A special guest post co-authored by Meena Krishnamurthy (Michigan) and Jessica Wilson (Toronto).
An Argument for Improved Citation Practices in Philosophy*
Here are two claims concerning scholarship practices in philosophy:
- In academic philosophical contexts, we ought to cite work that is clearly relevant to the topic at hand.
- In academic philosophical contexts, we ought to engage with work that is clearly relevant to the topic at hand.
In this post, we provide moral and philosophical reasons for thinking that claim 1 is true. Since claim 1 is widely violated, members of the profession should aim to improve citation practices in philosophy; we offer some potential ways of achieving this end, and address some potential barriers to doing so. We also think that there are moral and philosophical reasons for thinking that claim 2 is true; however, since claim 2 imposes a greater burden, we start with the less onerous, but still widely violated, claim 1. We focus on academic philosophical contexts (henceforth, just ‘philosophical contexts’, or ‘philosophy’) to fix ideas, because we are most familiar with these contexts, and because it is our impression that scholarship practices in philosophy are lax as compared to those in many other disciplines (e.g., linguistics, history, and the sciences).
In what follows, we assume that the works at issue have all been published in peer-reviewed venues and are readily available to members of the profession.
1. Three varieties of citation failure
We start by observing three sorts of (sometimes overlapping) cases in which citation failure manifests in philosophy.
First are cases where an individual (or individuals) fail to be properly credited for a specific argument, approach, or account, as when, e.g., philosopher X has a claim equal to or better than that had by philosopher Y to be considered an originator of account A, but A is typically attributed only to Y.
Second are cases where the presentation of a given dialectic ignores a large existing literature, as when, e.g., philosopher Z claims that approach A has been ignored or rejected in past decades, even though dozens of philosophers have been working on and developing A in past decades.
Third are cases where a philosophical wheel is reinvented, as when, e.g., philosopher Z proposes ‘new’ account A as the best way to accommodate phenomenon P, even though most or all existing accounts of P take A as their starting point.
2. Moral and philosophical problems with citation failure
There are at least two broad categories of reasons to think that citation failure is problematic—problematic enough that, it seems clear, members of the profession should individually and jointly strive to improve citation practices in philosophy.
2.1 Broadly moral problems with citation failure
Two broadly moral concerns are salient. First, citations of an individual’s work have concrete positive bearing on a variety of professional outcomes, and failures of an individual’s work to be cited, even when it has been published in a peer-reviewed journal or other venue, have a concrete negative bearing on these outcomes. As such, failure to cite work that is clearly relevant to the topic at hand is concretely injurious to the individual authors of that work.
Second, citation failure is most clearly a problem for members of groups that are subject to implicit bias, including women, persons of color, those at non-elite (especially ‘foreign’ non-elite) institutions (think ‘prestige’ bias), and/or members of other disadvantaged categories. The standard biases are moreover exacerbated in cases where the disadvantaged individual is comparatively junior or is not a member of the ‘in-crowd’ working on a given topic (membership in which strongly tracks advantageous demographic categories) whose work must be cited by anyone working on the topic at hand. Relatedly, each of the varieties of citation failure mentioned above is commonly associated with and exacerbated by what is reasonably seen as implicit bias. For example, in cases where philosopher X has a claim equal to or better than that had by philosopher Y to be considered an originator of account A, but A is typically attributed only to Y, it is common that X is bias-disadvantaged as compared to Y. And it is common that those misrepresenting a given dialectic are members of advantaged categories or associated ‘in-crowds’, who appear to read and cite mainly each other. Bias (especially prestige bias) is also operative in allowing inaccurate presentations to be published and encouraging them to be propagated; it is an underappreciated fact that implicit bias not only unfairly disadvantages members of disadvantaged categories, but also unfairly advantages members of advantaged categories. Citation failure is both a manifestation of and perpetuates implicit bias, and so is concretely injurious to those subject to such bias, both as individuals and as members of disadvantaged groups.
2.2 Broadly philosophical reasons
Two broadly philosophical concerns are salient. First, if a given author fails to cite work that is clearly relevant to the topic at hand, it is likely that said author has not read this clearly relevant work. This can lead to any number of philosophical wrong turns, even putting aside whether one should (as per claim 2 at the head of this post) engage with any work that is clearly relevant to the topic at hand. For example, suppose that philosopher Z’s topic is account A, but that Z only reads and cites Y’s presentation and defence of A, notwithstanding that X has extensively developed their own version of A. Z may conclude that A should be rejected on grounds of facing a significant objection, even though X’s version has the means of responding to that objection. For another example, failure to be familiar with work relevant to the topic at hand may lead to enthymematic argumentation (as with an argument-by-cases that neglects an existing case).
Second, failure to properly cite individuals or groups of individuals whose work is relevant to the topic at hand can lead, as above, to distorted presentation of a given dialectic and/or to the reinventing of existing wheels. In turn, these dialectical failures—especially when disseminated by influential individuals—can lead to the generation of literatures founded on false presuppositions, and the propagation of ignorance by those (in particular, graduate students, who may falsely assume that they can rely on dialectical claims made by those in positions of comparative prominence) working within the parameters of the ill-founded debates.
3. The way forward: improving citation practices in philosophy
The moral and philosophical problems with failing to cite work that is clearly relevant to the topic at hand indicate that claim 1 is true: in academic philosophical contexts, we ought to cite work that is clearly relevant to the topic at hand. But claim 1 is regularly violated, as our experience and attention to citation data indicate (see in particular this post by Kieran Healy, in which he argues that few articles by women end up being frequently cited). We conclude that practicing philosophers should commit to improving citation practices in our field—first, by each of us aiming, in our own work, to cite any work that is clearly relevant to the topic at hand, and second, by each of us taking steps to ensure (qua referee, for example) that others cite any work that is clearly relevant to the topic at hand. In what follows we expand upon these suggestions, considering possible barriers to carrying them out along the way.
First, however, we pause to consider an objection, according to which notwithstanding the canvassed moral and philosophical considerations, we don’t have an obligation to cite all the relevant literature, since some of it might not be of sufficiently good quality; relatedly, we have heard some say that they only feel obliged to read and cite papers that appear in ‘top’ journals (more generally, presses) in the area at issue. Here we stand firm, maintaining that any clearly relevant work that has been published in a peer-reviewed press should be cited. As with so much else, evaluations of quality are subject to implicit biases that, both on the path to publication and afterwards, unfairly operate against individuals in disadvantaged groups and unfairly operate in favor of individuals in advantaged groups. If an author doesn’t like a paper, then they can downplay it or, better yet, briefly say why they don’t like it.
3.1 Improving citation practices: citing the clearly relevant work of others
Ultimately, improving citation practices in philosophy requires that individual philosophers do basic scholarly due diligence in order to identify work that is clearly relevant to the topic at hand. Luckily, there are a number of readily available tools which greatly lessen the burden of engaging in literature surveys. To start, there are a number of sophisticated search engines, including PhilPapers, Philosophers’ Index, and Google Scholar: for each of these, one can input the topic at hand (often under different sorting mechanisms—e.g., by date, if one wants to check for recent contributions) and consider the results. (One can also sign up for PhilPapers updates in one’s areas of interest in order to be kept up-to-date on new contributions.) The bibliographies of papers that come up in such searches are rich sources of information about existing work on the topic at hand, as are encyclopedias (e.g., the Stanford Online Encyclopedia of Philosophy) and ‘state of the art’ articles (as in, e.g., Philosophy Compass). It is increasingly common for philosophers to post requests for pointers to relevant literature on Facebook and Twitter. And finally, philosophers can continue doing what they have always done: talk with people and ask for suggestions.
One might object that there is no precise standard of ‘clearly relevant’, and relatedly, that search engines may turn up irrelevant work, or fail to turn up relevant work. To this objection we respond that of course, search engines, encyclopedia entries, and the like are indicators, not infallible algorithms; as per usual, one has to bring one’s brain to the procedure’s application. One has to input a topic that is narrower than, e.g., ‘Properties’ or ‘Quantum Mechanics’. And even for topics that are narrowly defined, not all listings may in fact be clearly relevant to the topic at hand.
That said, when using one’s brain, assisted by search engines, to figure out which texts are clearly relevant to the topic at hand, one needs to err on the side of inclusion. This is especially important given that citation-based searches will incorporate existing citation neglect of women and/or other disadvantaged demographics. Taking the fact of longstanding bias into account means that proper due diligence, citation-wise, does not consist in citing only works that are published in the very top journals, or which are written by prominent members of the elite ‘in-group’, or which—reflecting these dimensions of citation valence—show up at the top of ‘impact’ or ‘number of citations’ searches. Relatedly, proper due diligence doesn’t consist in an author’s antecedently judging which works on the topic at hand to read and cite, since such judgments are subject to significant skew by implicit bias. Indeed, given the impact of implicit bias on who and what gets cited, many papers that are not just clearly relevant but objectively philosophically important to the topic at hand may not come up near the top of searches—and relatedly, given that implicit bias also unfairly advantages members of advantaged groups, many less-than-stellar papers on a given topic may come up right at the top. When doing literature surveys we need to ‘go deep’ down the list if we aim to identify all work that is clearly relevant to the topic at hand.
That we typically need to go deep in doing literature surveys gives rise to another concern—namely, that reading all of the work that is clearly relevant to one’s topic will be overly onerous. Here we have three responses. First, in many cases involving an appropriately fine-grained topic-at-hand, there is no reason to think that doing a thorough literature survey will be overly onerous. In many other disciplines (e.g., linguistics, history, physics and the other sciences), practitioners are expected to carry out, and do carry out, full literature reviews and to cite any clearly relevant literature accordingly. Philosophers do not appear to face any special burdens so far as literature that must be surveyed. If practitioners of other disciplines can carry out such surveys, then so can we. (Indeed, philosophers working in the history of philosophy are notably more conscientious about citation practices; if they can do it, so can other philosophers.) Second, in some cases doing a full literature survey may indeed involve a significant commitment of time. But this front-end commitment is small—less onerous, overall—when compared to the downstream waste of time, trees, and cognitive resources associated with problems (failure to register existing responses to objections, inaccurate presentation of dialectics, reinvention of wheels, etc.) commonly generated by failures to read and cite relevant literature. Third, in any case, part of what we are pushing for is a reconception of how much time philosophers devote to scholarship. Yes, given current practices whereby many do not engage in anything like full literature surveys, implementing our suggestion is going to involve some rearrangement in one’s work habits, whereby sufficient time is scheduled for and devoted to engaging in finding out the actual dialectical state of play as regards the topic at hand. The resulting paper and associated dialectical context will be better for it, however—as will our profession and its practitioners.
3.2 Improving citation practices: working to ensure that others cite clearly relevant work
Unless large numbers of philosophers commit to citing the clearly relevant work of others, citation practices in philosophy as a whole will continue to be problematic. Here we offer two ways in which individuals can act so as to ensure that the clearly relevant work of themselves and of others is cited.
Strategy 1. The first strategy concerns the actions of referees. We suggest that the quickest way to improve citation practices in philosophy as a whole would be for large numbers of individuals to commit, in their capacity as journal or other referees, to rejecting for publication papers or other submissions that fail to cite work that is clearly relevant to the topic at hand, on grounds of failing to meet basic standards of scholarship. The suggestion here is that it be generally considered a necessary condition on a submission’s getting a full review that the author engages in basic scholarly due diligence. The next quickest way to improve citation practices would be for large numbers of individuals to commit, in their capacity as referees, to not accepting any submissions that fail to cite work that is clearly relevant to the topic at hand. The suggestion here is that it be generally considered a necessary condition on a submission’s getting published that the author engages in basic scholarly due diligence.
Whether rejection or simply non-acceptance is at issue: if sufficient numbers of philosophers were to commit to the practice of rejecting/not accepting papers on these grounds, authors would, it may be reasonably anticipated, immediately work much harder at searching for and citing clearly relevant work sending their own work out.
One might object to Strategy 1 that papers that don’t do basic due diligence can still make good contributions. After all, some have said, many historically foundational philosophical texts hardly cite anyone at all—think of Descartes’ Meditations, or more recently, much of Quine’s work. As such, we shouldn’t be so draconian as to require that authors cite clearly relevant work in order to get published. To this call for ‘exceptionalism’ we have two responses. First, the contributions of authors who cogitate ex nihilo will be even better once they fulfill basic scholarly standards by reading and citing clearly relevant work. Second, the proposed brand of exceptionalism is just an invitation for implicit bias to enter in, and for those who falsely think they are introducing a dialectical shift or inventing a new wheel to justify their ignorance or ignoring of existing literature as in some grand historical tradition, to the disbenefit of the un-cited. In today’s professionalized climate, where citations are hugely important along every relevant dimension of career success, a fond remembrance of the citation-free historical figures of the past is not to the proper point.
Strategy 2. The second strategy is for individuals to be proactive on their own and others’ behalf, by contacting authors of papers, encyclopedia articles, etc., to let them know of citation failures. One convenient way to do this is to sign up to receive PhilPapers updates on one’s areas of interest in order to get a heads-up about recent work. If a piece of work shows up that is lacking appropriate citations, to oneself or to others with whom one is familiar, then one can take a few minutes to contact the author and nicely inform them of this. Often the work will be in draft form, in which case the author will have time to correct the citation failure; but even if the work is already published, it is worth contacting the author to request that the citation be included in future publications on the topic. In some cases (especially where priority for a given account is at issue) it might even be worth contacting the editor of the relevant publication to see if the citation failure can be corrected in an erratum. Similarly, one can read encyclopedia articles with an eye to seeing if clearly relevant literature has been cited, and if not, contacting the author of the entry. Engaging in these sorts of communications can be a bit of a pain, but over time can be quite efficacious.
One concern about Strategy 2 is that junior persons, in particular, may not feel comfortable contacting prominent persons in their area about failures of the senior person to cite the junior person’s clearly relevant work. In some cases, it may be possible for the information to be communicated by a friendly intermediary (or an editor, say); but in general our experience in implementing this strategy is that in nearly all cases authors are gracious and grateful when citation failures in their work are called to friendly attention. Here it can be useful to communicate, among other soothing facts, that citation failures are frequently a function of implicit bias, to which we are all subject. And the more people engage in the strategy, the less taxing it will be to contact others or be contacted in relation to citation failures, and the sooner people will work in ways ensuring that they don’t need to be so contacted.
5. Concluding remarks
We have here provided moral and philosophical reasons for thinking that academic philosophers ought to cite work that is clearly relevant to the topic at hand. We have also provided some practical suggestions about how members of the profession can better do this. We hope our remarks will not only stimulate further discussion, but also, and more importantly, inspire others to take action to bring about the desired change. To this end, it would be most effective if people committed publicly to aiming to implement one or more of the strategies that we have outlined. More minimally, we hope that individuals will in fact aim to implement some or all of these strategies, and encourage others to also do so—e.g., by sharing this post.
*For additional discussion of related issues, please see Marcus Arvan, “Philosophers Don’t Read and Cite Enough,” at Daily Nous:
Thanks for this post. Another welcome side-effect of your proposed change in citation practices would be that a paper’s cross-niche appeal would become the main driver of high citation counts — whereas at the moment one can get lots of citations just by belonging to the cited tip of the scholarly iceberg.
LikeLiked by 2 people
Here are two extra, purely self-interested reasons to cite all the relevant work: (1) Many of the extra philosophers you cite will get notified about the citation through things like Research Gate or Google Scholar citation alerts (and then might go on to read and engage with your work). (2) Many folks these days track down literature by following the citation chain from older articles on the topic they know of forward in time using things like Google Scholar, and the more webs your paper is attached to, the more likely it will be found. So being a good scholarly citizen can also increase the visibility and reach of your own work.
LikeLiked by 2 people
Thank you both for this article. As a librarian you can appreciate the fact that when I am questioned by students “what is the ‘citation format’ to use in philosophy papers,” I am often at a loss. Some professors are very clear on documenting sources, both in-text and references; but, it has been my experience that many philosophers are very vague about what standard to use. “As long as you are consistent” is not helpful to students who are not familiar with Chicago, MLA, APA, or any citation format.
Secondly, you cite excellent resources for doing research. One that you should mention for those in the dissertation stage is checking the dissertation databases (Proquest, Worldcat, Open Access, etc.). PhilPapers may have recent ones, (I am not sure about this), but as we all know subjects go in cycles. It is always worth a look digging away in dissertations. I did help one student search DAI (Dissertations Abstracts International) several years ago. Lo and behold, his topic had more than one dissertation written – much to his surprise.
LikeLiked by 1 person
I’m grateful to Professors Krishnamurthi and Wilson for their provocative post. I have three questions about their proposal, however.
First, how widespread is the phenomenon of citation malpractice actually? If the problem is widespread, then perhaps journal editors ought to take a more proactive role in enforcing these “best practices”. However, the post doesn’t accuse any specific individuals, or specific papers of malpractice (presumably because blaming individuals for participating in a collective action problem is not only unfair, but beside the point). Thinking of recent literature I’ve read, I can think of a few examples of citation malpractice, but not many. And so I’m not moved to think of implementing these best practices as a particularly urgent problem. But that’s just anecdotal. I wonder if Profs. K&W have more systematic data to share with us?
Second, can we formulate meaningful criteria for ‘relevance’? K&W acknowledge in the post ‘relevance’ is a slippery term, but then point out that people seem to figure out what is relevant in the sciences, without too much problem. But is that really an apples to apples comparison? It looks to me notion of relevance is well-defined in science, and poorly defined for philosophy. If I were writing a chemistry paper on some new reaction I discovered, it would be obvious that I should cite: each of the papers in which the methods I’m using were first described, each of the papers on similar reactions, any papers which predict my new reaction and its properties, and so on. However matters are less clear in philosophy. Suppose I think up a new objection to Frankfurt-style compatiblism. Off the bat, it’s clear should cite Frankfurt since I’m criticizing him (one paper); and any of the other incompatibilist arguments against Frankfurt to act as the contrast class to my argument (say there are five such papers). But after that it gets really murky. Do I need to cite each of the articles that proposes an objection to each of those other incompatibilist arguments (say there are five objections to each of the five arguments, that’s 25 new citations)? Those objections could be quite beside the point–after all, I’m proposing a NEW argument against Frankfurt. On the other hand, maybe they are relevant, since perhaps my incompatibilist argument is better than others precisely because there aren’t these objections to it. But then do I need to cite all the papers in which each of the proponents of each of the incompatibilist arguments have attempted to defend those arguments against those objections (say there are five incompatibilist responses for each of the 25 objections, that’s 125 new cites)? At this level of detail, I would have already cited 1 + 5 + 25 + 125 = 156 papers and that’s assuming that my argument for my new incompatibilist aren’t isn’t drawing on some other standard tool or trick in the literature (which would engender it’s own cascade of citations). Clearly this is unmanageable. It seems we need some standard of relevance to help us avoid these kind of cascades.
Third, why isn’t there any mention above of citing philosophical research that isn’t published in English? Perhaps K&W think that they already imply that relevant work not published in English should be cited when they say that “every clearly relevant piece published in a peer-reviewed journal” should be cited. However, it seems appropriate to make this explicit. All of the same considerations that K&W advance in favor of more extensive literature reviews would apply inter alia to extending that literature review into a variety of different languages. Historians of philosophy are already expected to cite relevant literature (at a minimum) in French and German. Who’s to say that something quite obviously relevant to my new argument against Frankfurt hasn’t been published in the Zeitschrift für philosophische Forschung, or the Revue de métaphysique et de morale? Most leading analytic philosophy departments don’t require any research languages at all, and I can’t recall the last time I read an “analytic” article and saw a citation of a non-English source, even when the author is herself not a native English speaker. Perhaps K&W should amend their best practices suggestion to include a call for the return of research languages. After all, Anglo-centrism does in fact disadvantage some people with respect to others. However, notice that the traditional idea of a research language is alsoeuro-centric. Perhaps K&W would acknowledge that my new paper on Frankfurt should cite relevant work from the Revue or the Zeitschrift, but should I be required to check the Tijdschrift voor Philosophie, or the Hebrew articles in Iyyun? What about articles written in Hindi, or Mandarin? Again, obviously we need some criteria of reasonableness, since learning a research language is obviously a huge time commitment and it’s simply not plausible to think that any practicing philosopher is going to be able to maintain an active command of more than three or four research languages.
LikeLiked by 2 people
Thanks for this article.
I think up and coming philosophers in particular are in a difficult spot. Ensuring that your research is sufficiently thorough so that you can cite all the relevant literature is going to be time consuming. For a philosopher with an established career, taking some extra time to finish an article has no great cost. But for someone early in his or her career, say, someone about to enter the job market, or someone going up for tenure, the cost of not getting papers published (or at least accepted for publication) within a relatively small time-frame may be high: failure to secure employment in the field after earning a doctorate, or failure to make tenure.
Of course, if every up and coming philosopher made sure to cite thoroughly, this would be less of an issue. Hence we appear to be in a sort of prisoner’s dilemma.
I am not sure if there is any better solution to this problem than a concerted effort by journals to ensure proper citation practices. If journals keep papers from being published that fail to cite relevant literature, then the amount and frequency of papers published will decrease. Not only will this lead to less redundant and higher quality literature, it will reward philosophers who want to do the right thing.
LikeLiked by 1 person
Enzo Rossi: “a paper’s cross-niche appeal would become the main driver of high citation counts”
Won’t the driver of high citation counts will be the number of papers that are written in the area? If I publish a paper in area X and one in area Y, and the expectation is that papers in areas X and Y will cite previous papers in X and Y, and it happens that the next year 100 papers are published on X and 50 are published on Y, then my paper on X gets 100 citations and my paper on Y gets 50. And if a couple of papers on area Z brings in my paper on Y, but all subsequent papers on Z are not expected to cite my paper on Y, then Y’s cross-niche appeal isn’t going to swamp the effect of the larger amount of publications on X.
(Reposting a comment that’s apparently been left in moderation–my WordPress account automatically uses my WordPress screenname, which is not my full name, so I have signed out in order to be able to use my full name with this comment)
Pingback: Featured Philosopher: Marcia Baron « Philosopher