Two professors at Loyola University in New Orleans “have created a computer program”:http://www.csmonitor.com/2005/0427/p17s01-usju.html which they claim can accurately predict 90% of the time whether a particular criminal defendant will be sentenced to death. The two professors fed 19 variables into the program, including age, race, sex, marital status, and the date and type of offense.
Unsuprisingly, the the CSM article also notes that “some observers” found it alarming that the variables fed into the program “contained no details of the case.”[1] One of the professors who authored the program, Dee Wood Harper, commented that “if this mindless software can determine who is going to die and who is not going to die, then there’s some arbitrariness here in the [United States justice] system.” A Texas lawyer is quoted as saying, “who gets executed still appears to be random and arbitrary.”[2]
However, the one variable singled out by the article does not support the argument that imposition of the death penalty is either arbitrary or random. Certainly the whole premise of the article is that the death penalty is *not* random; the program can apparently predict with 90% accuracy which defendants will be sentenced to die. Nonetheless, I expected the article to discuss the arbitrary character of the death penalty by showing that one of the variables with the most predictive power was race. Instead, the professors point to the state in which the defendant is tried and convicted as a key factor. Now, that fact should come as no surprise. California, for example, has executed only two people in the last 38 years, while “Texas has executed”:http://www.tdcj.state.tx.us/stat/annual.htm 23 prisoners in 2004 alone, and a total of 341 since 1982. If you are convicted of a capital crime in Texas, you are obviously much more likely to be executed than if you are convicted in California.
Harper says, “[y]ou can be one foot over the line in the next state and not be subject to this kind of penalty.” Well, of course, but I hardly see how this makes the imposition of the death penalty arbitrary. The citizens of each state are governed by a different set of criminal laws, with their own proscriptions and their own penalties. The difference in imposition of the death penalty from state to state is no more unjust than the difference in any other criminal law. The implications of Harper’s argument would be to create a unified criminal law for all the states. Now, a “uniform body of criminal law”:http://myweb.wvnet.edu/~jelkins/crimlaw/basic/mpc.html across all the states would certainly have some benefits, but the mere fact that some states choose to deviate from the model should not be considered arbitrary.
fn1. I quibble a bit with this point, as I’d say that the type of offense for which the defendant was convicted is certainly a relevant detail of the case. The article does not discuss the impact of each of the variables, as “the research is in its early stages,” but I wouldn’t be all that surprised if there is a least some correlation between the type of offense and the imposition of the death penalty. Indeed, it seems to me that in addressing discrimination in the imposition of the death penalty we should be most concerned with variability within the same category of offense, rather than across the entire range of potential offenses.
fn2. Will Baude “responds”:http://www.crescatsententia.org/archives/2005_04_27.html#005338 by arguing that, in fact, decisions motivated by prejudice–while certainly no less distasteful–are not arbitrary at all. He write, “it doesn’t seem fair to criticize predictable decision-making as ‘arbitrary’ rather than stereotyped, prejudiced, or otherwise incorrect.” This is true only if we accept that “arbitrary” means random, which is not necessarily true. I take arbitrary to mean that the decision was supported by a justification, but that the justification was specious. In other words, the imposition of the death penalty can be both predictable *and* arbitrary, when the variable with the greatest predictive power is, for example, race.