Peer review as a formal, routine part of scientific publishing is surprisingly recent, only becoming standard practice after World War II. But its roots stretch back to 1665, when the first scientific journal launched in London. The gap between those two dates is filled with a gradual, messy evolution from one editor’s personal judgment to the structured system researchers use today.
The First Scientific Journal in 1665
On March 6, 1665, Henry Oldenburg published the first issue of Philosophical Transactions of the Royal Society, the world’s first scientific journal. Oldenburg was the Secretary of the Royal Society of London, and the early issues consisted of edited letters from his correspondence, book summaries, and accounts of experiments from natural philosophers across the UK and mainland Europe.
For a long time, historians credited Oldenburg with inventing peer review, pointing to letters in which he sought expert opinions on submitted work. But when historians like Adrian Johns and Aileen Fyfe examined the evidence more carefully, they found that Oldenburg didn’t actually use anything resembling a modern review system. He maintained such tight personal control over the journal that contemporaries sometimes called him its “author” rather than its editor. Although members of the Royal Society were supposedly tasked with reviewing the journal’s contents, in practice Oldenburg decided what got published based largely on what he found interesting or important.
This pattern held across early scientific publishing more broadly. Journal editors published work they personally deemed worthy, often from members of the learned society associated with the journal. The assessment was inherently based on reputation: if the person submitting the work was considered an expert, that was often enough.
Early Steps Toward External Review
The first recognizable move toward organized peer review came in 1731, when the Royal Society of Edinburgh began publishing Medical Essays and Observations. That publication established principles of review that historians point to as the earliest structured attempt to evaluate submitted work before publication. It predates the widespread adoption of external review by roughly two centuries, making it more of an outlier than a trendsetter.
Throughout the 1800s, biomedical journals operated largely as personal platforms for their editors, following the model of general journalism. Editors saw themselves primarily as educators. Some journals did occasionally send individual articles to outside experts for opinions, a practice that could have started as early as the mid-19th century. But this was casual and inconsistent, not a systematic requirement.
Why World War II Changed Everything
The real turning point came after 1945. The end of the war triggered a massive expansion of scientific research, fueled by government funding agencies that poured money into universities and laboratories. The volume of research papers surged, and journals that once had excess page space suddenly couldn’t publish everything submitted to them. Editors needed a way to be more selective.
By this point, most reputable journals had adopted some form of peer review. But there was also a practical barrier that had limited the process for decades: copying manuscripts was difficult and expensive. Editors couldn’t easily share a paper with multiple outside reviewers without risking the only copy.
That changed in 1959 with the commercial photocopier. For the first time, editors could make multiple copies of a submitted article and mail them to external reviewers without fear of losing the original. This simple technological shift made large-scale peer review feasible. Over the following decade, as competition for journal space intensified, peer review transformed from an occasional editorial tool into a rigorous gatekeeping system designed to assess the quality of every submission.
Even Major Journals Were Late Adopters
One detail that surprises most people: Nature, one of the world’s most prestigious scientific journals, did not require peer review for all its published research articles until 1973. Before that, editors could and did publish papers based on their own judgment. This was not unusual. The shift to mandatory external review happened journal by journal over the course of the 1960s and 1970s, not all at once.
The version of peer review that researchers know today, where two or three anonymous experts evaluate a manuscript and recommend whether it should be published, only became the dominant model in the mid-1970s. That means the system governing how science gets validated is roughly 50 years old in its current form, despite the first scientific journal being over 350 years old.
Variations and Challenges Since the 1970s
Once peer review became standard, debates about how to do it better followed almost immediately. The traditional model uses single-blind review, where the reviewers know who wrote the paper but the authors don’t know who reviewed it. Double-blind review, where both sides are anonymous, emerged as an alternative meant to reduce bias based on an author’s reputation or institution.
The rise of the internet introduced entirely new models. In 1991, the preprint server arXiv launched, allowing physicists (and later researchers in mathematics, computer science, and other fields) to share papers publicly before any formal review. ArXiv now receives about 20,000 submissions per month. The papers are not peer reviewed, though volunteer moderators screen for plagiarism and nonscientific content. Preprint servers haven’t replaced peer review, but they’ve created a parallel track where findings circulate and receive informal scrutiny from the scientific community before, or sometimes instead of, going through traditional journal review.
Open peer review, where reviewer identities and comments are made public, has gained traction at some journals as a way to increase transparency and accountability. These experiments reflect an ongoing tension: the peer review system is both relatively young and already under pressure to evolve again.