Category Archives: Science

Normal Science and Abnormal Publishing

When the Large Hadron Collider locates its elusive quarry under the sofa cushion of the universe, Nature will be there to herald the news of the new particle and the scientists who found it. But below these headline-worthy discoveries, something fascinating is going on in science publishing: the race, prompted by the hugely successful PLoS ONE and inspired by the earlier revolution of arXiv.org, to provide open access outlets for any article that is technically sound, without trying to assess impact ahead of time. These outlets are growing rapidly and are likely to represent a significant percentage of published science in the years ahead.

Last week the former head of PLoS ONE announced a new company and a new journal, PeerJ, that takes the concept one step further, providing an all-you-can-publish buffet for a minimal lifetime fee. And this week saw the launch of Scholastica, which will publish a peer-reviewed article for a mere $10. (Scholastica is accepting articles in all fields, but I suspect it will be used mostly by scientists used to this model.) As stockbrokers would say, it looks like we’re going to test the market bottom.

Yet the economics of this publishing is far less interesting than its inherent philosophy. At a steering committee meeting of the Coalition for Networked Information, the always-shrewd Cliff Lynch summarized a critical mental shift that has occurred: “There’s been a capitulation on the question of importance.” Exactly. Two years ago I wrote about how “scholars have uses for archives that archivists cannot anticipate,” and these new science journals flip that equation from the past into the future: aside from rare and obvious discoveries (the 1%), we can’t tell what will be important in the future, so let’s publish as much as possible (the 99%) and let the community of scholars rather than editors figure that out for themselves.

Lynch noted that capitulation on importance allows for many other kinds of scientific research to come to the fore, such as studies that try to reproduce experiments to ensure their validity and work that fails to prove a scientist’s hypothesis (negative outcomes). When you think about it, traditional publishing encourages a constant stream of breakthroughs, when in reality actual breakthroughs are few and far between. Rather than trumpeting every article as important in a quest to be published, these new venues encourage scientists to publish more of what they find, and in a more honest way. Some of that research may in fact prove broadly important in a field, while other research might simply be helpful for its methodological rigor or underlying data.

As a historian of science, all of this reminds me of Thomas Kuhn’s conception of normal science. Kuhn is of course known for the “paradigm shift,” a notion that, much to Kuhn’s chagrin, has escaped the bounds of his philosophy of science into nearly every field of study (and frequently business seminars as well). But to have a paradigm shift you have to have a paradigm, and just as crucial as the shifting is the not-shifting. Kuhn called this “normal science,” and it represents most of scientific endeavor.

Kuhn famously described normal science as “mopping-up operations,” but that phrase was not meant to be disparaging. “Few people who are not actually practitioners of a mature science,” he wrote in The Structure of Scientific Revolutions, “realize how much mop-up work of this sort a paradigm leaves to be done or quite how fascinating such work can prove in the execution.” Scientists often spend years or decades fleshing out and refining theories, testing them anew, applying them to new evidence and to new areas of a field.

There is nothing wrong with normal science. Indeed, it can be good science. It’s just not often the science that makes headlines. And now it has found a good match in the realm of publishing.