DUI v. Cell-Phones: Don't Go There
---
Alex Tabarrok at Marginal Revolution has a provocative post about this study (pdf file) that "found" that people who use cell phones while driving are more impaired than those who drink and drive.
Now before the hyper-anarcho-libertarians (you know who you are) jump all over the study, allow me some hasty-stitchy observations:
1. Limited sample size (41 subjects).
2. The simulator only replicated daytime driving conditions (drunk driving tends to occur at night).
3. The simulator only replicated a "multilane interstate highway" (again, most drunk driving happens on local or even residential roads).
4. Any study that uses a simulator suffers from observation bias (i.e., the subjects know they're being observed and know that it's only a simulation, therefore their behavior is inevitably altered, even if only subconsciously, by the simulation itself).
My previous posts (here and here) on DUI laws sparked some controversy, but my basic premise holds -- if you believe it is conceptually permissible to criminalize sufficiently reckless conduct before there is a clearly-identifiable "victim" (i.e., if "reckless endangerment" is a valid crime) then you cannot argue, on philosophical grounds alone, that DUI laws are in any way an affront to libertarian principles. You can argue that they're inefficient, unfairly applied, costly, etc., but the same can always be argued about any particular law in any particular context.
But if you want to argue that all DUI laws are inherently "just plain wrong," in the sense that drug prohibition is "just plain wrong" or the draft is "just plain wrong" or smoking bans are "just plain wrong," then you need to do more than throw a temper tantrum, which tends to be the preferred form of argument on this issue.
Now before the hyper-anarcho-libertarians (you know who you are) jump all over the study, allow me some hasty-stitchy observations:
1. Limited sample size (41 subjects).
2. The simulator only replicated daytime driving conditions (drunk driving tends to occur at night).
3. The simulator only replicated a "multilane interstate highway" (again, most drunk driving happens on local or even residential roads).
4. Any study that uses a simulator suffers from observation bias (i.e., the subjects know they're being observed and know that it's only a simulation, therefore their behavior is inevitably altered, even if only subconsciously, by the simulation itself).
My previous posts (here and here) on DUI laws sparked some controversy, but my basic premise holds -- if you believe it is conceptually permissible to criminalize sufficiently reckless conduct before there is a clearly-identifiable "victim" (i.e., if "reckless endangerment" is a valid crime) then you cannot argue, on philosophical grounds alone, that DUI laws are in any way an affront to libertarian principles. You can argue that they're inefficient, unfairly applied, costly, etc., but the same can always be argued about any particular law in any particular context.
But if you want to argue that all DUI laws are inherently "just plain wrong," in the sense that drug prohibition is "just plain wrong" or the draft is "just plain wrong" or smoking bans are "just plain wrong," then you need to do more than throw a temper tantrum, which tends to be the preferred form of argument on this issue.
All Related Posts (on one page) | Some Related Posts:
Posted by KipEsquire on
28 August 2004
To comment on this post, please visit the new blogsite.
