Abstract
The ‘scourge of self-plagiarism’ has begun to find a place in the discourse of organization and management scholarship. Whether a real issue of concern or a moral panic, self-plagiarism has captured the attention of authors, editors, publishers, and plagiarism-detection software companies. The types of behaviors castigated as self-plagiarism and the severity of approach toward those behaviors vary, as power brokers in the publishing process argue they hold an ethical high ground. Yet, little has been done to problematize self-plagiarism as a concept and how, and why, it came to occupy such a central role in the academic discourse. In this article, I explore these issues and argue that self-plagiarism is a misnomer that has been retrospectively (im)moralized through regimes of power. I review the spectrum of behaviors that now fall under the self-plagiarism umbrella and problematize issues associated with self-plagiarism. I identify and challenge the power interests that are negotiating the spaces in which self-plagiarism has risen to the forefront and present a call to action to more transparently, and ethically, deal with issues that are currently labeled as ‘self-plagiarism’. Furthermore, in presenting this article, I engage in a form of ‘guerrilla plagiarism’ to resist the appropriation of my authorial voice by power elites in the institutional field of publishing.
The concept and enactment of ethics continue to grow in importance for organizations. Increasingly, organizations (to include publishers and higher education institutions) desire to be seen as ethical (Pullen and Rhodes, 2013) and compete to demonstrate the highest level of standards. And, yet, in the quest to be seen as ethical, socially constructed norms and codes of ethics may punitively target members of a community without considering alternative modes of understanding ethics from the perspective of the individual (Pullen and Rhodes, 2013; Randall, 2001). In the academic community, plagiarism and self-plagiarism are currently constructed as ethical issues (Elliott et al., 2013) and scholars themselves are the targets of reconstructed codes of conduct that address these issues.
As I have discussed elsewhere (Callahan, 2014), more and more publications are appearing about issues of self-plagiarism, and much debate has ensued about the ‘scourge of self-plagiarism’ (Green, 2005). In 2005, the number of hits on the keyword search ‘self-plagiarism’ was 8000 (Green, 2005); in 2013, that number had increased to 82,500 (Callahan, 2014). As of August, 2016, the same keyword search resulted in over 2 million results. Whether a real issue of ethical concern or a moral panic, self-plagiarism has captured the attention of authors, editors, publishers, and plagiarism-detection software companies.
I contend that the dominance of these power structures interrupts a scholar’s ability to express themselves as they so choose within the publishing space (Rhodes and Wray-Bliss, 2012). While others have addressed ethics with respect to the conduct of research (e.g. Jeanes, 2017), little has been done to problematize self-plagiarism as a concept and how, and why, it came to occupy such a central role in the discourse of academe. In this article, I extend the conversation about research ethics to explore these issues and argue that self-plagiarism is a misnomer that has been retrospectively (im)moralized (Bloom and White, 2016) through regimes of power.
To provide background for my argument, I first discuss plagiarism and self-plagiarism, reviewing the spectrum of behaviors that now fall under the self-plagiarism umbrella and problematizing issues associated with self-plagiarism. I then explore the power interests that are negotiating the spaces in which self-plagiarism has risen to the forefront and problematize the competing interests between these actors. Finally, I present a call to action to more transparently, and ethically, deal with issues that are currently labeled as ‘self-plagiarism’. In this article, I use the core ideas from my previously published short commentary on the topic (Callahan, 2014) and expand upon them by problematizing the concept of self-plagiarism, addressing the power interests that are driving the current focus on the issue, and offering a call to action for addressing ‘self-plagiarism’ within scholarly publications. In developing this article, I engage in a form of ‘guerrilla plagiarism’ (Randall, 2001) to resist the appropriation of my authorial voice by power elites in the field of publishing.
Plagiarism
To understand self-plagiarism, we must first understand plagiarism. Historically, the word ‘plagiarism’ is rooted in the Latin plagiarius, or kidnapper; in the 17th century, the word plagiary referred to a ‘kidnapper or a kidnapping, theft or a thief of ideas’ (McArthur, 1992: 784). Today, plagiarism is considered the theft of another’s words or ideas (Cronin, 2013). On rare occasions, this type of theft occurs when an individual presents as his or her own an entire paper published by another author, without attribution; it is much more common, however, for an individual to take sections of writing from one scholar and present them as his or her own.
Most scholars would contend that plagiarism is grievous because it ‘undermines the entire scholarly enterprise’ (Hexham, 2005: Section 1, paragraph 2). In academe, our ideas and the representation of those ideas (usually in the form of the written word) are our capital. Those words create the scholarly identities that form the basis of our credibility, reputations, and careers. Our ideas as scholars become the commodities that we trade for our livelihoods; however, as will be discussed later, our remuneration is indirect. Furthermore, identifying ‘ownership’ of those ideas can easily be uncertain territory.
And, yet, even highly esteemed scholars have been accused of plagiarism—from direct copying of language to appropriation of ideas without attribution. In his blog, Yiannis Gabriel (2014) posted his thoughts regarding the recently discovered plagiarism by major scholars such as Lewis Wolpert, Zygmunt Bauman, and Karl Weick. Wolpert accepted full responsibility for having used language taken directly from a variety of sources (to include Wikipedia) and apologized for having done so, claiming inadvertence and sloppiness (Davis, 2014). Bauman, as Gabriel (2014) notes, was not only sloppy in referencing and attribution, but he also seemed to have claimed ‘ownership of an idea that was very clearly someone else’s, in this case Erich Fromm’s’ (paragraph 9). Weick’s case has drawn both critics and supporters and is particularly relevant to the present argument because it is associated with ownership of language. Weick relayed a story from folklore using language nearly identical to a source, which he cited without using quotation marks. Many, such as Gelman and Basbøll (2014), contend that Weick’s actions were a grievous case of theft; others, such as the authors of the blog West Coast Stat Views (Can you plagiarize folklore? 25 April 2012, http://observationalepidemiology.blogspot.co.uk/2012/04/can-you-plagiarize-folkore.html), challenge whether or not folklore can be plagiarized, given the nature of repetition of oral history and lack of clear ownership of the original idea or story.
Such an argument is consistent with Randall’s (2001) deconstruction of originality and ownership in literature. Rose (2002) points out that Randall’s explanation of the history of plagiarism shows that ‘(a) plagiarist, then, is one who is by definition incapable of authorship’ (p. 271). Both authorship and plagiarism are ‘transhistorical’, and the definition and understanding of both change and evolve over time. As Ross (2002) notes, such evolution is based upon the decisions of any given ‘interpretive community’. It is institutions that have the power to affect evolution, creating a panopticon effect that encourages fellow academicians to surveil each other to ensure they follow the institutionally crafted, and institutionally benefiting, ‘rules’. In this context, it is this changing nature of language that has contributed to the popularization of another form of scholarly transgression—self-plagiarism.
Self-plagiarism
If plagiarism itself can be contested, then the uncertainty associated with the recently popular issue of self-plagiarism is surely even greater. The current positioning of the concept of ‘self-plagiarism’ began in the medical sciences before leapfrogging to the social sciences. A dominant player in the viral spread of the concept was the Committee on Publication Ethics (COPE), which was founded in 1997 by editors of medical journals (Yentis, 2010). Now, with more than 11,500 journals pledged to uphold the standards established by COPE (to include nearly 2500 social science and management related journals), the conceptions about what constitutes ‘self-plagiarism’ are grounded in natural and physical sciences. This is important because medical research follows inherently different practices to social science research, so definitions, understandings, and practices do not transfer cleanly and there has been limited effort to problematize how these meanings appear in social science spaces.
As a form of perceived transgression, self-plagiarism has instigated a change in guidelines of leading professional associations regarding ethical publishing behavior in organization and management studies. Honig et al. (2014) note that the Academy of Management’s code of ethical conduct has changed from ‘an implicit recognition that a certain amount of self-plagiarism is acceptable, as long as “different audiences and outlets” are employed’ (p. 128) to a more stringent and explicit exhortation to cite any and all words and ideas published, unpublished, or electronic, even if they are one’s own (Martin, 2013).
Although some have contended that self-plagiarism is a new concept (e.g. Bretag and Carapiet, 2007; Green, 2005), it has, in fact, been bandied about for more than a century; an early reference to self-plagiarism can be found in an anonymous response letter to the Atlantic Monthly in 1893 (Anonymous, 1893). While the origination of the term self-plagiarism is unclear (Bird and Sivilotti, 2008), what is clear is that the concept is once again gaining ground. Martin (2013) noted that scholarly journal articles regarding self-plagiarism and related misconduct increased significantly (from 170 to 820 articles) between 2000 and 2012. Even the latest edition of the Publication Manual of the American Psychological Association (American Psychological Association (APA, 2010)) includes a discussion of self-plagiarism for the first time (iThenticate, 2011); certainly, self-plagiarism is on the publishing dashboard right now.
In this spotlight, self-plagiarism is framed in at least four categories (Andreescu, 2013; Thurman et al., 2016) which occupy a spectrum of perceived (un)ethical behaviors on the part of authors. These are dual or duplicate publication, ‘salami slicing’, redundant publication, and textual recycling. In the paragraphs that follow, I will present the most rigorous definition of each of these categories; however, while there is general agreement in the literature that these four categories in the extreme can be defined as ‘self-plagiarism’, there is not agreement on the nature or threshold of behavior that would be considered self-plagiarism.
The first, and typically seen as the most egregious, is dual or duplicate publication (Andreescu, 2013). In this instance, the author publishes the same or substantially similar work in more than one outlet. Initially, this behavior was associated with authors publishing (nearly) the same work in its entirety in more than one journal. An example would be the widely publicized case of renowned economist Bruno Frey who, among other purported instances, published virtually the same critique of academic publishing in two different journals within 2 years (Storbeck, 2011). De Vasconcelos and Roig (2015) explore another example of duplicate publication in which a conference proceedings article was significantly expanded upon and then later published in a refereed journal. The article was subsequently retracted under the broad umbrella of plagiarism (as opposed to the more specific self-plagiarism).
The transgression of ‘salami slicing’ covers two types of self-plagiarising (Andreescu, 2013). One type is taking one large data set and dividing it into multiple projects to maximize the number of publications that can emerge from a single study. A recent example is work done about teachers, as employees of school organizations, and their perspectives of bullying among students. Sokol et al. (2016a, 2016b) used the same data set to explore ‘the impact of victims’ responses on teacher reactions to bullying’ (Sokol et al., 2016b: 78) and ‘teachers’ perspectives on effective responses to overt bullying’ (Sokol et al., 2016a: 851). The other type of salami slicing is to use a different portion of an existing data set for another paper. With this understanding of self-plagiarism, for example, Roulston’s (2001) insightful article that reanalyzed data from a 1991 study using a different methodological technique would have been an example of self-plagiarism despite having offered readers something ‘new’.
Roulston’s (2001) article also would have been guilty of redundant publication, another type of self-plagiarism (as it has currently been retrospectively constructed). This type of self-plagiarism occurs when an idea or an existing data set is used again in another publication with a different ‘thematic slant’ (Andreescu, 2013). Roulston’s (2001) re-analysis of the same data using a different theoretical lens resulted in a re-interpretation of the same findings. Examples of this type of publishing deemed as ‘self-plagiarism’ abound in the social science literature, especially because the nature of qualitative research encourages multiple subjectivities and, therefore, multiple interpretations of the same phenomenon.
Finally, textual recycling is frequently seen as self-plagiarism (Andreescu, 2013); it is probably what most people now think of with regard to self-plagiarism. This type of behavior is not so severe as to reach the level of copyright infringement, and indeed, it seems to contradict the very author protections that copyright was designed to afford (Bently, 1994). Textual recycling refers to reusing parts of one’s own published works in a subsequent publication without a formal citation to that earlier work. In 2009, the Academy of Management code of ethics specifically required, without calling it self-plagiarism, that Academy of Management members explicitly cite others’ work and ideas, including their own [emphasis added], even if the work or ideas are not quoted verbatim or paraphrased. This standard applies whether the previous work is published, unpublished [e.g., in-press or in-progress manuscripts, email communications, raw data (APA, 2010)], or electronically available. (Schmink, 2009)
Despite, or perhaps as a result of, this wide variation of what is seen to constitute self-plagiarism, there is much debate about the concept. The very definition of plagiarism as theft causes many to argue that one cannot steal from one’s self, and therefore, self-plagiarism is an oxymoron (Cronin, 2013) which purposefully ‘invoke[s] the pejorative tone of the root-word’ (Clarke, 2009: Section 2, paragraph 4). Some contend that self-plagiarism is academic fraud or misconduct (e.g. Bretag and Carapiet, 2007; Martin, 2013), while others have argued that scholars can, do, and even should reuse their written words and ideas, within reason and without citation (e.g. Andreescu, 2013; Anonymous, 1893; Chalmers, 2009; Hexham, 2005; Scanlon, 2007; Vermuelen, 2012). As the anonymous letter writer argued in 1893, reissuing one’s own ideas is much like the state reissuing money; ideas, like currency, have greater impact if widely circulated through multiple outlets by the same ‘issuer’, such as an author (Anonymous, 1893). Vermuelen (2012) and Neville (2005) extended this notion by contending that, today, pre-eminent seminal scholars with widely published, commonly themed works would be castigated as self-plagiarists, and our scholarly community would be the poorer for it.
For if plagiarism is theft of words or ideas, how can one steal one’s own words or ideas? Unless, of course, the author does not, in fact, have ownership of those words or ideas. Copyright laws were initiated as a means to protect the rights of authors (Bently, 1994). However, in the present era of publishing, those rights are consistently being called into question. Gennaro (2012) is particularly frank about how copyright law has come to privilege publishers at the expense of those who created the work in the first place: ‘Once you have transferred copyright to a journal [in order to publish] you cannot ethically use the words that you have written in another journal article; you no longer own those words’ (p. 109). Nevertheless, Bently (1994) remarks on Roland Barthes’ contention that once text has been published, the words no longer belong to that author or anyone else for that matter. It is this notion of intertextuality that leaves space for authors to challenge the demonization of using their own words, data, or ideas for different effects and toward different audiences.
Problematizing self-plagiarism
The concept of self-plagiarism can be critiqued on the whole as an example of retrospective moralization (Bloom and White, 2016) or, in this case, retrospective (im)moralization. This concept can be seen when a group retroactively reconstructs the norms of behavior to legitimize (or de-legitimize) already existing behavior. For much of the discourse surrounding self-plagiarism, this retrospective (im)moralization has explanatory power with respect to arguments about reach, rationale, and reductionism.
Reach
A key attribute of scholarly research is that it should have extensive reach in order to have maximum impact. This concept of reach can be understood in multiple ways. One way reach is manifested is through gaining the attention of members in multiple audiences. The notion here is that the readership of any given journal is likely at least to be somewhat different than the readership of another journal, trade magazine, or book. Each of these audiences is likely to be distinct, but may well have common interests; however, they are unlikely to read the same publications. The most rigorous interpretations of self-plagiarism would (now) contend that it is not appropriate to publish a similar manuscript that has been modified to address the needs of each different audience. Paraphrasing a well-written idea to avoid recycling may actually be worse because it is not honest that it is indeed the same idea (Andreescu, 2013). Consistency and clarity around a core set of concepts, constructs, ideas or data, methods, and design would be desirable to be in verbatim, or nearly so, language because it creates layered impact (Chalmers, 2009), and there is no question when one encounters the subsequent articles that this portion remains the same. Reach can also occur through first presenting at a conference or publishing a small portion of findings in a journal and then subsequently publishing a thoughtfully considered, well-developed manuscript that built upon previous versions of the manuscript. Such a journey would follow traditional guidelines regarding the way in which scholarly progression and new insights occur, and yet it is considered by some as self-plagiarism in the re-envisioned norms of academic behavior (Kokol et al., 2016).
Self-plagiarism associated with issues of reach is premised upon the privileged, and erroneous, belief that technology has rendered all publications accessible to all audiences. There are a number of reasons why this simply is not the case. First, databases which house journals are expensive; not every library can afford to maintain access to the work. Second, not all readers are privileged to be part of an institution that provides access to online copies of published manuscripts; these readers would have to purchase each work separately. Third, it assumes that readers always use targeted searches to identify works of interest and, therefore, will find the sole article published in a given outlet; it also assumes those targeted searches will be conducted within the database that houses the journal in question.
Rationale
The rationale for self-plagiarism can also be challenged. Some authors (e.g. Bruton and Rachal, 2015; Honig et al., 2014) contend that to recycle text is an indicator of laziness or deception. Another rationale is that the push for academe to operate more like a business has commodified scholarship, and new measures of success are predicated on quantity of output rather than quality of output. Quality is subjective and not easily captured by administrators, while quantity is easy to quickly identify. As noted earlier, Andreescu (2013) argued that it may be more ethical to simply use the same verbatim language when the content is common across two otherwise unique pieces of work. For example, the literature review or methods section are likely to be similar (or, historically, even identical) across multiple pieces of work within the broader body of work of an author’s research agenda.
Direct quotation of several sentences, or paragraphs, of one’s own work would be awkward and would be costly in terms of using limited word count space for self-citation instead of presenting truly new material. Andreescu (2013) suggests that, ethically, what is important is that each work offers new knowledge toward theory, research, or practice. Presenting the same language may be interpreted as more consistent and more honest, and ironically, given the earlier critique regarding the commodification of academe, such practices are also more efficient. Furthermore, allowing verbatim language precludes the need for excessive self-citation (which is also considered a transgression; see Brown-Syed, 2010) when large chunks of supporting language are easily identified as having been used in a previous work.
Reductionism
Finally, self-plagiarism is all too frequently subjected to reductionism when it is referred to as simply plagiarism in the quest for efficiency, lack of redundancy, or possibly conciseness of type font. Such reductionism places the author in a much more serious group of ‘offender’, one that is not retrospectively (im)moralized but that is genuinely and historically seen as a transgression by most scholars. Dropping off the modifier of ‘self’ conflates the construct with ‘plagiarism’, which is acknowledged widely as being an egregious offense in academe. For example, Honig and Bedi (2012) explore the nature of plagiarism among Academy of Management scholars, and in their description of methods, they qualified their stance toward self-plagiarism in such a way that implies it may well be perceived as traditional plagiarism by some: Since the focus of this study was on individuals plagiarizing others’ work without appropriate acknowledgement, we adopted a more conservative approach toward self-plagiarism. If authors used sections from their own previous work or cited the primary source, then it was not considered plagiarism. (p. 112)
But the relabeling also is connected to what appears to be a general misunderstanding of how to read findings from software detection reports that, once again, reduces the transgression to a percentage of a similarity index. However, these similarity index reports cannot be interpreted in isolation—references may be captured inadvertently, block quotes are identified as copying, common phrases are picked up (e.g. ‘the purpose of this study is to …’), and more. When scholars discuss the extent of plagiarism (and self-plagiarism) in the literature, they frequently express the severity of the incident in terms of percentage of similarity with other texts (e.g. De Vasconcelos and Roig, 2015; Honig and Bedi, 2012). Marik (2015) discusses this problem in depth as he reflects on his own ‘conviction’ as a self-plagiarist as a result, in part, of a plagiarism-detection software that picked up a high similarity index which included his name, affiliation, common phrases for conditions found in his field, and even attributed quotations.
Another reduction is the relegation of understandings of what constitutes quality in research to numeric performance indicators (Martin, 2013) at individual and institutional levels. Wade (1975) quotes a National Science Foundation (NSF) official: I am not looking forward to the day when Senator Proxmire’s assistant can get a printout from the Science Citation Index and says, ‘I see there has been a breakthrough in this field—I can’t quite pronounce it—and why haven’t you put half your money into it?’ (p. 430)
Fifty years ago, Janke (1967) noted the inherent concern of reducing the evaluation of the quality of work to the number of times it has been cited—namely that bad papers may be cited frequently, while outstanding papers ‘are too far ahead of their time’ (p. 891) and are not frequently cited. Concerns about advancement with citation indices expressed 50 years ago may well be coming to fruition.
Power interests and the social construction of self-plagiarism
In her book Pragmatic Plagiarism, Randall (2001) argued that plagiarism is contested territory. Power interests compete to define what constitutes plagiarism and to best profit from the existence and enforcement of the socially constructed phenomenon. She contends that [a]cademia is much prone to internal accusations of plagiarism, carried out from time to time in the public eye, as cleansing rituals in which the mechanism of scape-goating, raised to the level of purging the internal ranks of the elite, guarantees the honesty, or the appearance of honesty, among the ranks of the virtuous. (Randall, 2001: 121)
I contend that self-plagiarism is another step in this power struggle between corporate and scholarly interests.
Corporate
Corporate interests associated with self-plagiarism are represented not only by publishers but also by plagiarism-detection software companies and even institutes of higher education themselves. These groups all have a financial or managerial interest in promulgating the idea of self-plagiarism. Higher education is embedded in this corporatized audit society (Lorenz, 2012) and has fallen into Ritzer’s McDonaldization (Parker and Jary, 1995; Roberts and Donahue, 2000). As Margolis (2004) noted, ‘Where not so long ago professors “owned” the tools of scholarship, controlled the labor process, and certified the quality of our product, the process of McDonaldization has torn this relation asunder’ (p. 368). Self-plagiarism is an extension of the McDonaldization of academia.
The introduction of managerialism in higher education led to such things as ranking lists and league tables that stripped academics of their autonomy as scholars and reduced their ‘worth’ to quantifiable targets (Lorenz, 2012; Parker and Jary, 1995; Trowler, 2001). As universities continue to compete for recognition in world ranking lists, they pressure their faculty to produce more—more publications, more grant funding, more student graduations. The message delivered here is that more is better because quantity serves as a proxy for quality in an audit culture. In turn, universities create promotion and performance management systems linked to journal output that privileges quantity over quality. Such pressure lends itself to problems of rationale and reductionism noted above because anything that is not measured does not ‘count’ toward the proclaimed efficiency of the enterprise (Parker and Jary, 1995; Waring, 1988). Thus, cutting corners in a managerialized institution has created a culture which facilitates behaviors that have been described as self-plagiarism (e.g. ‘dividing one paper into many’ (Parker and Jary, 1995: 329)). Self-plagiarism merely becomes one more step in deprofessionalization, surveillance, and control that is guised in terms of ‘ethics’ in a corporatized university.
Perhaps more culpable in the creation of self-plagiarism as an ethical breach are publishers and plagiarism-detection software companies. Each of these entities has a financial interest in raising self-plagiarism as a transgression that needs to be monitored and curtailed. On one hand, corporate plagiarism-detection software companies want to create greater opportunity to increase markets for their products; manufacturing a new ‘need’ for detecting more instances of plagiarism suits that capitalist interest. On the other hand, publishers want to narrow the reach of each manuscript so that they have exclusive rights to make money from downloads and purchases of the research. Thus, technology has created instrumentalities that become meaningful only through their exchange value (Vieta, 2006).
Marcuse ([1941] 1997) foresaw this when he suggested that the commodification of products (e.g. journal articles) undermined the foundation of individuality. As a result, the less powerful (i.e. academics) were forced ‘under the dominion of the giant enterprises of the machine industry’ (Marcuse, [1941] 1997: 141) represented by corporate and university enterprises. Under their influence, the ‘… standardization of thought under the sway of technological rationality also affects the critical truth values. The latter are torn from the context to which they originally belonged and, in their new form, are given wide, even official publicity’ (p. 147).
As part of this technological growth, a platform that provided publishers even greater control over words and ideas appearing in their outlets emerged. The ability of software detection programs to find the minutest level of similarity has created a self-monitoring system in which people police themselves and their colleagues. However, early forays into the use of the leading plagiarism-detection software, Turnitin, were met with resistance because they were seen by academics to violate student rights under copyright law (Foster, 2002) and to unfairly take advantage of student–faculty power differentials (Vie, 2013). Bringing commercial demand full circle, Read (2008) argues that the CEO of Turnitin seeks to maximize profit by expanding into new markets. One such new market was through publishers, and self-plagiarism increased the ability for both commercial enterprises to increase their earnings. Vie (2013) suggests that these technological platforms ‘lull us into compliance and cause us to forget that there are larger issues regarding copyright law and ownership of ideas still up for debate’ (p. 3). This is a corporate solution to a non-corporate issue.
Another corporate solution to a non-corporate issue is the exchange of free labor by academics in the production of content for journals that serves as a ‘building block of [the publisher’s] strong market position’ (Nentjes, 2012: 427). The Intellectual Property Director at Cambridge University Press acknowledges ‘the legitimate economic interests of the rightsholders’ (Taylor, 2006: 262), recognizing that frequently the rightsholder is the publisher and sometimes the author. He contends that such economic interests are typically more substantive for publishers, particularly with regard to copyright piracy more so than plagiarism. Taylor (2006), the book publisher representative, frames the corporate role as protecting ‘the rights of the author in the commercial marketplace’ (p. 266). Others, such as Nentjes (2012), disagree and contend that (especially journal) publishers have been protecting corporate economic capital as their priority.
Academic
The priority of academics shifts based on whether one is a reader, an author, or an editor. All three share, with each other and with corporate entities, an interest in some type of efficiency with respect to achieving their goals. However, a key differentiation of academic interests from corporate interests is that reputational factors are far more significant for scholars (Nentjes, 2012; Parker and Jary, 1995).
Proponents of self-plagiarism as a scourge frequently argue that readers will be upset by reading the same words or ideas in multiple outlets without citation of the earlier sources (Bonnell et al., 2012; Nentjes, 2012). On one hand, they say, readers will feel duped if they pick up a subsequent article and find that portions of it are the same as another article they have read. This negative feeling is apparently mitigated if the author cites the previous work. Others contend, for efficiency’s sake, that the language from the original work may not even be used and that a reference or overview abstract should simply take the place of the common content (e.g. ‘For a description of the method, please see this article’.) (May, 1967). This would benefit the publisher of the original manuscript because many readers would want to know that information and would obtain the article. The efficiency argument can also be raised as rationale for not citing an author’s original use of an idea or language in every case because providing a citation encourages a reader to obtain that resource (a benefit to the publisher); however, if the use of particular language (background literature or methods) is not particularly germane to the new purpose or new audience of the present manuscript, taking the time to obtain and read the original is not an efficient use of the reader’s time.
Authors want to broaden the reach of each idea to many audiences because their performance is evaluated in terms of numbers and impact, which is made more likely by multiple publications that recycle text or build upon existing ideas. And, while doing these things, the expectation to avoid self-plagiarism is that authors cite their own ‘published, unpublished, and electronically available’ words and ideas (Schmink, 2009: 589). However, a dilemma here is that another transgression in publishing is frequent self-citation (Martin, 2013). Until recently, self-citation was seen as more unethical than reusing one’s own words and ideas to support a new project. Before the creation of the Science Citation Index in 1963, interest in citation analyses was very limited and concern about self-citation was generally in the context of egotistical and excessive reference to an author’s own works (Martyn, 1975). While this type of self-citation was most common, Tagliacozzo (1977) noted that there were, in fact, three different possibilities for self-citation—individual, journal, and institutional. Individual-level self-citation is an article in which the author(s) cites their own previous works; journal self-citation is an article in which other articles from the same journal are cited; and institutional self-citation is an article that cites other articles by authors from the same institution. Self-citation as primarily an individual transgression has been retrospectively moralized (Bloom and White, 2016). Self-citation is now more frequently associated with journal transgression (cf. Mavrogenis et al., 2010).
Finally, editors are in a precarious position of power—they are scholars laboring without pay for publishers for the honor of furthering the dissemination of the latest scholarly research. Editors have a vested interest in raising the perceived quality of their journals, as currently measured by various versions of an ‘impact factor’. This has led to concerns about coercive citations (Honig et al., 2014), a form of self-citation (Tagliacozzo, 1977), in which editors or reviewers pressure authors to cite articles from the journal itself to increase the prestige (and ranking) of the journal. Indeed, early writers on the topic warned that enthusiasm for citation analyses could lead to journals (and authors) attempting to manipulate their reputations and ‘impact’ by selective citation (e.g. Margolis, 1967; May, 1967; Wade, 1975).
This fear of self-plagiarism led to editors and publishing bodies to create formalized ethics policies (Jeanes, 2017) that resulted in rejections and retractions. In the spirit of the McUniversity (Parker and Jary, 1995) or McAcademy (Roberts and Donahue, 2000), there has emerged an emphasis on creation of and adherence to rules (compliance-oriented ethics) and ‘commitment to shared values and encouraging ethical aspirations (values-oriented)’ (Weaver et al., 1999). As more journal editors pick up the mantle of dealing with this once and future transgression, they write guidelines and codes. Codes are written by those in positions of relative power, relegating the transgression to the lowest denominator in the hierarchy (Farrell and Farrell, 1998). Such codes ‘do not liberate moral and ethical resources within the individual. Rather the codes use language to create or maintain a hierarchical power relationship’ (Farrell and Farrell, 1998: 598), thus disempowering the author.
Implications
In this power struggle, the author bears the brunt of a ‘blame the victim’ mentality. The work of Honig et al. (2014) reveals that at the end of the day, this ethical dilemma of self-plagiarism is focused upon the author. Our actions to address this relatively recently immoralized behavior are attempts to fix the scholars instead of challenging the structures that give rise to the behaviors and even to the phenomenon itself. To that end, I suggest that the field might reconsider its approach to self-plagiarism. I argue that authors and editors can create a healthier and more rigorous space for scholarly publishing by pursuing three courses of action. These three action items are to rename, recognize, and resist.
Rename
Multiple authors have noted the unjust and inflammatory nature of the label ‘self-plagiarism’ (Andreescu, 2013; Chalmers, 2009; Marik, 2015). Ironically, words matter. Painting (perhaps latently) as plagiarists authors who build upon their ideas to contribute to knowledge or who use their own previously published words to drive home a salient point does not strengthen the field. While clearly some actions captured under the current label of self-plagiarism are unethical or questionable, some may be argued as acceptable or even encouraged. Regardless of position on the spectrum of action within scholarly publication, to equate these actions with theft does not offer an accurate picture of what occurred. I concur with Andreescu (2013), Chalmers (2009), and others that self-plagiarism is a term that should be renamed or even banished from our lexicon. Thurman et al. (2016) offer the suggestion of identifying ‘unacceptable duplication’ instead of ‘self-plagiarism’. If one accepts that some or all of the types of behaviors are unethical, then I suggest it would be better to simply call those things as they are instead of creating a moral panic of self-plagiarism—excessive reuse of text, redundant publication without offering new knowledge, aggrandizing micro-segments of a study, and similar language would be more authentic ways of identifying problem areas.
Recognize
Extending from this need to rename the socially constructed offense of self-plagiarism is the need to recognize the spectrum of actions that fall into this category of action and establishing guidelines for assessing the soundness of the resultant manuscripts. I suggest that these guidelines be framed around standards associated with authenticity and audience.
Authenticity refers to vetting each work based on openness and honesty, by both author and editor, regarding how the newly submitted work builds upon existing work to create new knowledge. Following Andreescu (2013), I suggest that it is not necessary to formally quote one’s own words in each instance as this may well lead to excessive self-citation (which has potential negative implications for both the journal and the author (Martin, 2013)). However, I do concur that it is important to acknowledge what work(s) the present piece builds upon. May (1967), for example, suggests a covering abstract that contextualizes the manuscript within the scholarship of both author and field.
Audience refers to the extent to which the newly submitted work addresses a different group of readers who are less likely to have access to other similar published works. A cross-disciplinary study exploring a common phenomenon in two very different fields (say, for example, medicine and management) may well result in two publications with significant similarities of method and framing, but with substantive differences in purpose and interpretation of findings due to the different nature and needs of the readers. Another way of understanding unique audiences is if an author chooses to do similar variations of the write-up for a particular study as a refereed journal article, a trade magazine, and a book chapter, each of which is likely to reach a very different audience. With the Academy of Management exhortation that electronic and even unpublished ideas or words must be cited (Schmink, 2009), the scope of self-citation becomes massive—Twitter feeds, blogs, Facebook, LinkedIn, syllabi, and more. And the reach of ideas grows smaller and smaller as scholars try to balance the twin evils of self-plagiarism and self-citation.
Resist
Finally, as scholars who are producing works to extend knowledge of theory and practice and as editors who are shepherding those works into dissemination to their readers, we are called to resist those pressures to diminish the impact of our collective work. Authors and editors can both play a significant role here. Two methods of resistance are guerrilla plagiarism and editorial license.
Guerilla plagiarism
One way for authors to resist is to engage in ‘guerilla plagiarism’. Randall (2001) refers to ‘guerilla plagiarism’ as a means for marginalized groups to appropriate the language of oppressors in order to undermine their authority. Her conception of plagiarism rests upon the assumption that words and authorship have ‘ownership’ by another writer, and to engage in guerilla plagiarism is to reproduce those words in a ‘self-consciousness of contemporary appropriation’ (Randall, 2001: 220). However, I take a twist on this notion of guerilla plagiarism and suggest that authors can use their own words as a means of resistance against publishers who technically own the author’s words in today’s ‘property rights’–oriented (McArthur, 1992) society. Indeed, Chalmers (2009) implicitly argues for such guerilla plagiarism in his response to the Lancet editor when he acknowledges repeatedly using his own words in multiple publications as a means ‘to persuade readers and editors to take serious problems seriously’ (p. 1422). According to accepted copyright guidelines, up to 30% of one’s own words in subsequent publications is perfectly reasonable (Robinson, 2014; Samuelson, 1994). In this article, I have used some (less than 10%) of my own words from my previously published short commentary (Callahan, 2014) and a subsequent conference presentation that further developed and honed my reflections on this topic (Callahan, 2017) to acknowledge constructs that are relevant to understanding my arguments. These few paragraphs and sentences, largely verbatim to my original language, are merely foundational in nature and not central to the new points that I address here.
Editorial license
Editors can take the lead on meaningfully addressing these issues associated with what we currently call ‘self-plagiarism’ in a way that extends dissemination and application of new knowledge rather than joining the hue and cry of transgression. To be sure, there are examples of genuine breaches of ethical norms regarding plagiarism and other types of behavior currently clustered under the label of self-plagiarism. However, editors have significant power in determining how they choose to define what ethical scholarship is and to make situational choices about each submitted work.
The pressure on (especially) junior scholars to be prolific writers is substantial (Honig and Bedi, 2012; Honig et al., 2014). Yet, as Honig and Bedi (2012) discovered, junior scholars are actually less likely to engage in plagiaristic behaviors. Instead, they found that senior scholars were far more likely to have plagiarized. Rather than assume the worst about scholars who have shaped the field, I submit that this reinforces Randall’s (2001) contention that conventions regarding plagiarism change over time. This is not to say that high-volume publishing is necessarily bad, but if this high rate of publishing is in a narrow field of focus, there is a great likelihood that behaviors currently identified as the grievous case of ‘self-plagiarism’ will occur. Also, a similar idea or replication study by the same author may very well appear in a subsequent publication. The construct of interest or the data collection method can be explained with clarity in only so many ways—overlap with subsequent works is inevitable. Indeed, the idea that one could use the same data for more than one publication is claimed by some to be considered ‘salami slicing’ and, therefore, unethical. Finally, the pressure to be prolific shifts the emphasis of research from depth and breadth in long-term explorations to speed and efficiency in short-term assaults; this fundamentally changes the nature and impact of research. Editors can be important gatekeepers in the resistance to protect authors so that their words are still their own, to protect the field so that new knowledge is disseminated even if some of the authors’ words have been used elsewhere, and to protect their unique readership so that they have access to scholarship they might not otherwise have had access to.
Conclusion
The power interests within the field of scholarly publication compete, and I contend that the ultimate victim in this competition is the scholar author. In the words of Evans and Giroux (2015), the current trend of self-plagiarism has become ‘tantamount to a form of intellectual violence wrapped in objective scholarship that plagues the academy’ (paragraph 2). Thus, in this space of pursuit of scholarly knowledge and creativity, I offer a call to resist. A response for (especially senior) scholars to make is to take back copyright protections (Bently, 1994) and engage in ‘guerilla plagiarism’ (Randall, 2001) while working with editors to craft meaningful policies about when and how these actions retrospectively (im)moralized as self-plagiarism can be used in published works.
Footnotes
Acknowledgements
I extend my sincere thanks to Professor Carole Elliott, Professor Jim Stewart, and Dr. Mina Beigi for their helpful reviews and feedback on earlier drafts of this paper.
