Academic publishing

Based on Wikipedia: Academic publishing

The Business Where Authors Pay and Reviewers Work for Free

Here is one of the strangest business models in the modern economy: academics conduct research, often funded by taxpayers. They write up their findings and submit them to journals. Other academics review the work—without pay. Then the journal, typically owned by a for-profit corporation, publishes the article and charges universities enormous subscription fees to read it. The universities, of course, are also funded largely by taxpayers.

Everyone in this system works for free except the publishers. And those publishers enjoy profit margins around forty percent—making academic publishing one of the most profitable industries in the world.

How did we end up here? And why does it matter so much for how knowledge spreads through society?

The Unlikely Origins of Scientific Journals

The story begins in Paris on January 5, 1665. A twelve-page pamphlet called the Journal des sçavans appeared on that Monday, becoming the first academic journal published in Europe. Two months later, on March 6, the Philosophical Transactions of the Royal Society followed in London.

Publishing your research was controversial back then—even ridiculous to many. Scientists had good reason for secrecy. When you announce a discovery, you risk someone else claiming they thought of it first.

Isaac Newton and Gottfried Wilhelm Leibniz both used an ingenious workaround: they would publish announcements as anagrams or coded messages. This established a timestamp for their discovery while keeping the actual content secret from competitors. Clever, but it created chaos. The sociologist Robert K. Merton studied disputes over scientific priority and found that ninety-two percent of simultaneous discoveries in the seventeenth century ended in bitter conflict.

The Royal Society championed a different approach. They believed science could only advance through "transparent and open exchange of ideas backed by experimental evidence." It was a radical notion. But it worked.

As publishing became standard practice, disputes dropped steadily. By the eighteenth century, seventy-two percent of simultaneous discoveries sparked conflicts. By the late nineteenth century, fifty-nine percent. By the first half of the twentieth century, just thirty-three percent. Researchers estimate that around fifty million journal articles have been published since those first pamphlets appeared in 1665.

What Peer Review Actually Means

When scientists talk about peer review, they mean something specific: before a paper gets published, other experts in the same field read it critically. They check whether the methods make sense, whether the conclusions follow from the data, whether the work adds something new to existing knowledge.

This sounds obvious now. It was not always the standard.

Early journals operated in various ways. Some were essentially one-person operations where an editor decided what to print, often just publishing excerpts from letters sent by colleagues. Others used group decision-making that resembled modern peer review. But peer review as we know it—where anonymous experts systematically evaluate submissions—did not become standard practice until the middle of the twentieth century.

The Royal Society of London formalized routine peer review in 1752 when it took over official responsibility for Philosophical Transactions. Before that, there were scattered earlier examples, but nothing systematic.

Today, the process works roughly like this: An author submits a manuscript to a journal. The editor sends it to two or three reviewers who work in the same field. These reviewers read the paper carefully and send back comments—sometimes supportive, sometimes devastating. The author revises the paper in response. This cycle might repeat several times before the paper is finally accepted or rejected.

The whole process typically takes months. For popular journals, there is often an additional delay of many more months before an accepted paper actually appears in print, simply because more papers are accepted than can fit in each issue.

The Different Flavors of Anonymity

Peer review comes in several varieties, distinguished mainly by who knows whose identity.

In single-blind peer review, the reviewers know who wrote the paper, but the authors do not know who reviewed it. This is the most common system. In double-blind peer review, neither side knows the other's identity—the reviewers do not see the authors' names, and vice versa. The idea is to prevent bias based on reputation or institutional prestige.

Then there is open peer review, where everyone's identity is known. Some journals even publish the reviews alongside the final paper, so readers can see the entire discussion that shaped the work.

Each approach has tradeoffs. Single-blind review is simpler to administer but may let reviewers be harsher than they would be if their names were attached. Double-blind review sounds fairer, but in specialized fields, reviewers can often guess who wrote a paper anyway—there might only be a handful of people in the world working on that particular problem. Open review promotes accountability but might make reviewers hesitant to criticize powerful figures in their field.

The Uncomfortable Truth About Quality Control

Peer review is supposed to be the gatekeeper of quality. And it does catch a lot of poor work. Journal editors generally agree the system is essential for filtering out research that does not meet basic standards.

But here is the uncomfortable truth: peer review regularly fails to recognize important work.

Consider Edward Jenner's report on the first vaccination against smallpox in 1796. The editor of Philosophical Transactions rejected it. This was arguably one of the most important medical discoveries in human history—the technique that would eventually eradicate smallpox entirely—and peer review initially said no.

Jenner is not an isolated case. The history of science is full of landmark papers that bounced from journal to journal before finally finding a home. Peer reviewers are human. They have biases. They tend to favor papers that confirm their existing beliefs and to be skeptical of work that challenges the conventional wisdom. Researchers call this confirmatory bias, and experimental studies have documented that it really does affect peer review.

The most prestigious journals reject around ninety to ninety-five percent of submissions. The American Psychological Association's journals range from thirty-five percent to eighty-five percent rejection rates. High rejection rates are seen as a marker of quality—but they also mean a lot of good work gets turned away for reasons that may have little to do with scientific merit.

The Corporate Takeover

For most of its history, academic publishing was run by the academics themselves. Scientific societies and universities operated journals as a service to their fields. This changed dramatically in the 1960s and 1970s.

Commercial publishers began systematically acquiring what they called "top-quality" journals—the most respected publications in each field. Once they owned these journals, they raised subscription prices significantly. Libraries kept subscribing anyway. Researchers needed access to publish and to stay current in their fields. Demand was, in economic terms, highly inelastic: price increases did not reduce the quantity demanded.

By 2013, five for-profit companies controlled half of all academic articles published. These were Reed Elsevier (now just Elsevier), Springer Science+Business Media (which has since merged to become the even larger Springer Nature), Wiley-Blackwell, Taylor and Francis, and SAGE Publications. More than two thousand publishers exist, but these five dominated the market.

The profit margins are remarkable. Available data suggest these large publishers operate at around forty percent margins—extraordinary for any industry, and especially striking when you consider that their two most important inputs come essentially free.

Think about it. The articles themselves? Written by academics, usually on university salaries funded by tuition and grants. The peer review? Performed by other academics, unpaid. The publishers' contribution is editing, typesetting, printing, and web hosting. These are real services, but are they worth forty percent margins?

A 2005 analysis by Deutsche Bank put it bluntly: "We believe the publisher adds relatively little value to the publishing process... We are simply observing that if the process really were as complex, costly and value-added as the publishers protest that it is, 40% margins wouldn't be available."

The Serials Crisis

The consequences of this business model have been severe for universities and libraries.

From 1986 to 2005, total spending on journal subscriptions increased by 7.6 percent per year. But the number of journals libraries could afford grew by only 1.9 percent per year. Prices rose far faster than budgets. Researchers call this the serials crisis.

The effects rippled through academia in unexpected ways. As journals consumed ever-larger portions of library budgets, something had to give. It was books.

In 1986, libraries spent forty-four percent of their budgets on books and fifty-six percent on journals. By 1998, the ratio had shifted to twenty-eight percent for books and seventy-two percent for journals. This devastated the humanities, where books—not journal articles—are the primary form of scholarly communication. University presses, which publish most academic books, found their main customers unable to afford their products.

The timing was particularly cruel. Just as university presses lost their markets, tenure committees in the humanities increasingly expected candidates to have published a monograph—a full-length scholarly book. Scholars needed to publish books to keep their jobs, but publishers could not sell books because libraries could not buy them.

The crisis continued into the 2010s. A 2009 survey found that thirty-six percent of libraries in the United Kingdom had their budgets cut by ten percent or more. Libraries began fighting back, using the leverage of open access alternatives to negotiate better deals. Some institutions simply canceled their subscriptions entirely, finding that most of the content they needed was available through other means.

The Open Access Revolution

Starting in the early 1990s, a counter-movement emerged. What if research could be freely available to anyone with an internet connection?

This is open access publishing. In its purest form, the author publishes in a journal that makes the article immediately and permanently free online. No subscription required. No paywall. Just knowledge, available to whoever needs it.

The appeal is obvious. Taxpayers fund much of the research. Should they not be able to read the results without paying again? A physician in a developing country, with limited institutional resources, might desperately need access to medical research. A curious citizen might want to understand the science behind a policy debate. A researcher at a small college might lack the expensive subscriptions that their colleagues at wealthy universities enjoy.

But open access creates a new problem: if readers do not pay, who does?

Many open access journals charge authors an article processing charge, sometimes several thousand dollars per paper. This shifts costs from the reader to the researcher—or more accurately, to whoever funds the research. Grant agencies and universities often pay these fees. The model works reasonably well for well-funded fields like biomedicine. It works less well for scholars without grants, including many in the humanities and social sciences.

Some journals have found other approaches. They operate on institutional subsidies, volunteer labor, or foundation support. The landscape is diverse and still evolving.

The Preprint Phenomenon

There is another route to open access that does not depend on journals at all: self-archiving.

Authors can post copies of their papers online themselves, either on personal websites or on dedicated servers. The most famous of these is arXiv, pronounced "archive," a repository started in 1991 for physics papers. Today it hosts preprints in physics, mathematics, computer science, quantitative biology, statistics, and other fields.

Some important mathematical results have been published only on arXiv, never appearing in traditional journals at all. The most celebrated example is Grigori Perelman's proof of the Poincaré conjecture, one of the most famous unsolved problems in mathematics. Perelman posted his proof to arXiv and never submitted it to a journal. He was awarded the Fields Medal—the highest honor in mathematics—but declined it.

Preprints are papers that have not yet gone through peer review. They represent work in progress, shared with the community for feedback and priority. For decades, physicists have circulated preprints as a normal part of their workflow. During the COVID-19 pandemic, preprints became suddenly important in medicine too.

The Pandemic Publishing Explosion

COVID-19 transformed academic publishing almost overnight.

The urgency was unprecedented. A new disease was spreading globally. Doctors needed information about treatments. Public health officials needed data on transmission. Vaccine developers needed every scrap of relevant research.

The traditional publishing timeline—months of peer review followed by months of production—was simply too slow. Preprint servers suddenly became essential. Researchers posted findings as quickly as they could generate them, even if the work had not been formally reviewed.

By August 2021, according to a Royal Society study, at least 210,000 new papers on COVID-19 had been published. Of the roughly 720,000 authors of these papers, nearly 270,000 came from just four countries: the United States, the United Kingdom, Italy, and Spain. The pandemic deepened existing inequalities in who gets to participate in global scientific conversations.

The preprint boom had costs. Without peer review, some low-quality or even fraudulent work circulated widely. Studies with methodological flaws influenced public debate before experts could identify the problems. But the speed of information sharing also accelerated vaccine development and identified effective treatments faster than traditional publishing could have managed.

The Predatory Publisher Problem

Where there is money, there are scams.

Predatory publishing refers to journals and publishers that exploit the academic publishing system for profit without providing the services that legitimate publishers offer. They accept article processing charges but do minimal peer review, little editing, and sometimes no quality control at all. They flood researchers' inboxes with spam invitations to submit papers and serve on editorial boards.

For researchers desperate to publish—whether for tenure, promotion, or visa requirements—predatory journals offer an easy path. Pay the fee, and your paper appears in something that looks like a journal. But the work is not truly vetted. It adds little to scientific knowledge. And it pollutes the literature with unreliable findings.

Distinguishing legitimate open access journals from predatory ones is genuinely difficult. The predators have become sophisticated, creating professional-looking websites and adopting names that sound similar to respected journals. Young researchers, unfamiliar with the landscape, are particularly vulnerable.

The Digital Transformation

Academic publishing is still in the middle of a massive transition from print to electronic formats.

Paper journals still exist, but almost all are now available electronically as well. Many newer journals publish only online—no paper edition at all. For subscribers, electronic versions are typically available the moment they are ready, sometimes even before the paper version. For non-subscribers, many journals now offer what is called delayed open access: after an embargo period ranging from a few months to two years, articles become freely available.

The shift has changed how licensing works. Since the early 1990s, libraries have increasingly licensed electronic access rather than purchasing physical copies. This creates peculiar situations where a library might pay for access year after year but own nothing. If the subscription lapses, the access disappears.

Meanwhile, academics have increasingly turned to informal channels. The economist Glenn Ellison reported in 2011 that the dramatic growth of opportunities to share results online had led to declining use of peer-reviewed journals in economics. Why wait months for peer review when you can post a working paper immediately and get feedback from the entire world?

Publication Bias and the File Drawer Problem

There is another systemic problem in academic publishing that has nothing to do with business models or access.

In the sciences, researchers want their results to be statistically significant—that is, unlikely to have occurred by chance. Journals want to publish interesting, conclusive findings. This creates a bias: studies that find a clear effect are more likely to be published than studies that find nothing.

This is called publication bias, and its consequences are profound. Imagine ten research teams independently testing whether a new drug works. By chance alone, one team might find a statistically significant result even if the drug is actually useless. That team publishes triumphantly. The other nine teams, with their boring null results, face rejection after rejection. Their papers end up in what researchers call the file drawer—unpublished, unseen, unknown.

The published literature then gives a distorted picture. The drug appears effective because the only published study found an effect. The nine studies showing no effect are invisible. Decisions get made based on incomplete evidence.

Some fields have tried to address this through preregistration: researchers publicly declare their hypothesis and methods before collecting data. This makes it harder to cherry-pick results after the fact. Some journals have committed to publishing well-designed studies regardless of whether they find the expected results. But publication bias remains pervasive.

The Many Types of Academic Papers

Not all academic papers are the same. Several distinct types serve different purposes.

Research papers report original findings—new experiments, new data, new analysis. This is what most people imagine when they think of scientific publications.

Review articles synthesize existing research on a topic. Rather than presenting new findings, they survey what is already known, identifying patterns and gaps in the literature. Good review articles are immensely valuable, especially for researchers entering a new field.

Case reports describe individual patients or situations in detail, particularly common in medicine. A single unusual case might not prove anything statistically but can suggest new hypotheses or alert practitioners to rare conditions.

Position papers argue for a particular viewpoint on a controversial topic. They are more opinion than data, but grounded in the author's expertise and engagement with the literature.

Concept papers introduce new theoretical frameworks or definitions. They shape how a field thinks about its fundamental questions.

Technical papers describe methods, tools, or techniques that other researchers might use. They advance the infrastructure of science rather than answering specific research questions.

In taxonomy—the science of classifying organisms—species papers formally describe new species, following strict conventions that date back centuries.

A Note on Law Reviews

Legal scholarship follows completely different rules.

In the United States, law reviews are run by law students, not by established scholars. Student editors select which papers to publish and edit them for publication. This is radically different from peer review in the sciences, where experts in the field evaluate the work.

The system is peculiar, even within legal academia. Critics argue that students lack the expertise to evaluate sophisticated legal scholarship. Defenders argue that the system prevents established scholars from excluding challenging new ideas. Whatever its merits, the law review model is unique—a reminder that academic publishing takes many forms across different disciplines.

The Production Process

Once a paper clears peer review, it enters production—the process of turning a manuscript into a published article.

A production editor or publisher takes over. The manuscript goes through copy editing, where editors ensure it conforms to the journal's house style. They check that all references are correct, that labels match their figures, that the text is consistent and clear. This work can involve substantial editing and back-and-forth with authors.

Editors at journals often call themselves manuscript editors rather than copy editors because their work overlaps with what authors' own editors might do. The distinction matters because at this stage, copyright typically transfers from the author to the publisher—a controversial practice that means researchers often cannot legally share their own work.

Next comes typesetting: arranging the text, figures, and equations into the journal's format. The author receives proofs to review and correct. Historically, this was labor-intensive work involving handwritten corrections transcribed manually onto clean versions. Today, electronic annotation has streamlined the process somewhat, though the basic cycle of proof, correction, and verification remains.

Finally, the article is assigned to a specific issue of the journal and prepared for both print and online publication.

Grey Literature and the Edges of Publishing

Not everything academics write gets formally published. The term grey literature refers to scholarly output that is printed or posted online but never goes through traditional publication channels.

Working papers, technical reports, conference presentations, dissertations before they become books, government documents—all of these might contain valuable research but exist outside the peer-reviewed literature. Grey literature is harder to find, harder to cite, and carries less prestige. But it often contains the most current work in a field, unconstrained by publication delays.

What the Future Might Hold

The crisis in academic publishing has no obvious resolution.

Several models are being explored. Open access continues to expand, though questions remain about funding and sustainability. Some advocate for community-oriented features—post-publication commentary, collaborative revision, dynamic updating of articles as knowledge advances. The rigid boundaries between draft, preprint, and final publication are blurring.

Online scientific interaction outside traditional journals is becoming more important. Researchers discuss findings on social media, post data and code to public repositories, share analyses through blogs and forums. The formal journal article may become just one node in a larger network of scholarly communication.

Some experts have proposed evaluating papers based on the significance and novelty of their findings rather than where they happen to be published. The current system ties prestige to journals—publishing in Nature or Science carries enormous weight regardless of the specific paper's quality. A more rational system might assess each paper on its own merits.

But reforming systems is hard, especially when careers depend on the existing rules. Tenure committees still count publications in prestigious journals. Grant agencies still use publication records to evaluate proposals. Individual researchers, acting rationally within the current system, perpetuate that system.

The pandemic showed that rapid change is possible when urgency demands it. Whether that energy translates into lasting reform remains to be seen. For now, academic publishing remains caught between its idealistic origins—the Royal Society's vision of transparent, open exchange—and the realities of a multi-billion dollar industry built on the unpaid labor of scholars themselves.