1Posters are one of the research process genres enacted during the “macrogeneric event” (Shalom 2002: 52) of a conference, along with oral presentations, plenaries, round tables, etc. They have long been a staple feature of scientific conferences, but are only now beginning to appear at meetings in the softer fields of social sciences and the humanities. While countless poster templates and tips can be found on the internet and discipline-specific guidelines in many specialised journals, apart from some studies on the use of posters for teaching purposes, the genre has, rather surprisingly, received very little attention from discourse analysts. As pointed out by the author of Academic posters in her earlier study (D’Angelo 2011: 27):
the linguistic investigation of academic posters and the construction of multidisciplinary corpora including this genre is still in its infancy. In particular no analysis has yet been conducted of the semiotic code of text and images, that is of how posters exploit the visual as well as textual resources in their content.
2The present book aims to fill this gap through the textual and visual analysis of 120 posters in three disciplines: high-energy particle physics, law, and clinical psychology. The author is currently doing, or has very recently completed, doctoral research on academic posters at the University of Reading, UK, and has already published three articles on the topic: a programmatic article outlining an analytical framework for the study of posters (D'Angelo 2010), and two others on small corpora of posters.
3The book comprises six chapters. After a brief introduction giving the rationale for the study, chapter 2 is a classical review of the literature, covering academic discourse, academic genres, the poster genre, metadiscourse, multimodality, some general principles of corpus design and the four research questions asked, namely: the cross-disciplinary differences between the posters in the three corpora in terms of a) word count, orientation and layout; b) textual metadiscoursal resources (both interactive and interactional); c) visual metadiscoursal resources (interactive only). The final question, drawing on interview data with some of the poster authors, concerns the motivations for the cross-disciplinary differences found. Chapter 3 describes the data collected. To cover a spectrum of disciplines, the three subcorpora correspond to disciplines situated at different points on the hard-soft continuum, from particle physics (hard) at one pole to law (soft) at the opposite pole, with clinical psychology midway between the two. To ensure that the corpus was representative, the author first carried out a written survey of experienced and novice academics (32 respondents worldwide) to ascertain by whom and how often posters were presented in their field, and whether they used poster templates or not (online or provided by their university). The posters themselves were collected online in accordance with the survey results.
4Chapter 4 describes the methodology and framework of analysis used, based on the author's earlier programmatic article. The study focuses only on the posters themselves –the oral poster discussions are not included in the analysis– and applies Hyland's (2005) metadiscourse model, which distinguishes between interactive and interactional metadiscoursal resources, to analyse the text on the posters, and Kress and Van Leeuwen's (2006) grammar of visual design to categorise the interactive visual resources. Interactional visual resources (salience, size of frame, and perspective) are not considered. Chapter 5, the longest in the book, reports the results for each of the three subcorpora, and concludes with a cross-disciplinary comparison of the results. The final chapter discusses the results in the light of the four research questions asked at the outset, and concludes with some comments on the limitations of the study and recommendations for further research. In addition to the six chapters, the book also contains seven Appendices that reproduce the survey questions, transcripts of extracts from the interviews with poster authors, the list of textual metadiscourse markers searched for, and thirty tables giving the detailed breakdown of the occurrences of the textual metadiscourse markers found in the corpus and that are presented more succinctly in the results chapter (Chapter 5). This is followed by a reference section of 468 entries and an index of notions.
5In view of the paucity of research on academic posters, the present book could therefore have filled a significant gap in our understanding of academic genres and knowledge communication; it does have some strengths. The corpus is sufficiently large to enable generalisations to be made about the three disciplines, and has been carefully compiled to ensure its representativity. The objectives and analytical framework chosen are clearly described, and the coverage of the literature is extensive, though the author could have mentioned the study by Li (2014): while a much smaller-scale study, as befits an MA dissertation, it was conducted on exactly the same lines as D'Angelo's research so could have provided a relevant point of comparison for the results section. In terms of readability, Academic Posters contains a sizeable number of colour reproductions of the posters which enliven the presentation and illustrate the points being made. There are, however, some serious flaws in the book, which, disappointingly, cannot as a result be considered as a reliable scholarly contribution to ESP.
6The most glaring weaknesses are in the results chapter (Ch. 5), which contains innumerable discrepancies and contradictions between what is claimed in the text, given in numerical form in the tables, and listed in greater detail in the corresponding tables in Appendix 7. As a result, the reader is at a loss to know which figures and percentages are the correct ones and cannot lend any credibility to the author's conclusions as they are apparently based on flawed data. For reasons of space, a few examples will have to suffice here to illustrate this problem, which is pervasive throughout the chapter. On page 150, concerning textual metadiscourse in the particle physics subcorpus, the text declares that there is “a total of 1,389 instances of metadiscoursive devices (947 interactive resources plus 442 interactional resources) […]. As Table 5.1 shows, High Energy Particle Physicists tend to use far more interactive features than interactional.” This is in total contradiction with what is claimed later in the chapter for the physics subcorpus (“textual interactive and interactional resources are used in equal quantity”, p. 173) and with the data in Table 5.1 itself on page 150, where the number of interactive resources is 444, not 947, and both the raw and normalized frequencies of interactive and interactional resources are practically identical (444 & 1,372 vs 442 & 1,363 respectively). The story does not end there: Appendix 7 gives the full details of the counts in chapter 5, so the totals of the two should tally; in fact, raw counts of interactive resources in the appendix total 934 (neither 947 nor 444…), and normalized counts come to 2,623 (instead of 1,372 as in Table 5.1), and there are some unaccountable discrepancies between the totals for individual categories (163 occurrences of Transitions in Table 5.1 but 630 in the Appendix, for instance). The same holds for the figures in Table 5.5 on textual metadiscourse in the Law corpus, which do not correspond at all to the occurrences in Tables 11 to 20 in the Appendix, neither in the overall totals nor the figures for individual devices. Similar errors plague the counts of visual features on the posters: the number of occurrences of Framing in Table 5.8 is 34 but the text on the facing page asserts that “framing recurs 94 times” (p. 191) and that it is “the most recurring visual resource” whereas it comes fourth out of five resources in Table 5.8. These inconsistencies between tables and text are also found between tables and barcharts: while Framing accounts for 15.8% of resources used in Law posters in Table 5.8, the figure for the same feature unaccountably jumps to 35% in the barchart in Figure 5.15, and both the tables and barcharts comparing the three subcorpora at the end of the chapter are riddled with similar mistakes. It would be tiresome to pursue this catalogue of errors further, as a detailed list of the mistakes runs to several pages. In some cases the source of the error can be pinpointed: the author has confused percentages and frequency figures, or subtotals and figures for individual items, or mixed up the rows in the table, but in the majority of cases, the reader is left bewildered.
7The front matter indicates that "This publication has been peer-reviewed." It is difficult to comprehend, however, how all these inaccuracies in the calculation and reporting of the results managed to go undetected during the peer review and proof-reading process. The absence of proof-reading is also evident in the numerous typos (over 200 by my reckoning) scattered throughout the book: spelling and punctuation mistakes, lexico-grammatical errors (tense usage in particular), missing words in the text, missing totals in most of the tables in Appendix 7, incorrect gender reference to authors (Betty Samraj, p. 34), sentences printed twice (p. 238), a wrong title for Table 23 in Appendix 7 which should be Code Glosses, not Endophoric Markers, Tables 4.1 and 4.3 later repeated as Tables 4.5 and 4.6, etc. In the Reference section, some references are missing (e.g. Mann & Thompson 1988; Fahnestock 2003), others are printed twice (Hay & Thomas 1999; Rowley-Jolivet 1999), the reprint of (Miller 1984) is given two separate entries, the alphabetical order of entries is not always respected, and authors’ names are on occasion garbled: G. Garzone & J. Archibald (Eds.) is entered as G. a. Garzone, James (Ed); in Delin et al. 2002, author Allen P. appears as Patrick A.; editors E. Ventola, C. Shalom & S. Thompson are printed as T.S. Ventola E. & Shalom C. I also take the opportunity of this review to correct an inexact statement made by the author on page 54 about one of my own publications: Rowley-Jolivet (2002) did not "seek to identify and classify different visuals and communicative strategies present in poster presentations," but in conference talks. Moreover the author appears to have confused the article referenced in the bibliography –which did not deal with visuals– with another one published the same year that did. These types of errors are often encountered in the initial version of theses (sent to members of the jury before the viva itself) and are subsequently corrected before permission to print the thesis is given, but are totally unacceptable and unwarranted in a published book.
8Indeed, as is perhaps clear from the summary of the book's contents given above, the work in many ways resembles a thesis more than a book. Its overall structure is the canonical structure of a PhD dissertation, with first the research questions, then the literature review, the material (the corpus collected), the methodology, and finally the results, discussion, and closing comments on the limitations and perspectives of the research. The author's stance is also more akin in places to what is expected from a doctoral student than from a book author, with many passages of ‘knowledge display’ in the earlier chapters and a lot of page space taken up with announcing the structure of chapters. It would probably have been preferable for the author not to rush into print but to take time to stand back from her doctoral research and rework the material into an argument and structure more suitable for a book format and book readership. This would also have allowed her to correct all the serious errors in the results chapter and the typos in the book as it now stands. To bring out the specificities of the poster genre, it would also have been useful to compare the textual metadiscourse features far more systematically with the abundant results available for metadiscourse in other academic genres, the research article in particular.
9To conclude on a questioning, rather than a merely critical note, I would like to come back to the analytical framework chosen by the author. In Kress and van Leeuwen’s framework (2006), visual interaction is an important feature, expressed by resources such as salience, size of frame, and perspective. Visual interactional resources were initially included in fact as a major factor in the author's earlier programmatic article on poster analysis (D’Angelo 2010) and are also implicitly recognized as an important feature by conference participants in the attractivity of posters and their interest in viewing. In the book, however, while both the textual and visual analyses address the interactive dimension, the interactional dimension is considered only in the textual analysis, and is excluded from the visual analysis on the grounds that it could only be subjective (cf. p. 145). This results in an imbalance between the textual and visual analyses of the posters. Admittedly, it is a difficult dimension to analyse, both across-the-board and within different disciplines: we are very poorly informed about potential interdisciplinary differences about what counts as salience, for example, while changes in size of frame or perspective are more likely to have a field-specific scientific value rather than an interactional one in academic genres. This methodological problem of multimodal analysis in general, and in academic disciplines in particular, would have deserved a fuller discussion than it is given in the book, even if no precise answers to the question are at present forthcoming, for it raises the issue of how suitable some of the interactional parameters in visual analyses initially designed for adverts, newspapers, art, etc. are for academic discourses, and for the poster genre in particular. Academic Posters is a pioneering, though seriously flawed, attempt to tackle some of the issues raised by this multimodal genre and will hopefully stimulate further research into this neglected genre.