Wednesday, May 20, 2015

Rule 410 Waivers

Federal Rule of Criminal Procedure 410 generally prohibits the introduction at trial of a guilty plea, statements made during a guilty-plea proceeding, or statements made during the negotiation of a guilty plea, where the defendant ultimately did not plead guilty or withdrew a guilty plea. But a recent decision published by the Tenth Circuit -- United States v. Jim -- reminds us that this rule is not absolute.
Instead, a defendant can waive his Rule 410 protections, and he can do so in a plea agreement that is never accepted by the district court. Think of a situation where a defendant pleads guilty pursuant to a plea agreement, the court accepts the plea but defers acceptance of the agreement, then allows the defendant to withdraw the plea. And in the plea agreement the defendant waives his Rule 410 protections. Those are the essential facts in Jim. At trial, the government introduced the admissions in the plea agreement in its case-in-chief. Unsurprisingly, the defendant was convicted.
The Supreme Court set the stage for the Tenth's decision way back in 1995. But that case involved the introduction of statements made during a failed cooperation proffer to impeach the defendant's trial testimony. Jim goes much further. It allows the introduction of the defendant's prior admissions in the government's case-in-chief, and the admissions were made in a plea agreement never accepted by the district court because that court allowed the defendant to withdraw his plea.
This latter point played prominently in Jim. The defendant's primary point was that the district court found a "fair and just reason" to allow him to withdraw his plea. Because the plea was unknowing and involuntary, so was the plea agreement. And, if the plea agreement was unknowing and involuntary, it was improper to use admissions in the agreement at trial. 
The missing premise in all of this is whether the district court's reason to withdraw the plea hinged on the unknowing and involuntary nature of it. The district court suggested that this was so: the defendant earlier mentioned that he thought he could still go to trial after signing the plea (thinking the plea merely limited his sentencing exposure), and the magistrate judge who conducted the plea colloquy forgot to advise the defendant that his plea waived his right to a trial (seriously???). But the district court, prior to allowing the defendant to withdraw his plea, also advised the defendant of the Rule 410 waiver, and the defendant's trial counsel acknowledged the repercussions of it prior to the defendant's plea withdrawal. And at trial, in admitting the plea admissions, the district court found that the plea was knowing and voluntary. The Tenth agreed on this point. And that is that.
As a practical matter, watch out for plea agreements with Rule 410 waivers. If you have one, make sure to advise your client that, upon signing the agreement, any meaningful right to a jury trial is gone.
A few more points:
  • The Rule 410 waiver in the plea agreement expressly stated that it took effect when the defendant signed the agreement. So, the fact that the agreement was never accepted by the Court is surprisingly irrelevant. The difficulty we have with this analysis is that the Rule 410 waiver was part of a contract between two parties, but the contract was never enforced. Why does the government benefit from the Rule 410 waiver when it had no obligation to uphold its promises made in the plea agreement? Contracts should be enforced as a whole, not in parts, and one party should not get the benefit of a bargain without performing its reciprocal obligations.
  • The plea agreement advised the defendant that a guilty plea would waive a jury trial, and so the Court dismissed the magistrate's failure to so advise at the change-of-plea colloquy (we wonder whether this analysis is problematic considering the defendant's overall argument that he did not understand the plea agreement in the first place).
  • "A fair and just reason" need not be that the plea was unknowing and involuntary; apparently, something less is sufficient for a district court to allow a defendant to withdraw his plea (we are unsure what).
  • The voluntariness of a Rule 410 waiver is determined under the same general principles of voluntariness that apply in other contexts.
  • And make your record below. The Court was unconvinced with the arguments on appeal primarily because they were not developed properly below.

Tuesday, May 19, 2015

Tenth Circuit uses categorical approach (for the most part) to determine SORNA Tier classification

Last month, the Tenth Circuit decided United States v. White, No. 14-7031 (April 6, 2015), reversing Mr. White’s sentence and lifetime term of offender registration and holding that he was improperly classified as a tier III sex offender under SORNA. The court held that Mr. White's sentence was procedurally unreasonable because he should have been classified as a tier I sex offender under 42 U.S.C. § 16911 (which results in a lower base offense level under U.S.S.G. § 2A3.5). Deciding a matter of first impression for the circuit, the court held that a categorical approach should be used to determine the tier classification for those offenders required to register under SORNA (except when looking to the victim's age). (Editor’s note: the author of this post was the counsel of record for Mr. White, and Mr. White has consented to this blog post.)

In doing so, the court reversed Mr. White’s sentence, which was determined using the base offense level of 16 based on a tier III SORNA classification. The court also explained that Mr. White’s term of offender registration would be affected by its ruling, making his new registration term 15 years (for tier I status) instead of life (for tier III status).

The court explained SORNA’s tier classification as follows:

Under SORNA, a defendant’s tier classification is determined by comparing the defendant’s prior sex offense to statutory criteria. For example, if Mr. White’s prior offense “is comparable to or more severe than [the federal crime of]. . . (i) aggravated sexual abuse or sexual abuse . . . ; or (ii) abusive sexual contact . . . against a minor who has not attained the age of 13 years,” he was appropriately classified as a tier III sex offender. 42 U.S.C. § 16911(4)(A). But if he does not qualify as a tier III sex offender, he will be classified as a tier II sex offender if, as relevant here, his underlying offense is “comparable to or more severe than [the federal crime of] . . . (iv) abusive sexual contact,” irrespective of the victim’s age. Id. at § 16911(3)(A)(iv). And a sex offender who qualifies as neither a tier III nor a tier II sex offender is a tier I sex offender. Id. at § 16911(4)(C).

The court explained that there are two options for determining a defendant’s tier classification under this process: the court could use a categorical approach, looking only at the elements of the prior offense and comparing it to the listed tier II or tier III offenses; or the court could use a circumstance-specific approach, looking to the defendant’s specific conduct.

The Tenth Circuit ultimately held that the categorical approach, for the most part, was what Congress intended:
In light of the text of the statute, its legislative history, and these practical and equitable concerns, we conclude Congress intended courts to apply a categorical approach to sex offender tier classifications designated by reference to a specific federal criminal statute, but to employ a circumstance-specific comparison for the limited purpose of determining the victim’s age.

In reviewing Mr. White’s tier classification, the Tenth Circuit used a modified categorical approach because the statute for Mr. White’s prior sex offense was a divisible statute (to determine the exact subsection of the state statute for which Mr. White was convicted). In holding that Mr. White should have been sentenced as a tier I offender, the court explained that Mr. White’s prior conviction for indecent liberties with a child, under North Carolina law, was not comparable to the federal offenses designated by SORNA as tier II or tier III offenses (looking only at the elements of the prior offense). The court explained that the North Carolina statute for which Mr. White was convicted did not require physical contact with the victim. And the tier II and tier III offenses listed in 42 U.S.C. § 16911 all require physical contact with the victim. So Mr. White's prior conviction was not comparable to the tier II or tier III offenses, and he gets resentenced and his offender registration term reduced to that of a tier I offender. 
--Carl Folsom

Wednesday, May 13, 2015

Trust Us, Do Not "Advertise" Child Pornography

The Tenth published United States v. Franklin a few days ago. If you doubted the concept of a "trial tax" within the federal criminal justice system, this decision might change your mind. The case involves child pornography. The Giga Tribe version (one of many file-sharing programs apparently used to obtain child pornography). The defendant went to trial on five counts, was convicted, and received a 100-year term of imprisonment. Seriously. 100 years. As far as we can tell, the defendant had no history of inappropriate contact with children. Like most child pornographers, he traded images via a file-sharing program. And for that, he will die in prison.
The Tenth Circuit is okay with this, rejecting a substantive reasonableness argument. Because the sentence is within the Guidelines range (the Guidelines actually recommended a sentence of life), the Court engaged "in limited scrutiny of the sentence." In doing so, the Court applied a presumption of reasonableness, despite the fact that the Guideline at issue (USSG 2G2.2) lacks an empirical basis. It seems odd to defer to a Guideline without empirical basis, but there you have it. At least in the Tenth Circuit. We've mentioned before that not all courts apply presumptions to within-range-sentences. And there are Jurists who disagree that this type of sentence is reasonable.
There is a lengthy discussion on whether the sentence creates unwarranted sentencing disparities (something courts are required to consider under 18 U.S.C. 3553(a)(6) when imposing sentence). The Court suggests that a sentence within the Guidelines range cannot create a disparity. But it fails to note that, with respect to this Guideline, more than two-thirds of sentences fall outside the range, (meaning that a within-range-sentence creates more of a disparity than a sentence outside the range). The Court also criticizes the defendant for citing cases on appeal not cited below. But that criticism ignores the fact that the defendant did not know the Court would impose a 100-year sentence until after it did so. It is difficult to compare cases to a 100-year sentence without knowing that the Court intends to impose a 100-year sentence. For better or worse, this decision essentially stands for the proposition that any unwarranted-disparity argument on appeal will be a loser (at least where the defendant fails to make an analogous argument below).
The Court also rejects a substantive reasonableness argument based on the fact that the district court, and not a jury, found the facts necessary to enhance the sentence. We explained before that a recent Tenth Circuit decision hinted at the possibility of such an argument. Apparently, it did so only to shoot it down.

The decision involves another issue: the scope of 18 U.S.C. 2251(d)'s prohibition on an "advertisement" or "notice" seeking or offering child pornography. The defendant thought his posts on Giga Tribe were not advertisements or notices because not made to the general public (only to his "friends" within the program). The Court rejected that interpretation based primarily on the fact that dictionary definitions of the two terms do not limit their reach to public acts. And even assuming a public component, the defendant's posts on Giga Tribe met the definition because "membership in an informal group of like-minded individuals would constitute a subset of the public." It is not clear how far the Court would travel down this path, but it appears the government might have just stumbled upon a dangerous weapon (the provision carries a 15-year mandatory minimum sentence). Is an email from one person "seeking" child pornography from another an "advertisement" or "notice?" If so, the already difficult task of litigating a child pornography case could get even more difficult (and dependent upon the attorney prosecuting the case).
And that is all. Other than to note that the Court made a fairly obvious mistake when it rejected, as "invalid as a matter of law," the application of the canon of noscitur a sociis to interpret "notice" analogously to "advertisement." It did so under a mistaken belief that the "Supreme Court has held that a list of three words is too short for application of the canon of noscitur a sociis." It cites this case for that proposition, but, if you read the case, the Court merely concludes that the application of the canon there was "unpersuasive" and not "particularly illuminating." Follow the citations and you'll find a Supreme Court case applying the cannon to a list of two or three words. Think about it: we know the difference between the meaning of the term "yellow jackets" when it is grouped with "bees" or "rain boots" via these neighboring words. We do not need to add the words "wasps and hornets" or "hats and gloves" to decipher the term's meaning.         

Tuesday, May 12, 2015

(Very Nearly) 6 out of 10 Federal Judges Agree: Chapter Seven Stinks

More from Kirk --

We wrote yesterday about the problems inherent in the supervised release guideline. Dan has covered similar topics again and again. You know who agrees with us? Federal district court Judges.

In a survey released a few months ago, the Sentencing Commission asked federal district Judges about their views of the revocation guideline. Chapter Seven was widely panned.

59% agreed that Chapter Seven should be significantly revised.

82% believed that the client’s criminal history category should be recalculated under Chapter Four, instead of using the criminal history category applied at the original sentencing.

94% want specific sanctions short of incarceration to be incorporated into the guideline.

The idea of graduated sanctions short of a BOP commitment is especially intriguing. The National Institute of Justice has validated that short jail sentences for violations of release conditions are a superior way to ensure compliance with the conditions of release. This model has been successfully replicated by other jurisdictions.

Three takeaway thoughts. First, the majority of federal judges disagree with how we are conducting revocation hearings. Second, if prior convictions have timed out, recalculate the client’s criminal history. And third, let’s try something that doesn’t require a BOP commitment. Graduated sanctions with short jail visits have proven to be much more effective. Fashion something creative.

Monday, May 11, 2015

An Unapologetic Lack of Empiricism: Chapter 7 of the Guidelines

First Assistant Federal Defender Kirk Redmond --

Chapter Seven of the Guidelines governs supervised release violations. But the sentencing ranges prescribed by this chapter are utterly bereft of empirical analysis. A long time ago, the Commission threw darts when constructing the penalties for particular supervised release violations. In 1990, the Commission wrote:
At this time, the Commission has chosen to promulgate policy statements only. These policy statements will provide guidance while allowing for the identification of any substantive or procedural issues that require further review. The Commission views these policy statements as evolutionary and will review relevant data and materials concerning revocation determinations under these policy statements. Revocation guidelines will be issued after federal judges, probation officers, practitioners, and others have the opportunity to evaluate and comment on these policy statements.

U.S.S.G. Chapter 7, Part A.

For the last 25 years, the Commission has been sharpening their pencils (we guess). The promised review of the revocation policy statement has never occurred. Apparently, the review of “relevant data” and the issuance of empirically-based guidelines has been consigned to the dustbin of history. Thousands of clients have been sentenced based on darts thrown a quarter-century ago.
The introduction to Chapter Seven has been amended twice, though neither amendment implicates the substance of the Commission’s approach. The two listed amendments, 362 and 646, merely clarify when intermittent confinement can be a condition of supervised release. The promised review of Chapter Seven has never occurred.

Too harsh? Maybe. But remember that this guideline sits in a scheme the Supreme Court has generally praised as “the product of careful study based on extensive empirical evidence derived from the review of thousands of individual sentencing decisions” And this review is purportedly “ongoing. The statutes and the Guidelines themselves foresee continuous evolution helped by the sentencing courts and courts of appeals in that process.” That continuing empirical review, when it occurs, is what lends the Guidelines legitimacy. Where that empirical review is absent, so is the legitimacy of the guideline.
Consider how this plays on the ground. You have a client who is accused of violating supervised release by using methamphetamine. The criminal history category is V. Chapter Seven sets the advisory policy range for a B grade violation at 18-24 months. But under USSG 2D1.2, which penalizes possession of methamphetamine, the offense level would be 4 (before deducting acceptance of responsibility credits for stipulating) with a policy range of 4-10 months. The supposedly empirically based guideline range is 22% of the Chapter Seven Guideline range. Because, darts.

Or, say your client slugs a guy in a bar. Pursuant to USSG 2A2.3, the offense level would be 7. With a criminal history category of II, for instance, the policy range would be 2-8 months. If two levels were deducted for acceptance of responsibility (based on a stipulation to the alleged conduct), the policy  range would be 0-6 months. But because the violation is a Grade B, the policy range under Chapter Seven is 6-12 months. Empirical validation, indeed.

Chapter Seven stinks. We need to let the courts know this.

Friday, May 8, 2015

Mistakes of Law Post-Heien

"It has been held from the earliest days, in both the Federal and state courts, that a mistake of law, pure and simple, without the addition of any circumstances of fraud or misrepresentation, constitutes no basis for relief at law or in equity, and forms no excuse in favor of the party asserting that he made such mistake."
That is a quote from the Supreme Court's decision in Utermehle v. Norment, 197 U.S. 40, 56 (1905) (we cited this case, instead of another, because the name is pretty cool).

We learned last December, however, that this time-tested rule has an exception for police officers. We blogged about that case (Heien v. North Carolina) here. That's right. If our clients act upon a mistake in the law, too bad, but if a police officer does, well, so what. With one exception: unreasonable mistakes of law, even committed by police officers, would not form an excuse for improper conduct (in this context, an unconstitutional search or seizure).

Now, full disclosure, we doubted the day would come that a court actually found an officer's mistake of law unreasonable.
We were wrong.
It happened. Already.
In United States v. Alvarado-Zarza, the Fifth Circuit reversed the denial of a suppression motion because a traffic stop based on a failure to signal 100 feet in advance of a turn was premised on an unreasonable mistake of law. The problem (for the officer anyway) was that the vehicle did not actually turn; it simply changed lanes. Under Texas law, the signal-100-feet-in-advance requirement is for turns, not lane changes. Hence, a mistake of law. 
In determining whether this mistake was unreasonable, the Fifth Circuit distilled a two-pronged analysis from Heien: (1) whether the statute at issue was ambiguous; and (2) whether the state's appellate courts had interpreted the statute. It held that the statute was not ambiguous (it clearly applied only to turns, as confirmed by other statutes and even the Texas Driver's Handbook). Moreover, a Texas state appellate court recently issued a decision discussing the differences between turns and lane changes (calling the distinction "plain and unambiguous"). So, the officer made an unreasonable mistake of law. Suppression was warranted.

There was a secondary issue because the defendant ultimately turned (he went from a through-lane to a turn-lane without signaling, then signaled prior to turning from the turn lane). The question was whether the defendant actually signaled more than 100 feet prior to turning. The defendant provided testimony from an expert who measured the distance at 300 feet (the incident was caught on a dash cam, which allowed the expert to identify the relevant points (when the defendant signaled and when the defendant actually turned)). The government argued that the officer's mistake of fact (I thought it was less than 100 feet) was reasonable. The Court disagreed, noting that the officer testified that he acted quickly without paying much attention to the distance (and don't forget, the officer did not know the law either).
When Heien dropped, it had the feel of an historically bad decision (for fans of the Fourth Amendment). It still is, but the Fifth Circuit has lessened the sting a bit.   

Thursday, May 7, 2015


That is the number of requests for issuance of delayed notification (“sneak and peek”) search warrants or requests for extensions of delay in notification made in the District of Kansas in Fiscal Year 2013. So says the Director of the Administrative Office of the United States Courts. There were a total of 135 requests for sneak-and-peek warrants, and 420 requests for extension of notification. Each of the 555 requests was granted. Every single one.
Initially enacted as section 213 of the USA Patriot Act, it was intended to combat terrorism. While the use of sneak-and-peek warrants for purposes other than terrorism investigations has been well documented, see here, here, and here, the extent to which such warrants were used in the District of Kansas was not previously widely known. Recently, several Kansas newspapers recently published an Associated Press story describing Judge James’ redacted order, which we will talk about in just a minute.

The required annual report by the Director of the Administrative Office of the United States Courts does not appear to be available on the US Courts website. The Electronic Freedom Foundation has posted a copy of the report on its web site. According to the report, of the 6,480 sneak-and-peek warrants authorized in fiscal year 2013:
          80% were for drug cases
          5% were for fraud cases
          2% each were for
               sex cases
               immigration and

The number issued for terrorism cases? 39. Yes. THIRTY NINE. That is 0.006% of sneak and peek warrants authorized in fiscal year 2013. Of the 4,649 requests for extensions for delay in notification, 91% were for drug cases. Terrorism cases? 12, or .0025%
In the interest of complete disclosure, George Washington Law School Professor Orin Kerr has argued AO report has been misinterpreted, positing:

The AO numbers are not just reporting sneak-and-peek warrants. Instead, I think the numbers are combining several different kinds of warrants, only a small number of which are sneak-and-peeks.

The statutory authority for sneak-and-peek search warrants is found at 18 U.S.C. § 3103a. Subsection 3103(b) allows a delay in notification that a warrant was executed, provided:
(1) the court finds reasonable cause to believe that providing immediate notification of the execution of the warrant may have an adverse result (as defined in section 2705, except if the adverse results consist only of unduly delaying a trial)  [1]  ;
(2) the warrant prohibits the seizure of any tangible property, any wire or electronic communication (as defined in section 2510), or, except as expressly provided in chapter 121, any stored wire or electronic information, except where the court finds reasonable necessity for the seizure; and
(3) the warrant provides for the giving of such notice within a reasonable period not to exceed 30 days after the date of its execution, or on a later date certain if the facts of the case justify a longer period of delay.
Subsection 3103a(c) authorizes an extension in the delay of notification for:

Good cause shown, subject to the condition that extensions should only be granted upon an updated showing of the need for further delay and that each additional delay should be limited to periods of 90 days or less, unless the facts of the case justify a longer period of delay.

As noted above, every single request for extension of delay in notification requested in Kansas during fiscal year 2013 was granted. However, in February, Magistrate Judge Teresa J. James fired a shot across the bow of federal prosecutors, faulting them for untimely and incomplete requests for extensions. In an unusual move, Judge James published a redacted memorandum and order two cases, 14-mj--8219 and 14-mj-8116 (PACER subscription required). In United States v. Cellular Telephone, Judge James strongly criticized the government for missing the statutory deadline to obtain authorization to extend delay of notice(in one instance, by over 4½ months), and for blaming the failures on “calendaring oversight.” Judge James’ independent research also “uncovered 18 occasions in the past two years in which the government filed delinquent requests for such extensions.”
While Judge James found the missed deadline in the cases before her resulted from a clerical error, she criticized the government’s cavalier approach to complying with the statutory requirements for delayed notices:

The numerous delinquent filings identified through very limited investigation indicate a lax attitude on the part of the government with regard to the clear statutory delayed notice requirements. The government’s record in this regard is especially troubling in that in none of the 18 identified delinquencies does the government’s motion represent that it provided the notice required under 18 U.S.C. § 2703 once its authority to delay such notice had expired. In other words, the government repeatedly has disregarded the deadlines for seeking extensions while turning a blind eye to the subject’s right to notice.
Judge James ultimately determined the government had shown excusable neglect in failing to meet its statutory deadline, and granted the out of time motions to extend delay in notice. However, she made two other notable observations. First she commented on the apparent expansion in the use of sneak-and-peek warrants beyond what Congress initially intended, and the necessity of strictly complying with statutory requirements:

Congressional testimony on the Act focused on physical searches of spaces or packages. One commentator states that there is no mention, in any of the discussion of Section 3103a, “that this general authority would also apply to search warrants that did not involve physical intrusions into homes, businesses, and packages. . . [and] [a]s recently as 2011, a congressional report characterized ‘delayed notice search warrants’ as involving covert searches of homes and businesses.” The experience in this Court is that delayed notice warrants typically do not involve searches of places, but instead are directed primarily or perhaps exclusively to cell phone information. In other words, while the law was enacted with the intention that law enforcement would use it to combat terrorism by searching places, it has quickly morphed into a tool for prosecuting drug and other crimes through surveillance. The Fourth Amendment certainly holds a place in the discussion.

The proliferation of Section 3103a warrants and extensions thereof, the lack of Fourth Amendment protection, and the types of offenses in which they are most frequently an investigatory tool all point to the increased need for vigilance on the part of the judiciary. Prosecutors frequently obtain extensions – multiple extensions – of delayed notice merely by reciting in conclusory fashion the same words, that immediate service of a notice of the execution of the warrant would “seriously jeopardize an ongoing investigation.” With such a low threshold, it is “essential that an objective mind of a neutral magistrate weigh the need to invade [a person’s] privacy in order to enforce the law,” and “it is necessary for a § 3103a warrant to strictly comply with the requirements of that section.”
          Second, Judge James ordered:
Now that the government’s awareness of these delinquencies is a matter of record and the Court has provided instruction for future filings, the Court expects the government to be more diligent and puts the government on notice that future missed deadlines in delay orders are much less likely to be viewed as excusable neglect.

Because all other filings in these two cases are sealed, it is unknown whether the government appealed Judge James’ ruling to the District Court.

Senior Judge Monti L. Belot has also previously suppressed a search conducted pursuant to a “sneak and peek” warrant. While noting it “appears that [sneak and peak warrants] are likely constitutional based on Supreme Court precedent,” Judge Belot found the warrant defective because it failed to prohibit the seizure of tangible evidence, and did not include probable cause to believe that providing immediate notification of the execution of the warrant may have an adverse result. See United States v. Tyrone Andrews, D. Kansas case 07-10221-02, Doc. 122. (PACER subscription required).

For more information about the use of sneak-and-peek warrants see:
Sneak and Peek Search Warrants and the USA Patriot Act, The Georgia Defender, (September 2002), Donald E. Wilkes, Jr.
From Patriot Act Sneak and Peek to “Sneak and Steal”: Section 213 of the USA Patriot Act, 19 Regent U.L. Rev. 203, Brett A Shumate.

-- David Freund