Читать книгу Reason: Book I - Bo Bennett - Страница 17

Scientists are people, and like all people, they are susceptible to biases, have self-interests, and are corruptible. In fact, corporations, lobbyists, political parties and action committees, defendants and plaintiffs will hire their own team of science “experts” who agree with their claims (or will, for the right price). Given the number of studies that have been done on any given issue (especially the major ones), finding data to support virtually any conclusion is not that challenging. This is a good reason to be skeptical of the claims of scientists—especially those who have clear conflicts of interest or report findings that conflict with an overwhelming consensus. The good news is that science as a method, process, and academic endeavor makes this kind of behavior exceptionally rare, making scientists as a group extremely trustworthy—at least when it comes to their published research. And here are just a few reasons why.

Full disclosure of biases and potential conflicts of interest are taken very seriously in science. Research and journal publication require full disclosure of potential conflicts of interests. Harsh disciplinary action can be taken against academics who fail to disclose any potential conflicts of interests. While this does not guarantee that no particular scientist will break the rules, it does function as an effective deterrent.

Biases might affect the hypotheses and the area of research, but are far less likely to affect the results. A strong example of this is the largest study ever conducted on the effects of intercessory prayer, conducted by the Templeton Foundation. The current President of the Templeton Foundation, John M. Templeton, Jr., is an evangelical Christian. The funded study basically set out to provide evidence that praying for others to get better was demonstrably effective—an area of research and hypothesis clearly influenced by the group’s biases and personal beliefs. However, the results of the study not only failed to support their hypothesis, but also showed that when people know they are being prayed for, they were worse off physically¹. Although the scientific process transcends personal biases and beliefs, it is not impossible for personal biases and beliefs to cause scientists to deliberately and dishonestly manipulate the results. If science comprised just one scientist, this would be a major problem—but it doesn’t, so it isn’t.

Science is self-correcting. The methodology sections of papers require very clear details of how the research was conducted so that other scientists can replicate the research. If any errors, biases, or “shenanigans” that affected the results made their way into the research (but not the disclosed methodology), replication would produce different results. Repeated attempts at confirming the results of the study by different researchers would provide further evidence against the claims and conclusions of the initial study. Scientific consensus is based on an understanding of the collection of research on a given topic—not just a few cherry-picked studies. Researchers take part in the peer-review process who know how to read a scientific paper and spot errors in the methodology and even the occasional fraudulent study. A humorous (but also disturbing) example is a scientist who found 130 “gibberish” papers that were computer generated and successfully published in scientific journals². These papers were all subsequently pulled from the journals.