Raising standard of evidence
Raising standard of evidence is a propaganda technique sometimes used in refuting a convincing and appropriate argument. It consists of defining an unreasonable standard of evidence that must be met before the argument of one's opponent can be accepted.
This is most commonly seen with respect to conspiracy theory. Usually such a theory describes events where there is high motivation to coverup real events, and where reliable unbiased witnesses are hard or impossible to come by, e.g. espionage activities. Because two or more groups are in competition both to influence and explain events as each other's fault, it is necessarily the case that evidence cannot be as reliable as it could be in a scientific or criminal matter. Very often, the only conceivable case that could be made is circumstantial: those who had opportunity, resources, motive, and did in fact benefit from the outcome. But rather than fairly require that all conspiracy theory meet a rigid standard of evidence, propagandists are skilled at lowering it for their own side, and raising it for their opponents', to the point where effectively they are trusted to arbitrate the truth without limit.
With abstract or specialized subject matter, such as medicine or mathematics, it is usually not hard for an expert in related subject matter to pose as an expert in the controversial subject, and censor material they find uncomfortable or unconvincing.
The primary defense against arbitrary raising of standard of evidence is to determine what standard is being applied to the competing arguments, using representative cases made by one's opponents. In doing so it can be quite useful to refer to a general scale of standards of evidence themselves:
- axiomatic proof which is generally thought to be very reliable but narrow.
- quasi-empirical methods including highly trusted human arbitrators.
- empirical methods as employed in the 'hard' physical sciences, those focused on prediction, and which employ mathematics for modelling
- forensic standards
- statistical standards
- judicial standards