Linguistik-Klassifikation
Refine
Year of publication
Document Type
- Part of a Book (100) (remove)
Language
- English (100) (remove)
Has Fulltext
- yes (100)
Is part of the Bibliography
- no (100)
Keywords
- Phonologie (44)
- Intonation <Linguistik> (23)
- Prosodie (19)
- Phonetik (16)
- Artikulation (15)
- Deutsch (15)
- Optimalitätstheorie (13)
- Artikulatorische Phonetik (9)
- Bantusprachen (9)
- Relativsatz (9)
Institute
- Extern (1)
Símákonde is an Eastern Bantu language (P23) spoken by immigrant Mozambican communities in Zanzibar and on the Tanzanian mainland. Like other Makonde dialects and other Eastern and Southern Bantu languages (Hyman 2009), it has lost the historical Proto-Bantu vowel length contrast and now has a regular phrase-final stress rule, which causes a predictable bimoraic lengthening of the penultimate syllable of every Prosodic Phrase. The study of the prosody / syntax interface in Símákonde Relative Clauses requires to take into account the following elements: the relationship between the head and the relative verb, the conjoint / disjoint verbal distinction and the various phrasing patterns of Noun Phrases. Within Símákonde noun phrases, depending on the nature of the modifier, three different phrasing situations are observed: a modifier or modifiers may (i) be required to phrase with the head noun, (ii) be required to phrase separately, or (iii) optionally phrase with the head noun.
We measure face deformations during speech production using a motion capture system, which provides 3D coordinate data of about 60 markers glued on the speaker's face. An arbitrary orthogonal factor analysis followed by a principal component analysis (together called a guided PCA) of the data has showed that the first 6 factors explain about 90% of the variance, for each of our 3 speakers. The 6 derived factors, therefore, allow us to efficiently analyze or to reconstruct with a reasonable accuracy the observed face deformations. Since these factors can be interpreted in articulatory terms, they can reveal underlying articulatory organizations. The comparison of lip gestures in terms of data derived factors suggests that these speakers differently maneuver the lips to achieve contrast between /s/ and /R/. Such inter-speaker variability can occur because the acoustic contrast of these fricatives is shaped not only by the lip tube but also by cavities inside the mouth such as the sublingual cavity. In other words, these tube and cavity can acoustically compensate each other to produce their required acoustic properties.
Consonants exhibit more variation in their phonetic realization than is typically acknowledged, but that variation is linguistically constrained. Acoustic analysis of both read and spontaneous speech reveals that consonants are not necessarily realized with the manner of articulation they would have in careful citation form. Although the variation is wider than one would imagine, it is limited by the phoneme inventory. The phoneme inventory of the language restricts the range of variation to protect the system of phonemic contrast. That is, consonants may stray phonetically into unfilled areas of the language's sound space. Listeners are seldom consciously aware of the consonant variation, and perceive the consonants phonemically as in their citation forms. A better understanding of surface phonetic consonant variation can help make predictions in theoretical domains and advances in applied domains.
A visual articulatory model and its application to therapy
of speech disorders : a pilot study
(2005)
A visual articulatory model based on static MRI-data of isolated sounds and its application in therapy of speech disorders is described. The model is capable of generating video sequences of articulatory movements or still images of articulatory target positions within the midsagittal plane. On the basis of this model (1) a visual stimulation technique for the therapy of patients suffering from speech disorders and (2) a rating test for visual recognition of speech movements was developed. Results indicate that patients produce recognition rates above level of chance already without any training and that patients are capable of increasing their recognition rate over the time course of therapy significantly.
In Nłeʔkepmxcin, consonant-heavy inventories, lengthy obstruent clusters and widespread glottalization can make potential F0 cues to prosodic phrase boundaries (e.g. boundary tones or declination reset) difficult to observe phonetically. In this paper, I explore a test that exploits one behaviour of phrasefinal consonant clusters to test for prosodic phrasing in Nłeʔkepmxcin clauses. Final /t/ of the 1pl marker kt is aspirated when phrase-final, but not phraseinternally. Use of this test suggests that Thompson Salish speakers parse verbs, arguments and adjuncts into separate phonological phrases. However, complex verbal predicates and complex noun phrases are parsed as single phonological phrases. Implications are discussed, especially in regards to findings that (absence of) pitch accent is not employed to signal the informational categories of Focus and Givenness, even though Nłeʔkepmxcin is a stress language.
This paper examines how questions, both Wh-questions and yes-no questions, are phrased in Chimwiini, a Bantu language spoken in southern Somalia. Questions do not require any special phrasing principles, but Wh-questions do provide much evidence in support of the principle Align-Foc R, which requires that focused or emphasized words/constituents be located at the end of a phonological phrase. Question words and enclitics are always focused and thus appear at the end of a phrase. Although questions do not require any new phrasing principles, they do display complex accentual (tonal) behavior. This paper attempts to provide an account of these accentual phenomena.
We focus in this paper on two prosodic phenomena in Chimwiini: vowel length and accent (or High tone). Vowel length is determined in part by a lexical distinction between long and short vowels, and also by various morphophonemic processes that derive long vowels. Accent is penult in the default case, but final under certain morphosyntactic conditions. In order to account for the distribution of vowel length and the location of accents in a Chimwiini sentence, it is necessary to segment sentences into a sequence of phonological phrases. This paper examines the phonological phrasing of both canonical relative clauses and what we refer to as "pseudo-relative" clauses. An account of relative clause phrasing is of critical importance in Chimwiini due to the extensive use of pseudo-relatives in the language. Close examination of the pseudo-relatives reveals that their phrasing is not exactly the same as the phrasing of canonical relative clauses.
In this paper, I argue that this apparent problem is accounted for by the interaction of constraints. For the fixed segment [ɛ] in Cɛ-reduplication, I argue that [ɛ] is the second least marked vowel in Palauan, which appears when the default vowel [ǝ] cannot appear. I show that the Palauan facts are not only consistent with the proposals of Urbanczyk (1999) and Alderete et. al (1999), but they actually provide support of their claims. In the following section, I discuss Urbanczyk's (1999) arguments concerning ROOT faithfulness in reduplication and possible asymmetries between affix reduplicants and root reduplicants. In Section 3, I introduce Palauan reduplication and discuss Finer's (1986) observations on the resulting state verb (RSV) form. I show that the RSV forms support the classification that Cɛ-reduplicants are affixes, and CVCV -reduplicants are roots. In Section 4, I discuss the shape and vowel quality of the two reduplicants. The CVCV-reduplicant has three variants: CǝCǝ, CǝC and CV. I explain this variation, illustrating why [ǝ] appears in the first two variations. Then, I discuss the shape and vowel quality of the Cɛ-reduplicant, arguing that the fixed segment [ɛ] in Cɛ-reduplication is a special case of TETU. I show that root faithfulness constraints are crucial in determining the shape and vowel quality of the reduplicants. Section 5 is the conclusion.
We present the results of an experimental study which targets prosodic correlates of subclausal quotation marks. We found that written sentences containing passages enclosed by quotation marks were read aloud in a manner that significantly differs in prosody from spoken realizations of corresponding disquoted counterparts. However, we also observed that such prosodic marking of subclausal quotation wasn't strong enough to survive subsequent back-translation into written language: there was no correlation between the presence/absence of quotation marks in the original written examples, and the presence/absence of quotation marks in corresponding back-translations from oral renditions. We investigated three different kinds of uses of quotation marks and found no systematic difference between them with respect to prosodic marking.
The aim of this paper is to try to explain how the Tooro system, which phonologically lacks tone, has come into being, by examining comparatively the tone system of each language itself and also by closely looking at the differences which exist among the Haya, Ankole and Nyoro systems (Kiga data insufficient) in order to look for phonetic reasons of the tone changes.
The purpose of this paper is to provide a unified (i.e. independent of lexical categories) account of Persian stress. I show that by differentiating word- and phrase-level stress rules, one can account for the superficial differences exemplified in (1) above and many of the stipulations suggested by previous scholars. The paper is organized as follows. In section 1, I look at nouns and adjectives and propose a rule that would account for their stress pattern. In section 2, I extend the stress rule to verbs and show the problem this category poses to our generalization. The main proposal of this paper is discussed in section 3. I introduce the phrasal stress rule in Persian and show that by differentiating word-level and phrase-level stress rules, one can come to a unified account of Persian stress. Section 4 deals with some problematic eases for the proposed generalization and discusses some tentative solutions and their theoretical consequences. Section 5 concludes the paper.
The current paper explores these two sorts of phonetic explanations of the relationship between syllabic position and the voicing contrast in American English. It has long been observed that the contrast between, for example, /p/ and /b/ is expressed differently, depending on the position of the stop with respect to the vowel. Preceding a vowel within a syllable, the contrast is largely one of aspiration. /p/ is aspirated, while /b/ is voiceless, or in some dialects voiced or even an implosive. Following a vowel within a syllable, both /p/ and /b/ both tend to lack voicing in the closure and the contrast is expressed largely by dynamic differences in the transition between the previous vowel and the stop. Here, vowel and closure duration are negatively correlated such that the /p/ has a shorter vowel and longer closure duration. This difference is often enhanced by the addition of glottalization to /p/. In addition to these differences, there are additional differences connected to higher-level organization involving stress and feet edges. To make the current discussion more tractable, we will restrict ourselves to the two conditions (CV and VC) laid out above.
The purpose of this paper is to show how WH questions interact with the complex tonal phenomena which we summarized and illustrated in Hyman & Katamba (2010). As will be seen, WH questions have interesting syntactic and tonal properties of their own, including a WH-specific intonation. The paper is structured as follows: After an introduction in §1, we successively discuss non-subject WH questions (§2), subject WH questions (§3), and clefted WH questions (§4). We then briefly present a tense which is specifically limited to WH questions (§5), and conclude with a brief summary in §6.
"The documentation of... descriptive generalizations is sometimes clearer and more accessible when expressed in terms of a detailed formal reconstruction, but only in the rare and happy case that the formalism fits the data so well that the resulting account is clearer and easier to understand than the list of categories of facts that it encodes.... [If not], subsequent scholars must often struggle to decode a description in an out-of-date formal framework so as to work back to... the facts.... which they can re-formalize in a new way. Having experienced this struggle often ourselves, we have decided to accommodate our successors by providing them directly with a plainer account." (Akinlabi & Liberman 2000:24)
I argue in this study that consonantal strength shifts can be explained through positional bans on features, expressed over positions marked as weak at a given level of prosodic structure, usually the metrical foo!. This approach might be characterized as "templatic" in the sense it seeks to explain positional restrictions and distributional patterns relative to independently motivated, fixed prosodic elements. In this sense, it follows Dresher & Lahiri's (1991) idea of metrical coherence in phonological systems, namely, "[T]hat grammars adhere to syllabic templates and metrical patterns of limited types, and that these patterns persist across derivations and are available to a number of different processes ... " (251). [...] The study is structured as follows: section 1 presents a typology of distributional asymmetries based on data from unrelated languages, demonstrating that the stress foot of each of these languages determines the contexts of neutralization and weakening of stops. Section 2 elaborates the notion of a template, exploring some of its formal properties, while section 3 presents templatic analyses of data from English and German. Section 4 explores the properties of weak positions, especially weak onsets, in more detail, including discussion of templates in phonological acquisition. Section 5 summarizes and concludes the study.
In order to investigate the articulatory processes of the hasty and mumbled speech of clutterers, the kinematic variability was analysed by means of electromagnetic midsagittal articulography (EMMA). In contrast to stutterers, clutterers improve their intelligibility by concentrating on their speech task. Variability is an important criterion in comparable studies of stuttering and is discussed in terms of the stability of the speech motor system. The aim of the current study was to analyse the spatial and temporal variability in the speech of three clutterers and three control speakers. All speakers were native speakers of German. The speech material consisted of repetitive CV-syllables and foreign words, because clutterers have the most severe problems with long words which have a complex syllable structure. The results showed a higher quotient of variation for clutterers in the foreign word production. For the syllable repetition task, no significant differences between clutterers and controls were found. The extremely large and variable displacements were interpreted as a strategy that helps clutterers to improve the intelligibility of their speech.
We show that wh-words are a tool to investigate the prosodic structure of Bàsàa. Our claim is that the end of an Intonation Phrase (IP) can be identified by the presence of a long vowel on the wh-word. We propose that wh-words, which sometimes surface as C´V and sometimes as C´V´V, are underlyingly of the C´V form and they introduce a floating H. Whenever the association of this floating H with the first tone bearing unit that follows the wh-word is prevented by the presence of an IP boundary, a mora is created on the wh-word in order to realize the floating H. We briefly discuss the interface approach of Immediately After the Verb (IAV) focus (Costa and Kula, 2008) and we show that Bàsàa wh-questions and answers do not support this hypothesis. Finally, Bàsàa fronted whphrases, just like Hausa’s fronted foci (Leben et al., 1989), seem to provide support to the idea that intonational effects are also at play in the present tone language.
This study reports on the results of an airflow experiment that measured the duration of airflow and the amount of air from release of a stop to the beginning of a following vowel in stop vowel-sequences of German. The sequences involved coronal, labial and velar voiced and voiceless stops followed by the vocoids /j, i:, ı, ɛ, ʊ, a/. The experiment tested the influence of the three factors voicing of stop, place of stop articulation, and the following vocoid context on the duration and amount of air as possible explanation for assibilation processes. The results show that the voiceless stops are related to a longer duration and more air in the release phase than voiced ones. For the influence of the vocoids, a significant difference could be established between /j/ and all other vocoids for the duration of the release phase. This difference could not be found for the amount of air over this duration. The place of articulation had only restricted influence. Velars resulted in significantly longer duration of the release phase compared to non-velars. A significant difference in amount of air between the places of articulation could not be found.
The present article is a follow-up study of the investigation of labiodentals in German and Dutch by Hamann & Sennema (2005), where we looked at the perception of the Dutch labiodental three-way contrast by German listeners without any knowledge of Dutch and German learners of Dutch. The results of this previous study suggested that the German voiced labiodental fricative /v/ is perceptually closer to the Dutch approximant /ʋ/ than to the corresponding Dutch voiced labiodental fricative /v/. These perceptual indications are attested by the acoustic findings in the present study. German /v/ has a similar harmonicity median and a similar centre of gravity to Dutch /ʋ/, but differs from Dutch /v/ in these parameters. With respect to the acoustic parameter of duration, German /v/ lies closer to the Dutch /v/ than to the Dutch /ʋ/.
The present article illustrates that the specific articulatory and aerodynamic requirements for voiced but not voiceless alveolar or dental stops can cause tongue tip retraction and tongue mid lowering and thus retroflexion of front coronals. This retroflexion is shown to have occurred diachronically in the three typologically unrelated languages Dhao (Malayo-Polynesian), Thulung (Sino-Tibetan), and Afar (East-Cushitic). In addition to the diachronic cases, we provide synchronic data for retroflexion from an articulatory study with four speakers of German, a language usually described as having alveolar stops. With these combined data we supply evidence that voiced retroflex stops (as the only retroflex segments in a language) did not necessarily emerge from implosives, as argued by Haudricourt (1950), Greenberg (1970), Bhat (1973), and Ohala (1983). Instead, we propose that the voiced front coronal plosive /d/ is generally articulated in a way that favours retroflexion, that is, with a smaller and more retracted place of articulation and a lower tongue and jaw position than /t/.
The present study argues that variation across listeners in the perception of a non-native contrast is due to two factors: the listener-specic weighting of auditory dimensions and the listener-specic construction of new segmental representations. The interaction of both factors is shown to take place in the perception grammar, which can be modelled within an OT framework. These points are illustrated with the acquisition of the Dutch three-member labiodental contrast [V v f] by German learners of Dutch, focussing on four types of learners from the perception study by Hamann and Sennema (2005a).
Arguing against Bhat’s (1974) claim that retroflexion cannot be correlated with retraction, the present article illustrates that retroflexes are always retracted, though retraction is not claimed to be a sufficient criterion for retroflexion. The cooccurrence of retraction with retroflexion is shown to make two further implications; first, that non-velarized retroflexes do not exist, and second, that secondary palatalization of retroflexes is phonetically impossible. The process of palatalization is shown to trigger a change in the primary place of articulation to non-retroflex. Phonologically, retraction has to be represented by the feature specification [+back] for all retroflex segments.
Glide formation, a process whereby an underlying high front vowel is realized as a palatal glide, is shown to occur only in unstressed prevocalic position in German, and to be blocked by specific surface restrictions such as *ji and *ʁj. Traditional descriptions of glide formation (including derivational as well as Optimality theoretic approaches) refer to the syllable in order to capture its conditions. The present study illustrates that glide formation (plus the distribution of long and short tense /i/) in German can better be captured in a Functional Phonology account (Boersma 1998) which makes reference to stress instead of the syllable and thus overcomes problems of former approaches.
This article examines the motivation for phonological stop assibilations, e.g. /t/ is realized as [ts], [s] or [tʃ] before /i/, from the phonetic perspective. Hall & Hamann (2003) posit the following two implications: (a) Assibilation cannot be triggered by /i/ unless it is also triggered by /j/, and (b) Voiced stops cannot undergo assibilations unless voiceless ones do. In the following study we present the results of three acoustic experiments with native speakers of German and Polish which support implications (a) and (b). In our experiments we measured the friction phase after the /t d/ release before the onset of the following high front vocoid for four speakers of German and Polish. We found that the friction phase for /tj/ was significantly longer than that of /ti/, and that the friction phase of /t/ in the assibilation context is significantly longer than that of /d/.
In this article we propose that there are two universal properties for phonological stop assibilations, namely (i) assibilations cannot be triggered by /i/ unless they are also triggered by /j/, and (ii) voiced stops cannot undergo assibilations unless voiceless ones do. The article presents typological evidence from assibilations in 45 languages supporting both (i) and (ii). It is argued that assibilations are to be captured in the Optimality Theoretic framework by ranking markedness constraints grounded in perception which penalize sequences like [ti] ahead of a faith constraint which militates against the change from /t/ to some sibilant sound. The occurring language types predicted by (i) and (ii) will be shown to involve permutations of the rankings between several different markedness constraints and the one faith constraint. The article demonstrates that there exist several logically possible assibilation types which are ruled out because they would involve illicit rankings.
The present study offers an Optimality-Theoretic analysis of the syllabification of intervocalic consonants and glides in Modern English. It will be argued that the proposed syllabifications fall out from universal markedness constraints – all of which derive motivation from other languages – and a language-specific ranking. The analysis offered below is therefore an alternative to the traditional rule-based analyses of English syllabification, e.g. Kahn (1976), Borowsky (1986), Giegerich (1992, 1999) and to the Optimality-Theoretic treatment proposed by Hammond (1999), whose analysis requires several language-specific constraints which apparently have no cross-linguistic motivation.
Since the advent of nonlinear phonology many linguists have either assumed or argued explicitly that many languages have words in which one or more segment does not belong structurally to the syllable. Three commonly employed adjectives used to describe such consonants are 'extrasyllabic', 'extrametrical' or 'stray'. Other authors refer to such segments as belonging to the 'appendix'. [...] Various non-linear representations have been proposed to express the 'extrasyllabicity' of segments [...]. The ones I am concerned with in the present article analyze [...] consonants [...] structurally as being outside of the syllable [...]. For transparency I ignore here both subsyllabic constituency as well as higher level prosodic constituents to which the stray consonants are sometimes assumed to attach. For reasons to be made clear below I refer to syllables [...] in which the stray consonant is situated outside of the syllable, as abstract syllables.
The present study examines a particular kind of rule blockage – referred to below as an 'antistructure-preservation effect'. An anti-structure-preservation effect occurs if some language has a process which is preempted from going into effect if some sequence of sounds [XY] would occur on the surface, even though other words in the language have [XY] sequences (which are underlyingly /XY/). It will be argued below that anti-structure-preservation effects can be captured in Optimality Theory in terms of a general ranking involving FAITH and MARKEDNESS constraints and that individual languages invoke a specific instantiation of this ranking. A significant point made below is that while anti-structure-preservation effects can be handled straightforwardly in terms of constraint rankings they typically require ad hoc rule-specific conditions in rule-based approaches.
The distribution of trimoraic syllables in German and English as evidence for the phonological word
(2000)
In the present article I discuss the distribution of trimoraic syllables in German and English. The reason I have chosen to analyze these two languages together is that the data in both languages are strikingly similar. However, although the basic generalization in (1) holds for both German and English, we will see below that trimoraic syllabIes do not have an identical distribution in both languages.
In the present study I make the following theoretical claims. First, I argue that the three environments in (1) have a property in common: they all describe the right edge of a phonological word (or prosodic word; henceforth pword). From a formal point of view, I argue that a constraint I dub the THIRD MORA RESTRICTION (henceforth TMR), which ensures that trimoraic syllables surface at the end of a pword, is active in German and English. According to my proposal trimoraic syllables cannot occur morpheme-internally because monomorphemic grammatical words like garden are parsed as single pwords. Second, I argue that the TMR refers crucially to moraic structure. In particular, underlined strings like the ones in (1) will be shown to be trimoraic; neither skeletal positions nor the subsyllabic constituent rhyme are necessary. Third, the TMR will be shown to be violated in certain (predictable) pword-internal cases, as in Monde and chamber; I account for such facts in an OptimalityTheoretic analysis (henceforth OT; Prince & Smolensky 1993) by ranking various markedness constraints among themselves or by ranking them ahead of the TMR. Fourth, I hold that the TMR describes a concrete level of grammar, which I refer to below as the 'surface' representation. In this respect, my treatment differs significantly from the one proposed for English by Borowsky (1986, 1989), in which the English facts are captured in a Lexical Phonology model by ordering the relevant constraint at level 1 in the lexicon.
One of the most important insights of Optimality Theory (Prince & Smolensky 1993) is that phonological processes can be reduced to the interaction between faithfulness and universal markedness principles. In the most constrained version of the theory, all phonological processes should be thus reducible. This hypothesis is tested by alternations that appear to be phonological but in which universal markedness principles appear to play no role. If we are to pursue the claim that all phonological processes depend on the interaction of faithfulness and markedness, then processes that are not dependent on markedness must lie outside phonology. In this paper I will examine a group of such processes, the initial consonant mutations of the Celtic languages, and argue that they belong entirely to the morphology of the languages, not the phonology.
The phenomenon of phonological opacity has been the subject of much debate in recent years, with scholars opposed to the Optimality Theory (OT) research program arguing that opacity proves OT must be false, while the solutions proposed within OT, such as sympathy theory and stratal OT , have proved to be unsatisfying to many OT proponents, who have found these proposals to be inconsistent with the parallelist approach to phonological processes otherwise characteristic of OT. In this paper I reexamine one of the best known cases of opacity, that found in three processes of Tiberian Hebrew (TH), and argue that these processes only appear to be opaque, because previous analyses have treated them as pure phonology, rather than as an interaction between phonology and morphology. Once it is recognized that certain words of TH are lexically marked to end with a syllabic trochee, and that the goal of paradigm uniformity exerts grammatical pressure on phonology, the three processes no longer present a problem to parallelist OT. The results suggest the possibility that all crosslinguistic instances of apparent opacity can be explained in terms of the phonology-morphology interface and that purely phonological opacity does not exist. If this claim is true, then parallelist OT can be defended against its detractors without the need for additional mechanisms like sympathy theory and stratal OT.
Ida'an-Begak is a Western Malayo-Polynesian language spoken by approximately 6,000 people on the east coast of Sabah, Malaysia, Borneo and belongs to the Sabahan subgroup of the North Borneo subgroup (Blust 1998). Ida'an-Begak has three dialects, Ida'an, spoken in the villages of Segama to the west of Lahad Datu, Ida'an Sungai spoken in the Kinabatangan and Sandakan districts, and Begak spoken in Ulu Tungku, to the east of Lahad Datu (Banker 1984).1 Moody (1993) deals with Ida'an; this paper concentrates on the Begak dialect. In this paper I will present new data gathered in the field and provide an analysis of the allomorphy. The study is based on spontaneous data as well as examples elicited from my language informants.
This study examines the movement trajectories of the dorsal tongue movements during symmetrical /VCa/ -sequences, where /V/ was one of the Hungarian long or short vowels /i,a,u/ and C either the voiceless palatal or velar stop consonants. General aims of this study were to deliver a data-driven account for (a) the evidence of the division between dorsality and coronality and (b) for the potential role coarticulatory factors could play for the relative frequency of velar palatalization processes in genetically unrelated languages. Results suggest a clear-cut demarcation between the behaviour of purely dorsal velars and the coronal palatals. Moreover, factors arising from a general movement economy might contribute to the palatalization processes mentioned.
Low- dimensional and speaker-independent linear vocal tract parametrizations can be obtained using the 3-mode PARAFAC factor analysis procedure first introduced by Harshman et al. (1977) and discussed in a series of subsequent papers in the Journal of the Acoustical Society of America (Jackson (1988), Nix et al. (1996), Hoole (1999), Zheng et al. (2003)). Nevertheless, some questions of importance have been left unanswered, e.g. none of the papers using this method has provided a consistent interpretation of the terms usually referred to as "speaker weights". This study attempts an exploration of what influences their reliability as a first step towards their consistent interpretation. With this in mind, we undertook a systematic comparison of the classical PARAFAC1 algorithm with a relaxed version, of it, PARAFAC2. This comparison was carried out on two different corpora acquired by the articulograph, which varied in vowel qualities, consonantal contexts, and the paralinguistic features accent and speech rate. The difference between these statistical approaches can grossly be described as follows: In PARAFAC1, observation units pertain to the same set of variables and the observation units are comparable. In PARAFAC2, observations pertain to the same set of variables, but observation units are not comparable. Such a situation can be easily conceived in a situation such as we are describing: The operationalization we took relies on the comparability of fleshpoint data acquired from different speakers, which need not be a good assumption due to influences like sensor placement and morphological conditions.
In particular, the comparison between the two different approaches is carried out by means of so-called "leverages" on different component matrices originating in regression analysis, calculated as v = diag(A(A A)−1A ) and delivering information on how "influential" a particular loading matrix is for the model. This analysis could potentially be carried out component by component, but we confined ourselves to effects on the global factor structure. For vowels, the most influential loadings are those for the tense cognates of non-palatal vowels. For speakers, the most prominent result is the relative absence of effects of the paralinguistic variables. Results generally indicate that there is quite little influence of the model specification (i.e. PARAFAC1 or PARAFAC2) on vowel and subject components. The patterns for the articulators indicate that there are strong differences between speakers with respect to the most influential measurement as revealed by PARAFAC2: In particular, the most influential y-contribution is the tongue-back for some talkers and the tongue-dorsum for other speakers. With respect to the speaker weights, again, the leverage patterns are very similar for both PARAFAC-versions. These patterns converge with the results of the loading plots, where the articulator profiles seem to be most altered by the use of PARAFAC2. These findings, in general, are interpreted as evidence for the reliability of the PARAFAC1 speaker weights.
Four speakers repeated 8 times 15 sentences containing 'pVp' syllables (V being /a/, /i/ and /u/). The 'pVp' syllables were located in final, penultimate and antepenultimate position relatively to the Intonational Phrase (IP) boundary. They were embedded in lexical words of 1-3 syllables and were either word-initial or word-final. Results show that the closer the vowel in word-final position is to the IP boundary, the longer the duration and the higher the fundamental frequency of the vowel; it is also characterised by larger lip opening gestures. The potential reduction or coarticulation of vowels in wordinitial position compared to their counterparts in word-final position is discussed.
In order to understand the functional morphology of the human voice producing system, we are in need of data on the vocal tract anatomy of other mammalian species. The larynges and vocal tracts of four species of Artiodactyla were investigated in combination with acoustic analyses of their respective calls. Different evolutionary specializations of laryngeal characters may lead to similar effects on sound production. In the investigated species, such specializations are: the elongation and mass increase of the vocal folds, the volume increase of the laryngeal vestibulum by an enlarged thyroid cartilage and the formation of laryngeal ventricles. Both the elongation of the vocal folds and the increase of the oscillating masses lower the fundamental frequency. The influence of an increased volume of the laryngeal vestibulum on sound production remains unclear. The anatomical and acoustic results are presented together with considerations about the habitats and the mating systems of the respective species.
Studying kinematic behavior in speech production is an indispensable and fruitful methodology in order to describe for instance phonemic contrasts, allophonic variations, prosodic effects in articulatory movements. More intriguingly, it is also interpreted with respect to its underlying control mechanisms. Several interpretations have been borrowed from motor control studies of arm, eye, and limb movements. They do either explain kinematics with respect to a fine tuned control by the Central Nervous System (CNS) or they take into account a combination of influences arising from motor control strategies at the CNS level and from the complex physical properties of the peripheral speech apparatus. We assume that the latter is more realistic and ecological. The aims of this article are: first, to show, via a literature review related to the so called '1/3 power law' in human arm motor control, that this debate is of first importance in human motor control research in general. Second, to study a number of speech specific examples offering a fruitful framework to address this issue. However, it is also suggested that speech motor control differs from general motor control principles in the sense that it uses specific physical properties such as vocal tract limitations, aerodynamics and biomechanics in order to produce the relevant sounds. Third, experimental and modelling results are described supporting the idea that the three properties are crucial in shaping speech kinematics for selected speech phenomena. Hence, caution should be taken when interpreting kinematic results based on experimental data alone.
Introduction
(2010)
The papers in this volume were originally presented at the Bantu Relative Clause workshop held in Paris on 8-9 January 2010, which was organized by the French-German cooperative project on the Phonology/Syntax Interface in Bantu Languages (BANTU PSYN). This project, which is funded by the ANR and the DFG, comprises three research teams, based in Berlin, Paris and Lyon. [...] This range of expertise is essential to realizing the goals of our project. Because Bantu languages have a rich phrasal phonology, they have played a central role in the development of theories of the phonology-syntax interface ever since the seminal work from the 1970s on Chimwiini (Kisseberth & Abasheikh 1974) and Haya (Byarushengo et al. 1976). Indeed, half the papers in Inkelas & Zec’s (1990) collection of papers on the phonology-syntax interface deal with Bantu languages. They have naturally played an important role in current debates comparing indirect and direct reference theories of the phonology-syntax interface. Indirect reference theories (e.g., Nespor & Vogel 1986; Selkirk 1986, 1995, 2000, 2009; Kanerva 1990; Truckenbrodt 1995, 1999, 2005, 2007) propose that phonology is not directly conditioned by syntactic information. Rather, the interface is mediated by phrasal prosodic constituents like Phonological Phrase and Intonation Phrase, which need not match any syntactic constituent. In contrast, direct reference theories (e.g., Kaisse 1985; Odden 1995, 1996; Pak 2008; Seidl 2001) argue that phrasal prosodic constituents are superfluous, as phonology can – indeed, must – refer directly to syntactic structure.
This paper presents preliminary results of a phonetic and phonological study of the Ntcheu dialect of Chichewa spoken by Al Mtenje (one of the co-authors). This study confirms Kanerva's (1990) work on Nkhotakota Chichewa showing that phonological re-phrasing is the primary cue to information structure in this language. It expands on Kanerva's work in several ways. First, we show that focus phrasing has intonational correlates, namely, the manipulation of downdrift and pause. Further, we show that there is a correlation between pitch prominence and discourse prominence at the left and right periphery which conditions dislocation to these positions. Finally, we show that focus and syntax are not the only factors which condition phonological phrasing in Chichewa.
This paper tests three current theories of the phonology-syntax interface – Truckenbrodt (1995), Pak (2008) and Cheng & Downing (2007, 2009) – on the prosody of relative clauses in Chewa. Relative clauses, especially restrictive relative clauses, provide an ideal data set for comparing these theories, as they each make distinct predictions about the optimal phrasing. We show that the asymmetrical phase-edge based approach developed to account for similar Zulu prosodic phrasing by Cheng & Downing also best accounts for the Chewa data.
Introduction
(2006)
The papers in this volume reflect a number of broad themes which have emerged during the meetings of the project as particularly relevant for current Bantu linguistics. [...] The papers show that approaches to Bantu linguistics have also developed in new directions since this foundational work. For example, interaction of phonological phrasing with syntax and word order on the one hand, and with information structure on the other, is more prominent in the papers here than in earlier literature. Quite generally, the role of information structure for the understanding of Bantu syntax has become more important, in particular with respect to the expression of topic and focus, but also for the analysis of more central syntactic concerns such as questions and relative clauses. This, of course, relates to a wider development in linguistic theory to incorporate notions of topic and focus into core syntactic analysis, and it is not surprising that work on Bantu languages and on linguistic theory are closely related to each other in this respect. Another noteworthy development is the increasing interest in variation among Bantu languages which reflects the fact that more empirical evidence from more Bantu languages has become available over the last decade or so. The picture that emerges from this research is that morpho-syntactic variation in Bantu is rich and complex, and that there is strong potential to link this research to research on micro-variation in European (and other) languages, and to the study of morpho-syntactic variables, or parameters, more generally.
In this paper, I discuss four different verb forms in Ndebele (a Nguni Bantu language spoken mainly in Zimbabwe) - the imperative, reduplicated, future and participial. I show that while all four are subject to minimality restrictions, minimality is satisfied differently in each of these morphological contexts. To account for this, I argue that in Ndebele (as in other Bantu languages) Word and RED are not the only constituents which must satisfy minimality: the Stem is also subject to minimality conditions in some morphological contexts. This paper, then, provides additional arguments for the proposal that Phonological Word is not the only sub-lexical morpho-prosodic constituent. Further, I argue that, although Word, RED and Stern are all subject to the same minimality constraint – they must all be minimally bisyllabic - this does not follow from a single 'generalized' constraint. Instead, I argue, contra recent work within Generalized Template Theory (see, e.g., McCarthy & Prince 1994, 1995a, 1999; Urbanezyk 1995, 1996; and Walker 2000; etc.) that a distinct minimality constraint must be formalized for each of these morpho-prosodic constituents.
Much work on the interaction of prosody and focus assumes that, crosslinguistically, there is a necessary correlation between the position of main sentence stress (or accent) and focus, and that an intonational pitch change on the focused element is a primary correlate of focus. In this paper, I discuss primary data from three Bantu languages – Chichewa, Durban Zulu and Chitumbuka – and show that in all three languages phonological re-phrasing, not stress, is the main prosodic correlate of focus and that lengthening, not pitch movement, is the main prosodic correlate of phrasing. This result is of interest for the typology of intonation in illustrating languages where intonation has limited use and where, notably, intonation does not highlight focused information in the way we might expect from European stress languages.
The goal of this paper is to survey the accent systems of the indigenous languages of Africa. Although roughly one third of the world’s languages are spoken in Africa, this continent has tended to be underrepresented in earlier stress and accent typology surveys, like Hyman (1977). This one aims to fill that gap. Two main contributions to the typology of accent are made by this study of African languages. First, it confirms Hyman's (1977) earlier finding that (stem-)initial and penult are the most common positions, cross-linguistically, to be assigned main stress. Further, it shows that not only stress but also tone and segment distribution can define prominence asymmetries which are best analyzed in terms of accent.
This paper presents a preliminary survey of the positions and prosodies associated with Wh-questions in two Bantu languages spoken in Malawi. The paper shows that the two languages are similar in requiring focused subjects to be clefted. Both also require 'which' questions and 'because of what' questions to be clefted or fronted. However, for other non-subjects Tumbuka rather uniformly imposes an IAV (immediately after the verb) requirement, while Chewa does not. In both languages, we found a strong tendency for there to be a prosodic phrase break following the Wh-word. In Tumbuka, this break follows from the general phrasing algorithm of the language, while in Chewa, I propose that the break can be best understood as following from the inherent prominence of Wh-words.
Introduction
(2011)
In spite of this long history, most work to date on the phonology-syntax interface in Bantu languages suffers from limitations, due to the range of expertise required: intonation, phonology, syntax. Quite generally, intonational studies on African languages are extremely rare. Most of the existing data has not been the subject of careful phonetic analysis, whether of the prosody of neutral sentences or of questions or other focus structures. There are important gaps in our knowledge of Bantu syntax which in turn limit our understanding of the phonology-syntax interface. Recent developments in syntactic theory have provided a new way of thinking about the type of syntactic information that phonology can refer to and have raised new questions: Do only syntactic constituent edges condition prosodic phrasing? Do larger domains such as syntactic phases, or even other factors, like argument and adjunct distinctions, play a role? Further, earlier studies looked at a limited range of syntactic constructions. Little research exists on the phonology of focus or of sentences with non-canonical word order in Bantu languages. Both the prosody and the syntax of complex sentences, questions and dislocations are understudied for Bantu languages. Our project aims to remedy these gaps in our knowledge by bringing together a research team with all the necessary expertise. Further, by undertaking the intonational, phonological and syntactic analysis of several languages we can investigate whether there is any correlation among differences in morphosyntactic and prosodic properties that might also explain differences in phrasing and intonation. It will also allow us to investigate whether there are cross-linguistically common prosodic patterns for particular morpho-syntactic structure.
As work like McCarthy (2002: 128) notes, pre-Optimality Theory (OT) phonology was primarily concerned with representations and theories of subsegmental structure. In contrast, the role of representations and choice of structural models has received little attention in OT. Some central representational issues of the pre-OT era have, in fact, become moot in OT (McCarthy 2002: 128). Further, as work like Baković (2007) notes, even for assimilatory processes where representation played a central role in the pre-OT era, constraint interaction now carries the main explanatory burden. Indeed, relatively few studies in OT (e.g., Rose 2000; Hargus & Beavert 2006; Huffmann 2005, 2007; Morén 2006) have argued for the importance of phonological representations. This paper intends to contribute to this work by reanalyzing a set of processes related to vowel harmony in Shimakonde, a Bantu language spoken in Mozambique and Tanzania. These processes are of particular interest, as Liphola’s (2001) study argues that they are derivationally opaque and so not amenable to an OT analysis. I show that the opacity disappears given the proper choice of representations for vowel features and a metrical harmony domain.
It is one of the most highly debated issues in loanword phonology whether loanword adaptations are phonologically or phonetically driven. This paper addresses this issue and aims at demonstrating that only the acceptance of both a phonological as well as a phonetic approximation stance can adequately account for the data found in Japanese. This point will be exemplified with the adaptation of German and French mid front rounded vowels in Japanese. It will be argued that the adaptation of German /oe/ and /ø/ as Japanese /e/ is phonologically grounded, whereas the adaptation of French /oe/ and /ø/ as Japanese /u/ is phonetically grounded. This asymmetry in the adaptation process of German and French mid front rounded vowels and further examples of loans in Japanese lead to the only conclusion that both strategies of loanword adaptation occur in languages. It will be shown that not only perception, but also the influence of orthography, of conventions and the knowledge of the source language play a role in the adaptation process.
In this study, cross-dialectal variation in the use of the acoustic cues of VOT and F0 to mark the laryngeal contrast in Korean stops is examined with Chonnam Korean and Seoul Korean. Prior experimental results (Han & Weitzman, 1970; Hardcastle, 1973; Jun, 1993 &1998; Kim, C., 1965) show that pitch values in the vowel onset following the target stop consonants play a supplementary role to VOT in designating the three contrastive laryngeal categories. F0 contours are determined in part by the intonational system of a language, which raises the question of how the intonational system interacts with phonological contrasts. Intonational difference might be linked to dissimilar patterns in using the complementary acoustic cues of VOT and F0. This hypothesis is tested with 6 Korean speakers, three Seoul Korean and three Chonnam Korean speakers. The results show that Chonnam Korean involves more 3-way VOT and a 2-way distinction in F0 distribution in comparison to Seoul Korean that shows more 3-way F0 distribution and a 2-way VOT distinction. The two acoustic cues are complementary in that one cue is rather faithful in marking 3-way contrast, while the other cue marks the contrast less distinctively. It also seems that these variations are not completely arbitrary, but linked to the phonological characteristics in dialects. Chonnam Korean, in which the initial tonal realization in the accentual phrase is expected to be more salient, tends to minimize the F0 perturbation effect from the preceding consonants by taking more overlaps in F0 distribution. And a 3-way distribution of VOT in Chonnam Korean, as compensation, can be also understood as a durational sensitivity. Without these characteristics, Seoul Korean shows relatively more overlapping distribution in VOT and more 3-way separation in F0 distribution.
Vowel dispersion in Truku
(2004)
This study investigates the dispersion of vowel space in Truku, an endangered Austronesian language in Taiwan. Adaptive Dispersion (Liljencrants and Lindblom, 1972; Lindblom, 1986, 1990) proposes that the distinctive sounds of a language tend to be positioned in phonetic space in a way that maximizes perceptual contrast. For example, languages with large vowel inventories tend to expand the overall acoustic vowel space. Adaptive Dispersion predicts that the distance between the point vowels will increase with the size of a language's vowel inventory. Thus, the available acoustic vowel space is utilized in a way that maintains maximal auditory contrast.
The current study focuses on the prosodic realization of negators in Saisiyat, an endangered aboriginal language of Taiwan, and compares its prosodic realization of negation with that of English. The results of this study indicate that sentential subjects are the most acoustically prominent items in the Saisiyat negative sentences measured. This contrasts sharply with the English experimental sentences, in which the negator itself was the most acoustically prominent item. These findings suggest that Saisiyat is a pitch-accent language; thus, the presence of negators does not significantly change the prosodic parameters of surrounding words. English, in contrast, is an intonation language, so the presence of negation results in substantial prosodic modification. This suggests that the phenomenon of negation is universally prominent; however, languages with different prosodic systems will adopt different strategies for realizing prominence.
Tone as a distinctive feature used to differentiate not only words but also clause types, is a characteristic feature of Bantu languages. In this paper we show that Bemba relatives can be marked with a low tone in place of a segmental relative marker. This low tone strategy of relativization, which imposes a restrictive reading of relatives, manifests a specific phonological phrasing that can be differentiated from that of non-restrictives. The paper shows that the resultant phonological phrasing favours a head-raising analysis of relativization. In this sense, phonology can be shown to inform syntactic analyses.
This study focuses upon a detailed description and analysis of the phonetic structures of Paiwan, an aboriginal language spoken in Taiwan, with around 53,000 speakers, Paiwan, a member of the Austronesian language family, is not typologically related to the other languages such as Mandarin and Taiwanese spoken in its geographically contiguous districts, Earlier work on phonological features of Paiwan (Chang, 1999; Tseng, 2003) sought an account in terms of segments and isolated facts about reduplication and stress, without accounting for the possible roles of phrase-level and sentence-Ievel prosodic structures, Government Teaching Material (1993) listed 25 consonants and 4 vowels, without any description of phonetic features and phonological rules, Chang's (2000) reference grammar included 22 consonants and 4 vowels, with a very brief description of 5 phonological rules on single words, Regional diversity and 25 consonants have been mentioned in Pulaluyan's (2002) teaching material; however, no description of phonological rules was found in his material.
Rate effects on aerodynamics of intervocalic stops : evidence from real speech data and model data
(2008)
This paper is a first attempt towards a better understanding of the aerodynamic properties during speech production and their potential control. In recent years, studies on intraoral pressure in speech have been rather rare, and more studies concern the air flow development. However, the intraoral pressure is a crucial factor for analysing the production of various sounds.
In this paper, we focus on the intraoral pressure development during the production of intervocalic stops.
Two experimental methodologies are presented and confronted with each other: real speech data recorded for four German native speakers, and model data, obtained by a mechanical replica which allows reproducing the main physical mechanisms occurring during phonation. The two methods are presented and applied to a study on the influence of speech rate on aerodynamic properties.
Mechanisms of contrasting korean velar stops : A catalogue of acoustic and articulatory parameters
(2003)
The Korean stop system exhibits a three-way distinction in velar stops among /g/, /k'/ and /kh/. If the differentiation is regarded as being based on voicing, such a system is rather unusual because even a two-way distinction between a voiced and a voicless unaspirated velar stop gets easily lost in the languages of the world especially in the case of velar stops. One possibility for maintainig this distinction is that supralaryngeal characteristics like articulators' velocity, duration of surrounding vowels or stop closure duration are involved. The aim of the present study is to set up a catalogue of parameters which are involved in the distinction of Korean velar stops in intervocalic position.
Two Korean speakers have been recorded via Electromagnetic Articulography. The word material consisted of VCV-sequences where V is one of the three vowels /a/, /i/ or /u/ and C one of the Korean velars /g/, /k'/ or /kh/. Articulatory and acoustic signals have been analysed It turned out that the distinction is only partly built on laryngeal parameters and that supralaryngeal characteristics differ for the three stops. Another result is that the voicing contrast is not a matter of one parameter, but there is always a set of parameters involved. Furthermore, speakers seem to have a certain freedom in the choice of these parameters.
Articulatory token-to-token variability not only depends on linguistic aspects like the phoneme inventory of a given language but also on speaker specific morphological and motor constraints. As has been noted previously (Perkell (1997), Mooshammer et al. (2004)), speakers with coronally high "domeshaped" palates exhibit more articulatory variability than speakers with coronally low "flat" palates. One explanation for that is based on perception oriented control by the speaker. The influence of articulatory variation on the cross sectional area and consequently on the acoustics should be greater for flat palates than for domeshaped ones. This should force speakers with flat palates to place their tongue very precisely whereas speakers with domeshaped palates might tolerate a greater variability. A second explanation could be a greater amount of lateral linguo-palatal contact for flat palates holding the tongue in position. In this study both hypotheses were tested.
In order to investigate the influence of the palate shape on the variability of the acoustic output a modelling study was carried out. Parallely, an EPG experiment was conducted in order to investigate the relationship between palate shape, articulatory variability and linguo-palatal contact.
Results from the modelling study suggest that the acoustic variability resulting from a certain amount of articulatory variability is higher for flat palates than for domeshaped ones. Results from the EPG experiment with 20 speakers show that (1.) speakers with a flat palate exhibit a very low articulatory variability whereas speakers with a domeshaped palate vary, (2.) there is less articulatory variability if there is lots of linguo-palatal contact and (3.) there is no relationship between the amount of lateral linguo-palatal contact and palate shape. The results suggest that there is a relationship between token-to-token variability and palate shape, however, it is not that the two parameters correlate, but that speakers with a flat palate always have a low variability because of constraints of the variability range of the acoustic output whereas speakers with a domeshaped palate may choose the degree of variability. Since linguo-palatal contact and variability correlate it is assumed that linguo-palatal contact is a means for reducing the articulatory variability.
The unfolding discussion will focus on the internal representation of turbulent sounds in the phonology of German as well as pinpoint the special status of the prime defining the quality of turbulence. It will also be argued that this prime is capable of entering into special types of licensing relations, which results in specific phonetic manifestations of forms. We shall compare the effects of two processes attested in German: consonant degemination and spirantisation with a view to revealing the role of the turbulence-defining element in the two operations. Furthermore, our attention will be focused on the workings of the Obligatory Contour Principle which, as will be shown below, exerts decisive impact on prime interplay and consequently the phonetic realization of sounds and words. We shall see that segmental identity is contingent on the languagespecific interpretation of inter-element bonds.
Aware of the importance of prime autonomy in determining the manifestation of sounds, let us start with a brief outline of the fundamental segment structure principles offered by the theory of Phonological Government.
In this paper we focus on the similarities tying together the second segment of an onset cluster and a singleton coda segment. We offer a proposal based on Baertsch (2002) accounting for this similarity and show how it captures a number of observations which have defied previous explanation. In accounting for the similarity of patterning between the second member of an onset and a coda consonant, we propose to augment Prince & Smolensky's (P&S, 1993/2002) Margin Hierarchy so as to distinguish between structural positions that prefer low sonority and those that prefer high sonority. P&S's Margin Hierarchy, which gives preference to segments of low sonority, applies to singleton onsets; this is our M1 hierarchy. Our proposed M2 hierarchy applies both to the second member of an onset and to a singleton coda. The M2 hierarchy differs from the M1 hierarchy in giving preference to consonants of high sonority. Splitting the Margin Hierarchy into the M1 and M2 hierarchies allows us to explain typological, phonotactic, and acquisitional observations that have defied previous explanation. In Section 2 of this paper, we briefly provide background on the links that tie together the second member of an onset and a singleton coda. In Section 3, we review P&S's Margin Hierarchy, showing that it becomes problematic when extended to coda consonants. We then offer our proposal for a split margin hierarchy. Section 4 extends the split margin approach to complex onsets. We then show how it is able to account for various typological, phonotactic, and acquisitional observations. In Section 5, we will conclude the paper by briefly sketching how the split margin approach enables us to analyze syllable contact phenomena without requiring a specific syllable contact constraint (or additional hierarchy) or reference to an external sonority scale.