Obstructing Justice on the Pretense of Redressing Discrimination

The criminal justice system takes too long and costs too much. That is why the much-criticized practice of plea bargaining is a necessary evil. That is why actually carrying out the just, deserved sentences for the very worst murderers is the exception and not the rule.

So what did California Governor Gavin Newsom do Wednesday? He signed a bill to make the problem worse, layering a large, expensive, and time-consuming new burden on the already staggering system. The pretense for AB 2542 (now Chapter 317, Statutes of 2020) is redressing racial discrimination, but it will not do any good along those lines, while it will do much harm.

In the wake of the U.S. Supreme Court’s decision 33 years ago in McCleskey v. Kemp, bills titled “racial justice acts” were introduced in many legislatures across the country. At the time, almost all of the legislatures had the good sense not to pass them. Good sense has departed the scene in California.

The first operative section of AB 2542, new Penal Code § 745, begins by stating, “The state shall not seek or obtain a criminal conviction or seek, obtain, or impose a sentence on the basis of race, ethnicity, or national origin.” Of course. Hardly anyone would disagree with that principle. But as usual, the devil is in the details. What is “on the basis of”?

The new section lists five ways to prove “on the basis of,” none of which actually prove that. Here is number (4) (emphasis added):

The defendant was charged or convicted of a more serious offense than defendants of other races, ethnicities, or national origins who commit similar offenses and are similarly situated, and the evidence establishes that the prosecution more frequently sought or obtained convictions for more serious offenses against people who share the defendant’s race, ethnicity, or national origin in the county where the convictions were sought or obtained.

Number (5) is similar for sentence severity, and it include both race-of-defendant and race-of-victim components. So what is “more frequently sought.” Here is paragraph (h)(1) (emphasis added):

“More frequently sought or obtained” or “more frequently imposed” means that statistical evidence or aggregate data demonstrate a significant difference in seeking or obtaining convictions or in imposing sentences comparing individuals who have committed similar offenses and are similarly situated, and the prosecution cannot establish race-neutral reasons for the disparity.

Whenever people collect data on anything in our society and tally it up by any demographic characteristic, the numbers invariably some difference. How could we prevent there being any difference? One way would be to have a strict quota system, but quota systems themselves are unconstitutional discrimination.

The law requires a showing of different treatment of people who are similarly situated. That seems reasonable at first blush, but how do you determine “similarly situated”? The problem is mind-numbingly complex.

This issue has been studied and litigated most in the area of capital punishment. What “everybody knows” about these studies is quite wrong. For the long version, see my 2012 article, Rebutting the Myths About Race and the Death Penalty. In a nutshell, studies on racial discrimination and capital punishment have failed to make a convincing case that either the race of the defendant or the race of the victim is a substantial factor in determining which murderers are sentenced to death.

The best known study is the Baldus study in Georgia, litigated in the McCleskey case. After a full trial, the district court found as a matter of fact that Baldus had not made his case. Not even close.

“The best models which Baldus was able to devise which account to any significant degree for the major non-racial variables, including strength of the evidence, produce no statistically significant evidence that race plays a part in either [the prosecutor’s or the jury’s] decisions in the State of Georgia.” McCleskey v. Zant, 580 F. Supp. 338, 368 (N.D. Ga. 1984) (emphasis omitted).

This important finding is nearly unknown because the court of appeals took the highly unusual (for a case tried to judgment) step of assuming the facts in favor of the appellant, McCleskey, in spite of a finding against him. The Supreme Court followed suit on review of that decision, and ever since then people have treated the Baldus study as “proving” what Baldus claimed, even though it had been found not so.

In New Jersey, a very pro-defendant, anti-death-penalty court appointed a special master to investigate capital sentencing discrimination in that state. The defense side claimed to have proved discrimination, but the special master came to the same conclusion as the Georgia judge. In Maryland, an academic study, not adjudicated, found no effect based on the race of the defendant but claimed to find one on the race of the victim. Yet an independent scholar, one who had been a witness on the defense side in McCleskey, analyzed the same data and came to the opposite conclusion.

So what is going on here, that so many people can look at the same data and come to very different conclusions? The Supreme Court put its finger on one important aspect of the problem nearly half a century ago, when it rejected a challenge to the death penalty as it then existed and declined to constitutionalize the Model Penal Code’s suggestion of a defined list of aggravating and mitigating circumstances. “[F]or no list of circumstances would ever be really complete. The infinite variety of cases and facets to each case would make general standards either meaningless ‘boiler-plate’ or a statement of the obvious that no jury would need.’ ” McGautha v. California, 402 U.S. 183, 208 (1971).

Claims that similarly situated people of different races are being treated differently have been made with mathematical models that purport to take into account the relevant factors. The Baldus study litigated in McCleskey had models with 200, 230, and 250 variables. These are staggering high numbers of variables, resulting in unstable models. Different models give different results depending on which variables are included. The judge in that case found that one of Baldus’s models was completely invalid because it contained no factor for the strength of the evidence, an obviously valid consideration.

Compounding the difficulty of the very high number of variables is the difficulty of the input data. Models require numbers. How do you quantify, in an objective way, a factor like “strength of the evidence”? You really can’t. Even for variables you can quantify, how much in resources will be consumed by recording hundreds of factors in every case?

Unlike prior “Racial Justice Acts,” this one is not limited to capital punishment. It applies in every criminal case. Are we now going to have McCleskey-type litigation, with massive studies and extensive, expensive testimony of dueling experts in every criminal case? Or will we merely have to have one such huge case for each offense commonly charged in each county in the state?

This bill will produce massive expenditures in time and resources that, in the end, do not prove anything one way or the other. But this time, by shifting the burden to the prosecution, a great many criminals may get off with much less than they deserve or get off altogether.

The Legislature has not merely opened a can of worms; it has opened a freight train load of cobras.