A potted history of peer review - Part 1

Peer review is a staple of science, a seemingly inescapable barrier that every researcher must cross in order to share their work with the world. The internet is awash with complaints about Reviewer #2 and long, drawn out review processes. Despite this, most academics believe peer review to be a gold standard process that protects the scientific record from poor quality studies. 

This history is important to understand as it informs much of the current discourse around scholarly communication, peer review and trust in research. Despite this importance, there is a noticeable absence of training and teaching around how peer review actually works and what it does and does not achieve. These two posts will explore the history of peer review and discuss what the future may hold.

The historical context of peer review; ancient times - 19th century

For the majority of the time that science has existed as a pursuit, results have been shared without formal peer review. Most scholars believe in the myth that peer review began in 1665 with the Royal Society in the UK. However, there is a pre-history to peer review that dates back much further. 

Some of the earliest examples of scholars critiquing each other's work dates back to ancient Greece and the medieval Islamic world. In the 5th century BC, scholars in ancient Greece would publicly debate work in informal settings such as public forums. Later evidence from ~900 AD further demonstrates the concept of peer review in excerpts from “Ethics of the physician” by a Syrian author which states “The notes of the physician were examined by a local council of physicians, who would adjudicate as to whether the physician had performed according to the standards that then prevailed”. 

In many ways, this ancient concept of review is still in practice today, with many scholars informally sharing their manuscripts amongst colleagues and collaborators prior to submission to a journal. 

In 1665 the Royal Society launched the Philosophical Transactions of the Royal Society, and thus the scientific journal was born. The editor of this new journal, Henry Oldenburg is often mis-credited with being the first to undertake peer review. However, Henry did not generally use any external refereeing system and largely relied on his own judgement. Slightly later in 1731, other journals such as Medical Observations and Essays, “distributed [memoirs] according to the subject matter to those members who are most versed in these matters”. This remained an internal review process meant to safeguard the reputation of the society affiliated with the journal. These forms of early peer review were designed to aid editors in selecting articles to publish, rather than authenticating findings. 

Throughout the 18th century and into the 19th century, essays, journal articles, monographs, pamphlets and public presentations were all regular methods of disseminating scientific knowledge, with few exposing the contents to any form of external refereeing. 

The birth of peer review; a struggling idea

London, England, 1831. There is a Cholera outbreak and the new London Bridge is officially opened. The Royal Society is already a well established scientific institution and one of its Fellows, William Whewell, proposes a new system of refereeing. Whewell suggests that articles submitted to the Philosophical Transactions journal should be sent to 2 Fellows who would write a report about the article. Both the article and the report could be published in a new journal (Proceedings of the Royal Society of London) with the reports potentially being “more interesting than the memoirs themselves” and thus a great source of publicity for science. Interestingly, this very beginning suffered from a problem that continues to plague peer review today; the first two referees (one of whom was William himself) disagreed about the claims of the manuscript they were reviewing and struggled to draft a report they were both willing to sign - yes these early reports were signed by the reviewers. Ultimately, the Proceedings journal rarely printed any referee reports. However, Philosophical Transactions did adopt this external refereeing system but the reports became confidential and were not shared with the original manuscript authors. These reports were used internally to help the society decide what to print. It would take the 20th century before journals began to send authors the referee reports. 

“[reviewer reports may] often [be] more interesting than the memoirs themselves and thus a great source of publicity for science”

In the 1840s and 1850s, the role of the referee was to advise the editor. However, as the number of journals and scientific articles expanded, concerns about the quality of the literature began to change the role of the referee from an advisor to a gatekeeper, who was responsible for the quality of the scientific literature as a whole. 

By the 20th century, researchers were expected to publish their findings in a good, reputable, journal. However, this did not mean that the journal needed to use any refereeing system, indeed many of the most “reputable” journals did not. For most of this time, refereeing remained a primarily British and American practice.

To be continued...