Elsevier

Journal of Phonetics

Volume 23, Issues 1–2, January–April 1995, Pages 37-52
Journal of Phonetics

Velarization degree and coarticulatory resistance for /I/ in Catalan and German

https://doi.org/10.1016/S0095-4470(95)80031-XGet rights and content

Abstract

This paper compares formant values and patterns of linguopalatalcontact for [I] in the sequences [iIi] and [ala] in both Catalan and German, in order to determine whether differences in consonantal velarization between the two languages influence the degree of vowel-to-consonant coarticulation.

Significant F2 and F2- F, differences for [I] in the sequence [ili]indicate that the consonant is velarized in Catalan and non-velarized in German; F2 data for male speakers show a 1500Hz threshold between the two consonantal varieties. This language-dependent significant difference was not found to hold for the dorsopalatal contact as measured by electropalatography; there is, however, a clear trend for dorsal contact at the palatal zone for German [I] to exceed that for Catalan [I]. Moreover, F2 values suggest that the consonant may be less non-velarized in German than in other languages (Italian, Spanish, French).

The degree of coarticulation was not inversely related to the degree of velarization in the case analyzed in this paper, i.e., coarticulatory effects from [i] vs. [a] on [I] were not significantly different in German vs. Catalan. Velarized [I] in Catalan is highly resistant to coarticulatory effects from [i] since the two phonetic segments are produced with antagonistic tongue dorsum gestures, i.e., tongue dorsum lowering and retraction, and tongue dorsum raising and fronting. While being non-velarized, German [I] appears to be highly resistant to coarticulatory effects from [i] vs. [a]: F, values for [I] in both vowel environments are less extreme in German than in other languages showing a non-velarized variety of the consonant.

These findings suggest that the tongue dorsum for German [I] attains some target position, and are in support of the notion that the velarization/non-velarization distinction proceeds in a scalar, noncategorical fashion.

References (21)

There are more references available in the full text version of this article.

Cited by (41)

  • Acoustic characteristics of Korean-English bilingual speakers’ /l/ and the relationship to their foreign accent ratings.

    2021, Journal of Communication Disorders
    Citation Excerpt :

    Classic studies by Giles and Moll (1975) and Sproat and Fujimura (1993), however, showed that American English (AE) word-initial and -final /l/ both involve anterior as well as posterior constrictions, and the allophonic variation of light or dark /l/ is not a binary entity but is gradient in nature, which is affected by several factors, such as phonetic contexts, speaker age, gender, or dialects/languages. AE word-initial and -final /l/, although vary by the above listed factors, is in general reported to fall into the dark /l/ end of the continuum as compared to /l/ of other languages (e.g., Berkson, de Jong & Lulich, 2017; Chung, 2021; Lehiste, 1964; Recasens et al., 1995; Recasens, 2004; Recasens & Espinosa, 2005). In contrast to English /l/, Korean /l/ generally falls in to the light /l/end of the continuum.

  • An acoustic study of coarticulatory resistance in "dark" and "light" alveolar laterals

    2015, Journal of Phonetics
    Citation Excerpt :

    If the “dark” variant is specified for a particular tongue dorsum configuration (medio-dorsum lowering and post-dorsum raising and backing, for instance), one would expect for this /l/ variant to be less affected by contextual vowels, as its own dorsal specification would be in competition with that of the neighboring vowels. Since this view of the role of the tongue dorsum on the articulation of /l/ helps to successfully predict the attested difference between the two /l/ variants with respect to their patterns of resistance to coarticulation (e.g. Recasens, 1987; Recasens & Espinosa, 2005; Recasens et al., 1995, 1996, 1998), this is a view that deserves serious attention. Recent articulatory data suggest, however, that both “dark” and “light” /l/ recruit the tongue dorsum to some extent (Sproat & Fujimura, 1993; Browman & Goldstein, 1995).

  • Articulatory synergies in the temporal organization of liquid clusters in Romanian

    2014, Journal of Phonetics
    Citation Excerpt :

    Clear /l/ is characterized by a fronted, raised tongue body. Dark /l/s, due to their pronounced postdorsal retraction gesture show a high degree of coarticulatory resistance (Recasens & Espinosa, 2005; Recasens et al., 1995, 1996, 1998). English dark /l/ and its prosodically conditioned allophony has been extensively studied: coda /l/ is produced with strong tongue body retraction that temporally precedes the coronal constriction, whereas onset /l/s are characterized by a wider tongue body constriction degree and a synchronous timing of the two gestures (e.g., Delattre, 1971; Giles & Moll, 1975; Narayanan, Alwan, & Haker, 1997; Scobbie & Pouplier, 2010; Sproat & Fujimura, 1993).

  • Dark and clear laterals in Catalan and Spanish: Interaction of phonetic categories in early bilinguals

    2010, Journal of Phonetics
    Citation Excerpt :

    In other words, a tongue tip raising gesture towards the alveolar ridge is characteristic of both sounds. What distinguishes them is that dark laterals have a dorsal approximation at the velar region as well as considerable pre-dorsal lowering (Narayanan, Alwan, & Haker, 1997; Recasens, 2004; Recasens & Espinosa, 2005; Recasens, Fontdevila, & Pallarès, 1995, 1996; Recasens, Pallarès, & Fontdevila, 1998; Sproat & Fujimura, 1993). Active dorsal backing and lowering in dark laterals triggers a lowering of the second formant (F2) and a raising of the first formant (F1) as compared with clear laterals (Recasens, 2004).

View all citing articles on Scopus
*

Also of Universitat Autónoma de Barcelona.

View full text