Letter to the Editors

To the Editors ‘Reward or persuasion? The battle to define the meaning of a citation’, by Philip M. Davis,1 left me neither rewarded nor persuaded. Not rewarded because he failed to use my own article ‘Reward, persuasion, and the Sokal hoax: a study in citation identities’.2 Not persuaded because his discussion of the ‘reward vs. persuasion battle’ involves too many shallow arguments. He does cite many of the right things, but my article should have been among them; it presents arguments highly relevant to his and might even have helped him avoid some misunderstandings. It appeared in the issue of Scientometrics devoted to Robert K. Merton’s work, and addresses both the Mertonian ‘reward’ and the social constructivist ‘persuasion’ positions with empirical data and in considerable detail. It critiques many of the same studies he cites. Moreover, it is easy to find: if one enters ‘reward’ and ‘persuasion’ into Google, it is the first thing one sees, with his article now coappearing. It might also have led him to another article of mine that also discusses ‘the battle’.3 The two positions mentioned above involve conflicting beliefs about what motivates authors to cite: are they rewarding citees by acknowledging use of their work, or are they trying to persuade readers of their arguments with citations as part of their rhetoric? Actually, there is no doubt whatever that authors write to persuade (how could they not?), and deploy citations in making their case. This is a truism that studies by Susan Cozzens and Terrence Brooks, for example, simply reinforce. But Davis muddles the matter by not distinguishing unexceptionable attempts to persuade from what I called in my article ‘dark persuasion’, which is identified with constructivists such as G. Nigel Gilbert, Bruno Latour, Stephen Woolgar, and Michael MacRoberts. They claim that authors, like advertisers, routinely use citations to manipulate readers, for example, by overciting ‘big names’ so as to borrow luster for their own works, by tendentiously misrepresenting the works they cite, and by citing acquaintances or the powersthat-be in hope of reciprocal favors. The latter would be for Merton a kind of particularism; that is, basing citations on particular personal characteristics of the citees, such as their status, fame, or social ties with the citer. The Mertonian reward model, in contrast, assumes his norm of universalism; that is, acknowledging the relevant precedent work of citees regardless of their personal characteristics. The real conflict in explanatory models is thus between reward and ‘dark persuasion’. However, only by conflating all kinds of attempted persuasion, as Davis does, can the conflict between models be made to seem unresolved. As soon as one examines the case for dark persuasion, it falls apart. Its basic claim is that the entire system of citation is (perhaps amusingly) corrupt. The constructivists take this position without producing even anecdotal evidence for it. They can do this because they write essentially as satirists, relying on every academic’s stock of cynicism for assent. But they give no hard data; readers must supply their own examples of morally flawed citation practices. These are less readily found than one might think, and generally turn out to be anecdotes that in no way prove widespread corruption. Nor can one adduce as evidence studies of inadvertently flawed citation practices, such as that by Wright and Armstrong:4 ‘Faulty citations include omissions of relevant papers, incorrect references, and quotation errors that misreport findings.’ These are not provable attempts at dark persuasion; they are simply failings that must be set against millions of citations that can withstand criticism. There is also the problem of reflexivity: readers who are also citers would be remiss if they did not apply standards of manipulative intent to themselves as well as others. Those who believe that attempts at dark persuasion are everywhere in science and scholarship must show that they pervade the work of critics such as Latour, MacRoberts, Woolgar, and Gilbert, and that they are common in their own citations as well. Happily, one need not worry about these matters, because the case for dark persuasion undermines itself. Even if authors attempted it in great numbers, the data would look just like those resulting from efforts at Mertonian reward. That is because Letter to the Editors 253