It Is Time to Fix the Federal Criminal System

Blackstone famously declared that “it is better that ten guilty persons escape, than that one innocent suffer.”In today’s federal criminal system, that idea seems quaint. An acquittal is viewed as a sign the system does not work.Efficiency has become the highest value.

The federal criminal system in its current form affords the prosecution so many advantages that even an innocent defendant has little chance of acquittal. A defendant whose guilt is merely doubtful has almost no chance. Once in a great while the defense wins, but only because courageous judges and juries resist the systemic pressure to convict and imprison.

This article lists ways the system favors the prosecution, from investigation through appeal, and suggests a menu of reforms. The list is not exhaustive — just a few examples among dozens. Encountered in isolation, each imbalance may seem inconsequential. As a whole, they produce a system so heavily stacked for conviction that few defendants dare risk trial. Those who do (Jeffrey Skilling, Bernard Ebbers, Raj Rajaratnam) often pay for their temerity with decades in prison.3

It is time to fix the federal criminal system. Let’s start at the beginning and look at the process step-by-step, from pre-indictment investigation through appeal.

I. Obtaining Evidence

A. Grand Jury Subpoena Power

Federal prosecutors have the grand jury’s subpoena power at their disposal. Grand jury subpoenas are not subject to relevance objections; prosecutors can fish at will for evidence, subject only to privileges.The defense has no comparable power to compel the production of documents from third parties. Before indictment, it has no compulsory process. Even post-indictment it has only trial subpoenas duces tecum. Trial subpoenas will not be enforced unless the defense demonstrates the documents sought are relevant, admissible, and specifically identified.As a practical matter, this inequality means the prosecution can cast a broad net for inculpatory information, while the defense has far less ability to compel production of exculpatory information.6

B. Immunity

Potential witnesses — particularly those involved in the underlying facts — often assert the Fifth Amendment privilege. The government alone can immunize witnesses under 18 U.S.C. § 6002; the defense has no such power.

Federal courts have only limited power to immunize potential defense witnesses (or to compel the prosecution to confer immunity) and are reluctant to use the discretion they have.The prosecution can thus skew the evidence by immunizing witnesses with inculpatory testimony and refusing to immunize witnesses with exculpatory testimony. To make the most of this advantage, prosecutors sometimes admonish potential defense witnesses (or ask the court to admonish them) about the risks of self-incrimination if they testify. Courts have approved these admonitions, as long as they are not too heavy-handed.Witnesses whom the prosecutor cautions usually get the message and decline to testify.

The defense cannot obtain a missing witness instruction for a potential defense witness whom the prosecution refuses to immunize,and the hearsay rule generally bars introduction of prior statements the witness may have made exculpating the defendant.

C. Rewarding Witnesses

The government can reward witnesses for their testimony, as long as it takes care to specify that the testimony must be “truthful” — meaning, in practice, that it inculpates the defendant. Government largesse can take the form of favorable plea deals, money, immigration benefits, conjugal visits in jail, and so on. The government can even pay informants contingent on the conviction of specified persons.10

If defense counsel were to pay for “truthful” exculpatory testimony, they would open themselves to prosecution under 18 U.S.C. § 201 for bribing witnesses. But when a panel of the Tenth Circuit ruled that prosecutors had to comply with § 201, howls of outrage erupted from coast to coast. Acting on its own motion, the en banc court vacated the panel decision nine days after it was entered11 and ultimately held that § 201 does not apply to federal prosecutors.12 Other circuits took turns lambasting the Tenth Circuit panel for threatening the foundation of the federal criminal system.13 Today, the government remains free to pay witnesses for inculpatory testimony it considers truthful, and defense lawyers remain subject to prosecution for doing the same to procure exculpatory testimony.

D. Wiretaps

The government has the power to obtain wiretaps under Title III and the Foreign Intelligence Surveillance Act (FISA).14 The defense has no similar power. Wiretap evidence is extraordinarily powerful — but the government alone (with a court’s unvarying approval) determines whose phones will be tapped and for how long. Particularly in conjunction with the co-conspirator exception to the hearsay rule and the government’s power to select the portions of the defendant’s statements that will be admitted at trial, the power to wiretap gives the government broad power to shape the evidence.

E. Engaging in Illegal or Unethical Conduct

Prosecutors can have their informants and undercover agents participate in crimes with the targets of investigations. Limited only by entrapment and the outrageous government conduct defense (which courts “recognize” but almost never apply),15 the government can manufacture crimes and then prosecute those who participate. Defense counsel who use similar tactics to expose the criminality of prosecution witnesses will themselves be prosecuted.16
Prosecutors and agents can also engage in conduct that would be unethical if done by defense counsel. For example, rules of professional conduct in virtually every state prohibit lawyers and their agents from communicating with persons they know to be represented about the subject of the representation.17 The McDade Amendment18 subjects federal prosecutors to these “no-contact” rules. But the rules have an exception for contacts “authorized by law” — and courts have uniformly “authorized” pre-indictment undercover contacts with represented targets.19 Defense attorneys have no similar exception that permits them to conduct surreptitious interviews with represented prosecution witnesses.20

F. Discovery

In some states, liberal discovery rules offset the prosecutor’s overwhelming pre-indictment information-gathering advantage. In the federal system, however, discovery is limited. Defense counsel cannot compel prosecution witnesses to give statements or depositions. Expert discovery usually consists of a bare-bones statement of the expert’s qualifications, opinions, and the “bases and reasons for those opinions.”21 The Jencks Act22 and Rule 26.223 require both sides to produce prior statements by witnesses, but the disclosure does not have to be made until after the witness has testified on direct. Even when the parties agree to early production, it is common to receive statements in the middle of trial.

In many jurisdictions, the government’s notes or even reports of interviews are not considered Jencks material, so the defense receives no record at all of the witness’s prior statements to law enforcement. When the prosecution does produce FBI 302s or other reports, several interviews are sometimes combined into one report to obscure how the witness’s version changed over time across interviews.24

Under Brady and its progeny, prosecutors must produce exculpatory information, including information that can be used to impeach prosecution witnesses. But the Brady rule is largely toothless. Most courts require production only if the information is “material,” meaning that there is a reasonable probability it would change the outcome of the trial.25 Courts and prosecutors rarely find that information meets this standard.

The timing of Brady disclosure reduces the value of information the prosecution chooses to produce. Some courts hold that Brady material contained in a witness statement does not have to be disclosed until the Jencks Act deadline.26 And even non-Jencks Brady need only be disclosed in time for effective use at trial — which means, in the view of many courts, that midtrial disclosure suffices.27

G. Classification

Cases involving classified information present additional obstacles for the defense. Once information is classified, only persons with security clearances can view it, and then only in a specially protected room in the courthouse. If the defendant cannot obtain a security clearance — and he generally cannot, unless he previously held one — he cannot review the classified material. That presents significant practical problems. For example, the government routinely refuses to declassify recordings of the defendant’s conversations intercepted under FISA. This means that a defendant without a security clearance cannot review his own recorded statements, and his cleared counsel cannot discuss them with him.28

The defense can use classified material at trial only if it gives notice under the Classified Information Procedures Act (CIPA),29 the judge determines pretrial that the information is relevant and otherwise admissible,30 and the government has an opportunity to propose unclassified substitutions for the classified information. The court must accept the substitutions, and force the defense to use them at trial, if they give the defendant substantially the same ability to make his defense.31 This process places a largely one-sided disclosure burden on the defense and may lead to an artificial, stilted defense presentation. The more information the Executive classifies, the more the defense is forced into the cumbersome and one-sided CIPA process.

II. Evidentiary and Procedural Rules

The rules outlined in the preceding section provide the prosecution an overwhelming advantage in shaping the pool of information from which the parties draw their trial evidence. The defendant’s lot becomes worse when trial begins. A new set of prosecution-friendly evidentiary and procedural rules kicks in to compound the prosecution’s pretrial advantages.

A. The Co-Conspirator Exception

The defendant must overcome the hearsay rule to get his defense before the jury. The prosecutor, on the other hand, has a ready means of avoiding both the hearsay rule and the Confrontation Clause: the so-called co-conspirator exception.32 It takes scant evidence to convince a judge that a conspiracy existed and the defendant and the declarant were members. Judges routinely admit the co-conspirator evidence conditionally and wait until the end of the prosecution’s case to make these determinations. By that point, given the choices of mistrial or admissibility, admissibility generally wins. And courts find almost any relevant out-of-court statement by the alleged co-conspirator to be in furtherance of the conspiracy.

Unlike true hearsay exceptions, Federal Rule of Evidence 801(d)(2)(E) does not rest on the premise that co-conspirator statements are inherently reliable.33 Some courts justify the rule on the theory that every conspirator is the agent of every other conspirator and thus a statement by one is a statement by all.34 But if the basis for the co-conspirator exception were agency, it would be superfluous; Rule 801(d)(2)(D) expressly covers out-of-court statements of agents. And in most conspiracies, the conspirators have nothing that would pass for an agency relationship in any other context.

The advisory committee note to Rule 801(d)(2)(E) acknowledges what cases almost never do: “[T]he agency theory of conspiracy is at best a fiction and ought not to serve as a basis for admissibility beyond that already established.”35 In fact, the co-conspirator exception exists not because conspirators are agents of each other, but as a practical recognition that conspiracies are hard to prosecute because they operate in secret to conceal criminal conduct.36 In other words, the rule exists to make it easier for prosecutors to convict people they believe are conspiring.

Sometimes the government wants to admit out-of-court statements by a person associated with the defendant, but it cannot show that the defendant and the declarant were in a conspiracy, i.e., an agreement to do something unlawful. To solve this problem, prosecutors have argued with remarkable success that Rule 801(d)(2)(E) applies not merely to statements in furtherance of a conspiracy, but also to statements in furtherance of a “lawful joint venture.”37 The lawful joint venture theory has no footing in the language of Rule 801(d)(2)(E) or in its policy of facilitating prosecution of illegal agreements. Courts have embraced the “lawful joint venture” theory using the discredited agency rationale, overlooking that the alleged “joint venturers” are generally not agents of each other and that, if they were, another exception — Rule 801(d)(2)(D) — would cover their out-of-court statements.38

The co-conspirator exception and its “lawful joint venture” variant exemplify the “mission creep” that pervades federal criminal law. An ostensibly narrow rule designed to make it easier for prosecutors to convict conspirators has expanded to become a central feature in most federal criminal trials. The defense hunts, often in vain, for the hearsay exception that will allow a few exculpatory out-of-court scraps into evidence. The prosecutor incants the magic words — “co-conspirator exception” — and suddenly the courtroom fills with the recorded voices of persons captured on tape saying incriminating things that can never be tested through cross-examination.

B. Defendant’s Statements

The government can introduce the defendant’s prior statements as admissions of a party opponent.39 But if a defendant seeks to introduce his own prior statements, they will be excluded as hearsay unless they fall within the “state of mind” exception40 or are not offered for their truth.41 This imbalance permits the government to select inculpatory snippets from a defendant’s wiretapped conversations, FBI interview, or civil deposition testimony while barring the defendant from introducing the exculpatory portions except to the limited extent the rule of completeness42 requires. The defendant’s apparently inculpatory statements are thus wrenched out of context, and the jury is left with the distorted impression that the defendant spoke and thought about nothing but criminal conduct.

C. Rule 404(b)

When evidence of the defendant’s criminal intent is weak, prosecutors compensate by introducing evidence of his other, uncharged bad acts. Rule 404(b) is supposed to prevent the admission of evidence simply to show the defendant’s criminal propensity, and Rule 403 bars evidence where the unfair prejudice substantially outweighs the probative value. But courts routinely find even the most shocking evidence of other bad acts relevant to the defendant’s intent or knowledge (and thus not barred by Rule 404(b)), and the prejudice versus probative value balance often tilts the prosecution’s way.43 The limiting instruction that accompanies the admission of Rule 404(b) evidence, like most limiting instructions, is soon forgotten, and the damaging evidence becomes part of the mix of information by which the jurors judge the defendant’s guilt.

D. Expert Testimony

In civil cases, experts receive careful scrutiny. Courts zealously perform their “gatekeeper” role under Daubert and Kumho Tire. The parties subject experts to grueling depositions, and courts then hold exhaustive Daubert hearings. It is a different story for prosecution experts in criminal cases. Prosecutors routinely call FBI agents, police officers, and other members of the prosecution team as “experts” on the modus operandi of whatever kind of criminal the defendant is alleged to be. DEA agent experts opine on the usual characteristics of drug dealers, which always fit the alleged drug dealer defendant to a T.44 Law enforcement terrorism experts opine on the usual characteristics of terrorists, which always fit the alleged terrorist defendant.45 The most skillful of these witnesses marshal the evidence for the prosecution while maintaining the objective tone of a purportedly independent expert. Many courts allow these “experts” with only perfunctory inquiry into the reliability of their methods and the usefulness of their testimony to the jury.

E. Jurors Can’t Be Told The Punishment the Defendant Faces

Federal courts are haunted by the fear that jurors might acquit out of sympathy for the defendant or a belief that the punishment is too harsh. This fear produces bizarre results. Jurors have always had the power to nullify, but no one can tell them they have that power. Nor can jurors be given information that might lead them to nullify on their own. It is forbidden, for example, to tell jurors how long the defendant will spend in prison if they find him guilty.46 The fear is that jurors might be distracted from the task at hand by the human consequences of their decision. As the Supreme Court put it, telling jurors the potential sentence would “invite[] them to ponder matters that are not within their province, distract[] them from their fact finding responsibilities, and create[] a strong possibility of confusion.”47 To eliminate this danger, judges instruct jurors not to be concerned about the punishment if they find the defendant guilty.48 Some courts even forbid detailed cross-examination of cooperating witnesses about the incarceration they have avoided, for fear the jurors will deduce the defendant faces a similar punishment.49

F. Preference for Joint Trials

Anyone who has tried a multidefendant conspiracy case knows the disadvantages of a joint trial. The defendants and their counsel huddle around the defense table, directly across the courtroom from the jurors, looking for all the world like … conspirators. Evidence comes in against one defendant subject to limiting instructions that jurors quickly forget, and thus spills onto the other defendants. Defenses conflict, and efforts to avoid those conflicts may force defense counsel to forego promising arguments that might damage another defendant and produce a damaging response in turn. Cross-examination must sometimes be muted to avoid harming a co-defendant.

The prejudice to the defense from a joint trial of conspirators is powerful and undeniable. Yet so strong is the cry for efficiency50 that severances are as rare as unicorns. Mutually antagonistic defenses do not suffice for severance; efficiency outweighs a defendant’s right to avoid having his co-defendant join the prosecutor in trying to convict him.51 As a practical matter, severance of defendants in a conspiracy case is possible only under Bruton (although Bruton has lost much of its force as the courts have watered it down), or if one defendant has a right to introduce evidence that another has a right to exclude (and even then courts often find a limiting instruction sufficient to avoid severance), or if one defendant will clearly exculpate a co-defendant in a second, separate trial but will take the Fifth in a joint trial. These are rare circumstances; defendants indicted together can expect to be tried together, with all the attendant disadvantages.

G. Multiple Counts

In most price fixing cases, the Antitrust Division (to its credit) brings a single count indictment, which affords an up-or-down verdict of guilty or not guilty. Most other federal indictments split a single set of facts into multiple counts. Charging many counts creates the risk that jurors will compromise in close cases, acquitting on some counts but convicting on others that are indistinguishable. Inconsistent verdicts do not provide a ground for reversal,52 and the defendant who wins on most but not all counts generally suffers the same sentence as if he had lost them all — a sentence, of course, about which the jurors must remain ignorant.

III. Substantive Law

The reader understands by now that the deck is stacked for the prosecution. The prosecutor has the grand jury, wiretaps, the power to grant immunity, and other tools. The rules of evidence and other procedural rules tilt the playing field even more. What about the substantive law? Doesn’t the prosecutor face a big hurdle in proving the elements of federal offenses, especially criminal intent? Well, no. Some examples illustrate the point.

A. Conspiracy

Almost every multidefendant prosecution involves a conspiracy charge. To win a conspiracy conviction, the prosecution must prove an unlawful agreement; the defendant’s participation in the agreement; and, for some conspiracy offenses, an overt act by one of the conspirators in furtherance of the conspiracy.

A jury instruction could stop here and the jury would understand the elements of the offense. But standard conspiracy instructions detail what the prosecution does not have to prove. The Ninth Circuit pattern instruction is typical. It assures the jury that “it does not matter whether the crime agreed upon was committed”; that “it is not necessary that the conspirators made a formal agreement or that they agreed on every detail of the conspiracy”; that “[i]t is not necessary that all members of the conspiracy join it at the same time, and one may become a member of a conspiracy without full knowledge of all the details of the unlawful scheme or the names, identities, or locations of all the other members”; and that “[i]t is not a defense that a person’s participation in a conspiracy was minor or for a short period of time.”53 Some courts add that “once it has been shown that a conspiracy exists, the evidence need only establish, beyond a reasonable doubt, a slight connection between the defendant and the conspiracy to support the conviction.”54

More than 60 years ago, Justice Jackson described conspiracy as an “elastic, sprawling and pervasive offense” and warned that “loose practice as to this offense constitutes a serious threat to fairness in our administration of justice.”55 It is past time to heed his warning.

B. Pinkerton

Courts routinely give juries two routes to convict the defendant: as principal, and as aider and abetter. Courts too often add a third route: Pinkerton. The Pinkerton doctrine permits the jury to convict one conspirator for foreseeable crimes committed by other conspirators in the course and in furtherance of the conspiracy.56 Some courts have expanded Pinkerton to include “co-schemer” liability in mail and wire fraud cases for substantive offenses other defendants commit in furtherance of the common scheme.57 No statute supports the Pinkerton theory; it is a common law creation in a system that disavows common law crimes.

Pinkerton is especially pernicious given the broad sweep of the conspiracy statutes. Jurors first receive the conspiracy instruction, which emphasizes what the prosecution does not have to prove. Then they get the Pinkerton instruction, which tells them they can convict the defendant for substantive offenses in which he played no part, merely because he was in a conspiracy with the persons who did commit them.

C. Conscious Avoidance

Despite the prosecution’s advantages, it can be hard to prove the defendant’s culpable knowledge. To help the prosecution overcome that difficulty, courts began interpreting the “knowingly” element of federal criminal statutes to include conscious avoidance of knowledge (also known as willful blindness or deliberate ignorance).58 This was pure judicial legislation — Congress knows how to define knowledge as the avoidance of knowledge59 — but courts acted on their own to give the prosecution a leg up when Congress did not.

At first, courts worried about expanding the concept of knowledge to include the avoidance of knowledge. They warned that conscious avoidance should be “rarely” invoked because it might cause the jury to convict based on negligence or recklessness.60 That early caution yielded, however, to expediency.61 Although some courts still insist that conscious avoidance is reserved for extraordinary cases,62 the instruction enjoys wide acceptance in cases where knowledge is at issue.63 The Supreme Court recently read the doctrine narrowly but gave it an overwhelming stamp of approval, with only Justice Kennedy in dissent.64

Courts permit the prosecution to argue conscious avoidance in the alternative.65 A prosecutor can argue, in other words, that the defendant knew about (for example) his company’s improper accounting — but if he did not know, then he deliberately avoided knowledge. The prosecutor can argue, remarkably enough, that both of these mutually exclusive theories have been proven beyond a reasonable doubt. Furthermore, in a particularly prosecution-friendly twist, some courts of appeals hold that if the district court gives a conscious avoidance instruction without an evidentiary basis, the error is automatically harmless because the jury is presumed to have convicted on the correct ground (actual knowledge) and not the unsupported ground (conscious avoidance of knowledge).66 District courts in these circuits can give the instruction in every case where knowledge is disputed with no fear of reversal.

Conscious avoidance is thus another example of mission creep. A doctrine that took root as a narrow tool for helping prosecutors convict drug couriers (who often denied knowledge of the drugs they were carrying) has spread like kudzu to cover the entire field of federal criminal law. By judicial fiat, the conscious avoidance of knowledge now equals knowledge.

D. Reasonable Doubt

What about reasonable doubt? Isn’t the burden of proof the great equalizer?

The question itself is misplaced. The prosecution’s burden of proof is not some regrettable disability that requires compensating prosecution advantages elsewhere in the system. It is, instead, a bias built into the system to make convictions possible only when the evidence is close to conclusive. It is a manifestation of Blackstone’s admonition that we would rather free ten guilty men than convict one innocent man. So we should reject the notion that the prosecution should get any advantage, no matter how small, to offset its burden.

Beyond that, reasonable doubt is a less onerous standard than it seems. Most potential jurors believe that the government would not knowingly charge an innocent man, so the case would not have come to trial if it did not have some merit. The presumption of innocence challenges these basic beliefs, and a strong reasonable doubt instruction is necessary to overcome them. But the usual reasonable doubt instructions diminish the prosecution’s burden by emphasizing what it does not have to show. For example, the Fifth Circuit pattern instruction admonishes the jury that “[w]hile the government’s burden of proof is a strict or heavy burden, it is not necessary that the defendant’s guilt be proved beyond all possible doubt. It is only required that the government’s proof exclude any ‘reasonable doubt’ concerning the defendant’s guilt.”67 The Eleventh Circuit instruction begins: “The government’s burden of proof is heavy, but it doesn’t have to prove a defendant’s guilt beyond all possible doubt. The government’s proof only has to exclude any ‘reasonable doubt’ concerning the defendant’s guilt.”68 These instructions are like a wink and a nod to the jury. Sure, they say, the government’s burden of proof is heavy, but it is not all that heavy.

IV. Sentencing

Assume these advantages have their usual effect and the defendant is convicted. The sentencing process brings more bad news.

The Sentencing Guidelines are advisory after Booker, but they still exert a powerful “gravitational pull” on federal sentences.69 For white collar cases, each new scandal brings a wave of specific offense characteristics and adjustments that function as a ratchet: guidelines sentences go up, but they rarely go down. The fraud guideline — U.S.S.G. § 2B1.1 — is so heavily encrusted with often-overlapping enhancements that any significant white collar offense produces a double-digit (years, not months) guidelines sentence. For a major case, such as an Enron or a Worldcom or an Adelphia, the guidelines will recommend an effective life sentence for a middle-aged defendant. The Presentence Reports that recommend those sentences often rely on hearsay upon hearsay, which the defendant can never confront, and which bears no independent reliability beyond the fact that the defendant was convicted.

In the post-Booker world, district courts do not have to impose a guidelines sentence. But prosecutors invariably advocate a sentence within (or sometimes above) the guidelines range, and even district judges who loathe the guidelines must worry that the government will appeal a sentence it considers too lenient.70 The safe course for a district court is to sentence within the guideline range. For a white collar defendant who has been convicted at trial, that means catastrophe — years and years in prison during which children become estranged, spouses divorce and remarry, professional skills dull, and careers vanish.

It is not surprising that so many defendants elect to plead and cooperate. For many that route, however distasteful, offers the only hope of preserving a semblance of the family and professional life they know.

V. Appeal

The federal courts of appeals stand as the last line of protection for most defendants.71 But two appellate doctrines — standard of review and harmless error — weaken that protection considerably.

The courts of appeals review de novo properly preserved, purely legal issues such as statutory interpretation. But most issues are reviewed under deferential standards that excuse much error by the district court. Evidentiary rulings are generally reviewed for abuse of discretion. The district court’s factual findings (in a suppression hearing, for example) are reversed only if clearly erroneous. The evidence is reviewed for sufficiency in the light most favorable to the prosecution, with all inferences and credibility determinations drawn in the prosecution’s favor. Under this standard, the testimony of even the most blatantly perjurious cooperator must be taken as true on appeal.

The harmless error doctrine works in tandem with the standard of review to insulate district court rulings from reversal. With few exceptions, even the most grievous constitutional errors will not result in reversal if appellate judges conclude that the defendant would have been convicted anyway. The recent Skilling and Black cases illustrate this point. Both men convinced the Supreme Court they had been convicted in part on an honest services fraud theory that the Court held was not a crime.72 Back in the courts of appeals, though, they discovered that the theory the prosecution fought to present at trial and defended tooth and nail on appeal was harmless beyond a reasonable doubt — in other words, it made no difference.73 Even without the honest services theory, the appellate judges ruled, Skilling and Black would have been convicted. No matter that a jury never had a chance to make that decision.

The standard of review and harmless error rules form a coda to a process that is skewed from start to finish. Appellate courts often remark when invoking these doctrines that “the Constitution entitles the criminal defendant to a fair trial, not a perfect one.”74 Fairness may suffice, but the federal criminal system rarely produces it.

VI. How to Fix the System

The federal criminal system is broken. In the ways outlined above, and a dozen others that any experienced defense lawyer can recite, the system has become skewed in favor of the prosecution. How can it be fixed? Here are some ideas.

  1. Give judges the statutory authority to immunize defense witnesses.
  2. Revamp the discovery rules along the lines of many states, to include the right of both parties to depose or otherwise obtain pretrial statements from the other side’s witnesses.
  3. Codify the Brady rule without a requirement of materiality and with a pretrial disclosure deadline.
  4. When a case involves co-conspirator statements, courts should (a) require the prosecution to lay the foundation for the exception in a pretrial evidentiary hearing, (b) read “in furtherance” strictly to include only statements that are clearly intended to advance the conspiracy, and (c) jettison the “lawful joint venture” variant.
  5. Permit depositions of experts for both sides and subject experts to the same Daubert scrutiny they receive in civil cases.
  6. Modify the rule of completeness for defendants’ statements to provide that if the prosecution places in evidence an out-of-court statement by the defendant, the defense may introduce any other out-of-court statement by the defendant of the same general kind and from approximately the same time that is relevant to the statement the prosecution introduced.
  7. Eliminate the judge-made conscious avoidance and Pinkerton doctrines. Congress can enact them as statutes if it wishes.
  8. Rewrite pattern instructions to eliminate the litany of facts the government does not have to prove. Similarly, revise pattern reasonable doubt instructions to remove what reasonable doubt does not mean.
  9. Get rid of the Sentencing Guidelines (and, of course, mandatory minimum sentences). The guidelines have become so encrusted with sentencing enhancements that, particularly in white collar cases, they bear no resemblance to justice.


These few steps would go far toward fixing a broken system.


1. 4 William Blackstone, Commentaries on the Laws of England 352 (1769); see, e.g., United States v. Doyle, 130 F.3d 523, 538 (2d Cir. 1997) (“[U]nder the American system of justice, it is preferable to let ten guilty men go free than to convict one innocent man.”) (quotation omitted).
2. For example, when a New York jury acquitted Mohamed Ghailani on 284 out of 285 counts, many politicians and talking heads decried the verdict as evidence that federal courts were the wrong forum for terrorism cases, rather than praising it as a triumph of American justice. Of course, Ghailani received a life sentence on the single count of conviction — the same sentence, in practical terms, that he would have received had he been convicted on all counts — which underscores the near-impossibility of achieving a favorable result for a defendant in federal court.
3. See generally Ellen S. Podgor, White Collar Innocence: Irrelevant in the High Stakes Risk Game, 85 CHI.-KENT L. REV. 77 (2010).
4. E.g., United States v. R. Enterprises, 498 U.S. 292 (1991). The grand jury process as a whole is in desperate need of reform. See, e.g., National Association of Criminal Defense Lawyers, Federal Grand Jury Reform Report & ‘Bill of Rights’ (2007) (describing deficiencies in process and proposing legislation to provide basic protections for grand jury witnesses and targets).
5. E.g., United States v. Nixon, 418 U.S. 683, 700 (1974).
6. Of course, prosecutors also have search warrants available to obtain documents and other items. Defense counsel have no comparable tool.
7. See, e.g., United States v. Straub, 538 F.3d 1147 (9th Cir. 2008); United States v. Ebbers, 458 F.3d 110, 117-22 (2d Cir. 2006); United States v. Serrano, 406 F.3d 1208, 1216-18 (10th Cir. 2005).
8. See, e.g., United States v. Johnson, 437 F.3d 665, 677-79 (7th Cir. 2006); United States v. Jackson, 935 F.2d 832, 846-47 (7th Cir. 1991).
9. See, e.g., United States v. Rios, 636 F.3d 168, 171-72 (5th Cir. 2011).
10. See, e.g., United States v. Levenite, 277 F.3d 454, 458-64 (4th Cir. 2002); United States v. Cervantes-Pacheco, 826 F.2d 310, 312-16 (5th Cir. 1987) (en banc).
11. United States v. Singleton, 144 F.3d 1343 (10th Cir. 1998) (en banc).
12. United States v. Singleton, 165 F.3d 1297 (10th Cir. 1999) (en banc).
13. See, e.g., United States v. Hernandez-Albino, 177 F.3d 33, 42 (1st Cir. 1999) (noting that panel decision in Singleton was “roundly questioned,” “swiftly vacated and reversed,” and faced “overwhelming condemnation” ); see also, e.g., United States v. Anty, 203 F.3d 305, 307-12 (4th Cir. 2000) (§ 201(c) did not prohibit paying $10,000 to an informant for his testimony, “so long as the payment is not for or because of any corruption of the truth of the testimony”).
14. 18 U.S.C. § 2516 (Title III); 50 U.S.C. § 1804 (FISA).
15. See, e.g., United States v. Sanchez, 138 F.3d 1410, 1413 (11th Cir. 1998) (Eleventh Circuit and Supreme Court “recognize” outrageous government conduct defense but have never found it established).
16. See, e.g., Schalk v. State, 943 N.E.2d 427 (Ind. App. 2011) (defense counsel who arranged controlled buy with prosecution witness to prove witness was still involved in drugs prosecuted and convicted for marijuana possession; court rejects argument that defense lawyer stands on the same footing as law enforcement in making controlled buys).
17. See, e.g., ABA Model Rules of Professional Conduct, Rule 4.2; Cal. Rules of Prof. Conduct, Rule 2-100.
18. 28 U.S.C. § 530B.
19. See, e.g., United States v. Carona, 630 F.3d 917, 920-24 (9th Cir. 2011).
20. Cf. Midwest Motor Sports v. Arctic Cat Sales, Inc., 347 F.3d 693, 697-98 (8th Cir. 2003) (imposing evidentiary sanction where attorney in civil case sent private investigator to question surreptitiously adversary’s employees); United States v. Sierra Pacific Industries, 759 F. Supp. 2d 1206 (E.D. Cal. 2010) (defense counsel in civil case contacted Forest Service employees); San Diego County Bar Association Legal Ethics Opinion 2011-2 (May 24, 2011) (attorney who sends “friend” request to person he knows is represented violates no-contact rule).
21. FED. R. CRIM. P. 16(a)(1)(G).
22. 18 U.S.C. § 3500.
23. FED. R. CRIM. P. 26.2.
24. See, e.g., United States v. Skilling, 554 F.3d 529, 577-91 (5th Cir. 2009), aff’d in part, rev’d in part, 130 S. Ct. 2896 (2010).
25. See, e.g., United States v. Coppa, 267 F.3d 132, 140 (2d Cir. 2001). But cf., e.g., United States v. Safavian, 233 F.R.D. 12, 16-17 (D.D.C. 2005) (materiality requirement does not apply pretrial; prosecution must disclose all favorable information); United States v. Carter, 313 F. Supp. 2d 921, 924-25 (E.D. Wis. 2004) (same); United States v. Sudikoff, 36 F. Supp. 2d 1196, 1198-99 (C.D. Cal. 1999) (same).
26. See, e.g., United States v. Bencs, 28 F.3d 555, 561 (6th Cir. 1994).
27. See, e.g., United States v. Rittweger, 524 F.3d 171, 180-82 (2d Cir. 2008), cert. denied, 129 S. Ct. 1391 (2009).
28. See, e.g., Joshua L. Dratel, Sword or Shield? The Government’s Selective Use of Its Declassification Authority for Tactical Advantage in Criminal Prosecutions, 5 CARDOZO PUB. LAW, POL’Y & ETHICS J. 171, 175-79 (2006).
29. 18 U.S.C. App. 3 § 5(a). See generally John D. Cline & K.C. Maxwell, Criminal Prosecutions and Classified Information, 29 LOS ANGELES LAWYER 35 (2006) (describing CIPA process).
30. 18 U.S.C. App. 3 § 6(a).
31. Id. § 6(c).
32. FED. R. EVID. 801(d)(2)(E).
33. See Bourjaily v. United States, 483 U.S. 171, 189-92 (1987) (Blackmun, J., dissenting) (discussing theoretical basis for rule).
34. See, e.g., United States v. Moon, 512 F.3d 359, 363 (7th Cir. 2008); United States v. Weisz, 718 F.2d 413, 433 (D.C. Cir. 1983).
35. FED. R. EVID. 801, Advisory Committee Note.
36. See id.; United States v. Perez, 989 F.2d 1574, 1577-78 (10th Cir. 1993).
37. See, e.g., United States v. Gewin, 471 F.3d 197, 201 (D.C. Cir. 2006); United States v. Russo, 302 F.3d 37, 45 (2d Cir. 2002), United States v. Layton, 855 F.2d 1388, 1397-1401 (9th Cir. 1988); Government of the Virgin Islands v. Brathwaite, 782 F.2d 399, 403-04 (3d Cir. 1986).
38. See generally Ben Trachtenberg, Coconspirators, “Coventurers,” and the Exception Swallowing the Hearsay Rule, 61 HASTINGS L.J. 581 (2010) (criticizing the “lawful joint venture” theory).
39. FED. R. EVID. 801(d)(2)(A).
40. FED. R. EVID. 803(3).
41. FED. R. EVID. 801(c).
42. FED. R. EVID. 106.
43. E.g., United States v. Hardy, 643 F.3d 143 (6th Cir. 2011).
44. E.g., United States v. Sanchez-Hernandez, 507 F.3d 826, 831-33 (5th Cir. 2007).
45. E.g., United States v. Damrah, 412 F.3d 618, 625 (6th Cir. 2005); United States v. Hammoud, 381 F.3d 316, 335 (4th Cir. 2004) (en banc).
46. E.g., United States v. Brewer, 624 F.3d 900, 907 (8th Cir. 2010), cert. denied, 131 S. Ct. 1805 (2011).
47. Shannon v. United States, 512 U.S. 573, 579 (1994).
48. E.g., Fifth Circuit Criminal Jury Instructions, Instruction 1.20 (2001).
49. E.g., United States v. Cook, 776 F. Supp. 755, 756-57 (S.D.N.Y. 1991).
50. See, e.g., Zafiro v. United States, 506 U.S. 534, 537 (1993); United States v. Lighty, 616 F.3d 321, 350-51 (4th Cir.), cert. denied, 131 S. Ct. 846 (2010).
51. E.g., Zafiro v. United States, 506 U.S. 534 (1993).
52. See, e.g., United States v. Powell, 469 U.S. 57 (1984); United States v. O’Connor, 2011 U.S. App. LEXIS 12388, at *40 (2d Cir. June 16, 2011).
53. Manual of Model Criminal Jury Instructions for the District Courts of the Ninth Circuit, Instructions 8.20, 8.23 (2010 ed.).
54. United States v. Jeffers, 570 F.3d 557, 569 (4th Cir.) (quoting instruction), cert. denied, 130 S. Ct. 645 (2009); see, e.g., United States v. Cardona-Rosario, 285 Fed. App’x 20, 25 (3d Cir. 2008), cert. denied, 129 S. Ct. 1387 (2009).
55. Krulewitch v. United States, 336 U.S. 440, 445-46 (1949) (Jackson, J., concurring).
56. See, e.g., Pinkerton v. United States, 328 U.S. 640 (1946); United States v. Vazquez-Castro, 640 F.3d 19, 24-25 (1st Cir. 2011).
57. See, e.g., United States v. Leahy, 445 F.3d 634, 655-56 (3d Cir. 2006); United States v. Stapleton, 293 F.3d 1111, 1115-20 (9th Cir. 2002).
58. E.g., United States v. Jewell, 532 F.2d 697 (9th Cir. 1976) (en banc).
59. E.g., 15 U.S.C. § 78dd-2(h)(3)(B) (Foreign Corrupt Practices Act); 31 U.S.C. § 3729(b)(1)(A)(ii) (False Claims Act).
60. E.g., United States v. Lara-Velasquez, 919 F.2d 946, 951 (5th Cir. 1990).
61. E.g., United States v. Heredia, 483 F.3d 913, 924 n.16 (9th Cir. 2007) (en banc) (disavowing statements in past cases that conscious avoidance instruction should rarely be given).
62. E.g., United States v. Lighty, 616 F.3d 321, 378 (4th Cir.), cert. denied, 131 S. Ct. 846 (2010).
63. E.g., United States v. Reyes, 302 F.3d 48, 54 (2d Cir. 2002).
64. Global-Tech Appliances, Inc. v. SEB S.A., 131 S. Ct. 2060, 2068-72 (2011).
65. E.g., United States v. Heredia, 483 F.3d 913, 922-24 (9th Cir. 2007) (en banc); United States v. Hopkins, 53 F.3d 533, 541 (2d Cir. 1995).
66. E.g., United States v. Geisen, 612 F.3d 471, 486-87 (6th Cir. 2010), cert. denied, 131 S. Ct. 1813 (2011); United States v. Stone, 9 F.3d 934, 937-42 (11th Cir. 1993).
67. Fifth Circuit Criminal Jury Instructions, Instruction 1.05 (2001).
68. Eleventh Circuit Pattern Jury Instructions (Criminal Cases), Instruction 3 (2010) (emphasis in original).
69. Rita v. United States, 551 U.S. 338, 390 (2007) (Souter, J., dissenting); see, e.g., United States v. Vega-Santiago, 519 F.3d 1, 5 (1st Cir. 2008) (en banc).
70. See, e.g., United States v. Stewart, 590 F.3d 93, 134-52 (2d Cir. 2009), cert. denied, 130 S. Ct. 1924 (2010).
71. Supreme Court review and relief under 28 U.S.C. § 2255 are theoretically possible, but rarely of any practical significance.
72. Black v. United States, 130 S. Ct. 2963 (2010); Skilling v. United States, 130 S. Ct. 2896 (2010).
73. United States v. Skilling, 638 F.3d 480 (5th Cir. 2011); United States v. Black, 625 F.3d 386 (7th Cir. 2010), cert. denied, 131 S. Ct. 2932 (2011).
74. E.g., United States v. Adams, 628 F.3d 407, 419 (7th Cir. 2010); United States v. Benally, 546 F.3d 1230, 1240 (10th Cir. 2008), cert. denied, 130 S. Ct. 738 (2009); United States v. Ramirez, 426 F.3d 1344, 1353 (11th Cir. 2005).

via National Association of Criminal Defense Lawyers


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s