Language  (Vol. 92, No. 1) March 2016

Modally hybrid grammar? Celestial pointing for time-of-day reference in Nheengatú

Simeon Floyd, Max Planck Institute for Psycholinguistics

From the study of sign languages we know that the visual modality robustly supports the encoding of conventionalized linguistic elements, yet while the same possibility exists for the visual bodily behavior of speakers of spoken languages, such practices are often referred to as ‘gestural’ and are not usually described in linguistic terms. This paper describes a practice of speakers of the Brazilian indigenous language Nheengatú of pointing to positions along the east-west axis of the sun’s arc for time-of-day reference, and illustrates how it satisfies any of the common criteria for linguistic elements, as a system of standardized and productive form-meaning pairings whose contributions to propositional meaning remain stable across contexts. First, examples from a video corpus of natural speech demonstrate these conventionalized properties of Nheengatú time reference across multiple speakers.  Second, a series of video-based elicitation stimuli test several dimensions of its conventionalization for nine participants. The results illustrate why modality is not an a priori reason that linguistic properties cannot develop in the visual practices that accompany spoken language. The conclusion discusses different possible morphosyntactic and pragmatic analyses for such conventionalized visual elements, and asks whether they might be more cross-linguistically common than we presently know.

On measurement and quantification:The case of most and more than half

Stephanie Solt, Zentrum für Allgemeine Sprachwissenschaft (ZAS)

The quantifiers most and more than half pose a challenge to formal semantic analysis. On the one hand, their meanings seem essentially the same, prompting accounts that treat them as logically equivalent. On the other hand, their behavior is known to diverge in a number of interesting ways. This paper draws attention to some previously unnoticed contrasts in distribution and interpretation between the two, and develops a novel semantic analysis of them, based on principles of measurement theory. Most and more than half have logical forms that are superficially equivalent (per Hackl 2009), but which place different requirements on the structure of the underlying measurement scale: more than half requires a ratio scale, while most can be interpreted relative to an ordinal scale or one with a semiordered structure. The latter scale type is motivated by findings from psychophysics, and by psychological models of humans’ approximate numerical abilities. A corpus analysis is presented which confirms the predictions of the present account. These findings add to other recent work demonstrating the relevance of measurement theoretic concepts to natural language meaning.

The A-map model: Articulatory reliability in child-specific phonology

Tara McAllister Byun, Steinhardt School of Culture, Education and Human Developmen; Sharon Inkelas, University of California and Yvan Rose, Memorial University of Newfoundland

This paper addresses a phenomenon of longstanding interest: the existence of child-specific phonological patterns which are not attested in adult language. We propose a new theoretical approach, termed the A(rticulatory)-Map model, to account for the origin and elimination of child-specific phonological patterns. Due to the performance limitations imposed by structural and motor immaturity, children’s outputs differ from adult target forms in both systematic and sporadic ways. The computations of the child’s grammar are influenced by the distributional properties of motor-acoustic traces of previous productions, stored in episodic memory and indexed in the eponymous A-map. We propose that child phonological patterns are shaped by competition between two essential forces: the pressure to match adult productions of a given word (even if the attempt is likely to fail due to performance limitations), and the pressure to attempt a pronunciation that can be realized reliably (even if phonetically inaccurate). These forces are expressed in the grammar by two constraints that draw on the motor-acoustic detail stored in the A-map. These constraints are not child-specific, but remain present in the adult grammar, although their influence is greatly attenuated as a wide range of motor plans come to be realized with a similar degree of reliability. The A-map model thus not only offers an account of a problematic phenomenon in development, but also provides a mechanism to model motor- grammar interactions in adult speech, including in cases of acquired speech impairment.

The role of indirect positive evidence in syntactic acquisition: A look at anaphoric one

Lisa S. Pearl and Benjamin Mis, University of California, Irvine

Language learners are often faced with a scenario where the data allow multiple generalizations, even though only one is actually correct. One promising solution to this problem is that children are equipped with helpful learning strategies that guide the types of generalizations made from the data. Two successful approaches in recent work for identifying these strategies have involved (i) expanding the set of informative data to include INDIRECT POSITIVE EVIDENCE, and (ii) using observable behavior as a target state for learning. We apply both these ideas to the case study of English anaphoric one, using computationally modeled learners who assume one’s antecedent is the same syntactic category as one and form their generalizations based on realistic data. We demonstrate that a learner who is biased to include indirect positive evidence coming from other pronouns in English can generate 18-month-old looking preference behavior. Interestingly, we find that the knowledge state responsible for this target behavior is a context-dependent representation for anaphoric one, rather than the adult representation, but this immature representation can suffice in many communicative contexts involving anaphoric one. More generally, these results suggest that children may be leveraging broader sets of data to make the syntactic generalizations leading to their observed behavior, rather than selectively restricting their input. We additionally discuss the components of the learning strategies capable of producing the observed behavior, including their possible origin and whether they may be useful for making other linguistic generalizations.

Short Report: On ergativity in Bumthang

Cathryn Donohue, The University of Hong Kong and Mark Donohue, The Australian National University

This report investigates the uses of the ergative case marker in transitive clauses in Bumthang, an East Bodish language of Bhutan. We discuss the conditions under which the ergative is required and show that a simple analysis involving multiple influences models the data, which we present in the form of a decision tree. Variable case marking has been shown to be determined by syntactic, semantic, or pragmatic factors, but in Bumthang we see that all these factors play a role in determining the use of the ergative case marker. We hope that our analysis may prove useful for understanding the variable uses of the ergative case marker in other Himalayan languages, as well as providing interesting challenges for formal models of case marking.

Individual-level contact limits phonological complexity: evidence from bunched and retroflex /ɹ/

Jeff Mielke, North Carolina State University; Adam Baker, University of North Dakota
and Diana Archangeli, University of Arizona and University of Hong Kong

We compare the complexity of idiosyncratic sound patterns involving American English /ɹ/ with the relative simplicity of clear/dark /l/ allophony patterns found in English and other languages.  For /ɹ/, we report an ultrasound-based articulatory study of 27 speakers of American English. Two speakers use only retroflex /ɹ/, 16 use only bunched /ɹ/, and nine use both /ɹ/ types, with idiosyncratic allophonic distributions. These allophony patterns are covert, because the difference between bunched and retroflex /ɹ/ is not readily perceived by listeners. We compare this typology of /ɹ/ allophony patterns to clear/dark /l/ allophony patterns in 17 languages. On the basis of the observed patterns, we show that individual-level /ɹ/ allophony and language-level /l/ allophony exhibit similar phonetic grounding, but that /ɹ/ allophony patterns are considerably more complex. The low complexity of language-level /l/ allophony patterns, which are more readily perceived by listeners, is argued to be the result of individual-level contact in the development of sound patterns. More generally, we argue that familiar phonological patterns (which are relatively simple and homogeneous within communities) may arise from individual-level articulatory patterns, which may be complex and speaker-specific, by a process of koineization. We conclude that two classic properties of phonological rules, phonetic naturalness and simplicity, arise from different sources.

Online-only: Historical Syntax
Growing syntax: the development of a DP in Northern Germanic
Kersti Börjars Mail, The University of Manchester
 
Grammaticalisation as standardly conceived is a change whereby an item develops from a lexical to a grammatical or functional meaning, or from being less to more grammatical. In this paper we show that this can only be part of the story; for a full account we need to understand the syntactic structures into which grammaticalising elements fit and how they too develop. To achieve this end we consider in detail the history of definiteness marking within the noun phrase in North Germanic, and in particular in Faroese. We show how this change requires us to distinguish between projecting and non-projecting categories, and how a category can emerge over time and only subsequently develop into a head with its own associated functional projection. The necessary structure, rather than being intrinsic to an aprioristic universal grammar, grows over time as part of the grammaticalisation process. We suggest that this in turn argues for a parallel correspondence theory of grammar such as the one adopted here, Lexical-Functional Grammar, in which different dimensions of linguistic structure can change at different rates.

Online-only: Phonological Analysis
Cumulativity and ganging in the tonology of Awa suffixes
Laura McPherson, Dartmouth College

This paper revives old descriptive data on Awa, a Papuan language of the Kainantu group. The tonal system was described in detail by Loving (1973), where he reports a series of toneless noun suffixes, falling into six classes depending on their tonal alternations when combined with a noun root. This paper demonstrates that the suffixes are best understood as carrying lexical tone; the alternations in form arise from the interaction of typologically natural phonotactic constraints. While the system can be described in autosegmental terms without much difficulty, a formal constraint-based analysis is less straightforward. I show that strict ranking, as in Optimality Theory (Prince and Smolensky 1993) fails to capture the data patterns due to cumulativity effects. The data are successfully modeled in Harmonic Grammar (Legendre et al. 1990).

Future Issues

Grammar and social agency: the pragmatics of impersonal deontic statements

Giovanni Rossi, Max Planck Institute for Psycholinguistics and Jörg Zinken, University of Portsmouth

Any grammatical system will provide its speakers with alternative sentence structures for different pragmatic functions. Among these, some languages include impersonal deontic constructions like ‘it is necessary to X’ (or NTX), which are used to assert the existence of a need or obligation without tying it to any particular individual. These constructions are, at first glance, pragmatically equivocal, in that they can have a range of functions including getting others to take on a task, rationalising the speaker’s own engagement in the task, or both of these together. But why would languages provide such a multifunctional structure? And how do people know which function the structure has in a given context? We address these questions using video recordings of informal interactions among speakers of Italian and Polish. Our analysis results in two findings. The first is that the pragmatic equivocalness of impersonal deontic statements is systematically resolved by, on the one hand, the relative responsibility of participants for the necessary task, and on the other hand by the nonverbal behaviour of the speaker at the time of the statement, which can increase or decrease the chances of another’s involvement in the task. The second finding is that, although the context and the nonverbal behaviour of the speaker can influence their function, impersonal deontic statements generally afford an open response space, where participation in the necessary task may be shared, negotiated, or avoided. This leads us to reassess the pragmatic equivocalness of an impersonal deontic statement as useful ambiguity. By virtue of its grammar, the statement has the potential to both mobilise and legitimise an act by different participants in a speech event, which makes it a versatile tool for the management of social agency.

Old English *motan, variable-force modality, and the presupposition of inevitable actualization

Igor Yanovich, Universität Tübingen

Old English *motan and Middle English *moten, the ancestors of modern must, are commonly described as ambiguous between a possibility and a necessity reading. I argue instead that in the Alfredian Old English prose, *motan was a non-ambiguous, “variable-force” modal with the modal force different from both possibility and necessity. I propose that *motan’s variable-force effect was due to the presupposition of a collapse between possibility and necessity. Informally, motan(p) presupposed “if p gets a chance to actualize, it will”. I then trace the development of *motan into a modal genuinely ambiguous between necessity and possibility in Early Middle English.

Short Report: Repair organization in Chinantec whistled speech

Mark A Sicoli, Georgetown University

Recent work on language in interaction has aspired to incorporate more communicative modalities, like the visual and tactile, into the analysis of talk. This short report however points out that studying stripped down communicative channels can also help inform our understanding of human language. Here I describe how conversational repair is organized in the whistled speech register of San Pedro Sochiapan Chinantec of Oaxaca, Mexico. While repair in whistled speech shares the same universal sequence organization as repair in spoken speech, the affordances of the linguistically reduced channel result in drastic differences in the preference organization and the typology of repair initiations in whistled speech. I propose that these patterned differences are motivated by an organizing principle where single whistled turns are preferentially constructed to enact only one communicative action and introduce the concept of a semiotic carrying capacity for guiding studies of language in different interactional environments.

Reactive effort as a factor that shapes sign language lexicons

Nathan Sanders and Donna Jo Napoli, Swarthmore College

Many properties of languages, including sign languages, are not uniformly distributed across the lexicon. Some of this distribution can be accounted for by appeal to articulatory ease, with easier articulations being found more frequently in the lexicon than more difficult articulations. We define a previously unstudied aspect of articulatory ease, reactive effort (the effort of resisting incidental movement elsewhere in the body induced by an articulation), which we distinguish from active effort (the extensively-studied effort internal to the articulation itself). Because reactive effort is needed to resist incidental movement of the torso induced by path movement of the manual articulators in sign languages, we argue that reactive effort has a significant effect on which types of manual movements are more or less common. We support this argument with evidence from the lexicons of Italian Sign Language, Sri Lankan Sign Language, and Al-Sayyid Bedouin Sign Language.  This work adds to a growing body of functionalist literature which claims that some linguistic patterns are extralinguistic in origin, emerging from more fundamental factors of the physical world.

Learnability shapes typology: the case of the midpoint pathology

Juliet Stanton, Massachusetts Institute of Technology

The midpoint pathology (in the sense of Kager 2012) characterizes a type of unattested stress system in which the stressable window contracts to a single word-internal syllable in some words, but not others. Kager (2012) shows that the pathology is a prediction of analyses employing contextual lapse constraints (e.g. *ExtLapseR; no 000 strings at the right edge), and argues that the only way to avoid it is to eliminate these constraints from Con. This paper explores an alternative: that systems exhibiting the midpoint pathology are unattested not because the constraints that would generate them are absent from Con, but because they are difficult to learn. This study belongs to a growing body of work exploring the idea that phonological typology is shaped by considerations of learnability (e.g. Boersma 2003, Blevins 2004, Alderete 2008, Heinz 2009).

Modification of Stative Predicates

Thomas Ernst, University of Massachusetts, Amherst, and Dartmouth College

Manner and locative expressions modifying stative predicates, as in own (something) honestly and (be) quiet in the car, are rare compared to those modifying dynamic predicates, and it has been claimed (for example by Maienborn 2005 and Katz 2008) that they are systematically excluded on semantic grounds. I argue here that this is not so: in fact they are perfectly acceptable once the restrictions on them are understood. I propose further that these restrictions take the form of (a) a pragmatic condition that generally bans locative modification of stative predicates, but which may be overridden in certain defined contexts, and (b) regular semantic incompatibilities between adverbs and stative predicates, which, being semantically ‘impoverished,’ have relatively few modifiable semantic features compared to dynamic predicates. These proposals are supported by extensive examples. The conclusions indicate that there is no need to treat states as fundamentally different from other eventualities, whether by invoking Kimian states or by avoiding eventuality variables altogether in their representations.

On the cognitive basis of contact-induced sound change: Vowel merger reversal in Shanghainese

Yao Yao, The Hong Kong Polytechnic University and Charles B. Chang, Boston University

This study investigated the source and status of a recent sound change in Shanghainese (Wu, Sinitic) that has been attributed to language contact with Mandarin. The change involves two vowels, /e/ and /ɛ/, reported to be merged three decades ago but produced distinctly in contemporary Shanghainese. Results of two production experiments showed that speaker age, language mode (monolingual Shanghainese vs. bilingual Shanghainese-Mandarin), and crosslinguistic phonological similarity all influenced the production of these vowels. These findings provide evidence for language contact as a linguistic means of merger reversal and are consistent with the view that contact phenomena originate from cross-language interaction within the bilingual mind.

The discourse basis of ergativity revisited

Geoffrey Haig, University of Bamberg and Stefan Schnell, The University of Melbourne

Since Du Bois’ (1987b) seminal paper, ergative alignment of morphological argument encoding (S=P vs. A) has been claimed to correlate with a characteristic constellation in discourse: both S and P arguments serve to introduce new referents into discourse via full NPs, while A arguments are dispreferred for this function, and are thus mostly realised as pronouns or zero (cf. Du Bois 1987b; Dixon 1995; Du Bois 2003a, 2003b; Goldberg 2004). These claims have recently been questioned by Haspelmath (2006) and Everett (2009), who point to a number of empirically and conceptually unresolved issues. The present paper draws on the largest data base yet compiled (19 spoken language corpora from 15 typologically diverse languages) with the aim of testing the claim for an ergative bias in discourse, and the explanations that have been advanced to account for it, in a more rigorous manner. We find that with the exception of Du Bois’ original Sakapultek data, there is little evidence for the postulated S=P (‘ergative’) pattern crosslinguistically. As for the robustly attested low lexicality levels of A, we confirm Everett’s (2009) findings that the semantic feature of [+/–human] provides an empirically sounder and conceptually more economical account than earlier explanations framed in terms of information management. Finally, we assess the plausibility of emergentist claims regarding ergative patterning in discourse as a diachronic source of ergative alignment in morphosyntax.

The evolution of medial /t/ over real and remembered time

Jennifer Hay, New Zealand Institute of Language Brain and Behaviour, and University of Canterbury and Paul Foulkes, University of York

This paper follows a change in pronunciation of word-medial intervocalic /t/ in New Zealand English, as it unfolds over 120 years. Data are analysed in the context of questions about the role of experience-based lexical representations and their potential impact on the time-course of sound change in progress. Three major results are reported. First, frequent words lead the change. Second, the distributions of individual words affect their participation in the change: words favoured by younger speakers are produced with newer variants. Finally, the topic of conversation affects which variant is favoured: older topics elicit older variants. Together, these findings provide evidence that phonetic distributions of word-level representations are implicated in the course of sound change.

Asymmetries in the representation of categorical phonotactics

Gillian Gallagher, New York University

A comparison of speakers’ treatment of two categorically unattested phonotactic structures in Cochabamba Quechua reveals a stronger grammatical prohibition on roots with pairs of ejectives, *[k’ap’u], than on roots with a plain stop followed by an ejective, *[kap’u]. While the distribution of ejectives can be stated as a single restriction on ejectives preceded by stops (ejective or plain), *[-cont, -son][cg], speakers’ show evidence of having learned an additional constraint that penalizes cooccurring ejectives more harshly, *[cg][cg]. An inductive learning bias in favor of constraints with the algebraic structure of *[cg][cg] is hypothesized (Berent et al. 2002; Berent et al. 2012; Marcus 2001), allowing such constraints to be preferred by learners over constraints like *[-cont, -son][+cg], which penalize sequences of unrelated feature matrices.