holding that courts may not rely on an arrest without a conviction or other "independent proof of conduct"Summary of this case from United States v. Reyes-Gomez
Rick Nemcik-Cruz, Rio Grande, PR, on brief for appellant. Rosa Emilia Rodríguez-Vélez, United States Attorney, Mariana E. Bauzá-Almonte, Assistant United States Attorney, Chief, Appellate Division, and Julia M. Meconiates, Assistant United States Attorney, on brief for appellee.
Rick Nemcik-Cruz, Rio Grande, PR, on brief for appellant.
Rosa Emilia Rodríguez-Vélez, United States Attorney, Mariana E. Bauzá-Almonte, Assistant United States Attorney, Chief, Appellate Division, and Julia M. Meconiates, Assistant United States Attorney, on brief for appellee.
Before Torruella, Boudin, and Barron, Circuit Judges.
BOUDIN, Circuit Judge.
Police officers in Puerto Rico received a tip as to the whereabouts of Josue Marrero-Pérez ("Marrero"), who had left Delaware to evade arrest on an outstanding warrant. Locating him at his premises in Puerto Rico, they discovered in plain view in the building a Glock pistol and a Smith & Wesson semiautomatic pistol, each loaded with ammunition. They arrested him and a grand jury indicted him on two counts of possessing a firearm while prohibited from doing so because of a prior felony conviction, 18 U.S.C. § 922(g)(1), and because of his fugitive status, id. § 922(g)(2).
Without a plea bargain, Marrero pled guilty to both charges. The final presentence investigation report ("PSR") recommended that the guidelines range be fixed at 37-46 months in prison, based on a recommended total offense level of 17 and a recommended criminal history category of IV. The probation officer's report set out an ominous criminal record, some of whose detail and characterizations Marrero disputes in this appeal. According to the report, as a juvenile Marrero behaved violently and was arrested regularly, and as an adult he had been convicted twenty times including for assault, drug violations, resisting arrests and possessing a firearm with an obliterated serial number.
The PSR advised that Marrero's category IV designation substantially understated his criminal history and likelihood of further crimes, suggesting as options an upward departure or a variant sentence. The PSR had been available to Marrero and his lawyer but no objection was made to the report, and at sentencing defense counsel stated that he had reviewed the PSR prior to that hearing. He did not object to the report at any time before the sentence.
At sentencing, after hearing Marrero's allocution and his lawyer's request for a sentence at the lower end of the proposed range, the court said that Marrero's record called for an upward departure. See U.S.S.G. § 4A1.3(a)(1). As the judge concluded at sentencing,
Mr. Marrero's vast prior record, as reflected in the pre-sentence investigation report and history of convictions, indicates that his Criminal History Category substantially underrepresents the seriousness of his offense or the likelihood that he will commit other crimes.... Mr. Marrero has been arrested at least 44 times during his life.... As an adult, Mr. Marrero has more than 20 convictions.... According to documentation received from the probation officer, in the District of Delaware, Mr. Marrero has the following known history of warrants: 25 failure to appear, 13 violations of probation, and 28 failures to pay, all of which is consistent with Mr. Marrero's utter disregard for the law and high[ ] likelihood of recidivism.
The district court imposed a variant sentence of 72 months, itself in excess of the 60-month upward variance proposed by the government. The most difficult issue on this appeal concerns the possibility that the judge at sentencing relied in his upward variance on prior arrests of the defendant which did not result in convictions; Marrero also objects to a lack of detail in other cases where the conviction is conceded.
In addition to the PSR, documents concerning outstanding warrants in Delaware were apparently provided by the probation officer to the court; this information was not included in the PSR but was referred to by the court at sentencing. Marrero now suggests that it is a "fair inference" from the sentencing transcript that these documents were provided to the court ex parte and should have been tested at a hearing.
Marrero's strongest arguments are that the sentencing judge (1) relied on past arrests listed in the PSR (usually involving at worst trivial conduct), many of which did not result in prosecution or conviction, and (2) relied on ex parte documents which were not timely disclosed to the defense.
We agree that as a matter of judicial policy, in this case and henceforth, no weight should be given in sentencing to arrests not buttressed by convictions or independent proof of conduct. See U.S.S.G. § 4A1.3(a)(3). In certain perhaps rare cases, a reasonable person might in particular circumstances assign some weight to a collection of arrests, but no such argument is made or available here. To rely on acquitted conduct in sentencing was troublesome enough even where the government needed to provide only preponderant proof rather than proof beyond a reasonable doubt, see United States v. Watts, 519 U.S. 148, 152, 117 S.Ct. 633, 136 L.Ed.2d 554 (1997) (per curiam); proof only of an arrest is no proof of guilt.
As for lack of detail, the PSR offers nothing as to the substance of the juvenile conduct for seven of the ten arrests; and no facts are provided for ten of the twenty adult convictions beyond the fact of conviction; whether in most cases more is required is more doubtful. But other arrests resulted in proceedings that were nolle prossed or otherwise terminated without a trial or a guilty plea.
Subtracting allegations in the PSR for which there are no detailed records or for which no conviction resulted, much remains that is unhelpful to Marrero, and serious prior crimes and recidivist behavior are of course proper considerations at sentencing. 18 U.S.C. § 3553(a)(1)-(2)(C). How unhelpful is a judgment call largely confided to the sentencing judge, who is in a good position to make such individualized assessments. See Gall v. United States, 552 U.S. 38, 51-52, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007).
Reliance on arrests alone or other dubious inferences to enhance a criminal sentence might be thought to offend the often invoked "presumption of innocence," but historically the presumption has governed only the trial itself. United States v. Doyle, 348 F.2d 715, 721 (2d Cir. 1965) (Friendly, J.). Thus, at sentencing judges have long been free to rely on anything useful that is established by a preponderance of the evidence, without regard to ordinary rules of trial evidence.
See McMillan v. Pennsylvania, 477 U.S. 79, 91, 106 S.Ct. 2411, 91 L.Ed.2d 67 (1986) ; Taylor v. Kentucky, 436 U.S. 478, 483 n.12, 98 S.Ct. 1930, 56 L.Ed.2d 468 (1978) ; United States v. Gonzalez-Vazquez, 34 F.3d 19, 25 (1st Cir. 1994) ; see also In re Whitney, 421 F.2d 337, 338 (1st Cir. 1970) (presumption of innocence does not apply to probation revocation).
The line between the criminal trial itself and almost everything afterwards is sharp, enshrining the care and protection afforded to the defendant at the trial. The comparative informality of everything after the guilty verdict has practical advantages and is too well settled for debate. The Supreme Court may on a rare occasion arguably seem to extend the presumption of innocence beyond its historical core, see, e.g., Nelson v. Colorado, ––– U.S. ––––, 137 S.Ct. 1249, 197 L.Ed.2d 611 (2017), but not yet in any discernable pattern directly relevant to a sentencing hearing.
Marrero's failure to object to the sentence based on the judge's possible reliance on arrests not resulting in conviction does not insulate the district judge's purported error from review by this court—even under the plain error test—and this circuit has several times warned over a three-year period against reliance on arrests as a proxy for criminal culpability or the likelihood of recidivism. United States v. Rondón-García, 886 F.3d 14, 25-26 (1st Cir. 2018) ; United States v. Delgado-Sánchez, 849 F.3d 1, 13 (1st Cir. 2017) ; United States v. Cortés-Medina, 819 F.3d 566, 570 (1st Cir. 2016). In sum, to equate arrest with guilt is by now both error and obviously so.
Material about past offenses may be furnished wholesale to the probation officer by the government's lawyer and probably as often adopted by the probation officer without much scrutiny—counting on defense counsel to object if warranted. So far as appears, no such objection was advanced here, so the judge ended up with a package of allegations, correctly believing that defense counsel never disputed the underlying facts or the negative inferences one might draw from them. The crimes proved by adult convictions, without more, may well have justified an upward variance. U.S.S.G. § 4A1.3(a)(1). This is a classic issue for the district judge and would be tested, even if the objection had been made and preserved, only for abuse of discretion. See United States v. Vázquez-Martínez, 812 F.3d 18, 26 (1st Cir. 2016).
Given the lack of objection, the ordinary test is for plain error, United States v. Sosa-González, 900 F.3d 1, 4 (1st Cir. 2018), which requires that a defendant show that (1) an error occurred, (2) the error was obvious, (3) the error affected substantial rights, and (4) the error seriously impaired the fairness, integrity, or public reputation of judicial proceedings, see United States v. Constant, 814 F.3d 570, 580 (1st Cir. 2016). Other statements by this court stress particular aspects of the test pertinent to the case at hand, but the stress is almost always on one or more of the factors set forth in Constant.
See Rondón-García, 886 F.3d at 24 (noting that for an error to have affected a defendant's substantial rights, a defendant must show that there is a "reasonable probability that, but for the error, the district court would have imposed a more favorable sentence" (internal quotation marks omitted)); United States v. Saxena, 229 F.3d 1, 5 (1st Cir. 2000) (describing the final prong of the plain error test as a "miscarriage-of-justice standard" (citing United States v. Alicea, 205 F.3d 480, 484 (1st Cir. 2000) ) ).
Applying the Constant formulation, we think that an error occurs when a district judge relies on an arrest report, without some greater indicia of reliability that the conduct underlying the arrest took place; the "obviousness" of the error is a matter of degree but becomes greater with every new case; whether the error affects the outcome depends on the facts peculiar to the case (e.g., the presence or absence of corroborating evidence in the record concerning the conduct underlying the arrests without convictions); and the fourth factor—the "miscarriage of justice" standard, Alicea, 205 F.3d at 484 —is implicated if the sentence has truly been altered by the arrests, itself a matter that the district judge can readily lay to rest on remand.
While the record permitted a variance upward on the basis of convictions, we think that Marrero has made a strong enough case that the sentencing judge relied on these past arrests in determining the sentence. Here, the specific conduct underlying the dismissed charges was not, as sometimes happens, set forth in undisputed portions the PSR. See United States v. Mercer, 834 F.3d 39, 50 (1st Cir. 2016). Finally, the sentencing judge stated on the record that "Mr. Marrero's vast prior record" and the fact that he "has been arrested at least 44 times during his life," along with other factors, were "consistent with [his] utter disregard for the law and high[ ] likelihood of recidivism."
It is enough to warrant a remand that the reliance on such arrests—now an obvious error—"set the wrong framework for the sentencing proceedings," Molina-Martinez v. United States, ––– U.S. ––––, 136 S.Ct. 1338, 1345, 194 L.Ed.2d 444 (2016). Marrero is not in turn required to show that this obvious error in fact influenced the length of the sentence, a showing that in most cases will be nearly impossible for a defendant to make given the kitchen-sink approach to sentencing pronouncements adopted by many judges. Cf. id. at 1347.
While the lack of objections throughout usually would defeat a defendant's effort to seek a remand, this course would merely defer the matter to a likely habeas proceeding directed at trial counsel's competence. A predictable habeas petition would prolong and further complicate proceedings and ultimately cost time and trouble for judge and counsel alike, and the defendant would already be serving his sentence when, a year or two later, he would discover its duration.
Our concerns are reinforced not by the lack of detail regarding the facts underlying the convictions—that is common, see, e.g., United States v. Turbides-Leonardo, 468 F.3d 34, 37 (1st Cir. 2006), and only occasionally troublesome—but by the second argument raised by Marrero, that is, the still not fully explained material submitted ex parte by the probation officer to the court concerning certain outstanding warrants in Delaware. The record is not clear on this point, but it appears that defense counsel did not object at the sentencing hearing. Rather, at the end of miscellany (e.g., which prison the judge might recommend for Marrero), defense counsel said only that he wanted some documentation that the probation officer had employed (it appears that defense counsel had declined earlier to review it). He did ask for his own copy at the end of the hearing and this the court granted.
A remanding appeals court has authority to shape a remand as circumstances warrant, see United States v. Correy, 570 F.3d 373, 378 (1st Cir. 2009). Whether or not the conduct underlying the prior arrests can be shown by a preponderance of the evidence —and whether that matters to the district judge—is for him to say on remand, but either way public confidence will be served by an explicit answer. The defendant can also pursue the issue of the basis and impact of the ex parte material; but here the district judge is free to take into account the lack of timely objection.
Ex parte communication between the probation officer and the court is usually permissible where the court is merely seeking advice or analysis, United States v. Bramley, 847 F.3d 1, 7 (1st Cir. 2017) ; to some extent the probation office is an extension of the district court, see United States v. Saxena, 229 F.3d 1, 5 n.1 (1st Cir. 2000), and the probation officer and the court may consult privately about certain issues incident to criminal sentencing, see United States v. Fraza, 106 F.3d 1050, 1056 (1st Cir. 1997). But where the probation officer discloses new facts that bear on the judge's sentencing calculus, cf. United States v. Craven, 239 F.3d 91, 102 (1st Cir. 2001), the general rule requires disclosure to the defense in advance of the sentencing hearing and an opportunity to subject the new material "to whatever adversarial testing may be appropriate," Bramley, 847 F.3d at 7.
The ex parte material in this case, the outstanding warrants in Delaware, are pretty thin as against defendant's own criminal record but might suggest a proclivity for lawlessness or at best a disrespect for the court system. If the information about the warrants was not included elsewhere in the record or made known to Marrero prior to sentencing, reliance on this information at sentencing would also have been error. See United States v. Millán-Isaac, 749 F.3d 57, 70 (1st Cir. 2014).
Marrero's counsel may well have received adequate advanced notice of the ex parte material and simply declined to review it, as was true with other material; Marrero says only that it is a "fair inference" from the sentencing transcript that the material was provided ex parte. Whatever happened, the facts and the consequences are for the district court's consideration on remand.
The sentence is vacated and the matter remanded for resentencing in accordance with this decision. It is so ordered.