Post-Truth Debates in Science

Have you ever heard the phrase 'well, that's just politics.' In the public discourse, this well-worn statement and many like it get my hackles up. I frequently interpret that statement as an admission of intellectual laziness.

Intellectual laziness, here, refers to when someone would rather not make the effort to dig deeper on a problem in order to identify and evaluate potential solutions. Instead, they chalk it up to politics and usually state their predetermined view or conclusion in a bald or unsupported manner. This unwillingness to examine the facts behind the opinions one holds is counterproductive to all of public policy.

In the more complicated scientific debates occuring right now, an additional element is mixed in: getting around the critique of intellectual laziness by stating facts to support the argument that may not be sound. These facts, on the surface, sound authoritative and based on rigorous analysis. Of course, we expect statistics and studies conducted by experts to be reliable and credible. What do we do when different sides of a debate have polar opposite facts, from groups of experts that are presented as credible? That's where most people pick a side and get entrenched.

What if we came to find out that somehow (through sloppy work usually, and maligned intent only sometimes) pseudo-science and view-centered data created solely to support one side of the argument is being injected into the discourse. What can this pseudo-science or misinformation do to a thorny and complicated scientific debate? It takes the dialogue into a very murky place where it is transformed into a post-truth debate.

In a post-truth debate, each side has their own evidence, their own experts, and their own views on what we should do to resolve a problem. However, there could be some ways to investigate whether or not a source or side in the debate has flaws.

The first way to dig a little deeper is to look no further than the order in which people approach the problem. If there is a firmly held view on the best solution, then a search for experts who already agree with that conclusion, then developing a study that likely is designed to only return the evidence that supports the conclusion; you might be on your way to a flimsy post-truth argument.

In short, this is what tobacco companies did to defend against lawsuits from people who developed lung cancer after smoking for many years. Thankfully, it was difficult for the public to get confused about whether or not smoking contributed to people developing lung-cancer. However, the collusion and cherry-picked experts with scientific sounding statements that support their claims is still deeply troubling.

Another way to examine the facts brought to the table is to look at the organization that funded the studies. There are many reputable organizations in the sciences. However, a point of confusion arises since not everything that ever comes out of a reputable organization is unable to be contradicted. This has more to do with the level of confidence that a given finding has (and I'll come back to that in a moment). More reputable organizations in the scientific arena include big-named journals and the larger professional societies.

Some think tanks are quite reputable. However, a think tank is usually less reputable when it comes to areas where they stand to benefit from a conclusion. If you are a think tank with a political lean, it undermines the credibility of your findings if the data you produce uniformly lean in one direction. Why? Because the question of whether the research was designed soundly, or whether you have suppressed evidence becomes relevant. These questions are relevant because doing both of those things (biasing research design, and suppressing evidence) have become enticing incentives that would support the overall mission of the organization.

Oh boy, we're deep into talking about research design and evidence. This is the last way to investigate whether there are flaws in a source: examining the critical nuts and bolts of the facts and statements used as fodder for arguments in many of the more technical and scientific debates our society faces.

Doing this kind of analysis is Yeoman's work. Frequently, it is only the researchers (scientists, engineers etc.) that are familiar enough with the nuances of research design and how to determine the level of confidence that can be ascribed to information or conclusions.

One particularly sobering problem with doing this kind of work is how difficult it is to correct the record when a flaw is discovered. In the public discourse, the lightning fast pace of the dialogue can pass this information by because the topic is 'old hat.'

Now, what about the cases where we are lucky enough to expose a flaw, and bring it out in the open for all to see? Here is an example. A politician makes the statement that a risk assessment shows there will be a biological attack on our country. This sounds scary!

Upon deeper investigation of the important nuts and bolts used in this assessment, it turns out that the model they use to do the assessment is severely flawed. In fact, the model will produce the result that an attack will happen on the country, no matter what the inputs are. In other words, it cannot produce a result that would say anything other than 'an attack will happen.'

Whether it was sloppy work or maligned intent, most reasonable people might be distressed by this and have good reason to distrust the assessment. At the very least, we would want a more rigorous and analytically sound method used. We'd also like to make sure we examine any important decisions that were based on this analysis. What if it was used to justify going to war?

Taking it one step further, what if we were to find out that the politician knew the model would always return the same result. At this point, people might get really angry.

In short, I think the exposure of the critical flaws is received by society as a betrayal or loss. Elizabeth Kubler Ross wrote in her work about grieving that people go through stages when confronted with this type of event. Denial, Anger, Bargaining, Depression, Acceptance.

I think our modern society has not moved past denial or anger when faced with this type of information. Further, I think it is becoming more common and a principle element of post-truth debates.