Month: January 2013

The Fallacy of Incommensurability II – Gun Control

In the past months there has been a lot of talk about various gun control measures. And a lot of the argumentation has presupposed the fallacy of incommensurability I mentioned in the previous post.

The idea that one has to do everything possible in order to save even one innocent life naturally leads people to the conclusion that, if the gun control measures X, Y, and Z will save even one person from being murdered, one has to take those measures.

Now, some of those measures can also be expected to result in the death of innocent persons who are killed because they, or someone around them, did not have a gun as a result of legislation or other pressures imposed by government or media. We may suppose, though, that at least some measures can be taken that will per year save more lives than those lost as a result of the same measures.

But what about the possibility that personal ownership of guns is a safeguard against a possible tyrannical government? Should that be a significant consideration in deliberation about gun control laws? Or is it ridiculous, as some have claimed, to avoid taking current measures helpful to save lives in order to safeguard ourselves from some imaginary future scenario of a tyrannical government?

I'll run through the numbers, but before I do that, let me state the summary result for those less interested in the numbers behind it. Basically, when one works out the various risks involved, one comes to the conclusion that a rational legislator considering any significant gun-control laws, is obliged to consider the risk of an increased probability of tyranny, and to take that risk into account in considering those laws.

Let's look at the numbers. Looking history, it is a pretty conservative estimate to say that there is a 2% chance that the government of the USA will within the next 100 years have become a tyrannical power that has murdered at least 5% of the civilian population (15,000,000 persons) due to personal characteristics (race, beliefs, infirmity, etc.) or in order to maintain its own power. Note that by this 2% chance I do not mean there is a 2% chance of tyranny if the country keeps going in such and such a direction, but that the chance is this high all things considered.

Statistically this risk of murder by a future tyrannical government is equivalent to the certain murder of 300,000 persons in that same 100 years, or 3000 persons per year. There would of course be a great deal of harm done to the other 95% of the population being a part of or living under tyranny, which would increase the evil of tyranny and the importance of considering the possibility of a future tyrannical government, but let's set that aside.

Now suppose a set of gun control laws increases the mentioned risk of murderous tyranny by a small amount, but enough that the increased risk is perceivable and plausible, say 5% (for a total of 2.1%). That additional risk is the equivalent of an expected 150 persons per year murdered.

So, if it is even plausible that a set of gun-control laws will increase the risk of an eventual tyranny by even a very small amount, those gun-control laws would need to have an expected outcome of at least 150 lives saved from murder per year. And there is, indeed based on historical evidence a good deal of plausibility, indeed probability for the opinion that the enaction of serious gun-control increases the chance of tyranny.

Consequently, the argument of gun-control opponents that the second amendment and the possession of guns is a safe-guard against tyranny is a significant argument, that a rational legislator is obliged to take seriously. It could only be set aside on the premise that there is no plausibility to the opinion that gun-control laws increase the chance of a future tyranny, or that it is just as probable that gun-control laws decrease the chance of a future tyranny.

Note that I make no claim to evaluate here the prudence of any particular legislation on guns, only a kind of meta-claim about what is necessary in order to establish the prudence of such legislation.

The Fallacy of Incommensurability

"if there’s even one life that can be saved, then we’ve got an obligation to try."

"If there's anything we can do to save even one life… we should do that."

"One life lost on the job is one too many."

"If even one innocent person would be convicted, that's too many."

Frequently in speech about policies pertinent to human life and safety, one hears absolute claims such as the above. The true principle that "one may not do evil that good may come," is transformed into the seemingly plausible, but false principle "one must do everything possible to prevent such-and-such evil [e.g., innocent human death]". The plausibility of this principle derives from the fact that human life is, in a sense, of infinite value, so it seems that the loss of innocent human life can never be weighed up and considered as an altogether acceptable risk.

President Obama has on several occasions invoked such a principle to justify his positions. He even gives the impression that he actually believes the principle is valid, which would be a strong indication that he is quite incompetent to lead any community, let alone a country.

It is hard to know how genuinely any politician holds such a principle, since it seems to have settled so deeply into popular consciousness that it may be difficult for a politician to truthfully and rationally justify such decisions without risking being widely regarded as inhuman and inconsiderate. Imagine a politician publicly stating something like the following:

"This health-care legislation could be expected to save 1000 lives per year, but it would cost 12 billion dollars per year, and would a burden for doctors and other medical professionals; it's not worth it."

In actual fact, it is quite impossible to apply this principle consistently in practice, and attempting to do so leads to many contradictions. In almost every case, one of the things one could do to prevent certain bad things from happening would be to refrain from certain actions one is purportedly obliged to do in order to hinder other bad things.

Imagine, for example, that a city is deliberating whether to build a bridge over a large river. As the situation stands at present, 10,000 persons are driving daily on average 40 miles to work and back; with the new bridge, they would only need to drive an average of 20 miles. The construction of the bridge is expected to cost 20,000,000 dollars, and is estimated to have an 85% chance of involving at least one fatality for a construction work, and probably several.

Given that at least one death is expected in the construction of the bridge, it seems the city must forbid its construction. On the other hand, by constructing the bridge, 100,000,000 miles less would need to be driven per year, which over just ten years would save 15 lives. So, according to the principle that if something can be done to save even one life, one is obliged to do it, the city must build the bridge.

What will someone do who pretends or thinks that he abides by this principle of incommensurability? Who thinks that, because human life has a kind of infinite value, one can never consider an expected loss of human life to be an acceptable consequence of some policy? Faced by such a dilemma, he will either choose randomly, or in accordance with personal or basic human prejudices.  Most frequently, he will be inclined to over-rate probable and proximate events, and under-rate large but improbable and remote events. In the case of the bridge construction, he will likely underrate the importance of the deaths in traffic accidents that would be avoided by building the bridge, and overrate the importance of the deaths involved in the construction of the bridge.