Writing in the Huffington Post recently, Karl Giberson, a prominent proponent of theistic evolution, appealed to the well-known argument that science is a self-correcting mechanism. He writes,
Science – and this includes evolution – is a self-correcting enterprise. I know little of psychiatry, but I am not shocked to discover that critical voices have emerged and are being heard. This is the norm for science. Seemingly secure science is often modified – think Newtonian physics – and entire fields even disappear, like phrenology (studying personality via bumps on the skull). Anyone who understands the scientific community knows it to be full of renegade individualists only too eager to overturn the status quo. This aggressive self-examination is the reason why we now understand the world so well…
The reality is different from this idyllic description, and informed consumers of science know that, public relations aside, there are serious doubts as to the extent to which science is a self-correcting enterprise. For example, the epidemiologist John Ioannidis wrote a paper in 2012 entitled Why Science Is Not Necessarily Self-Correcting. The abstract begins as follows:
The ability to self-correct is considered a hallmark of science. However, self-correction does not always happen to scientific evidence by default. The trajectory of scientific credibility can fluctuate over time, both for defined scientific fields and for science at-large. History suggests that major catastrophes in scientific credibility are unfortunately possible and the argument that “it is obvious that progress is made” is weak.
Ioannidis proceeds to mention one mechanism which renders self-correction less than perfect:
Efficient and unbiased replication mechanisms are essential for maintaining high levels of scientific credibility. Depending on the types of results obtained in the discovery and replication phases, there are different paradigms of research: optimal, self-correcting, false nonreplication, and perpetuated fallacy. In the absence of replication efforts, one is left with unconfirmed (genuine) discoveries and unchallenged fallacies.
What the last sentence means is that, if replication of research results is not a ubiquitous feature of science, there will be unchallenged fallacies. They will not be corrected. And, as we have discussed several times in this forum, replicability of research is a major weakness in contemporary science.
Ioannidis is too savvy about problems with contemporary science to swallow Karl Giberson-type propaganda:
The self-correction principle does not mean that all science is correct and credible. A more interesting issue than this eschatological promise is to understand what proportion of scientific findings are correct (i.e., the credibility of available scientific results).
Even if we believe that properly conducted science will asymptotically trend towards perfect credibility, there is no guarantee that scientific credibility continuously improves and that there are no gap periods during which scientific credibility drops or sinks (slightly or dramatically). The credibility of new findings and the total evidence is in continuous flux. It may get better or worse.
The paper by Ioannidis is enlightening. I was particularly pleased to discover that arguments I made in Genesis and Genes mirrored those made by Ioannidis. So I reproduce here the section of the book which deals with the issue of science as a self-correcting mechanism:
Jonathan: I’ve heard it said that the fact that new theories replace old theories only proves that science is a self-correcting enterprise. Do you agree?
YB: That’s a nice way to put a happy face on it. But there are two serious problems with this suggestion. Firstly, even if science were this gigantic super-tanker that eventually turns around, it might be too slow for the individual who lived while the old paradigm prevailed. Let’s consider the demise of the eternal universe paradigm. Until 1965, most scientists believed that the universe had never been created – it was eternal. This stood in total contrast to the Torah view that the universe was created at a specific point. By 1965, the old paradigm had collapsed, and was replaced by the Big Bang model, according to which the universe came into existence, apparently ex nihilo. Now imagine a person who died in 1950. Does it help him that science is a self-correcting mechanism? His entire life was spent in the shadow of the monolithic scientific consensus that the universe is eternal. Since he, like all of us, was not a prophet, he could not foresee that some time after his death, the scientific paradigm that dominated his life would crumble and be replaced with a radically different picture. If this person had been a Jew, he would have lived his entire life with unresolved tension between the scientific paradigm that the universe is eternal, and Jewish belief in the creation of the universe. So this business of self-correction, even if it were true, is only good for historians. It won’t help your average individual struggling with a particular issue and having only one lifetime.
Jonathan: I see. But you mentioned that there were two problems with this suggestion.
YB: Yes. The second problem is this: Why do you believe that science is a self-correcting mechanism? It is because we know that in specific cases, certain beliefs that the scientific community subscribed to turned out to be wrong and were discarded. But there is no way to estimate in what percentage of all cases science indeed reverses its course. Oh, I know the party line about how scientists constantly scrutinise the evidence, compare their hypotheses to experimental results and the rest of it. But we saw enough in the previous chapters to appreciate that in real life, it hardly ever reaches this ideal. I described some stories that had happy endings, like the one involving Dr. Robin Warren, who established that bacteria cause some ulcers. But do you know how many stories had a sad ending? Can you estimate how often in the past a researcher had a hunch but abandoned his line of research when he was subjected to ridicule? Do you have any way of estimating which ending happens more frequently, the sad or the happy? What if for every case like Dr. Warren’s, there were a hundred scientists who had a promising insight or idea, but were deterred by the initial rejection they experienced? We only hear the stories with a happy ending. But scientists are human beings, and most human beings don’t have a thick skin.
The post Dr. John Ioannidis and the Reality of Research:
The post Dr. Ben Goldacre and the Reproducibility of Research:
Retrieved 9th June 2013.
Retrieved 9th June 2013.