Read the whole book free at the MIT Press
The last part deals with fakes that borrow a famous name or look, which the book calls "brand appropriation", and with parody and hoax (p. 239). Fake universities take names that sound like the old elite universities of the American Ivy League, and fake journals copy the look of real ones. What is copied is the brand, not a single product (p. 239). Fittingly, some of the people who expose these fakes use imitation themselves: computer-generated papers, invented scientists, joke pen names (p. 239).
Brunton asks why some journals are called "spam journals", meaning journals that hunt for authors through junk email, and who gains and who loses from them (p. 239). Jacob treats the original and the counterfeit as tied together: wherever there is a valued original, a copy will follow, so the fake is to be expected, not an exception. Even a watchdog like COPE, the Committee on Publication Ethics, has to protect its own trademark (p. 240). Delfanti shows imitation used as protest. viXra.org is a preprint server, a website where researchers post papers before or instead of formal publication. It spells the name of the leading one, arXiv, backward and parodies its design to protest the way arXiv decides who may post (p. 240). Lippman looks at the watchdogs' humour. Classic carnival mocked the powerful; this humour mocks the fraudsters instead (pp. 240–241).
Chapter 18. Finn Brunton, "Making People and Influencing Friends: Citation Networks and the Appearance of Significance"
Brunton, who has written a history of spam, asks why these journals are called "spam journals" (p. 243). The obvious answer is that they find authors through unsolicited bulk email and use the spammer's playbook: lower fees, faster turnaround, friendly "reviewers", good "impact metrics" (numbers that claim to show how widely a journal is cited), mail-drop addresses, made-up editorial boards, and website text copied word for word from one journal to the next, typos included (pp. 243–245).
The deeper answer is that spam and spam journals both exist to create the look of importance (p. 245). A spammer has to win over the counting systems that decide what matters, because the content would not survive a human reading it (p. 245). There is a historical irony here. Google's PageRank, the method that ranked web pages by the links pointing to them, borrowed its idea from academic citation analysis, the study of which papers cite which. Its authors noted that web pages, unlike academic papers, were not carefully reviewed, so their counts could be inflated (p. 245). An arms race between link spammers and search engines followed. Early "link farms", clusters of fake pages that linked to one another, were "densely interlinked but lonely": they pointed at each other and little else pointed at them, which made them easy to spot. So the spammers adapted (p. 246). Spam accounts, like papers in spam journals, are rarely meant to be read. They exist to move numbers (p. 246). The methods change, but the goal stays the same, in Brunton's phrase: "making people, and influencing 'friends,' in the eyes of the machines" (p. 246).
The first lesson he draws from spam is to watch the side businesses. In a gold rush, the money is in selling shovels. Here the shovels are the companies that supply misleading metrics and identifiers to spam journals, "a crooked ancillary economy" (p. 247). The second lesson is that the question of who the victims are is harder than it looks (p. 247). Spam bots on Twitter inflate Twitter's own numbers. In the same way, take an adjunct (a lecturer on a short, part-time contract) who pays the equivalent of about $90. The adjunct gets a line on a CV, the department and the administration get something to count, and no one has to read the paper (p. 248). Spam journals also make the journals of the Global North, the wealthy countries mostly in North America and Europe, look serious by contrast. The top journals and universities already work as an unofficial citation cartel, a club whose members cite one another and keep each other's numbers high, and the cheap copies strengthen it (pp. 248–249). He stresses that this is not a defence of spam journals but an account of who gains from them (p. 249).
Chapter 19. Marie-Andrée Jacob, "Crack Open the Make Believe: Counterfeit, Publication Ethics, and the Global South"
Jacob, a legal scholar, has sat in on the quarterly forum of COPE, where journal editors bring hard cases for advice, since 2010 (p. 251). She is as interested in the watchdogs as in the people they accuse. She approaches fake journals through the idea of the counterfeit. She looks at the skill and craft that go into it, rather than condemning it or romanticizing it (p. 251). Counterfeiting, she argues, is a predictable response to the conditions described elsewhere in the book (p. 252).
On harm to the public, she argues that the danger bad journals pose is assumed rather than shown (p. 252). Watchdogs often "uncover" fakes with mocking humour, and the mockery gives the person doing the revealing an air of moral and intellectual superiority (p. 252). In 2013 John Bohannon ran a sting: he sent a deliberately flawed paper to open-access journals (journals that are free to read and often charge authors a fee). She reads it as an attack on the Global South, the poorer countries of Africa, Asia and Latin America. It begins, she notes, with the African-sounding pen name he used (pp. 252–253). Research by Xia and colleagues suggests that the real audience of such journals is the people who make decisions in Global South institutions, where authors need publications to keep their jobs (p. 253). She compares this to the documents issued by the Turkish Republic of Northern Cyprus, a state most of the world does not recognize: paperwork that works locally but is not "real" anywhere else (p. 253). She also notes that voices from the Global South are almost absent from the debate, "including this book" (p. 254). She asks where a hoax ends and fraud begins, and cites work that sees no clear line between them (p. 255).
On authenticity, she argues that counterfeiting helps those who are copied. It strengthens the model of elite science and keeps attention off the mainstream (p. 256). At COPE, she observes, disputes over who deserves to be listed as an author are discussed with attention to funding, rank and relationships. Cases of fabrication, meaning invented data, are handled differently: the "research record" is treated as a self-contained object to be put right (p. 257).
Finally, ethics itself can be faked. In China, stall owners sell fake bags with fake certificates of authenticity (p. 258). A certificate of good ethics is likewise a template anyone can copy. COPE has responded by giving each fee-paying member a logo with its own unique number (p. 258). She ends:
It is unhelpful to see the "make believe" as anomalous. (p. 259)
Chapter 20. Alessandro Delfanti, "Fake Archives: The Search for Openness in Scholarly Communication Platforms"
Delfanti studies viXra.org, a preprint server that copies the design, logo and layout of arXiv.org and spells its name backward (p. 261). The physicist Philip Gibbs launched it in 2009, frustrated with arXiv's dominance and what he saw as its abuses (pp. 261, 265). It holds a great deal of fringe work, and it uses some spam-like tricks, such as a mirror site that catches people who mistype arXiv's address (p. 261). Delfanti argues that it is not misconduct. What it shows is how digital platforms can work against their own talk of openness (p. 261).
ArXiv was founded in 1991 and is now run by Cornell. It holds more than 90 per cent of articles in particle physics and passed a million papers in 2014 (p. 262). In parts of physics, a researcher whose work is not on arXiv hardly exists (p. 262). ArXiv also feeds services that rank papers from "renowned" down to "unknown" (pp. 262–263). Anyone can read it, but it screens authors in three ways: an email address at a research institution or a recommendation from an existing author, automated checks, and human moderators who may reject a paper or move it into a general-physics category (p. 265). Critics accuse it of secrecy and blacklisting, which arXiv denies (p. 265).
ArXiv is open for readers, but not for authors. (p. 265)
Delfanti uses brand theory to read viXra as a "subverted" brand, like the mock versions of corporate logos that activists make, rather than a counterfeit meant to deceive. Its site says the copied design is parody, and that the reversed name and colours prevent confusion (pp. 263–264). By June 2017 it held more than eighteen thousand articles, mostly unorthodox, with some by recognized scholars (p. 266). It says that a paper accepted there does not count as academic publication (p. 266). Neither Google Scholar nor the physics service inSPIRE lists its papers, so it keeps its own download counts (p. 267). He is unsure what viXra will achieve. Comparisons with Linux or Wikipedia, he says, would be amiss (p. 268). Its main value may be the criticism it stands for. Preprint platforms such as arXiv, SSRN and Academia.edu now decide who gets in and keep their own metrics, and that power needs examining (p. 268).
Chapter 21. Alexandra Lippman, "Humor, Hoaxes, and Software in the Search for Academic Misconduct"
Lippman contrasts the humourless Office of Research Integrity, the US government office that rules on research misconduct, with the new watchdogs, who use jokes, pranks, pen names and hoaxes (p. 271). Her argument is that criticism of academic gaming, meaning working the system to score well on the numbers, has taken a carnival turn, with crowds, masks and mockery (pp. 271–272). It differs from classic carnival in its target. Older academic hoaxes, such as Alan Sokal's 1996 parody article that the journal Social Text published as serious work, mocked authority. The new pranks mock fraudsters (p. 271). Metrics, she argues, have produced a new, "unambitious" kind of misconduct, aimed only at meeting basic requirements, and the watchdogs follow it down (p. 272).
Retraction Watch, a blog about retractions (papers that journals withdraw after publication, usually for serious error or misconduct), relies on reader tips and puns. Its leaderboard of the authors with the most retracted papers is a "snide inversion" of the impact factor, the best-known score of how often a journal's articles are cited (pp. 272–273). The site Science Fraud credited its tipsters by pen name (p. 272). Anonymity works like the carnival masks that the theorist Bakhtin described, and lets junior scholars challenge senior ones (p. 273). PubPeer, a website where readers discuss published papers, is anonymous too but keeps a fairly serious tone (p. 273). Invented scholars such as Ike Antkare and Hoss Cartwright are jokes with a point (p. 274).
She then compares two text generators released ten years apart. Andrew Bulhak's Postmodernism Generator of 1996 mocked famous theorists by producing essays in their style; the online version had turned out 14,440,299 essays since 2000 when she used it (pp. 274–275). It aimed high, in the traditional way. SCIgen, written in 2005 by three MIT students, produced fake computer-science papers and aimed low, at spammish conferences that would accept almost anything. One of its papers was accepted by a conference, which then withdrew it, and the students held their own session, in disguise, in the conference hotel (pp. 275–276). They wrote jokes into the program's grammar, down to "a testbed of Gameboys" (p. 276). Then people began to use their tool in earnest. The publisher Springer hired Labbé to build SciDetect, software that spots generated papers. One SCIgen author criticized it as a way to avoid real peer review, the checking of a paper by experts before it is published (p. 277).
She ends with Bohannon's 2013 sting, in which 157 of 304 journals accepted a fatally flawed paper, among them journals from large established publishers (p. 279). SCIgen, she says, announces not the death of the author, as literary theory once put it, but "the death of the reader" (p. 279). Then she turns to the laughter itself. Are we laughing at people whose first language is not English, or at scholars from the Global South? Does the laughter confirm the standing of publishers who charge $10,000 per article (p. 279)?
Laughter judges but offers little analysis (p. 279)