Introduction: new dimensions in genitive variation

This special issue concerns English genitive variation: the choice between the s-genitive and the of-genitive. It has grown out of the workshop ‘Genitive variation in English’, which was held at the conference of the International Society for the Linguistics of English at the University of Boston in June 2011. While previous research on genitive variation has already unearthed a wealth of factors predicting the variation, the aim of this volume is to add new dimensions to existing parameters. More specifically, the papers presented in this volume extend previous research in four important ways:¹

1. (a) they provide new insights into the effects that the syntactic parameter of weight has on genitive variation and explore different operationalisations thereof

2. (b) they add to our understanding of how a phonological variable can impact on the variation, taking into account the Principle of Rhythmic Alternation (see Schlüter 2005)

3. (c) they provide a new sociolinguistic dimension by investigating change in progress in vernacular Canadian English

4. (d) they extend the discussion of genitive variation to include the much rarer oblique genitive of the type the most recent book of John’s.

For a comprehensive overview of research on genitive variation from its beginnings to the present day we refer the reader to Rosenbach (this volume), who extensively discusses the methodological challenges researchers have faced over the years and who summarises the main findings. Not only does Rosenbach refer to the theoretical frameworks which have been adopted to explain these findings, but she also provides discussion of several controversial issues and points out which questions need to be addressed by future research. In this introduction, we will therefore limit ourselves to introducing the research papers presented in this volume and highlight what is new about the parameters analysed.

In the contribution by Ehret, Wolk & Szmrecsanyi the effects that rhythm and weight have on the distribution of the s-genitive and of-genitive are investigated. While both parameters are in origin speech-based, the authors’ data set is unconventional in that it is based on written historical data, namely the news and letter section of ARCHER, covering the period between 1650 and 1999. Their measurement of weight effects includes both word-based length measures and character-based counts (both in single and multi-constituent measurements), while their notion of rhythm goes back to Schlüter's Principle of Rhythmic Alternation (2005: 18), which predicts that ‘an ideal rhythm alternates between maximally contrasting units, i.e. between stressed and unstressed syllables’. In their regression analysis, which also includes such other factors as animacy of the possessor, semantic relation, definiteness, final sibilancy, time and variety of English, the authors show that weight has a strong effect on the distribution of the variants (being surpassed only by animacy of the possessor), while rhythm comes out as being only marginally important in this particular data set.

Their most important findings with respect to weight effects are, first of all, that short genitives are different from long ones in the sense that length does not have a linear effect on the distribution of the variants. By contrast, for individual lengths, this difference is a quadratic one, which means that longer constituents are more strongly affected than shorter ones. Secondly, they show that character counts add to the predictive accuracy of the statistical model, which means that they should be included as an independent variable. When word counts and character counts are highly correlated, character counts still show a small independent effect in the model. Thirdly, if the total length of the constituents is included in the model, the ratio of the approximated constituent lengths can substitute for individual lengths given that the two are allowed to interact.

If rhythm, as operationalised in terms of the Principle of Rhythmic Alternation, is included in the statistical model, it hardly adds to its predictive accuracy. By contrast, it produces quadratic effects which are theoretically unexpected, showing that speakers do not opt for a consistent alternation between stressed and unstressed syllables. This finding not only suggests that the Principle of Rhythmic Alternation is, if at all, only a minor player in the variation, it also raises the question of whether different operationalisations of rhythm might not have to be taken into account for a fuller understanding of the genitive variation.

In Jankowski & Tagliamonte's contribution, the focus is on how sociolinguistic factors determine the genitive variation. In particular, the authors investigate the distribution of the s-genitive and of-genitive in vernacular Canadian English, basing their research on socially stratified corpora that represent data from speakers of all age groups. In essence, they show that use of the s-genitive has been growing with possessors that represent collectives or organisations. Testing for those factors which successfully predict the variation in a set of (a) non-prototypical human possessors and (b) non-prototypical collectives/organisations and place possessors, they find that the two sets are functionally differentiated. While the s-genitive is stable in apparent time for human possessors and correlates only with the final sibilancy of the possessor (depending on whether the possessor ends in a final sibilant or not), collectives/organisations show an increasing use of the s-genitive in apparent time and are sensitive to the length of the possessor, persistence effects and the occupation of the speaker (blue-collar vs white-collar jobs). Place possessors do not show this development, but it is suggested that these too might potentially form a locus for expansion of the s-genitive. To summarise, Jankowski & Tagliamonte's contribution attests to incipient change in progress in vernacular Canadian English. They show that the spread of the s-genitive has almost been completed for human possessors and that it is gradually encroaching on collectives and organisations.

The volume is rounded off by Payne & Berlage's contribution on the oblique genitive. The authors first provide a quantitative account of the properties typical of the oblique genitive as compared with the s-genitive and of-genitive, showing that the possessor is almost exclusively human, that it is mostly represented by pronouns and that, if it is noun-headed, its mean length hardly exceeds one word. In terms of the determiner in the possessor, it is overwhelmingly indefinite and the core semantic relation that holds between possessor and possessum is the interpersonal one (of the type a friend of mine). The authors then show that the s-genitive, the of-genitive and the oblique genitive are not in complementary distribution exclusively but that variation exists in contexts in which all three constructions function as predicative complements of the clause (as e.g. in he is John's friend / a friend of John / a friend of John’s) and in environments where the oblique genitive and the of-genitive are introduced by the determiner the. While oblique genitive constructions introduced by the definite article are extremely rare in the data provided by the BNC, it is theoretically intriguing to see that these constructions are not confined to pre- or postmodification of the possessum but can occur without any modification (as e.g. in the explosion of hers). Payne & Berlage's qualitative comparison of the s-genitive, of-genitive and oblique genitive additionally concludes that the semantic relations which can hold between possessor and possessum in all three constructions are best described by the following principle of inclusion: the semantic relations compatible with the oblique genitive construction are a subset of those available to the s-genitive, and these latter again are a subset of those that can occur in the of-genitive construction.

To summarise, the first paper in this volume presents an in-depth state-of-the-art survey of research on genitive variation, while the remaining papers explore new dimensions. The second and third papers add to our understanding of how weight, rhythm and sociolinguistic variables impact on genitive variation, while the final paper shows that, at least theoretically, the oblique genitive construction should be included in the comparison where it alternates with the s-genitive and of-genitive.