On the Origins of Phonology

The Handbook of Language Emergence, 2015

Language, 1999

Rule Opacity 130 5.2 Opacity of Phonetic Motivation 132 5.3 Opacity by Parasites 135 6.1 Number of Syllables per Word (General Corpus) 153 6.2 Number of Syllables per Word (Neologisms) 153 6.3 Stressed Syllables (General Corpus) 155 6.4 Stressed Syllables (Neologisms) 155 6.5 Voicing in Word-Initial Position (General Corpus) 157 6.6 Voicing in Word-Initial Position (Neologisms) 157 6.7 Distribution of Active Articulators in Word-Initial Position (General Corpus) 158 6.8 Distribution of Active Articulators in Word-Initial Position (Neologisms) 159 6.9 The Favoring of Visible Phonemes in Word-Initial Position (General Corpus) 160 6.10 The Favoring of Explosive Phonemes in Word-Initial Position (General Corpus) 161 6.11 The Favoring of Explosive Phonemes in Word-Initial Position (Neologisms) 161 6.12 The Favoring of Apical Phonemes in Word-Final Position (General Corpus) 162 6.13 Percentage of Active Articulators in Word-Initial and Word-Final Position (General Corpus) 162 6.14 The Favoring of Apical Phonemes in Word-Final Position (Neologisms) 163 6.15 Percentage of Active Articulators in Word-Initial and Word-Final Position (Neologisms) Preface therapy, and audiology. During the last decade, we have attempted to apply the theory of phonology as human behavior to all aspects of the speech clinic. We are not attempting to sell the theory of phonology as human be havior as the best or the ultimate theory of phonology, nor do we view it as a panacea for all linguists and speech and hearing clinicians. This volume merely documents our attempts to apply a specific approach to language in general and to phonology in particular to both language and languages as well as to another area of linguistic research-clinical phonology and audiology. At this initial stage of our research, we basically view both the theory and its clinical applications as being reasonably well founded empirically and potentially promising. Like all students of language and linguistics who optimistically attempt to apply linguistic theories to the clinic, we have a long way to go. The answers-if they ever come-will come only when we know more about the complexities of the human brain, human development and cognition, and human behavior. We view our research as a tenuous first step in achieving a better understanding of these mysteries.

Language Sciences, 2009

Since the before publication of The Sound Pattern of English . The Sound Pattern of English. Harper & Row, New York], phonological theory has been struggling with finding ways to express both the surface and the deep phonological generalizations that are inherent in language structure. In the 1970's, Kisseberth and Kenstowicz, recognizing that rules often ''conspired" to maintain phonotactic constraints, pointed out the descriptive value of representing both rules and constraints explicitly, at the cost of redundancy in the formalism. However, phonological theory has continued to choose to make either rules or constraints easy to apprehend, at the cost of obscuring the other. This paper presents a major extension of the exemplar-based approach of radical templatic theory [Vihman, M.M., Croft, W., 2007. Phonological development: toward a 'radical' templatic phonology. Linguistics 45, 683-725] and new data from early phonological acquisition, and supports the psycholinguistic appropriateness of frequency-sensitive representations. In order to account insightfully for the striking regularities as well as the irreducible irregularities of early phonology, our Linked Attractor model maintains both underlying and output representations for each lexical entry, and additionally posits both input-to-output rules and feedback (output-to-input) rules to account for patterns of fossilization and developmental progress. Our long-term goal is to make the Linked Attractor model explicit enough for statistical modeling and testing, and to offer it to phonological theory as the basis for a psycholinguistically viable model for adult phonology as well.

Current Anthropology, 2010

1976

This paper is concerned with the-Aristotelian notion of nuniversalli as applied to phonological phenomena. It is claimed ,that speech production in children and adults, in normal end deviantspeakers, and in a variety of languages, can all be.descrO.bed according to the same universal phonological rules which 'constitute 'the universal process of grammar optimalization, that is, theprocess 'bf working toward the replication of some.standard adult scdel. For diverse reasons, the linguistically deviant fail to optimalize their grammars. It is concluded that the:phonological rules of language acquisition are universal and finite:in number, that these same rules are found in deviant language acquisitlon, that they are:also found in adult dissolution of language-facility, and that.they are the motivation for the diachronic phenomena studied by historical : linguists since the time of Bask, Grimm, Bopp, Pott, and Schleicher.

From the middle of the 20th century until now the relations between motor processes and language development have been discussed. This problem is discussed as related to both normal and deviant speech and language development. At the same time it may be helpful for solving the riddle of language origin.

This paper will discuss the ways in which Cognitive Grammar (CG) has integrated the fundamental concepts of phonology. Phonology has traditionally been the neglected stepchild of CG, in part because the initial excitement of CG revolved around the insightful semantic analyses of what had previously been thought to be purely syntactic or arbitrary lexical puzzles, and phonology is, by definition, about meaningless units. Additionally, phonology deals with the coordination of motor activity and auditory perception, areas that initially didn’t seem to lend themselves to the conceptual tools of CG. Unlike much work on syntax and semantics, research within several of the generative and functional traditions turns out to be adaptable to CG’s view of the nature of phonological processing. The primary source of useful insights is the work of the Natural Phonologists (Donegan and Stampe) who argued for non-modular cognitive realism and against Chomskyan innatism independently from CG theorists. Furthermore, some work by recent offshoots of Generative Phonology turn out to have useful things to say about how CG might understand the structure of sounds in language. The earliest writing on the subject (Nathan) examined ways in which the core categorization concepts in early CG (radial prototype categories, image schema transformations, basic level categorization) could be utilized to explain traditional phonological concepts such as the phoneme, allophones and (natural) phonological processes. In more recent research, the conceptual tools of usage-based models have also been used to account for some aspects of phonological behavior, leading some researchers to question the relevance even of such traditional phonological constructs as the phoneme/allophone contrast (Bybee) or the distinction between abstract phonemic representation and representations of individual instances of particular utterances (Pierrehumbert). Additionally, the question of the boundary between phonology and morphology has been raised, with some researchers (such as the present author) arguing for a sharp, functional boundary based on the varying cognitive and physiological resources involved, while others (Nesset, Kemmer) arguing for a more generalized schematization model covering all aspects of phonological as well as morphological and syntactic generalizations. After reviewing relevant issues this paper argues that evidence from early aspects of child language acquisition, such as the onset and development of babbling (MacNeilage), the embodied nature of perception (Johnson), and other research on the acquisition and processing of complex motor skills, shows that phonology requires a more active conception of the storage and production of stored heard instances. Phonology is based upon speakers’ knowledge of the nature and capabilities of their vocal tracts and deals with how speakers actively construct utterances based on their knowledge of their individual language(s)’ conventionalized responses to those physiological and acoustic constraints.

2012

Journal of Psycholinguistic Research, 2017

Across languages, certain syllable types are systematically preferred to others (e.g., blif bnif bdif lbif, where indicates a preference). Previous research has shown that these preferences are active in the brains of individual speakers, they are evident even when none of these syllable types exists in participants' language, and even when the stimuli are presented in print. These results suggest that the syllable hierarchy cannot be reduced to either lexical or auditory/phonetic pressures. Here, we examine whether the syllable hierarchy is due to articulatory pressures. According to the motor embodiment view, the perception of a linguistic stimulus requires simulating its production; dispreferred syllables (e.g., lbif) are universally disliked because their production is harder to simulate. To address this possibility, we assessed syllable preferences while articulation was mechanically suppressed. Our four experiments each found significant effects of suppression. Remarkably, people remained sensitive to the syllable hierarchy regardless of suppression. Specifically, results with auditory materials (Experiments 1-2) showed strong effects of syllable structure irrespective of suppression. Moreover, syllable structure uniquely accounted for listeners' behavior even when controlling for several phonetic characteristics of our auditory materials. Results with printed stimuli (Experiments 3-4) were more complex, as participants in these experiments relied on both phonological and graphemic information. Nonetheless, readers were sensitive to most of the syllable hierarchy (e.g., blif bnif bdif), and these preferences emerged when articulation was suppressed, and even when the statistical properties of our materials were controlled via a regression analysis. Together, these findings indicate that speakers possess broad grammatical preferences that are irreducible to either sensory or motor factors.

Journal of Linguistics, 1997

Reviewed by P D, University of Hawai'i at Ma, noa Recent interest in language learnability has prompted interaction between acquisition studies and syntactic theory ; Phonological acquisition and phonological theory aims to provide a similar interchange between acquisition studies and phonology. It includes nine chapters, plus a preface and introduction by the editor, a unified author index, and a subject index. The volume is attractively printed, and it is generally free from typographical errors." The authors share a clear idea of what the object of phonological acquisition is ; their theoretical framework is that of generative phonology, with a focus on autosegmental, featuregeometric, underspecified, and prosodic representations. The volume is a much-needed attempt to interrelate the representations of modern phonological theory, the principles and parameters approach to acquisition, and the problems encountered by a child acquiring a first-language phonology, but the chapters reveal some difficulties in achieving the desired synthesis. John Archibald's introduction describes phonology as a system of interconnected levels of representation which also includes processes that map one type of representation onto another. His emphasis on representations and his view of phonology as a system entirely independent of phonetics are consistent with the rest of the volume. Elan Dresher & Harry van der Hulst discuss ' global determinacy '-their term for the cases where a single phonetic form corresponds to multiple possible phonological representations and a ' global ' knowledge of the language is required to determine a phonological representation. They cite various sources of global determinacy in modern phonological theory, including abstract underlying representations, floating material, underspecification, and variable grouping and dependency. These are problems of learnability ' because a lot must be known about the phonology of specific languages in order to determine which representation is adequate ' (). They note that much modern phonology assumes that the main burden of explanation rests with representations, not rules, and that the rules should follow from the segmental inventory, and they add that other hypotheses are implied : that representations can be established before any rules have been learned, and that representations increase in complexity by adding contrast. Dresher and van der Hulst formulate questions of learnability which phonological analyses should answer, but they offer no solutions here. The authors who deal with segmental acquisition (Keren Rice & Peter Avery, David Ingram, E. Jane Fee, Daniel A. Dinnsen & Steven B. Chin) almost uniformly attribute children's production errors to limitations on their phonological representations. However, they do not reconcile these limited representations with children's perceptual accuracy, or interpret what this accuracy implies about children's phonologies. Some of the authors (like Rice & Avery) disregard this problem ; others (Dinnsen & Chin) take it seriously. Most assume that adults' and children's phonological representations are underspecified, but none actually present any evidence from acquisition data for underspecification. In particular, no evidence is provided for the sorts of perceptual confusion that underspecification would be expected to cause. Underspecification is accepted here despite the fact that it makes incorrect predictions about children's perceptual abilities, differential phonological effects of sounds that children pronounce the same, and the nature of changes in children's productions. A child with representations which lack information regarding Place, for example, should fail to distinguish tea from key perceptually, but does not. Identical (underspecified) representations should not be treated differently by phonological rules, but they often are treated differently. And correction of an error pattern should correct inappropriate forms as well as the appropriate forms if their [] One surprising error, however, () involves ' McCawley () following Hayes … , p. ) ' in his definition of the mora-when, of course, it is Hayes who followed McCawley.     underlying representations do not differ, but this rarely happens (for example, in a child who says [t] for adult [k], [t] and [k] are both supposedly unspecified for Place, but when the child learns to pronounce velars, only the adult-[k] words change to [k]). These indications regarding children's representations have been pointed out repeatedly (Stampe , , Smith , Barton , etc.). Daniel A. Dinnsen & Steven B. Chin are to be credited for recognizing and describing these problems. But, having criticized underspecification, Dinnsen & Chin make a surprising attempt to save it with a theory of ' shadow specification '. They say that, for any merged pair (or series), the segments the child produces correctly are specified for the distinguishing feature, and the segments the child produces incorrectly are ' shadow ' specified for the distinguishing feature (for example [t] is unspecified for Place, and [k] is specified for Place as [Coronal]). This proposal is hard to swallow : the child's feature specification reflects his or her own pronunciation,      . That is, the child is said to specify a segment as coronal only when he or she knows it to be -coronal. Further, no basis is provided for the regularity of children's mispronunciations-since they are not determined by universal processes or by redundancy rules, we still need an explanation of the fact that a child specifies adult velars  as coronals, and not randomly as coronals, labials, glottals, etc. The proposal also predicts that a child who produces a three-way merger (such as [t] for [p, t, k] or [t] for [t, H, s] would mark both ' erroneous ' segments with the same wrong feature ([Coronal] or [kContinuant], respectively). When a new articulation is required, a change rule that affected the mis-specified segments would necessarily change them all to the newly acquired sound-which does not always happen. When the only proposal in the volume which confronts the problems of underspecification has these effects, one may well look again at a possibility Dinnsen & Chin reject : that the child's underlying forms are accurately specified and that articulatory constraints cause substitutions that prevent their accurate realization. Prosodic and tonal issues are not neglected : Jane Fee provides a discussion of how UG rules for building melodic and prosodic structures may dictate the shape of children's early productions, John Archibald surveys the acquisition of stress in first and second languages, focusing on whether stress is learned lexically or by rule and whether there are default parameter settings regarding foot type, extrametricality, etc., and Katherine Demuth presents a case study of the acquisition of some morphologically conditioned tone rules in Sesotho. Both Fee and Archibald propose a bisyllabic trochee as the unmarked metrical unit, instead of a bimoraic trochee, on the grounds that open syllables are more common in many children's productions ; neither considers early monosyllables with long vowels as potential evidence for a bimoraic unit. On the second-language acquisition front, Ellen Broselow and Hye-Bae Park present an interesting analysis of vowel epenthesis in the Korean pronunciation of English words, which considers Korean speakers' perception and production in terms of parameter settings, and Thomas Scovel offers a chapter on the discrimination of foreign accents. The learnability questions that Dresher & van der Hulst raise, and the objections to underspecification outlined by Dinnsen & Chin are important problems for phonological theory and for acquisition, and it is not clear how they can be resolved. Questions about the relationship between perception and representation-how a phonetic signal is processed into a phonological representation-go unanswered (Rice & Avery) or are insufficiently answered (Ingram, Archibald). And important aspects of the relationship between phonological representation and production are glossed over (as in Fee's claim that although phonology constrains production, ' lexically learned ' forms may be produced which violate UG constraints on phonological form ()). Through much of the volume, there is an assumption (stated most clearly by Ingram ()) that the principles that govern the acquisition of phonology should be the same as those which govern the acquisition of syntax. Some learning principles may, of course, be common to both, but this assumption overlooks the more direct relation of phonology to the learner's perception and production abilities-to phonetics. The authors hold the learner's phonological system responsible for his production errors, but the features, representations, prosodic structures, and rules of this system are unconnected to their phonetic realizations, where many of the learner's difficulties lie. Default parameter settings and the redundancy rules for default feature specification are simply attributed to UG, as if no articulatory or perceptual explanation were involved. Phonetic inabilities (absence of particular phonetic skills) are not even considered as a potential source of negative phonological constraints. This isolation of phonology from phonetics is assumed without discussion. Phonological theory and studies of language acquisition often seem unconnected ; this book    works from the phonology side to bridge that gap. Most of the chapters are attempts to show how the segmental, prosodic or tonal systems of a current model of phonology can be acquired. The attempt is welcome, but the bridge is incomplete ; many difficulties regarding the connection remain.

2015

Phonology is traditionally defined as the description of the grammatical competence of native speakers concerning the sound patterns of their language (Chomsky and Halle 1968). The field has subsequently been broadened to account for the abstract properties of phonology that are not specific to spoken language, but extend to the manual-visual modality in sign languages. A lot of attention has thus been devoted to the properties of the sensory-motor system of grammar - i.e. phonology - that are not specific to the auditory modality, that is, to spoken language, but extend to the visual modality. Abundant literature has explored this issue and showed that Sign Languages share many grammatical properties with spoken languages (among others, Brentari 1998; Nespor and Sandler 1999, Wilbur 1999; Sandler and Lillo-Martin 2006). This body of research has shown that parallel segmental and suprasegmental representations exist in the two linguistic modalities -- the auditory and the visual. DO...

2001

This paper discusses diverse aspects of the acquisition of phonology. These topics are examined using a wide range of tasks and experimental paradigms across different ages. Various levels of processing and representation are thus involved. The goal is that such data can be coherently interpreted only within a particular informationprocessing model. it first present the basic architecture of a model of speech perception and production, justifying it with psycholinguistic and neuropsychological data. then use this model to interpret data from the target articles relative to the acquisition of phonology