Inferring and Explaining

127 t ′ 1 . All the data comes from a state in the Deep South where there is a long history of racial discrimination. t ′ 2 . All the data comes from the late 1970s, racial discrimination has greatly dissipated in the ensuing two generations. I concede that these are legitimate counter- arguments, but I still believe that t ′ 0 is the best explanation. It’s true that gathering all one’s data from a single state or region of the coun- try is a less-than-ideal polling technique, but sometimes one has to take the data that are available. Remember that medical researchers took the Framingham data very seriously, even though all of it came from the Northeast. As for the claim that attitudes about race have greatly improved since the latter half of the 1970s, I am of two minds. I would like to believe that your generation, and your parent’s, is less racist than those who came of age in the ’40s, ’50s, ’60s, and early ’70s. I think there is some evidence for this. But at the same time, one need only turn on the TV or radio and be aware of what is going on in our country right now to see that whatever improvements we see with respect to race, we still have a hell of a long way to go. Mymain reason for continuing to support t ′ 0 , however, is that the Baldus study is not all the data that is available. In 1990, the United States General Accounting Ofce released a report, “Death Penalty Sentencing: Research Indicates a Pattern of Racial Disparities,” that reviewed research from across the entire county. Here is their summary of their fndings. Our synthesis of the 28 studies shows a pattern of evidence indicating racial disparities in the charging, sentencing, and imposition of the death penalty afer the Furman decision. In 82 percent of the studies, race of the victim was found to infuence the likelihood of being charged with capital murder or receiving the death penalty, i.e., those who murdered whites were found to be more likely to be sentenced to death than those who murdered blacks. Tis fnding was remarkably consistent across data sets, states,datacollectionmethods,andanalytictechniques. 6 A Causal Explanation of the Correlation In Furman , the Supreme Court was concerned with the arbitrary and capricious actions of trial judges and particularly juries. It appears now, however, that the judgments, both arbitrary and prejudicial, of other legal ofcials are evenmore problematic. An obvious explanation of the Bal- dus data is the following. CaPItal PunIshment and the ConstItutIon t ″ 0 . The race of the murder victim causally infuences the decision whether to seek the death penalty. Given my earlier interpretation of the Eighth Amendment, this account of the murder statis- tics inGeorgiaseems todemandthat theSupreme Court declare capital punishment, at least in the state of Georgia, to be unconstitutional. As I hope by this point you are all already thinking, the crucial question is whether t 0 is really the best explanation. I, personally, cannot see any way that this could be a case of “reverse causation.” Tus I reject any possibility that the following needs to be considered at all: t ″ 1 . The decision to seek the death penalty causes the race of the victim.