EDITORS: Daniel Recasens, Fernando Sánchez Miret; Kenneth J. Wireback TITLE: Experimental phonetics and sound change SERIES TITLE: LINCOM Studies in Phonetics 05 PUBLISHER: Lincom Europa YEAR: 2010
Erin Ament, Department of English, College of William and Mary
SUMMARY This is an edited volume of articles based on a workshop on sound change that took place in Salamanca, Spain in May 2009. The workshop was focused on gaining a better understanding of sound change through the study of articulatory, acoustic, and perceptual data. The majority of the articles combine a discussion of historical sound changes with experimental data that either supports or contradicts the existing historical analyses.
The first article, by Silvia Calamai and Irene Ricci, is titled “Speech rate and articulatory reduction in Italian alveolar and velar nasal + stop clusters”. It describes the use of electropalatography to examine regressive place assimilation in nasal + stop clusters in Italian. The interaction between speaking rate and degree of coarticulation in these clusters was tested, along with the effect of the location of the primary stress in the word, the location of the cluster (word internal versus across a word boundary), and the voicing of the post-nasal stop. The results showed that there was assimilation in all three places of articulation and that the consonants were indeed produced for a shorter duration in faster speech. Interestingly, the articulation of the consonants was not as consistently affected by the speaking rate. Speaker-specific strategies regarding the amount and the location of contact during the stop, particularly for the alveolar clusters were found. The presence of these strategies sheds light on the fact that even carefully controlled laboratory speech is subject to a high degree of variation based on how the speakers interpret the instructions given. This raises the issue of the perhaps invalid dichotomy between clear laboratory speech on the one hand and natural speech on the other. The article was clearly written and easy to follow.
The second article is Chiara Celata’s “Rhotic retroflexion in Romance. Acoustic data for an articulation-driven sound change”. It investigates the retroflexion of /t(:)r/ clusters in Sicilian. The article states that /t(:)r/ clusters are often realized as [ʈ(:)ʂ] in spontaneous speech. This paper focuses on the origin of retroflex consonants in Romance languages as part of a diachronic process based on phonetic variability. This is in contrast to the majority of research on retroflexes in these languages which view them as a typological phenomena. Five speakers were recorded and acoustic measurements of the duration of the affricate, the intensity of the frication, the spectral characteristics of the frication, and the F3 and F4 values of the preceding and following vowels were measured. While the articulatory data is extensive in this article, the link between this data and the possibility for an articulation based diachronic change in the cluster is not made completely clear. The author argues that the cluster changed from a stop-trill combination to a retroflex-fricative combination due primarily to the transformation of the trill to a retroflex. This contrasts with the more assimilation-based proposal of Sorianello & Mancuso (1998). The lack of standard IPA for the Italian examples is somewhat frustrating, but overall the articulatory data is well presented.
The third article by Juan Felipe García Santos is an updated summary of a book by the same author in 2002. In “Experimental analysis of some acoustically driven phonetic changes in Medieval Spanish”, the author argues that historical analyses of sound changes should be supported with data from current phonetic and acoustic research. The article goes on to detail changes in 16th century Castilian Spanish and draws parallels between these historical changes and a current change in progress in Castilian palatals. The experimental data on the change in progress is somewhat hidden in the historical analysis and is addressed relatively briefly in the article, but the general claim is that the experimental evidence suggests a single process for the 16th century change of /b/ to /v/ and the current change of /j/ to /x/. Both changes are claimed to be due to a perceptual change resulting from the shortening of the segments as suggested by A. Alonso (1967-1969).
The fourth article is Daniel Recasens and Aina Espinosa’s “A perceptual analysis of the articulatory and acoustic factors triggering dark /l/ vocalization”. This article examines the perception of dark /l/ as /w/ in Catalan. The authors looked at both an articulation-based hypothesis related to the loss of clear alveolar contact and an acoustic-based hypothesis based on the similar F2 values of dark /l/ and /w/. While both hypotheses predict the presence of dark /l/ word finally and before labials and velars, only the acoustic hypothesis predicts that dark /l/ will also be present before alveolars due to the similarity in F2 between /w/ and dark /l/. A combined production and perception study found that the dark /l/ target was most often perceived as a /w/ when it was produced with a small amount of alveolar contact and a low F2 value. The paper was very clearly written and convincing in its argumentation.
The fifth article is “The effect of word final unstressed high vowels on stressed vowel duration and its consequences for metaphonic diphthongization in Southern Italian” by Fernando Sánchez Miret. This clearly written paper presents preliminary results of an experiment focusing on the possible origin for the change of mid-open vowels to diphthongs in Northern Calabrian Italian. This change occurred when the mid-open vowel was followed by an unstressed high vowel. A general claim that long stressed vowels will frequently become diphthongs is evaluated as an alternative to Schürr’s (1936) hypothesis that the diphthongization is due to anticipatory raising and long distance assimilation. The results from three speakers are discussed and the general trend seen in these speakers is that phonemically long stressed vowels before low unstressed /a/ are shorter than before unstressed /i/, supporting the stress based origin for the sound change.
The sixth article, by Kenneth Wireback, is titled “A reexamination of the palatalization of Latin /kt/ in the light of phonetic research”. In this paper a historical account of the palatalization of the Latin /kt/ cluster is reanalyzed with respect to current experimental evidence regarding the gestural demands of producing the cluster itself. The author states that historical accounts often propose forms that contain the palatal guide /j/, yet forms containing this segment are largely unattested, making such accounts problematic. The author argues instead for an account based on gestural blending and regressive assimilation due to acoustic factors. Several articulation studies done by others are cited to support this claim. In general, the claim made by the author about the use of the experimental evidence to help validate historical analyses is well taken, but the extensive footnoting and the writing make the paper somewhat hard to follow.
The seventh and final article is by Marzena Żygis, “On changes in Slavic sibilant systems and their perceptual motivation”. The article clearly presents extensive articulatory, acoustic, and perceptual data about sibilants in several Slavic languages. The results of these studies show that perceptual based accounts do well at explaining the distribution of sibilants in Slavic languages. In languages with complex sibilant inventories, (i.e. those containing more than one post-alveolar segment) one of these segments is very likely to be a retroflex. The author argues that this is because the retroflex /ʂ/ is gesturally and acoustically distinct from /ʃ/ and /ç/ and provides good evidence in support of this argument. The article ends with an argument for the use of perceptually-based features, such as sibilant tonality, in the description of sibilants to help explain the distribution of these segments in a language’s phonemic inventory. The article is clearly written and makes excellent use of several types of experimental data to argue its point.
EVALUATION This volume fits well within the current trend in linguistics of using laboratory and experimental methods to evaluate, test, and expand upon more abstract analyses. It also shows the value of working at the intersection of several fields, in this case phonetics, phonology, and historical linguistics. Current work on the phonetics / phonology interface is beginning to challenge the long standing tradition of using careful laboratory speech to test phonological theories due to the difference between laboratory speech and conversational speech. This challenge is supported by the work reported by Calamai and Ricci who note that there was a fair amount of variation amongst their speakers even in a laboratory setting. While such variation may be problematic from the perspective of drawing concrete generalizations about language behavior, it serves to highlight why language change happens and the importance of looking at acoustic data from a number of individuals.
The article by García Santos presents a well stated argument for the role of acoustic data in evaluating rival theories of historical changes, but the reporting of the data collected by this author leaves something to be desired. In contrast, the article by Recasens and Espinosa and the article by Żygis are excellent examples of using laboratory techniques to test theories of sound change. As a whole, the volume is cohesive and shows the growth of experimental research in linguistics and the importance of using multiple approaches in any sort of linguistic analysis. One problem with the volume is the somewhat inconsistent use of IPA in some of the articles, making the arguments hard to follow if one does not speak with the languages being discussed. Also, for someone who is not familiar with the literature on historical sound changes, some of the arguments for specific theories may be hard to follow. However, the potential for future work using laboratory techniques to aid in historical analyses is vast and this volume is a good step in the direction of a more integrated approach to linguistic analyses.
REFERENCES Alonso, Amado. 1967-1969. De la pronunciación medieval a la moderna en español. Madrid: Gredos.
Schürr, Friedrich. 1936. Umlaut und Diphthongierung in der Romania. Romanische Forschungen 50: 275-316.
Sorianello, Patrizia & Antonella Mancuso. 1998. Le consonanti retroflesse nel consentino: analisi preliminare. In Atti delle VIII Giornate di Studio del Gruppo di Fonetica Sperimentale, Pisa 18-19 dicembre 1997. Pisa: Edizioni ETS: 142-154.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Erin Ament is currently a Visiting Assistant Professor of Linguistics at
William and Mary. She is interested in the intersection of experimental
phonetics, language acquisition, and psycholinguistics.