[written in Fall 2021]
Introduction
The goal of science is to understand the natural world, often in its native language of mathematics. The endeavour used to be led by isolated brilliant minds for several centuries, and often the theories of a few were accepted as gospel by several generations. This is not surprising, for the principles of Nature really are confusing to an untrained intelligence trying to grasp them through its limited senses. Any illumination on Nature's conspicuous riddles (Why do things fall? What are those shiny specks moving in the sky? Where and who are we? What is life? Why do some things rot?) must have appeared like the truth in those days; the typical philosopher of Nature was not yet used to the notion of verifying their cherished ideas with experiments. Then Galileo happened, who may be rightfully called the father of experimental science. Since his time (late 16th to early 17th century), any scientist worth the name had to keep in the forefront of their mind that Nature had a vote — the only one, in fact, that counted.
But in the middle of the 17th century another beautiful thing happened, a development almost as important as the deification of Experiment. The Royal Society (“The Royal Society of London for Improving Natural Knowledge”) was founded in London, and with it the first scientific journal, Philosophical Transactions of the Royal Society, to this day in circulation. This was no ordinary journal, for before you get your findings published you must endure the opinions and criticisms of fellow experts around Europe. The idea was that Science came before Individual, that human nature was unreliable in the pursuit of objective knowledge, and therefore one’s claims on any aspect of reality had to be articulated well and defended against scrutiny. The practice of peer review was born.
Today scientific research is a violent avalanche of discoveries following in the long wake of Isaac Newton (yes, even the “non-physics” pursuits like biology — the general guidelines for carrying out any scientific investigation were explicitly etched for posterity in Newton’s Prinicipia). And the workhorse of this enterprise is peer review. Broadly speaking, this is how it works. You conduct thorough research on a topic, write up your findings in classical prose (using equations and figures if you must) clear enough for others in your field to follow you, and send the manuscript to a journal, whereupon an editor will assign one or more of your peers around the world to, well, peer at it. As in my field of particle physics, the identity of the reviewer is usually a secret, and in some fields, to minimize bias, so is the author’s. The first style is called single-blind peer review and the latter, double-blind. The referee prepares a report and makes recommendations to the editor on the publishability of the work, and often referee and author exchange a few rounds of communication mediated by editor. This process not only ensures quality of scientific output, but also plays a vital role in the scientist’s career, whose academic worth is measured and tracked in their ability to get their work published in impactful peer-reviewed journals. It is typical to hear an academician say: “Publish or perish”.
The advent of pre-prints has altered the process somewhat. Since the early 1990s scientists of many fields have gotten into the practice of uploading an early version of their work on a website, arXiv (founded by a particle physicist), about a week before sending it to a journal. This has several advantages. For one, it gets your work to your readers much quicker than a journal does (where the peer review process could take months to conclude), often crucial for a rapidly moving field like mine. For another, it minimizes confusion on the question of priority — often scientists work on similar ideas and make similar discoveries around the same time, and pre-prints provide a clearer picture of who got there first. But the greater advantage in my view is that it serves as a grand peer review. Your work is open to attack from anybody in the world, not just from those who have access to specific journals behind paywalls. That puts added pressure on authors to scrutinize their study for holes, and revise and re-revise their manuscript until it is nearly unimprovable. In my field, for instance, people frown heavily upon significant differences between the arXiv and journal version of a paper: a carefully written paper would not give that much ammunition to a referee. Thus arXiv, via its panopticon-like nature, enforces good, publishable science. I can attest to this personally. I have thus far written 31 papers, and all of them have been published in leading journals of the field, due in large part to the terror of arXiv.
The system, like most things, is far from perfect, and you will often find long-embattled professors declaring no faith in it. But it is the best thing we have for ensuring that science ultimately meets the standards required of it. There is an amazing story to illustrate this. In 1936, twenty years after he had formulated the theory of general relativity — the spacetime picture of gravity — Albert Einstein co-wrote with Nathan Rosen a paper claiming that his theory does not accommodate the existence of gravitational waves, i.e. propagating ripples of space. Their submission to Physical Review came back with an extensive calculation by the anonymous referee coming to opposite conclusions, that gravitational waves must exist. The editor asked the authors to revise the manuscript. Emperor Einstein flew into a rage and shot back, “I see no reason to address the, in any case erroneous, opinion expressed by your referee”, and vowed never to submit to the journal again. Shortly after, physicist Leopold Infeld, suspecting that the referee might actually be right, consulted with cosmologist H. P. Robertson, who was able to pinpoint Einstein’s mistake. When Infeld communicated this to Einsten, he reversed his stance, revised the paper, and got it published in a different journal. What nobody knew at that time — and was only revealed by Physical Review after the deaths of all parties — was that the referee was H. P. Robertson. And of course, gravitational waves do exist. They were observed in 2015 by the ultra-sensitive behemoth detectors of the LIGO experiment in USA, an achievement that garnered its designers the Nobel Prize the following year. (This occasioned a couple of interesting coincidences in my life.)
In this piece1 I wish to share some anecdotes of my own experience participating in the process as both author and referee. By so doing I hope to have entertained all with the unusual turns these stories take, and to have given my non-STEM friends a flavour of life in science. Before I do that, let me set the stage a bit by recalling an episode from mythology showing how peer review could go wrong, spectacularly wrong. I allude of course to the picture above from Thiruvilayadal, depicting the referrral of Lord Shiva by the poet Nakkeerar, overseen by King Shenbagapandian, who plays editor. To briefly recollect the relevant bits: the king puts to the public a question for which he wants an answer in verse. Shiva submits one via proxy, which satisfies the king, but not court poet Nakkeerar, who spots an error in the poem. Upon hearing this Shiva storms into the court in human disguise thundering for an explanation, upon which Nakkeerar, admitting that he was the referee, asserts that the content of the answer was factually incorrect. To put him in his place, Shiva then reveals himself by opening his Divine Eye #3, and we get this immortal line from Nakkeerar: “Nice, nice, but a fail is a fail.” At this point Shiva resorts to insulting Nakkeerar’s family trade. Not one to back down, Nakkeerar retorts that his clan at least does not beg, unlike He who depends on offerings. Unable to take it any more, the god annihilates the poet.
So what went wrong? The root of the problem was that the reviewer’s identity was compromised. Had the king been a better moderator/editor, he would have decreed that the process be double-blind, and that all arguments to the other side go through him first, ensuring civil dispute minus ad hominem attacks. Sure, Shiva would have flown off the handle at first, but instead of eliminating his dissident would been compelled to provide empirical evidence to support his stance. And even better if there had been an appeal process whereby Shiva, if left dissatisfied, could have asked for a new expert to weigh in on the debate. Imagine how events may have unfolded had the identity of Physical Review’s referee been revealed to Einstein. By all accounts he would not have as readily admitted his mistake and rewritten his paper when Infeld conveyed to him the knowledge learnt from H. P. Robertson.
Epilogue
Let me leave you with another Nakkeerar snippet, this time from the movie Naan Petra Selvam. In this version, Sivaji Ganesan plays both Lord Shiva and Nakkeerar. And from that we may interpret a lesson: you are your own best referee. In the words of my hero Richard Feynman, “Nature cannot be fooled,” or as he put it more effectively: “The first principle is that you must not fool yourself, and you are the easiest person to fool.”
