« A modest(?) proposal for filling the bench from the ivory tower | Main | Another prominent white-collar defendant gets a big variance »

December 16, 2008

When can guideline errors be harmless (and when is procedural error is not "significant") after Gall?

An otherwise ordinary Sixth Circuit opinion today in US v. Shor, No. 07-2334 (6th Cir. Dec. 16, 2008) (available here), has this extraordinary little footnote:

Both parties cite pre-Bookercases for the proposition that our review of the district court’s interpretation of the Guidelines is de novo.  This is incorrect: the Guidelines are advisory and thus a district court’s interpretation of the Guidelines is only one factor in our holistic review for procedural reasonableness.  The Supreme Court was quite explicit on this point in Gall. Here, however, this difference is immaterial, for the Supreme Court was also quite explicit in Gall that miscalculating the Guidelines range is a “significant procedural error” that requires reversal.  See Gall, 128 S. Ct. at 597, 598.  It is unclear in the wake of Gall when a procedural error is not “significant.”

I find this footnote extraordinary in part because, to my chagrin, circuit courts have largely embraced and encouraged heavy reliance on pre-Booker circuit jurisprudence even in settings where there is every reason to think Booker (and now Gall and Kimbrough) alters the validity and value of that pre-Bookercircuit jurisprudence.  It is nice to see a circuit court poke at the parties for all-too-ready reliance on pre-Booker cases.

That said, courts and litigants can be excused for heavy reliance on pre-Booker circuit jurisprudence given that Booker (and now Gall and Kimbrough and Rita) have left unresolved many practical questions about the operation and application of reasonableness review.  The issue spotted by the Sixth Circuit in Shor concerning the possibility of non-significant procedural error is part of a larger set of questions I have about the review of post-Booker sentencing errors.  Specifically, I often wonder about the potential value (and potential drawbacks) from the idea that some guideline errors should be reviewed for harmlessness.  (I sense that some circuit panels have brought harmlessness concepts into the mix of reasonableness review, but not with any consistency.)  

In the end, the potential interaction of the array of potentially applicable appellate review terms and ideas  — reasonableness (procedural and substantive), abuse of discretion, harmless/significant/clear/plain error, de novo review — makes my head hurt.

December 16, 2008 at 11:23 AM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference When can guideline errors be harmless (and when is procedural error is not "significant") after Gall?:


I am not convinced as the court is that a miscalculation of the guidelines is always significant error. It seems to me to a large extent to depend on what the particular attitude of the circuit is towards reasonableness in the first instance. If a within guidelines sentence is per se reasonable then a miscalculation is a significant error. But if the circuit does not give such deference to within guideline sentences, then the fact that a guideline sentence was miscalculated by a month or two could certainly be seen as harmless error.

Posted by: Daniel | Dec 16, 2008 11:35:05 AM

Ever since Gall, I have wondered about the 'significance' of the first adjective in the phrase "significant procedural error." I have had AUSA's argue in reply briefs that because the claimed error does not fit within one of the categories listed in Gall, then “no significant procedural error occurred," thereby requiring the Court to affirm. I therefore agree with Prof. Berman that an important issue after Gall is whether a procedural error must be "significant" in order to make the sentence reversible as a "procedurally unreasonable sentence."

The point that the 6th Cir. seems to be making is that when reviewing a sentence, the standard of review is "reasonableness." In turn, the court implies, a sentence may be procedurally reasonable even if a "procedural error" occurred, so long as that error was not a "significant" procedural error.

IMO, the 6th Cir. was a little mistaken in its comments about de novo review being inapplicable, even assuming the Cir's premise (i.e. that a procedural error must be "significant" in order to reverse; or that an error itself will not always make the sentence "procedurally unreasonable" so as to require reversal).

The first step is to determine whether "error" occurred. This would require deciding whether the dist. ct's interpretation of the GL was correct. This should be reviewed de novo both before and after Booker, and the parties were correct to advocate such an approach. If "error" is found under de novo review, then -- under the 6th Cir's premise -- the second step is to determine if the sentence can nonetheless be affirmed because "holistic review" shows that the sentence was still procedurally reasonable. Or, put another way, whether that error constituted "significant" procedural error, requiring reversal.

My point is that the 6th Cir. is incorrect when they assert that de novo review no longer has a place in reviewing sentences. That standard must still be used to determine whether "error" occurred, regardless of whether that "error" requires reversal under "reasonableness" review.

In response to Daniel's comments, I believe he is mistaken due to the distinction between "procedural" and "substantive" reasonableness. A miscalculation of the guideline will always be "significant procedural error" requiring reversal for procedural unreasonableness. Daniel disagrees, citing some Cir's use of the presumption of reasonableness. This presumption, however, only applies to the "substantive" reasonableness of the sentence, and would have no bearing on whether a sentence is "procedurally" reasonable.

(NOTE: My head hurts too now).

Posted by: DEJ | Dec 16, 2008 12:36:07 PM

DEJ. You may wish to go back and read US v. Cavera (2nd Circuit). Your argument is not the argument they are making. While it is certainly true that there is a difference between procedural and substantial reasonableness, these concepts don't exist in isolated boxes. While as a procedural matter the guidelines need to be calculated first by the TJ, any error in that calculation must be evaluated by the appeals court in light on its impact on the substantive reasonableness of the sentence.And given that TJs have wide discretion to impose sentences (because there is no assumption of reasonableness for within guideline sentences in the 2nd) there was no procedural error.

US v. Cavera is not a substantive reasonableness case; it is a procedural reasonableness case. Indeed, the difference between the majority and the dissent in the case is over that exact issue issue you raise. The dissent argues that as a factual matter, there is no basis for saying that gun crimes are more dangerous in an urban area. The majority's opinion doesn't hinge, however, on a factual rejection of the dissents claim . They say that even if the dissent is correct, it doesn't matter. It's not a procedural error for the TJ to consider such factors.

Now, if it's not procedural error for a TJ to consider such policy factors, then it's beyond me how a miscalculation of one or two months in a guideline sentence can be considered procedural error in the 2nd. It would be a rather strange position to take that a mere technical error results in a sentence being overturned but an error in a broad policy calculation results in an affirmation.

Posted by: Daniel | Dec 16, 2008 1:31:14 PM

Daniel, you are wrong on three inter-related levels.

First, Cavera, while certainly not a substantive reasonableness case, is not a case involving an incorrect application of the GL. Accordingly, Cavera has very little to do with the issue of whether "a miscalculation of the guidelines is always significant error" requiring reversal. Your belief as to Cavera's relevance likely stems from your mis-reading of the opinion, see below.

Second, and more troubling, the majority opinion in Cavera does not "hinge" on what you claim it does. If I am reading your post correctly, then you believe the majority in Cavera held that "[i]t's not a procedural error for the TJ to consider [guns being more dangerous in an urban area]." That is not what the majority held; in fact, the majority explicitly withheld holding on that issue. The key point of the majority's decision is that the dist. ct. had "an independently sufficient justification" to grant the variance it did. In other words, even if there was error in factor A, it was harmless procedural error b/c the same sentence would have been imposed based on factor B, which was a valid reason.

Third, contrary to the underlying argument of your first post, the various opinions of Cavera would have been the exact same had the 2d Cir. imposed a presumption of reasonableness on within GL sentences. There is nothing in any of the many opinions to suggest the lack of a presumption had any impact the opinions' outcomes.

Posted by: DEJ | Dec 16, 2008 2:17:54 PM

My head hurt before I even read the post... now it really hurts. But I do agree with DEJ that the courts should be careful about conflating traditional de novo review of legal errors with whether there will be a remedy for any errors found. The appellate courts should always review a trial court's interpretation of rules and guidelines de novo.

For example, say a trial court held that a certain out-of-court statement was inadmissible under FRE 801(d)(2)(D) because it was made by an independent contractor for the defendant, not an employee. Although admission of evidence is generally reviewed for abuse of discretion, the appellate court should review de novo the lower court's interpretation of the FRE---that is, the appellate court should decide the legal question of whether a contractor is, in fact, categorically excluded from being considered a "agent or servant" under FRE 801(d)(2)(D), or whether the trial court should have undertook a more context-specific inquiry. Such de novo review is necessary for the development and consistency of the law. However, even if the appellate court decides that the trial court was wrong about the law and should have admitted the testimony, the court should still affirm if the admission was harmless (i.e., it did "not affect substantial rights" under Fed. R. Crim. P. 52(a)).

Note that, in some cases, if the testimony was clearly of a tertiary nature, the appellate court might go straight to harmlessness. (Essentially saying, even if the lower court got the law wrong, any error was so minor and inconsequential that we don't think it is important to analyze the legal question at length in this case.) Such an approach, though, should still make clear the distinction between the question of error and the question of remedy. That is why it is, in my view, slightly concerning that the Sixth phrased its footnote in a way that implies it is "incorrect" to state that "our review of the district court’s interpretation of the Guidelines is de novo."

Posted by: Observer | Dec 16, 2008 2:35:47 PM

That's an impressive (and important) little point. Judge Boyce Martin (the author of the opinion) has an impressive track record in sentencing cases. Here's to hoping that courts and counsel follow this lead and are more perceptive of these issues going forward.

Posted by: MJG | Dec 16, 2008 3:22:05 PM


As to point one. That's true, it's not an application of the GL case, but that doesn't mean its logic is case-bound. The reason we study cases is to draw forth logical principles that can be applicable to many different cases. So while it true that its fact pattern has little to do with a miscalculation of the GL, its underlying logic certainly is applicable to that scenario.

As to point two, you and I read that differently. "That is not what the majority held; in fact, the majority explicitly withheld holding on that issue." The court withheld judgment on that as *factual* matter, not as a legal matter. It didn't need to decided whether guns were *in fact* more dangerous in an urban area, so it withheld judgment on that issue. But it didn't withhold judgment on whether or not the judge could consider these type of policy issues. In fact, it explicitly endorsed the evaluation of policy consideration by trial judges.

As to point three I just found this comment bizarre, "There is nothing in any of the many opinions to suggest the lack of a presumption had any impact the opinions' outcomes." This totally ignores the fact that majority opinion on page 13 lines 15-16 and lines 20-21 the court explicitly states that it's grounding its decision in just that distinction.

Posted by: Daniel | Dec 16, 2008 5:15:20 PM

Daniel: First, thank you oh so much for explaining "The reason we study cases." Everything is so much more clear now. (sarcasm). My point was the last sentence I wrote: that your belief as to the relevance of Cavera's principles stems from your mis-reading of the opinion. Got it?

Second, we seem to agree that the majority withheld judgment on whether guns actually "in fact" cause more harm in urban areas. But the reason why they withheld judgment was because a separate rationale existed which *in fact* was accurate, and thereby the variance was justified under that alternative rationale. The majority held that "policy issues" could be considered; nobody doubts that. But those policy positions must be examined for their reasonableness/accuracy (for lack of better words). The majority explicitly did not examine the reasonableness/accuracy of rationale A because rationale B was reasonable/accurate and independently supported the variance.

To put this more clearly for you, had the dist. ct. rested its variance solely on "guns create more harm in urban areas," (and never proffered any other justification for the variance), then the 2d Cir. certainly would have had to evaluate the factual reasonableness/accuracy of that policy view.

Finally, for someone who obviously knows so much about studying cases (per your earlier comment), I don't understand how you can claim that in "page 13 lines 15-16 and lines 20-21 the court explicitly states that it's grounding its decision in just that distinction." Those lines are nothing but statements of broad principles concerning an appellate court’s sentencing review. (You commonly find those in en banc opinions). They have no unique application to the case, and there is nothing at all to indicate the opinions are grounded in these lines.

In fact, one might say it is "bizarre" to believe that those lines have any "grounding" to the opinion.

Posted by: DEJ | Dec 16, 2008 6:59:11 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB