- 1 According to some scholars indeed (Atkinson, 1999; Valle, 1999; Gross et al., 2002), it is in Janua (...)
1The practice of book reviewing in academia is as old as the scientific community itself. Indeed, the earliest journals that appeared in the major European countries in the latter part of the 17th century consisted for the most part of book notices because at that time (and up to the 19th century) books played a fundamental role in the dissemination of new knowledge. The Journal des Sçavans, for example –the first periodical to provide information on scientific matters–1 was composed entirely of summaries of scholarly or scientific works. At that time, the book review (BR) function was very conservative and consisted in “giving readers (and scholars) a universal account of the state of learning” (Roper, 1978). Referring to the importance of books by the end of the 18th century, Andrew Duncan, then editor of The Medical and Philosophical Commentaries, the first medical journal entirely devoted to the publication of abstracts of medical books, wrote:
The last section (of Medical Commentaries) will consist of a list of new medical books … for the satisfaction of those who may be deprived of other methods of information … published both in this and othercountries, during the preceding months. We cannot indeed pretend that this list will in any case be a complete one; but it will be our endeavour to render it as much so as our situation will allow; and we are hopeful we shall be able to obtain intelligence of every material book. (Duncan, 1773)
- 2 This rhetorical feature seems to be a distinctive feature of early scientific BR in general (i.e., (...)
2In these early book reviews (BRs) book reviewers were not giving any personal opinion with regard to the content of the books reviewed. As a matter of fact, Duncan urged book reviewers to avoid, as much as possible, either applauding or condemning any author, because, as he put it, the chief aim of that section was “to give such a view of books as may enable every reader to judge for himself” (Duncan, 1773, p. 9).2
3A few years later, discussions on private opinions and exchange of ideas regarding controversial issues became parts of BRs, and the simple recording of published scholarship was abandoned. Indeed, the second decade of publication of the Medical and Philosophical Commentaries favoured a more critical appraisal of the books reviewed. Book reviewers were then encouraged not to confine themselves to a mere analysis of new books but to “candidly offer their opinions of the book contents” which should be expressed “with that respect which is due to merit and that diffidence which the nature of the subject demands” (Duncan, 1786, p. vi). It is only at the end of the 18th century when BRs underwent what Hyland (2000, p. 42) refers to as “a rhetorical shift”when, as a consequence, the common practice of transcribing long verbatim passages from the reviewed books without comment started being replaced by the reviewer’s own stance (Roper ,1978, cited in Hyland, 2000, p. 42).
4BRs have been examined from different perspectives. The pioneering study is that of Motta Roth (1998) who considered 60 BRs from chemistry, economics and linguistics journals. Following Swales’ (1990) genre analysis framework, she examined the rhetorical macrostructure of the genre and could identify 10 steps or sub-functions grouped into 4 moves, the rationale of which is similar to that observed in research article construction. Another quite comprehensive cross-disciplinary research on BR is that of Hyland (2000) who counted speech acts encoding praise and criticism in 160 BRs from 8 different disciplines, including applied linguistics. One of the main findings of Hyland’s research is that BRs published in social sciences journals are much more critical than BRs from the hard sciences (engineering, for example).
5These studies were followed by more sociologically-oriented research, such as that of Burgess (2000), on the one hand, who examined the role of the book reviewer in two distinct types of BRs: those solicited by journal editors and those published freely on the net, and that of Gea Valor (2000), on the other, who explored the issue of courtesy markers and modalised statements in linguistics BRs. A diachronic and cross-linguistic approach was adopted by Salager-Meyer (2001) Salager-Meyer and Alcaraz-Ariza (2004), Salager-Meyer et al. (2005, 2006) who studied the evolution of the expression of criticism in medical BRs.Moreno (2008) and Moreno and Suárez (2008) also examined literary BRs from a cross-linguistic perspective but in a corpus of literary BRs, whereasGiannoni (2002) explored the pragmalinguistic features associated with expressions of overt praise and criticisms in software reviews and Alcaraz-Ariza (2009) analysed compliments in a corpus of medical book reviews from a socio-pragmatic point of view.
6All these studies have shown that BRs are one of the most face-threatening genre of scholarship because, on the one side, their primary audience is the book author him/herself and, on the other, their primary communicative function is precisely to assess the worth, integrity and professional value of books. Moreover, from a Hallidayian standpoint, this quite extensive body of research has put forward the double function of BRs: 1) an ideational function where the book reviewer expresses his/her opinion regarding the content of the book reviewed, and 2) an interpersonal function that refers to the cognitive behaviour of the book reviewer who must strike a balance between collegiality, social harmony and the conveying of critical speech acts in as polite a way as possible. The emphasis in BRs being not on knowledge construction but on the actual impact of such knowledge on scientific and professional communities, these studies also reveal how disciplinary discourse communities encode and negotiate interpersonally sensitive information in a manner that is both socially and epistemically acceptable and congruent.
7The present paper builds on the findings of our previously published research on scholarly BRs by examining how the protagonists’ identification (i.e., the book author and the book reviewer) has evolved over a 120-year period. More specifically, we will examine how book authors were identified in the BR by-lines; how book reviewers (hereafter also called “the judge”) used to sign their BRs; how they used to refer to themselves in their reviews, and finally, which entity (a human being or an object) was held responsible for the flaws mentioned in the BRs.
8In order to analyse these variables, we examined a corpus of 60 BRs drawn from English-medium leading medical journals (see Appendix 1) published in three different time periods: 1) Block A that corresponds to the last decade of the 19th century (1890–1899); 2) Block B that consists in BRs published in the mid-20th century (1950–1960); and 3) Block C or the beginning of the 21st century (2005–2008). The following variables were recorded in each BR:
- 3 To simplify matters and because it is quite frequently impossible to make an educated guess as to t (...)
The book author’s/editor’s identification in the BR by-lines. Was he identified with the initial of his name followed by his surname or with both his name and surname written out in full?3 What kind of data was provided along with his name (e.g., his institutional affiliation, his degrees, his field of expertise, etc.)?
The judge’s (i.e., the book reviewer’s) identification at the end of the BR. Was he identified at all? If so, how was he identified? With his initials only? With the initial(s) of his name followed by his surname written out in full or with both his name and surname in full? What kind of data was provided along with his name/surname?
The judge’s identification in the BR itself. Did he clearly identify himself by using personal pronouns such as “I”, “we”, “us”, and their derivative expressions such as “in my/our opinion”, “it is my/our contention”, “this reviewer” or was he an “invisible” judge”?
The “accused”, i.e., the entity held responsible for the flaws mentioned. Was it a person, i.e., the book author(s)/editor(s), the book itself or a book chapter?
9Results will be analyzed by means of Chi-square tests and interpreted in the light of a socio-constructivist approach to discourse the basic tenet of which is that cultural knowledge and representations of reality are interactionally constructed, socially transmitted, historically sedimented and often institutionally congealed, and finally communicatively reproduced in situ. It thus assumes that writers do not communicate in a vacuum but are embedded in a constraining sociolinguistic setting from which they make the lexical, grammatical and rhetorical choices in order to indicate the purpose of their statements and their point of view. The approach adopted here is thus multi-perspectival and interdisciplinary in the sense that it draws on a number of different research traditions: discourse analysis, pragmatics, rhetorical text analysis, genre analysis and sociology of science.
Graph. 1. – Authors’/editors’ identification (1890–2008).
10Graph. 1 shows a clear distinction between the frequency of the variables recorded in Blocks A and B, on the one hand, and those recorded in Block C, on the other. Indeed, in the end-of-19th century BRs, the book author was mentioned mainly with the initial of his first name followed by his surname (60% of the cases) and less frequently with his name and surname written out in full (40%). Very detailed biographical information followed the author’s surname, such as the degrees held (100% of the time), membership and/or fellowship to scientific societies (75%), and past and present positions (85% and 90%, respectively) to the point that, when the list was too long, it ended up with “etc.”. Here is a typical example of the time drawn from a BR published in 1893:
- 4 The year indicated at the end of each example indicates the year in which the BR was published.
1. The after-Treatment of Cases of Abdominal Section, by C. Martin M.D. Oxon., F.R.C.S. Ed., formerly President of the Royal College of Physicians in Ireland. Obstetric Physician and Gynaecologist, Mater Misericordiae Hospital; ex Examiner, Conjoint Board Royal College of Surgeons and Apothecaries’ Hall; Consulting Physician, Hospital for Children, Fellow of the New York Academy of Medicine, Member of the Association of American Physicians, etc., etc., etc. … (1893)4
This example shows that at that time academic degrees were followed by the abridged name of the university that had awarded them, for example: M.D. Oxon means that the degree was awarded by Oxford University, then called academia oxioniensis: F.R.C.S. Ed. means Fellow of the Royal College of Surgeons at Edinburgh.Quite frequently too, degrees were abbreviated according to their Latin names, e.g., A.M which means artium magister (Masters of Arts).
11As can also be seen on Graph. 1, over time, book authors were increasingly mentioned with their names and surnames written out in full (in 75% of the cases in Block B and 90% in Block C), but mention of academic degrees abruptly disappeared in Block C (80% in Block B and only 10% in Block C). In other words, the book authors’ biographical data that were so abundant at the turn of the 20th century disappeared over time to the point that they are almost non-existent in today’s BRs.
Graph. 2. – The judge’s identification (1890–2008).
12What strikes the eye when comparing Graph. 1 and Graph. 2 is what could be called an “inverse relationship” in the sense that Graph. 1 exhibits a sharp decrease in the frequency of the variables it displays, whereas Graph. 2 shows a sharp increase in the frequency of these same variables.
- 5 Book reviewers’ e-mail addresses were indicated in all the BRs published in 2008, but not in all of (...)
13In effect, a clear evolution can be observed in the way book reviewers were identified in BRs over the 120 year-period studied here. At the end of the 19th century, the majority of them (60%) was anonymous, 35% “signed” their BRs with the initials of their names and surnames, whereas only 5% signed their reviews by indicating the initial of their names followed by their surnames written out in full. Book reviewers became more “visible” by the mid-20th century when they were mainly identified with their names and surnames in full (80% of the cases). Today, not only is the great majority of book reviewers (80%) identified with their names and surnames written out in full, but also with a lot of additional –personal and professional– information, such as the institution they work at (this information is present in all the BRs examined in Block C), the degrees they hold (80%), their medical speciality (70%) and, less frequently, their e-mail addresses (56%)5.
14The judge was “invisible” (i.e., not mentioned by any personal pronoun like “I”, “we”, “one” or expression such as “this reviewer”) in 30% of the BRs making up Block A, 40% of those making up Block B, and 56% of those from Block C (cf. Graph. 3).
Graph. 3. – The “invisible” judge.
15A statistically significant difference was detected between the frequency of this “judge invisibility” phenomenon recorded in Block A and that recorded in Block C (p = .004).
16Let us now examine how the book reviewer identified himself in the BRs where he was “visible”.
Graph. 4. – The judge’s presence in the book review (1890–2008).
17Graph. 4 displays an interesting evolution in the use of the personals pronouns that refer to the judge. Indeed, in end-of-19th century BRs, book reviewers used to refer to themselves with the almost exclusive use of the first person plural pronoun “we” (97% of the time). As examples 2 to 6 below illustrate, “we” (or “us”) referred to the book reviewer only, i.e., it was an exclusive “we” (Kuo, 1999; Hyland, 2005):
2. We are at a total loss to understand why the author introduces a complete dietary. (1894)
3. We do not see the utility of Mann’s division into median tears and transverse tears. (1896)
4. We regret we cannot agree with Polk. (1896)
5. The opinion of American surgeons is not expressed by Dr. Alligham, and we regret to have it appear as being so in an American book. (1898)
6. We hope that the author is misinformed when he states that “sponge tents” are the best means of dilating a contracted cervix in cases of abortion. (1898)
18In a few examples only (ex. 7 and 8 below), it was not easy to determine whether the personal pronouns were inclusive (i.e., included the readers as well the judge) or exclusive (referring to the judge only), although we are inclined to believe that they belong to the former category:
7. Our knowledge on these subjects has greatly increased during recent years. (1899)
8. This book, we are told by the editors, is designed as a working text book. (1899)
19Finally, as Graph. 3 clearly shows, the frequency of “We” significantly decreased over time: from a high 97% in Block A, it fell to 40% in Block B and 10.8% in Block C (p = .0001).
20The frequency of the first-person singular pronoun “I” and its variants –“for me”, “I am not sure”, “in my opinion”, “my only concern”, “my initial reaction” (see examples 9 and 10 below) significantly increased over time: it jumped from non-existent in Block A to 50% in Block B and 87% in Block C (p = .0016). As a matter of fact, as can be seen on Graph. 3, the more frequent “we”, the least frequent “I”, and vice versa.
9. The discussions are balanced and critical, but in my opinion, they are not critical enough. (2007)
10. I was also disappointed that the clinical aspects received relatively little space. (2008)
21Although never commonly used, the indefinite pronoun “one” was more frequent in the mid-20th century (10%) than in the other two periods (3% in Block A and 2.2% in Block C). Here are three examples drawn from the different Blocks:
11. We hardly need to state that the printing and the paper are of a degree of excellence one would expect of a work from this peerless institution. (1894)
12. From the pen of Dr. Wabaro, one is of course entitled to expect the highest quality, and one is not disappointed. (1955)
13. One hopes that the third edition will build on our progress in thinking about cancer as a disease of genes. (2005)
22A delightful example (ex. 14 below) was found in a 19th century BR where the book reviewer feels it is a sacrilege to criticize the book:
14. It seems a sacrilege to criticize such a work; one hesitates long before plunging his dissecting knife into a body once loved, once admired. But the wheel of time moves on relentlessly. (1896)
23As previously stated in relation with “we” above, it is not always straightforward to determine whether the indefinite pronoun is exclusive or inclusive, although in example 11 above, “one” is clearly exclusive.
Graph. 5. – On the bench of the accused (1890–2008).
24Graph. 5 shows some interesting features regarding the evolution of the entity representing “the accused”.
25In Block A that responsibility almost exclusively fell on animated/personal entities, i.e., authors/editors. Indeed, at that time personal accusation represented almost 70% of all the accusation cases: the accused was either unnamed and simply referred to as “The author/editor” (examples 2 and 6 above) or clearly identified with his surname (examples 3, 4 and 5 above). Conversely, the impersonal accusation type (books and books chapters) was much less frequent (15.2% and 16.9%, respectively). The difference between both accusation types was found to be statistically significant (p = .0001).
26This state of affairs changes drastically in the mid-20th century (cf. Graph. 5) when book authors/editors progressively leave the bench of the accused to be replaced by books (55.7% of the cases). The difference between both types of accusation was here too found to be statistically significant (p = .0006).
15. For me, the book is rather disappointing. (1956)
16. The book is very absorbing, but it is not easy reading and necessarily of limited appeal. (1957)
27Interestingly, in Block C, the impersonal/personal accusation ratio is exactly the reverse as that observed in Block A. In effect, in the 21st century BR sample, the frequency of occurrence of the impersonal accusation type (67.7%) is significantly greater than that of the personal type (32.3%, p = .006).
28Our quantitative data revealed that, over the 120 years analyzed, book authors are more and more laconically identified in the BR by-lines, whereas the exact opposite can be said about book reviewers who are identified with an increasing amount of personal and professional information at the end of the BR.
29In order to determine whether the way book authors/editors and book reviewers are today identified in medical BRs abide by existing editorial guidelines or house-style, we examined the “Instructions to contributors” or “Guidelines for authors” section of the 21st century medical journals we consulted. No specific information could be found, except sometimes a brief mention of word limit (e.g., between 1 000 and 2 000 words). Unsatisfied with the results of our enquiry, and in order to get first-hand information, we conducted a small-scale ethnographic study by contacting a few journal editors by e-mail.
30The majority of the replies we received confirmed that there are indeed no specific guidelines on how to present book authors/editors and book reviewers in the BR section of scientific journals.An editoreven wrote that only a few days ago he had been looking at the way reviewers identified themselves and their affiliations in the BR section of his journal. “I don’t believe we have a policy,” he wrote. “Perhaps we should adopt a consistent house style. I had started to wonder about that…”Some editors, though, stated that book reviewers are sent an invitation letter asking them to provide such bio-data as institutional affiliation, city, state and (optional) e-mail address.
31We would therefore like to argue that the decreasing amount of book author’s bio-data and the parallel increasing amount of reviewer’s personal and professional data provided in medical BRs are related to the following intrinsic and extrinsic factors that are all, one way or another, related to the growing professionalization and institutionalization of scientific medicine:
1. The increasingly laconic way of identifying book authors/editors in the by-lines of the BR is very likely related to the fact that extensive bio-data is today provided in the book itself. Moreover, because today’s book reviewers have to respect the word limit imposed by journal editors/publishers (space is money!), all unnecessary, redundant or superfluous information is to be avoided.
2. The size of the scientific community, we believe, has a direct bearing on the evolution of the way book reviewers were identified in the past and the way they are identified today. Indeed, 19th century scientific community was quite small to the point that scientists knew who was who (e.g., Bazerman, 1988; Atkinson, 1999; Valle, 1999; Gross et al., 2002). We can thus safely assume that the indication of the book reviewer’s initials was then sufficient to know who had written the BR. Our speculation does not seem to be so far-fetched because quite a few 19th century BRs were “signed” by the same initials, i.e., written by the same reviewer. We even wonder whether it was not the journal editor who wrote the BRs. Unfortunately, and in spite of all our efforts to find an answer to that question, we were unsuccessful.
32Nowadays, by contrast, scientists are mostly “invisible colleagues”, especially in a discipline as vast, inter-institutional and international as the medical one. It is therefore important to know who is who. In this respect, a journal editor we consulted wrote: “Personally, I like to know who the reviewer is as that will affect my reaction to what is said. I want to read reviews by experts not tyros, and I guess that journals provide information on their reviewers so that the reader has some sense of the writer’s credentials and credibility.”
33It is certainly true that the writing of a BR does not require the original research, creativity and intellectual innovation that are involved in the writing up of a scientific journal paper, but we cannot deny the fact that a well-written, judicious review requires mature judgement and an up-to-date knowledge on the subject dealt with in the book, i.e., a high level of expertise in a specialized field. It is therefore important, as the above mentioned journal editor asserted, to know whether the reviewer’s credentials are up to the task required.
3. We also strongly believe that the pressure of the academic world and its related curriculum vitae “bodybuilding exercises” have influenced the way book reviewers are identified. Although, as Felber (2002) rightly mentions, no scientist ever made a professional reputation on the basis of the BRs he wrote, and although BRs have a marginal status in the world of today’s scholarship, the writing of that “intermediate genre” (Felber, 2002, p. 166) can help junior researchers (who all compete for attention) build up their curriculum vitae in the pursuit of some “coin of recognition” in the academic marketplace. In other words, BR writing can increase junior researchers’ stock of symbolic capital (Bourdieu, 1991). This, certainly, was not a concern for 19th and mid-20th century researchers. The anonymity of book reviewers back in the 19th century tends to corroborate this assertion.
34Our quantitative data showed a progressive increase in the “invisible judge” phenomenon. A close look at those BRs where the judge is “invisible” allows us to classify them into two categories:
those from Block A and B that remind us of 18th century BRs (see Introduction) in the sense that they are purely descriptive and do not contain any critical –positive or negative– appraisal at all. In these usually very short BRs (they rarely exceed 300 words), the protagonist is the book or a book chapter, certainly not the book reviewer;
those from Block C that are called “Book Notes” or “structured BRs” (see Appendix 2). These highly structured BRs –that remind us of the so-called “structured abstracts”– are a requirement today in some medical journals such as Annals of Internal Medicine. In these Book Notes, there is no place at all for the emotional and/or personal element. Here too, the protagonist is the book or a book chapter, not the book reviewer.
35We have seen that in 19th century BRs, the bench of the accused was mainly occupied by personal/animate entities (named and unnamed authors/editors), and that over time impersonal/inanimate objects (i.e., the book or a book chapter) replaced the persons. This personal/impersonal or animate/inanimate evolution adds further support to the now well-documented increasing depersonalization of scientific discourse (e.g., Webber, 1998; Hyland, 2005; Cronin, 2005).
36Moreover, we could speculate that today’s judges, by preferring to address their criticisms to objects –and, to a lesser degree, to subjects– wish to demonstrate their cooperative, respectful and harmonious relationships with their peers. It is as though today’s book reviewers were trying to minimize the interpersonal damage of their critical comments, negotiate the social relations inherent to the BR genre and establish a solidarity framework with the book’s author. Such concerns were totally foreign to the 19th and mid-20th century judges.
37Our quantitative data also showed that 19th century judges used to identified themselves with the exclusive “we” personal pronoun, whereas today’s judges are either (apparently) totally absent from their BRs (see “The judge’s growing ‘invisibility’” above or refer to themselves with the “I” personal pronoun).
38One could argue that the “we”/“us” involves several reviewers. Not so, because all the BRs, except 3 in Block B, were written by one reviewer only. This is why we rather believe that the exclusive “we” pronoun was extensively used by 19th century judges because it expressed the voice of assertive experts talking on behalf of the scientific community. Its high frequency at that time could also be a calque from French rhetoric that loath the arrogant-sounding “je” (“I”). We should indeed not forget that French was an important language of scientific knowledge dissemination by the end of the 19th century.
39From a purely quantitative standpoint, our data thus clearly showed that the frequency of “we” significantly decreased over the 120 years analyzed, but an interesting qualitative finding should also be mentioned here. Indeed, the semantic reference of “we” has also changed over time. The semantic reference of 19th century “we” in examples 2 to 6 above contrasts sharply with that of example 17 below drawn from a 20th century BR in the sense that the former are expert-sounding exclusive “we” ‘s (that involved the judge only) and the latter is an inclusive “we” that invites, engages and binds the reader into the judge’s persuasive arguments (Hyland, 2005, 2008):
17. Should we rely on transvaginal ultrasonography or shall we proceed to magnetic resonance imaging? (2005)
40Finally, with his almost exclusive use of the first singular personal pronoun “I”, today’s judge not only wishes to reflect coyness but also to project himself as an ordinary reader. Nonetheless, the extensive bio-data with which today’s BRs are signed remind the reader of the BR that the book reviewer is not an ordinary reader at all, but an expert.
41This research on the quantitative and qualitative evolution of the variables that identify the book author and the book reviewer in English-language BRs published between 1890 and 2008 showed that the former are accompanied by a decreasing amount of bio-data, whereas the latter signed their BRs with an ever-increasing amount of personal and professional information. It has also shown that judges are less “visible” in their BRs today than they were in the past. Regarding their identification or visibility in the BR itself, 19th century judges used to identify themselves with the exclusive “we” (hence projecting the image of an expert assertive voice), mid-20th century judges with the inclusive “we” (probably in order to show solidarity with the reader), whereas today’s judges rather identify themselves with the “I” singular personal pronoun, thus playing the role of an ordinary reader. Finally, an increasing depersonalization or objectivization has been observed in the entities accused of the flaws mentioned in the BRs.
42The changes observed reflect changes not only in the scientific enterprise but also in the scientific society in general and should not be considered as a sign of progress or improvement but as a process of selection and adaptation to the increasing number of scientists –hence, the increasing volume of scientific papers and books–, to the needs and the increasing complexity of the context in which scientific activity develops and to the changes that, as a consequence, the scientific enterprise has been subjected. The evolution evidenced by our results finally mirrors the fact that the scientific community in general has become increasingly competitive, specialized and professional, thus underscoring the increasing standardization and institutionalization of science in general.