Welcome to what we expect will be a very interesting and productive discussion of Michael Cholbi and Alex Madva‘s, “Black Lives Matter and the Call for Death Penalty Abolition.” The paper is published in the most recent edition of Ethics and is available through open access here. Erin Kelly has kindly agreed to contribute a critical précis, and it appears immediately below. Please join in the discussion!
Erin Kelly writes:
Michael Cholbi and Alex Madva’s paper, “Black Lives Matter and the Call for Death Penalty Abolition,” argues that capital punishment wrongs black defendants and black communities, and that the proper remedy for this wrong is abolition of the death penalty. In developing this argument, they make an interesting case for understanding the racial wrongs of capital punishment in political terms—as instances of distributive injustice—rather than (simply) in terms of a failure to achieve retributive justice. I will explore both the nature of their claims about distributive justice and their criticism of retributive justice. I won’t address the case for abolition, which flows naturally from their conclusions about the harm done by the death penalty. Instead I will suggest, briefly, how their argument against the retributive theory could be stronger.
The authors argue that black communities suffer a distributive injustice. Specifically, racial bias and outright discrimination in American legal practices compromise the promise of equal status and equal protection under the law for African Americans. As the authors explain, this represents a harm to the black community, generally speaking, not just to some individual members. The harm the authors are particularly concerned with is related to the death penalty. Their claim is that the existence of the death penalty elicits race-based biases against black defendants and, more generally, against black people.
Empirical evidence supports the conclusion that murders are treated as more heinous when they are committed by blacks and/or against whites. Black defendants are more likely to receive the death penalty than defendants of other races or ethnicities, and defendants overall are more likely to receive the death penalty when their victims are white. Studies citied also show that these two effects interact: cases involving black defendants and white victims are treated the most punitively.
The authors hypothesize that the perception that murders committed by blacks and/or against whites are worse is influenced by the eligibility of defendants for the death penalty. The death penalty requires juries to consider whether some murders are worse—bad enough to receive the death penalty—and this task triggers racial bias regarding the moral status of these murders, or so the authors plausibly argue. Racial disparities in sentencing are not as starkly produced by the possibility of other harsh penalties, including life without parole. For example, one study found that mock jurors who were as likely to impose life without parole on whites as blacks were significantly more likely to impose death on blacks. Another study showed that whites are actually more, not less, likely to support the death penalty when informed about racial bias in capital sentencing. These studies cited by the authors suggest that the prospect of imposing death exacerbates racial bias in sentencing.
This raises a question about why this should be the case. The author’s propose an explanation. They point to anti-black stereotypes that include the belief that blacks are less sensitive to pain, that the family members of black murder victims are less worthy of sympathy, and that blacks are considered either ‘subhuman’ or ‘superhuman,’ in various contexts. The authors argue this: “These specific biases suggest that anti-Black biases are closely linked psychologically with the infliction of bodily violence, i.e., that Blacks are perceived to both suffer less harm from it and are more prone to inflict it. Although such biases are apt to exert some influence across a variety of contexts, e.g., in police officers’ split-second decisions to use force as well as clinicians’ deliberations about prescribing pain medication to Black patients, they are likely to be more salient in capital cases than in cases that do not involve murder or capital punishment.” This is a perceptive and interesting analysis of the impact of particular racialized social attitudes that are animated by the death penalty.
Still, evidence that “the death penalty arouses race-based biases that distort judgments regarding the justifiability of imposing death as a punishment” would not in itself show that the death penalty harms the black community. That requires a further step in the argument, a step the authors take. They argue that the racial bias that is elicited by the death penalty has consequences for how black defendants and black people are treated more generally. They argue that the death penalty elicits and aggravates racial bias that in turn influences “intuitive judgments of guilt, appraisals of incriminating evidence, charging decisions, assessments of the severity of pain and suffering, and general moral intuitions related to punitiveness and desert.” In other words, they argue, “the possibility of a capital sentence may function as a powerful contextual cue that both activates and amplifies the effects of specific anti-Black biases, and second, that the bias- amplifying power of capital contexts is not restricted to capital-sentencing contexts.” The “social meaning” of murder, which varies with the races of the people involved, influences how members of the black community are perceived more generally, at least in relation to criminal law. The bias that is elicited by the legal requirement that murders be “graded” for severity, loops back to reinforce more pervasive biases against black people as less worthy, insensitive to pain and loss, and dangerous, and this affects the likelihood of their being subjected to arrest, prosecution, conviction, and harsher sentences.
In short, the case the authors make, and the empirical findings on which it rests, support the conclusion that “capital punishment is not simply another manifestation of racial discrimination but a context that activates biases that make such discrimination more likely, with cascading effects downward into other aspects of the criminal justice system, such as conviction rates, police use of force, mistrust of criminal justice in Black communities, etc.” (28) These are concrete harms to the black community.
Indeed, some of the empirical findings referred to above suggest that part of the point of the criminal justice system is to intimidate African Americans. There is a long history, at the highest levels, of using the law to denigrate blacks to second-class citizenship, or worse. In the Dred Scott decision (1857), Supreme Court Justice Taney declared unapologetically that blacks in America “had no rights the white man was bound to respect.” 100 years later, the Supreme Court’s reversal, in the Brown decision, on the constitutionality of racial segregation in schools, resulted in its flabby requirement to desegregate “with all deliberate speed.” De facto racial segregation in schools is rampant to this day. The Supreme Court’s death penalty adjudication expresses the same lack of determination to eliminate racial injustice. Evidence for the racially uneven application of the death penalty, considered by the Furman Court in 1972, led to its formulation of facially race-neutral “guidelines” that opened the way for a reinstantiation of the death penalty, affirmed by the court in its 1976 Gregg decision. The predictable persistence of a racially-biased death penalty should be viewed as nothing less than an outcome willfully accepted by the courts.
Cholbi and Madva take a somewhat milder view of the criminal justice system and American society, arguing that failure to address the systematic wrongs done to Black Americans in the criminal justice system constitutes a form of societal or institutional recklessness. They argue that societal support for, or even mere toleration of the death penalty, displays unjustified risk-taking with regard to Black lives. The authors apply the concept of recklessness self-consciously. The legal concept of recklessness does not imply intentional harmdoing or knowledge that harm will in fact occur, though, unlike negligence, it does involve a conscious disregard of an unreasonable risk.
The authors unpack the recklessness claim along two lines. They point to reckless disregard for the impact of the death penalty on both the equal protection and equal status of blacks under law. The equal protection claim is less convincing since it seems to depend on the disputed notion that the death penalty has deterrent value. The authors claim, “Because would-be murderers are justified in believing that individuals who murder Blacks are likely to face lesser costs (the presumptively less severe punishments of prolonged or lifelong incarceration rather than execution) than they would for murdering individuals of other races, the law thus fails to penalize killings of blacks in a manner consistent with their having the equal protection of the law.” To be clear, when aimed at the police, this claim is believable, since police officers risk no liability to capital punishment when, in their official capacity, they kill blacks. What is questionable is the broader claim the authors seem to have in mind: the deterrent value of the death penalty (vs. prolonged incarceration) when directed to ordinary white would-be murderers.
The idea that the death penalty further undermines the already unequal legal status of blacks is more obviously a serious concern. As the authors emphasize, blacks who are neither charged with nor victimized by murder are affected by how blacks accused of murder are treated by the legal system. (Note that in many states, over 90% of murders are death-penalty eligible.) The legal system’s treatment of the alleged “worst of the worst” affects how black defendants, more generally, will be treated by juries, police, and prosecutors. Blacks can expect to be regarded as less innocent, interactions with the police to be potentially very dangerous, and evidence or mere suspicions against them to be more vigorously pursued. The authors argue, plausibly, that this unequal treatment represents a demotion of legal status. They might have added that it also represents a diminishment of moral status.
These serious harms are missed by popular moral and legal theories whose treatment of matters of criminal justice has a retributive focus. Retributivists might believe that black defendants get more punishment than they deserve, or that white defendants get less punishment than they deserve. But retributivists neglect the social implications of sentencing practices, because those implications are not considered relevant to determining what an individual defendant deserves. The authors make this point, but do not otherwise criticize retributivism, except to express mild skepticism about notions of individual desert. Here I think the authors miss an opportunity to be more critical. A focus on retributive justice does not merely overlook distributive injustices. It obfuscates them. Its individualistic, character-oriented focus amounts to a form of racial bias that diverts attention from the state’s role in perpetuating systematic and institutional racial injustice.
This excellent article makes a compelling case that the abolition of the death penalty would increase the fairness (or at least reduce the unfairness) of the penal system by eliminating the opportunity for implicit bias in an instance in which its effects are particularly striking. The argument that doing so would also contribute to overall improvements in perceptions of black people, and consequently in how they are treated in a variety of circumstances, is also plausible. I agree with these arguments. I would go further. While there may be some rare instances in which implicit bias plays no role in arrest, prosecutorial decisions, quality of legal representation, verdict, or sentencing, every form of social injustice in the U.S. disproportionately affects people of color. Injustices in the distribution of wealth and opportunity directly affect the probability that those subject to them will turn to crime. The composition of our prison population thus reflects both implicit bias and social injustice. In failing to rectify social injustices, we as a society not only choose that preventable crimes will be committed, but also that those who commit them will disproportionately be people of color. We will not achieve fairness in criminal justice until social injustices, as well as the effects of implicit bias, are rectified.
Thanks very much to PEA Soup for organizing this discussion and to Erin Kelly, whose precis captures very elegantly the reasoning of our article and its importance.
We’d like to kick things off with two clarifications raised by remarks in Erin’s precis. The first concerns the role of the deterrent effect of capital punishment in our reasoning, the second how our position relates to a retributive critique of race-based discrimination in the administration of capital punishment.
1. Erin quotes us: “Because would-be murderers are justified in believing that individual who murder Blacks are likely to face lesser costs (the presumptively less severe punishments of prolonged or lifelong incarceration rather than execution) than they would for murdering individuals of other races, the law thus fails to penalize killings of Blacks in a manner consistent with their having the equal protection of law” (523). This claim is central to the case we mean to make about the distributive injustice of the American capital punishment regime, i.e., that this regime does not accord Blacks as such the equal protection of the law, even if they are not murder victims. However, Erin takes this reasoning to be weakened by its dependence on the “disputed notion that the death penalty has deterrent value.” While we concur that this is disputed, our reasoning does not (as best as we can discern) rest on whatever deterrent value the death penalty may have. Would-be murderers are justified in believing that the race of their victims influences the probability of their being executed. But whether or not these differences in the ‘cost’ of murder affect the death penalty’s capacity to deter, would-be murderers need not have any beliefs regarding that capacity in order for our ‘equal protection’ argument to stand. As we point out (pp. 525-26, note 20 in particular), the effects that racial discrimination has on capital punishment’s deterrent value may be difficult to predict. In any event, it’s crucial that our position doesn’t turn on the consequences, deterrent or otherwise, of race-based differences in capital sentencing. An analogy: A facially neutral poll tax might disproportionately burden low income Black voters and thereby be unjust, even if it did not depress Black voter turnout (perhaps Black voters turn out in greater numbers due to backlash against the poll tax itself).
2. Erin suggests that we could have made a more forceful criticism of retribituvism. We share her underlying concerns, in particular, about the “individualistic, character-oriented” focus of much retributivist discussion, and we certainly agree that retributive justice is inadequate for capturing the full range of wrongs done to black communities by racially discriminatory criminal-justice practices, including capital punishment. Hence our emphasis on distributive justice. Retributivism, however, need not be construed individualistically. A case can be made against such practices on grounds of comparative desert, where the subjects of comparison are again whites and blacks as classes, rather than (merely) as individuals. And Ben Yost makes precisely this case in an excellent recent paper, arguing that racially differential punishment is retributively unjust, in comparative terms, because it contributes to racial oppression. We take Yost’s argument to be at least consistent with ours, but there may be richer conceptual connections to draw between the distributive and retributive cases against racially differential punishment (see also our note 14 on p.523).
Yost, B. S. (2017). What’s Wrong with Differential Punishment? Utilitas, 29(3), 257–285. https://doi.org/10.1017/S095382081600039X
Deirdre – Thanks for your comment! Alex and I agree that there’s likely to be bias at many points in the administration of criminal justice. I’d be interested to know if you think our claims regarding capital punishment’s “cascading effects” on the rest of criminal justice administration are plausible.
Thanks, Deirdre. I wholeheartedly agree about the importance of unjust structural conditions across numerous dimensions, and that fairness “outside” criminal justice is necessary for fairness “inside” it. I use the scare quotes there because I take it that many of these central social institutions are integrated in ways that make the inside/outside comparison somewhat misleading, as exemplified in phrases like “school-to-prison pipeline.”
I might, however, pick some nits about the sort of contrast you’re drawing between implicit (and for that matter, explicit) bias and social injustice. Consider your final sentence: “We will not achieve fairness in criminal justice until social injustices, as well as the effects of implicit bias, are rectified.” Fair enough, but what exactly is preventing us from rectifying social injustices? Aren’t racial biases partly to blame? As in the “Southern Strategy,” as in invocations of “welfare queens” in order to dissolve our social safety nets, and as in the demonstrable role of racial resentment in support for the current administration and in driving economic anxieties, etc. (https://www.vox.com/identities/2017/12/15/16781222/trump-racism-economic-anxiety-study).
There are a variety of reasons to think, then, that changing individuals’ implicit and explicit biases will be integral to bringing about the more substantive changes necessary for true social justice.
I’m going to come at this from the opposite direction from Erin. Your (excellent) paper works hard to assuage worries about the putative injustices generated by underpunishing death-deserving murderers. The main appeal is to considerations of distributive justice. But I wonder if expanding the normative framework beyond retributivism opens you up to a new objection. An expressivist in the vein of Jean Hampton might reply that abolition harms the victim and his or her community. On the expressivist view, part of the point of punishment is to communicate society’s disapprobation of the wrongdoer’s implicit claim to superiority over his or her victim. If we assume that death might be an appropriate punishment for murder, refusing to execute murderers would thus wrong the victim, victim’s family, and society at large (including Black communities), by failing to sufficiently condemn the crime. (US legal practice sides with the expressivist to the degree that it construes crimes as wrongs against the state.) This isn’t some implausible van den Haagian claim about noncomparative justice trumping comparative justice, but a claim about a problem with underpunishment over and above noncomparative injustice. I wonder if this changes the moral calculus at all.
Let me put this question slightly differently. In your response to van den Haag, you state that it is not obvious that abolition involves a loss in noncomparative justice (539). Those who, like me, have serious misgivings about capital punishment will agree, but anyone who worries about a loss of noncomparative injustice is going to be unimpressed by your maneuver. So if we set this burden-of-proof-type strategy to the side, we are left with the task of weighing the wrongs involved in underpunishment against those your paper identifies as associated with capital punishment. And then we seem to have to weigh the additional wrong I suggested above.
Great question, Ben. I for one will certainly need more time to think it through, so this comment is by way of preparing a fuller response. Suppose we do adopt an expressivist view (we don’t, but our references to the social meaning of murder are in that ballpark). Then in the current situation, there’s a comparative expressivist injustice in that the US is systematically communicating a greater level of disapprobation for certain crimes (those committed by blacks and against whites, for example) than others. The findings that race influences how juries take the considerations of victims and their families into consideration seems particularly relevant here. This is clearly unjust.
Then the question is what (if anything) to do about it, and there are various options to consider. When it comes to considering these options, we don’t have to argue that there are no downsides to abolition, only that the downsides are less severe than the downsides for other strategies. In that vein, we spend some time in the paper (leading up to the section you cite) considering some of the other options and find them either seriously unjust or simply unfeasible. Realistic strategies to meet van den Haagian aims of executing more white defendants and murderers of black victims run into obvious problems. So it seems to me that we can grant (for the sake of argument) some non-trivial loss in crime-condemning while it still remaining the case that abolition is the least bad of the options.
But honestly I think I’m not completely clear on the work that the distinction between underpunishment and noncomparative desert is doing. We talk about weighing comparative vs. noncomparative, and your suggestion is that we’d instead (or also?) have to weigh underpunishment. So the idea is that we have two separate considerations weighing against abolition– one is losses in noncomparative justice and the other is underpunishment and maybe failure to sufficiently condemn, and might that tip the scales of the moral calculus? It’s a little hard for me to see why that wouldn’t amount to double-counting closely related concerns (or rather, treating two different moral idioms or normative-theoretic approaches to effectively the same issue as if they are two different reasons). But even if it’s not double-counting, perhaps we could just run an analogous argument specifically about underpunishment vs. comparative punishment? (So, if we introduce expressivist considerations, then on the pro-abolition scale, there’s the comparative story I rehearsed in the previous paragraph to weigh against the loss in crime-condemnding.)
Of course, this is all assuming that there are absolute punishments that are precisely calibrated to disapprove of certain crimes, and so on. And we are certainly skeptical. For one thing, we suspect that racial biases are shaping intuitions about how much disapprobation is required to condemn crimes.
Also, it’s been a while since I dug into the research on “closure” for families, but insofar as that’s relevant here, my recollection is that execution doesn’t actually bring closure for families. So I also strongly suspect there’s an empirical case to be made that the death penalty does not uniquely communicate the relevant sorts of condemnation over and above life in prison (see also 540-1). If it did, then there ought to be some evidence for it.
If “family closure” is relevant to assessing what the death penalty communicates, in contrast to life imprisonment, then there’s at least some evidence in favor of life imprisonment over the death penalty — as practiced in the current American legal system, which includes extremely lengthy appeals processes for the death penalty.
https://www.pbs.org/newshour/nation/death-penalty-bring-closure-victims-family
But again, I haven’t done a deep-enough dive to see if there’s countervailing evidence.
Ben, I agree with the move Alex makes in his first paragraph: It looks to me like we (he and I) can translate our distributive argument into expressivist terms, i.e., that the US capital punishment regime ‘undercondemns’ murders of Blacks and ‘overcondemns’ murders by Blacks. That said, I’m not charitably disposed toward expressivism for reasons that could be extracted from our paper: Expressive considerations seem always to require some non-expressive basis. In this case, I’d propose that what’s ‘expressed’ in the US capital punishment regime is failure to accord Blacks equal respect under, and the equal protection of, the law. But if so, what’s the ‘expression’ doing here that’s not done by equal respect and equal protection? That line of argument generalizes.
As to underpunishment: I’d follow Alex in responding that we need not deny there is some loss in non-comparative justice via abolition, more specifically, by failing to adequately condemn murder. So long as this loss is, on balance, less objectionable than the maintenance of the existing CP regime, then our case for abolition stands. (Though even here, I wonder about the magnitude of the loss: Life imprisonment — in effect, exile within one’s own national borders — expresses pretty harsh disapprobation, doesn’t it?)
Just to clarify what I meant by underpunishment and noncomparative injustice… If we grant that some murderers deserve death, then LWOP will underpunish those murderers. For some retributivists, the wrong of underpunishment is free-floating – it doesn’t wrong anyone in particular. For expressivists, underpunishment *also* wrongs the victim and wrongs or morally harms the community at large. (I think a plausible case could be made that these are distinct kinds of wrongs, though this is doubtlessly controversial.) My claim that that abolition might involve an injustice “over and above” noncomparative injustice was meant to capture this latter point. My thought is that if we acknowledge the wrongs done to the victim and community, we might have to recalculate the moral cost of abolition. If it’s a close call in favor of abolition without accounting for expressivist considerations, it’s an even closer call with them, and perhaps a different call altogether. At any rate, thanks to both of you for your responses – I’ll need to think a bit more about them.
I really enjoyed the article and the discussion. I had a badly thought out question about justice and deterrence. I take it that DP is generally justified only if it deters compared with lesser punishments (I am not a retributivist or expressivist,but many retributivists and expressionists will agree, for many retributivists and expressivists think that desert or expression are values that can be pursued only conditionally). Suppose that it does deter substantially compared with LWOP. Now note that black people are significantly more likely to be victims of homicide than white people. As DP tends to punish those who kill blacks more harshly than those who kill whites, retaining DP will tend to disadvantage blacks compared with whites in terms of protection. And there is some reason to think that it disadvantages them overall – those who are inclined to kill someone have reason to select black victims over whites to avoid DP. But in considering whether to abolish DP, we must consider the overall effect on blacks of abolition compared with retention, even with racist application, including both protection and other values, and they may be disadvantaged by abolition if the overall deterrence effect is great enough. Furthermore, they may be comparatively disadvantaged because the overall reduction in deterrence will fall more heavily on them. Couldn’t this effect be great enough to outweigh the other disadvantages to black people of retaining DP that you illuminatingly explore?
Of course, this argument depends on the idea that DP deters, and ought to be retained if it deters. I have doubts about both claims – the empirical evidence on the first claim is contested, and I think that there are constraints on the use of death for deterrence which are not met except in very exceptional cases. Furthermore, an overall just package of policies to tackle violent crime does not, I think, include DP even if DP is effective in very unjust social circumstances such as those found in the US. But these are more general arguments against DP, and it seems very hard to make a complete case for DP that does not rely on deterrent justice.
Two comments.
First, retributivism is primarily a theory concerned with proportion, not with setting specific penalties for specific crimes. The only exception seems to be the death penalty and murder, making the exception look ad hoc, not something retributivists need take responsibility for. One can, for example, be a retributist and oppose the death penalty for all the reasons Cholbi and Madva offer. Cholbi and Madva make this point but some of the discussion above seems to miss it.
Second, I wish the discussion so far had not gone on as if all American states had the death penalty. Roughly a third does not. Few defenders of the death penalty seem ready to claim that they are states of injustice–the same for many countries that have no death penalty. I therefore wish Cholbi and Madva had used the data from those states–Michigan,Illinois, and so on–to support (or refute) their empirical claim for the effect of abolition on distributive justice. Do courts in fact discriminate appreciably less on the basis of race when the maximum sentence is something other than death?
Michael, I agree about the importance of your second question. We tried to look into it, reaching out to various places like the Sentencing Project, and came up empty. ***If anybody out there has any access to relevant data on this, please share it!***
What I did come up with was that racial sentencing disparities vary dramatically from state to state (here’s a map: https://www.sentencingproject.org/the-facts/#map?dataset-option=BWR), and certainly, when you just eyeball it, side to side with a map of states that do and do not have CP (https://deathpenaltyinfo.org/states-and-without-death-penalty), the correlation would not pop out at you. (If anything, going by the eye test, it looks like disparities are worse in some of the non-CP states.) Of course, this is an extremely unscientific way to evaluate the claim. There are, obviously, many other variables affecting sentencing patterns and a proper analysis would need to hold them constant in a regression analysis.
Victor: You raise a good set of questions. On the question of deterrence more generally: As we said in response to Erin, our argument doesn’t turn on the consequences, deterrent or otherwise, of the DP. Still, you’re right that in considering abolition, we should think about the overall effects on blacks vis-a-vis retention. These things are tough to predict, but if (as we argue) in the article, the kinds of bias that infect the DP are not likely to show up in LWOP administration, then the result would be little or no racial bias in the administration of punishments for murder. Worth keeping in mind that while blacks are more likely to be victims of homicide, it *isn’t* the case that the DP tends to punish those who kill blacks more harshly than those who kill whites (the reverse actually). So it seems likely that from the overall standpoint of deterrence, Blacks would fare better under LWOP as the harshest sanction for murder (rather than DP).
Sorry, there was a slip of the pen, or keyboard, in my first comment – I meant to acknowledge that DP tends to punish those who kill blacks less harshly than those who kill whites, but miswrote. My question was whether this comparative effect is what we should focus on, or focus on alone. Some might argue that what matters most is the overall protection that blacks (or even just people generally) receive from being murdered when we compare retaining or abolishing DP. If DP saves lives, some might think, and no one is punished beyond what they deserve or are liable to, any distributive questions or questions of discrimination are secondary. I’m not intending to defend this view, but I wonder whether you think that such questions retain their importance even when DP reduces the death toll overall, and if you do, why.
Thanks, Victor. First, I must say that it’s a bit difficult for me to get into the hypothetical head space of taking deterrence seriously, for some of the reasons Michael (Cholbi) and I discussed above (and also in pages 525-6 and notes 19-21, and in some of our responses to van den Haag in section V). In particular, (1) I’m concerned that our intuitions about when and how to weigh the value (i.e., mattering!) of lives against questions of equal treatment, upholding rights, etc., are themselves racially inflected, and (2) social models get really complex once we take deterrence seriously.
Re (2), I strongly recommend Jack Glaser’s Suspect Race (OUP 2014), which focuses on the question of deterrence and profiling (esp. Ch.5, and the book is chock full of other important references). One of the most salient examples here is the problem of “reverse deterrence”: if we’re going to posit that under conditions of differential enforcement, the more targeted group is going to be deterred, then we also have to suppose that the *less targeted* group will know they can offend with relative impunity. In conditions where the more targeted group is a numerical minority, then the net effect could be an *increase* in the relevant crime. And Glaser offers lab data, computer models, and drug enforcement field data to suggest that this is a very live possibility. Long story short, it simply can’t be a question of a simple weighing of lives saved vs. distributive justice, because the variables interact.
OK, all that said, putting my armchair-philosopher’s cap back on, if I imagine that the data on deterrence were really compelling, such that we had very strong reason to believe that racially disparate applications of the death penalty were actually protecting more lives than would abolition of the death penalty, that fact would certainly play a role in my moral assessment of the overall practice. How much is hard to say (and now I circle back to concern #1 above!), especially if we suppose that we’re protecting “people generally” at the expense of the black community in particular. Practices that perpetuate racial oppression are morally weighty indeed! If we instead suppose that the overall saving of lives disproportionately benefits the black community, that would carry more weight in my mind. But again, it’s really hard to see how that latter supposition could be close to realized under current conditions, given that cases with black victims are among those punished least harshly. We simply aren’t protecting black lives as much as white lives.
I’m going to take the opportunity to ask a question that has less to do with the stated aims of your paper and more with anti-death penalty arguments in general. One persistent problem for the objections to capital punishment that rely in part on procedural shortcomings is the possibility of cases where the relevant procedural issue fails to materialize. (I realize your paper isn’t exactly an exercise in procedural abolitionism, though you clearly foreground the proceduralist value of equal protection.) This brings me to your response to Matthew Kramer… Kramer’s vindication of capital punishment against proceduralist attacks gets much of its strength from its restricted scope, applying, as it does, only to defilingly evil offenders. You contend that while a Kramerian capital punishment regime might seem to avoid the racial injustice of our current system, Kramerian reforms would still leave the system vulnerable to implicit bias (536). I agree with you that notions like “defilingly evil” are likely to be interpreted in a discriminatory fashion. But it is always open to Kramer’s friends to raise the standards of defiling evil. In this vein, the committed retentionist might suggest that we build a Kramerian floor to the definition of “defiling evil” such that any white person committing such a crime would (very, very likely) be sentenced to death. A suitable definition might be extracted from the particulars of the crimes of Richard Ramirez, Ted Bundy, etc., along with those guilty of genocide and the like. These super-charged “worst of the worst” criteria would potentially eliminate the implicit bias issues identified on p. 535, because any DA worth her salt (and with a taste for reelection) would seek the death penalty in cases where the criteria apply. Would this Kramerian regime be satisfactory to you? If so, it looks like it’s possible – in a practical sense – to have a system of capital punishment that does not devalue black lives.
Ben, I’m afraid I have to remain skeptical that even the “super-charged” Kramerian criteria would result in a DP regime free of racial bias (or even one with a tolerably small level of bias).
For one, I’m not sanguine about DA’s pursuing only cases where the criteria are seemingly met. Even if that were true, so long as they have the sort of discretion they have in US jurisdictions, DA’s might also opt *not* to pursue DP for some white offenders who meet the criteria. (Consider the different sympathies elicited by white mass shooters, who tend to get labelled ‘troubled’ or ‘mentally ill,’ and unarmed blacks shot by police, who are nearly always subsequently exonerated for those shootings on the grounds that the individual seemed threatening, etc..) But supposing we could figure out the offender effect, I’d be even more concerned about the victim effect — how would we ensure that victim race doesn’t enter into judgments about who to charge, etc.?
Beyond this, my memory of Kramer’s book is a bit hazy, but I don’t recall him saying much to operationalize ‘defilingly evil.’ I’d be concerned that the usual language surrounding capital murder statutes (‘heinous,’ etc.), if incorporated into jury instructions, etc., will invariably make salient the racial biases Alex and I appeal to in our paper.
You’re right that Kramer doesn’t do much in the way of operationalizing, and he places some rather interesting burdens on appellate courts – like enforcing a second-order rule of closure that prohibits all murder cases exhibiting epistemic or moral uncertainty from cashing out in execution. But at some level, this doesn’t matter: even if the Kramerian system isn’t absolutely free from bias, it might be free enough to meet your objections. If only a few people every decade are subject to capital trials, and if those trials are far less vulnerable to bias (due to the heightened criteria, the fact that most defilingly evil offenders Kramer names are white, and the more general fact that most mass murderers are white), wouldn’t this change the social meaning of murder? Garden variety murders, even multiple murders (or cruel ones like those of Robert Alton Harris), would not be death-eligible, and garden variety murderers would thus be subject to the more egalitarian LWOP. And those garden variety cases are the ones, it seems to me, that fuel the arguments of your paper.
Problems of vagueness are not solved by moving the threshold; they’ll arise wherever you set it. We see this regularly play out in terms of debates over how what counts as a “mass shooting.” (Similarly for problems of rule-following… we say you have to kill at least 10 people to get DP, but then what about the person convicted of killing 9 and attempting to kill 3… etc.) The more profound obstacle here, however, is all the ways that bias can, before the sentencing stage, and especially in the detective work and evidence-gathering stages, make it the case that some defendants will come to seem to have committed defilingly evil crimes while others will not. Less evidence will be dedicated to cases with white defendants and black victims, making it less likely to turn up evidence related to serial patterns, and so on. And Netflix’s *Making a Murderer* offers a vivid demonstration of how coerced confession tactics can lead to trumped-up charges of heinousness.
Perhaps Cholbi and Madva, having dealt with the death penalty, can go on to the penalty of life without parole (sanitized above as “LWOP”), a penalty that strikes me as good a candidate for “cruel and unusual” as death, especially when (as in fact it often is) combined with solitary confinement in a maximum security prison (or “supermax”). Indeed, reducing the maximum for penalty for murder to ten or twenty years in prison (and reducing the penalties for other crimes accordingly) would have a humanizing effect on American justice. Even if it did nothing for the amount of racial disparities, it still would take some of the bite out of it. It would also do much to reduce the cost of criminal justice without (as far as I can tell from the empirical literature) significantly affecting the crime rate. Just a thought.
Hear! Hear!
Hi Alex, sorry to beat on this horse some more… I agree with everything you say about the problem of vagueness. But one of keystones of the paper is the claim that what is wrong with unequal sentencing is the negative effects it has on black communities in the various ways you guys describe so well. If, however, there are only a handful of death penalty cases every decade (due to the stringent nature of the criteria), how would the associated inequalities have any effect at all on attitudes and practices? The number of capital cases would be too small to do any practical or normative work. Furthermore, because almost all (or all) mass-murderers are white, the very few death-eligible murderers would be white. (Though the potential problem of underpunished racist serial murderers, as you note.)
These are all good questions, for sure. To dip my toes into some super speculative waters, it’s not immediately obvious to me that the rarity of the cases would necessarily mean that they stop embodying and perpetuating unequal status and protection. Rare events can be extremely powerful drivers of public sentiment and policy (cf. a certain large-scale terrorist attack on American soil and its downstream effects on the equal status and protection of Muslim Americans). Death penalty cases are already relatively rare. Police shootings of unarmed black men are also relatively rare, yet they’ve been incredibly powerful catalysts for activism (and hopefully eventually meaningful social change). Many social theorists complain that we are *too* captivated by isolated, rare events rather than the humdrum workings of structural-institutional forms of oppression (consider also the public outcry about high-profile instances of animal cruelty against gorillas or lions next to the inuring to the everyday horrific realities of the meat industry, or the effects of isolated, unseasonably hot or cold days on perceptions of climate change). OK, enough examples. That’s all by way of highlighting that rarity does not necessarily prevent, or even impede, the problematic practical or normative consequences, but again, that’s all super speculative.
Hypothetically, if we did revise death eligibility upward to the levels you describe, and found that almost everybody on death row was white (bracketing the grounds for skepticism about this result that Michael and I raised above about prosecutorial discretion, evidence-gathering biases, etc.), then it would be hard to see how the practice could continue having the same racial salience it does now. One response here is that our paper is not intended to be an argument against DP in all possible worlds, but specifically in ours. Here and now, DP is racially coded. (Compare: there are possible worlds where American mobs only lynch/lynched white people. In those worlds, one couldn’t make an argument against lynching on the grounds of its being racially discriminatory against blacks, although one could probably make a case against its being discriminatory against whites, and one could of course object to it on other grounds. But it’s not clear what significance such possibilities should have for people considering whether to end or reintroduce public lynchings in our world.)
Another way at this issue is that our theory basically predicts that the precise upward revision you describe simply won’t happen. As in the evidence suggesting that support for DP and other punitive policies is *caused* by racism (specifically, white support increases when they learn about racial disparities). And if it did happen (such that whites were now disproportionately more likely to get DP), it wouldn’t be stable. The prediction is that people would either become less invested in perpetuating the practice under these conditions or that, as soon as another high-profile instance of violence occurred, at the hands of (say) an immigrant, person of color, or a member of another salient social minority, but where that high-profile case failed to meet the super-Kramerian threshold, then a public outcry would lead to reducing the threshold again.
Lastly, related to the instability point, and as alluded to in the lynching analogy above, there’s a case to be made that a practice that resulted in disproportionate DP for whites would violate whites’ equal status and protection. Michael and I are not of the view that you can only be racist against otherwise-oppressed groups.
Just to add a little to Alex’s previous comment: It’s important to underscore how local the conclusions of our article are meant to be — abolition of DP in, and in light of, the specific racial context of the U.S. One thing that we don’t have, and it would have been very salutary to have had as we conducted our research for this article, is information about the DP in nations similar to the U.S. demographically, etc., but without its fraught racial history. Unfortunately (from a research standpoint only, mind you!), the other nations where the DP is used are very different from the U.S.:
https://www.theguardian.com/news/datablog/2011/mar/29/death-penalty-countries-world
it’s reasonable to wonder whether the sorts of bias that infect the DP in the US play a role in other settings, but again, it seems impossible to know.
I’m sure this is dealt with in the paper, but how do you respond to arguments like those from van den Haag that go something like this: suppose only Pintos get speeding tickets, whereas Mercedes don’t. Certainly the answer isn’t to stop ticketing the Pintos, it’s to ticket more Mercedes. And so, analogously, if the death penalty disproportionately affects minorities, why isn’t the answer to execute more non-minorities?
Assuming the point of the argument has to do with distributive justice, you could fix inequities in *either* of these two ways (i.e., executing fewer minorities *or* executing more non-minorities). And then if that isn’t the answer the authors want, the problem with the death penalty wouldn’t trade on distributive justice at all, but would rather be that the death penalty should be banned regardless. Either way, it’s hard to see what distributive justice has to do with anything. Or something like that.
And then, just in way of clarification, how’s this line different from the holding in *Furman* as subsequently clarified in *Gregg*? The point of that jurisprudence was, iirc, that there’s nothing inherently discriminatory about capital punishment and that we just have to carefully tend to the details. (Finally, congratulations on the paper!)
Thanks, Fritz. We agree that these are two of the central questions for the sort of distributive-justice argument we develop. Van den Haag looms large over the second half of our paper. Rather than rehearse all the gory details here, let me just refer to you a few of the most pertinent passages
Section IV (pp.534-7) considers and rejects alternative responses to the current injustices (feasible options for putting more white folks, and folks convicted of killing black folks, on death row run into other difficult-to-surmount obstacles; doable reforms are either (or both) unlikely to address the injustice or (and) introduce unjust consequences of their own)
Section V (pp.537-541) responds specifically to van den Haag on the point you allude to
Section VI (541-543) then addresses the “proves too much” objection (does our argument ultimately entail prison abolition?) and the question whether eliminating DP involves an intolerable loss in noncomparative justice
Of course, our responses here may or may not be satisfying and some of the questions above, I think particularly from Victor and Ben, pushed us the related issues.
We discuss Furman and the history of jurisprudence on this issue especially in Section I (p.520) but then at various other places throughout the document