Literal quotations can be used by reviewers to substantiate their value judgements on the novel[i] they discuss. Quoting is one of the four ways of presenting data that support these judgements. Besides quotations, reviewers can also use non-literal examples, such as ‘The characters are stereotypical’ to argue that the book is not very original (stereotypical characters being an example of an unoriginal book). A summary of the story is another way of supporting the judgement. Showing that a book is not very original can also be done by paraphrasing and thus summarising the story; the same story which may have been used in other novels. A fourth kind of data reviewers can present, is an ‘abstracted summary’. In that case reviewers abstract from what is going on in the novel to what the novel is about. They focus on themes and motives: ‘The novel is about personal freedom, conflicting with social norm and values”.
Quotations can be seen as the ‘purest’ kind of data that can be used to substantiate the judgement in literary reviews because they are the most factual and less interpreted. Therefore, theoretically, quotations are ‘necessary’ for resolving the dispute between reviewers and the readers of their article. The reason for this necessity is that the readers do not know anything about the object being judged (a new novel). Therefore, reviewers should present the data on which they base their judgement. Otherwise, readers cannot adopt a critical attitude towards the arguments that reviewers present to support their value judgements, nor can they decide whether they accept the reviewer’s argumentation, disagree with reviewers, or form their own opinion about the reviewed novel. In addition, reviewers cannot make their standpoint acceptable if their discussion partners do not know the data on which the judgement on the novel is based.
According to F.H. van Eemeren and R. Grootendorst it is essential for resolving a dispute that the discussion partners share common starting-points: jointly accepted propositions or propositions which can be made acceptable by testing them, for example by consulting an encyclopedia or a dictionary. Without these common starting-points it will not be possible to decide when the antagonist is obliged to accept the argumentation of the protagonist, and thus the protagonist will not be able to defend her/his standpoint successfully and the dispute cannot be resolved.
“If applying this procedure (intersubjective identification procedure or intersubjective testing procedure, TU) produces a positive result, the antagonist is obliged to accept the propositional content of the illocutionary act complex argumentation performed by the protagonist. If on the other hand it produces a negative result, then the protagonist is obliged to retract his illocutionary act complex” (Van Eemeren & Grootendorst, 1984: 165-168).
Quotations can also be used purely informative, and interpretative. For example, a reviewer may quote a passage to give the reader an impression of the style of the book. Informative quotations characterise of the book without judging it. Quotations can also be used to support an interpretation. In that case a quotation is an argument that substantiates a claim about a characteristic which does not clearly appear from the novel, without judging it. Interpretative quotations are presented to show that the interpretation is correct, that it is allowed to characterise the (aspect or part of the) book in this way.
In this paper I restrict myself to data supporting a judgement on the novel, which are essential for resolving the dispute between reviewer and reader. Therefore, I shall focus on argumentative quotations. Argumentation by quotations is a form of argumentation by example (Perelman & Olbrechts-Tyteca, 1969: 352). However, in contrast to the descriptions and definitions of argumentation by example in most theoretical overviews of argumentation types, argumentation by quotation is not a form of inductive reasoning, in which the example is one of the many observations that leads to a certain standpoint.
Hastings, Schellens, Kienpointner, Kelley and Reinard typify argumentation by example as inductive argumentation. I do not think that argumentation by example necessarily has to be inductive. An example can be used as an argument to answer implicit questions such as ‘What makes you think that?’, ‘What have you got to go on?’ and ‘Could you give an example?’
The value judgements in literary reviews are not based on several examples from the reviewed book. Reviewers make up their minds about the book while reading, or sometimes even while writing the review. The general claims are not based upon random taken examples from the book but on the whole book. Of course, reviewers mark parts of the book while reading and make notes, but their value judgement on the novel is not derived at several random taken quotations. Quotations are used to justify claims. Reviewers pick them out after they have decided what they will write about the book, or while writing. The quotations are presented as answers to the implicit question of a critical antagonist ‘Could you give an example?’. In terms of S. Toulmin an answer to the question ‘What have you got to go on’(Toulmin, 1969: 98). Quotations as arguments are not examples in a ‘context of discovery’ but in a ‘context of justification’.
2. Critical questions for argumentation based on quotations
Quotations can easily be ‘abused’. It is possible for a reviewer to quote the only awful sentence to substantiate a negative judgement. A reviewer could also change the content of a text fragment by taking the quotation out of context. On the basis of the ‘norms’ for argumentation by example that Ch. Perelman & L. Olbrechts-Tyteca (1969), A.C. Hastings (1962), P.J. Schellens (1985), M. Kienpointner (1992), D. Kelley (1988) and J.C. Reinard (1991) present in their theories, I’ve formulated four critical question for the correctness of argumentation by quotation.
Is the quotation
2.1 Preliminary question
Before I deal with these four questions, I shall briefly discuss one preliminary: Is it possible to substantiate all evaluative claims on novels by means of quotations? Actually, all judgements in literary reviews, on the novel as a whole as well as on specific features of a novel, should in the end be supported by data taken from the book that is being reviewed. Ideally, the evaluative claims should be supported by means of quotations, because, as I have mentioned, they are the ‘purest kind’ of data. The question is whether it is possible to realize this ideal. It is interesting to know which types of evaluative claims can be substantiated by quotations.
In literary reviews the value judgement on a novel is, in general, supported by various so-called sub-standpoints: evaluative claims about features of novels such as style, originality, comprehensibility and moral values. My survey of about 500 literary reviews shows that all 22 types of sub-standpoints[ii] that reviewers use to support the main value judgement, are substantiated by means of quotations.
In the following example, the reviewer uses a combination of an argument from reality and an argument from economy. She states that the author, Maarten ’t Hart, has used too many words, sentences and chapters in his novel. (The last part of the claim, “the novel contains too many chapters”, could never be substantiated by the quotation.) This claim is an argument from economy. However, by expressing that the Red Hot Chili Peppers are not as bad as the author suggests, the reviewer turns her argument at the same time to an argument from reality. According to her, the author does not correctly represent reality. As an argument for these claims, the reviewer has quoted the fragment in which the author writes about the Red Hot Chili Peppers. This quotation should show the lack of economy and the lack of reality of the description.
And as always there is to much. Too many words, too many sentences, too many chapters. Where economy would really be a virtue, there is a lack of it. Maarten ’t Hart suffers from what the English call ‘overkill’. I am not particularly a lover of the Red Hot Chili Peppers, but they are not as bad as he suggests. “Then, through the speakers in the car something could be heard, for which the word ‘roaring’ was definitely an understatement. It was not human anymore, it was frightening, appalling, it seemed to come from cellars where hungry deceased after an atomic war go for each other with cannibalistic intentions.“ (Luis, 1991).
Quotations are being used to substantiate all the 22 types of sub-standpoints. However, this does not mean that reviewers could always easily present quotations to support all these evaluative claims. The frequency of the use of quotations as arguments for these 22 types differs tremendously per type of sub-standpoint. Reviewers rarely present quotations to substantiate sub-standpoints about the social engagement of the author, the moral values of the book, the authors poetics, the degree of identification of the reader, the relative value of the novel considering other books of the author, the fantasy in the novel, the theme of the book or the value of the novel considering that it is part of a trend. On the other hand, quotations as arguments for a sub-standpoint about the style of the novel appear very often. In almost every review that I have examined, the argument from style was being supported by one or more quotations.
Sub-standpoints about autonomous, immanent characteristics of the novel, such as style, composition and ‘para-aesthetic value’ (for example humour), are often supported by quotations. Those sub-standpoints relate only to the book itself and can therefore only be justified with data taken from this book. Abstract characteristics that reviewers ascribe to novels, such as the social engagement of the author or the moral values of the book, do not easily show from quotations. When reviewers use data to support sub-standpoints about these types of abstract characteristics, they often present an ‘abstracted summary’[iii].
There is another, more simple reason why quotations are frequently being used to support sub-standpoints about characteristics like style, and are rarely being used to support sub-standpoints about characteristics like moral values. Reviewers use certain types of sub-standpoints more frequently than others. Sub-standpoints about the degree of realism, the emotional effect on the reader, the originality and the composition often appear in the reviews that I have examined. The sub-standpoint about style can be found in almost all literary reviews.
2.2 Is the quotation correct?
A quotation is, in the first place, only a correct argument if the text that is being quoted is in accordance with the text in the book. The quotation must be correct: the text must be verbatim. All kinds of changes, like inversion, must be specified and should not change the nature or meaning of the quotation.
2.3 Is the quotation representative?
Secondly, a quotation is only a correct argument if it is representative of (the parts of) the book that is being reviewed. If a quotation is not representative, the range of the claim is, in general, wider than the range that is being justified by the quotation. A non-representative quotation only justifies an evaluative claim on itself, not on the novel in general.
However, M. Kienpointner’s examples of holiday and restaurant experiences show that a non-representative example can also justify a judgement. On the basis of the rule that certain things are not allowed even once (lousy dinners in a restaurant), one exception (one lousy dinner) can be sufficient ground for a negative judgement (the dinners in this restaurant are lousy, therefore I am not going there anymore) (Kienpointner, 1992: 366-367).
On the basis of the rule that in a novel not one cliché passage may occur, one quotation of a cliché passage could be sufficient to justify a negative judgement on the originality of the novel. One awful sentence, one grammatical mistake or one ugly metaphor can, on the basis of such rules, be quoted to substantiate a negative claim on the style.
2.4 Is the quotation sufficient?
Thirdly, a quotation (or quotations) is only a correct argument if it is a sufficient argument. However, the number of quotations that is required to support a claim sufficiently, cannot be determined. In general, one quotation, either as a representative example, or as an exception, will be sufficient to justify a judgement on a feature of the book.
2.5 Is the quotation relevant?
Fourth, a quotation is only a correct argument if it is typical. The question whether the quotation is a relevant example, depends on two sub-questions. First, is the quotation typical for the evaluated characteristic of which it is an example? When a quotation sub-stantiates a sub-standpoint on the social engagement of the author, the social engagement should appear from the quotation. There should be a plausible relation between the quotation and the characteristic. Second, the suggested relation between the quotation and the judgement on the characteristic should be plausible. The question is whether the quotation is justifying the evaluation. Is quotation X an example of a beautiful style? Is the quotation that is supposed to show how cliché the book is, really cliché?
When the reviewer presents a quotation to substantiate the claim that the style is beautiful, the readers can decide themselves whether the quotation supports this claim, whether a beautiful style does appear from the quotation. It should therefore be clear what is supposed to appear from the quotation. However, quotations do not always speak for themselves. The reviewer will sometimes have to make the relation between the quotation and the claim explicitly clear. That can be done by commenting on the quotation. In a comment reviewers can, for example, make clear how the social engagement of the author shows from the quotation or they can indicate the awfulness of the quotation that supports the negative evaluation of style.
Whether the quotation really shows what it is supposed to show, is related to what Quiroz and others call the ‘argumentative direction’ of an argument. One could question the ‘argumentative direction’ of an argument and state that the argument is actually substantiating the opposite conclusion (Quiroz, ea., 1992: 174-175).
When the reader finds a quotation an example of beautiful style, when the reviewer meant it as an example of awful style, the argumentative direction of this quotation is opposite. The argumentative direction is also opposite when the reviewer finds a quotation extremely funny and the reader does not. This difference has got to do with the subjective criteria for judging novels and with taste. However, it could also be a consequence of misunderstanding if the reviewer does not make clear what is so funny about the quotation.
I assume that quotations in literary reviews are correct and representative. I have made this assumption not only because it is impossible to answer these two questions without analysing the novels that are reviewed, but also because the readers assume the quotations to be correct and representative. Readers trust reviewers. Reviewers are not supposed to mislead their readers. It can be seen as a kind of Gricean sincerity condition that reviewers present correct and representative quotations.
3. Difficulties in using quotations as arguments
Quotations cannot always sufficiently justify the claim they are supporting. In the first place, some sub-standpoints cannot be totally justified by quotations because they do not only relate to features of the novel. For example, the quotation that substantiates the claim that the Red Hot Chili Peppers are not as bad as the author Maarten ’t Hart suggests, only shows that the author presents an exaggerated description of this music. A quotation could never show what this music is really like. Secondly, the range of the claim can be so wide, that it cannot be supported by quotations. For example, it cannot show from quotations that a novel contains too many chapters. Thirdly, some claims can only be substantiated by more than one quotation. One quotation does not suffice, for example, to show that a certain phenomenon occurs ‘repeatedly’ in the book.
The quotation in the next example is supposed to show that the story is continuously being interrupted by turns from the third person singular to the second person singular, a case of excess. However, it does not appear from this short quotation that the story is continuously interrupted by turns from the third person singular to the second, and that excess is the case. In addition, this quotation only shows that the second person singular is used in the book, not that also the third person singular is used, nor that there are turns from the third to second person.
Perhaps the text should have it from its structure? The story is continuously being interrupted by turns from the third person to the second person, which addressed Hanna as it appears: “You had a clear desire to grow up, you were looking forward to that time, you were not afraid of it“. But also in this case: you can have to much of a good thing (Schouten, 1990).
It is impossible to determine, in general as well as in a specific case, how many quoted words are necessary to support a claim sufficiently. Due to lack of space it is not always possible to present as many quotations as needed to justify a claim sufficiently. There is always a lot of argumentation ad verecundiam in reviews, even if the reviewer quotes. The readers will have to trust the reviewers and will have to assume that they have sufficient grounds for their claims. “That is true: everybody who reads a review knows it, and the demand for thousand-and-one arguments is an absurd demand, because not even endless space will be enough to remove distrust of the judgement of the reviewer“ (Van Deel, 1982: 22).
In some cases, it could be unclear what a quotation is supposed to show or a quotation may not show what it is supposed to show. For example, a quotation that is used by the reviewer to show that a book is funny, may not be funny to the reader. The reader does not understand what is so funny about it.
A quotation can also not show what it is supposed to show because the argumentative direction is called in question. In that case the reader understands why the quotation is funny, but (s)he thinks the quotation is not funny, but silly or dull. In this case, the opposite of what the reviewer meant appears from the quotation.
For example, the last quotation in the next fragment is supposed to show the ‘irony of the stopgaps’. However, this does not appear from the quotation. It is unclear which word is the stopgap because all of the words in the quotation only occur once, and it is unclear what is so ironic about these words.
Everything in this novel is ‘in a manner of speaking’: the childish and distant way of narrating, the old-fashioned chapter titles (…), as well as the irony of the stopgaps (“All men only think about one thing: sleeping“) (Goedegebuure, 1991).
In literary reviews, evaluative claims are presented about books which are unknown to the readers. To substantiate and justify the evaluative claim on the novel and to resolve the dispute with the readers, reviewers should present data from the novel that is being reviewed. Literary reviewers can provide factual data because books consist of words. They can copy material from the book into their reviews by quoting. Quotations, as a mean of presenting factual data, can only be used in book reviews. In reviews about theatre, sculpture or painting, no factual material from the work of art that is reviewed can be added. After all, a picture of a painting is not the painting itself.
i. I have restricted my research on quotations as arguments to Dutch reviews on Dutch novels from 1990 until 1997 in daily and weekly newspapers.
ii. Argument from reality, abstraction, engagement, moral, expressive, intentional, authors poetics, composition, stylistic, emotional, identification, didactic, originality, tradition, relativity, fantasy, comprehensibility, development, economy, theme, ‘para-aesthetic value’ and trend (Boonstra, 1979 & Praamstra, 1984).
iii. Reviewers could support these kinds of arguments of course also by summarising (retelling) the story of the novel or by non-literal examples. However, I have seen very often that an ‘abstracted summary’ is presented as an argument in these cases.
Boonstra, H.T. (1979). Van waardeoordeel tot literatuuropvatting. De Gids, 243-253.
Deel, T. van (1982). Over recenseren. Voorburg: Protestantse Stichting tot bevordering van het Bibliotheekwezen en de Lectuurvoorlichting in Nederland.
Eemeren, F.H. van & R. Grootendorst (1984). Speech acts in argumentative discussions. Dordrecht: Foris Publications.
Goedegebuure, J. (1991). Gewapende vrede. HP / De Tijd, 18-01-1991.
Hastings, A.C. (1962). A reformulation of the modes of reasoning in argumentation. Evanston: unpubl. diss. Kelley, D. (1988). The art of reasoning. New York / London: W.W. Norton & Company.
Kienpointner, M. (1992). Alltagslogik. Struktur und Funktion von Argumentationsmustern. Stuttgart – Bad Cannstatt: Frommann-Holzboog.
Luis, J. (1991). De vorstelijke geur van natta jassen. NRC Handelsblad, 01-11-1991.
Perelman, Ch. & L. Olbrechts- Tyteca (1969). The New Rhetoric. A Treatise on Argumentation. London: University of Notre Dame Press.
Praamstra, O. (1984). De analyse van kritieken. Voortgang. Jaarboek voor de Neerlandistiek 5, 241-264.
Quiroz, G., D. Apothéloz & P.Y. Brandt (1992). How counter-argumentation works. In: F.H. van Eemeren e.a (eds.): Argumentation Illuminated. Amsterdam: SicSat, pp. 172-178.
Reinard, J.C. (1991). Foundations of argument. Effective communication for critical thinking. Dubuque: W.C. Brown Publishers.
Schellens, P.J. (1985). Redelijke argumenten. Een onderzoek naar normen voor kritische lezers. Dordrecht: Foris Publications.
Schouten, R. (1990). Godsdienst en hitte op het platteland. Trouw, 29-03-1990.
Toulmin, S. (1969). The Uses of Argument. Cambridge: Cambridge University Press.