A moving decision boundary will create unpredictable results and poor decisions. A fixed decision boundary can be created by having a set of criteria defining when the decision is on this side or that side of the boundary, which define the consistency of the decision. The quality of the criteria defines the quality of the decision. The fallacy is to have a poorly defined boundary, allowing for favouritism and cherry picking, leading to a poor argument.
Let’s unpack that a little. As you approach the traffic lights, there comes a point where if the light were to turn amber, attempting to stop becomes dangerous. A practised driver learns where that boundary is and will consistently know when they have reached it. Outside influences can change that boundary – the weight of the vehicle, the speed of the vehicle, whether it is raining or dry, the density of traffic and so on. However given the same circumstances the decision point is the same – on this side of the line it is safe to stop, on the other it is safer to keep going. For most people, the boundary is intuitive – we learn through experience when we have hit that point.
Science is a philosophy that attempted to take intuition out of the equation. If you cannot explain how you know that this is the decision point, then you cannot teach it, or test if that is actually accurate. After all, you may have a very unsafe decision point to stop or start… it would be nice to know where that is before finding out the tragic way. As such, the philosophy behind science requires various questions to be asked, ideas to be tested and so forth to create rules behind defining the boundary and testing if these rules create a good boundary or a poor one. One could say that if the traffic light turns amber (from green on the way to red) then if your vehicle is 50 metres before the solid white line, it is a good place to break, after the 50 metre line it is not. This rule can work very well for certain loads, road conditions and speeds, but very poorly for others. The rule is consistent – but not necessarily creating a good quality outcome. Thus the rule needs to be refined, if you weight is less than 2,000 kilograms and your speed is less than 80 kilometres per hour, then 50 metres may be the best spot for the decision boundary, however for every 500 kilograms add an additional 10 metres and for every 10 kilometres per hour add another 10 metres… and so on. Testing these parameters is quite easy and following the set of rules leads to a consistent good quality outcome.
This example seems like a simple one to analyse and quite obvious. Another obvious set of rules is around medication. A medication should have a primary desired effect for a dose for a certain percentage of the population with minimal negative side effects, otherwise it should not be prescribed by a medical practitioner. The creation of the pharmaceuticals should meet certain standards to ensure that the effect of the medication is what is predicted by the clinical trials rather than being biased by some random element introduced to pack out the active ingredient to make the drug easier to administer. Seems sensible – you know what you are taking, you know what it does and you know how to measure its effectiveness. The rules for governing effectiveness and for creating the medication are consistent, which should give a consistent outcome. These rules have been refined over time to ensure a high quality result. Combined the vast majority of prescription medication is both consistent and good quality. The factors the interfere with the consistent quality is for more variations in humans rather than medications.
People take chemicals for medical conditions that are not prescribed. These are often in the form of supplements or “alternate medicines”. When the medical professionals or governments ask for these to be controlled by the same set of rules, the alternate professionals raise arms citing this is unfair or oppressive. Independent tests of supplements have found that a concerning proportion of supplements do not contain the active ingredient listed on the container, or do not have the dosage listed (either too high or too low). There is no clinical evidence that the alternate medication or supplement has any therapeutic effect and no control on manufacture methods. If the rules used to govern these supplements and alternate medicines were applied to the prescribed medications, people would die, professionals would be sued and their would be a government inquiry. So why are the rules for the goose not being applied to the gander?
As an aside, to quote Tim Minchin in Storm – “Do you know what we can alternative medicine that works?… Medicine.”
Back to the point in all this. Any time an argument is made that uses inconsistent rules, that argument has an inherent logical fallacy. Keep in mind, that while consistent rules will give a consistent outcome for both the goose and the gander, if those rules are terrible, the result will be also terrible.