Science Under the Microscope Looks Infected
Surgical changes are coming in scientific institutions and methods, due to chronic maladies coming to light.
Warning: the following news items may be disturbing to those raised on scientism (the belief that science is the only objective way to generate knowledge). Others thinking the “scientific method” is an unchanging process may also be adversely affected by news of what is going on.
Tradition is like a hot bath; a person gets in, becomes accustomed to the temperature and touch, relaxes, and stops thinking. That’s good sometimes, but science is about thinking. Its rationale is to observe phenomena and think about causes that produce them. It’s about testing and sharing results. Scientists have been engaged in these traditions for centuries. The “scientific method” has been refined to a science itself: observe, make a hypothesis, test, publish in peer-reviewed journal. Why break tradition?
There is a science of science, called philosophy of science. Some philosophers of science realize that every aspect of the “scientific method” incurs bias. There are even disputes about what is meant by common scientific terms, such as law of nature, evidence, and explanation. But those are just quibbles about details, aren’t they? If they were, then we would not see headlines like this.
Study finds scientific reproducibility does not equate to scientific truth (Phys.org). What a shock! A key rationale for the superiority of scientific knowledge has been reproducibility. As we have reported for a few years now, though, many scientific results are not reproducible—especially in psychology and other social sciences.
Reproducible scientific results are not always true and true scientific results are not always reproducible, according to a mathematical model produced by University of Idaho researchers. Their study, which simulates the search for that scientific truth, will be published Wednesday, May 15, in the journal PLOS ONE.
Independent confirmation of scientific results—known as reproducibility—lends credibility to a researcher’s conclusion. But researchers have found the results of many well-known science experiments cannot be reproduced, an issue referred to as a “replication crisis.”
The article ends, “Insisting on reproducibility as the only criterion might have undesirable consequences for scientific progress.” What? Is that a post-hoc rationalization for the failure of reproducibility? Readers may wish to scour the open-access journal article to see the reasoning behind that! We don’t want to slow down the advance of science by too difficult a criterion, do we? Or is relaxing the reproducibility criterion a recipe for multiplying fake science? The play’s the thing. Onward to discover more and more things that may not be so!
Facing Plan S, publishers may set papers free (Science Magazine). A major revolution in scientific publishing is going on. The journals are running scared. Safe behind high-priced subscriptions, journals were able to filter what was deemed “legitimate” science. They could even control peer review. Now, the trend toward preprints and open access is accelerating. Preprints allow scientists to publish their ideas and findings online before peer review! Open access forces journals to make their articles visible to the public online for free. Did you know this?
Plan S, the funder-backed scheme to require free online access to scientific literature, aims to shake up the subscription journals that have long dominated scholarly publishing. Now, some publishers are considering an approach they hope will both comply with the plan and maintain their subscription income: allowing authors to post manuscripts in public archives as soon as their papers are published.
Paywalls may become a thing of the past. The justification for preprints and open access is that taxpayers who funded research should have access to it without having to pay a second time. Journal editors are scrambling to safeguard their incomes by justifying the “value added” that subscribers get. These trends are radically changing the old peer review model and, consequently, the very meaning of peer review itself. No more will good-old-boys clubs be able to censor a maverick’s ideas. No more will an anonymous peer reviewer be able to steal or block the ideas of his rival. The democratization of science is trending! —which, of course, calls in to doubt whether the old method was even superior to begin with.
Springer Nature journals unify their policy to encourage preprint sharing (Nature). Like the fable of the Senator noticing which way the crowd is moving, then running ahead of them and telling them he is their leader, the Springer-Nature publishing goliath is telling their rivals, “Hey, we’re all for this new trend!” In physics and astronomy, Cornell’s arXiv server has a long record of success in preprints. This past year, Cold Spring Harbor’s bioRxiv server started posting preprints for biologists. Noticing their success, Springer-Nature wants to become drum major for the bandwagon, showing the crowd they were for it all along!
For more than two decades, Nature and its sister journals have supported pre-publication sharing of manuscripts on preprint servers. Nature’s first editorial on this goes back to 1997 — although, back then, the practice was common only among physicists. By making early research findings accessible quickly and easily, preprints allow researchers to claim priority of discovery, receive community input and demonstrate evidence of progress for funders and others.
Recognizing these benefits, we are now pleased to announce an updated policy encouraging preprint sharing for Springer Nature journals.
The times, they are a-changing. On the sad side, people may become aware that science is not the rock-solid edifice of reliability its reputation has claimed. On the bright side, more people will be able to watch the sausage-making progress, taste the product, and discern who does it best.
Because science is mediated by fallible humans, it is inherently fallible. Someone quipped, “It is impossible to make anything foolproof, because fools are so ingenious.” Much science is reliable, and is as impartially done as humanly possible. That goes for non-controversial subjects that are observable, testable, and repeatable. Other research that grabs the “science” label and runs with it, especially unobservable, untestable and non-reproducible research about the past (like Darwinism), should come with an FDA warning, “The claims made in this paper may be worldview-biased and unreliable.” Caveat emptor.