concurring:
I fully concur in Judge Wallace’s careful analysis, and in his conclusion that the sentence in this case was flawed by procedural error. I add a few words only to put into ordinary language the common-sense basis of the sometimes arcane or technical analysis required by our somewhat complex law of sentencing.
On a superficial level, Jamar Preacely seems to have gotten a break. The Sentencing Guidelines recommended a sentence of at least 188 months in prison, and the district court, in recognition of the value to society of his cooperation with the authorities, sentenced him to just half that. That is not an atypical reduction in this circuit for turning state’s evidence in a narcotics case.
On closer examination, however, the guideline recommendation used by the district court as a baseline before rewarding Preacely for his help to the government appears distinctly inflated. Preacely pled guilty to possessing with intent to distribute more than five grams of crack cocaine; the total amount of the drug proved to have been in his possession was approximately twenty-one grams. I have no intention of minimizing the serious criminality of dealing in dangerous drugs— Preacely’s own sad, addicted life before his arrest illustrates the harm inflicted by narcotics trafficking. Nor should we be naive about the likelihood that what Preacely was caught doing was the full extent of his misbehavior. But even for a man with a history of multiple (if mostly minor) criminal convictions (almost exclusively tied to the possession and sale of narcotics), a sentence of nearly sixteen years in prison for the possession of a few thousand dollars worth of cocaine seems remarkably severe.
*84That conclusion is not merely my personal opinion. It is in fact the conclusion warranted by the ordinary application of the Sentencing Guidelines themselves. As Judge Wallace explains, under the complicated formulas applied by the Guidelines to determine the seriousness of offenses and the severity of an offender’s prior criminal history, Preacely’s offense would be treated as “level 23” and his criminal history as “category V.” For someone with these scores, the ordinary minimum recommended sentence at the time of Preacely’s sentencing (before any reduction for cooperation) was 84 months in prison, not 188. I do not pause to quibble over the fact that even this recommendation is affected by what many have believed, and what the Supreme Court, Congress, and the Sentencing Commission itself have now come to agree, is an excessively harsh response to the danger of crack (cocaine base) as compared to other illegal drugs. See Fair Sentencing Act of 2010, Pub.L. No. Ill— 220, 124 Stat. 2372 (Aug. 3, 2010); Kimbrough v. United States, 552 U.S. 85, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007); U.S.S.G. supp. to app. C., amend. 713 (2008); id. amend. 706 (2007). I note only that no one has ever accused the ordinary Sentencing Guidelines for narcotics offenses of being anything other than stringent.
What drove Preacely’s guideline recommendation to a level more than double that severe guideline outcome is his treatment as a “career offender.” In ordinary English, “career offender” is a reasonable description of Jamar Preacely. Although he is still a young man, to the point of his arrest on the present charge the only “career” he ever had seems to have been selling drugs. But in the technical jargon of the Guidelines, Preacely qualifies as a “career offender,” and therefore earns an extraordinary sentence for a relatively mundane offense, based merely on two prior narcotics offenses. To put those offenses in perspective, the most serious of all his prior crimes resulted in a sentence of two to four years, but that record now puts him in line to have an already sharp seven-year prison term escalated to nearly sixteen years.
There is a significant debate about the desirability of this sort of “three-strikes” sentencing, which dramatically escalates a sentence that is not only already severe, but that has in fact already been substantially increased by reason of Preacely’s criminal record. (A level 23 offense would draw a recommended sentence starting at only 46 months for a first offender; the 84-month recommendation that would apply to Preacely absent the “career offender” enhancements already adds more than three years to a less than four-year jail term in recognition of the increased risk of recidivism posed by someone with his criminal history.) Once again, however, I do not address that debate, and accept that the relevant authorities have decided that such a simplistic three-strikes policy should trump the more nuanced treatment of criminal history embodied in the Guidelines’ ordinary criminal history calculations.
But here’s the point: the Sentencing Reform Act, as interpreted and rendered constitutional by the Supreme Court’s decision in United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), requires sentencing courts to treat the Guidelines only as a starting point, and then to craft an appropriate sentence taking full account of “the history and characteristics of the defendant.” 18 U.S.C. § 3553(a)(1). The career offender guideline is based on the likelihood that someone falling into that category presents an extreme danger of committing more crimes. Preacely, however, presented *85substantial evidence that he had reformed, and no longer presented such a danger. If that evidence were fully credited — and I note that the prosecution seems not to have contested its accuracy — Preacely’s “history and characteristics,” including his history of rehabilitation after this offense as well as his criminal history before it, would suggest that the 188-month guideline was not after all the appropriate baseline sentence for him. Yet the district court seems not to have recognized this point, and in weighing what sort of sentence was appropriate for him in light of his cooperation emphasized only that (as a result of the career offender designation) the guidelines treated him as a “category VI” offender, which the district court appears to have treated as a definitive declaration of Preacely’s likelihood of returning to criminality.
Let me be clear about what I am not saying. I am most emphatically not saying that the district court was required to believe that Preacely was a changed man. Once caught, every criminal knows that it is time to feign remorse and rehabilitation, and every judge has seen many defendants who do exactly that. The district judge has had the opportunity to observe Preacely, as we have not. That is why great deference is due to the district court’s judgment as to sentencing. Had the district court concluded that Preacely’s claims of reform were exaggerated, or altogether unworthy of belief, we would have a very different case.
But the district court evidently did not so conclude. Rather, the district court refused to take a starting point other than 188 months, or to consider a more dramatic reward for his cooperation, not because of a specific judgment that Preacely the man was a danger to society, notwithstanding his claims of rehabilitation, but because of Preacely’s general classification, under the career offender guideline, as a “category VI” offender.*
That is what Judge Wallace’s opinion correctly concludes was error. The law of the land, reflecting ordinary common sense, requires a judge to start with the Guidelines, and then consider whether their recommendation — which we can assume reflects a reasonable judgment about the appropriate sentence in a typical case^ — comports with what is reasonable for the particular offender. If the judge concludes that the offender before the court is indeed the incorrigible criminal for whom the career offender guideline was written, the guideline most often should be followed and the severe sentence necessary to protect the public imposed. But if the offender can convince the judge that he is no longer that kind of person, the guideline recommendation is no longer the sentence “sufficient, but not greater than necessary,” to protect the public, id. § 3553(a) (emphasis added), but rather is excessive.
*86We do not hold today that Preacely’s sentence was excessive, or that the district court should have accepted his claims of reform. We do hold that the district court was required to consider those claims, and not merely to rely on the fact that Preacely was classified (somewhat artificially) as a “category VP’ offender, and thus to assume that he is dangerous, so as to require (before being rewarded for his cooperation) a sentence more than twice what his actual crime and criminal record called for. The apparent failure to consider this evidence, and decide whether Preacely is or is not the changed man he claims (with the support of some genuine evidence) to be, warrants reconsideration of his sentence.
I add one note in response to Judge Raggi’s thoughtful dissent. I would be the last to suggest any mandatory sequencing of consideration of grounds for departure, for just the reasons Judge Raggi identifies. I do think, however, that it is extremely useful to give separate consideration to that aspect of the cooperation departure that operates as a pure reward to induce testimony, regardless of the defendant’s character, and the quite separate basis for mitigation that relates to a defendant’s potential reformation (which might be evidenced in part by his cooperation with the authorities). A sentence such as the one in this case that seems to have been calculated as a percentage reduction of a guideline sentence gives the appearance of operating only in the former dimension, and (absent a clear statement of reasons addressing both issues) supports Preacely’s argument that the district court did not consider other reasons for mitigating his sentence.
It is possible, of course, that we misread the district court’s intentions here, by taking overly literally his references to Preacely's classification under the Guidelines. Perhaps the experienced judge was using a kind of shorthand, and meant to say that a further reduction in Preacely's sentence was inappropriate not simply because of his guideline level, but because an individualized assessment of his character, including the evidence of reform proffered on his behalf, led the judge to conclude that he remained a high-risk offender. If that is so, no harm is done by remanding the case and giving the judge an opportunity to explain his conclusion. But great harm would be done if we upheld a sentence that imposed long years in prison on an offender who no longer presents a danger, when a lesser sentence would better serve the purposes of the criminal law, because the judge had given excessive weight to a guideline category that was not appropriately applied to the defendant.