How Nice Is Too Nice? Australian Book Reviews and the ‘Compliment Sandwich’

By Emmett Stinson

This article responds to an ongoing public debate about whether Australian book reviewing is ‘too nice’, which started in the literary journal Kill Your Darlings in 2010 and has continued in other literary publications. It takes up Ben Etherington’s claim that ‘too nice’ reviewing is characterised by the ‘compliment sandwich’ in which critique is surrounded by mollifying praise. It offers a ‘distant reading’ of two years of fiction reviews in the Australian Book Review, applying a manual appraisal analysis to demonstrate that book reviews in Australia’s flagship reviewing publication do often adhere to the compliment-sandwich form. The article then returns to the question of ‘too nice’ reviewing, and applies a modified Bourdieusian analysis to examine how reviewing debates have served as proxies for larger disputes between institutions and interlocutors in the literary field.

1.Critiquing the Too-Nice Review

The Melbourne journal, Kill Your Darlings (KYD) opened its inaugural issue in March of 2010 with Gideon Haigh’s polemic, ‘Feeding the Hand that Bites’, which bemoaned the ‘demise of Australian literary reviewing’ (9). Haigh accused reviewers of timidity, arguing that, since they are neither well-paid nor highly esteemed, ‘there’s little incentive for sticking one’s neck out, for actually taking a position, for arguing that a book is bad, or sloppy, or stupid’ (10). For Haigh, negative criticism is counterproductive, since ‘the author might be reviewing us one day… [i]n which case, it may, of course, be payback time’ (10). He argues reviewers avoid critical judgment, preferring to ‘summarise the contents, recapitulate the blurb, describe the author’s reputation, or examine the author’s politics’ (10).

Book reviewing seems an unlikely flashpoint for controversy, but Haigh’s essay served as a proxy for larger literary debates. When he argues that book reviews ‘have become hodgepodges of conventional wisdom and middlebrow advertorial’ (9), he rehashes old antipathies between highbrow and middlebrow. He raises concerns about how economics impinge on notions of literary value (however such a term might be construed), when he bemoans the ‘the lacklustre infomerciality of so much Australian reviewing’ that gushes ‘over the latest vogue’ (11). He laments Australian literary culture’s insularity by claiming reviewing practices support ‘vested interests in Australia’s small, snobbish, fashion-conscious, self-celebrating literary scene’ (11). Craven book reviews become symptoms of an unhealthy literary culture. Haigh’s refraining from naming specific reviewers further created anxiety among critics who wondered if they were the essay’s secret target (although this failure to name names—common in critical pieces on book reviewing—may itself be a symptom of a ‘too nice’ literary culture).

KYD published Haigh’s essay to generate controversy and establish the journal as a locus of edgy and important literary conversations. Haigh was an inspired choice for generating media buzz: he had access to wider media networks from his popular sports writing but still possessed a highbrow appeal among literary insiders; in this sense, he crossed the domains of popular and high culture that Pierre Bourdieu describes as the key opposition within the literary field (Bourdieu 53). When Haigh read the essay at KYD’s launch on March 10, 2010, it was already familiar to much of the audience, who had heard it discussed across a range of media. The Sydney Morning Herald ran a short piece about Haigh’s essay on February 20, and Haigh appeared on ABC Radio on February 22. KYD made an excerpt available online on February 24th and published a response to Haigh by Martin Shaw (then the books division manager of Readings Books Music & Film) on their website the following day. Stephen Romei published another response in The Australian on March 2. This was the intention, since this controversy promoted the journal; KYD followed up by holding a panel discussion on reviewing with Haigh, Jo Case, and Anthony Morris at Readings on March 17.

Haigh’s essay, though tied to the marketing strategies of a new literary journal, also instigated an ongoing debate about Australian literary reviewing. In August of 2011, Louise Pine revisited Haigh’s concerns in the Overland essay ‘To Review or Not to Review’, and Melbourne writer Mel Campbell responded in a Crikey essay the next day, arguing that ‘hatchet job’ reviews constitute ‘a lazy form of reviewing’ (‘Hugging with Knives’ 2011). Concerns about too-nice reviews then received their most detailed articulation in Ben Etherington’s 2013 essay ‘The Brain Feign’, which critically describes Haigh’s essay as ‘glint[ing] with aphorisms’ but too ‘brief when it comes to articulating what is at stake’. Rather than lamenting generalities, Etherington examines the reception of Anna Funder’s 2011 novel, All That I Am, in a case study of the book’s reviews, which locates a ‘structural’ problem within an insular Australian contemporary literary field[1], caught between ‘print’s decline and the self-promotion attending so much activity online’ (2013).

The context of Etherington’s essay matters, because it was used, along with five others, to launch The Sydney Review of Books (SRB). Like Haigh’s article, Etherington’s essay had been surreptitiously passed around literary insiders before publication (it was sent to me by the editor of a literary journal who had received it from a festival director), thereby introducing the journal to those in the field with significant stories of social and symbolic capital. The confrontational nature of Etherington’s article—which strenuously criticised a much-lauded Australian novel—generated controversy, producing wider interest from readers beyond the field of cultural production. The irony of Etherington’s piece (though one he is certainly aware of) is that it criticises the deleterious effects of social media on literary culture while leveraging those same forces to promote SRB. The key distinction here is that ‘The Brain Feign’ also served as the first installment of ‘Critic Watch’ an ongoing column meant to expose the self-serving reviewing practices Haigh identified.

Despite the appearance of Critic Watch, questions about the niceness of book reviewing have persisted, as evidenced by Kerryn Goldsworthy’s 2013 ‘Everyone’s a Critic’ in the Australian Book Review and the 2015 Monash University conference, Critical Matters, which presented perspectives on book reviewing from academics and practitioners. But rather than tracking this debate, my interest is in substantiating whether or not book reviews are ‘too nice’. Haigh does not produce any evidence to support his claims. Etherington’s ‘The Brain Feign’ employs a case study of Anna Funder’s All That I Am, a process he repeated in a 2013 examination of the reception of Hannah Kent’s Burial Rites (‘The Real Deal’ 2013). But case studies—while they may be able to highlight or substantiate particular lapses of critical reception—work through a large-scale synecdoche, in which the specific instances stand in for a broader set of failings. But individual instances of critical failure are just that, and do not necessarily reflect wider practices.

Etherington implicitly acknowledges the limitations of exemplary case studies in ‘The Poet Tasters’, which examines the 247 reviews of Australian poetry that appeared in 2013. In this survey, Etherington notes reviewers’ frequent use of the ‘compliment sandwich’, in which critical comments are bookended by vague affirmation according to a set formula:

1.Introduce the volume, the poet and their previous publications.
2.Describe the poet’s overall aesthetic with reference to European and/or North American antecedents.
3.Quote approvingly from two or three choice poems with some technical commentary.
4.Express reservations about one or two poems.
5.Affirm, nevertheless, the worthiness of the volume as a whole.

For Etherington, the compliment sandwich is not just lazy, but ‘inverts good critical practice’. In the compliment sandwich, criticisms are not ‘patiently explained’, so readers must take reviewers’ claims on trust rather than on the strength of their analysis. This muted critique also ‘weakens the praise’ in a review, because such praise is the default rather than having ‘been won from a determinedly critical disposition’ (‘The Poet Tasters’).

Etherington states that the ‘obvious and probably accurate conclusion’ is that compliment-sandwich reviews are mostly written by writers afraid of making enemies in a small literary scene. But Etherington’s essay expands the critique of too-nice reviewing in two important ways. Instead of using illustrative examples, it analyses aggregated cultural practices of reviewing. Secondly, in identifying the compliment sandwich, Etherington suggests that too-nice reviewing is not simply dispositional, but also formal. The compliment sandwich thus constitutes a formal criterion for assessing reviewing—and this is what I have set out to do in a provisional mode by applying a modified form of appraisal theory to a small sample of literary reviews from The Australian Book Review to see whether or not they are compliment sandwiches; this pilot study—which applies a novel methodology to a limited sample—gestures toward ways in which aggregate literary practices might be analysed. I will then re-examine the results of my analysis through a Bourdieusian frame, arguing that disputes over ‘niceness’ reflect key distinctions between agents in the Australian literary field, and thus are of broader significance for understanding contemporary Australian literary culture.

2.Measuring the Compliment Sandwich

I surveyed two years of issues of Australian Book Review (ABR), from September 2013 through August 2015, examining reviews of fiction to see if they matched the formal characteristics of the compliment sandwich. This produced a relative small sample of 78 reviews, which I then analysed for polarity (a linguistic term referring to the orientation of an opinion as positive, negative or neutral) to see whether or not they conformed to the ‘compliment sandwich’. This admittedly small and non-random sample would not meet the evidentiary criteria of formal statistics, and I am not claiming that it meets this burden. Given both the novelty of the method I have applied and the time-intensive nature of manual appraisal analysis (which, as I will demonstrate, requires careful, close reading), I worked with a smaller sample to ensure precision, rather than a broader sample that might include significant errors. As a result of this limited sample, however, my results are neither indicative of other reviewing outlets in Australia, nor, necessarily, of ABR reviews outside of this designated time period. Nonetheless, I do think these results are a way of partially substantiating Etherington’s claims, and also represent another (and, within the field of literary studies, original) mode of distant reading for analysing aggregates of cultural praxis.

I decided to examine ABR (founded in 1961), because it is the longest-running publication devoted to Australian book reviews, although it has had competitors like SRB, as well as the now-defunct Australian Literary Review (2006-11) and Australian Review of Books (1996-2001), both of which ran as inserts in News Corp’s The Australian newspaper. While ABR is the standard-bearer of Australian book reviewing, it often publishes emerging critics, who might produce more formulaic reviewing than in the broadsheets, which employ established reviewers.

A few caveats need to be addressed about the reviews I sampled. I did not consider all reviews in the ABR from this period, but restricted my analysis to reviews of fiction. I did this because I wanted to be sure my analysis was not affected by the logics of different cultural subfields (such as when Etherington argues that poetry reviews are affected by the closely-knit nature of the poetry community). Moreover, debates about ‘niceness’ in book reviews have mostly been about fiction reviewing—and literary fiction reviewing in particular. Popular fiction reviews don’t seem to have entered the debate—though I did not make distinctions between literary and popular works in this survey.[2]

I excluded non-fiction, because it can be considered valuable (by containing unusual information, for example) even if it’s deemed faulty or wanting in aspects of style, language or structure. Fiction, on the other hand, is usually assessed in relation to its style, language, characterisation, and narrative, rather than content. Admittedly, this binary breaks down for some works; I excluded reviews of Robert Dessaix’s What Days Are For (2014) and Martin Edmond’s Battarbee and Namatjira (2014) because both are technically non-fiction, even though they share many stylistic qualities with fiction. Moreover, many fictional works do refer to important political or social events and issues, and thus encompass more than purely stylistic or formal concerns.

I also excluded reviews of overseas works and republished ‘classics’, since attacks on the ‘niceness’ of Australian reviewing typically claim the smallness of the local literary scene is what discourages robust criticism. I also did not analyse all ‘capsule’ reviews of fiction, which tend to be 300-500 words long; my method of analysis requires each review to have at least four paragraphs, so capsule reviews that were three paragraphs or less had to be excluded. The final sample comprised 56 feature reviews and 22 capsule reviews.

To determine whether or not these 78 ABR reviews were compliment sandwiches, I had to assess where negative and positive comments appeared within the reviews. Determining this orientation—known as ‘polarity’ within the field of sentiment analysis—presents intractable problems because it is inherently subjective. Analyses of polarity vary significantly between readers, and this variation has led to the rise of computational sentiment analysis or opinion mining, which makes use of software to determine polarity. Digital sentiment analysis, which examines responses from relatively short and simple texts (such as social media posts or customer-generated product reviews), can be useful for analysing certain kinds of cultural responses, as Beth Driscoll has recently demonstrated in her analysis of tweets about and survey responses to the Melbourne Writers Festival (Driscoll).

But Driscoll’s method, which employs the program SentiStrength, cannot easily be applied to book reviews. SentiStrength analyses polarity based on the most positive or negative words in a passage, and longer texts therefore distort its results. SentiStrength functions by evaluating texts against an internal bank of words with pre-determined polarity scores; complex rhetorical works, such as book reviews, express polarity in ways that do not correspond to SentiStrength’s word bank. Alison Broinowski’s review of Dominique Wilson’s The Yellow Papers, includes the statement ‘I cavil at half a dozen typos’ (45). This is unambiguously negative, but SentiStrength does not recognise ‘cavil’ as such. In the same review, Broinowski notes the novel depicts ‘racial prejudice’ (45), but SentiStrength assesses this as negative. Computational sentiment analysis—at least the kind enabled by off-the-shelf software tools—does not yet present an adequate means for determining polarity of book reviews.

I have instead applied a ‘manual’ analysis of polarity, following a method partially derived from appraisal theory in corpus linguistics as articulated by Martin and White’s The Language of Evaluation: Appraisal in English (2005). In Martin and White’s taxonomy, book reviewers’ evaluations constitute ‘appreciation’, which refers to ‘evaluations of “things”’ (56); ‘judgment’ applies to persons and actions that are oriented towards the social (52). For Martin and White, appreciation can be subdivided into three categories: ‘“reactions” to things (do they catch our attention; do they please us?), their “composition” (balance and complexity), and their “value” (how innovative, authentic, timely, etc.)’ (56). There are valid objections to these categories: compositional notions of balance and complexity seem culturally specific value judgments in their own right. Nonetheless, Martin and White’s typology reflects the insights common to book reviews and mirrors Haigh’s claims that book reviews should be should be ‘engaging’ (10), examine what ‘makes good books good’ (11), and present context that ‘deepens understanding and clarifies debate’ (11).

Martin and White also usefully discuss the inherent subjectivity of manual appraisal analyses: such analyses are ‘inevitably interested’ and ‘can never be the final word’ (206), but, rather than being an unfortunate artifact of examining polarity, this is its natural result. Appraisal is subjective by nature, and can only be grasped subjectively. Nonetheless, attempting to gain more objective purchase on appraisal through analytical methods brings to light trends that otherwise might be obscured. This point recalls John Frow’s argument about sociological readings of texts, which are never objective, but provide an essential vantage point that undermines ‘the apparent coherence of the literary’ by revealing its embedment within determining and contingent social and cultural structures (Frow 242).

Martin and White’s appraisal theory establishes some guidelines for analysing polarity, but manual methods for marking appraisal are still in flux. This is due to the novelty of appraisal theory, and the fact that much work in the field has focused on digital sentiment analysis. My method applies aspects of appraisal theory, but greatly simplifies manual analysis to focus on evaluative statements at the levels of the sentence and the paragraph. This approach would be too simplistic for corpus linguistics scholars, but I would argue it is sufficient to establish polarity in ABR reviews.

My method is as follows. Rather than analysing entire reviews, I examined the polarity of opening, median, penultimate, and final paragraphs of reviews, which reflect the key structural points of the complement sandwich. Where there were two median paragraphs (because a review had an even number of paragraphs), I examined the first of them. This method requires reviews to be four paragraphs long, so shorter reviews were excluded.

This method constitutes a mode of ‘distant reading’, a term Franco Moretti has defined as ‘a condition of knowledge’ that ‘allows you to focus on units that are much smaller or much larger than the text: devices, themes, tropes—or genres and systems’ (‘Conjectures’, 57). By examining only sections of the text, this approach participates in modes of literary analysis that suspend the usual relationship between part and whole, which forms the basis of traditional hermeneutics. Perhaps the most famous such example from Moretti’s Distant Reading (2013) involves his analysis of changes in literary culture using a database containing only the titles of books. As Caroline Levine has pointed out, the privileging of wholeness in hermeneutics itself relies on a set of assumptions that cannot be separated from contingent political, social, and cultural circumstances (Levine 24-5).

Nonetheless, I have also read each review in its totality, and my belief is that the given selection of paragraphs generally represents the distribution of evaluation in ABR reviews. If this had not been the case, reviews’ polarity should not correlate to the compliment sandwich form. There is a formal objection to this method: it may be that reviews of this length (between 300-1500 words) naturally follow a pattern that moves from summary to close analysis and ends with evaluation, which lends itself to the compliment sandwich. Further research would be needed to determine whether this is a generic commonplace of most reviewing.

My polarity analysis employed two different levels of what corpus linguists call ‘unitisation’ (i.e., the granular level at which polarity is assessed); I assessed polarity at the level of each sentence, and then averaged the results to determine the overall polarity of the paragraph. There were four possible assessments of polarity: units assessed as ‘negative’ received a score of 0.0; units assessed as ‘positive’ received a score of 1.0; units assessed as ‘mixed’ received a score of 0.5; units with no polarity (i.e., sentences that contained only summary, non-evaluative analysis, and neutral statements of fact) were excluded from averages.

So, if a paragraph contained six sentences, three of which were neutral (excluded), one of which was negative (a score of 0.0), one of which was mixed (a score of 0.5) and one of which was positive (a score of 1.0), the polarity average would be calculated by dividing the total score (1.5) by the total number of evaluative sentences (3). Following this method, this paragraph would receive a score of 0.5, which would suggest that it was mixed. This method might overstate the polarity of some paragraphs; if a paragraph contained five sentences, four of which were non-evaluative and one of which was negative (0.0), then the entire paragraph score would be 0.0. I would argue, however, that evaluative sentences following neutral analysis or description often determine the tone of paragraphs.

I will briefly outline the criteria I used to assess polarity. Assessments of polarity are affected by subjective perception and background knowledge: a reader familiar with book reviews’ evaluative lexicon will probably be more sensitive to such judgments. Despite the subjectivity of such assessments, I was surprised that most evaluations of polarity seem straightforward. Positive evaluations often employ explicit modifiers. For example, Catriona Menzies-Pike’s 2015 review of Lisa Gorton’s The Life of Houses states that the novel ‘is a nuanced and intelligent reflection on the spaces mothers and daughters share’ (19). Amy Ballieu’s 2015 review of James Bradley’s Clade notes that he ‘elegantly evokes the subtleties of his characters’ evolving relationships’ (36). Felicity Plunket’s 2015 review of Amanda Lohrey’s A Short History of Richard Kline notes that Lohrey’s ‘perceptive analysis irradiates each of the novel’s questions’ (37). Chris Flynn’s 2015 review of Steve Toltz’s Quicksand praises not only the novel under considerations but also three other novels and the publishing house that produced them:

Penguin Australia’s recent fiction output has been remarkable. Ceridwen Dovey’s Only the Animals, Omar Musa’s Here Come the Dogs, and James Bradley’s Clade have all been idiosyncratic and inventive reads, bristling with energy and ideas. Steve Toltz’s Quicksand proves to be the cherry on the cake—a beguiling novel that confounds and astonishes in equal measure, often on the same page. (30)

The rampant praise here is so expansive that a cynical reader might be forgiven for wondering if Flynn—himself a novelist—is hoping to sign a contract with Penguin Random House in the future. In each case, though, modifying adjectives—‘nuanced’, ‘intelligent’, ‘perceptive’, ‘idiosyncratic’, ‘interesting’, ‘beguiling’, ‘bristling’—indicate a positive polarity.

I deemed as ‘mixed’ those evaluations in which criticisms were both advanced and ameliorated. For example, in his 2014 review of Rohan Wilson’s To Name Those Lost, David Whish-Wilson both questions and praises the dark tone of the novel:

Wilson’s vision of Launceston town is hellish, and some readers will question the relentlessness of his vision, his refusal to heighten the dark with contrasting moments of light (one episode in which children torture a cat felt like overkill), but this is not to detract from the novel’s vitality or its perfectly rendered dialogue. (55)

Here, the novel’s ‘relentlessness’ is queried with the suggestion that the novel may be too dark, but this criticism is paired with mitigating praise about the work’s ‘vitality’ and ‘perfectly rendered dialogue’; Whish-Wilson also suggests this criticism is a matter of personal disposition rather than a technical failing (although attributing this perspective to ‘some readers’ could also pass off a subjective critique as a more objective one). In a 2015 review of Anson Cameron’s The Last Pulse, Catriona Menzies-Pike similarly notes, regarding the novel’s objectivising portrayal of women, that ‘In a gleeful and inclusive romp, this strikes a dud note’ (32). While the comment is critical, it is alleviated by praise and cannot be considered wholly negative.

The distinction between ‘mixed’ and ‘negative’ assessments seems more subject to variation across readers. Some negative assessments are clear, as in the case of Rachel Robertson’s 2014 review of Riding A Crocodile: A Physician’s Tale by Paul Komesaroff when she notes that ‘Like the characters, the dialogue can be stilted and unconvincing, all too obviously serving the novel’s themes’ (32). Such wholly negative assertions are relatively uncommon in the sample of ABR reviews I analysed; the rarity of unqualified criticism goes some way to substantiating the claim that ABR reviews are nice, or at least aim to be civil.

Many of the criticisms I judged as negative were still hedged, as in Sarah Holland-Batt’s review of When the Night Comes by Favel Parrett, which notes that ‘While the novel integrates its two halves evenly, they do not always feel equally balanced or pressing’ (12). Here the criticism—that novel’s two halves lack equal weight—is mitigated by the claim they are ‘integrated’—a fairly opaque distinction. The hedging of negative judgments may simply mean book reviewers finely calibrate their judgments to be sensitive to a novel’s form and the author’s apparent intentions. Nonetheless, Holland-Batt’s critique here differs in intensity from ‘mixed’ evaluations because it does not counterpoise its criticism with strongly positive language.

Another source of potential variation requires consideration: several prominent critics studiously avoid the modifiers that typically signify evaluation. Such sentences, if not read carefully, can be incorrectly deemed ‘non-evaluative’. James Ley—recipient of the Pascall Prize and founding editor of the Sydney Review of Books—claims that ‘Whenever I write a sentence that sounds like the kind of thing that gets plastered across a book cover, I cross it out’ (Ley 29). Kerryn Goldsworthy, winner of Pascall Prize and one of Australia’s most eminent critics, has also noted an aversion to overtly evaluative language:

I try to avoid direct expressions of evaluation—except in extreme cases, I don’t think the worth of a book can be confidently quantified—and, as a result, can sometimes find that I haven’t made my judgement as clearly as readers might have liked; I prefer to make more indirect comment on the book’s value by using descriptive terms with positive or negative connotations. (‘Everyone’s a Critic’ 22)

Goldsworthy’s 2014 review of Joan London’s The Golden Age demonstrates this precept when she states that ‘The curse of the Old World is invoked in flashbacks; although the word “Jewish” appears in this book only once… a handful of scenes from wartime Europe tell us all we need to know in this respect’ (11). The praise is implicit: London’s novel is understated and alludes indirectly to larger issues, such as anti-Semitism, which demonstrates her technical mastery. Goldsworthy’s observations positively reflect London’s craft and restraint without evaluative adjectives.

3.The Dominance of the Compliment Sandwich

My expectation was that ABR reviews would not overwhelmingly conform to the compliment sandwich, but a significant proportion were ‘compliment sandwiches’ in my analysis. This finding becomes more significant when considering some other trends revealed in the analysis. For one, overwhelmingly positive reviews make up a large portion of the sample: 31 of 78 reviews (39.7%) did not contain any significant negative criticism in sampled paragraphs. The high prevalence of positive reviews to some degree substantiates the idea that ABR reviewing is often ‘nice’. Moreover, only two of the wholly positive reviews were compliment sandwiches—which is logical since a wholly positive review would normally not have any criticism requiring mitigation.

This high proportion of positive reviews was not balanced out by an equal number of harshly critical reviews. There are only seven reviews (08.9%) that could be viewed as significantly negative (having an overall evaluation score of 0.25 or lower). But there was a significant differential in the percentage of negative reviews in relation to form: while 4 of the 22 (18.2%) of capsule reviews were negative, only 3 of the 56 (05.3%) feature reviews were negative. I will consider the significance of this difference later in the essay.

Of the three negative feature reviews, Alison Broinowski’s review of The Yellow Papers by Dominique Wilson seems the most critical, raising significant concerns with little compensatory praise. Delia Falconer’s review of Mark Henshaw’s The Snow Kimono offered some significant critiques of the novel, which I will discuss at the end of this essay. Susan Lever’s 2015 review of Merciless Gods by Christos Tsiolkas, though it praises the author’s adherence to naturalism as ‘admirable’, ultimately concludes that ‘it makes for a severely confined literary art’ (23). Such reviews are outliers, however, since 71 of the 78 reviews are mixed or positive; my analysis thus applies within a context of reviewing practices that are often positive but rarely negative.

Another key finding—which differs from the model of the compliment sandwich that Etherington describes—is that evaluations almost always occur at the end of reviews, but are less frequent in early paragraphs. Only 34 of the opening paragraphs (43.5%) from my sample contained any evaluations, which is logical, given that opening paragraphs often provide summary information. Only 7 of these 34 reviews (08.9%) had mixed or negative evaluations in the first paragraph, while 27 of the 34 (79.4%) first paragraphs with evaluations were positive. Of the median paragraphs, 46 (58.9%) contained evaluative language; again, many of the reviews seem to engage in thematic or formal analysis at this stage, but refrain from evaluation. By contrast, 55 of the penultimate paragraphs (70.5%) contained explicit evaluations (and 7 of the 13 (53.8%) reviews whose penultimate paragraphs were non-evaluative were wholly positive). Seventy-three of the 78 reviews (93.5%) contained evaluations in the final paragraph. Only four (05.1%) of these final paragraphs had a negative polarity overall. Again, this suggests that ABR reviews tend to privilege a civil criticism that refrains from ending reviews on a negative note.

The inconsistent appearance of evaluation in early paragraphs means that—while I could not locate the form of the compliment sandwich Etherington describes—I could identify another form of what might be called the ‘open-face’ compliment sandwich (‘OFCS’). The OFCS leads with summary and formal analysis that has no evaluative polarity. It deploys negative or mixed criticism in the penultimate paragraph, which is qualified and alleviated in the final paragraph. Such reviews are not necessarily wholly positive in the final paragraph, but rather the polarity of final paragraph is higher than in the penultimate paragraph. Of the 78 reviews I examined, 35 (44.8%) conformed to the OFCS.

The OFCS is even more predominant than this suggests, because wholly positive reviews generally do not employ this form. Of the 47 reviews that were not wholly positive, 33 (70.2%) met the criteria of the OFCS. Moreover, only two of the seven negative reviews adhered to the OFCS. Perhaps unsurprisingly, the OFCS is most common among mixed reviews, comprising 31 of the 40 (77.5%) mixed reviews. Regardless, it is clear the wholly positive review and the OFCS are the most common mode of ABR reviewing in my sample, since 64 of the 78 reviews (82.1%) belonged to one or both of these categories.

In this sense, then, my survey of ABR fiction reviews between September 2014 and August 2015 suggests that the OFCS is both a major form and the dominant form among reviews that are not wholly positive. The survey also suggests that such reviews, on balance, tend to follow a rough formal pattern: they open with summary information about the author or work, offer detailed non-evaluative analysis of the text in the middle of the review, present their most stringent critiques in the penultimate paragraph, and then qualify or soften such criticisms in the final paragraph. I now want to examine the consequences of these findings by considering their significance, offering some possible explanations, and returning to the question of whether or not ABR reviews are ‘too nice’.

4.Do Open-Faced Compliment Sandwiches Matter?

My survey does suggest that the OFCS is prominent in ABR fiction reviews, and, following the claims applied by Etherington in ‘The Poet Tasters’, this would suggest that ABR reviews are ‘too nice’. Its ‘niceness’ is further underscored by the prominence of overwhelmingly positive reviews, and the relative paucity of negative reviews. If one believes that reviewing should be relentlessly critical and strongly negative where appropriate, then ABR appears to fail this test. What I want to examine now is how such criticism is motivated by what Pierre Bourdieu might describe as the field-position of different agents. In particular, I will make two claims: 1) attacks on overly nice reviews tend to ignore the commercial function of book reviewing, which cannot be easily separated from its critical task, and 2) the niceness (or not) of reviewing seems to be at least partially tied to the field-position of agents and organisations in important, and arguably determining, ways.

Etherington’s critique of too-nice reviewing seems motivated by the idea that it undermines genuine criticism. In ‘The Poet Tasters’ he suggests that the compliment sandwich blunts critical praxis. In ‘The Brain Feign’ he argues that cordial reviewing practices do not apply adequate scrutiny and generate illegitimate praise that can harden into received opinion: ‘Critical acclaim compels us to entertain the idea that this novel’s distinction should be regarded universally to be true’ (2013). Etherington elaborates on these claims in a comment left on an Overland essay also about reviewing practices, arguing that:

I think decline polemics arise out of a keen and justifiable sense of despair—that so many of the public performances of artistic experience (aka ‘criticism’) fail to articulate well the truth of those experiences; and nearly always under the predictable pressures of the distribution of real and symbolic capital. (Brooker 2014)

Here, his frustration—presented in explicitly Bourdieusian terms—is directed towards criticism that reflects the social prestige of an author or publishing house, rather than applying a rigorous analysis outside of the commercial and symbolic valuations of the publishing industry. Etherington worries that reviewing practices might be informed by the commercial imperatives of the book trade rather than a disinterested or at least distanced application of critical rigor.

James Ley makes a similar point in ABR’s own ‘Critic of the Month’ column from 2014 by arguing that ‘so much alleged “reviewing” is transparently chicken-hearted and insipid’ (37). But he diverges from Etherington in arguing that ‘niceness’ is not an adequate metric since it reflects the ‘misperception… that the salient aspect of a review is the critic’s final verdict’ (37). As Ley argues, this undue emphasis on evaluation obscures the fact that the ‘quality of analysis is always more important than one’s personal impressions…. The primary concern of criticism is the meaning of the work, so whatever evaluations might follow are secondary concerns’ (37).

Presumably, the pernicious reviews Ley refers to are both overly evaluative and reflect commercial imperatives. Ley’s suggestion for combatting ‘chicken-hearted’ reviewing is not through Kantian disinterest, but the formation of strong critical dispositions: ‘a critic needs to have some kind of traction, some point of view. A perfectly even-handed critic would resemble the proverbial liberal who refuses to take his own side in an argument’ (37). Ley’s arguments indicate a larger structural critique: the need for critics with ‘traction’ presupposes the existence of a cultural and commercial structure that will foster the growth of what I will describe as ‘strong’ critics—a term that is not meant as a form of subjective praise, but rather an objective description of position characteristics.

I am sympathetic to Etherington’s and Ley’s concerns, and have myself written an essay (‘In the Same Boat’, 2013) in The Sydney Review of Books, which articulated similar concerns about reviewing practices.[3] It needs to be noted that my analysis of ABR reviews potentially substantiates aspects of their claims. It is notable, for example, that, as I mentioned earlier, 18.1% of capsule reviews are negative, while only 5.3% of feature reviews are. Moreover, 41.1% of feature reviews were wholly positive, while only 31.8% of capsule reviews are. This distinction matters, because capsule reviews are much more likely to examine works by debut or lesser-known writers than feature reviews. Indeed, of the three negative feature reviews, one examines a debut work (Dominique Wilson’s The Yellow Papers). The difference in polarity between capsule and feature reviews suggests that the symbolical capital possessed by established authors might affect reviewers to some degree; when taken in aggregate, reviews of works by more established writers, which appear in feature reviews, are more likely to be overwhelmingly positive and less likely to be overwhelmingly negative. At the same time, it could be argued that more established authors are simply more likely to produce works of high quality and less likely to produce bad works.

So, while there may be merit to Etherington’s and Ley’s claims that social and economic capital affect reviewing, their claims largely ignore the fact that book reviews are inextricably tied to the book’s status as a commodity—as evidenced by the fact that virtually all book reviews cover new releases. Book reviews straddle a divide between economic and ‘literary’ notions of value, a distinction already made ambiguous given that published works of literature are always already commodities. Book reviews may contain incisive analysis—it may appear as if they only contain such analysis—but reviews are absolutely a form of indirect marketing presented as a specialised kind of informed consumer recommendation.

Ley’s claim that reviewing is not primarily evaluative runs counter to the commercial conditions that underwrite virtually all forms of book reviewing. Books reviews are a hybrid genre, combining literary criticism, advertising and news reporting (since the publication of a book is a newsworthy ‘event’); this hybridity produces a schizoid split because the genre has its feet planted in two irreconcilable notions of value (the economic and the literary). Ongoing debates about reviewing practices derive from this internal contradiction, which explains why such debates serve as a proxy for questions about the difference between commercial and literary regimes of value. From this perspective, reviewing practices, because they are enmeshed in the commerce of the book trade, inevitably reflect economic and symbolic capital. The campaign against ‘niceness’ in reviewing often does not adequately grasp the intractability of this situation.

Moreover, attacks on ‘nice’ reviews have not always adequately grappled with how such views derive from positions in the literary field. As I noted at the beginning of this article, attacks on literary niceness by Haigh and Etherington have been used to create controversy and discussion around the launch of new literary journals (Kill Your Darlings and The Sydney Review of Books). More recently, The Saturday Paper attracted attention for its book reviews by instituting a policy that its reviewers remain anonymous. These provocations suggest that existing publications—such as ABR—are staid and ‘too nice’, whereas the new publications will offer different and more objective forms of criticism. In other words, the argument against niceness enables new entrants to the field of literary journals to justify their existence and differentiate themselves in a crowded market. Attacks on niceness reflect the position characteristics of upstart journals seeking to challenge the legitimacy of powerful agents or organisations that determine the structure of the field.

A journal like ABR seems to have little to gain from publishing overly critical reviews. It is already established as a prominent outlet for literary reviewing, has an active subscriber base, various forms of institutional support and recognition, and attracts significant private donations. This last fact suggests that many ABR subscribers and stakeholders have attachments—whether formal, informal, or emotional—to established literary institutions. Given this, why would ABR disrupt the circuits of reviewing that underwrite its influence? I also suspect that ABR’s generally civil reviewing practices reflect the expectations of its audience (both subscribers and donors), who want informed cultural recommendations and restrained analysis, rather than literary provocations. In other words, the ‘niceness’ of ABR reviews probably cannot be separated from the position the journal occupies in the field and the concomitant expectations of its readers and stakeholders.

It is also interesting to note that many of the most vocal critics of ‘nice’ reviewing have positions that are related to, but not directly involved with, the book trade. Etherington is an academic. Haigh is known primarily as a sports writer. I am an academic, and James Ley, although he is an active freelance book-reviewer, has a PhD and has written an academic monograph on literary book reviewing (The Critic and the Modern World, 2014). On the one hand, this outsider status enables the capacity to look at the functioning of literary symbolic capital without economic self-interest. On the other hand, the ‘outsider’ status of such critics means they are not subject to the same penalties for violating the rules of the game as those directly engaged with literary commerce. In this sense, outsiders’ criticism of ‘insiders’ ignores the precarious nature of making a living through the publishing industry.

Professional reviewers similarly experience economic precariousness: very few people in Australia can make a living from book reviewing, because the work is typically undertaken on a freelance basis (which is always feast or famine) and there is a paucity of outlets for reviews. As a result, few book reviewers have the economic liberty to cast aside or ignore the conventions of book reviewing, which probably often do encourage civility (or niceness) and discourage overly critical reviews, at least in the case of well-known authors.

There are, however, some exceptions to this tendency. These exceptions are critics who write for publications like ABR but are still able to offer negative assessments even of works that have been highly regarded. I term these individuals ‘strong critics’, both because of their capacity to express negative evaluations outside of the accepted OFCS form and because this capacity is backed by a store of symbolic capital. From this perspective, stringent reviewing would not simply reflect the strength of personal convictions, but rather an agent’s position in the field, which enables him or her to make such claims without fear of reprisal, losing face, undermining relationships, or simply being ignored. In this sense, being a ‘strong critic’ still requires one to ‘play the game’ within the literary field, since the reviewer in question needs to be esteemed (i.e., to have an adequate amount of symbolic capital) and to publish in an outlet or journal that will be sufficiently read (either by other cultural producers or by the broader public, or by both) to have an impact on the field.

One potential example of strong criticism in the sample I analysed comes from Delia Falconer’s review of Mark Henshaw’s The Snow Kimono (2014), which was generally well-reviewed and won the Christina Stead Prize for Fiction in the NSW Premier’s Literary Awards. Despite such accolades, Falconer highlights a variety of problems in the book with analytical rigor, by noting both its intellectual aims, and simultaneously describing the problematic results of the novel’s employment of ‘an oddly affectless, flat prose’:

The effect is like watching the kind of arthouse film in which everything receives lingering attention from the camera—the rain on a window pane, light on a flagstone park—and especially women’s sufferings, as a highly aestheticized element of the mise en scene. (10)

Here, these reservations, along with questions about the novel’s sexual politics are raised in the review’s final paragraph, and the criticism builds towards the final sentence which offers a clearly negative evaluation: ‘I finished The Snow Kimono with a queasy sense of discomfort, and not, I sense, of the sort intended’ (10). Nonetheless, it is worth noting that even this strong opinion is articulated in a highly-personalised and hedged (e.g., ‘I sense’) mode that employs affect as a strategy for softening stringent criticism. Thus, even moments of ‘strong’ criticism—which make critical assertions about books against the grain of broader reception—still employ hedges. Perhaps, then, the debates about the ‘niceness’ of literary reviewing are about field position in the sense that they raise the question of who is allowed to offer strong critiques of novels and under what circumstances.

Emmett Stinson is a Lecturer in English and Writing at the University of Newcastle. His research interests include modern and contemporary literature, Australian literature, publishing studies, and post-digital literature. His monograph, Satirizing Modernism, will be published by Bloomsbury in 2017. He is a Chief Investigator on the Australian Research Council Discovery Project, ‘New Tastemakers and Australia’s Post-Digital Literary Culture’.

Notes

[1] The nature of the Australian literary field remains contested. David Carter argues that the Australian literary field increasingly resembles ‘media cultures’, such as ‘cinema, television and popular music’ rather than Bourdieu’s oppositional model of high and low cultures (Carter 141). Beth Driscoll claims that literary discourse is generated by a middlebrow circuit of production and reception that sits between the notions of avant-garde and popular fiction. I have argued that the Australian literary field comprises a set of producer-consumers who are both audience and participants (Stinson 36-7), a mode of cultural praxis Bourdieu identified with the avant-garde. At the moment, however, I am applying this term in Bourdieu’s sense to refer to the ‘space of literary or artistic position-takings’ that comprises ‘the structured set of the manifestations of social agents’ in relation to writing and literary culture (Bourdieu 30).

[2] Ken Gelder, in Popular Fiction (2004), has argued that popular fiction might be viewed as the ‘opposite of Literature’ (11). While it’s worth noting the differences between these fields, I have not made distinctions between popular and literary works for the purposes of my survey. Indeed, many reviewed works seem to be popular fiction, but the Australian Book Review mostly reviews what appear to be literary titles.

[3] It’s worth emphasising that the criticisms of reviewing offered by Etherington, me and others are hardly objective or disinterested. The Sydney Review of Books was founded at Western Sydney University, and runs out of the same offices as Giramondo publishing; both arguably reflect a set of highbrow literary practices and preferences.

Works Cited

Ballieu, Amy. ‘Wonder and Grief.’ Australian Book Review 369 (March 2015): 36.

Bourdieu, Pierre. The Field of Cultural Production. Ed. Randal Johnson. Columbia UP, 1993.

Broinowski, Alison. ‘Cultural Resonances.’ Australian Book Review 363 (August 2014): 45.

Brooker, Ben. ‘Bad Writing and Criticism: A Response to Alison Croggon.’ Overland. 16 May 2014. <https://overland.org.au/2014/05/bad-writing-and-criticism-a-response-to-alison-croggon/>. 12 May 2016.

Campbell, Mel. ‘Hugging with Knives: The Viciously Cosy Culture of Reviewing.’ Crikey 9 August 2011. <http://www.crikey.com.au/2011/08/09/hugging-with-knives-the-viciously-cosy-culture-of-reviewing/>. 6 May 2016.

Carter, David. ‘Good Readers and Good Citizens: Literature, Media and the Nation.’ Australian Literary Studies 19.2 (1999): 136-51.

Driscoll, Beth. ‘Sentiment Analysis and the Literary Festival Audience.’ Continuum: Journal of Media and Cultural Studies 29.6 (2015): 861-73.

Etherington, Ben. ‘The Brain Feign.’ Sydney Review of Books 29 January 2013. <http://www.sydneyreviewofbooks.com/the-brain-feign/>. 15 May 2016.

—. ‘The Poet Tasters.’ Sydney Review of Books 30 January 2015. <http://
sydneyreviewofbooks.com/australian-poetry-reviewing/>

Falconer, Delia. ‘Dark Activities.’ Australian Book Review 364 (September 2014): 9-10.

Flynn, Chris. ‘Cherry on Top.’ Australian Book Review 371 (May 2015): 30.

Frow, John. ‘On Mid-Level Concepts’. New Literary History 41.2 (Spring 2010): 237-52.

Gelder, Ken. Popular Fiction: The Logics and Practices of a Literary Field. New York: Routledge, 2004.

Goldsworthy, Kerryn. ‘Everyone’s a Critic.’ Australian Book Review 351 (May 2013): 20-30

—. ‘Liminality.’ Australian Book Review 364 (September 2014): 11.

Haigh, Gideon. ‘Feeding the Hand that Bites.’ Kill Your Darlings 1 (March 2010): 9-11.

Holland-Batt, Sarah. ‘Rough Seas.’ Australian Book Review 364 (September 2014): 12.

Levine, Caroline. Forms: Whole, Rhythm, Hierarchy, Network. Princeton, NJ: Princeton UP, 2015.

Lever, Susan. ‘Confined Art.’ Australian Book Review 368 (January/February 2015): 23.

Ley, James. ‘Critic of the Month.’ Australian Book Review 361 (May 2014): 37.

Martin, J. R., and P. R. R. White. The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave Macmillan, 2005.

Menzies-Pike, Catriona. ‘Champagne Flutes.’ Australian Book Review 372 (June/July 2015): 19.

—. ‘Floating Vengence.’ Australian Book Review 371 (May 2015): 31-2.

Morreti, Franco. ‘Conjectures on World Literature.’ New Left Review 1 (January/February 2000): 54-68.

Morreti, Franco. Distant Reading. London: Verso, 2013.

Pine, Louise. ‘To Review or Not to Review.’ Overland 8 August 2011 <https://overland.org.au/2011/08/to-review-or-not-to-review/>. 5 May 2016.

Plunkett, Felicity. ‘Hauntings.’ Australian Book Review 369 (March 2015): 37.

Robertson, Rachel. ‘Listening to Stories.’ Australian Book Review 365 (October 2014): 32.

Romei, Stephen. ‘Are Our Book Reviewers Duds?’ The Australian 2 March 2010. <http://web.archive.org/web/20100306075731/http://blogs.theaustralian.news.com.au/alr/index.php/theaustralian/comments/are_australian_book_reviewers_crap>. 2 May 2016.

Shaw, Martin. ‘“It’s One Thing to Kill Your Darlings, Quite Another to Throw the Baby Out with the Bathwater”—Martin Shaw’s Response To Gideon Haigh.’ Killings: The Kill Your Darlings Blog 25 February 2010 <http://web.archive.org/web/20100301080957/http://www.killyourdarlingsjournal.com/2010/02/%E2%80%98its-one-thing-to-kill-your-darlings-quite-another-to-throw-the-baby-out-with-the-bathwater%E2%80%99-martin-shaws-response-to-gideon-haigh/>. 10 May 2016.

Stinson, Emmett. ‘Small Publishers and the Emerging Network of Australian Literary Prosumption.’ Australian Humanities Review 59 (April/May 2016): 23-43.

Whish-Wilson, David. ‘Dangerous Times.’ Australian Book Review 366 (November 2014): 55.

Wyndham, Susan. ‘A Welcome Edition.’ Sydney Morning Herald 20 November 2010. <http://blogs.smh.com.au/entertainment/archives/undercover/
022465.html>. 10 May 2016.