May 30, 2003
Libertarianism and the Harm Principle
CalPundit has tried to take Eugene Volokh to task on the "Harm Principle". Kevin writes
This is a good example of why I've never been able to take libertarianism seriously: it simply doesn't provide any meaningful real-world guidance for what governments should and shouldn't do. Once you agree that "harm" also means "potential harm" or "harm done to third parties down the road" or "unintentional harm" or — well, or anything, really, then you no longer have a principle at all. Virtually every human action there is can plausibly be supposed to cause harm of some kind, which in turn means that we are left to judge policies by balancing their effects on personal liberty with the protections they provide us against harmful behavior by others. Which is exactly what Eugene proposes.
This is sloppy thinking that comes from just not understanding probability very well. The problem is here in this part of the quote
Virtually every human action there is can plausibly be supposed to cause harm of some kind, which in turn means that we are left to judge policies by balancing their effects on personal liberty with the protections they provide us against harmful behavior by others.
Is it plausible that a man having intercourse with his wife is going to cause an injury? No. Is it plausible that a man who is also having sex with many male partners could cause injury to his wife? I think the answer is that it is much more likely to result in injury.
Probabilistically it would look like this.
P(Injury|SMMP) >> P(Injury|NSOP)
Where SMMP is Sex with Many Male Partners, and NSOP is No Sex with Other Partners.
The idea here is that we can differentiate between different levels of risk. But in good "politically correct" fashion Kevin has decided that nobody can make such a ranking as we do above.
So we could refine Prof. Volokh's argument to fit that when the probability of injury falls below a given level, the government has no reason to be interfering.
Ideally, I wouldn't want the government interfering at all in people's sexual lives, but let consider one important fact: information is not perfect. That is people do not have all the relevant information necessary for making the best decision. If for example, men who had sex with many male partners were to disclose this information to each new partner, then the new partner would have important information for making decisions. But, people rarely reveal such information.
Similarly with other activities. Sitting in my living room reading a book is highly unlikely to pose a risk to anybody. Driving my car 125 MPH down a residential street on the other hand....
The mistake comes with the Kevin's introduction of the word plausible into the sentence. No, not every human action is plausibly capable of causing harm. Those that are not should not be of any concern to the government.
Posted by Steve at May 30, 2003 01:19 PMSteve, you just repeated almost exactly what I said. Different things have different levels of (potential) harm and different (potential) benefits. You have to weigh them. Or rank them. Or balance them.
And I said "virtually," not "all."
I know you like to disagree with me, but what's the point when we're actually saying the same thing?
Posted by: Kevin Drum on May 30, 2003 11:10 PMI don't know Kevin, but I'd say lately I have been disagreeing with virtually all of your posts.
See the word virtually means practically all, almost all, those that do not count is a small percentage.
I take words as having a meaning. In this case virtually plausible means that pick any action at random and it is highly probably to pose a risk to another. I don't think this is the case, or if it is you should try to make an arguement supporting your claim.
I think actually the reverse of you claim is true. Most of what I do has little or no impact on people. Is there a risk when I get up, take a shower, get dressed, and get in my car? Now driving my car, yeah that has a very definite potential risk, so it is regulated. Now, I get to work. Is walking in to the building a risk? How about putting a couple of sodas in the fridge? Sitting at my desk and turning on the computer? Eventually heating up my lunch in the microwave? Opening that dreaded tupperware?
Most of our lives are unregulated there isn't somebody standing around watching everything you do. This says to me, most of what you do is not a danger. Thus, libertarians are arguing about teh areas that are regulated.
Further, it is you who are essentially agreeing with Volokh in the end. It all becomes a question of what is pragmatic.
But that's just what everyone does, liberals and conservatives alike. So exactly how does libertarianism help us make these decisions?
The difference is degree. A libertarian argues that instead of assuming virtually every human action is not plausibly a danger. It is the opposite of you, Kevin.
Granted the approach is subjective. Even cloaking it with probability still leaves the big subjective question of what probablitity of danger is too high.
Posted by: Steve on May 31, 2003 10:12 AM