Abstract
Since its launch in the early 1990s, the World Wide Web has provided a fruitful ground for discourse analysis. A substantial body of literature, covering diverse topics such as Internet language, digital identity, hypertext genres and multimodal communication, has been produced, to which Hoffmann’s Cohesive Profiling is a welcome addition. The central premise of the book is a simple yet elegant one: ‘meaning-making is a negotiable practice with texts consisting of operative devices to coordinate the meaning-making activities of different interlocutors. The semiotic currency with which they mainly organize their discursive moves is textual connectivity, i.e. cohesion’ (p. 9). In the space of 10 chapters, the book explores, in the theoretical tradition of text linguistics (Halliday and Hasan, 1976; Schubert, 2008; Tanskanen, 2006) and through the methodological lens of corpus linguistics, the ‘cohesive profiles’ of 10 personal blogs written in English. By focusing on quantitative and qualitative cohesive profiling, the book offers interesting insights into the textual collaboration between authors and users on the blogosphere as well as the challenges involved in analysing and theorizing classic linguistic categories, cohesion (and coherence) in this case, in the hypertext environment.
After outlining the background, themes and central questions of the research in Chapter 1, the book proceeds to an extensive discussion of the linguistic characteristics of weblogs in the following three chapters. In Chapter 2, a working definition of a weblog (pp. 18–19) is first mapped out through a comprehensive review of the literature. The basic organizational structure of a blog is then explained, which consists of first-level elements, that is, entries and comments (the sole focus of the book) and second-level or paratextual elements such as tags and links. The focus of Chapter 3 is the categorization of blog genres. It is in this chapter that readers will have a first glimpse into Hoffmann’s commitment to empirical rigorousness. The generic types are identified based on the functions as stated by the blog authors in an online survey. The topical consistency has also been assessed through random word frequency counts of different entry sections in each blog. Subsequently, four types of personal blog genres or ‘super-genres’ (p. 38) are recognized – diary, friendship, career and commercial. Chapter 4 starts with a review of the notion of hypertext – its origin and history, and its organizational patterns and textual characteristics, followed by a discussion of the roles authors and readers play in creating and consuming hypertext. While the discussion of hypertextuality and the prosodic discussions of cognitive theory of memory are engaging and informative, the chapter bears only loose connections with the preceding and following chapters. Many concepts discussed are not elaborated until Chapter 9. A more explicit forward referencing or perhaps a minor change in the structure of the book might have provided a smoother navigation for readers, although the book otherwise operates with a clear and explicit logic.
Chapters 5 and 6 introduce, in turn, the analytical categories of cohesion and the corpus – the Augsburg Blog Corpus (AWC), consisting of 143 entries, 137 comment sections and a word count of 123,242 tokens. These two chapters showcase the strength of the book. In developing a cohesive framework in Chapter 5, the author navigates through classic literature and engages with each category and subcategory of cohesion (i.e. grammatical and lexical) in great theoretical depth and with acute critical awareness. The detailed recount of the analytical procedure in Chapter 6 provides a particularly fascinating read. Hoffmann’s surprising candidness about the detours and complications in combining computer-assisted auto-analysis using two corpus tools – an automatic concordancer (Antconc) (auconc©) and POS-parser (TreeTagger©) – with manual analysis will no doubt resonate with many, as developing a computer-assisted and empirically reliable mixed-methods approach remains a key challenge for discourse studies.
The ‘cohesive profiles’ of the personal blogs in the AWC are presented in Chapters 7 and 8, with the former focusing on grammatical cohesion and the latter on lexical cohesion. The empirical findings offer insights into the texture of personal weblogs, for example, high frequency of collocation, high frequency of personal and demonstrative reference and prominence of total recurrence. While some of the individual findings are revealing in their own right, for example, that visual semiosis triggers demonstrative reference more than verbal referents, the majority of the findings perhaps need to be interpreted against findings from comparable studies. Nevertheless, the central conclusion – that the AWC blogs demonstrate quantitatively similar cohesive profiles across blog entries and comments, both of which are closer to written monologue (academic prose) than spoken dialogue (face-to-face conversation) – offers a convincing refute to the dominating view of personal blogs as some form of dialogue between authors and users.
In Chapter 9, Hoffmann ventures beyond corpus-based discourse analysis and actively engages with cognitive theory, specifically Tulving’s (1983, 2002) works on episodic and semantic memory, in order to explain ‘how cohesive collocations give rise to the negotiation of common ground between interlocutors in blog texts and invite an interactive attribution of participation roles by interlocutors’ (p. 12). Here, in this theoretical move lies the biggest potential of the research, which understandably does not have the chance to be fully developed in the book. After illustrating the main cognitive frames (e.g. Technology, Entertainment, Travel and Kinship) observed in the AWC blogs and their respective collocational pairs, the chapter ends with the introduction of a theoretical framework for interpreting interaction and the negotiation of knowledge in blog space (an overview is provided in Figure 41 on p. 196). The discussion of this highly sophisticated framework feels rushed, failing to elucidate and reveal the full implications of the theory. 1 There is also a lack of explicit connection with the empirical findings in the previous two chapters. The chapter leaves readers wanting, which Hoffmann is certainly aware of when he closes the book in Chapter 10 with a promising invitation for the future: ‘The key to a better understanding of blog cohesion and discourse is therefore likely to be found at the crossroads of discursive and cognitive interaction’ (p. 215).
Cohesive Profiling will no doubt stand out among similar titles for its meticulous scholarship, its devotion to empirical rigour as well as its willingness to take theoretical risks. Paradoxically, the strength of the book also gives rise to its limitations. Well-versed in the traditional linguistic notions of text and analytical categories, the book (deliberately) overlooks elements of the paratextual plane, such as the functions of tags and searchability, which invite different textual practices (e.g. Zappavigna, 2011) and produce inevitably different notions of cohesion and coherence. Similarly, in its preoccupation with providing a theory of the classic author–reader dichotomy, the book disregards the increasingly critical role semiotic technologies (i.e. blog platforms such as WordPress or Tumblr) play in mediating the author–reader interaction (Zhao et al., 2014). The limitation described here is, ultimately, a reflection not on an individual researcher but on the current state of discourse analysis as a field. If ‘the most fundamental patterns of scientific discovery is the revolution in thought that accompanies a new body of data’ (Goldenfeld and Woese, 2007: 369), the ever-evolving body of web-based texts should offer us a chance to re-examine and challenge some of our basic assumptions such as cohesion/coherence rather than simply function as a testing ground for established theoretical categories. So, it is perhaps high time to take a step back and observe emerging digital discursive practices without a tainted theoretical lens before we can re-engage with classic linguistic theories in a critical way.