This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
As the author states in the preliminaries, this monograph is intended for scholars or university students who want to become acquainted with salience as a theme in linguistics, and, particularly, in sociolinguistics. Rácz believes that, so far, the concept of salience in sociolinguistics has been regarded as having a wide range of properties, from being equated with high token frequency to being attributed an extra-linguistic property (see Siegel 2010: 120-127 for examples). In sum, the author argues that it has been loosely defined. Thus, one of the general objectives of this book is to give a narrow definition of salience in sociolinguistics. Among the other objectives are: putting forward a singular scheme to make salience operational, and ascertaining the theoretical relevance of salience, particularly, in the field of language change, which is understood here as the way that novel variants are transmitted in a speech community.
With these objectives in mind, the book is divided into three parts (excluding the preliminaries chapter). Part I (pp. 23-54) reviews for us how the notion of sociolinguistic salience has been practiced so far, builds up a narrow definition of salience, and introduces a method to make salience operational. Part II (pp. 55-128) consists of four case studies (definite article reduction, glottalisation in the south of England, hiatus resolution in Hungarian, and derhoticisation in Glasgow) in which the mechanism of making salience operational is used. Part III (pp. 129-156) puts forward the importance of the notion of salience to a theory of language change.
This section can be divided into four subsections. In the first one (pp. 23-25), Rácz provides an overview on how salience has been traditionally defined in phonology, as well as in morphology. He argues that, whereas in phonology a salient feature is one that stands out among others and that will be noticeable both for linguists and language users, in morphology, a pattern can be considered more salient if it is less marked from the point of view of complexity, thus, “the more regular the pattern, the more it adheres to the structure, the more salient it will be” (p. 24), in comparison to others, of course.
In the second section (pp. 25-31), Rácz outlines how the concept has been generally framed in the field of sociolinguistics, where it “aspires for a status as a stand-alone theoretical concept” (p. 25). He claims that it has been interpreted in two ways. On the one hand, it can be used to allude to linguistic variables that bear social indexation, and thus behave differently, for example, in language contact situations. In this sense, salience is not more than a synonym for ‘marker’ (cf. Labov 1972). On the other hand, it can be a tag applied to variables that have perceptual and/or cognitive prominence, that is, salience has an external basis and it cannot be justified only by social dynamics. According to Rácz, opting for the second possibility enables us to find a general ground for salience, whereas opting for the first one means that we give up on finding a universal definition of it.
In the third section (pp. 31-35), Rácz looks at salience in visual cognition, where it is seen as a notion connected to surprise in human perception. To put it simply, one could consider an entity surprising (thus salient) if its presence has a high information value compared to its surroundings. In other words, according to the field of visual cognition, an entity is salient when its presence among other entities is unexpected.
As we can see in the fourth section of Part I (pp. 36-54), Rácz turns to salience in visual cognition to implement his own definition of salience in sociolinguistics, and to subsequently make it operational. Based on his immersion in the aforementioned field, he defines it as follows: “A segment is cognitively salient if it has a large surprisal value when compared to an array of language input” (p. 37). One of the main points that the author underlines is that salience is a relative concept, that is to say, it is always interpreted in a context. A linguistic variable can be considered salient when its behaviour in a dialect is different enough from another one (usually the norm) to be surprising for language users of the latter. Particular emphasis is also placed on the fact that socially salient variables are always cognitively salient, but that the opposite is not always the case. As for making salience operational, one must measure surprisal through the comparison of the transitional probabilities of the segmental realisations of the variable under study, that is, through the contrast of differences in the distributions of the segmental realisations of a variable in two dialects.
The first case study presented by Rácz is on definite article reduction (DAR) in the North of England (pp. 55-69), which he defines as a glottal stop variant of word-initial “the”, as in, for example, ‘the day’ [ʔdeɪ] or ‘the order’ [ʔodə]. The author argues that even if glottal stops are already present in Standard English (that is, they exist in the speech string and thus will not be totally surprising for listeners), DAR’s salience can be explained by its unlikelihood of occurrence in the aforementioned position. Therefore, it can be said that DAR’s salience lies in the distribution of its realisations, not in its presence. The author also points out that DAR’s cognitive salience makes it an ideal candidate for being a marker. Indeed, he stresses that it has been used for a long time in England as a feature to illustrate typical Northern speech; thus “one can safely says that DAR is a dialectal marker, used to express social or regional differences” (p. 59).
Rácz continues with the case of the glottalisation of /t/ in Southern English (pp. 71-86), a well-documented feature that has been reported by various authors (e.g. Williams & Kerswill 1999). Nowadays, it can occur in all positions, but it is vigorously avoided among the middle class and upper middle class before vowels -- both word-finally and word-medially. According to the author, one possible explanation for this avoidance is that, unlike pre-consonantal glottalisation -- a phonetically natural phenomenon whose occurrence has been reported for a while now, pre-vocalic glottalisation is rather new. Thus, its transitional probability is basically equal to zero. In other words, any glottal stop in this environment is extremely unexpected, thus it has a large surprisal value compared to pre-consonantal glottalisation. Consequently, the author suggest that the glottalisation of /t/ has developed salience as it spread to a new environment.
The third case study deals with hiatus resolution in Hungarian (pp. 89-100). Rácz argues that while it is obligatory in vowel clusters including [i], and variable in clusters with [eː], it is quite innovative in clusters with [ɛ]. Hence, for Educated Colloquial Hungarian (ECH) speakers, who are said to be non-innovative, the third hiatus resolution pattern is cognitively salient, since its occurrence has a large surprisal value compared to the first two hiatus resolution patterns. In order to evaluate if cognitive salience translates to social salience, Rácz performed an attitude study with native speakers of ECH. The results show that ECH speakers are aware of the innovative hiatus resolution pattern regarding the conservative one. According to the author, the innovative hiatus resolution pattern’s salience emerges from a difference in transitional probabilities, which “renders the pattern less familiar to the listeners” (p. 99). The low probability of occurrence for the pattern was confirmed by data that he draws from the Hungarian Webcorpus.
The last case study presented by Rácz deals with derhoticisation (the vocalisation of coda /r/) in Glasgow (pp. 101-128), which is, probably, one of the most widely researched sociolinguistic features in English studies (cf. Stuart-Smith et al. 2007, for example). Rácz shows that this feature lacks social salience despite meeting the requirements for being sociolinguistically salient, namely, it displays differences in segmental distributions and clear social stratification; and that this dearth of salience can be linked to an absence of cognitive salience, that is to say, to a lack of surprisal in its distributions. The latter can be tied to the important extent of inherent variation exhibited by derhoticisation. Therefore, this is a sort of counterexample to the previous case studies.
This section considers the relevance of both cognitive salience and its social indexation to linguistic theory. One of the questions that Rácz puts forward is whether we should assume a distinction between salient and non-salient variables outside the field of sociolinguistics. The author argues that cognitive salience has strong consequences to models of linguistic competence, in that it grants us to depict salient variation and its counterpart, non-salient variation, and explain why the latter receives no consideration from language users. Rácz also shows salience’s relevance to linguistic change, namely sound change. He shows that for the case of network-based models of change (cf. Milroy 2007, Eckert 2000, among others), which pay considerable attention to speaker/variant prestige in sound change, the theory of salience that he defends provides additional support, because it puts at one’s disposal self-reliant tools to confirm the salience of a variable.
This book is sure to appeal to scholars and students that are particularly interested in the notion of salience in sociolinguistics. It introduces a corpus-based (i.e. empirical) approach to appraising the salience of phonological variation -- both at a cognitive and a social level, and contributes a procedure and case studies for this approach. Moreover, it is also a work that can be appealing for researchers of fields such as linguistic change and general linguistic theory, since that, in the third part of the book, the author took the trouble to show how the notion of salience that he puts forward can be used in these fields. It can thus be said that the author has achieved the goals set at the beginning of this book.
In spite of that, this book is not without a few (very) minor issues. For one thing, Rácz introduces a new terminology in place of such conventional terms as ‘indicators’ and ‘markers’ (cf. Labov 1972), for ‘cognitive salience’ and ‘social salience’, respectively. Although, thanks to the author’s entrance in the field of visual cognition, this is totally justified, and his approach ends up making these concepts’ understanding a lot easier, and their operationalization more practical. The other minor issue might be that of all the case studies presented, none analyzed salience in so-called standard dialects. Indeed, the four case studies presented by Rácz describe how a feature from a nonstandard dialect is salient from the perspective of a standard one. I am really curious to know how the author would use these concepts and their operationalization in cases that adopt the opposite perspective, that is to say, in cases where salience is analyzed from a nonstandard dialect and its speakers.
These minor issues aside, this book is an interesting and useful resource for corpus-based research in the field of sociolinguistics.
Eckert, Penelope. 2000. Linguistic variation as social practice: The linguistic construction of identity in Belten High. Hoboken: Wiley-Blackwell.
Labov, William. 1972. Sociolinguistic patterns. Philadelphia: University of Pennsylvania Press.
Milroy, Leslie. 2007. Off the shelf or under the counter? On the social dynamics of sound changes. Studies in History of the English Language 3. 149-172.
Siegel, Jeff. 2010. Second Dialect Acquisition. Cambridge: Cambridge University Press.
Stuart-Smith, Jane, Claire Timmins & Fiona Tweedie. 2007. Talkin’ jockney? Variation and change in Glaswegian accent. Journal of Sociolinguistics 11 (2). 221-260.
Williams, Ann & Paul Kerswill. 1999. Dialect levelling; change and continuity in Milton Keynes, Reading and Hull. In Paul Foulkes & Gerard Docherty (eds.), Urban voices. Accent studies in the British Isles, 141-162. London: Edward Arnold.
ABOUT THE REVIEWER:
Víctor Fernández-Mallat is currently an associate researcher for the Research training group GRK DFG 1624/1 “Frequency effects in language” and a lecturer for University of Basel’s Institut für Iberoromanistik. His research interests include sociolinguistic variation and change, dialects in contact, and Hispanic linguistics. His current research project focuses on the interplay of convergence and divergence processes in dialect contact situations.