This paper investigates the transfer of spirantization into the English of native Spanish-speaker... more This paper investigates the transfer of spirantization into the English of native Spanish-speakers. Typologically, spirantization is a kind of lenition in which a stop is reduced to a fricative or an approximant (Kirchner, 2004: 313). In Spanish, spirantization refers to an allophonic process in which the voiced stops /b d g/ surface as approximants in certain linguistic environments (Harris, 1969). Previous studies (Zampini 1995 and Algara, González, & Boada 2004) have shown that native Spanish-speakers do exhibit the transfer of spirantization when speaking English, however, these studies relied on impressionistic transcription and were carried out under the assumption that native English-speakers do not lenite voiced stops.
In this paper, I use intensity ratios and a qualitative ‘spirantization score’ to re-examine the spirantization of target English voiced stops by L1 Spanish speakers. This paper looks at a subset of data from a larger corpus: six female speakers of Mexican Spanish (3 intermediate and 3 advanced students of English) and a control group of three female speakers of American English. Subjects produced voiced stops in the intervocalic environment (both word-initially and word-medially) and in utterance-initial position. The former (e.g. the ball [ðəˈbɑl]) is an environment where Spanish-speakers are likely to spirantize target English voiced stops because this environment is known to condition more approximant-like phones across all dialects of Spanish (Piñeros 2001). The latter (e.g. ball [ˈbɑl]) is an environment where Spanish speakers are less likely to spirantize English stops because, as studies have shown (e.g. Eddington 2011), the utterance-initial environment is a significant predictor of stop-like pronunciations.
The data were submitted to a linear mixed-effects statistical model. Results show that Spanish-speakers do exhibit the transfer of spirantization, however, the advanced learners of English improved significantly, signifying that they were able to suppress the transfer of this allophonic process to some degree. Furthermore, results suggest that the same factors that condition more lenited voiced stops by native Spanish-speakers (stress status, word-position, task formality), also, to a much lesser extent, condition more spirantized voiced stops by native English speakers.
By means of a matched-guise study, this paper examines the attitudes of L2, heritage, and native ... more By means of a matched-guise study, this paper examines the attitudes of L2, heritage, and native Spanish speakers in the state of Washington toward Mexican-accented and English-accented Spanish. We interpret our findings in the wake of previous research on language attitudes and ideologies related to Spanish in the United States which shows that Spanish and those who speak it as a first or heritage language are thought to have a lower socioeconomic status than English and Anglophones. 97 Spanish-speaking participants residing in Washington (N=95) and the Pacific Northwest (N=2) rated 4 voices along six-point semantic differential scales falling into the dimensions of superiority, solidarity, language competence, and physical characteristics. We submitted mean scores to a linear mixed-effects model. Contrary to our expectations, all groups rated the Mexican-accented guises higher than the English-accented guises in the dimension of superiority. Also unforeseen, the L2 speakers rated the Mexican-accented voices higher in the dimension of solidarity. We consider the high level of education of the respondents and, for the L2 subjects, their experience as advanced Spanish language speakers, as likely explanations for the observed attenuation of well-documented prevailing stereotypes directed at Latinos from the monolingual community at large.
This paper investigates the transfer of spirantization into the English of native Spanish-speaker... more This paper investigates the transfer of spirantization into the English of native Spanish-speakers. Typologically, spirantization is a kind of lenition in which a stop is reduced to a fricative or an approximant (Kirchner, 2004: 313). In Spanish, spirantization refers to an allophonic process in which the voiced stops /b d g/ surface as approximants in certain linguistic environments (Harris, 1969). Previous studies (Zampini 1995 and Algara, González, & Boada 2004) have shown that native Spanish-speakers do exhibit the transfer of spirantization when speaking English, however, these studies relied on impressionistic transcription and were carried out under the assumption that native English-speakers do not lenite voiced stops.
In this paper, I use intensity ratios and a qualitative ‘spirantization score’ to re-examine the spirantization of target English voiced stops by L1 Spanish speakers. This paper looks at a subset of data from a larger corpus: six female speakers of Mexican Spanish (3 intermediate and 3 advanced students of English) and a control group of three female speakers of American English. Subjects produced voiced stops in the intervocalic environment (both word-initially and word-medially) and in utterance-initial position. The former (e.g. the ball [ðəˈbɑl]) is an environment where Spanish-speakers are likely to spirantize target English voiced stops because this environment is known to condition more approximant-like phones across all dialects of Spanish (Piñeros 2001). The latter (e.g. ball [ˈbɑl]) is an environment where Spanish speakers are less likely to spirantize English stops because, as studies have shown (e.g. Eddington 2011), the utterance-initial environment is a significant predictor of stop-like pronunciations.
The data were submitted to a linear mixed-effects statistical model. Results show that Spanish-speakers do exhibit the transfer of spirantization, however, the advanced learners of English improved significantly, signifying that they were able to suppress the transfer of this allophonic process to some degree. Furthermore, results suggest that the same factors that condition more lenited voiced stops by native Spanish-speakers (stress status, word-position, task formality), also, to a much lesser extent, condition more spirantized voiced stops by native English speakers.
By means of a matched-guise study, this paper examines the attitudes of L2, heritage, and native ... more By means of a matched-guise study, this paper examines the attitudes of L2, heritage, and native Spanish speakers in the state of Washington toward Mexican-accented and English-accented Spanish. We interpret our findings in the wake of previous research on language attitudes and ideologies related to Spanish in the United States which shows that Spanish and those who speak it as a first or heritage language are thought to have a lower socioeconomic status than English and Anglophones. 97 Spanish-speaking participants residing in Washington (N=95) and the Pacific Northwest (N=2) rated 4 voices along six-point semantic differential scales falling into the dimensions of superiority, solidarity, language competence, and physical characteristics. We submitted mean scores to a linear mixed-effects model. Contrary to our expectations, all groups rated the Mexican-accented guises higher than the English-accented guises in the dimension of superiority. Also unforeseen, the L2 speakers rated the Mexican-accented voices higher in the dimension of solidarity. We consider the high level of education of the respondents and, for the L2 subjects, their experience as advanced Spanish language speakers, as likely explanations for the observed attenuation of well-documented prevailing stereotypes directed at Latinos from the monolingual community at large.
Uploads
Drafts by Max Carey
In this paper, I use intensity ratios and a qualitative ‘spirantization score’ to re-examine the spirantization of target English voiced stops by L1 Spanish speakers. This paper looks at a subset of data from a larger corpus: six female speakers of Mexican Spanish (3 intermediate and 3 advanced students of English) and a control group of three female speakers of American English. Subjects produced voiced stops in the intervocalic environment (both word-initially and word-medially) and in utterance-initial position. The former (e.g. the ball [ðəˈbɑl]) is an environment where Spanish-speakers are likely to spirantize target English voiced stops because this environment is known to condition more approximant-like phones across all dialects of Spanish (Piñeros 2001). The latter (e.g. ball [ˈbɑl]) is an environment where Spanish speakers are less likely to spirantize English stops because, as studies have shown (e.g. Eddington 2011), the utterance-initial environment is a significant predictor of stop-like pronunciations.
The data were submitted to a linear mixed-effects statistical model. Results show that Spanish-speakers do exhibit the transfer of spirantization, however, the advanced learners of English improved significantly, signifying that they were able to suppress the transfer of this allophonic process to some degree. Furthermore, results suggest that the same factors that condition more lenited voiced stops by native Spanish-speakers (stress status, word-position, task formality), also, to a much lesser extent, condition more spirantized voiced stops by native English speakers.
Talks by Max Carey
Papers by Max Carey
Thesis Chapters by Max Carey
In this paper, I use intensity ratios and a qualitative ‘spirantization score’ to re-examine the spirantization of target English voiced stops by L1 Spanish speakers. This paper looks at a subset of data from a larger corpus: six female speakers of Mexican Spanish (3 intermediate and 3 advanced students of English) and a control group of three female speakers of American English. Subjects produced voiced stops in the intervocalic environment (both word-initially and word-medially) and in utterance-initial position. The former (e.g. the ball [ðəˈbɑl]) is an environment where Spanish-speakers are likely to spirantize target English voiced stops because this environment is known to condition more approximant-like phones across all dialects of Spanish (Piñeros 2001). The latter (e.g. ball [ˈbɑl]) is an environment where Spanish speakers are less likely to spirantize English stops because, as studies have shown (e.g. Eddington 2011), the utterance-initial environment is a significant predictor of stop-like pronunciations.
The data were submitted to a linear mixed-effects statistical model. Results show that Spanish-speakers do exhibit the transfer of spirantization, however, the advanced learners of English improved significantly, signifying that they were able to suppress the transfer of this allophonic process to some degree. Furthermore, results suggest that the same factors that condition more lenited voiced stops by native Spanish-speakers (stress status, word-position, task formality), also, to a much lesser extent, condition more spirantized voiced stops by native English speakers.