In this post, I want to briefly explain and discuss a logical blunder known commonly as the “nirvana fallacy.” This fallacy occurs when you suggest either that a solution should not be used because it is imperfect or that a solution should not be used because there is some underlying issue that is not being addressed, but you fail to provide a plausible alternative. That may seem a bit confusing at first, so I will use several examples that are commonly used by opponents of science.
Let’s start with one of the most basic and most obvious examples. Anti-vaccers often like to claim that we should not vaccinate because vaccines aren’t 100% effective. This is an extremely clear cut instance of the nirvana fallacy, because the fact that something isn’t 100% effective does not mean that we should not use it. Partial effectiveness is still better than no effectiveness. Indeed, almost nothing is 100% effective. For example, seat belts, helmets, condoms, parachutes, etc. are all less than 100% effective, but they are still very useful. Even so, a life-saving medical marvel like vaccines doesn’t need to be 100% effective to be useful, because every life that is saved is important.
Now let’s look at a slightly more complex example: the peer-review system. This is the system that scientific papers have to pass before being published, and it is admittedly imperfect. Indeed, I have devoted numerous posts on this blog to problems with it (for example here, here, here, and here). This has led some to suggest that it is worthless and should be abandoned entirely. In reality, however, an imperfect quality control system is still better than no quality control system at all. For example, we can all agree that a quality control system at a candy bar facility that limits cockroach legs to one per every ten chocolate bars is better than no quality control system at all (note: those aren’t actual statistics). Even so, the peer-review system is imperfect, and bad papers do sometimes get through, but an awful lot of bad papers never make it.
This brings me to an important point about nirvana fallacies: if you are going to argue that something should be abandoned because it is imperfect, then you must simultaneously propose a more effective alternative. To go back to my chocolate bar example, there would be nothing wrong with saying, “we should stop using the current method that limits cockroach legs to 1 per 10 bars, and switch to method B, which limits legs to 1 per 20 bars.” It is, however, invalid to simply say, “the current quality control system doesn’t stop every cockroach leg (i.e., it isn’t 100% effective), therefore we should remove the quality control altogether.” Even so, if you can propose a better alternative the current peer-review system, then by all means do so, but it is ridiculous to argue that we should either let any “study” pass as valid science or just abandon science altogether.
A similar example often occurs in climate change debates. I frequently encounter people who admit that we are probably causing the climate to change, but they argue that we will never be able to curtail our greenhouse gas emissions in time to truly stop climate change, therefore we shouldn’t bother to do anything. Once again, however, an imperfect solution is better than no solution. Yes, we probably won’t be able to fully prevent climate change, but we can prevent the worst consequences of it (Schleussner et al. 2016), and that makes it worth taking action.
Another version of this fallacy also occurs during climate change debates, and it can be summarized as, “but not everyone else will do it.” For example, I often talk to Americans who argue that America should not try to limit its fossil fuel use because even if America switched to renewable energy sources, many other countries wouldn’t, so there is no point. Once again, the problem is that even if America was the only country to take climate change seriously (which is currently almost backwards of reality), that would still have an impact on the amount of warming that occurs. Further, the actions of others have no bearing on your own responsibility. In other words, the fact that everyone else is doing something unethical does not mean that it is ok for you to do it (yes, I know that was a philosophical argument not a scientific one, but I think that it is relevant here).
A final variant of the nirvana fallacy occurs when you argue that a solution should not be used because it does not address some underlying issue. A good example of this comes from GMOs. Despite all of the anti-GMO propaganda, not all GMOs are about money. Some, such as golden rice and GMO bananas, are being developed solely as humanitarian endeavors. These GMOs are rich in vitamin A, which is currently lacking in the diets of some developing countries. Anti-GMO activists often respond to this by saying that the vitamin deficiency in developing countries is actually just a symptom of poverty, and those deficiencies would go away if we took care of economic inequality and food distribution. Really think about that response for a minute. They are actually arguing that all that we have to do to fix the problem is solve world hunger and poverty. Sure, fixing those things would solve the vitamin problem, and we should be trying to fix those problems, but we clearly aren’t going to find the solution any time soon. In contrast, we could be using vitamin rich GMOs within a few years or even months. Asking people who are suffering and even dying from vitamin deficiencies to wait for us to fix world hunger rather than using a GMO is absurd. In other words, it is true that the GMOs don’t address the underlying issue, but the underlying issue is a nearly impossible problem to solve. Therefore, we should use the solution that is available to us, even though it’s not perfect.
In short, an imperfect solution is generally better than not having any solution at all. Therefore, it is generally not valid to argue that a solution should not be used simply because it is imperfect or incomplete, unless you can provide a feasible and superior alternative.
Schleussner et al. 2016. Differential climate impacts for policy-relevant limits to global warming: the case of 1.5C and 2C. Earth Syst. Dynam. 7:327-351.