, Baerman 2012). We show that all of the apparent surface variety can be reduced to a handful of operations. The proliferation of inflection classes is due to a property we call paradigmatic saturation: practically every combination of inflectional operations is attested, yielding the maximum variety with the minimum of means.

Subjacency characterizes a set of phenomena whose acquisition must be explained by any proposal for human language learning. We take a broader perspective than previous responses to Ambridge, Pine, and Lieven (2014), arguing that they have not shown that this UG principle is ‘redundant’ because their proposed alternative does not take into account firmly established constraints on A-bar dependencies. We illustrate a range of challenges for theories hoping to reduce subjacency to independently motivated, primarily nonsyntactic constraints: they must include a way to account for attested crosslinguistic variation in island effects, the cross-construction generality of island effects, and the effects of resumption and of WH-in-situ on island behavior.

The following report presents some of the findings of several years' research on the relations between standard English and the non-standard English used by Negro speakers in various urban ghetto areas. The immediate subject is the status of the copula and auxiliary be in Negro non-standard English. The approach to the problem combines the methods of generative grammar and phonology with techniques for the quantitative analysis of systematic variation. The notion ‘rule of grammar’ is enlarged to include the formal treatment of inherent variation as a part of linguistic structure. Furthermore, a model is presented for the decisive solution of abstract questions of rule form and rule relations, based upon the direct study of linguistic behavior.

The relevance of the concepts of local syntactic selection and constraints on locally selected dependents in modeling how arguments are syntactically realized has been virtually unchallenged and is assumed to be universal. In this article, we examine more closely these assumptions and ask whether there is anything invariant in the realization of arguments. We argue that the facts of Oneida, and Iroquoian in general, suggest the answer to this question is No. There is no use in Oneida for local selection of phrases realizing semantic arguments (and lexical records of this selection), and words are, as a result, functionally complete. We also argue that there is no use for a level of argument structure or any analogous notion in Oneida. The kinds of processes that justify positing such a level are systematically absent. The facts of Oneida suggest that there is less universality in the syntax/semantics interface (particularly when it comes to the realization of semantic arguments) than is typically assumed and that languages may vary widely in the way semantics is mapped onto syntax. We end with a formal model of the relevant fragment of Oneida within head-driven phrase structure grammar (HPSG).