The phonological status of geminate consonants has long been debated between two antithetical explanatory hypotheses, one bisegmental (Trubetzkoj, 1939; cited in Loporcaro, 1990) and one monosegmental (Martinet, 1975; cited in Loporcaro, 1990) . In addition, authors who to date espouse the former hypothesis, such as Loporcaro (1990), consider the latter not yet definitively outdated. Significantly, the "facts" used by phonologists to support either hypothesis have been mainly of two types, both "internal" to the language: on the one hand, the observation of distributional regularities, and on the other, derivation from Latin. In contrast, "facts" of an "external" nature, related to acoustic or kinematic analysis, which quantitatively measure locutions' realizations, have rarely been investigated. Acoustic analysis, however, can only confirm the significant difference between scimes and geminates without providing information on the mode of segment realization. Kinematic analysis, on the other hand, has the advantage of revealing which of at least four strategies by which a duration contrast can be created (Cho, 2001) is the one actually used by the speaker (and prescribed by the language). The first portion of the work aims to conduct a review of the different theories, with the purpose of highlighting how they describe the phonological representation of geminates. In particular, it is intended to devote a section on the status of geminate consonants in the Italian language. Next, the theory of Articulatory Phonology (Browman & Goldstein, 1986) will be introduced, which assumes that the fundamental phonological units are not abstract symbols such as phonemes, but 'articulatory gestures'. In the following, the gestural organization of geminates from the perspective of Articulatory Phonology will be expounded, making special reference to studies inherent in the Italian language. The second portion of the thesis is devoted to experimental analysis, in which data from four Italian locutors, collected by Zmarich et al. (2011), will be used to verify the coordination of Italian geminates from the perspective of Articulatory Phonology. To do this, we intend to use the formalization of Gafos (2002), which identifies four basic landmarks in the trajectories of consonantal gestures, used by the phonologies of individual languages for the coordination of intra- and inter-syllabic gestures. First, we will take up the experimental hypothesis underlying Zmarich et al (2011), inherent in the prediction of Ohman's V-toV model (cited by Smith, 1995). Next, we intend to test whether the Italian geminates are coordinated according to the alternative model of Articulatory Phonology (Browman and Goldstein, 1986), the predicts a different coordination of the consonantal gesture with respect to the vowel gesture depending on whether the consonantal gesture is onset (in-phase coordination) or coda (out-of-phase coordination). Then, the two main and antithetical hypotheses of traditional phonology about the phonological status of geminates will be evaluated: that of Trubetzkoj (bisegmental) and that of Martinet (monosegmental). These theories have been variously reinterpreted by Autosegmental Phonology, according to which geminates constitute a single segment occupying two X- slots at the level of the temporal tier (Marotta and Vanelli, 2021). We will test their validity of the two hypotheses by constructing a 'virtual geminate,' simulated from articulatory data related to the dumb consonants found in Zmarich et al. (2011). Finally, the veracity of the results of Tilsen and Hermes (2021) about specific time intervals of Italian singletons and geminates will be tested.
Lo statuto fonologico delle consonanti geminate è stato a lungo oggetto di dibattito tra due antitetiche ipotesi esplicative, una bisegmentale (Trubetzkoj, 1939; citato in Loporcaro, 1990) ed una monosegmentale (Martinet, 1975; citato in Loporcaro, 1990) . In aggiunta, autori che ad oggi sposano la prima ipotesi, quali Loporcaro (1990), reputano la seconda non ancora definitivamente superata. Significativamente, i "fatti" usati dai fonologi per supportare l'una o l'altra delle due ipotesi sono stati principalmente di due tipi, entrambi "interni" alla lingua: da una parte l'osservazione delle regolarità distribuzionali, dall'altra la derivazione dal latino. Al contrario, i “fatti” di natura “esterna”, relativi all’analisi acustica o cinematica, i quali misurano quantitativamente le realizzazioni dei locutori, sono stati di rado oggetto di indagine. L’analisi acustica, tuttavia, può solamente confermare la differenza significativa tra scempie e geminate senza fornire informazioni sul modo di realizzazione del segmento. L'analisi cinematica, invece, presenta il vantaggio di rivelare quale delle almeno quattro strategie mediante le quali si può creare un contrasto di durata (Cho, 2001) è quella realmente utilizzata dal parlante (e prescritta dalla lingua). La prima porzione del lavoro si prefigge di effettuare una rassegna delle diverse teorie, col fine di mettere in luce come esse descrivano la rappresentazione fonologica delle geminate. In particolare, si intende dedicare una sezione sullo statuto delle consonanti geminate nella lingua italiana. Successivamente, verrà introdotta la teoria della Fonologia Articolatoria (Browman & Goldstein, 1986), la quale assume che le unità fonologiche fondamentali non siano simboli astratti come i fonemi, bensì i ‘gesti articolatori’. Di seguito, si esporrà l’organizzazione gestuale delle geminate secondo la prospettiva della Fonologia Articolatoria, facendo particolare riferimento agli studi inerenti alla lingua italiana. La seconda porzione della tesi è dedicata all’analisi sperimentale, nella quale si utilizzeranno i dati di quattro locutori italiani, raccolti da Zmarich et al. (2011), per verificare la coordinazione delle geminate italiane nell’ottica della Fonologia Articolatoria. Per far questo, si intende utilizzare la formalizzazione di Gafos (2002) che individua quattro landmark fondamentali nelle traiettorie dei gesti consonantici, usati dalle fonologie delle singole lingue per la coordinazione dei gesti intra- e inter-sillabici. In primo luogo, si riprenderà l’ipotesi sperimentale alla base di Zmarich et al (2011), inerente alla predizione del modello V-toV di Ohman (citato da Smith, 1995). Successivamente, si intende verificare se le geminate italiane siano coordinate secondo il modello alternativo della Fonologia Articolatoria (Browman e Goldstein, 1986), la prevede una diversa coordinazione del gesto consonantico rispetto a quello vocalico a seconda che il gesto consonantico sia in onset (coordinazione in-phase) o in coda (coordinazione out-of-phase). Quindi, si valuteranno le due ipotesi principali ed antitetiche della fonologia tradizionale circa lo statuto fonologico delle geminate: quella di Trubetzkoj (bisegmentale) e quella di Martinet (monosegmentale). Tali teorie sono state variamente reinterpretate dalla Fonologia autosegmentale, secondo la quale le geminate costituiscono un unico segmento che occupa due X- slots a livello del tier temporale (Marotta e Vanelli, 2021). Si testerà la loro validità delle due ipotesi mediante la costruzione di una ‘geminata virtuale’, simulata a partire dai dati articolatori relativi alle consonanti scempie presenti in Zmarich et al. (2011). Infine, si proverà la veridicità dei risultati di Tilsen e Hermes (2021) circa specifici intervalli temporali di singleton e geminate italiane.
"Geminate nasali bilabiali dell’italiano: stato dell’arte ed applicazione della Fonologia Articolatoria ad uno studio sperimentale sulla sincronizzazione gestuale"
FONTANELLA, FRANCESCA
2022/2023
Abstract
The phonological status of geminate consonants has long been debated between two antithetical explanatory hypotheses, one bisegmental (Trubetzkoj, 1939; cited in Loporcaro, 1990) and one monosegmental (Martinet, 1975; cited in Loporcaro, 1990) . In addition, authors who to date espouse the former hypothesis, such as Loporcaro (1990), consider the latter not yet definitively outdated. Significantly, the "facts" used by phonologists to support either hypothesis have been mainly of two types, both "internal" to the language: on the one hand, the observation of distributional regularities, and on the other, derivation from Latin. In contrast, "facts" of an "external" nature, related to acoustic or kinematic analysis, which quantitatively measure locutions' realizations, have rarely been investigated. Acoustic analysis, however, can only confirm the significant difference between scimes and geminates without providing information on the mode of segment realization. Kinematic analysis, on the other hand, has the advantage of revealing which of at least four strategies by which a duration contrast can be created (Cho, 2001) is the one actually used by the speaker (and prescribed by the language). The first portion of the work aims to conduct a review of the different theories, with the purpose of highlighting how they describe the phonological representation of geminates. In particular, it is intended to devote a section on the status of geminate consonants in the Italian language. Next, the theory of Articulatory Phonology (Browman & Goldstein, 1986) will be introduced, which assumes that the fundamental phonological units are not abstract symbols such as phonemes, but 'articulatory gestures'. In the following, the gestural organization of geminates from the perspective of Articulatory Phonology will be expounded, making special reference to studies inherent in the Italian language. The second portion of the thesis is devoted to experimental analysis, in which data from four Italian locutors, collected by Zmarich et al. (2011), will be used to verify the coordination of Italian geminates from the perspective of Articulatory Phonology. To do this, we intend to use the formalization of Gafos (2002), which identifies four basic landmarks in the trajectories of consonantal gestures, used by the phonologies of individual languages for the coordination of intra- and inter-syllabic gestures. First, we will take up the experimental hypothesis underlying Zmarich et al (2011), inherent in the prediction of Ohman's V-toV model (cited by Smith, 1995). Next, we intend to test whether the Italian geminates are coordinated according to the alternative model of Articulatory Phonology (Browman and Goldstein, 1986), the predicts a different coordination of the consonantal gesture with respect to the vowel gesture depending on whether the consonantal gesture is onset (in-phase coordination) or coda (out-of-phase coordination). Then, the two main and antithetical hypotheses of traditional phonology about the phonological status of geminates will be evaluated: that of Trubetzkoj (bisegmental) and that of Martinet (monosegmental). These theories have been variously reinterpreted by Autosegmental Phonology, according to which geminates constitute a single segment occupying two X- slots at the level of the temporal tier (Marotta and Vanelli, 2021). We will test their validity of the two hypotheses by constructing a 'virtual geminate,' simulated from articulatory data related to the dumb consonants found in Zmarich et al. (2011). Finally, the veracity of the results of Tilsen and Hermes (2021) about specific time intervals of Italian singletons and geminates will be tested.File | Dimensione | Formato | |
---|---|---|---|
Fontanella_Francesca.pdf
accesso riservato
Dimensione
4.11 MB
Formato
Adobe PDF
|
4.11 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/51056