Thursday, May 26, 2011
Article of Interest: Recognizing Constitutional Rights at Sentencing, by Carissa Hessick and Andy Hessick
In Tinker v. Des Moines Independent Community School Dist., 393 U.S. 503, 506 (1969), the Supreme Court famously found that "[i]t can hardly be argued that either students or teachers shed their constitutional rights...at the schoolhouse gate." In their terrific new article, Recognizing Constitutional Rights at Sentencing, 99 Cal. L. Rev. 47 (2011), Carissa Hessick and Andy Hessick, professors at the Arizona State University College of Law, argue that the same applies to criminal defendants who have been convicted and then go through the courthouse doors to the sentencing hearing. The purpose of this post is two-fold: (1) to lay out the reasoning of their article; and (2) to apply the reasoning to victim impact statements.
Might Mean Rights
In Part I of the article, the authors begin by tracing the history of sentencing in this country and then note that "[t]hough most modern sentencing schemes afford judges discretion in imposing sentences, in recent times courts have nonetheless begun to recognize some constitutional limitations to that discretion."
As they note, "[m]ost of the constitutional rights recognized at sentencing have been procedural," including:
•the right to effective representation by counsel;
•(a limited version of) the privilege against self-incrimination;
•the due process right to advance notice of facts that a court relies on to impose a higher sentence; and
The authors go on to point out, though, that "[c]ourts have also placed some substantive limitations on what courts may consider in imposing sentence," such as precluding courts from:
•basing a sentence on "materially untrue" assumptions about the defendant’s criminal record;
•considering a successful appeal of a conviction at resetencing;
•imposing a higher sentence to punish the defendant for successfully appealing her original conviction;
•considering race, national origin, or gender at sentencing;
•giving harsher sentences to nonresident defendants; and
•(according to some courts) allowing information about a defendant's religion (or lack thereof) at sentencing.
Few and Far Between
In Part II of the article, the authors begin by cautioning that while
the preceding Part might convey the impression that courts enforce the Constitution as rigorously at sentencing as in criminal trials, that is not accurate. Constitutional limitations on sentencing factors are the exception rather than the rule. For the most part, courts have not recognized constitutional limitations on what information may be considered at sentencing....The consequence is that courts have systematically failed to recognize a number of constitutional rights at sentencing that are otherwise recognized during trial....Instead of engaging in ordinary constitutional analysis when defendants challenge these factors, courts have swept constitutional concerns under the proverbial rug based on the ungrounded conclusion that the sentencing process is somehow unique and thus shielded from constitutional review.
The authors then highlight five sentencing factors, with the purpose of revealing "that there is at least a colorable constitutional objection to a number of traditional sentencing factors, and that courts have failed to grapple with these objections."
•Double Jeopardy: A number of jurisdictions permit the consideration at sentencing of criminal conduct for which the jury entered an acquittal. These courts conclude that such consideration does not violate the Double Jeopardy Clause by reasoning, inter alia, that such a sentencing enhancement does not constitute "punishment" for the acquitted conduct. For the authors, this explanation is unsatisfactory because when a sentence is enhanced based on acquitted conduct, the same conduct is the basis for both (1) the government’s first, unsuccessful effort to punish the defendant; and (2) the sentence enhancement applied after the government's second, successful attempt to punish the defendant for a different crime (for more on this issue, check out their article, Double Jeopardy as a Limit on Punishment (forthcoming Cornell Law Review));
•Guarantee of a Fair Trial: While the Sixth Amendment guarantees defendants the right to a fair trial, courts consistently allow sentencing enhancements for the aforementioned acquitted conduct as well as uncharged conduct. Such enhancements are troubling under the bedrock principle of the Sixth Amendment: that punishment for criminal conduct be based only on conviction by a jury. The same goes for courts, including the Supreme Court, upholding the imposition of longer sentences on those who refuse to plead guilty and have the temerity to go to trial, i.e., the "trial penalty." Courts defend these sentences by conceptualizing these longer sentences not as additional punishment, but instead as the absence of leniency afforded to those who plead guilty. But, according to the authors, the refusal to grant a defendant leniency that is given to others because the defendant performed some act—exercising his right to a jury—seems analytically indistinct from increasing that defendant’s punishment because he performed that same act;
•Self-Incrimination: In Mitchell v. United States, 526 U.S. 314, 328 (1999), the Supreme Court did find that under the Fifth Amendment a judge could not draw an adverse inference about the amount of drugs the defendant was involved in selling based on her silence. The Court, however, limited its holding "to factual determinations respecting the circumstances and details of the crime." Based on the Court's limited guidance, many courts have since concluded that judges can impose longer sentences when defendants fail to express remorse for their crimes. But for a defendant to express remorse for committing a crime, she must admit to having committed that crime, creating a possible Fifth Amendment violation;
•Free Speech: These remorse rulings as well as sentencing enhancements for defendants who affirmatively testify to a lack of remorse also raise First Amendment free speech issues. Courts have navigated around the First Amendment by not conceptualizing lack of remorse as a free speech issue, but the authors contend that a regulation of speech designed to reduce secondary effects -- such as recidvism -- is still a regulation of speech;
•Due Process: Courts frequently enhance sentences based upon predictions of future dangerousness, which the authors claim violates the Due Process requirements (1) that individuals only be punished based upon past acts; and (2) that individuals be given notice such that they can conform their behavior and avoid increased punishment.
Breaking Down the Defense(s)
In Part III, the authors challenge the three commonly cited justifications "for why constitutional rights ought not be recognized at sentencing."
•Historical Practice: As the authors note, courts first defend these sentencing practices by claiming that courts have historically used them. But, as they also note, this is not an originalism argument because courts at the founding engaged in determinate sentencing. The authors acknowledge that the Supreme Court has held that a "universal and long-established tradition" of allowing certain conduct may be reason to presume that the conduct is constitutional, but they find no such entrenched tradition with regard to the aforementioned sentencing enhancements. Indeed, they note that the recent shift from courts viewing sentencing as a wholly informal proceeding to courts increasingly recognizing procedural rights at sentencing suggests a growing commitment to protecting rights at sentencing rather than to disregarding them;
•Forfeiture: Courts used to find that defendants, by being convicted, forfeited their constitutional rights at sentencing, but the authors are hardly surprised that modern courts have not relied on the forfeiture theory in rejecting constitutional challenges to sentencing factors. They point to a variety of reasons for this reluctance, such as the fact that the Eighth Amendment prohibition on cruel and unusual punishments can only apply after a defendant has been convicted. The main thrust of their argument is that "[a]t its core, the theory of forfeiture rests on the assumption that the Constitution’s protections against the government apply only at criminal trial and not at sentencing after the defendant is found guilty of a crime. This assumption may have made sense when sentences were determined largely by the crime of conviction, but it is no longer sound now that sentences depend on factors other than the fact of conviction;"
•Incompatible With the Goals of Sentencing: (1) Information Maximization: Relying on Williams v. New York, 337 U.S. 241 (1949), courts have upheld reliance on questionable sentencing factors by arguing that more information is better and that a judge's sentencing ability would be hampered if the information stream were cut off. In addition to pointing out that Williams has been overruled, the authors level three criticisms against this information maximization defense: (a) it turns constitutional law on its head by sacrificing individual rights for governmental goals; (b) it ignores the due process limitation on unbridled judicial discretion; and (c) it falsely assumes that there is a single correct sentence for each defendant
(2) The Theories of Punishment: One final defense of constitutionally questionable sentencing factors is that consideration of those factors is necessary to impose sentences that appropriately punish defendants under the various theories of punishment. (a) One main theory of punishment is retributivism, but the authors contend, inter alia, that future dangerousness is inconsistent with this theory of punishment, and they allege more generally that retributivism does not comfortably allow for punishment based on constitutionally protected conduct. (b) The other main theory is utilitarianism, and the authors contend that courts have not articulated any clear theory as to why questionable evidence should be more readily admissible at the sentencing stage of trial than the guilt stage of trial under this theory.
Based upon this analysis, the authors propose two possible solutions: (1) simply to exclude consideration of constitutionally doubtful sentencing factors; and (2) for judges to undertake a close analysis of the constitutionality of the sentencing factors that they identified. The authors advocate for this second solution, which I think makes a great deal of sense. As they note,
a judicial finding that a particular sentencing factor impinges on a defendant’s constitutional rights need not automatically result in the exclusion of that sentencing factor. Constitutional rights are not absolute. Courts could subject constitutionally doubtful sentencing factors to the same scrutiny that would apply in other contexts. For example, in determining whether to enhance a sentence for lack of remorse, courts would apply heightened scrutiny, the standard usually applied in assessing the constitutionality of evaluating laws coercing speech.
As they go on to note, there are costs to society if courts cannot consider sentencing factors because it would mean a shift back to more determinate sentencing, with some defendants being punished too leniently and others being punished too harshly. But exposing sentencing factors to heightened scrutiny,
may actually reduce these costs to society. To return to the previous example involving lack of remorse, if courts concluded that enhancements for lack of remorse infringe the First Amendment, courts could still impose the enhancement if the government demonstrated a sufficiently high correlation between a defendant’s refusal to express remorse and a propensity to commit future crime. Making this showing would require studies on whether lack of remorse is, in fact, an accurate predictor of recidivism. Such studies may reduce social costs if they reveal that some traditional sentencing factors that have not been subjected to empirical study are not accurate predictors of recidivism. This effort would lead the government to identify those factors that are reliable predictors of recidivism instead of relying on intuition.
I think that, as with many of the authors' previous articles (see, e.g., here, here, and here), this piece makes a significant contribution to the tumultuous state of the law surrounding sentencing, and I strongly recommend that readers check out the full article for the authors' complete analysis.
Victim Impact Evidence
The authors consciously chose to steer clear of procedural, issues, so I thought that I would spend the second part of this post preliminarily applying their analysis to victim impact evidence.
In Booth v. Maryland, 482 U.S. 496 (1987), the Supreme Court held that the Eighth Amendment per se prohibits a capital sentencing jury from considering victim impact evidence. Four years later, in Payne v. Tennessee, 501 U.S. 808 (1991), a defendant was convicted of murdering a 3 year-old boy's baby sister and mother. At the capital sentencing hearing, the boy's grandmother testified that the boy missed his baby sister and mother, and "the prosecutor commented on the continuing effects on [the boy] of his experience and on the effects of the crimes upon the victims' family." In affirming the defendant's death sentence, the Payne Court overruled Booth, found that the Eighth Amendment did not per se prohibit the consideration of victim impact evidence, and held that "[i]n the event that evidence is introduced that is so unduly prejudicial that it renders the trial fundamentally unfair, the Due Process Clause of the Fourteenth Amendment provides a mechanism for relief."
I see victim impact evidence as similar to future dangerousness evidence. As the authors noted in their article, a few defendants have challenged judicial findings of future dangerousness on due process grounds, but they'e run into a wall: Jurek v. Texas, 428 U.S. 262 (1976), in which the Supreme Court held that "prediction of future criminal conduct is an essential element in many of the decisions rendered throughout our criminal justice system."
The score's somewhat the same with defendant's challenging victim impact evidence. While the Payne Court did declare that victim evidence can violate the Due Process Clause, it also held that "[i]n the majority of cases,...victim impact evidence serves entirely legitimate purposes." Inevitably, then, when defendants claim that victim impact evidence violates the Due Process Clause, courts cite to Payne, not to hold that the evidence violated the Due Process Clause, but to find that the evidence was presumptively admissible.
Indeed, the Payne Court's due process language has lay fallow for the last two decades, with courts viewing the Clause as a speed bump that victim impact evidence almost always clears rather than a hurdle that must be carefully crossed. And this might be okay if victim impact evidence in most cases today bore any resemblance to the grandmother's words in Payne. But it doen't. For example, take a look at the 20+ minute professionally produced victim impact statement video from Kelly v. California, a case in which the Supreme Court did not grant cert.
I've written about Kelly on a few occasions (see here here, and here). It presents a nice microcosm of the problems that have arisen as courts try to apply the Payne due process analysis in the absence of any real guidance. In Kelly, the prosecution's victim impact evidence was accompanied by music from Enya. The defendant cited to United States v. Sampson, 335 F.Supp.2d 166, 191 (D. Mass. 2004), one of the rare cases in which a court has found that victim impact evidence violated the Due Process Clause. In Sampson, the videotaped victim impact statement had music from the Beatles and James Taylor, and this "evocative contemporary music" rendered the sentencing hearing fundamentally unfair according to the court. The court in Kelly, however, found that the Enya music in the video before it was "generally soft, not stirring," with most of the words unrecognizable, meaning that it didn't render the sentencing hearing fundamentally unfair.
While, as noted, the Supreme Court denied cert in Kelly, Justice Stevens dissented, arguing, inter alia, that
even under the rule announced in Payne, the prosecution's ability to admit such powerful and prejudicial evidence is not boundless. These videos are a far cry from the written victim impact evidence at issue in Booth and the brief oral testimony condoned in Payne. In their form, length, and scope, they vastly exceed the 'quick glimpse' the Court's majority contemplated when it overruled Booth in 1991. At the very least, the petitions now before us invite the Court to apply the standard announced in Payne, and to provide the lower courts with long-overdue guidance on the scope of admissible victim impact evidence. Having decided to tolerate the introduction of evidence that puts a heavy thumb on the prosecutor's side of the scale in death cases, the Court has a duty to consider what reasonable limits should be placed on its use.
So, why do I agree with Justice Stevens' based upon the above article? Well, let's start with the issue of why victim impact evidence might violate the Due Process Clause, a question that courts have not clearly answered.
Obviously, the use of manipulative music and images can render a sentencing hearing fundamentally unfair, which seemed to the focus of the Payne Court. But what about the issue of notice? The authors claimed that future dangerousness evidence might violate the due process requirement that individuals be given notice such that they can conform their behavior and avoid increased punishment. Are defendants given such notice of victim impact evidence?
I think that the answer is a clear "no." Victim impact evidence is a non-statutory aggravating factor, which means that many defendants might not know that their punishments can be increased based upon the impact that their killings have on the victims' families. Moreover, "[t]he courts of appeals have split on the need for advance notice of an upward departure based onvictim impact statements." Paul G. Cassell, Treating Crime Victims Fairly: Integrating Victims Into the Federal Rules of Criminal Procedure, 2007 Utah. L. Rev. 861, 937 (2007). Finally, even if we think that most people are now aware, and thus have notice, that victim impact evidence is admissible,
notice that a sentence may rest on pernicious factors does not cure the factors' perniciousness. It would not, for example, cure an equal protection problem to give defendants notice that they are more likely to receive death sentences if they or their victims are black, rather than white. Moreover, in the capital sentencing context, even aggravating factors that are permissible must be specified in advance, by the legislature. Susan Bandes, Reply to Paul Cassell, What We Know About Victim Impact Statements, 1999 Utah L. Rev. 545, 546 n.8 (1999).
Let's look next at the historical practice defense. As the authors noted, some courts defend the continued use of certain sentencing factors such as future dangerousness because they have historically considered such factors. But, as noted, victim impact evidence is not nearly so firmly rooted. It wasn't until 1979, that the first court allowed information regarding the impact of a crime, and the Supreme Court did not approve of victim impact evidence until Payne in 1991. See Michael A. Johnson, Note, The Application of Victim Impact Statements in Capital Cases in the Aftermath of Booth v. Maryland: An Impact No More, 13 T. Marshall L. Rev. 109, 111 (1987-1988).
In terms of forfeiture, victim impact evidence is only potentially admissible at sentencing, so, applying the authors' analysis, it makes no sense to say that the defendant forfeits his Eighth Amendment and due process objections to such evidence by being convicted.
So, what about theories of punishment? A frequent defense of victim impact evidence is that it is necessary to punish individuals pursuant to the retributionist theory of punishment; indeed, some have claimed that "he only justification to which victim impact evidence directly relates is retribution." Catherine Bendor, Defendants' Wrongs and Victims' Rights: Payne v. Tennessee, 11 S.Ct. 2597, 27 Harv. C.R.-C.L. Rev. 219, 234 (1992). But other disagree. For instance, in State, Be Not Proud: A Retributivist Defense of the Commutation of Death Row and the Abolition of the Death Penalty, 40 Harv. C.R.-C.L. 407, 455 (2007), Dan Markel argues that
most retributivists would demur at lending victims a direct role in sentencing, say, by permitting a victim impact statement to be read prior to sentencing. The concern is that the judge or jury would be swayed to harshness or leniency as a result, and the problem with that, from a retributive perspective, is the disruption of the equality norm. Specifically, if victims or their survivors have a role that directly influences sentences, then the sentence that an offender receives may hinge on whether the jury finds the victim or his allies persuasive or sympathetic. Just as one victim's family may urge especial harshness against an offender, another victim may be surpassingly compassionate, in which case the offender may receive an unwarranted request for no punishment or unusually lenient punishment. This advantages him as compared to similarly situated offenders who committed the same offense.
And what about the utilitarian theory of punishment? Well, if we believe the Bender quote from above, victim impact evidence doesn't directly relate to utilitarianism. Indeed, in The Impact of Information Overload on the Capital Jury's Ability to Assess Aggravating and Mitigating Factors, 17 Wm. & Mary Bill Rts. J. 1089, 1125-26 (2009), Katie Morgan and Michael J. Zydney Mannheimer argue that capital "jurors are often asked to weigh incommensurables: the utilitarian-based idea that very dangerous offenders must be permanently incapacitated by death, added to the harm-based retributivist values reflected by victim impact evidence, balanced against the intent-based retributivist notion of diminished culpability often underlying the mitigating evidence."
It thus seems like there is a solid argument that victim impact evidence cannot be defended on either retributionist or utilitarian grounds. It therefore seems to be on somewhat equal footing with future dangerousness evidence. So, what's the solution?
Well, as the authors note, there are two possible solutions. The first is to ban victim impact evidence altogether (see, e.g., here). The other, probably preferable, solution is to subject such evidence "to the same scrutiny that would apply in other contexts." In other words, the solution is to finally flesh out the test first laid out by the Court in Payne. Let's return to the Markel article from above. He notes that:
Undoubtedly, it is better that victims or survivors not be re-traumatized and that they find the psychological healing they need. A good state and the rich panoply of societal institutions within it should make this therapy available to the extent possible. But it is purely speculative—if not illusory—to assert that the death or cruel punishment of the offender will necessarily achieve that healing, or that other means will not provide the closure victims seek. Indeed, there "is no evidence that families of murder victims in non-death states such as Michigan or Wisconsin endure more lasting pain than families of murder victims in death states such as Texas or Ohio." In other words, the death penalty is not a means narrowly tailored to satisfy the state's legitimate end of healing the wounds of crime victims. Moreover, to the extent that victims participate in the decision to impose the death penalty, and to the extent the death penalty is imposed because of the positive personal characteristics of the victim or the economic and social dislocations caused by a particular victim's death, then the imposition of death will be distributed on the basis of morally immaterial features of the victim or her family. Thus, the use ofvictim impact evidence undercuts the retributivist commitment to the fair and equal application of criminal sanctions. (emphasis added).
Now, a court may agree or disagree with Markel about whether victim impact evidence is narrowly tailored to satisfy legitimate state interests, but the point is that they should apply a similar type of scrutiny to victim impact evidence rather than just glibly citing to Payne and engaging in matador style judging. And again, that's exactly the point of the authors' article. I asked the authors what led them to write the article, and Carissa Hessick responded that
Andy and I have collaborated on a few different articles --- being married to another law professor means that we spend *a lot* of time talking about law with one another. A few times, when talking through a problem with our research or a case one of us read, we've hit upon a topic on which we've both been able to contribute. This article, for example, arose out of a problem that I confronted writing another paper. Andy and I were having dinner, I was complaining about a series of cases that I'd been reading that day, and he observed that the issue bothering me was the fact that government actors were permitted to use facts and factors to adjust the amount of punishment a person receives, even when that fact or factor, standing alone, would have been an unconstitutional ground for punishment.
I'm very interested to see your post on due process and victim impact statements. Our article you read specifically skirted procedural issues, but I'm working on a new paper about sentencing inputs and procedures so I'm quite interested in seeing what you have to say.