A Reasonable Degree of Scientific Certainty

Over the years I have heard people say, "I believe/know to a moral certainty that such-and-such (is true)." I have said it myself but have no idea what it means. I think it means the same thing as "I really, really, really believe such-and-such (is true)." Interestingly, when people use it they may be talking about a moral issue as in "I know to a moral certainty that screwing goats is a sin." But it doesn't have to be a moral issue. Right now, I know to a moral certainty that George Bush has lost his fu__ing mind. Selling the operating rights to something like 6 or so of our most important ports to some dudes in Dubai is what one of my acquaintances calls "dirt fu__ing stupid." Maybe some Dubai citizens involved in the purchase of these operating rights are reliable but if you are familiar with the 9/11 Commission Report, you will know that

Bin Laden relied on the established hawala networks in Pakistan, Dubai, and throughout the Middle East to transfer funds efficiently.

Hawala networks are informal banks that allow persons to transfer money around and hide the interactions from government investigators charged with making us as safe as possible. Unfortunately our President doesn't seem to give a damn about the safety of the American people when there is money to be made.

However, it is not my primary intention to bash Bush though it is always good fun but to worry about phrases like "to a moral certainty" and the widely used legal phrase "to a reasonable degree of scientific certainty." I don't know of any other similar phrases. I have never heard, for instance, things like "to a religious certainty," for instance or "to an academic certaint" or any other such phrase.

Attorneys using expert witnesses who are scientists ask them things like "Do you know to a reasonable degree of scientific certainty that OJ's blood was present at the crime scene?" The scientist replies, "Yes" and he or she might give the reasons. I have argued in an earlier blog that scientific certainty is not attainable. We can say with great certainty if not total scientific certainty things like "I know to a scientific certainty that John Jones' heart was beating at noon because he showed a blood pressure reading of 120/70 at that time." However, our speaker could not say, "I know to a reasonable degree of scientific certainty that John Jones' blood pressure was 120/70 at the time in question." The problem is that the hearing of the person taking John Jones' blood pressure could be bad or the mercury manometer employed might be inaccurate. Only the most trivial of facts, if any, are knowable to a scientific or any other certainty.

If you buy into this argument, then we are forced to talk about degrees of scientific certainty as the phrase we are looking at suggests. In some cases, as in experiments that provide levels of confidence experts may say what the level of confidence of some finding was. However, the experiment may have been predicated on problematic assumptions or the sample of subjects (mice, people, etc.) might have been defective in one way or another.

In some cases, as when dueling psychiatrists give expert testimony they cannot honestly say that they know to a reasonable degree of scientific certainty that the defendant did (or did not) know the difference between right and wrong or that the defendant was (or was not) legally insane. The problem here is that no measuring is being done and without that, there can be no degree of certainty, much less scientific certainty. When I gave expert testimony during depositions about alleged deceptive advertising by some eight or nine major oil companies in regard to the inclusion of alcohol in gasoline ("gasohol," as it is sometimes called) I did say that it was my expert opinion that the language of any sign put up at a gasoline filling station saying, "No Alcohol Added" or "No Alcohol in Our Gasoline" implicates that adding alcohol to gasoline is somehow bad. I could say that I knew with a reasonable degree of scientific certainty that some people would do so because I did and the claim is so phenomenally weak that my drawing the inference is enough to verify it. (Another linguist and a psycholinguist said the same thing.) I could have said that I was reasonably sure that the oil company ad agencies and the oil executives involved in deciding to use signs like that also must have believed that people would draw that inference or they wouldn't have bothered creating and putting up the signs. In fact, I believe that most people would draw that inference but I couldn't assert that with any specific degree of scientific certainty.

The link associated with the title of the present blog also notes that we cannot be certain about the claims of scientists, and goes on to say that there are those who exploit this lack of certainty for their own ends. The anti-Darwinian Creationist believes this leaves room for Intelligent Design in the classroom, as if the latter (non)theory actually could be as certain or more certain than the theory of evolution (or theories of evolution, if you prefer). Or a conservative politician like Bush might exploit the existence of scientific uncertainty to delay action on some environmental regulation when acting on it will be expensive.

To exploit the fact that scientists cannot be certain about their claims (not counting trivial claims based on observations such as "John is alive") simply because this uncertainty exists is intellectually dishonest. Its like blaming a rose for having thorns. Unlike roses, however, where we have alternatives, there is no alternative to science as a means of understanding the physical universe (including our minds but alas, not our souls, whatever those things are).

I am more than a little troubled by use of the word "reasonable" in "to a reasonable degree of scientific certainty." There are some published articles discussing this notion but I would have to go to the library to find the publications since they are all at pay to view sites. I did run into Blog 702, which concerns legal issues, that noted that the Third Circuit held that a "Handwriting Expert's Testimony Need Not Be Given to "Reasonable Degree of Scientific Certainty."

Once one puts the word "reasonable" in "to a reasonable degree of scientific certainty" an unacceptable level of subjectivity is involved in assessing the claim because what might be reasonable to one scientist or judge or jury might not be reasonable to another. In short, use of "reasonable" and "scientific certainty" in such a phrase makes it into a kind of oxymoron. We have at best an illusion of scientific credibility. It wouldn't be the first time a court has deemed credible quite sloppy science. In my expert opinion, experts should restrict themselves to saying "It is my expert opinion that ...." and add whatever measures might have been used or defend the opinion and then shut up.

I fear I have meandered a bit in this blog. I trust most will forgive me.