Turning a blind eye to the reality of academic publishing

Abstract

There is a fundamental divide in academic publishing between those who create papers and those who publish them. Power lies with the publishers, who have persuaded academics that they should contribute a free supply of papers along with editing and refereeing services. Publishers make enormous profits from academic publishing: academics get to keep their jobs. Both like to be seen, and to see themselves, as advancing knowledge for public benefit: in fact, both produce indicators of scholarship rather than scholarship itself, and then for their own private benefit. We explore how this travesty has arisen, specifically the impact of Robert Maxwell in demonstrating how scholarship might be exploited and of Eugene Garfield in showing how it might be replaced by indicators, especially the journal impact factor. Their contributions were instrumental in making academic publishing viciously competitive while still presenting itself as collegial. Citation was presented as a dependable measure of academic quality requiring only the counting of cited papers. It was not—it was ruthlessly gamed, though the very possibility of gaming was dismissed out of hand. So, too, were the thoroughly negative observations that papers were produced to be counted rather than read, even by their authors, who were not expected to have written them either. Academic publishing chose to ignore such scurrilous suspicion, relying on peer review to sort out any problems that might have escaped its notice. The reluctance of academic publishing to recognise reality was challenged by the influx of predatory publishers with their open access model. Predatory publishers were quite open about publishing papers to generate performance indicators in return for the academic’s cash. Established publishers, insulated from reality by their profitability, could not fail to notice that more money could be made if lofty academic values were taken less seriously. The largest also began to see themselves, and to present themselves, as information companies producing not academic papers, but indicators to measure and rank all manner of academic activity. They seemed to forget that these indicators had been gamed for years, even though they themselves had been responsible for much of the manipulation. While the blind eye saw nothing and left academic publishing in ignorance, the reality seen by academic publishing’s good eye was no more than the creation of its own indicators.

Introduction

Academic publishing is dominated by two pervasive myths. The first of these is that academics conduct research in areas of interest and importance, write papers announcing their findings to a waiting world, publish their papers where they will be eagerly read and appreciated by the most appropriate audience, and reap rewards befitting those who contribute to knowledge. In reality, little of this happens. Academics publish to get and keep their jobs; visiting the frontiers of research is not a requirement:

. . . one respondent told us that he was explicitly requested by the editor of a 4-star journal to ‘dumb down’ his argument: ‘They actually used the expression ‘dumb down’, they did not use ‘simplify’ . . . I never thought you would actually get that . . . (Editor’s report to submitting author quoted in Butler and Spoelstra, 2014: 544–545)

The academic paper is now written less to be read than to be counted (Biagioli, 2020). It is designed to be publishable, which means – above all else – citable, and it is written for private rather than public benefit (Macdonald and Kam, 2010), contributing less to knowledge than to citation metrics:

We . . . [used] . . . to make our acceptance criterion those articles that we felt would make a contribution to the international literature. Now our basis for rejection is often ‘I don’t think this paper is going to be cited’. (Editor of a medical journal as quoted in Chew et al., 2007: 146)

The second myth is that academic publication has inbuilt safeguards which ensure that what is published adds to the sum of knowledge. Chief of these is peer review, more valued in principle than in practice:

. . . the ease of producing good-looking but vacuous articles risks exposing how thin the peer review barrier separating real science from nonsense truly is. (Seife, 2023)

‘I’ve published over 200 peer-reviewed research articles’, a candidate recently announced to a professional appointment panel I was sitting on, ‘and I have never had a manuscript rejected’. (Quoted in Flinders, 2023: 27)

Peer review serves less to examine scholarship than to lend a patina of propriety to the process of publication. No matter how nominal and ineffective, peer review still has about it the aura of old school respectability that academic publishers crave:

Academic publishing is the process of sharing . . . findings with the scientific community at large, thereby enabling the peers in one’s field to benefit from learning one’s findings and stimulating them to conduct their own research to fulfill the gaps [sic] in the current knowledge. (Misra et al., 2017: 1774)

Such idealism helps hide the primary purpose of academic publishing, which is to make money for publishers (often a great deal of money) through measuring the performance of academics.

Focus in this paper is on academic journals rather than books or the various forms of more ephemeral publishing offered by social media. We apply a Northern Hemisphere and generally UK perspective to academic publishing. We refer occasionally to Robert Maxwell and often to Eugene Garfield because each has wrought radical change in the industry. Robert Maxwell enticed chosen academics with a rosy glimpse of commercial rewards in academic publishing that would never be theirs, and Eugene Garfield concocted a measure of academic performance derived from journal publication that would poison the scholarly well for decades.

It’s a racket so audacious that if you pitched it as a business model, investors would laugh you out of the room. Yet for decades this scheme has siphoned billions from research budgets. (Avasthi, 2025)

Throughout, we exploit in quotation the personal experiences of those who know academic publishing well and can bear witness to the innovation it has undergone. No change has been greater than the last explored in this paper, the introduction of open access (OA) and article processing charges (APCs), making academics the customers of publishers, paying them directly for publication in their journals and for the performance metrics academics require. Established publishers, which had long presented at least a semblance of propriety in academic publishing and were accustomed to their comfortable and profitable interpretation of reality, were quite unprepared for what was coming their way.

The powerful and the pitiful in academic publishing

The academic publishing industry is dominated by five multinationals – Elsevier, Springer, Taylor & Francis, Wiley and Sage. The growth of the Big Five has been prodigious: 20% of the world journal market in 1973, 30% in 1996, 50% in 2006, 53% in 2013 and 60% in 2023 (Crotty, 2023; Hagve, 2020; Larivière et al., 2015). No other academic publisher has more than a 3% share. With profit margins approaching 40%, the Big Five are able to enrich themselves at levels even the pharmaceutical and armaments industries might envy (Hagve, 2020). Their growth has continued through takeovers of other publishers and through expansion into associated businesses, especially data gathering and analysis.

Key to the strategy of academic publishers has long been distancing themselves from the academic side of the business. Researching, writing, editing and the like have remained the preserve of academics, supposedly driven by the search for knowledge rather than for the search for profits of publishers. Beyond the possibility of a parsimonious stipend, a mean little honorarium, academics are paid nothing above their basic salary to produce papers for journals, serve on editorial boards, and referee submissions (see Van Noorden, 2013). To be sure, the relationship was symbiotic: academic freedom lent huge value to the publisher’s journal, and the publisher’s journal provided the performance metrics on which the academic depended. It is hard to imagine a multinational in any other industry being content to leave its supply chain to forces over which it had no direct control. Did publishers really believe what they chose to see when they declared that academics wanted no more than to improve the publisher’s product?

Most researchers give up time to review papers for no charge . . . Why do they do it? . . . [They] quite simply do it because they enjoy being able to improve papers. (Robert Campbell, chairman of the Publishing Research Consortium as quoted in Sense About Science, 2009)

The satisfaction of refereeing has done little to forestall a pressing shortage of referees (Forrester, 2023; Meyerson et al., 2025; Tropini et al., 2023). An editor might now expect to issue 40 invitations to secure the services of just a couple (Wosen, 2025):

I like staying up to date and be challenged in my thinking because of reviewing, but I don’t like the fact that my free labour helps to increase profit for major publishers. (Respondent in survey by Bonn and Pinxten, 2021)

Academic labour freely available to the publisher is freely available for the publisher to squander. Every re-submission to another journal has demanded hours of re-formatting to meet the esoteric requirements of publishers’ manuscript submission systems (Khan et al., 2018). Estimated wastage there is 52 hours a year for the average academic (at the rate of 14 hours per manuscript; LeBlanc et al., 2019). Publishers still impose apparently random word limits on authors, a relic of a distant time when more words cost more to print. In a world of electronic publishing, such limits are a senseless restriction – and an indication that publishers turning a blind eye to the plight of academics was beginning to tip over from not knowing into not caring:

I’m starting to keep a list of the publishers I will not referee for. If they don’t see value in serving the academic community, I don’t see value in donating my work to them – and I do see harm. (Academic quoted in Kincaid, 2023a)

The seeds of change

Robert Maxwell

Maxwell was an outsider in Britain, a Czech Jew with neither knowledge of academic publishing nor love of learning. Robert Maxwell was never an establishment figure and his entry into the genteel world of academic publishing in the years immediately following the Second World War may well have been, at least in part, an attempt to compensate. Maxwell ‘turned scientific journals into a spectacular money-making machine that bankrolled his rise in British society’ (Buranyi, 2017). By injecting energy and dynamism into the stuffy environment of academic publishing, he made the fusty fashionable. His Pergamon Press exploited the academic status of an Oxford base where Maxwell could pay court to academics. Schmoozing editors and authors was central to Maxwell’s strategy:

Many Pergamon authors and editors appreciated Maxwell’s support in their academic careers. . . . He provided them with kudos by publishing important but esoteric material which did not sell in large numbers. He also accepted for publication applied material that the learned societies did not consider, until much later, as proper for gentlemen of science, and made available attractive locations for regular editorial meetings. (Cox, 2002: 276)

In his 40 years at the helm of Pergamon, Maxwell launched 700 journals, mainly in Engineering, Medicine and the Sciences. His tactics included acquiring smaller publishers and university presses, organisations which tended to put scholarship before profit; for Maxwell, scholarship was entirely a route to profit.

In 1973, the UK Department of Trade and Industry declared Maxwell unfit to run a publicly quoted company. It mattered not a jot. In 1991, Pergamon was taken over by Elsevier, now the behemoth among the world’s gargantuan academic publishers. Maxwell was a decorated war hero. He was also a revolting individual, making off with the pension funds of his employees and finally disappearing while urinating from the stern of his luxury yacht. It is convenient to attribute to Robert Maxwell many of the post-war changes in academic publishing (see Braley, 2005), but more realistic to see them in the context of what was happening elsewhere, especially in higher education. Until the 1960s, something like 3% of UK school leavers went on to university. Forty years later, the Blair government was proposing that 50% of the cohort should be entering higher education, a target that was rapidly met (Morgan, 2023). Similar targets were set in many other countries, developed and developing alike, and higher education became very big business indeed. The market for journals soared, as did the number of journal titles, and the profits of academic publishers. There is little doubt that Robert Maxwell was a disruptive influence in academic publishing. Domination by an intellectual elite was gone forever, though its aura remained good for business and was allowed to linger. In practice, the elite had become an intellectual underclass serving the requirements of an international publishing industry in return for performance metrics. Maxwell’s strength lay in his ability to describe a reality in which others would believe.

Eugene Garfield

Eugene Garfield is best known for devising the journal impact factor (JIF), based on how often a journal’s contents are cited in the years immediately following publication (Larivière and Sugimoto, 2019). The effect this has had on academic life and on all parts of academic publishing has been profound, greater than any other development during the same period except, perhaps, the Internet. In consequence, it is difficult not to see Garfield in terms of the JIF: he devised the term in 1955 (Garfield, 1955) and founded the Institute for Scientific Information (ISI) in Philadelphia in 1956 to exploit his brainchild. Thomson Reuters acquired ISI in 1992 with Garfield remaining chairman until his death in 2017. Clarivate ISI still calculates journal impact factors annually (Bohorquez et al., 2024). ¹

You have to understand that there are some people [working for academic publishers] for whom, when Impact Factors are published, it’s their life . . . it’s what they’ve been waiting for all year. It’s bigger than Christmas. And people will do their utmost to get the highest Impact Factor possible. (David Tempest of Elsevier quoted in Burbules, 2015: 722)

Finding a use for the impact factor took decades. Garfield’s original intention was that his impact factor would let authors know who was using their work (Garfield, 1955: 109).

By referring to the listings for his article, an author could readily determine which other scientists were making reference to his work, thus increasing communication possibilities between scientists. (Garfield, 1955: 109)

Or perhaps a use in historical research:

[The impact factor] would clearly be particularly useful in historical research, when one is trying to evaluate the significance of a particular work and its impact on the literature and thinking of the period. Such an ‘impact factor’ may be much more indicative than an absolute count of the number of a scientist’s publications . . . (Garfield, 1955: 109)

Garfield thought the impact factor might allow Nobel Prizes to be awarded objectively by citation count – the more citations, the better the paper and hence its author – rather than subjectively by committee (Garfield and Malin, 1968). And authors seeking suitable journals in which to publish might find the JIF useful (Bensman, 2007: 136). By 1972, Garfield had still not settled upon a satisfactory purpose for his JIF. Nothing daunted, he explored the logic that linked the citing paper to the cited one:

[Garfield] pointed out that a citation index is based upon the principle that there is some meaningful relationship between one paper and another that cites it and, thus, between the work of the two authors or two groups of authors who published the papers (Bensman, 2007: 128)

But this led only to the further problem that the citation data collected by ISI to calculate a JIF was for journals and not individual papers. As was fashionable at the time, Garfield looked to the mystical properties of peer review to overcome the problem; after all, citation to a paper was a recommendation of peers:

Even in high IF journals there are some poorly cited papers. But even these poorly cited papers are usually good, professional papers. They went through the sieve of thorough refereeing and editing. (Garfield and Pudovkin, 2015: 10–11)

Garfield, says Bensman (2007), was a man able to live with contradiction, to entertain opposing ideas. His logic stemmed from bibliometrics rather than business or academe and was presented weekly to subscribers in Current Comments, his own early form of social media. Garfield’s ruminations reveal him to have been something of an elitist. Why bother to point the way to academic quality when any academic worth his salt would already know which were the discipline’s leading journals and which authors published in them? For Garfield, the JIF was less useful for identifying the best than for weeding out the worst. He often observed that most scientific papers are cited rarely and some 25% never: such papers and their authors were obstacles to citation data disclosing what was truly significant.

Decades were to pass during which even Garfield occasionally conceded that there were problems with his JIF as a measure of quality (e.g. Garfield, 2006). Notwithstanding, Garfield’s faith in his own fiction that the most cited papers are the best papers became inexorably entrenched not only in his own mind but in the business models of both higher education and academic publishing:

Unwittingly or not, Garfield had created the tools for academic game-playing and institutional performativity. (Mills, 2023: 13)

Relying on an unnatural alliance of collegiality and bibliometrics for justification, Garfield insisted that the JIF could not possibly be gamed because any attempt would be noticed by the scientific community and science would self-correct (Mills, 2023: 13):

The combination of a long bibliography of papers published in obscure journals and an abnormally high self-citation count would make the intent so obvious that the technique would be self-defeating. (Garfield, 1979: 362)

Even while he was insisting that gaming of the JIF would be impossible, Garfield must have realised that the more influential his indicator of performance, the more it would tempt gaming. And so gaming proliferated until no part of academic publishing could afford not to game. Even in his own company (as Clarivate ISI), the gaming that Garfield insisted could not happen was rife (Bohorquez et al., 2024). Clarivate has deliberately never been precise about what bits of a journal’s contents should be included in the JIF calculation (Metze, 2010; Moed and Van Leeuwen, 1995), which gives considerable scope for negotiation and adjustment (Brembs et al., 2013). Clarivate could readily exclude self-citation from its calculations of JIF, but it does not: all citation contributes to the JIF and so to author and journal value.

For years, Garfield (2006) hailed a paper by Oliver Lowry and colleagues (1951), cited 300,000 times by 2005, as an example of brilliance having a major impact on science (Garfield, 1977). The paper had 355,968 citations by 2025 and was still the most cited ever (Van Noorden, 2025). Lowry et al. (1951) is cited not because it is brilliant, but to demonstrate adherence to a discipline’s mainstream, as Garfield should have known. Many papers are cited simply because they always are cited. When Garfield (1974) came to update his list of the 50 most cited papers published between 1961 and 1967, only 10 had been displaced in a decade. Lowry himself protested that his work dealt simply with laboratory methods, than which nothing could be more mundane:

It is flattering to be ‘most cited author’, but I am afraid it does not signify great scientific accomplishment. . . .. (Lowry, 1969)

The most cited papers are not necessarily the best papers (not unless ‘best’ is defined as ‘most cited’; Larivière et al., 2016). Absent this condition there is ‘no correlation between number of citations and expert ratings of article quality’ (Nieminen et al., 2006).

I’ve found little correlation between articles that changed the world and number of citations to them . . . (Editor quoted in Chew et al., 2007: 147)

Garfield showed little interest in actually testing the link between JIF and academic performance. Like the impossibility of gaming, it was a convenient delusion. Any remaining problems would somehow be solved by peer review, yet another delusion. In 1996, Garfield had shrugged off any responsibility for the accuracy of citation indicators, declaring that ‘impact and other citation measures merely report the facts’. What they did not report could not be relied on as fact (Garfield, 1996: 412). Garfield’s obsession with the hyper-citation of the Lowry paper did not prevent him spelling Lowry with an ‘e’ (e.g. Garfield, 1996), extraordinary for someone so engaged with the homograph problem in identifying authors, but understandable for someone who created his own facts.

The publisher’s view

Universities use a range of threats and incentives less to foster research among their employees than to encourage their publication in journals deemed ‘best’ by Clarivate. The JIF of academic journals is still calculated by the descendant of Garfield’s original company (ISI). Its venerable formula is ripe for manipulation. Citations are the numerator, valid content the denominator. So, the larger the numerator and the smaller the denominator, the higher the JIF and the ‘better’ – at least the more prestigious and expensive – the journal.

Prices [of journals] are positively correlated with quality measured by the number of citations they receive . . . (Dewatripont et al., 2007: 409)

Clarivate works on the commercial side of academic publishing. Its decisions are not made on academic grounds and its algorithms have long been secret – even from the academics who create the data Clarivate uses. Academic fortunes have been ‘placed into the hands of a private firm that has failed to respond to increasingly numerous calls for action’ (Archambault and Larivière, 2009: 648). Clarivate’s lists of JIFs are its own intellectual property, not to be revealed by others (Willmott, 2011), though many journals advertise their own (unverified) JIFs. Clarivate claims that any journal with 20% of its citation to the journal’s own papers will be investigated for possible manipulation of citation data (to boost JIF), but this seems to be subject to commercial considerations (Siler and Larivière, 2022). The World Journal of Gastroenterology reached not 20% internal citation but 85% before being dropped by Clarivate (Begley, 2006).

The most powerful publishers are able to meet with Clarivate to appeal its JIF determinations. Algorithms can be set aside; the JIF is negotiable:

Publishers often forget that the IF [impact factor] was designed as a measure of quality, innovation, and contribution to scientific advance and, instead of trying hard to improve their journals in terms of scientific quality, they take advantage of certain design flaws and disadvantages of the IF that permit a degree of artificial and arbitrary inflation of this index. (Falagas and Alexiou, 2008: 223)

After their meeting with Clarivate (then Thomson Reuters), the publishers of PloS Medicine discovered that the journal’s IF might be 3 – or perhaps 11:

. . . there was no explicit process for deciding which articles other than original research articles it deems as citable. We conclude that science is currently rated by a process that is itself unscientific, subjective and secretive. (Quoted in Metze, 2010: 939)

Through negotiation, the JIF of Current Biology was increased from 7.0 to 11.9 in the space of a year (Brembs et al., 2013), suggesting that a JIF is really whatever the largest publishers want it to be (see Mason and Singh, 2022). Even the tiniest tweak could make a huge difference to the indicator: the simple reclassification of ‘meeting abstracts’ as ‘academic papers’ in the Journal of the Federation of American Societies for Experimental Biology boosted its JIF from 0.24 in 1988 to 18.3 in 1989 (Gowrishankar and Divakar, 1999). COVID alone was responsible for an increase in the Lancet’s JIF from 79.3 in 2021 to 202.7 the following year, a more modest leap in quality of just 255% (Kincaid, 2022). Measured quality also varied by academic discipline: the JIF of the British Journal of Medicine is now above 30. This compares with 6 for Research Policy, 3 for Human Relations, and just 0.1 for Notes and Queries, a prominent publication in the Arts, evidently of no quality at all.

Much like dowsing, homeopathy or astrology, journal ranking seems to appeal to subjective impressions of certain effects . . . (Brembs et al., 2013: 7)

Large publishers are powerful and influential beasts. Elsevier alone spends 400,000 euros each year lobbying the European Commission (Grossman and Brembs, 2021). In the UK, the Finch Report of 2011 on expanding access to published research findings joined with academic publishers in arguing that the private benefit of the publisher would always trickle down to benefit the public (see Macdonald, 2013). The publisher’s role has changed quite a bit:

. . . traditional publishing activities like copy-editing, type-setting and managing the peer review process account for only a minority of the overall costs. In practice, being an academic publisher today is primarily about maintaining a technology platform, running an organisation and competing for article submissions. (Johnson, 2024)

The academic publisher still requires academic input, of course, and cannot direct this workforce without undermining the academic independence and thus the value of the journal. What the publisher can do is keep the academic workforce in its place. Abuse flourished while the trappings of academic scholarship, peer review in particular, were used to provide assurance that all was well. Having reminded its editors and authors that ‘unethical practices should be avoided at all times’, Taylor & Francis told its editors that ‘selecting articles with the most potential to be highly cited for the first issues of the year maximises their time in the two-year Impact Factor window’, and its authors that ‘if four researchers collaborated on a paper, the article would be entitled to 200 eprints to share (50 eprints for each author). That’s a lot of readers, and a lot of potential citations’ (Nicholas, 2015). Gaming had become endemic, enabling the creation of an image of reality with which publishers were comfortable.

The academic’s view

It is hard for the academic to see beyond citation and the calculations made from it. The indicator has surpassed in importance what it is supposed to indicate. According to Peter Higgs (of Higgs boson fame), no university using modern measures of performance would ever have employed Peter Higgs (Aitkenhead, 2013). The curriculum vitae of the modern academic is at its most impressive when it dazzles with numbers rather than ideas:

H-index: 29; total citations: 2962 (2505 without self-citations); citing articles: 1953 (1852 without self-citations). Average citations per item: 25.18. The most highly cited paper has received 442 citations. From Google Scholar: H-index: 35 (22 since 2012); i-10 index: 88 (53 since 2012), total citations 4377 (2398 since 2012). (CV of an academic job applicant cited in Wood, 2021: 13)

Papers are made more publishable by making them more citable, which means ensuring they are compatible with prevailing ideas, what Finley (2023) calls the ‘preapproved narrative’. The ideal paper is one that echoes what everyone else is saying:

. . . papers that supported commonly held beliefs were cited nine times as frequently as those that conflicted with common beliefs. (Wright and Scott Armstrong, 2008)

Radical new research is hard to cite – papers announcing major discoveries may not be cited for years (van Raan, 2004) – and is best avoided:

Every university is pushing publishing so hard that this results in significantly lower quality research in total . . . the vast majority of work is derivative, or makes observations that amount to ‘this tiny square of the sky is blue . . . unlike the sky over there which is blue’. (Editor quoted in Karabag and Berggren, 2016)

. . . [I]mpact factors are deeply perverse as the less a journal publishes primary research, the higher its impact factor will be. . . . [T]he most successful journal is one which publishes almost no empirical research at all. (Heathers and Grimes, 2022: 21)

Manipulation of citation has become so extensive and ubiquitous that it is accepted as normal. Even the most egregious gaming attracts little attention. For instance, ‘Highlights of the year’ for 2011 in the Journal of the American College of Cardiology comes with 292 references, 272 of them to papers in that very journal, increasing its JIF by 250% and instantly making it the world’s best heart journal (Opthof, 2013). With 60% of citations in the Journal of Marketing Research being to papers in the Journal of Marketing Research, the journal shows little interest in anything beyond the Journal of Marketing Research. Administrative Science Quarterly and the Academy of Management’s Journal and Review are as introverted as the house magazine of some large corporation, and therefore score well (Macdonald, 2011). Coercive citation (the editor insisting that submissions cite papers in her own journal no matter how irrelevant) is now a practice so embedded that authors know full well what is expected of them (Fong et al., 2023). The most prestigious journals are the most likely to coerce authors to cite the journal’s own papers, perhaps because they have most to lose if authors do not (Wilhite and Fong, 2012).

We have noticed that you cite Leukemia [just once in 42 references]. Consequently, we kindly ask you to add references of articles published in Leukemia to your present article. (Editor of Leukemia to an author as quoted in Smith, 1997)

There is nothing new about sloppy citation habits:

How often are citations simply lifted from the bibliography in someone else’s work without either reading or giving credit to the man who did the original search of the literature? How often are citations tacked on after the paper is completed as an afterthought and window dressing? (Kaplan, 1965: 181)

What is new is that a wrong citation (inappropriate, irrelevant or simply non-existent), being every bit as countable as a right citation, is just as valuable. Many citations are very wrong (Pavlovic et al., 2021; Wright and Scott Armstrong, 2008). When Elsevier provided an example to demonstrate its preferred citation style, this totally fictitious reference went on to be cited nearly 400 times by 2017 and nearly 500 by 2023 (Harzing, 2023). In top journals, about a quarter of citations do not support the relevant argument (e.g. Smith and Cumberledge, 2020). About 20% of authors have actually read what they cite (Simkin and Roychowdhury, 2003; see also Knothe, 2006).

The context in which citation errors are made is one in which reading no longer complements writing (Macdonald and Kam, 2010). Much electronic copy now carries a consumer warning of the minutes it takes to read. Summaries minimise the burden of reading:

Based on your entire viewing history, you could have saved 20,406 minutes by using Summaries. (Email to author from Academia.edu, April 2024)

Academic papers have become steadily less readable (Wen and Lei, 2022), perhaps because the less readable and more impenetrable a paper, the more likely it is to be cited (Ante, 2022; Wang et al., 2022). Everybody writing and nobody reading rather suggests that everybody is writing for nobody (Walls, 2018). Even peer review can be detached from reading:

I am sure there are still people who are printing papers and reading them by the fireplace on a Saturday . . . but AI [artificial intelligence] tools are helpful for those who need to be efficient. (Szymon Machajewski quoted in Swaak, 2023)

The really efficient academic may never read any more than a paper’s title, keywords and abstract (O’Leary et al., 2023).

I’ve been on a number of search committees. I don’t remember anybody looking at anybody’s papers. Number of IF of pubs are what counts. (Terry McGlynn as quoted in Smaldino and McElreath, 2016)

The crudeness of citation cartels (arrangements among authors to cite each other) has been supplemented with more sophisticated and extensive gaming. Single-authored papers have become rare, but not, as is often claimed, because research has become more complex and collaborative. Co-authors can always be relied upon to cite themselves (Marušić et al., 2011) meaning more citations, a higher JIF and a ‘better’ journal. Some papers, in Physics for instance, have hundreds of authors. The author list can be longer than the paper (e.g. Khachatryan et al., 2010). Not all of these authors have had much – indeed, anything – to do with creating the paper (Macdonald, 2023). Authorship has descended from a claim to have written a paper, to a sign of some sort of involvement with the paper, to an entitlement in recognition of an individual’s authority and influence (Macdonald, 2026). When authors value a paper for its metrics, they easily become detached from its contents. Accused of plagiarising, one first author responded:

After careful checking, I noticed that I am not the author of this paper despite my first authorship since it has been written by our previous medical writer [sic]. (Romaric Loffroy as quoted in Joeling, 2023)

Even papers published by the most respectable learned societies have not always been written by their listed authors (Karabag and Berggren, 2012).

The concept of authorship has escaped from academic confines (Chawla, 2023). Authorship can be simply bought and sold (e.g. Ansede, 2023a). For instance, an organisation trading as Scopus Today issues weekly lists of available slots, charging so much to be first author and a bit less to be a middle author of a paper already accepted for publication. Increasingly, buying authorship is simply a transaction, devoid of moral implications:

. . . many scholars consider authorship and citation to be fungible attributes, components of a project one can alter to improve their publication and funding record or to increase journal impact factors. (Fong and Wilhite, 2017)

In the field of medicine, commercial ghost writing (paying an anonymous company to produce a paper) is rife. Some years ago, 16% of papers published in the prestigious New England Journal of Medicine had ghost authors, and no fewer than 44% had honorary authors (Flanagin et al., 1998). Of papers describing industry-initiated randomised trials, nearly all were ghost-written (Gøtzsche et al., 2007). Illustrious academics front papers produced by unacknowledged others, often with the full knowledge of their universities:

An option would be to add Greim and Kier or Kirkland to have their names on the publication, but we would be keeping the cost down by us doing the writing and they would just edit & sign their names so to speak. Recall that is how we handled Williams Kroes & Munro, 2000. ² (Email from William Heydens, regulatory product safety assessment lead at Monsanto, February 2015) ³

Even behaviour as outrageous as this attracts little attention. It has been normalised by an academic publishing industry that views scholarship, much like it sees peer review, not as an antibiotic fighting disease, but as an anaesthetic ensuring that disease is hardly noticed.

Predatory publishing – Let there be light

The open access movement tackled the problem of society paying several times over for academic publications: once through governments (and so taxpayers) covering most university costs; again through governments (and so taxpayers again) awarding specific grants for academic research; and then through direct payment to the publisher for access to journals.

Rather elegantly labelled the ‘triple-pay’ model by some commentators, this business model has long looked bizarre without ever changing. (Braley, 2005: 16)

The argument for open access is that the public should pay but once to use academic journals. A portion of the research costs paid by government and research funders would be ringfenced to pay publishers an ‘article processing charge’ (APC), an arrangement that would still disadvantage unfunded academics and those in the developing world (Akabayashi, 2024; Anderson, 2023). Established academic publishers were less enthusiastic about open access than governments and research funders, being quite comfortable with their familiar and highly profitable subscription model (Arunachalam et al., 2014).

Meanwhile, Jeffrey Beall, a librarian at the University of Colorado, was also unhappy about open access. It seemed to him that open access publishers would publish papers in return for payment, regardless of their quality. In one notorious sting, a paper consisting entirely of ‘Get me off your fucking mailing list’ repeated 800 times was submitted to the International Journal of Advanced Computer Technology and accepted with no requirement for revision (see Safi, 2014). ⁴ One referee apparently rated the paper ‘excellent’ (Sandlin, 2025), which says much about the distance between the peer review of theory and the peer review of practice. Beall christened the industry’s new open access entrants ‘predatory publishers’ and brazenly compiled a list of them. Legal threats from listed publishers ensued, followed by the sudden removal of Beall’s blacklist from the Internet in 2017 and of Beall from his job (Butler, 2013). Even so, the list survives and is still used, there being nothing better and the problem of identifying publishers who will accept any paper in return for cash remaining as intractable as ever.

At first, established publishers countered this intrusion into their reality by bathing the sins of predatory publishers in the light of their own virtue. Established publishers argued that predatory publishers would undermine the academic values they held so dear and had struggled for so long to protect. Crucially, the predators showed little respect for the peer review that established publishers – and Garfield – considered a guarantee of academic quality. With all the evangelical zeal they could muster, and as guardians of all that was good in academic publishing, established publishers gave dreadful warning of the fate that awaited those who succumbed to the blandishments of predatory publishers.

The hypocrisy is astounding and would have been derided had knowledge of the workings of academic publishing been more widespread. Established publishers cautioned authors to be wary of websites that were vague about a publisher’s location and ownership – as if their own convoluted, tax-efficient ownership and location arrangements were transparent to all. They accused predatory publishers of duping young authors in the developing world (Kurt, 2018) when they themselves had done their best for decades to suppress emergent publishers in Africa (Mills et al., 2021). They lambasted predatory publishers for negligence in archiving the scientific record when all the while their own manipulation of metrics was distorting the same scientific record (Verkhratsky and Petersen, 2023). Established publishers claimed that predatory journals would publish just about anything, but then so would many established journals. When John Bohannon (2013) submitted his spoof paper on the use of lichen to cure cancer, it was as acceptable to the journals of established publishers as to those of predatory ones. Coming from some of the world’s wealthiest organisations, grown fat at least in part from exploiting the free labour of academics, the accusation from established publishers that predatory publishers were in it only for the money was less than convincing:

. . . if the essential features of predatory publishing are unethical behavior and a focus on profit, then many publishers using the subscription model are even bigger predators than some new OA publishers. (Krawczyk and Kulczycki, 2021: 8)

The mire of metric manipulation that academic publishing had become blurred the distinction between established and predatory publishers. One unhelpful checklist of no fewer than 108 tell-tale signs of predation produced by Cabells, a publishing intelligence company (see Anderson, 2017), includes many indicators equally applicable to established publishers. Basically, a predatory publisher was simply any publisher that threatened the entrenched position of established publishers (Houghton, 2022). Interest in what made a publisher ‘predatory’ and in which publishers were predatory and which were not stirred new curiosity in academic publishing as a whole. Could there possibly be an alternative to what established publishers had been offering for decades? The APC made academics the customers of publishers and academics began to think about shopping around for a better publishing deal. The struggle of predatory publishers to justify their publishing activities and of established publishers to defend theirs revealed more about the industry than either sort of publisher would have wanted to disclose. Beall’s efforts as an individual to defend the public interest seemed to suggest that the public interest was not a burning concern of any of the parties in academic publishing. All put private benefit before public.

The demand for the services of predatory publishers proved insatiable and was certainly not limited to the naïve young academics identified as at risk. Authors – even the most senior - published for the metrics and were glad to patronise journals that supplied them quickly, reliably and cheaply (Pyne, 2017). Even large pharmaceutical companies, looking for rapid publication, could prefer predatory outlets to publication with established publishers (Deprez and Chen, 2017). The relationship of established publishers with pharmaceutical companies was already tainted: their journals had disguised drug advertising as academic content (Friedberg et al., 1999) and had taken cash from pharmaceutical companies to report favourably on their drugs (Lundh et al., 2017). But then established publishers had resorted to peer review to excuse even this egregious behaviour:

. . . what’s wrong with publishing an industry-funded editorial or review article as long as it gets appropriate peer review? (Elliott, 2004: 21)

There seemed to be no clear distinction between the APC extorted from authors by established publishers for open access and the pay-to-publish fee of predatory publishers (Brockington et al., 2023):

Are the real predators those that charge excessive APCs or is this difference in price considered a premium for publishing in a ‘quality’ journal via impact factors that can be manipulated by the so called ‘legitimate’ journals? (Teixeira da Silva et al., 2019)

The conversion of established publishers from a subscription model to pure open access has been delayed by their consideration of the strategic possibilities of a hybrid model. Subscription could mean waiting years for Clarivate to recalculate a journal’s JIF, which would eventually be reflected in the subscription price: on the other hand, open access links the APC directly to the JIF (which could always be manipulated: the higher the JIF, the higher the APC). So, for example, the Journal of Inflammation, with a JIF of 4.4, had an APC of US$3,290 in 2012 according to Solomon and Björk (2012) while Nature, also from Springer, with a JIF of 50.5, had an APC of US$12,290. There is no relationship to the actual cost of publishing a paper (Khoo, 2019), which Grossman and Brembs (2021) put at about US$400. The hybrid model presented authors with a choice: the subscription route (a second class service, free to the author, but with slow publication) or open access (a first class service offering fast publication and the likelihood of more citation, but charging an APC).

Far from distinguishing established publishers from predatory publishers, open access made clear just how much they might have in common:

New innovative publishers repositioned themselves as service providers to the authors, publishing with them, rather than seeing themselves as content providers to readers. (Shen and Björk, 2015)

Though some established publishers clung to the fanciful notion that academic endeavour was sacrosanct and that they were the conduit through which knowledge passed from the cloister to the outside world, APCs made a mockery of the ideal. Given that the income from APCs is a function of the number of papers published rather than the marginal cost of publishing them, it made commercial sense to publish as many papers as possible (Ioannidis et al., 2023). This is the strategy that attracted predatory publishers to academic publishing, and established publishers could not help but see its advantages, demanding that their journal editors increase production, sometimes both massively and instantly. Wiley, for example, pressured its editors to accept many more papers (whatever their quality) on threat of instant dismissal (Weinberg, 2023):

. . . Wiley told [the editor of the Journal of Political Philosophy] that in order to improve its performance to the publisher’s satisfaction, it needed to accept 35 articles within 60 days. Historically the journal has published between 12 to 16 articles per year . . . (Pettit, 2023)

The tactic has forced the resignation of many an editorial board, unable to accept the predictable decline in scholarly standards (e.g. Kincaid, 2024; Upton, 2023).

Some of the more enterprising publishers, such as Frontiers and MDPI, publish mega-journals (Mills, 2023) with hundreds of papers in each issue. The International Journal of Molecular Sciences, one of the many journals published by MDPI, was inviting submissions to more than 3,000 special issues in 2025. Its APC for publishing a single paper was £2,600 (Sample, 2025). Springer’s Cureus and Elsevier’s Heliyon, unburdened by any disciplinary focus, publish thousands – 17,000 in 2024 in the case of Heliyon (Orrall, 2026). To accommodate such mass production, peer review, that shibboleth of the academic publisher, could be sacrificed after all. Rather than decide whether a submission would make a contribution to the scientific record, a new ‘peer review lite’ would determine only whether it was technically sound, not whether it was of any importance (MacKinney, 2023). While the old peer review might take months, even once referees were found, the new version could be completed in days, even minutes.

The commercial imperative to go one step further and combine publishing forces proved irresistible, established publishers lending the predatory a degree of lingering respectability and predatory publishers sharing their invaluable experience of mass production in new markets. The saga of Wiley’s takeover of the Egyptian publisher Hindawi in 2021 for nearly US$300 million suggests that more worldly experience is certainly needed. Wiley’s press release boasts that Hindawi was a hugely profitable leader in open access publishing (Wiley, 2021). Perhaps, but the retraction of thousands of papers produced by paper mills and scarcely refereed was required to cleanse the Hindawi stable (Van Noorden, 2023), an exercise that cost Wiley dear (McMurray, 2023). Wiley may not be alone in its innocence: it is not at all unusual for discussion of paper mills in academic publishing to be accompanied by pictures of paper mills producing not academic papers for needy academics, but huge rolls of paper for the commercial paper market (e.g. Abalkina and Christopher, 2025; Soriano and Ruano-Ravina, 2025). Wiley was sufficiently disingenuous to place blame for this catastrophe squarely on Hindawi and its publishing ethics (Flintoft et al., 2023):

The [Wiley] report presents Hindawi as a victim of an ‘academic culture of “publish or perish”, which has incentivized unethical behaviour’. What it omits is the influence of the commercial publisher culture of greed which aims for massive growth in the number of published papers, with associated growth in profits. (Dorothy Bishop as quoted in Kincaid, 2023c)

Wiley announced in December 2023 that it would no longer be using the Hindawi name (Kincaid, 2023b). Hindawi had been suspected – including by Beall – of being a predatory publisher for more than a decade.

Prospects for academic publishing – Glimmering hope

So, where do we go from here? Exposure of both publishers and academics to predatory publishing has swept away much of the indolent ignorance that once pervaded academic publishing. It is no longer the case that academics have little idea of what happens to their papers after peer review, and that publishers know nothing of what goes into writing a paper. But enlightenment has had consequences for just about all the industry’s players. By far the most serious stems from the commercial invasion of academic territory. For example, editors were once sovereign in the selection of appropriate referees (Bell et al., 2022); now the task can be handled, after a fashion, by the publishers’ automated manuscript submission systems. Once editors edited papers before their publication: now publishers sell authors editing services. Springer, for instance, offers ‘Author services that help you do your best work’ (whatever that means). ⁵ Sage language services will ‘suggest alternate [sic] phrasing when poor wording is used’. ⁶ Publishers are increasingly anxious to develop extracurricular relationships with authors (‘Where are you in your next publication journey?’), recommending papers (‘based on your reading history’) – all from the publisher’s own journals, of course. Where once learned societies organised academic conferences, publishers (predatory and established alike) now schmooze in Maxwell fashion in order to hook paying authors (Heasman, 2019). Long gone are the days when the author’s only direct encounter with a publisher was over a trestle table at the back of a conference hall.

Even the insistence of academic publishers on manuscripts conforming to specific formats – all different, all imposing hours of frustration on authors, all quite pointless – is disappearing. Now these same publishers make a selling point of easy submission. Elsevier, for instance, boasts of its ‘Your Paper, Your Way’ scheme (Fennell and Barrie, 2014). Publishers will even re-submit rejected submissions – un-reformatted – to another and then another of their own journals, automatically cascading submissions from the highest JIF (with the highest APC) to the next highest. Springer, for example, publishes Nature, but also some 40 other Nature titles that feed off Nature’s scraps (Khelfaoui and Gingras, 2020).

But increasing knowledge of other parts of academic publishing has not necessarily led to greater understanding. The problem is that views of what goes on in academic publishing are actually views of what decades of manipulation have produced precisely to be seen. There is now a real danger of reality being obscured by indicators of this reality. Indicators are ubiquitous in academic publishing and may be furiously defended by interest groups that have come to depend on them. Trubnikov and Trubnikova (2023), for example, accuse ‘leading citation databases and global ranking organisations’ of benefitting from the manipulation of corrupt universities and governments (see also Chaudhary, 2026; Gadd, 2025; Mohan, 2025).

Actually, producing papers has become a back-office activity hidden from sight in an industry geared to generating metrics, no longer simply for academics to meet a university target, but for the information companies that now own the largest academic publishers to analyse and sell. Academic publishers, already hugely profitable, have found they can make greater profits still by selling information (Tennant, 2018). The paper itself has become a contribution less to knowledge than to meta-data (Biagioli, 2020).

Over the course of seventy years, academic journal articles have become commodities, and their meta-data a further source of revenue. (Mills, 2023)

Publishers, universities and academics look to metrics to understand each others’ activities, forgetting that they have all been mercilessly manipulating the metrics for decades (see Hallonsten, 2021). There is doubt about the accuracy of even academic publishing’s two major databases, Clarivate’s Web of Science (WoS), and Elsevier’s Scopus (Cram and Docampo, 2014; van Eck and Waltman, 2019). Many predatory journals are listed among their established academic journals (Abalkina, 2023; de Lange, 2023; Kulczycki, 2023):

Scopus (and WoS) should certainly put more effort into developing suitable tools for improving their data accuracy . . . at least as much as the effort they usually put in marketing campaigns. (Franceschini et al., 2016: 182)

Clarivate has just trimmed its annual list of highly cited researchers, its original total of 6,849 being reduced by the removal of 1000 fraudulent highly cited researchers – but only after pressure from Retraction Watch, a whistleblowing publication (Jack, 2022).

Elsevier plans to exploit what it calls ‘Scopus AI’, using artificial intelligence to provide summaries of papers in 27,000 of its journals to determine which are, according to its press release, ‘foundational and influential papers’ (Grove, 2024). As the most bland and undistinguished papers tend to be the easiest to cite, Scopus AI metrics may well pronounce these the most foundational and influential. Scopus AI will also identify star authors for Elsevier. It may be spoilt for choice among the hyper-authors who produce several papers a week. These include a psychologist at Nottingham Trent University who has accrued 50,000 citations in just 5 years in this way (Grove, 2020). Scopus AI may also find that a good few of its top authors have had nothing to do with producing the papers bearing their names. Author affiliations are another target of manipulation: many are fictitious, intended to impress an editorial team or to boost affiliated universities in international rankings (Biagioli, 2019, 2020). The existing Scopus list of prolific authors includes quite a few who were publishing decades before they were born (Abduh, 2024; Dillman, 2008). Some (thinking of Labbé’s Ike Antkare) have never lived at all (Labbé, 2020), though they continue to publish (Antkare, 2020). Clarivate reveals that Princeton and UCLA have been ousted from the top of the highly cited mathematicians list (Catanzaro, 2024) by universities in China and Saudi Arabia, some without even a Mathematics department. According to rankings published in US News & World Report, King Abdulaziz University in Jeddah has already overtaken Cambridge University in Mathematics (Biagioli, 2020). Ranking remains as compulsive as ever and ranking tempts gaming (Smaldino and McElreath, 2016). The recent loss of just one hyper-author brought the University of Cordoba down 150 places in the Shanghai university rankings (Ansede, 2023b).

But we are where we are, dealing with the legacies of Maxwell and Garfield and acceptance that citation is an indicator of quality in journal papers and gaming a valid way of increasing citation and therefore quality. Did anyone really believe this? Not even Garfield in the end. As long as indicators were inherent in the business models of universities and publishers, it was dangerous to question whether they made sense. Universities are full of people who will not tolerate ‘academics willing to speak truth to power, to be outspoken on issues of social justice and be committed to serving the public good’ (McKenna, 2024: 7).

. . . never before in the history of humanity have so many social scientists written so much with so little effect or of benefit to anyone. (Alvesson et al., 2017: 3)

The same might be said of other disciplines. It is a huge step to recognise existing indicators for what they are – manipulations of the academic publishing industry – and to replace them with reality. It is a step taken recently by the French CNRS (2025), the Sorbonne (Portala, 2025) and other academic institutions brave enough to face up to the powerful interests that would resist any transition, not least the academics who have been most successful at gaming their citations:

. . . what gets written, especially by more senior and successful scholars, becomes increasingly safe, almost mechanical, leading to publications . . . which do little to challenge the thinking of either the author or audience. (Spoelstra, 2024: 649)

Academics have rarely exercised their agency in academic publishing. They have been dutiful in supplying their employers with the publication metrics required, and have gamed enthusiastically to submit even better metrics. They have meekly offered up their manuscripts for publication at little cost to the publisher, and they have obediently supplied publishers with editing and refereeing services – for free, of course. It is hard to understand their cowardice and harder still to forgive it.

. . . behind even the big fancy journals there is essentially volunteer academics reading stuff between . . . driving the kids around and feeding the cat . . . There’s got to be a higher degree of respect for what we do. (Journal editor quoted in Luca et al., 2026)

But there are a few rays of hope: editorial boards resigning en masse because they are unwilling to fill their journals with mediocre papers to satisfy the greed of publishers suggests new stomach for questioning and resistance (e.g. Kincaid, 2025). In ‘Taking credit for stupidity’, Spoelstra (2024) observes that the academic must ask questions to find out what is going on. When indicators are allowed to substitute for knowledge, indicators flourish, and questioning to acquire knowledge falters. We are still a very long way from academics – and others in academic publishing – asking questions.