Systemic functional linguistics in the digital age

Gardner and Alsop adapt a quote from Shakespeare’s Twelfth Night to organize the collection of work in Systemic Functional Linguistics in the Digital Age into three parts: “Texts that are Born Digital: New Digital Genres,” “Texts that Achieve Digitality: Professional Genres Recontextualised,” and “Texts that Have Digitality Thrust Upon Them: Super Powers in Text Analysis.” Their aim is to explore from a Systemic Functional Linguistic (SFL) perspective how advances in computational technologies and “the democratizating reach of new forms of social and mass media” have influenced discourses and “ways of meaning” (p. 1). Given the “long history of SFL interaction with computational linguistics” (Bateman et al., 2017), this volume is a welcome addition to SFL work which utilizes recent computational linguistic tools, specifically in the study of digital(ized) texts.

The first part deals with social media texts, including online comments on newspaper stories (Montecino and Arancibia, Chapter 3 and Cambria, Chapter5), fandom blogs (Sindoni, Chapter 1), and social media features such as “like” and “poke” buttons (Petroni, Chapter 2). I particularly liked Harju’s (Chapter 4) analysis of a YouTube video tribute to Steve Jobs and comments on the video. Her analysis combines the Appraisal framework which is used to identify and measure patterns of evaluative strategies and a “netographic” methodology—“a form of digital ethnography” (p. 68). She considers how the comments align or not with the positive stance taken in the video and also how the comments build “imagined communities” (p. 78) of those who admire Steve Jobs and those who do not. She comments on the dialogic and dynamic nature of the comments and explains that the video tribute is “a multilayered recontextualisation of a lived life” (p. 77).

Part 2 focuses on a variety of professional texts/situations opening with Clarke’s (Chapter 6) diachronic study of the present progressive form of “hear” across different corpora. He suggests that an increase in its use over time may correspond with the introduction of 24-hour news coverage. Bloor (Chapter 7) analyzes medical information websites and discusses the use of lexical and grammatical metaphors to avoid the word “death” or “dying.” She frames her conclusions using the concept of “social distance,” explaining that there is a wide social distance between the website information and the reader. Drury (Chapter 8) presents a very clear explanation of the genre approach to academic writing in an online teaching and learning context. She relates how the teaching–learning materials have changed with changes in technology and points out the advantages and disadvantages of the online platform. She claims that “user evaluation and feedback are essential for improving design for learning” (p. 144). Color screenshots from the teaching/learning materials enhance the content and purpose of this chapter. The term “recontextualisation” is picked up by both Fryer (Chapter 9) and Garcia Riaza (Chapter 11). Fryer relates how a medical journal diagram is cut, modified, and used in new ways in several different Internet contexts. For Garcia Riaza (Chapter 11), “recontextualisation” has to do with the merging of scientific and newspaper discourses into popular science reports in The Guardian newspaper online. She analyzes the lexicogrammatical differences between the first paragraph of the reports (that which attracts readers and from which they choose to continue to read or not) and the rest of the report. Kaltenbacher’s (Chapter 12) focus is on hotel websites and the trend to use more linear text formats by decreasing the number of hyperlinks potentially leading viewers to competitors’ sites. While the content is easy to follow, I felt that a more substantial database would have better supported his claim for this trend.

Section 3 revolves around texts that are digitalized primarily for analytical purposes. The section opens with Miller’s (Chapter 13) description of a corpus-assisted Appraisal analysis of the word “noble” in Shakespeare’s Coriolanus. She finds that corpus tools are useful in pointing out items for “manual probing” (p. 221) but that a manual analysis is needed to understand the symbolic articulation of the theme of a work of verbal art. This observation is echoed by Bayley and Bevitori (Chapter 14) and Luporini (Chapter 16). In their analysis of a corpus of State of the Union addresses, Bayley and Bevitori point out that a keyword analysis can indicate where analyses might concentrate, but that “reading texts from beginning to end” is important to adequately account for changes in meaning over time. In her study of grammatical metaphor in a corpus of articles from the Financial Times, Luporini notes that corpus tools “smooth the way for the analyst” but that “the analyst’s contribution remains essential on both the quantitative and the qualitative plane” (p. 272).

Chapters 15 (Fusari), 19 (Menzel), and 20 (Shagalov and Fine) were highlights of Part 3 for me. Fusari draws the reader’s interest into wanting to test two corpus tools, the UAM CorpusTool and the Halliday Centre Tagger. She clearly outlines some of the problems of “marrying” SFL with Corpus Linguistic approaches in the analysis of large corpora and summarizes the advantages and disadvantages of doing so. An interesting, if perhaps somewhat provocative, point she makes is that while a meaning-based theory of language such as SFL provides a rich set of tools for describing and interpreting language in its social contexts, to take advantage of corpus analytical tools and taggers, SFL scholars need to agree on some kind of simplification. That may not sit well with those SFL scholars who do not wish to compromise on theoretical features and analytical tools. Menzel presents a very careful and thoughtful comparison of ellipsis in cohesive ties in an English-German corpus. She notes the difficulty in finding comparable text types and results and notes the need to manually transcribe spoken data. Shagalov and Fine work through clinical interviews of speakers with schizophrenia and mania, annotate and analyze the corpus using the UAM CorpusTool, and develop 46 linguistic variables to determine the “best subset” of combinations of variables which can more accurately classify the language of speakers with schizophrenia or mania. The importance of this kind of work for clinical diagnosis is made clear.

Perhaps due to the large number of chapters (21 including the introduction) and likely constraints on space, some chapters lacked detailed analyses and discussions. Nevertheless, as a whole, the volume will appeal to a wide range of researchers in discourse analysis due to the variety in content, computational methodologies, and corpus tools. There is also a fairly broad spectrum of SFL concepts and analytical levels and frameworks covered, including lexicogrammar, register, method of development, multimodality, genre, appraisal, and cohesion. One area of work that might have been included, however, is O’Halloran’s (2015) work on “multimodal digital humanities.” For those new to SFL theory, most chapters do not require too much background in the theory. A theme particularly evident in the latter part of the volume is that there remain challenges in developing and adapting computational linguistic tools to text analysis from an SFL perspective because of the multifunctional nature of the lexicogrammar and the meaning-based focus of the theory.