Microsoft Word - 34.2Carrascal.pubverion.doc © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. The Authority of Citations and Quotations in Academic Papers BEGOÑA CARRASCAL Dept. of Logic and Philosophy of Science, University of the Basque Country, UPV/EHU Avda. de Tolosa 70, 20018 Donostia-San Sebastian, Spain b.carrascal@ehu.es Abstract: I consider some uses of citations in academic writing and analyze them as instances of the “appeal to expert opinion” argumentative scheme to show that the critical questions commonly linked to this scheme are difficult to apply. I argue that, by considering citations as special communicative and argumentative situated acts, their use in real practice can be explained more adequately. Adaptation to the audience and to the social constraints is common and necessary in order to collaborate with others and to advance in a discipline, but also to attain rhetorical goals that differ from strictly cognitive ones. Résumé :J’analyse les usages de citations dans les travaux académiques comme des exemples du schème argumentatif de «l’appel à l’opinion d’expert» et montre que les questions critiques couramment liées à ces schèmes sont difficiles à appliquer. Je soutiens qu’en interprétant des citations comme des actes communicatives et argumentatives dans un contexte académique on peut mieux expliquer leurs usages. L’adaptation à un auditoire et aux contraintes sociales est courante et nécessaire afin de collaborer avec les autres et de progresser dans une discipline, mais aussi afin d’atteindre les objectifs rhétoriques qui diffèrent des objectifs strictement cognitifs. Keywords: act of arguing, argument from expert opinion, authority, citing, communication, testimony 1. Introduction1 The amount of information available on almost any topic has increased enormously and access to it has become easier thanks to new technologies of information and communication. However, it is not always straightforward to sort through the entire, complex range of literature in order to find appropriate information to make a practical decision about a matter directly and on our own. Moreover, this information is often 1 A preliminary version of this work entitled “Authority Arguments in Academic Contexts in Social Studies and Humanities” was presented at the OSSA 9 conference. Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 168 technical or there are too many options to weigh up and assess in sufficient depth. To overcome these difficulties, we rely on the opinions of so-called experts. Experts can roughly be described as specialists in a particular field or profession. This expertise or knowledge in the field gives them special authority and identifies them as being reliable and able to help in making many decisions in everyday life. We rarely question their expertise; we just rely on them and use their opinions or guidance to progress and decide among different possibilities present in many facets of life. In other words, we accept experts as a kind of authority on which we can depend. Nevertheless, the appeal to authority as an argument has traditionally been considered weak or even fallacious, mostly because of authoritarian and non-authoritative applications of it, but also due to the predominance of classical deductive logic as a normative model of reasoning and/or of arguing in classical textbooks. There is still ample discussion on how to define and assess expertise. Common considerations of expertise include what it means to possess knowledge about an issue and whether the expert should be recognized as such by the community. In this work, my concern is not the general problem of expertise. This issue only arises incidentally because of the consideration of citations in academic papers as assertions of experts in a field. As a consequence, the problem of expertise is dealt with only insofar as needed to evaluate citations as appeals to expert opinion. Herein, I attempt to analyze some uses of citations as instances of the argumentative scheme “appeal to expert opinion”, as presented by Walton (1997), and investigate whether the common list of critical questions linked to this scheme is applicable here. The scheme has been considered mostly in dialogs between an expert in a field and a layperson; in contrast, I evaluate whether special requirements are needed when the discussion is among experts, as is the case for academic papers. That is, I consider the so-called “expert/expert” problem (Goldman 2001), in which one expert seeks to appraise another expert’s assertions. In Section 3 of this paper, I consider citations as sources of information and knowledge via the words of others; that is, I think of them as instances of testimony, as in the case of epistemological studies. There have been lengthy discussions about the justification of testimony, given the fact that speakers are not Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 169 always sincere and truthful when talking with others. Recently, interest in this subject has increased (Lackey and Sosa 2006; Lackey 2010) and has involved attempts either to find a common definition of the many different practices under the name of testimony, or to find ways of combining classical views on testimony by situating it in its actual context of use. I believe that current positions that consider testimony a social and communicative practice, subjected to special conditions defined by the context of utterance in which the testimony is given, may enable us to develop a better approach to this phenomenon. As a consequence, by discussing the case of citations, I attempt to give more support to the contextualist view on testimony, considering how the special and institutional context in which academic citations appear helps to explain our epistemic responsibility in accepting them. That is, I support the need to consider testimony as a social practice subjected, first, to general conditions of communication; second, if used in an argumentative setting, to the special conditions of “the game of giving and asking for reasons” (Brandom 1994, p. 15); and third, to specific conditions defined by the communicative and social setting in which this practice is used. 2. Quotations as instances of “the appeal to expert opinion” argument? The appeal to previous work on an issue is difficult to avoid in academic writing and in conference presentations and has become standard practice in all fields of knowledge. Moreover, the use of citations is associated with and rooted in the methodology of writing academic papers. In many books on critical thinking, we can find generalized and similar advice on writing a paper, for example: “Base your paper on research rather than on your own unfounded opinions” or “Don’t present other people’s ideas as your own” (Ruggiero, 2003, p. 110). This is standard and part of institutional requirements regarding clarity over intellectual ownership2. As Willard (1990) states, citations can be used strategically to fulfill different functions other than to support one’s argument: 2 I would like to thank an anonymous referee for pointing out this issue and for their thorough review, which significantly contributed to the improvement of this paper. Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 170 Citation, of course, serves many functions—to acknowledge debts, identify allies and opponents, clarify and illustrate claims, display competence, and (especially in journals) to acquiesce to editors. These functions may have unintended side-effects on a par with the most explicit function for citation—as a mode of proof (Willard 1990, p. 11). Walton and Macagno (2011) present many insightful examples of different strategies in which quotations (and misquotations) function not only as reported assertions but also, for example, as reminders of past commitments of the quoted. They show that not all the uses of quotations imply that the proponent commits necessarily to the quoted words because, for instance, she can use the quotation only to show the existence of a particular result or theory in a field, or even to attack it (p. 29). Nevertheless, when a quotation is used, it places the burden to disprove it on the reader because the quotation, maybe for the simple fact of having been chosen, is usually considered true (p. 46). This can be illustrated by some examples. The first one comes from a paper by Sperber (2001, p. 1)3: There may be cases and situations where it is adaptive for a cognitive system to introduce systematic biases, for instance of excessive caution or on the contrary of overconfidence (see Stich 1990), but such cases are, I believe, marginal. We should generally expect the beliefs produced by an evolved cognitive system to be true. In other terms, cognitive systems are basically producers of knowledge. In this quotation, Sperber cites a paper to concede a point: he thinks that what Stich says may be true (or acceptable for many), but he does not accept it in general and states that what Stich says applies only to some marginal cases. The quotation shifted the burden of proof to his reader (Sperber) who feels compelled (in the paper) to argue otherwise to resist a generalization. Even in the case of a unique citation, as in the former case, the burden of proof can shift to the reader, mainly when 3 The references of the citations in the examples are given in the quoted work and do not appear here because they are merely illustrations. Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 171 the citation or quotation comes from an authoritative source that is associated with knowledge or esteem. Walton and Macagno (2011) state that “the strength of these arguments lies on the association of knowledge or esteem and plausibility or truth” and that quotations deployed in this way can help to shift the esteem from the quoted to the defended claim (p. 31). Walton and Macagno remark that quotations can be used to support an appeal to authority argument designed to support one’s own argument (p.30). Trust in the authority of the quoted source may give us a reason to accept an argument without examining how it was originally defended. There are many examples of this in forensic and public debates; as an example of a scholarly work, we present the following quote from Tindale (2004, p. 29): As Kerferd (1981) notes, various concerns have been deemed the distinguishing mark of a Sophist, and one of these was clearly “the educational ideal of rhetoric” (35). Tindale quotes Kerferd as an expert and, by doing so, he is presenting the statement as true (or as justified in Kerferd’s work) to his readers and of use to justify his several ulterior claims, for example, that it may be difficult to attribute to the Sophists the teaching of rhetoric as it was understood by Plato and Aristotle (p. 29). Another example illustrating how citations are authoritatively used to support a point comes from Mercier and Sperber (2011, p. 60)4: Reasoning enables people to exchange arguments that, on the whole, make communication more reliable and hence more advantageous. The main function of reasoning, we claim, is argumentative (Sperber 2000a; 2001; see also Billig 1996; Dessalles 2007; Kuhn 1992; Perelman and Olbrechts-Tyteca 1969; for a very similar take on the special case of moral reasoning, see Gibbard 1990 and Haidt 2001). The citations in this passage are presented to justify the authors’ claim that the main function of reasoning is argumentative. If a reader wanted to refute the claim, she would have to look at all these citations to identify, first, whether what they say supports this claim and, then, whether 4 This example is similar to the one used by Willard (1990, p. 11). Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 172 those assertions are sufficiently justified. As the number of cited works is huge and authoritative, the pragmatic presumption towards the claim is powerful (Willard 1990). Another somewhat different example, in which the authority of well-known philosophers is clearly used to support a point that, at least at the time of writing, was not universally accepted in the field, can be seen in the following example by Tymoczko (1986, p. 48): For mathematical justification turns on the ability to speak and use a language of mathematics. However, according to Quine, language is a social art and according to Wittgenstein and Kripke, private languages are impossible. Thus mathematical language must be a public language. It follows that mathematical justification is a social art and requires a community to practice it. Quine, Wittgenstein, and Kripke, analytical philosophers, are cited to support a point that at the time was (and still is) controversial. Their assertions are taken as true and used to justify Tymoczko’s claim. He knows about the controversy of the claim, but defends himself against it by appealing to distinguished philosophers in the field. A reader may find herself overwhelmed by the citations and so be prompted by the authority of Quine, Wittgenstein and Kripke to accept Tymoczko’s claim. It is well known that appeals to the authority of classical renowned philosophers are common in philosophical works. In spite of the many cases of misquotations and fallacies we may encounter in texts, quotations rest mainly on the presumption of being true (Walton and Macagno 2011, p. 52). We want to explore whether we can explain this fact by considering the two acts that a writer performs when quoting: first, the writer displays the statement, assertion, utterance, or saying of the quoted source or, more precisely, her interpretation of it; then, the writer invokes the authority of the quoted source, an expert in the issue at hand, as a reason to consider it true (or, at least, presumptively true). Brandom (1994) sees in the appeal to the authority of another asserter a way of vindicating a commitment to a claim by demonstrating entitlement to it (p. 175). He calls this form of support the legitimacy of an assertion by “deferring to the claim of another” a “person-based authority” in contrast to what he calls “content-based authority”, which is invoked by Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 173 justifying the claim through the assertion of other sentences from which the claim can appropriately be inferred (p. 175). In this way, a user discharges her responsibility to demonstrate her entitlement to an assertion by passing along to the authority any demands for its demonstration. Traditional accounts of authority distinguish between authority held because of social status or the position the person holds in society, and authority associated with the person’s knowledge in a given field of expertise. In scholarly contexts, it is clearly the latter that should be involved, but as Tindale (2011) states, the distinction is far from clear because the authority of experts in a field is associated with their previous performance, knowledge, and track record, and this entails experts exhibiting “authority” not only on the basis of their actual work, but also because of their academic status. Walton (1997) proposes that authority of the cognitive kind should be used in a fairly weak sense, as a challengeable source and open to critical questioning, but “one that is given a certain standing or weight of presumption where direct access to knowledge (or “the facts”) is not available within the practical constraints of arriving at a prudent conclusion on how to proceed in argumentation” (p. 85). In this sense, he does not contemplate authority in the internal scientific reasoning itself that led to a particular conclusion within the scientific discipline or research investigation because he thinks that, in a scientific investigation, “an impersonal testing and cumulation of evidence is supposedly what matters in an argument” (p. 17). Nevertheless, this is clearly not the case in philosophical discussions in which philosophical questions are time and again challenged and consensus is rarely attained. Time and continuous debate are necessary to reach an agreement about the interpretation of a given passage or a corpus of reasons and acceptable procedures in the field of philosophy. This is the reason why we want to analyze the use of authority in citations (mainly) in philosophical academic papers. Walton’s account identifies authority of the cognitive type with expertise or knowledge, linking argumentative uses of this appeal in a field with the “appeal to expert opinion” argumentation scheme.5 In the “appeal to expert opinion” scheme, Walton considers mainly an expert and a user of the information provided by the expert, who is not herself an expert, because he is thinking of uses of (scientific) expertise in public or 5 Walton considers also the argumentation scheme “from position to know” for a less specialized type of cognitive authority, an authority not linked to a specific domain of knowledge. Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 174 forensic contexts. However, are citations or quotations in academic papers (in philosophy, for example) not assertions made by experts in the field? Moreover, Walton considers that “the notion of any designated group, such as philosophers as being wise or sources of wisdom, over and above any expertise based on knowledge of a discipline is not one that has general acceptance” (p. 40), so I attempt here to determine whether the “appeal to expert opinion” argumentation scheme can be applied to analyze and evaluate those uses of citations in academic philosophical papers. There are several forms of the general “appeal to expert opinion” argument in the literature, but a general form of it is as follows: An expert E in subject domain D asserts a statement S (in domain D). Therefore, S is true. This argumentation scheme could be deductively closed by assuming an implicit premise that would more or less state: “Everything expert E says about a subject in domain D is true”. As this premise is usually false, the scheme is more often considered a defeasible argumentation scheme and the implicit assumption is stated as follows: “Everything expert E says about a subject in her field of expertise D is presumptively true”, so, the conclusion comes to “S is presumptively true”. Hastings (1963) presented the first systematic analysis of common argumentation schemes (Walton 2005). After his work, it is typical to present a set of critical questions matching the scheme together with the premises and the conclusion representing it. This normative account can be used to analyze and to evaluate any instance of the scheme and also, in an argumentative exchange, to cast doubt on its cogency or validity. When one of these questions is raised in a discussion, the burden of proof automatically shifts back to the proponent of the argument. An adequate answer to these questions may be a way to repair poor arguments, and the failure to do so a sign of the need to discard bad or fallacious ones. Common questions associated with the argumentation scheme “appeal to expert opinion” in the literature refer either to the assessment of several aspects of the content of the assertion of the expert or to the assessment of the expert herself. The list proposed by Walton (1997, p. 223) is as follows: Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 175 (1) Expertise Question: How credible is E as an expert source? (2) Field Question: Is E an expert in the field that S is in? (3) Opinion Question: What did E assert that implies S? (4) Trustworthiness Question: Is E personally reliable as a source? (5) Consistency Question: Is S consistent with what other experts assert? (6) Backup Evidence Question: Is E’s assertion based on evidence? If citations are instances of the “appeal to expert opinion” argumentation scheme, in order to ensure its proper application and our commitments as critical thinkers, we should be aware of the critical questions associated with it, so we should have them in mind to ensure the good application of the scheme. However, when writing a paper about a philosophical issue, are these critical questions adequate to assess and to improve our argumentation and thus to write a better argumentative text? Question (1) or expertise question, about the credibility of an expert source, is usually linked to the identification and track record of the source (Tindale 2007; Walton 1996). Identification of a source is a requirement of the institutional context in which academic papers are published and is (or should be) always met. However, I do not think that the credentials of the source should be considered a necessary condition to choose a citation in the academic context. This may be a factor explaining why well-known authors are more frequently cited, but in an academic context, the reason to choose a citation should be, in my opinion, more related to its content than to the track record of the author. Moreover, a reader of the citation may want to check it in order to learn more about how it was justified in the original paper, but I do not think that a lack of credentials should necessarily be considered as a reason for rejecting a citation. In fact, we would not want to say that all citations of statements made by young researchers are to be put aside as unreliable. Question (2) is related to the field in which the expert has to be situated with respect to the citation. This is also usually linked to the consideration of the field as a body of knowledge. It is true that, in philosophy, it is difficult to set the boundaries of the discipline and that, in contrast to the case of scientific domains, there is not a general standard consensus Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 176 about many parts of it. As a consequence, this question is more difficult to answer than in cases related to technical matters, but would we really want to state that philosophy does not constitute a body of knowledge? Moreover, in communications between experts, as in academic papers, how should we define being a member of a field? For example, how should we set such boundaries for the field of argumentation: in a narrow sense (pragma-dialectics, informal logic, argumentation, logic, etc.), or in a wider sense, so as to include substantial parts of the history of philosophy or cognitive psychology? It is true that, as Walton and Macagno (2011, p. 39) state, “wrenching a proposition from context is often used to exaggerate a position or draw inferences from the quoted words that do not really represent the arguer’s position”. As writers, we know that being truthful is part of the requirements of any communicative act, so we have to be aware of Question (3) in Walton’s list when quoting. There may be cases in which the obscurity of a paper makes us interpret it differently from the intended meaning of the writer, but this is difficult to avoid if we cannot directly confront her. Moreover, the number of citations in academic papers makes it difficult to check them all. Are we being irresponsible when acting in this way? Trustworthiness of an academic source may well arise as a result of her previous performance in the field, so we could link the answer to Question (4) with that to the expertise question. Nevertheless, all experts have their own (not necessarily vested) interests and, in philosophy, this means that they may have particular modes of focusing on and approaching a philosophical problem. It could be a particular case of, for example, a well-known (political) philosopher who uses her knowledge to defend in a paper a certain political or moral position that is rejected by many others. If somebody quotes her, the quotation could well be judged biased by many and, as a consequence, not worthy of use in defending any other claim. However, in philosophy, there is a multiplicity of positions about almost every important philosophical question. Does this invalidate the use of any quotation? In my opinion, what it means is that many philosophical discussions may never end, but this does not mean that, in the process, many points or positions about the issue will not be clarified or explained. As Gelfert (2011) points out, “as long as A and B engage in a shared practice of tracking and coordinating their differences in epistemic outlook, the mere lack of substantial agreement does not render such attempts at resolution futile” (p. 309). Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 177 Questions (5) and (6) are also difficult to apply in the case of citations in philosophical papers: as already mentioned, consensus is not generally expected and direct knowledge about an issue is not, in general, available and, in most cases, it would take us through a chain of citations that would eventually lead to a circular argument. A possible sub-question proposed in the literature (Walton 1996, Tindale 2007) involves the relevance of the expert’s assertions with respect to the defended claim. The fact that a quotation has been chosen raises the expectation of it being relevant. However, the relevance of the quote considered by the writer and the interpretation of it by a reader may differ. Nevertheless, we should be careful because, if the quotation is from a well-known source, the expectation of relevance increases, even when its interpretation may be difficult. An opaque formulation of a well-known source may lead us to accept it even if we do not fully understand it because it is associated with high expectations of deep and important thoughts (Sperber 2010). We may thus be led to interpret this obscure claim as somehow relevant to prove the claim in the paper, even when the original thought behind the quotation was not so. Another sub-question relates to claims of inconsistency to the previous performance of the expert. In academic philosophical work, a change of position with respect to an issue could surely be explained as evolution in the development of one’s position or as a consequence of developments in the field, although it could also be invoked by a critic as invalidating the original quote. However, inconsistency with respect to other statements in this field is, in my opinion, related to the non-expected consensus. Recently, in several studies on expertise from the fields of both argumentation (Zenker 2011; Goodwin 2011) and sociology of science (Collins and Weinel 2011), doubts have been expressed about the suitability and applicability of all or some of the questions on the list. Nevertheless, the questions can be of use to analyze and evaluate cases of appeals to expert opinion in some definite public contexts, for example, to assess cases of testimony in a court of law, or cases in which a doctor’s diagnosis differs from other diagnoses or data. However, the use of some of these questions in the case of citations in scholarly papers in the field of philosophy is difficult. As a consequence, should we consider that the special institutional constraints imposed by the academic context and the discipline require special considerations in this case? Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 178 The gap between a normative theory of argument and its practice has been considered in several papers (for example, Pinto 2001; Johnson 2005; Kvernbekk 2012). It is clear that a normative account such as Walton’s should not be solely criticized on empirical grounds, but I think that by applying the method of analysis, we lose track of the way in which these kinds of arguments are dealt with in actual practice and how we grasp the idea of defeasibility and subsequently respond to it in an active way, when we do not agree with the supported claim. Thus, in the next section, I relate citations to testimonial and communicative practices because I think that, by doing so, we can obtain some helpful ideas to develop greater awareness of the argumentative use of citations in academic papers and, as a consequence, a possibility of improving the theory. 3. Testimony and communication In epistemological studies, testimony is generally defined as one of the classical sources of acquiring knowledge and “to come to know that p on the basis of someone’s saying that p” (Steup 2010). Steup states that the term “saying that p” should be understood broadly, so as to include many ordinary utterances in conversational settings and also in most of the ordinary forms of written communication. This definition includes an overly broad range of phenomena under the general heading of testimony, but it allows us to begin the discussion because statements in academic papers, considered as means to convey or to communicate knowledge on the basis of someone’s words, fit the above definition nicely. Thus, in this section, we consider quotations and citations as instances of testimony as defined in epistemology, that is, as sources of justified beliefs based on what others say. Coady (1992) set the agenda for work on testimony for years to come; since then, numerous papers have been published on this topic. Many of them present the traditional divide regarding views on the problem of testimony (see for example, Lackey and Sosa 2006 or Lackey 2010 for references). The discussion roughly circles around whether accepting or believing the assertions told to us by a speaker (or a writer) is or is not sufficient for those beliefs to be justified or to constitute knowledge. The problem arises due to the fact that our dependence on other’s words is central in many facets Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 179 of our lives and that not all kinds of communicative act are truthful. To solve this puzzle, there are two main classical proposals, the so-called reductionist view and the opposite or anti-reductionist one (Lackey 2010). For reductionists, a testimony should be backed up by a chain of knowledge claims that would eventually end up in a speaker who knows about the issue directly and by a more reliable source, say, by perception. On the other side, the anti-reductionists claim that the person who gives testimony is a warrant by herself, and that testimony should not be reduced to other classical sources to ensure its reliability as a source of knowledge. Both positions have been challenged; that of reductionists because they somehow underestimate the many contributions made by testimony by reducing them to other sources of knowledge and that of anti-reductionists for the several cases of gullibility that can occur if recipients of testimony are not required to do any epistemic work. To accommodate these critics, both views have had to consider non-testimonially grounded positive reasons for testimony acceptance, and hybrid versions of both of these views have been developed (Lackey 2010). Lackey states that such positive reasons include elements about the speaker’s reliability, the interpersonal relationship between the two parties in the exchange, or about the particular constraints of the social context of the utterance. Not every communicative act can be considered an instance of testimony because we do not always come to know something from the words of others: there are conversation fillers, for example, that do not convey information, or it can be the case that the information may already be known or inferred by other means (Lackey 2006, p. 2). However, every instance of testimony occurs via a communicative act and, as a consequence, as with other communicative acts, testimony is subjected to the general characteristics of human communication. For communication to be successful, both speaker and hearer have to play their parts in the exchange. Moreover, Lackey considers the interaction of both parts to be necessary to obtain knowledge by testimony. Communication is not always truthful and it should not be analyzed solely in terms of truthfulness; indeed, the different theories of pragmatic communication have widely supported this fact. If we take Grice’s cooperation principle and its associated maxims then, although the quality maxim asks us to be truthful, it is by flouting this maxim that the actual sense of the utterance, that is, what is communicated, can be conveyed in many cases. If, instead, we adopt Sperber Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 180 and Wilson’s “relevance theory”, the communication is defined in terms of maximizing benefits and minimizing effort. This commits us to a general view on the truthfulness of the communication because, otherwise, communication would be impracticable, but this general principle does not entail that any communicative act has to be truthful. It depends on the balance between benefits and costs of a particular act between two particular actors. In this view, when testimony is given, first we have to interpret the speaker’s intended meaning, taking into account that her utterance is expected to be beneficial for both parts. Testimony is widely used in everyday communication and, as a consequence, it is unavoidable as a way of acquiring new beliefs; if pragmatic principles of communication apply to the uses of testimony, it is not realistic to say that, in the flow of communication in a dialogue, we will have the time or the means to check the truthfulness of all the utterances to which we are continuously being exposed. Communication occurs rapidly, so this would be uneconomical and psychologically implausible (Sperber and Wilson 1986/1995). However, acting in this way, are we being epistemically responsible when we accept what we are told as true or at least reliable? Adler (2012) proposes a kind of default rule, which he considers as an explicit norm that serves to give epistemic support to testimony in general: If the speaker S asserts that p to the hearer H, then under normal conditions, it is correct for H to accept (believe) S's assertion, unless H has special reason to object (Adler 2012). As explained by Adler, the epistemic support for this norm derives from another thesis proposed by several epistemologists who claim that we only assert p if we know about p (see Adler 2012, p. 9 for references). Adler considers the acceptance of testimony by the default rule: “as true issues in full, all-out, or unqualified belief, as contrasted to some degree of belief”. However, as he states, it is also well known that there are numerous circumstances and examples that do not seem to accord with the above-mentioned claim and so to which the default rule does not seem to apply: the testimony of a witness in a law court is not by default accepted by the jury; a claim in a paper is not accepted without backing. In many of such cases, internal constraints can be found because these practices have been established and evolved in formal Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 181 institutional contexts (Kauffeld and Fields 2003; van Eemeren and Houtlosser 2005). Testimony considered as a communicative act does not come out of the blue. Communication arises in a definite and particular conversational context in which a lot of background information is assumed, so testimonial acceptance (or not) could be explained by having a look at the communicational context surrounding the utterances. This context or “shared cognitive environment” (Sperber and Wilson 1986/1995) includes both participants in the exchange and the shared mutual knowledge that is manifest to them at the time of the utterance, which may include knowledge relative to the social or cultural group of which they are part. For example, in ordinary circumstances, we usually trust an unknown person who gives us directions to go somewhere in an unfamiliar city. This is a typical example in which the shared knowledge of both participants is minimal and so there are not, in principle, unusual conditions for not trusting the other person. However, if the directions are very difficult to follow or they direct you very far away from where you are, you may try to confirm the testimony given by this person because you may think that you were closer or that it is worth checking before going so far. Another example is someone who needs to take a train to go to an important meeting. In this case, she may want to double-check the information in the itinerary, which she would otherwise consider reliable because trains do not usually change their schedule. A witness in a law court cannot be trusted without questioning because many important decisions could be dependent on her testimony and because special norms of the activity apply to the case. Political leaders, chosen by citizens in their respective countries, are generally trusted, at least by those who vote for them, but continuous contradictions in their assertions and difficult circumstances can make everybody take a stance of distrust towards what they say: for example, pronouncements by many of the European leaders about the end of the economic crisis are believed by almost no one. A sentence is uttered in a definite context in which two actual actors are communicating. In principle, both speaker (or writer) and listener (or reader) can benefit from the exchange: the speaker’s expectations of an intended change in the listener’s attitudes or beliefs increase, and the listener benefits, under normal conditions, from the access to new and presumably reliable information without too much cost. Therefore, in normal circumstances, I think that it is epistemically reasonable to accept the speaker’s assertions as Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 182 presumptive truths. That is, trustworthiness can be a reasonable presumption in normal circumstances in any conversational exchange if no special circumstances or norms have to be taken into account. However, this presumption of truth can be canceled at any moment because speaker and listener are part of a conversational context that is continuously changing. In the course of the same conversation, we are continuously adding new information that comes to us through the interchange and, therefore, both participants have to dynamically adjust their epistemic standards to the conditions of the ongoing conversation (Origgi 2004). For example, through communication, new information is added to the information we shared beforehand with our interlocutor. This information includes all the beliefs in the shared cognitive environment in which we can situate the exchange, and makes it possible to contrast the relative coherence of the new information with that already stored. When adding new information to their belief systems, both participants have to constantly readjust those systems for coherence. An excess of non-demanded information may also put us on guard about the assertions that arise in the exchange. All kinds of conditions can influence the outcome of the communication and adaptation to those conditions has to be taken into account to explain whether and how we are justified in believing the testimony of a person. Thus, acceptance of testimony, in most communicative and social settings, should be considered only in terms of justified beliefs or presumptive truths and taken into account in the context in which a communicative act occurs. We know that those beliefs will, almost certainly, change or be modified when new information is added to our system. Vasallo (2006) presents several more examples that show how the context helps us to strengthen or lower the standards that we apply to an act of testimony. We may be justified in trusting a friend who tells us about the quality of a product because she has already used it, but if we know that she has begun working as a seller of the product, our epistemic standard will be adjusted accordingly because we may think that she is trying to sell it to us instead of guaranteeing its quality. We may trust the work of an historian before we learn that she has a strong ideological affiliation (or similarly distrust her for the same reason). We may be justified in accepting a doctor’s diagnosis if what she says does not have important consequences, but may ask for a second opinion if her diagnosis necessitates a major operation. Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 183 For Daukas (2006), epistemic authority is granted or withheld by virtue of whether or not we consider that an individual is part of a socially constructed group; that is, she maintains that epistemic trustworthiness is socially inculcated and that attitudes about an individual’s epistemic competence, in part, determine who is granted full membership of an epistemic community. For Daukas, epistemic trustworthiness is developed through time and practice and through interactions with others in the context of normative social practices (p. 14). Moreover, recent experiments in the field of cognitive psychology prove that a trustful (or not distrustful) attitude towards an interlocutor in a communicative exchange does not mean that we are ready to accept blindly everything that the other says because, in any communicative act, we have activated an attitude of epistemic vigilance (Sperber et al. 2010). For example, as noted above, we react critically if we consider that the source is not trustworthy enough or if the content of the utterance has important consequences for our system of beliefs. Nevertheless, Bondy (2010), Daukas (2006), and Sperber (2010) warn us of the danger of credulity when our interlocutor is close enough to the community or group to which we belong. The fact that I am considering here not only a communicative exchange but also a truly argumentative discourse in which testimony is used to support a claim can be of help to avoid this problem. A discussion begins because there is a disagreement about a point and, as a consequence, the mechanisms of vigilance should be more readily activated in argumentative exchanges than in other types of dialog. If we consider argumentation in the sense of an interchange of reasons about a disagreement in a social setting, the statements exposed by testimony have to refer first to a language structure having a particular form (how we assert, say, or utter things), but then they have to be considered also as reasons to support a claim. In this sense, they have to be assessed in the actual communicative and social context they are uttered as contributions to an ongoing discussion that takes place in this setting. Moreover, argumentation is public in the sense that many of our words can be and will be used, in some cases, to convey information in a future communicative exchange. The (communicative, dialectical, and social) context in which an utterance first appeared already sanctioned it; therefore, if it is relevant in a new context, it could be used to support new claims (Olmos 2007). This is the case, I believe, Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 184 for scholarly citations, which are analyzed in the remainder of this paper. The context of an utterance and the social relationships between the agents in the interchange help to define the “normal conditions” appealed to in the default rule. As a consequence, the application of this rule as a basis for justifying the acceptance of what is said by testimony depends on our adjustment to the epistemic standards of the communicational setting. As a result, considering the default rule (mainly stated by anti-reductionists) in this way, we are giving, at the same time, possible positive reasons to accept or not accept an instance of testimony (as reductionists ask for) because we situate communicative acts in their definite communicational context with its many conditions and constraints. In the text below, I try to apply the above considerations to the case of citations in academic papers in order to explain the use we make of them in practice and, if possible, to extract a possible norm of use that more closely resembles that practice. 4. Citations in argumentative practice Scholarly papers make use of instances of testimony as ci- tations to defend claims. They are, first of all, communicative acts that take place in a definite communicative and social context, namely, an academic paper, and, as a consequence, as in other contexts, communicational rules should be applicable to them: for any communicative act to be successful, we have to invest some effort, but we should expect some benefit in return (Sperber and Wilson 1986/1995).   Through communication, we communicate and look for relevant information given the context in which the act takes place. Willard (1990) remarks that most scholarly papers are seldom read and very few of them are quoted. Here too, we do not have the means to read and even less to check all the possible positions and information that can be found in the literature in our field6. To put it in Willard’s words, there is a problem of “literary management”: 6 The Philosopher’s Index Cumulative Annual Edition #45 (2011) has a total of over 490,000 journal article and book citations covering over 1400 journals from 85 countries in 39 languages: http://philindex.org/. Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 185 Exponentially expanding literatures proceeding along multiple paths make for a frustrating indefiniteness. One never knows whether one has fully heard a position out, or seen it in its best form, for there’s always more to read (Willard, p. 15). Therefore, when a source is cited, in many cases, we apply the default rule, that is, in normal circumstances, we feel justified to rely on the citation for many reasons that have to do with the communicative and social setting in which the citation is used: the explanation given by the quotation is compatible with our beliefs; we agree with the claim or the position taken in the paper; we think that the fact that the work has been published automatically gives the author the status of being trustworthy, reliable, or at least worth considering; we trust the writer regarding the reliability of the quote; and so on. Some of these considerations are gathered in the list of critical questions of Walton (1997), while others are not, for example, the issue of compatibility with our previous beliefs. Moreover, I consider that trustworthiness is the main mechanism to accept a quote within the scholarly context and I link it with the fact that the quote was previously sanctioned in an academic paper and that both writer and reader are part of this context. Furthermore, many of our assertions in scholarly papers are not merely expressions of our thoughts but reasons to support claims; as a consequence, besides the problem of true beliefs or of reasonably justified beliefs, we have to consider their use to support claims and how they are going to be handled by a reader or the audience of the argumentative exchange. When the reader or audience is opposed to the main claim in a paper or other work, her objections will rarely be about the falsity or unacceptability of the citations. If an exchange on this disagreement progresses, in most cases, she will not even try to prove that citations, considered as reasons to support the claim, are false or inaccurate, but would rather try to rebut the claim and look for alternative citations or ways to support her own claim. That is, a critical reader may react to the claim, but not necessarily because she does not concede what is said by the citation. It could well be that citations are not exact or do not accord with the original intended idea of the quoted, as Walton and Macagno (2011) show via numerous examples. As they illustrate, misquotation can occur for many reasons: it can be caused by the ambiguity of natural language, or because our interpretation of the quotation is loose or even Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 186 distorted and does not correspond to the intended meaning of the proponent, or even because we wrench it from the context, among other reasons. Nevertheless, quotations are mostly considered reliable (especially in academic papers) and are seldom checked for truth given our adjustment to the institutional context in which citations are used. Exceptions could occur in those cases in which the content of the quotation is close to our positions and of real interest to us or in the case in which the quoted author reacts to the misquotation, the case that is mostly considered by Walton and Macagno. As already stated, some authors (Origgi 2004; Hardin 2002; Daukas 2006) define the acceptance of testimony in terms of social relations as trust. We trust citations because we are part of the scholarly context in which they are used and to which all of us are supposed to contribute in the search for a better explanation of a problem. We trust citations because the academic context in which academic papers appear already sanctioned them. If we consider, as Rouse (2007) does, that normativity in practice involves a complex pattern of interrelations among performances through time (p. 8), the fact that the academic context had already sanctioned an academic work should be included as a criterion for evaluating the acceptability of a posterior citation of this work. Nevertheless, as stated above, practices are complex and, as a consequence, there are also other social factors we keep in mind when citing or accepting citations. These factors can be related to one’s own interests and are also subjected, in part, to social contextual constraints of the academic context. For example, we can consider the actual constraints that publishers place on the works they accept for publication based on the indexes against which the journal will be evaluated, the success of the author, and the number of publications she has or the prevalence with which her ideas have been discussed over time. If we think of disciplines in social terms, we, as parts of these social institutions, know about the requirements of each of them and adapt our needs to their standard practice. As a consequence, a careful choice of quotations can include more than showing competence or good support for an argument. It can be explained as a case of adjustment to the social context or, in other terms, as a rhetorical maneuver that looks towards a better performance in this context, for example, to arouse the interest of the more recognized members of the epistemic community, to increase our likelihood of being published, to Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 187 validate our not always original thoughts, and to fill out the required pages. Origgi (2004, p. 42) states that “there are no purely unbiased informants” and that “there are no naïve receivers of information”. The way in which we adjust our needs to accept or reject the authority of a (quoted) writer is dynamic and context-dependent; this context depends on our interests, or on how we define the disciplinary field. An author’s bias can be easily seen by observing the list of different references she uses. Even when writing about the same subject, these references differ because the amount of actual information available about any subject is enormous and spreads into different fields. We all make our own choice of authors to quote, and this selection is due to not only epistemological reasons to strengthen our claim, but is also influenced by practical factors related to the particular epistemic community of which we are a part. For example, if we consider the proceedings of any of the conferences of the International Society for the Study of Argumentation (ISSA)7, even when the authors are dealing with the same subject, by only glimpsing at the references in the papers, we can categorize them according to a narrower field, be it from the pragma-dialectical current, according to the ideas of informal logic, coming from the historical tradition of rhetoric, and so on. It is also possible to notice differences according to the countries of origin of the authors: authors from non-English-speaking countries cite and refer more frequently to authors from their own countries than those from English-speaking countries. Sociological studies on references or on citations in papers should be carried out, but we think that the results of those empirical studies would favor the thesis that testimony should be studied in a socially situated environment. It is necessary to integrate frameworks that focus on the communicative properties of the statements used in the acts of arguing and the consideration of them as part of social normative practices with the more philosophically oriented dialectical approaches to argumentation. These interdisciplinary frameworks, such as the “normative pragmatics” proposed by Jacobs (2000), emphasize the adaptation and evaluation of argumentation in the context (both dialectical and social) in which it was used and should 7 The choice of this conference is due to the huge number of papers presented there. Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 188 inspire the design of protocols and other tools for the different tasks involved in the practice of arguing. 5. Conclusions In this paper, I have considered the case of citations in academic work and tried to show that some of the questions linked to the “appeal to an expert opinion” argumentative scheme cannot be applied in cases of citations in philosophical academic papers. I have also tried to show that more elements are involved in accepting or rejecting an actual quotation. I have suggested that, to improve the analysis and the use of quotations, it would be better to consider them as part of the practice of arguing in academic contexts. In academic discussions in philosophy, both participants in a dialogue make their choices regarding citations, but collaborate by giving reasons or confronting the other participant with doubts, questions, or counterarguments. From this perspective, citations and quotes are part of this game in which social elements are included. Moreover, in philosophy, there is almost always a multiplicity of positions about an issue, so real arguments are better considered as part of an ongoing dialog improved by constant interaction between two or more participants by means of a collaborative practice that aims to advance towards a consensus or to the improvement of our knowledge about different issues. I have also analyzed citations as instances of testimonial practices, as is carried out in epistemology; by doing so, I have tried to show that the particular (institutional) context of which participants in the exchange are a part has to be taken into account to explain why and when it is reasonable to accept some beliefs on the word of others. Testimony as a communicative practice has to follow general communicative rules, which require interpreting instances of testimony in the particular context in which they were uttered, in order to look for new available information compatible with our beliefs. Limited resources and trust, extended from the institutional context, can in normal circumstances justify the acceptance of citations in academic work. In many cases, if a claim in a paper is not accepted, instead of trying to prove the falsity of the quotations supporting it, we engage in an ongoing process in which each participant puts forward new plausible reasons (or citations) to defend an alternative point of view. The choice of citations by the author depends on the audience and includes Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 189 different ways of accommodating rhetorical aspects of the context that are always present in argumentative exchanges. Acknowledgements This work was supported by the Research Project FFI 2010- 20118 of the Ministry of Economy and Competitiveness of the Spanish Government. References Adler, J. (2012). “Epistemological Problems of Testimony.” In: E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Spring 2013 Edition), http://plato.stanford.edu/archives/spr2013/entries/testimony -episprob/ Bondy, P. (2010). “Argumentative Injustice.” Informal Logic Vol. 30 (3): 263-278. Brandon, R. (1994). Making it Explicit: Reasoning, Representing and Discursive Commitment. Cambridge, London: Harvard University Press. Coady, C.A.J. (1992). Testimony: A Philosophical Study. Oxford: Clarendon Press. Collins, H. and Weinel M. (2011). “Transmuted Expertise: How Technical Non-Experts Can Assess Experts and Expertise.” Argumentation 25 (3): 401-413. Daukas, N. (2006). “Epistemic Trust and Social Location.” Episteme: A Journal of Social Epistemology 3 (1): 109-124. Eemeren, F.H. van and Houtlosser, P. (2005). “Theoretical Construction and Argumentative Reality: an Analytic Model of Critical Discussion and Conventionalised Types of Argumentative Activity.” In: D. Hitchcock and D. Farr (Eds.), The Uses of Argument. Proceedings of a Conference at McMaster University 18–21 May 2005 (pp. 75-84), Hamilton: Ontario Society for the Study of Argumentation. Gelfert, A. (2011). “Expertise, Argumentation and the End of Inquiry.” Argumentation 25 (3): 297-312. Goldman, A. (2001). “Experts: Which Ones Should We Trust?” Philosophy and Phenomenological Research 63, 85-110. Goodwin, J. (2011). “Accounting for the Appeal to the Authority of Experts.” Argumentation 25 (3): 285-296. Grice, H. P. (1975). “Logic and Conversation”. In: P. Cole and J.L Morgan (Eds.) Syntax and Semantics: Speech acts. Vol. 3 (pp. 41–58), New York: Academic Press. Begoña Carrascal © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 190 Hardin, R. (2002). Trust and Trustworthiness. New York: Russell Sage Foundation. Hastings, A. C. (1963). A Reformulation of the Modes of Reasoning in Argumentation. PhD Dissertation, Evanston, IL. Jacobs, S. (2000). “Rhetoric and Dialectic from the Standpoint of Normative Pragmatics.” Argumentation 14, 261-286. Johnson, R. H. (2005). Theory and Practice again: Challenges from Pinto and Toulmin. In: D. Hitchcock and D. Farr (Eds.), The uses of argument: Proceedings of a conference at McMaster University 18–21 May 2005 (pp. 222-231), Hamilton: Ontario Society for the Study of Argumentation. Kauffeld, F.J. and Fields J.E. (2003). “The Presumption of Ver-acity in Testimony and Gossip.” Proceedings of the 2003 IL@25 Conference. http://web2.uwindsor.ca/faculty/arts/philosophy/ILat25/pap ers.htm. Kvernbekk, T. (2012). “Argumentation in Theory and Practice: Gap or Equilibrium?” Informal Logic, 32 (3): 288-305. Lackey, J. (2006). “Introduction.” In J. Lackey and E. Sosa (Eds.), The Epistemology of Testimony (pp. 1-21), Oxford: Oxford University Press. Lackey, J. (2010). Learning from Words: Testimony as a Source of Knowledge. Oxford: Oxford University Press. Lackey, J. and Sosa, E. (2006). The Epistemology of Testimony. Oxford: Oxford University Press. Mercier, H. and Sperber, D. (2011). “Why do Humans Reason? Arguments for an Argumentative Theory.” Behavioral and Brain Sciences, 34 (2): 57-74. Olmos, P. (2007). “Making it Public: Testimony and Socially Sanctioned Common Grounds.” Informal Logic Vol. 27 (2): 211-227. Origgi, G. (2004). “Trust, Authority and Epistemic Responsibility.” Theoria 61, 35-44. Pinto, R.C. (2001). Argument, Inference and Dialectic. Dordrecht: Kluwer Academic Publishers. Rouse, J. (2007). “Social Practices and Normativity.” Philosophy of social Sciences 37 (1): 1-11. Ruggiero, V. (2003). Beyond Feelings: A Guide to Critical Thinking. New York: McGraw-Hill Sperber, D. (2001). “An Evolutionary Perspective on Testimony and Argumentation.” Philosophical Topics 29, 401-413. Authority of Citations and Quotations in Academic Papers © Begoña Carrascal. Informal Logic, Vo. 34, No. 2 (2014), pp. 167-191. 191 Sperber, D. (2010). “The Guru Effect.” Review of Philosophy and Psychology 1 (4): 583-592. Online at: http://www.dan.sperber.fr/ Sperber, D. and Wilson D. (1986/1995). Relevance: Communication and Cognition. 2nd edition. Oxford: Blackwell. Sperber, D., Clément F., Heintz Ch., Mascaro O., Mercier H., Origgi G. and Wilson D. (2010). “Epistemic Vigilance.” Mind & Language Vol. 25 (4): 359-393. Steup, M. (2010). “Epistemology.” In: E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Spring 2010 Edition), http://plato.stanford.edu/archives/spr2010/entries/epistemol ogy/ Tindale, Ch. (2004). Rhetorical Argumentation: Principles of Theory and Practice. Cornell, NY: Sage Publications. Tindale, Ch. (2007). Fallacies and Argument Appraisal. Cambridge: Cambridge University Press. Tindale, Ch.W. (2011). “Character and Knowledge: Learning from the Speech of Experts.” Argumentation 25 (3): 341- 353. Tymoczko, Th. (1986). “Making Room for Mathematicians in the Philosophy of Mathematics.” The Mathematical Intelligencer 8 (3): 44-50. Vasallo, N. (2006). “Testimonial Justification and Contexts.” Epistemologia XXIX, 127-144. Walton, D. (1996). Argumentation Schemes for Presumptive Reasoning. Mahwah, NJ: Lawrence Erlbaum. Walton, D. (1997). Appeal to Expert Opinion: Arguments from Authority. Pennsylvania: Pennsylvania State University Park. Walton, D. (2005). “Justification of Argumentation Schemes.” Australasian Journal of Logic (3): 1-13. Walton, D., Reed Ch. and Macagno F. (2008). Argumentation Schemes. Cambridge: Cambridge University Press. Walton, D. and Macagno F. (2011). “Quotations and Presumptions: Dialogical Effects of Misquotations.” Informal Logic Vol. 31 (1): 27-55. Willard, C.A. (1990). “Authority.” Informal Logic 12, 11-22. Zenker, F. (2011). “Experts and Bias: When is the Interest- Based Objection to Expert Argumentation Sound?” Argumentation 25 (3): 355-370.