Date: Wed, 4 May 2005 20:02:34 +0200 From: Visnja Josipovic Smojver Subject: Lexical, Pragmatic, and Positional Effects on Prosody in ... Croatian and Serbian
AUTHOR: Smiljanic, Rajka TITLE: Lexical, Pragmatic, and Positional Effects on Prosody in Two Dialects of Croatian and Serbian SUBTITLE: An Acoustic Study SERIES: Outstanding Dissertations in Linguistics PUBLISHER: Routledge YEAR: 2004
Visnja Josipovic Smojver, Department of English, University of Zagreb, Croatia
SUMMARY: This book is a slightly revised version of the Ph.D. dissertation which the author did at the University of Illinois. It investigates the interaction of lexical, pragmatic, and prosodic factors in determining the pitch contours and durational patterns in Belgrade and Zagreb speech. These two dialects are spoken in the capital cities of Serbia and Croatia and thus represent the two respective cognate and mutually understandable pitch accent languages, Serbian and Croatian. Although being essentially an acoustic study, this work is no less a phonological analysis of the prosodic systems of the two dialects under consideration. The theoretical framework of this analysis is autosegmental/metrical phonology combined with an intonational approach using the 'Tone-and-Break Indices' (ToBI) notation. The author relies, mainly for Serbian, on the earlier studies by Lehiste and Ivic (1986) and Godjevac (1999), but largely extends and to a certain point contradicts their findings.
The book consists of six chapters. Chapter 1 includes introductory remarks, offering an account of the historical, typological, and general linguistic background of the two Slavic languages under consideration, as well as the book outline. Chapter 2 presents the results of an acoustic experiment focusing on the alignment of tonal peaks in the two dialects. By this experiment the author establishes that Belgrade and Zagreb speech belong to different prosodic types. While the Belgrade dialect clearly makes use of the four lexical prosodic patterns, based on the combination of the long vs. short and rising vs. falling parameters, its Croatian counterpart lacks them. Consequently, there is a difference between Belgrade and Zagreb speech in the expression of focus. In Belgrade, there are asymmetric patterns of lengthening and pitch-peak retraction, resulting in the exaggeration of the lexical prosodic contrasts. In the speech of Zagreb subjects, however, narrow focus is expressed through a uniform enhancement of pitch and duration affecting the stressed syllable.
Chapter 3 investigates the pitch-range and tonal alignment of the valleys preceding and following the peaks. The alignment patterns discovered here are argued to be relevant acoustic correlates of lexical accents and pragmatic narrow focus. The acoustic data obtained here indicate that in both Belgrade and Zagreb speech the low tonal target preceding the peak (L1) forms a composite accentual gesture, L+H with the peak. However, the phonological function of these tonal valleys differs in the two dialects. In Belgrade speech, the L1 alignment serves to differentiate between the rising and falling lexical patterns (accents). In Zagreb speech it is used to express narrow focus. As for the low target following the accentual peak (L2), acoustic and phonological arguments are offered for treating it as a word-final boundary tone, rather than an integral part of a composite H+L accent. In short, the phonological analysis of the accentual patterns proposed for the two dialects is the following:
In Chapter 4 the author presents another acoustic experiment, by which she investigates the effect of pragmatic narrow focus on word non-initial accents. The data obtained here extend the findings from Experiment I. In Belgrade speech, except for the H tone in the short-rising series, both L- alignment and H- alignment are correlates of accent/vowel length, rather than of pragmatics. The high correlations between the alignment of H and L1 supports the phonological analysis of the rising accents as L*+H, proposed in the previous chapter. In Zagreb speech, by contrast, both L- and H- alignment is determined largely by pragmatic factors. Once again, the acoustic data obtained through the experiment under consideration support the phonological analysis of the two Zagreb pitch accents proposed in the previous chapter as L*+H (expressing broad focus) and L+H* (narrow focus).
Generally, based on the data presented in this chapter, the author identifies four strategies of acoustic cue manipulation for expressing narrow focus: vowel lengthening, pitch expansion, peak retraction and valley retraction. These strategies are used in both dialects, though to different extents.
Chapter 5 reports on yet another acoustic experiment, which serves to compare the expression of narrow focus in utterance-initial and utterance- final position. It is shown that in both dialects the pitch peak alignment in utterance-final position is affected by the upcoming intonational boundary. This is manifested as peak retraction, which is suggested to be a response to tonal crowding under close proximity to the following boundary tones. Notably, despite such additional tonal adjustment, the lexical pitch contrast is still maintained in Belgrade speech under narrow focus in final position.
Chapter 6 is the final chapter, which summarizes the main findings presented in the book and suggests future lines of research.
This book is a valuable source of insights into the principles governing the phonetic implementation of lexical, prosodic, and pragmatic information in pitch-accent languages. It is probably even more interesting as a contribution to Croatian and Serbian dialectology. What makes it particularly praiseworthy in this latter sense is the fact that the phonological analysis of the prosody of the two dialects offered here represents the first extensive study on the subject based on a scientifically rigorous acoustic study combined with equally serious statistical expertise in the interpretation of data. Thanks to these features, this book is free from any impressionism. On the other hand, however, being recorded in perfectly controlled laboratory conditions, the utterances which are used to represent contemporary colloquial Zagreb/ Belgrade speech are necessarily somewhat artificial, as the author herself in a way acknowledges when thanking her subjects for having done the tedious job of repetitively uttering the 'boring' sentences. Of course, the somewhat artificial nature of the corpus is perfectly normal for this type of study and reference to it is not meant as a criticism, but, rather, as a remark for the reading public, who perhaps should have been made more aware of the limitations of the scope of this study. Likewise, it should be noted that there is much more to the pragmatics of intonation than expressing the two types of focus studied here. In addition, for non- native speakers of Croatian who are potential readers of this book it would be useful to stress that the Zagreb dialect is the prosodically simplest variety of the Croatian language in that it lacks the four lexical prosodic patterns ('accents'), as opposed to many other varieties, including the standard type of Croatian pronunciation. Like Belgrade speech, Standard Croatian has the four accents, but nevertheless unmistakably sounds prosodically different from Belgrade speech or any other variety of Serbian speech for that matter. This suggests that there must be other major factors giving identity to particular varieties of Croatian and Serbian, apart from the presence vs. absence of the lexical accent contrast and the resulting consequences in the strategies of pitch manipulation. To make pitch-accent prosody even more elusive, the findings obtained by this acoustic study still need to be confirmed by perceptual studies, as Smiljanic correctly observes in her final chapter.
Although a somewhat more explicit account of the limitations of the scope of the book could have made the reading public more aware of the extreme complexity of Croatian and Serbian prosody, the above remarks are certainly not meant as criticism. Quite to the contrary, it must be observed that everything set out in the title of the book has been dealt with thoroughly, correctly, and nicely. However, in order not to sound too complimentary, at this point I will make a little critical comment concerning a claim made in 1.2.1. in the context of a general description of Croatian and Serbian. It must be either a misinterpretation of some source or a major slip of the pen when it is stated that the Glagolitic alphabet 'continues to be used in parts of Dalmatia to this day'. To the best of my knowledge, this is not true. The Glagolitic alphabet is no longer used for any practical purposes in any part of Croatia. If I am wrong, I would be grateful to anyone who could name a locality in Dalmatia where it is used in normal communication and thus clarify the point. At any rate, even if this piece of information is incorrect, this little detail cannot seriously detract from the value of this precious piece of work. Summed up in two words, my appraisal of Rajka Smiljanic's book is still: 'Well done!'
Godjevac, Svetlana (2000) Intonation, word order, and focus projection in Serbo-Croatian. Unpublished Ph.D dissertation. Ohio State University, Columbus.
Lehiste, Ilse & Pavle Ivic (1986) Word and sentence prosody in Serbocroatian. Cambridge, MA: MIT Press.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Visnja Josipovic Smojver is an Associate Professor in the Department of English, University of Zagreb, Croatia. She teaches undergraduate and graduate courses of phonetics and phonology, as well as an undergraduate course entitled 'Accents of English'. Her major research interests include the phenomenon of foreign accent, contrastive English/ Croatian phonetics and phonology, and, most recently, the speech of twins. She has published a number of works on these topics and is also the author of a textbook of phonetics and phonology for university students, Phonetics and Phonology for Students of English, Zagreb: Targa (1999).