By the statute’s literal terms, there can be no doubt that the prosecutor in United States v. Williams (No. 09-1924) failed to comply.  In the § 851 notice he served on Williams, the prosecutor identified only one prior conviction (not the requisite two) and then merely stated, “Further information concerning the defendant’s criminal history can be obtained from the United States Probation Office and specifically the Pretrial Services Report in this matter . . . .”  The Pretrial Services Report, which listed a second drug conviction, was not actually served on the defendant until after trial.  Indeed, it appears that the prosecutor himself had not even received and read the Report before his attempt to incorporate it by reference into the § 851 notice.  This was very sloppy work, and the Seventh Circuit righly chastised both the individual prosecutor and his office (the Northern District of Indiana), which lacked any protocol on how to make § 851 notices.  But sloppiness, even inexcusable sloppiness, is not the same thing as reversible error, and the court (per Judge Posner) affirmed Williams’ life sentence. 

Relying on prior cases, the court characterized the legal standard this way: “[A]s long as the defendant has actual notice of the intended use of a prior conviction to enhance his sentence, the statute has been substantially complied with and that is good enough.”  And Williams could not prevail on this “substantial compliance” test:

[T]he notice says that the government would rely on all “applicable” convictions in the pretrial services report, and it was apparent that the second felony drug conviction was “applicable,” that is, a basis for enhancement.  It was the only other felony drug conviction in the list of 19 dispositions [contained in the report]; . . . a lawyer reading the list would notice that in about fifteen seconds; and if the defendant’s lawyer had had any doubt about which conviction the prosecutor was planning to rely on for an enhancement he could have sought clarification from the prosecutor, and he didn’t — doubtless because it was clear.

Although Williams lost on his § 851 claim, the court made clear that it was not adopting a rule that any attempt at notice would always be sufficient:

We can imagine a notice and a list of convictions that were so confusing that the defendant and his lawyer could not be expected to pick out the one or ones that the government might try to use to enhance the defendant’s sentence. . . .

The government takes a risk by sloppy compliance . . . : the risk that either the court will hold that the government failed to provide the defendant with adequate notice or that the defendant will have a claim that by failing to interpret a confusing notice correctly his lawyer rendered ineffective assistance of counsel.

Interestingly, the Eleventh Circuit seems to be taking a very different view of § 851 compliance.  For instance, United States v. Bowden, No. 08-11935, 2009 WL 32755 (11th Cir. Jan. 7, 2009), apparently employs a strict compliance standard and holds that a failure to meet the standard deprives the sentencing court of jurisdiction to impose the mandatory life sentence.  Williams reports that the Solicitor General recently filed a petition for certiorari in Bowden, so the § 851 issue may be in the news again soon.

But now the court has significantly limited the significance of Head in United States v. Anderson (No. 09-1958). 

For conditions of supervised release, 18 U.S.C. § 3583 (in its pre-2008 form) authorized a sentencing court to select just about any of the permissible conditions of probation (except assignment to a halfway house) and “any other condition it considers to be appropriate.”  In Anderson, the Seventh Circuit (per Judge Wood) held that this latter language, the “catch-all provision,” permits assignment to a halfway house.  In Head, the court had rejected this reading of the catch-all provision, reasoning that the inclusion of the halfway-house condition in the probation statute and the lack of specific authorization for the condition in the supervised release statute decided the question; a general catch-all provision could not overcome the negative implication created by the statute’s failure to authorize the condition expressly. 

Anderson downplayed this aspect of Head:

Head indicated in a footnote that the catch-all provision did not recapture the power to impose the halfway-house condition . . . . This theory, however, had not been pressed by the Government . . . .

In effect, Anderson seems to treat Head’s analysis of the catch-all provision as mere dicta and adopts a quite different interpretation: the “any other condition” language means what it says, “confer[ring] broad discretion on the district courts to fashion appropriate conditions of release that compl[y] with the broad goals of sentencing.”  Thus, it turns out (per Anderson) that Head only held that the halfway-house condition was not expressly authorized by the supervised-release statute; Head did not really decide whether the condition was prohibited.  Anderson now tells us that the condition is not prohibited.

Anderson’s interpretation of Head strikes me as rather strained.  But the Anderson panel circulated its opinion to the full court, and not one judge voted to hear the case en banc.  It seems odd that the court has so thoroughly undermined such a recent decision as Head, which is only nine months old, without a single voice raised in protest.  I suppose this is a rather arcane aspect of sentencing law, and perhaps no one really cares.  Or perhaps something in the new cases convinced the judges that Head had sacrificed too much by way of sound corrections policy in the name of textualist purity.  Perhaps the overwhelming weight of contrary precedent in the other circuits also contributed to the judges’ discomfort with Head.

But the intermediate federal appellate courts have been slow to follow Booker to its logical conclusion — which is why Kimbrough was necessary in the first place.  Even after Kimbrough, several circuits, including the Seventh, have maintained that policy choices contained in § 4B1.1, the career offender guideline, remain binding on district court judges.  This is particularly important, and unfortunate, to the extent that § 4B1.1 contains the infamous 100:1 disparity in the treatment of crack and powder forms of cocaine.  That is a policy choice that district court judges ought to reject, and many doubtlessly would reject, if they were free to do so.

Last year, in United States v. Liddell, 543 F.3d 877 (7th Cir. 2008), a panel of the Seventh Circuit suggested that the court might be willing to reconsider its precedent on § 4B1.1.  But then Friday’s decision in United States v. Welton (No. 08-3799) slammed the door shut. 

Writing for the Welton panel, Judge Bauer expressly disavowed Liddell “to the extent that Liddell is inconsistent with [earlier holdings] that a district court may not rely on the 100:1 crack/powder disparity embedded in  § 4B1.1 as a basis for imposing a non-Guidelines sentence.”  Moreover, because of the avowed overruling of Liddell, Welton was circulated to the entire court, and only three judges voted to rehear the case en banc.  (On behalf of these three dissenters, Judge Williams wrote what seems to me a quite persuasive opinion in Welton arguing the basic “advisory means advisory” point.)

The majority’s position seems to boil down to this: the Sentencing Commission did not come up with § 4B1.1 on its own, but instead followed a directive contained in 28 U.S.C. § 994(h) that the guidelines “specify a sentence to a term of imprisonment at or near the [statutory] maximum” for career offenders.  The court thus sees  § 4B1.1 as embodying a congressional policy choice, which is binding on district judges because it is not merely a Commission policy choice.

There is no question that Congress could write a statute requiring district judges to sentence career offenders “at or near the statutory maximum.”  But Congress has not written such a statute.  By its terms, § 994(h) is a directive to the Sentencing Commission, not district judges.  And, if there is anything we have learned from Justice Scalia and his New Textualism, it is that Congress’s mere “policy choices” are not binding on anyone; it is only through the written law, as contained in statutes enacted through constitutionally mandated procedures, that Congress can make binding policy decisions. 

Instead of issuing a directive on career offenders to district judges, Congress chose to issue its directive to the Sentencing Commission, recognizing that its policy choice regarding career offenders would be embedded within, and necessarily qualified in all sorts of uncertain ways, by a larger structure of sentencing guidelines.  Rather than pass a stand-alone mandatory minimum statute, Congress evidently believed that it was best for the specifics of career-offender sentencing to be worked out in a way that cohered with the rest of the federal sentencing system.  Post-Booker, such coherence means that the career offender guideline should be treated as no less advisory than any other guideline.

Moreover, treating § 994(h) as binding on judges risks reversing Booker by the back door.  Section 994(h) is not the only congressional directive to the Sentencing Commission; the whole of § 994 is chock full of them.  Indeed, very nearly all of the guidelines could be fairly characterized as embodying one congressional policy choice or another.  (There is, in fact, an interesting debate in the scholarly literature over whether the worst features of the federal guidelines are due more to Congress’s or the Commission’s policy choices.)  Once we start saying that congressional policy choices embodied in the guidelines are binding, it is not clear where a principled line is to be drawn to save Booker.

In fact, the whole point of Booker is that Congress does not get to have its way when it comes to sentencing guidelines.  The Sixth Amendment trumps congressional preferences.  And the Sixth Amendment prohibits courts from treating the policies contained in sentencing guidelines as binding in the absence of jury fact-finding.

Judge Williams concluded her dissent with a call for the Supreme Court to address the § 4B1.1 problem in light of the disagreement it has sparked within the lower courts.  I heartily agree.

In the meantime, I take solace in the fact that there remains some interest and ability on the part of the Seventh Circuit to impose other sorts of limits on the mandatory application of the § 4B1.1 100:1 ratio, as evidence by the court’s recent decision in United States v. Knox, which held that the ratio is not binding on defendants convicted merely of conspiracy to commit a crack offense.  (My post on Knox is here.)

When a person is caught in a lie, we normally assume that he is covering something up.  But, if a defendant in a criminal case lies on the witness stand, is it fair to assume that he actually did what he was accused of doing?  Such was the question in United States v. Edwards (No. 08-1124).

Edwards was arrested after making arrangments to sell crack to a government informant.  The intended sale did not actually take place, but that is no barrier to conviction for drug trafficking.  And, once convicted, a drug dealer becomes responsible under the federal sentencing guidelines for the entire quantity of drugs he has ever sold that counts as “relevant conduct.”  (For an earlier post on the pitfalls of relevant conduct, see here.)  In order to establish the amount that Edwards sold, the sentencing judge relied on, among other things, $765 in cash that Edwards was carrying at the time of his arrest.  Edwards tried to explain away the cash with an unsubstantiated and seemingly implausible story about selling his minivan, but the judge was not convinced.  If the minivan story was fabricated, then Edwards must have earned the money from selling crack, right?  The sentencing judge concluded as much, and increased Edwards’ drug quantity accordingly.

On appeal, however, the Seventh Circuit held that the judge moved to this conclusion too quickly.   Judge Posner, writing for the panel, sensibly observed,

The falsity of the defendant’s testimony makes reasonably clear that the $765 was proceeds of an illegal transaction of some sort, but does not show that it was proceeds from selling crack.  For all one knows, the defendant sold other illegal drugs (he had been convicted in the past of possession of marijuana) or other contraband, such as guns, but did not want to acknowledge other illegal behavior, which he might have thought would get him into even worse trouble than he was in.

In light of these possibilities, it appeared that too much crack might have been attributed to Edwards when his guidelines sentence was calculated.  The Seventh Circuit thus vacated Edwards’ sentence and remanded for resentencing.

It is true that Edwards’ false testimony does not logically require an inference that the $765 came from crack sales, but it is still surprising to see the Seventh Circuit not permit the inference.  After all, what was at issue was a district court’s finding of fact, to which appellate courts are supposed to show some deference.  Indeed, the First Circuit upheld a district court’s decision on similar facts in United States v. Jackson, 3 F.3d 506 (1st Cir. 1993).  However, the Seventh Circuit felt that Jackson was inconsistent with other cases that reject “the automatic inference from disbelief in one part of a witness’s testimony to disbelief in the rest.”

Although the Seventh Circuit may not have shown its typical deference to district-court fact-finding in Edwards, the lower court’s piling on of relevant conduct is a reminder of why the drug sentencing guidelines have provoked so much criticism.  Quantity tends to overwhelm other considerations in drug sentencing, and the determination of quantity under the guidelines makes no distinction between what was proven to a jury beyond a reasonable doubt and what was proven to a judge by a preponderance of the evidence.  The diminished burden of proof under the guidelines invites dramatic increases in punishment on the basis of the sort of flimsy inferences used in Edwards.  This is especially troubling when the defendant is being sentenced under the crack guideline, which is widely recognized to be unjustifiably harsh.

In United States v. L.E. Myers Co. (No. 07-2464), the defendant corporation was convicted of criminal OSHA violations in connection with the electrocution death of one its employees.  The Seventh Circuit (per Judge Sykes) reversed and remanded for a new trial in light of erroneous jury instructions.  The errors related to mens rea issues.  Myers was convicted under a statute that bases liability on the knowing creation of a hazardous condition in knowing violation of an OSHA requirement. 

The problem is that a corporation, as a legal construct, cannot really know anything; the only way a corporation knows something is to the extent the law is willing to impute the knowledge of particular employees to the corporation.  Seventh Circuit precedent indicated that “corporations ‘know’ what their employees who are responsible for an aspect of the business know.”  More specifically, the corporation was said to know what an employee knows if the employee has a duty to report that knowledge to someone higher up in the corporation.

In light of this precedent, the Seventh Circuit determined that the knowledge instruction in Myers was too broad.  The instruction permitted the jury to convict based on any knowledge obtained by any employee within the scope of his or her employment, regardless of whether there was a duty to report the knowledge up the corporate ladder.

The trial court also erroenously gave an ”ostrich” instruction.  Such an instruction permits conviction where a defendant has “affirmatively avoided” obtaining information that would otherwise provide a basis for liability.  The Seventh Circuit determined that the instruction should not have been given because there was no evidence that Myers affirmatively acted to avoid learning the truth: “there is evidence of deliberate indifference to the facts, but there is no evidence of deliberate avoidance, and the latter is required for the ostrich instruction.”

In United States v. Ozuna (No. 07-2480), the defendant was convicted of drug trafficking based on the discovery of cocaine in his truck by DEA agents.  Ozuna argued that the search of his truck violated the Fourth Amendment.  At a suppression hearing, the government relied on a written consent to the search that was allegedly signed by Ozuna.  The defendant, however, denied the signature was his, and the district court initially granted his motion to suppress.  Following this decision, the government requested reconsideration based on a new handwriting analysis of the signature.  At a second suppression hearing, handwriting experts testified for both the government and the defense, and the court ultimately reversed its earlier suppression ruling.  On appeal, Ozuna argued that the government should not have been given a second hearing and that the court should have performed a Daubert analysis regarding scientific reliability before admitting the testimony of the government’s handwriting expert.

The Seventh Circuit (per Judge Kanne) affirmed.  The court acknowledged that at least three other circuits impose special requirements for the government to justify a second hearing on suppression.  (Such a standard would have been hard for the government to satisfy in Ozuna, as there was no apparent reason why the government could not have performed a handwriting analysis before the first suppression hearing.)  The Seventh Circuit, however, joined two other circuits that leave the matter more squarely within the discretion of the trial court.

Turning to the Daubert question, the Seventh Circuit held that the court need not conduct a special analysis of the reliability of scientific evidence in the context of a suppression hearing.  Daubert was intended to keep junk science from being presented to juries, but there is no jury at a suppression hearing.  It makes little sense to require a judge to act as a gatekeeper with respect to evidence that will be presented only to that judge.

In United States v. Huffstatler (No. 08-2622), the defendant was convicted of producing child pornography and given a sentence above that called for by the federal sentencing guidelines.  He appealed the sentence, arguing that the child pornography guidelines should not be followed because they were not the product of empirical research by the Sentencing Commission.  Although the child pornography guidelines do indeed reflect politics more than research (a topic covered in the most recent issue of the Federal Sentencing Reporter), and although sentencing judges are authorized to take such considerations into account under Kimbrough v. United States, 128 S. Ct. 558 (2007), Huffstatler’s argument was an odd one.  After all, the judge in his case did not follow the guidelines.  Huffstatler thus seemed to be advancing the facially dubious claim the because the child pornography guidelines are poorly designed, child pornography defendants must always get a sentence below the guidelines level.  Not surprisingly, the Seventh Circuit rejected this argument in a short per curiam opinion.

With uncertainty still surrounding whether sentencing judges may issue Kimbrough variances in child pornography cases, I imagine that many Seventh Circuit defense lawyers were concerned about Huffstatler presenting a challenge to the child pornography guidelines in such an unfavorable procedural and factual context.  In rejecting the proposition that the sentencing judge must vary downward in child pornography cases, the Seventh Circuit might have (intentionally or unintentionally) said things that would discourage sentencing judges from using their Kimbrough discretion in such cases.  It appears, however, that the Seventh Circuit was careful not to send such signals: “Even assuming that district courts may exercise their discretion based solely on policy disagreements with the child-exploitation guidelines (an issue we need not decide here), Huffstatler’s argument is without merit” (emphasis added). 

Indeed, there is even some language in the opinion that is helpful to defendants:

Huffstatler correctly submits that the child-pornography sentencing guidelines, U.S.S.G. §§ 2G2.1-.2, like the drug guidelines at issue in Kimbrough v. United States, 128 S. Ct. 558 (2007), are atypical in that they were not based on the Sentencing Commission’s nationwide empirical study of criminal sentencing.

Finally, in United States v. Kingcade (No. 08-2447), the Seventh Circuit considered whether the defendant properly preserved his Fourth Amendment issue for appeal notwithstanding his guilty plea.  Federal Rule of Criminal Procedure 11 permits defendants to enter conditional guilty pleas that preserve the ability to appeal particular rulings by the trial court.  Such a conditional plea must include a written agreement between the defendant and the government stating which issues are preserved.  Kingcade’s problem was that he and his attorney made several pretrial suppression motions, and the conditional plea agreement referenced a different set of pretrial rulings than the ones Kingcade actually pursued on appeal.  The Seventh Circuit (per Judge Kanne) cut him no slack.  The court relied on principles of contract law to determine the scope of the conditional plea agreement.  Because the court found no ambiguity in the language of the agreement, the court was unwilling to consider extrinsic evidence of any different intent than that which was expressed in the agreement.