In the hierarchy of scientific principles, the scientific consensus – collective opinion and judgement of scientific experts in a particular field – is an important method to separate real scientific thought from junk science, pseudoscience, cargo cult science, and other fake beliefs.
I often discuss scientific theories which “are large bodies of work that are a culmination or a composite of the products of many contributors over time and are substantiated by vast bodies of converging evidence. They unify and synchronize the scientific community’s view and approach to a particular scientific field.”
A scientific theory is not a wild and arbitrary guess, but it is built upon a foundation of scientific knowledge that itself is based on evidence accumulated from data that resulted from scientific experimentation. Generally, a scientific consensus eventually leads to a scientific theory.
I have written frequently about the scientific consensus, because it is one of the most powerful pieces of evidence in a discussion about critical issues of our day – evolution, climate change, vaccines, GMOs, and many other scientific fields.
Scientific consensus – collective opinion
Let me repeat myself. The scientific consensus is based on the consensus of experts in the field based on the evidence, almost exclusively from peer-reviewed and published data. Since science is not a democracy, there is no formal vote, although respected scientific societies will often publish a public statement on the scientific consensus for a particular issue.
It’s important to note that a consensus, or its more formal cousin, a theory, are not dogma. They aren’t incontrovertible. They aren’t irrefutable.
And, it’s not a debate. The only thing that matters to forming a consensus, or alternatively, demolishing said consensus, is evidence. And as I like to say, it’s the quantity and quality of evidence that matters.
In fact, one of the fundamental aspects of a scientific consensus is that those who disagree with it make up a minority of the scientists in a given field. Their disagreement only becomes critical if the breadth and depth of their evidence can help refute the consensus.
When Charles Darwin described the basics of evolution 150 years ago, he didn’t know about DNA and genes. Our current description of evolution is more detailed and includes DNA. The level of evidence that would be required to debunk evolution is so high, I cannot comprehend what it would take to challenge it.
The same with most other consensuses in science. Because I am openminded to information that might change the consensus, I can imagine a theory or consensus being overturned eventually. It’s just that the possibility is vanishingly small, because the evidence level that needs to be exceeded is so high, that it will require hundreds of published articles to begin the discussion.
How that collective opinion is formed
There are three criteria for an evidence based consensus, as distinguished from a typical political or group consensus (see Note 1):
- Consilience of Evidence – Many different fields of science all contribute to the understanding of major scientific principles, such as anthropogenic climate change–ranging across biology, geology, chemistry, and other natural sciences. Our acceptance of the safety and effectiveness of vaccines is supported by research in diverse fields like epidemiology, public health, microbiology, immunology, virology and many others. It’s not one subspecialty of science that builds the consensus, it’s several.By contrast, the support for homeopathy, as an example, is restricted to just advocates of that pseudoscience. They don’t published their findings in any outside journals because they’re lacking a plausible mechanism to explain inverse dose-response, and they are incapable of providing favorable results in double-blind randomized control trials. They can establish nothing more than a placebo effect.
In fact, since homeopathy would require a wholesale restructuring of our knowledge of quantum mechanics, chemistry, physics and other non-biological fields, the only way we would ever see consensus that homeopathy works would require broad support in numerous fields of science. And frankly, no one is on board, except the pseudoscience pushing crowd.
- Social calibration – The experts involved in the consensus agree on standards for evidence–and this standard is ridiculously high, which is why it is so powerful. Those in the pseudoscience world reject certain gold standards of clinical research, like double-blinded, randomized clinical trials, which places a large barrier to having their “alternative medicine” ever reaching a scientific consensus.Moreover, there is broad and powerful evidence that creates a scientific consensus, something that is lacking in pseudoscience. We have consensus that homeopathic remedies lack any active ingredients and have no plausible biological mechanism. So the consensus does work both ways, to support scientific knowledge that is evidence based, and reject beliefs that lack similarly powerful evidence.
- Social Diversity – Having researchers from many cultural and economic backgrounds provides diversity that helps eliminate social biases as a cause of error. For example, the published literature on the safety of GMOs has provided agreement from researchers in countries around the world from various cultural backgrounds.If the evidence of safety only came from middle-class white scientists, whose parents are farmers, and who live in St. Louis, Missouri only 5 blocks from Monsanto headquarters, we might rightfully suspect the consensus. But in general consensus is formed by the weight of evidence, and to get that weight requires research from nearly everywhere on the planet.
Evidence, all about the evidence
Let’s be clear, gathering evidence that supports any scientific idea isn’t easy. And this evidence isn’t derived from a Google search, with no critical analysis of the quality. A scientific consensus is formed by experts, who are substantially more knowledgeable than most of us in the field being discussed.
As Steven Novella, MD, writes in his NeuroLogica blog:
Let’s take vaccines as an example. To understand just a part of the science of vaccinations requires research into the fields of immunology, virology, epidemiology, infectious diseases, physiology, and possibly dozens of other areas of the biomedical sciences. It is simply impossible for one person to be an expert across all of those fields.
Thus, to develop just one vaccine, numerous experts, across a large number of biomedical disciplines, are involved. I have tremendous respect for Paul Offit – not only is his knowledge of vaccines based on his own education and research but also on the backs of research from hundreds of other researchers over a century (at least).
Evidence doesn’t come from a few hours on Google. Proper evidence is derived from researchers who spend years looking at small parts of the larger picture. And then there are hundreds or thousands of researchers working on their small part of the picture.
The fact of evolution doesn’t rely upon one article or book – in fact, Charles Darwin forms only one small part of the story of evolution. There are probably hundreds of thousands of published research that builds the story of evolution. Maybe you’ll find an occasional error here or there in the evidence for evolution, but it’s still overwhelmed by all of the other studies.
The same for the other big scientific issues of our time – climate change, GMOs, vaccines, and many others. The scientific consensus for anthropogenic climate change, the safety of GMOs, and the safety and effectiveness of vaccines is huge. These consensuses will not collapse with one error here or there. They will not collapse because some gadfly publishes a contrary article in a minor journal.
If there’s one thing that I wish I could impress upon the science deniers is that cherry picking one or ten articles that seem to contradict the consensus doesn’t do anything. If there is something that will cause the collapse of a scientific consensus, it will be research from many different locations, published in the highest quality journals, over a long period of time.
Because the consensus was formed by research from many different locations, published in the highest quality journals, over a long period of time.
Scientific consensus as appeal to authority
Sometimes, the scientific consensus is criticized as being a logical fallacy – an appeal to authority. However, it’s misused in this case.
Generally, an appeal to authority becomes a logical fallacy when a presumed authority makes an argument outside of the authority’s field of expertise, in a topic in which the authority is biased.
Clearly, one can appeal to authority if the evidence supports the claims of the authority.
For example, a biochemist can be an authority on evolution. In fact, I would expect nothing less. However, there are PhD level biochemists who are creationists – these are false authorities, because they have no or disputed evidence for their claims.
Climate change denialism is filled with false authorities. Over 97% of papers published about climate change support the fact of human caused climate change. If science were actually a democracy, this would be a landslide. Yet, climate change deniers rely upon false authorities and bad research to claim that there’s a “debate.” There’s no debate.
Not only are 97% of published papers in support of climate change, but also those papers are published by leading researchers in dozens of different fields. There are geologists, climatologists, archeologists, paleontologists, meteorologists, physicists, chemists, botanists, and dozens of other fields of science that publish tiny bits of evidence that form the whole of the consensus. Once again, like evolution, the evidence is the size of a literal mountain.
And these authority figures in all of these fields do not suddenly rise to prominence after a couple of hours Google searches. To become a competent immunologist, one of the fields that are basic to vaccine research, It takes years of education and research to understand, let alone describe, the nearly infinite complexity of the immune system.
To understand even the basics of immunology, one needs to grasp complex sciences like biochemistry, cell biology, physiology, and probably a half-dozen other fields of biology. Understanding the human immune system, an incredibly complex physiological response to the environment, isn’t something that can be grasped in a few hours.
Even after eight years of formal education immunology, it probably takes another decade of research, writing, presentation, criticism, and publication to be declared an expert. But even then, it’s all about the evidence.
There are actual false authorities in immunology, like Tetyana Obukhanych, who, on the surface, appear to be real experts or authorities, but they have no published evidence that supports their public opinions. That’s where the logical fallacy arises.
That’s why, as Dr. Novella states, we rely upon the expert opinion, though as skeptics, we still have to weigh the evidence, even if presented by experts. As I wrote in articles on skepticism and debunking pseudoscience, large amounts of high quality evidence must be used to form a consensus, or, alternatively, shift us away from the consensus.
What makes the scientific consensus so powerful, ideas that should be respected, is that the conclusions are crowd-sourced from many experts across many disciplines. Moreover, the conclusions are self-correcting as new evidence is provided. Opinions that aren’t included in the consensus rarely get supported not because of some conspiracy, but because the level of evidence has not reached the quality or quantity to be considered.
Criticisms of the scientific consensus
Generally, there are two broad condemnations of the scientific consensus – science has been wrong before, and corporate influence.
Let’s start with the false flags and tropes that try to claim that science gets it wrong all the time. But that’s just not true. Scientific evidence showed that smoking was harmful, despite the myth that “all” scientists thought that smoking was safe.
“Science” never claimed that world was flat. Or that the universe orbited around the earth. Or that there were canals on Mars. Or that thalidomide was safe. There was never a consensus about any of that, some of them, in fact, were religious beliefs, not scientific ones.
It’s ironic for me to say this, but a scientific consensus is provisional. Forming a consensus is difficult, but it can be overturned – with more and more powerful evidence.
Now, this doesn’t mean aspects of science don’t change over time. For example, I use evolution as an example of a fact – yet it’s constantly being “fine-tuned,” as new information is found. However, evolution itself is probably the most powerful consensus in science.
The second condemnation of the scientific consensus is corporate influence. This belief is so profoundly ridiculous, I find it difficult to even waste a nanosecond in refuting it. Thankfully, Dr. Novella saves me some time in his recent article:
Is that clear?
Let’s put these in numbers that you can hug and caress. The top 15 “Big Pharma” companies had a total revenue of slightly greater than US$500 billion in 2014. That sounds massive, doesn’t it? Exxon-Mobil has that level of revenue all by itself. And yet, despite a secret policy of climate change denial, it failed miserably to influence the scientific consensus on climate change. Influence from Big Pharma and Big Agra have even a lower chance of having any effect on scientific consensus.
Why is it so difficult for corporations to influence the scientific consensus? It goes back to how the the system works. The consensus is built on evidence, which is repeated over time, it is self correcting, and minority “opinions” (not evidence) lack the standing in the field to overwhelm the actual evidence supporting the consensus.
The scientific consensus is one of the cleanest examples of well-formed crowd sourcing!
Scientific consensus examples
In some cases there are independent scientific groups that proclaim the scientific consensus. The American Association for the Advancement of Sciences is an international non-profit organization that which promotes cooperation among scientists, to defend scientific freedom, encourages scientific responsibility, and supports scientific education and science outreach for the betterment of all humanity. It is the world’s largest and most prestigious general scientific society, and is the publisher of the well-known scientific journal Science. It occasionally makes statements on the scientific consensus.
There’s even a scientific consensus on vaccines, written by the the most prestigious scientific academy in the world, the National Academy of Sciences:
Despite the prestige of these institutions and the power of the evidence supporting the consensus, there will be a group that picks and chooses what they believe or don’t believe. It’s hysterical.
If you accept the consensus on climate change or evolution, but reject it for GMOs because each fit your pre-conceived view of the world, you are a science denier. Yes, I’m tough about this, because it shows the level of hypocrisy in these individuals.
You don’t get to accept the power of the consensus behind anthropogenic climate change but reject it for GMOs, just because you feel like it. The evidence for both is so huge, so overwhelming, your gut feeling or opinion is completely irrelevant.
I keep finding it ironic that GMO deniers and climate change deniers use the same exact “arguments” to reject science. They accept the pseudoscience that supports their a priori conclusions, while rejecting the science that doesn’t fit. That’s why many many writers agree that GMO deniers are the climate change deniers of the left.
Finding evidence for a consensus
This isn’t easy, but a consensus is not built on magic, debates, or even magical debates. It’s built on high quantity, along with high quality, of evidence.
There is a hierarchy of evidence (see Note 2) from utterly useless (anything in Natural News) to really bad (primary research described in an abstract presented at a scientific meeting as an example) to outstanding.
At the top of the list of scientific research are systematic reviews, which really are the foundation of powerful scientific consensuses. The systematic review examines all of the evidence, attempts to determine which evidence is weaker (based on statistical results and experimental design which might influence the results), and finally assembles it all into one study of studies that is complete and representative of all the research that has been done in one specialized field.
Instead of consciously or unconsciously (or deviously or innocently) choosing the papers that support your point of view, meta-reviews use science on science. They allow us to logically build and support a consensus.
This isn’t easy work (as is everything else in science, which is something that I believe is always ignored by science deniers). Each study ever published in the field must be closely reviewed. Data that might not have been published needs to be tracked down.
Authors of previous research might be contacted to explain information. Then all of the data is assembled into one huge statistical analysis. So instead of a clinical trial containing only 250 or 2500 patients, if you take all the trials ever done, you might have one of hundreds of thousands of patients.
When the authors of the systematic analysis assemble all of the data, they find the summary effect. If it is near the line of “no effect”, the researchers can conclude that the analysis shows no effect from a drug. Above is an example of a systematic analysis in a visual form called a “blobbogram,” although more formally as a Forest plot. Each of the horizontal lines represents an individual study, with the solid dot equal to the mean, and the horizontal line itself the error around the mean. The one vertical line, at 1, is the line of no effect. In this graph, the diamond point is the summary effect, which implies that it this imaginary drug has some effect.
Now, this type of systematic review is specific to clinical medicine. There are, however, large statistical reviews of scientific knowledge in many fields. A recent powerful review article discussed how climate change would effect public health.
One of the best organizations to produce systematic reviews is the Cochrane Collaboration, a non-profit academic organization that has produced meta-reviews of evidence in important questions in medicine for over 30 years. I consider their reviews to be the gold standard of research, though I think they have some tarnish on the gold at some times. A few of their reviews have tended to show strong selection bias of their authors.
Not all systematic reviews are published by Cochrane. There are many other published in major medical and scientific journals which help establish a scientific consensus. And just as one can cherry-pick individual articles, without critical analysis, to support or refute a point of view, one can do the same with these larger reviews.
Summary or TL;DR
- The scientific consensus is a powerful tool to determine what is scientifically accurate and what is not.
- The scientific consensus is based on evidence, generally published in high quality peer-reviewed journals. The evidence is repeated across many locations and over a large amount of time.
- The scientific consensus can be described as a meritocracy, that a consesnus only forms based on the merit of the experts and their evidence.
- Logic and evidence are the only qualities that matter.
- A scientific consensus represents the best knowledge of the best experts in a particular field. This is not a logical fallacy of the appeal to false authority, because evidence is always the basis of the consensus.
- Pointing out past examples of rejected scientific consensus is ridiculous. Most of the “claims” didn’t even represent the scientific consensus of the time. Even more, what is presented as “scientific consensus” were not at the level of scientific knowledge that we have today.
So, let’s be clear.
The scientific consensus says that evolution is a fact.
The scientific consensus says that climate change was caused by humans and will probably be dangerous to the planet.
The scientific consensus says that vaccines are safe and effective.
The scientific consensus says that GMOs are safe.
If you want to reject any of those facts because it hurts your feelings? Get over it.
If you want to reject any of those facts because it contradicts your belief set? You’re simply a science denier, and get over it.
If you want to reject any of those facts because it doesn’t fit your preconceived political view? Time to learn real science.
Editor’s note: This is an updated article that was originally published in December 2013. It has been revised and updated to include more comprehensive information, to improve readability and to some new ideas about the scientific consensus.
- Thanks to commenter Mike Lewinski for giving me information to add to this section.
- I have written several articles on how to distinguish and rank good research from bad. See this article, this article, and if you’re really energetic, this one.
- Cook J, Nuccitelli D, Green SA, Richardson M, Winkler B, Painting R, Way R, Jacobs P, Skuce A. Quantifying the consensus on anthropogenic global warming in the scientific literature. Environmental Research Letters 2013 May 15;8(2). doi:10.1088/1748-9326/8/2/024024.
- Demicheli V, Rivetti A, Debalini MG, Di Pietrantonj C. Vaccines for measles, mumps and rubella in children. Cochrane Database Syst Rev. 2012 Feb 15;2:CD004407. doi: 10.1002/14651858.CD004407.pub3. Review. PubMed PMID: 22336803. Impact factor: 5.720.
- Derry CJ, Derry S, McQuay HJ, Moore RA. Systematic review of systematic reviews of acupuncture published 1996-2005. Clin Med. 2006 Jul-Aug;6(4):381-6. Review. PubMed PMID: 16956145.
- Hess JJ, Eidson M, Tlumak JE, Raab KK, Luber G. An evidence-based public health approach to climate change adaptation.Environ Health Perspect. 2014 Nov;122(11):1177-86. doi: 10.1289/ehp.1307396. Epub 2014 Jul 8. Review. PubMed PMID: 25003495; PubMed Central PMCID: PMC4216160.
- Mills EJ, Wu P, Chong G, Ghement I, Singh S, Akl EA, Eyawo O, Guyatt G, Berwanger O, Briel M. Efficacy and safety of statin treatment for cardiovascular disease: a network meta-analysis of 170,255 patients from 76 randomized trials. QJM. 2011 Feb;104(2):109-24. doi: 10.1093/qjmed/hcq165. Epub 2010 Oct 7. Review. PubMed PMID: 20934984.
- Retraction–Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. Lancet. 2010 Feb 6;375(9713):445. doi: 10.1016/S0140-6736(10)60175-4. PubMed PMID: 20137807. Impact factor: 39.060.
- Schwartz LM, Woloshin S, Baczek L. Media coverage of scientific meetings: too much, too soon? JAMA. 2002 Jun 5;287(21):2859-63. PubMed PMID: 12038934. Impact factor=30.000