People v. Williams

Pigott, J. (dissenting in Williams, Hernandez, Lewis and Rodriguez and concurring in Echevarria).

According to the majority, a defendant who knows that the sentence he received is illegal, because it contravenes Penal Law § 70.45, nevertheless acquires a legitimate expectation in the finality of his sentence simply by virtue of being released from prison. Because that holding conflicts with United States Supreme Court jurisprudence, I dissent.

To begin with, I agree with the majority that there was no statutory impediment to the imposition of postrelease supervision on the defendants in these cases, after they had completed their terms of imprisonment (see majority op at 213). But I cannot accept the majority’s conclusion that imposing postrelease supervision on defendants in Correction Law § 601-d resentencing proceedings violates their rights under the Double Jeopardy Clause of the United States Constitution.

The majority concedes that “[sjince criminal defendants are charged with knowledge of the relevant laws that apply to them, they are presumed to be aware that a determinate prison sentence without a term of [postrelease supervision] is illegal and, thus, may be corrected by the sentencing court at some point in the future” (majority op at 217 [citation omitted]). But the majority fails to follow these propositions to their logical conclusion. A defendant who knows that the sentence he was given is illegal and is subject to correction cannot claim to have a legitimate expectation that the sentence will remain uncorrected. There can be no reasonable expectation of finality in a sentence that is less severe than required by the law.

The Supreme Court expressly held in Bozza v United States ( 330 US 160 [1947]) that a defendant has no legitimate expectation of finality in an unlawful sentence. In Bozza , the Court *226 ruled that a court does not violate a defendant’s double jeopardy protections when it merely “set[s] aside what it had no authority to do and substitute[s] directions required by the law” ( 330 US at 167 ). Bozza was convicted of operating an illegal still, a crime carrying a mandatory sentence of a $100 fine and a term in prison. The trial court originally sentenced the defendant only to the term of imprisonment. When the court realized its mistake, it resentenced defendant, imposing the $100 fine as well as the prison term. The Supreme Court charged the defendant with knowledge that the court lacked the authority to impose only imprisonment, and held that he had no legitimate expectation of finality in the original, illegal sentence (id.). “The Constitution does not require that sentencing should be a game in which a wrong move by the judge means immunity for the prisoner” ( 330 US at 166-167 ; see also Jones v Thomas, 491 US 376, 387 [1989]). *

Applying Bozza , it follows that the defendants in the cases before us cannot claim a legitimate expectation of finality in their sentences. All are, as the majority concedes, presumed to be aware that their determinate prison sentences lacking post-release supervision are illegal and, thus, subject to correction (see majority op at 217). Therefore, none may claim objectively good reason to believe that his sentence would not be corrected. And defendant Williams can have had no expectation of finality for another reason; he acknowledges that he agreed in his plea agreement to serve postrelease supervision. A defendant who was “mistakenly sentenced to a lesser term than he agreed to” does not “acquire a vested interest in the error so that it would be unfair, under the double jeopardy clause, to correct the error and make the defendant serve out the term of his own sentencing agreement” (People v Minaya, 54 NY2d 360, 366 [1981]; see also People v Williams, 87 NY2d 1014, 1015 [1996]).

In United States v DiFrancesco ( 449 US 117 [1980]), the Supreme Court held that a defendant was charged with *227 knowledge that his sentence was subject to appeal by the government—under a federal statute permitting sentence enhancement of “dangerous special offenders”—and therefore could not have a legitimate expectation of finality (DiFrancesco, 449 US at 136, 139 ). DiFrancesco did not involve correction of an illegal sentence, but its rationale applies here, where the defendants are charged with knowledge that their sentences, if illegally imposed, are subject to correction (see generally United States v Crawford, 769 F2d 253, 257 [5th Cir 1985]).

Given the Supreme Court’s holdings in Bozza and DiFrancesco , it is clear that a defendant has a legitimate expectation in the finality of his sentence—that is, a reasonable expectation that the sentence he received will not be modified so as to become more severe—only when he has objective reason to believe that it will not be changed or corrected. The majority’s holding that “there is a legitimate expectation of finality once the initial sentence has been served and the direct appeal has been completed (or the time to appeal has expired)” (majority op at 217) fundamentally misunderstands the concept of an expectation of finality in double jeopardy jurisprudence. A defendant cannot acquire a legitimate expectation of finality from the mere fact that he has been released from prison. Factual or legal circumstances may exist that would undermine a reasonable person’s expectation in the finality of the sentence imposed, even after release.

Moreover, the majority’s requirement of two conditions for the attachment of a legitimate expectation of finality—the completion of “the initial sentence” and the completion of the direct appeal or time to appeal (majority op at 217)—implies that an incarcerated defendant cannot acquire a legitimate expectation of finality until his prison sentence is complete. This is also incorrect. Where circumstances do not undermine the expectation, a defendant will acquire a legitimate expectation of finality in his sentence as soon as the appeal process is complete and the one-year period has expired during which the People may move to set aside his sentence as invalid under CPU 440.40 (compare DiFrancesco, 449 US at 136 [expectation of finality attaches when time for government appeal has expired]). That period may expire before he is released.

The practical aspects of the litigation before us support the view that the defendants had no expectations of finality in their sentences. It is noteworthy that not one defendant has alleged that he was not informed of the postrelease supervision *228 component of his sentence upon commencement of his imprisonment (s ee Correction Law § 803 [6]), and not one denies that he was informed of the postrelease supervision before being conditionally released (see Penal Law § 70.45 [3]; Executive Law § 259-g [2]). Yet, defendants did not seek to challenge their post-release supervision when their prison sentences started or even when they were released. All the evidence suggests that from the moment they entered the prison system, if not before, defendants expected to serve postrelease supervision.

The majority insists that “there must be a temporal limitation on a court’s ability to resentence a defendant” (majority op at 217, citing DeWitt v Ventetoulo, 6 F3d 32, 34-35 [1st Cir 1993]). But DeWitt did not involve double jeopardy protections at all. That case held that a tardy, procedurally unique reinstatement of a life sentence, six years after it was suspended, violated a defendant’s due process rights, and the First Circuit went out of its way to note that although

“notions of fundamental fairness do place some temporal limit on later increases in sentence . . .

[o]nly in the extreme case can a court properly say that the later upward revision of a sentence, made to correct an earlier mistake, is so unfair that it must be deemed inconsistent with fundamental notions of fairness embodied in the Due Process Clause” ( 6 F3d at 35 [emphasis added]).

The majority does not reach defendants’ due process arguments, and I will not address them here, except to say that if the majority believes that notions of fundamental fairness embodied in the Due Process Clause mandate that an expectation of finality originates as soon as a defendant’s initial sentence has been served, it should say so.

In any case, the statement that “there must be a temporal limitation on a court’s ability to resentence a defendant” (majority op at 217) again misconstrues the notion of legitimate expectation of finality, which is constrained not by time, but by whether or not it would be reasonable for defendant to believe that his sentence will not be modified. If, in a particular case, a defendant cannot be hauled back to prison 10 years after he has been released, on the ground of an error in his sentence, that may be because defendant cannot reasonably be charged with having known that his sentence was in any way defective. Or, more likely, it may be because of fundamental fairness considerations embodied in due process rights. It is never merely because of a temporal limitation on resentencing that emerges from double jeopardy protections.

*229 Selecting a defendant’s conditional release date to be the time when an expectation of finality attaches is simply arbitrary.

“The Double Jeopardy Clause does not provide the defendant with the right to know at any specific moment in time what the exact limit of his punishment will turn out to be. Congress has established many types of criminal sanctions under which the defendant is unaware of the precise extent of his punishment for significant periods of time, or even for life, yet these sanctions have not been considered to be violative of the Clause.” (DiFrancesco, 449 US at 137 .)

Finally, the Supreme Court has often had occasion to repeat that “a sentence does not have the qualities of constitutional finality that attend an acquittal” (DiFrancesco, 449 US at 134 , citing Chaffin v Stynchcombe, 412 US 17 [1973]; North Carolina v Pearce, 395 US 711 [1969]; Bozza ; and Stroud v United States, 251 US 15 [1919]; see also Caspari v Bohlen, 510 US 383, 391-392 [1994]). Resentencing “does not involve a retrial or approximate the ordeal of a trial on the basic issue of guilt or innocence” (DiFrancesco, 449 US at 136 ). We should be more cautious before we deviate from the Supreme Court’s “traditional refusal to extend the Double Jeopardy Clause to sentencing” (Caspari, 510 US at 392 ).

For these reasons, I dissent in People v Williams, People v Hernandez, People v Lewis, and People v Rodriguez, and I concur in result in Matter of Echevarria v Marks.

Chief Judge Lippman and Judges Ciparick, Read and Jones concur with Judge Graffeo; Judge Smith dissents and votes to affirm in a separate opinion; Judge Pigott dissents in another opinion.

In People v Williams, People v Hernandez, People v Lewis and People v Rodriguez: Order reversed, etc.

Chief Judge Lippman and Judges Ciparick, Read and Jones concur with Judge Graffeo; Judge Smith concurs in result in a separate opinion; Judge Pigott concurs in result in another opinion.

In Matter of Echevarria v Marks: Judgment affirmed, without costs.

Although the majority suggests that Bozza should be limited to situations in which the error is “promptly” corrected (majority op at 215; see also majority op at 219), federal courts have given Bozza a much wider interpretation. “The full import of Bozza is that a trial court not only can alter a statutorily-invalid sentence in a way which might increase its severity, but must do so when the statute so provides” (Thompson v United States, 495 F2d 1304, 1306 [1st Cir 1974] [omission of statutorily mandated special parole term from sentence raised during appeal process]; see also e.g. Caille v United States, 487 F2d 614, 615 [5th Cir 1973] [same]; United States v Rourke, 984 F2d 1063, 1066 [10th Cir 1992] [same]).