FRIENDLY, Circuit Judge:

After trial in the District Court for the Southern District of New York, Philip Peltz, an attorney, was convicted on the four counts of an indictment which Judge Weinfeld submitted to the jury. Count One charged that Peltz had conspired with an employee of the Securities and Exchange Commission to defraud the United States and the Commission of their rights to have their business conducted impartially and according to law and to have the securities laws administered impartially and according to their terms, in violation of 18 U.S.C. § 371. Counts Three and Four charged willful violations of § 10(b) of the Securities Exchange Act of 1934 and the SEC’s Rule 10b-5 in that Peltz had told two brokerage houses that sales of the common stock of Georgia Pacific Corporation which he had ordered were long whereas in fact they were short. Count Five charged that Peltz willfully violated § 10(a), relating to short sales, and the Commission’s Rule 10a-l(a). The court imposed four concurrent one-year sentences, three months of which were to be served in prison; the remainder of the term was suspended as to each count.

The first count charged that from about January 1, 1966, Peltz had conspired with others, including an employee of the SEC, to obtain confidential inside information about matters under consideration by the Commission and use such information for private profit. Specific mention was made of the Georgia Pacific Corporation and Peltz’ sales of its stock. The proof identified the employee as Murray B. Weiner, a branch chief of the SEC’s corporation finance division.

Essential background was furnished by the testimony of Ira Pearce, who at the time was a branch chief of the SEC’s trading and markets division and whose integrity was not in question. On or about March 11, 1966, Pearce had rec*50ommended that the SEC bring suit against Georgia Pacific because the company had apparently been purchasing its own securities through a pension fund it controlled, thereby influencing the market for its own stock in violation of the securities laws. The Commission approved the suit on or about March 24. During April Weiner asked Pearce about the current status of the matter, and Pearce kept him posted as to the target date for the filing of the action. This was originally April 15 but was postponed to April 27.

On April 7, Peltz telephoned a registered representative at Bache & Co., told him he was thinking of selling 1,000 shares of Georgia Pacific which he owned because he had heard there was a lawsuit pending against the company, and directed him to sell the shares at no less than a specific price. In fact, Peltz did not own the shares at that time. On the same day, he instructed a registered representative at Loeb, Rhoades & Co. to sell 2,000 shares allegedly belonging to his mother but, according to the proof, not in fact owned by her. Four days later Weiner asked a former SEC attorney to recommend a broker for Peltz; the attorney suggested a representative at Sterling, Grace & Co., with whom Peltz placed sell orders for 300 shares as long sales. The proof again was that Peltz did not own these shares. All of Peltz’ sales instructions were carried out. He made various excuses for not delivering the shares on the settlement dates but, after the SEC suit was announced and the price of Georgia Pacific fell, he borrowed enough money to enable him to cover his short sales and make a tidy profit.

The most damaging testimony against Peltz was given by a friend, Norman Horwitz, an attorney with experience in securities matters, and Horwitz’ wife, Sheila, who in April 1966 was his fiancée. According to them, Peltz indicated that he had a “contact” at the SEC who had previously given him information about several companies that were to be or had been investigated by the agency. On April 14, Peltz told them, in Sheila’s apartment in New York City, that he had obtained information from his “contact” at the SEC concerning proposed action against Georgia Pacific and had sold several thousand shares, but that he was having trouble “buying in” because the SEC had not made its announcement and the, price of the stock had not yet dropped. Peltz then accompanied Horwitz to the latter’s office and placed a call to a number identified as Weiner’s residence in Maryland. After completing the call, he declared that “Everything .is okay, it’s going to be released in a day or two, they’re working on the final papers.” Sheila testified that Peltz telephoned her apartment several times during the following week. Once he asked her to inform Horwitz that he thought news of the investigation would shortly break; on other occasions he simply provided progress reports. In a subsequent conversation, he told her that in his law practice he had defended prostitutes and “sometimes he would get the company of these girls for his friend in Washington.” Early in May, Peltz expressed concern about being watched by the FBI and, in answer to an inquiry whether he had paid money, said he had not but “expected this fellow would be in shortly” and hoped to avoid paying or seeing him.

A Miss Parkhurst testified that in February 1966 Peltz had phoned her to say that he had a friend called Mel Fein (a name evidently used for Weiner1), to whom he owed a favor and who was at another friend’s apartment. Peltz was unable to entertain him and asked Miss Parkhurst to help. She obliged by going to the apartment and having sexual relations with “Fein.” Some two months later, she reported to Peltz that she had received a call from “Fein.” Peltz became hysterical and told her not to see *51“Fein” again because he was involved in something and that she should not discuss “Mr. Fein or anything” with the FBI if they came to investigate.

I.

The statute, 18 U.S.C. § 371, on which Count 1 was based, provides:

If two or more persons conspire either to commit any offense against the United States, or to defraud the United States, or any agency thereof in any manner or for any purpose, and one or more of such persons do any act to effect the object of the conspiracy, each shall be fined not more than $10,000 or imprisoned not more than five years, or both.

The portion of the enactment relating to defrauding the Government is peculiar, perhaps even unique, in the federal penal code in that it punishes a conspiracy to defraud although the fraud may not constitute a substantive offense under any of the statutes dealing with various forms of corrupting government employees.

Peltz’ first argument is that the evidence was insufficient to show an agreement between himself and Weiner, as distinguished from a disclosure solely at the latter’s initiative, which the Government concedes would not constitute a violation of the conspiracy statute however offensive such conduct would be. But the testimony permitted a rational conclusion that there was a working relationship between the two men. Apart from the evidence of earlier tips, the recipient of unsolicited information would normally not feel entitled to pursue the informant with inquiries why the tip had not been made good or request his help in getting a broker. While it is unnecessary to a conspiracy that the relationship contemplated mutual benefit, however unlikely the contrary might be, the jury could have found an agreement that Weiner was to be rewarded by Peltz’ procuring female company,2 and, indeed, that he had been given some reason to expect money as well.

Peltz’ second challenge to his conspiracy conviction is that the statute is inapplicable when, as here, the Government suffered no pecuniary harm and there was no intention to prevent its taking the action contemplated. This meets the seemingly insurmountable obstacle of the holding in Haas v. Henkel, 216 U.S. 462, 476-480, 30 S.Ct. 249, 54 L.Ed. 569 (1910), that Rev.Stat. § 5440, the predecessor of 18 U.S.C. § 371, prohibited an alleged conspiracy to obtain information from an employee of the Department of Agriculture in advance of the regular release dates of its cotton crop reports and to use such information for speculation upon the cotton market.3 Peltz’ contention that Hammerschmidt v. United States, 265 U.S. 182, 44 S.Ct. 511, 68 L.Ed. 968 (1924) casts doubt upon this highly apposite decision misreads the opinion in the later case. Relying on Mr. Justice Lurton’s statement in Haas that “The statute is broad enough in its terms to include any conspiracy for the purpose of impairing, obstructing or defeating the lawful function of any department of Government,” 216 U.S. at 479, 30 S.Ct. at 254, the United States in Hammerschmidt sought to bring within the statute a conspiracy to urge persons subject to the Selective Service Act to disobey it. In ruling that the indictment did not charge an offense within the statute, the Court narrowed the language just quoted from Haas but not the holding. Noting that in Haas the element of trickery was clear and the only issue was whether pecuniary loss *52to the Government was also required, Chief Justice Taft explained, 265 U.S. at 188, 44 S.Ct. at 512 :

To conspire to defraud the United States means primarily to cheat the government out of property or money, but it also means to interfere with or obstruct one of its lawful governmental functions by deceit, craft or trickery, or at least by means that are dishonest.

Despite academic criticism, see A. Gold-stein, Conspiracy to Defraud the United States, 68 Yale L.J. 405 (1959), the Court has shown no disposition to recede from that view. In United States v. Johnson, 383 U.S. 169, 172, 86 S.Ct. 749, 15 L.Ed.2d 681 (1966), the Court thought it clear that Haas would cover the case of a Congressman’s acceptance of money for the use of his influence in endeavoring to induce the Department of Justice to dismiss mail fraud indictments, although no pecuniary loss to the United States would result from such action and the defendants “were not charged with ‘any false statement, misrepresentation or deceit,’ ” see Dennis v. United States, 384 U.S. 855, 861 n. 5, 86 S.Ct. 1840, 16 L.Ed.2d 973 (1966). That the receipt of consideration likewise is not essential is shown by the statement in Dennis v. United States, supra, 384 U.S. at 860-861, 86 S.Ct. 1840, at 1844 (1966), that a conspiracy to obtain the services of the NLRB by filing false non-Communist affidavits came within 18 U.S.C. § 371. An agreement whereby a federal employee will act to promote private benefit in breach of his duty thus comes within the statute if the proper functioning of the Government is significantly affected thereby.

It scarcely needs argument that the high repute and effective functioning of the SEC — conspicuous for its zeal in preventing the misuse of inside information, see, e.g., Cady, Roberts & Co., 40 SEC 907 (1961); SEC v. Texas Gulf Sulphur Co., 401 F.2d 833 (2 Cir. 1968), cert. denied, 394 U.S. 976, 89 S.Ct. 1454, 22 L.Ed.2d 756 (1969) — would be significantly compromised by arrangements whereby an individual could obtain information about its impending action from one of its employees and profit from having such knowledge before this became available to the public generally.4 Public confidence essential to the effective functioning of government would be seriously impaired by any arrangement that would enable a few individuals to profit from advance knowledge of governmental action. The very making of a plan whereby a government employee will divulge material information which he knows he should not is “dishonest” within Chief Justice Taft’s language in Hammerschmidt,5 regardless of whether such plan is secured by consideration.

II.

If this opinion had been written prior to June 23, 1969, it could have stopped at this point. Peltz' sentences on the two § 10(b) and the § 10(a) counts were concurrent with the sentence on the conspiracy count, and we see no likelihood that submission of the latter would have had a prejudicial “spill-over” on the conspiracy conviction. See United States v. Hines, 256 F.2d 561, 562-563 (2 Cir. 1958). While we do not pretend to un*53derstand exactly where Benton v. Maryland, 395 U.S. 784, 788-791, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969), leaves the “concurrent sentence doctrine,” see United States v. Febre, 425 F.2d 107, 113-114 (2 Cir. 1970), the question is so obscure and the issues with respect to the §§ 10(b) and 10(a) counts are sufficiently likely to recur that we think it best to consider Peltz’ attacks on these convictions.

Section 32(a) of the Securities Exchange Act, discussed in more detail in Part III of this opinion, makes criminal any willful violation of any provision of the act or any rule or regulation thereunder the violation of which is made unlawful. Section 10(b) of the Act and the Commission’s Rule 10b-5 thereunder, both too familiar to warrant quotation, admittedly qualify under this test. The Government’s submission was that by Peltz’ assertions to Loeb, Rhoades & Co. and to Bache & Co. with respect to ownership of the stock of Georgia Pacific whose sale he directed, he willfully and knowingly made untrue statements of material facts and engaged in acts that would operate as a fraud or deceit in connection with the sale of a security.

Prima facie this would seem entirely clear. The immediate consequence of the lies was that the brokerage houses did not demand the 70% collateral required by the then applicable regulation of the Federal Reserve Board, 12 C.F.R. § 220.3, for short sales. As a result the firms were exposed to risk of serious loss when, as turned out to be the case, Peltz did not deliver the stock on the normal settlement date. Also, as we shall see, the lies caused the brokerage firms to violate § 10(a) and Rule 10a-l relating to the price at which short sales can lawfully be made. Even though these violations of the margin and short sale rules were unwitting, the brokers were subject to possible disciplinary proceedings by the New York Stock Exchange. Despite the caveat in SEC v. Capital Gains Research Bureau, Inc., 375 U.S. 180, 192 n. 40, 84 S.Ct. 275, 11 L.Ed.2d 237 (1963), we see no more reason for requiring proof of actual (as distinguished from potential) injury in a criminal prosecution than in a suit for an injunction in a securities fraud case.

Indeed, Peltz does not so contend. His position is rather that there can be no criminal liability for violations of § 10(b) and Rule 10b-5 “where the fraud was not related to the investment value of securities.” Recognizing our decision in A. T. Brod & Co. v. Perlow, 375 F.2d 393, 397 (2 Cir. 1967), that the placing of purchase orders with the intent to pay only if the securities had appreciated by the settlement date gave rise to civil liability under the statute and the rule, he argues that the same words must have a narrower application in criminal prosecutions. We are not convinced of this, Herlands, Criminal Aspects of the Securities Exchange Act of 1934, 21 Va.L. Rev. 139, 162 (1934), nor do we believe that Peltz’ formulation would be appropriate if any narrowing were to be done. Short selling without compliance with the margin and short sale price rules can have a materially larger adverse effect on the public than a seller’s hoodwinking a buyer into an unfortunate purchase of a few hundred shares.6

*54III.

Section 10(a) of the Securities Exchange Act of 1934 makes it unlawful “for any person, directly or indirectly, by the use * * * of any facility of any national securities exchange” to effect a short sale of any security registered on a national securities exchange in contravention of such rules and regulations as the Commission may prescribe as necessary in the public interest or for the protection of investors. The Commission’s Rule 10a-l(a), sometimes called the “down-tick” rule, provides:

No person shall, for his own account or for the account of any other person, effect on a national securities exchange a short sale of any security (1) below the price at which the last sale thereof, regular way, was effected on such exchange, or (2) at such price unless such price is above the next preceding different price at which a sale of such security, regular way, was effected on such exchange * * *.

The Government established that Rule 10a-l(a) had been violated in the case of 1700 of the 3000 Georgia Pacific shares Peltz ordered to be sold on April 7, 1966.

Peltz’ first contention, that the statute and rule apply only to brokers and dealers, requires little discussion. The statute speaks of “any person” and the intention to go beyond the “person” actually arranging the trade is made manifest by the phrase “directly or indirectly.” The rule is even clearer in covering the person for whose account the sale is made as well as the person making it. See United States v. Mandel, 296 F.Supp. 1038, 1040 (S.D.N.Y. 1969). We thus would have no doubt that a customer explicitly instructing a broker to make a short sale in contravention of Rule 10a-l(a) would be as much subject to criminal liability as the broker who carried out the instruction.

However, there was no evidence that Peltz gave any such instruction and we must confront the issue whether a customer “willfully” violates Rule 10a-l(a) when he deliberately makes a false statement to a broker that he owns securities directed to be sold and there is a resulting violation of the rule, because the broker, having no reason to think that the rule was applicable, failed to take the precautions that would prevent such a violation. In considering this question it will be helpful to have the full text of § 32(a) of the Securities Exchange Act before us:

Any person who willfully violates any provision of this chapter, or any rule or regulation thereunder the violation of which is made unlawful or the observance of which is required under the terms of this chapter, or any person who willfully and knowingly makes, or causes to be made, any statement in any application, report, or document required to be filed under this chapter or any rule or regulation thereunder or any undertaking contained in a registration statement as provided in subsection (d) of section 15 of this title, which statement was false or misleading with respect to any material fact, shall upon conviction be fined not more than $10,000, or imprisoned not more than two years, or both, except that when such person is an exchange, a fine not exceeding $500,000 may be imposed; but no person shall be subject to imprisonment under this section for the violation of any rule or regulation if he proves that he had no knowledge of such rule or regulation.

The language makes one point entirely clear. A person can willfully violate an SEC rule even if he does not know of its existence. This conclusion follows from the difference between the standard for violation of the statute or a rule or regulation, to wit, “willfully,” and that for false or misleading statements, namely, “willfully and knowingly.” It follows also from the proviso whereby lack of knowledge of a rule or regulation pre*55vents imprisonment but not a fine.7 All this was pointed out in a penetrating article written by the late Judge Her-lands many years before his appointment to the federal bench, Criminal Law Aspects of the Securities Exchange Act of 1934, supra, 21 Va.L.Rev. at 148-149.

Decision that the Government need not prove knowledge of Rule 10a-l(a)8 in order to establish criminal liability still leaves the question what mental state it must prove. The Herlands article concluded it was necessary only that “the prosecution establishes a realization on the defendant’s part that he was doing a wrongful act,” 21 Va.L.Rev. at 149. We accept this with the qualifications, doubtless intended by the author, that the act be wrongful under the securities laws and that the knowingly wrongful act involve a significant risk of effecting the violation that has occurred. See A.L.I. Model Penal Code, § 2.03 (1962); Study Draft of a New Federal Criminal Code, § 305 (1970).

On this analysis the court was clearly justified in submitting the § 10(a) counts to the jury under proper instructions. There was ample evidence that Peltz was aware “he was doing a wrongful act,” namely, telling the brokers he was long when he knew he was not. His false statements, negating the facts that would have induced the brokers to take the requisite steps to avoid a violation of 10a-l(a), created a serious risk that such a violation would occur, as it did. Cf. Pereira v. United States, 347 U.S. 1, 8-9, 74 S.Ct. 358, 98 L.Ed. 435 (1954).

The conviction on all four counts is affirmed.

GRUENDER, Circuit Judge.

Bryan Behrens pled guilty to one count of securities fraud in violation of 15 U.S.C. §§ 78j(b), 78ff and 17 C.F.R. § 240.10b-5 (“Rule 10b-5”). The district court1 sentenced Behrens to five years’ imprisonment, three years of supervised release, and restitution in the amount of $6,841,921.90. Behrens appeals his sentence, arguing that because he had no knowledge his conduct violated Rule lob-5, imprisonment is not a permissible sentencing option. We affirm the district court’s sentence.

I.

Behrens owned and operated 21st Century Financial Group, Inc., a life insurance agency and financial investment advising business. He later expanded his business dealings by becoming the sole owner and operator of National Investments, Inc. (“Nil”). Behrens promoted Nil to his clients as a safe and lucrative investment opportunity. In exchange for their loans to Nil, Behrens issued the investors prom*928issory notes, which indicated the bearer would receive a seven to nine percent fixed rate of return. However, Behrens did not invest Nil’s funds in real estate as promised. Instead, he effectuated a Ponzi scheme, using investor funds to prop up his other business interests and support his profligate spending habits. After an SEC investigation, a federal grand jury returned a twenty-one-count indictment in April 2009.

Behrens and the Government reached a plea agreement, but it did not include a provision regarding Behrens’s final sentence. The presentence investigation report calculated the advisory guidelines imprisonment range as 121-151 months. At the sentencing hearing, Behrens argued he was ineligible for imprisonment under § 78ff(a)’s “no knowledge” defense. Section 78ff(a) imposes criminal liability for “[wjillful violations” of securities laws or SEC rules or regulations but allows defendants who prove they had “no knowledge” of the rule or regulation they violated to avoid imprisonment:

Any person who willfully violates any provision of this chapter ... or any rule or regulation thereunder ... shall upon conviction be fined not more than $5,000,000, or imprisoned not more than 20 years ... but no person shall be subject to imprisonment under this section for the violation of any rule or regulation if he proves that he had no knowledge of such rule or regulation.

The district court initially ruled that the “no knowledge” defense applies only to those convicted of violating SEC rules, rather than securities statutes, and because Behrens had pled guilty to violating the latter (§ 78j(b)), he accordingly could not take advantage of the “no-knowledge” provision. We vacated the district court’s imposition of sixty months’ imprisonment, holding that individuals who are charged with violating both § 78j(b) and Rule 10b-5 are still “entitled to assert the no-knowledge defense to imprisonment at sentencing” because one of the elements of the conduct proscribed by § 78j(b) is the violation of an SEC rule or regulation.2 United States v. Behrens, 644 F.3d 754, 757 (8th Cir.2011). We noted that Behrens bore the “burden of showing no knowledge” and remanded the matter to the district court for consideration of that question in the first instance. Id.

On remand, the district court imposed the same sentence.. The court found that, because Behrens “was aware of the verbatim provisions of Rule 10b-5 and that they proscribed illegal conduct,” he could not meet his burden of proving he had no knowledge of Rule 10b-5. On appeal, Behrens argues that even if he was aware of the provisions of Rule 10b-5 as they relate to securities, he can still obtain shelter under the “no knowledge” defense if he can show that he did not know the promissory notes he issued through Nil constituted securities within the scope of Rule 10b-5. In effect, Behrens insists he had “no knowledge” of Rule 10b-5 because he was ignorant of its applicability to his conduct. The Government, in contrast, interprets the no-knowledge provision as allowing Behrens to avoid prison only if he can show by a preponderance of the evidence that he was unaware of Rule 10b-5’s very existence.

II.

“As with any question of statutory interpretation, our analysis begins with the *929plain language of the statute.” Behrens, 644 F.3d at 755 (quoting United States v. Jeanpierre, 636 F.3d 416, 425 (8th Cir.2011)). The level of knowledge referenced in § 78ff(a)’s “no knowledge” defense to imprisonment is a matter of first impression for us, but we do not approach this topic upon a blank slate. “Statutory language must be read in context and a phrase ‘gathers meaning from the words around it.’ ” Jones v. United States, 527 U.S. 373, 389, 119 S.Ct. 2090, 144 L.Ed.2d 370 (1999) (quoting Jarecki v. G.D. Searle & Co., 367 U.S. 303, 307, 81 S.Ct. 1579, 6 L.Ed.2d 859 (1961)). Section 78ff(a) first makes imprisonment a possibility for “[a]ny person who willfully violates” an SEC rule or regulation but then carves out a safe harbor for those with “no knowledge of such rule or regulation.” We have interpreted “willfully violates” in this context as requiring proof of “the intentional doing of the wrongful acts,” but not as requiring proof that the defendant knew of a particular securities law or SEC rule prohibiting his actions. United States v. O’Hagan, 139 F.3d 641, 647 (8th Cir.1998); accord United States v. Tarallo, 380 F.3d 1174, 1187 & n. 3 (9th Cir.2004). “A person can willfully violate an SEC rule even if he does not know of its existence,” so long as “the act be wrongful under the securities laws” and “the knowingly wrongful act involve a significant risk of effecting the violation that has occurred.” United States v. Peltz, 433 F.2d 48, 54-55 (2d Cir.1970). In order for it to be meaningful, then, the “no knowledge” defense must offer at least some but not all of these willful violators the added protection of avoiding imprisonment. Cf. Ratzlaf v. United States, 510 U.S. 135, 140-41, 114 S.Ct. 655, 126 L.Ed.2d 615 (1994) (describing how judges should “hesitate” to treat statutory terms as surplusage). The question before us concerns the precise scope of this subset.

There is some initial appeal to the Government’s plain meaning interpretation of “no knowledge of such rule or regulation” as requiring a complete absence of knowledge of the pertinent SEC rule or regulation; in other words, no knowledge of its very existence. However, language from the Supreme Court’s opinion in United States v. O’Hagan, 521 U.S. 642, 117 S.Ct. 2199, 138 L.Ed.2d 724 (1997), indicates we should read the no-knowledge provision more broadly. O’Hagan involved whether trading on misappropriated information violated Rule 10b-5 because it constituted “deceptive” conduct “in connection with” securities transactions. Id. at 653, 117 S.Ct. 2199. After concluding that such conduct did violate Rule 10b-5 and therefore could lead to criminal liability under § 78ff(a), the O’Hagan Court characterized the “no knowledge” defense as one of two “sturdy safeguards Congress has provided regarding scienter” for criminal convictions for securities fraud. Id. at 665, 117 S.Ct. 2199. If the “no knowledge” defense to imprisonment were available only to those who deal in securities with a complete lack of awareness of the existence of an SEC rule or regulation, it would hardly be a “sturdy safeguard[ ] ... regarding scienter.” As a result, “no knowledge” cannot be limited, as the Government urges, to no knowledge of the existence of the pertinent SEC rule or regulation.

At the same time, there is little support for Behrens’s highly specific interpretation of “no knowledge” as no knowledge that one’s conduct actually violates the particular SEC rule he is convicted of transgressing. The meaning of “willfully violates” in § 78ff(a) is consistent with the traditional principle that “ignorance of the law is no excuse.” Bryan v. United States, 524 U.S. 184, 193-96, 118 S.Ct. 1939, 141 L.Ed.2d 197 (1998) (rejecting *930petitioner’s proposed definition of the willfulness requirement in 18 U.S.C. § 924(1)(D) as creating an exception to this traditional rule by “requir[ing] knowledge of the law”). The “no knowledge” defense, by requiring some level of regulatory awareness, undeniably departs from this traditional rule; the only question is how broadly the exception sweeps. We will not “interpolate words into a statute when the Congress has not seen fit to use such words itself.” Ark. Valley Indus., Inc. v. Freeman, 415 F.2d 713, 718 (8th Cir.1969). Section 78ff(a) does not say that defendants can avoid imprisonment if they can prove they had no knowledge that their conduct actually violated the pertinent SEC rule or regulation. If Congress so intended, it could have chosen to say, for example, that no person shall be imprisoned “if he proves that he had no knowledge he was violating such rule or regulation.” Instead, § 78ff(a) simply speaks of “no knowledge of such rule or regulation.” Therefore, we believe the better reading of the no-knowledge provision is to allow individuals to avoid a sentence of imprisonment if they can establish that they did not know the substance of the SEC rule or regulation they allegedly violated, regardless of whether they understood its particular application to then-conduct.

Our interpretation accords with the majority of other courts to have addressed the issue. The Ninth Circuit has reviewed a similar challenge to the mens rea required under the “no knowledge” defense. United States v. Reyes, 577 F.3d 1069, 1080-82 (9th Cir.2009). After concluding that the jury convicted one of the defendants of violating Rule 13b2-l, which prohibits the falsification of books and records, the district court at sentencing had to determine whether the defendant fell within the purview of the “no knowledge” defense. Id. at 1074. The analogue to the Government’s suggested interpretation of the no-knowledge provision would have been to ask whether the defendant was aware of the existence of Rule 13b2-l; the analogue to Behrens’s suggested interpretation would have been to ask whether the defendant knew that her conduct violated Rule 13b2-l. In lieu of either approach, the district court identified the relevant question as “whether [the defendant] has satisfied her burden of proving by a preponderance that she was unaware of a SEC rule or regulation prohibiting the falsification of books and records.” Id. at 1081 (quoting United States v. Jensen, 537 F.Supp.2d 1069, 1075 (N.D.Cal.2008)). The Ninth Circuit confirmed “[t]his was the proper formulation,” id., in effect approving the interpretation of “no knowledge” we adopt today.3 A defendant seek*931ing to take advantage of the no-knowledge provision need not prove that he was unaware of the existence of the SEC rule or regulation he is accused of violating, but it is insufficient to show that, although he was aware of the substance of a rule, he simply did not know that it applied to his particular course of conduct. Instead, a defendant will succeed in establishing the no-knowledge defense if he can show that he had no knowledge of the substance of the SEC rule or regulation he is convicted of violating. See also United States v. Johnson, 413 Fed.Appx. 151, 154-55 (11th Cir.2011) (affirming imposition of prison sentence where sufficient evidence indicated that the defendant “was aware of the, general substance of the securities laws and regulations” he had been convicted of violating).

This rule protects from imprisonment individuals who truly are unaware of the substance of an SEC rule or regulation, but it does not go so far as to completely vitiate the principle that ignorance of the law is no defense. Behrens asks that we fully depart from this principle and hold that only those with a specific intent to violate a particular SEC rule or regulation are eligible for imprisonment. In certain narrow circumstances, such as when interpreting criminal tax statutes, the Supreme Court has adopted such a definition of the term “willfully.” See, e.g., Cheek v. United States, 498 U.S. 192, 200-02, 111 S.Ct. 604, 112 L.Ed.2d 617 (1991) (holding that willfully “requires the Government to prove that the law imposed a duty on the defendant, that the defendant knew of this duty, and that he voluntarily and intentionally violated that duty”). The difference between the mens rea requirement the Court adopted in Cheek and the one we adopt today is that under our interpretation of the “no knowledge” defense, a defendant cannot raise a claimed good faith misunderstanding of the duties imposed on him, so long as he understands the substance of the SEC rule he was convicted of violating. To the extent the Court’s heightened mens rea requirement in criminal tax cases is driven by concerns related to the risks of “convicting individuals engaged in apparently innocent activity,” Bryan, 524 U.S. at 195, 118 S.Ct. 1939, the same fears do not animate our interpretation of the no-knowledge provision. For the most part, we are not confronted with “average citizen[s]” who must struggle to understand “the extent of the duties and obligations imposed by” securities laws. Cf. Cheek, 498 U.S. at 199-200, 111 S.Ct. 604. Individuals must pass licensing exams in order to lawfully deal in securities and related products; Behrens himself held four securities licenses (Series 6, 26, 63, and 65). For those laymen who— intentionally or otherwise — dabble in securities and run afoul of SEC rules or regulations, their rudimentary training would tend not to provide them with the requisite knowledge of the substance of the regulatory provisions they violated, therefore protecting them from imprisonment. Beh-rens’s interpretation would offer added protection primarily to relatively savvy individuals who have a baseline understanding of SEC rules and regulations but who claim to misunderstand some nuanced application of these rules. The purpose of the no-knowledge provision is certainly to “soften[] the impact of the common-law *932presumption” that “mistake of law is no defense,” cf. id. at 199, 111 S.Ct. 604, but Behrens’s interpretation deviates too far from this deeply rooted principle.4

Behrens, therefore, is ineligible for a term of imprisonment if he can prove by a preponderance of the evidence that he had no knowledge of the substance of Rule 10b-5. We review the district court’s factual findings for clear error. United States v. Hyatt, 207 F.3d 1036, 1038 (8th Cir.2000). During the resentencing hearing, the Government questioned Behrens about each of Rule 10b-5’s subsections.

Q [Government]. Right. In any event, based on your experience, you knew that it was illegal to fraudulently take money from investors in connection with the purchase or sale of a security, didn’t you?

A [Behrens]. Yes.

Q. You knew that it was illegal to make a misrepresentation of a material fact or an omission of a material fact related to the sale or purchase of a security, didn’t you?

A. As it relates tó a security, yes.

Q. You know that it was illegal to engage in a course of conduct which operates as a fraud or deceit upon a person relating to the purchase or sale of a security, didn’t you?

A. As it relates to securities, yes.

Q. And you know that — do you know what I’ve just read to you is Rule 10b-5?

A. As it relates to securities.

Based oir Behrens’s responses, the district court concluded that Behrens “had knowledge of the existence of a SEC rule or regulation addressing the illegal conduct memorialized in the provisions of 10b-5.” His admissions are sufficient to support the district court’s conclusion that Behrens had knowledge of the substance of Rule 10b-5. See Lilley, 291 F.Supp. at 993-94 (“[Defendants were familiar with the import of the rule.... [Defendants admitted that they knew securities fraud was prohibited, which is the substance of Rule 10b-5. No more knowledge is required.”). It is of no moment that, as Behrens claims, he did not know Nil’s promissory notes were securities for purposes of Rule 10b-5, because we refuse to go so far as to hold that Behrens had “no knowledge” so long as he did not know that his specific conduct actually violated Rule 10b-5. Beh-rens admitted to knowing the substance of Rule 10b-5, and this removes him from the protection of the no-knowledge provision.

Because Behrens failed to carry his burden of showing that he had no knowledge of the substance of Rule 10b-5, we affirm the district court’s sentence of imprisonment.

OPINION OF THE COURT

[561 U.S. 367]

Justice Ginsburg

delivered the opinion of the Court.

In 2001, Enron Corporation, then the seventh highest-revenue-grossing company in America, crashed into bankruptcy. We consider in this opinion two questions arising from the prosecution of Jeffrey Skilling, a longtime Enron executive, for crimes committed before the corporation’s collapse. First, did pretrial publicity and community prejudice prevent Skilling from obtaining a fair trial? Second, did the jury improperly convict Skill-ing of conspiracy to commit “honest-services” wire fraud, 18 U.S.C. §§ 371, 1343, 1346?

Answering no to both questions, the Fifth Circuit affirmed Skilling’s convictions. We conclude, in common with the Court of Appeals, that Skill-ing’s fair-trial argument fails;

[561 U.S. 368]

Skill-ing, we hold, did not establish that a presumption of juror prejudice arose or that actual bias infected the jury that tried him. But we disagree with the Fifth Circuit’s honest-services ruling. In proscribing fraudulent deprivations of “the intangible right of honest services,” § 1346, Congress intended at least to reach schemes to defraud involving bribes and kickbacks. Construing the honest-services statute to extend beyond that core meaning, we conclude, would encounter a vagueness shoal. We therefore hold that § 1346 covers only bribery and kickback schemes. Because Skilling’s alleged misconduct entailed no bribe or kickback, it does not fall within § 1346’s proscription. We therefore affirm in part and vacate in part.

I

Founded in 1985, Enron Corporation grew from its headquarters in Houston, Texas, into one of the world’s leading energy companies. Skilling launched his career there in 1990 when Kenneth Lay, the company’s founder, hired him to head an Enron subsidiary. Skilling steadily rose through the corporation’s ranks, serving as president and chief operating officer, and then, beginning in February 2001, as chief executive officer. Six months later, on August 14, 2001, Skilling resigned from Enron.

Less than four months after Skill-ing’s departure, Enron spiraled into bankruptcy. The company’s stock, which had traded at $90 per share in August 2000, plummeted to pennies per share in late 2001. Attempting to comprehend what caused the corporation’s collapse, the U. S. Department of Justice formed an Enron Task Force, comprising prosecutors and Federal Bureau of Investigation agents from around the Nation. *636The Government’s investigation uncovered an elaborate conspiracy to prop up Enron’s short-run stock prices by overstating the company’s financial well-being. In the years following Enron’s bankruptcy, the Government prosecuted dozens of Enron employees who participated in the scheme. In time, the Government worked its way up

[561 U.S. 369]

the corporation’s chain of command: On July 7, 2004, a grand jury indicted Skilling, Lay, and Richard Causey, Enron’s former chief accounting officer.

These three defendants, the indictment alleged,

“engaged in a wide-ranging scheme to deceive the investing public, including Enron’s shareholders, . . . about the true performance of Enron’s businesses by: (a) manipulating Enron’s publicly reported financial results; and (b) making public statements and representations about Enron’s financial performance and results that were false and misleading.” App. ¶5, p. 277a.

Skilling and his co-conspirators, the indictment continued, “enriched themselves as a result of the scheme through salary, bonuses, grants of stock and stock options, other profits, and prestige.” Id., ¶14, at 280a.

Count 1 of the indictment charged Skilling with conspiracy to commit securities and wire fraud; in particular, it alleged that Skilling had sought to “depriv[e] Enron and its shareholders of the intangible right of [his] honest services.” Id., ¶87, at 318a.1 The indictment further charged Skill-ing with more than 25 substantive counts of securities fraud, wire fraud, making false representations to Enron’s auditors, and insider trading.

In November 2004, Skilling moved to transfer the trial to another venue; he contended that hostility toward him in Houston, coupled with extensive pretrial publicity, had poisoned potential jurors. To support this assertion, Skilling, aided by media experts, submitted hundreds of news reports detailing Enron’s downfall; he also presented affidavits from

[561 U.S. 370]

the experts he engaged portraying community attitudes in Houston in comparison to other potential venues.

The U. S. District Court for the Southern District of Texas, in accord with rulings in two earlier instituted Enron-related prosecutions,2 denied the venue-transfer motion. Despite “isolated incidents of intemperate commentary,” the court observed, media coverage “ha[d] [mostly] been objective and unemotional,” and the facts of the case were “neither heinous nor sensational.” App. to Brief for United States 10a-11a.3 Moreover, “courts ha[d] commonly” favored “ef*637fective voir dire ... to ferret out any [juror] bias.” Id., at 18a. Pretrial publicity about the case, the court concluded, did not warrant a presumption that Skilling would be unable to obtain a fair trial in Houston. Id., at 22a.

In the months leading up to the trial, the District Court solicited from the parties questions the court might use to screen prospective jurors. Unable to agree on a questionnaire’s

[561 U.S. 371]

format and content, Skilling and the Government submitted dueling documents. On venire members’ sources of Enron-related news, for example, the Government proposed that they tick boxes from a checklist of generic labels such as “television,” “[newspaper,” and “[r]adio,” Record 8415; Skill-ing proposed more probing questions asking venire members to list the specific names of their media sources and to report on “what st[ood] out in [their] mind[s]” of “all the things [they] ha[d] seen, heard or read about Enron,” id., at 8404-8405.

The District Court rejected the Government’s sparer inquiries in favor of Skilling’s submission. Skilling’s questions “[we]re more helpful,” the court said, “because [they] [we]re generally . . . open-ended and w[ould] allow the potential jurors to give us more meaningful information.” Id., at 9539. The court converted Skilling’s submission, with slight modifications, into a 77-question, 14-page document that asked prospective jurors about, inter alia, their sources of news and exposure to Enron-related publicity, beliefs concerning Enron and what caused its collapse, opinions regarding the defendants and their possible guilt or innocence, and relationships to the company and to anyone affected by its demise.4

[561 U.S. 372]

In November 2005, the District Court mailed the questionnaire to 400 prospective jurors and received responses from nearly all the addressees. The court granted hardship exemptions to approximately 90 individuals, id., at 11773-11774, and *638the parties, with the court’s approval, further winnowed the pool by excusing another 119 for cause, hardship, or physical disability, id., at 11891, 13594. The parties agreed to exclude, in particular, “each and every” prospective juror who said that a preexisting opinion about Enron or the defendants would prevent her from impartially considering the evidence at trial. Id., at 13668.

On December 28, 2005, three weeks before the date scheduled for the commencement of trial, Causey pleaded guilty. Skilling’s attorneys immediately requested a continuance, and the District Court agreed to delay the proceedings until the end of January 2006. Id., at 14277. In the interim, Skilling renewed his change-of-venue motion, arguing that the juror questionnaires revealed pervasive bias and that news accounts of Causey’s guilty plea further tainted the jury pool. If Houston remained the trial venue, Skilling urged that “jurors need to be questioned individually by both the Court and counsel” concerning their opinions of Enron and “publicity issues.” Id., at 12074.

The District Court again declined to move the trial. Skilling, the court concluded, still had not “establish [ed] that pretrial publicity and/or community prejudice raise[d] a presumption of inherent jury prejudice.” Id., at 14115. The questionnaires and voir dire, the court observed, provided

[561 U.S. 373]

safeguards adequate to ensure an impartial jury. Id., at 14115-14116.

Denying Skilling’s request for attorney-led voir dire, the court said that in 17 years on the bench:

“I’ve found ... I get more forthcoming responses from potential jurors than the lawyers on either side. I don’t know whether people are suspicious of lawyers—but I think if I ask a person a question, I will get a candid response much easier than if a lawyer asks the question.” Id., at 11805.

But the court promised to give counsel an opportunity to ask followup questions, ibid., and it agreed that venire members should be examined individually about pretrial publicity, id., at 11051-11053. The court also allotted the defendants jointly 14 peremptory challenges, 2 more than the standard number prescribed by Federal Rule of Criminal Procedure 24(b)(2) and (c)(4)(B). Id., at 13673-13675.

Voir dire began on January 30, 2006. The District Court first emphasized to the venire the importance of impartiality and explained the presumption of innocence and the Government’s burden of proof. The trial, the court next instructed, was not a forum “to seek vengeance against Enron’s former officers,” or to “provide remedies for” its victims. App. 823a. “The bottom line,” the court stressed, “is that we want... jurors who ... will faithfully, conscientiously and impartially serve if selected.” Id., at 823a-824a. In response to the court’s query whether any prospective juror questioned her ability to adhere to these instructions, two individuals indicated that they could not be fair; they were therefore excused for cause, id., at 816a, 819a-820a.

After questioning the venire as a group,5 the District Court brought prospective jurors one by one to the *639bench

[561 U.S. 374]

for individual examination. Although the questions varied, the process generally tracked the following format: The court asked about exposure to Enron-related news and the content of any stories that stood out in the prospective juror’s mind. Next, the court homed in on questionnaire answers that raised a red flag signaling possible bias. The court then permitted each side to pose followup questions. Finally, after the venire member stepped away, the court entertained and ruled on challenges for cause. In all, the court granted one of the Government’s for-cause challenges and denied four; it granted three of the defendants’ challenges and denied six. The parties agreed to excuse three additional jurors for cause and one for hardship.

By the end of the day, the court had qualified 38 prospective jurors, a number sufficient, allowing for peremptory challenges, to empanel 12 jurors and 4 alternates.6 Before the jury was sworn in, Skilling objected to the seating of six jurors. He did not contend that they were in fact biased; instead, he urged that he would have used peremptories to exclude them had he not exhausted his supply by striking

[561 U.S. 375]

several venire members after the court refused to excuse them for cause. Supp. App. 3sa-4sa (Sealed).7 The court overruled this objection.

After the jurors took their oath, the District Court told them they could not discuss the case with anyone or follow media accounts of the proceedings. “[E]ach of you,” the court explained, “needs to be absolutely sure that your decisions concerning the facts will be based only on the evidence that you hear and read in this courtroom.” App. 1026a.

Following a four-month trial and nearly five days of deliberation, the jury found Skilling guilty of 19 counts, including the honest-services-fraud conspiracy charge, and not guilty of 9 insider-trading counts. The District Court sentenced Skilling to 292 months’ imprisonment, 3 years’ supervised release, and $45 million in restitution.

On appeal, Skilling raised a host of challenges to his convictions, including the fair-trial and honest-services arguments he presses here. Regarding the former, the Fifth Circuit initially determined that the volume and negative tone of media coverage generated by Enron’s collapse created a presumption of juror prejudice. 554 *640F.3d 529, 559 (2009).8 The court also noted potential prejudice stemming from Causey’s guilty plea and from the large number of victims in Houston—from the “thousands of Enron employees

[561 U.S. 376]

. . . [who] lost their jobs, and . . . saw their 401(k) accounts wiped out,” to Houstonians who suffered spillover economic effects. Id., at 559-560.

The Court of Appeals stated, however, that “the presumption [of prejudice] is rebuttable,” and it therefore examined the voir dire to determine whether “the District Court empaneled an impartial jury.” Id., at 561 (internal quotation marks, italics, and some capitalization omitted). The voir dire was, in the Fifth Circuit’s view, “proper and thorough.” Id., at 562. Moreover, the court noted, Skilling had challenged only one seated juror—Juror 11—for cause. Although Juror 11 made some troubling comments about corporate greed, the District Court “observed [his] demeanor, listened to his answers, and believed he would make the government prove its case.” Id., at 564. In sum, the Fifth Circuit found that the Government had overcome the presumption of prejudice and that Skilling had not “show[n] that any juror who actually sat was prejudiced against him.” Ibid.

The Court of Appeals also rejected Skilling’s claim that his conduct did not indicate any conspiracy to commit honest-services fraud. “[T]he jury was entitled to convict Skilling,” the court stated, “on these elements”: “(1) a material breach of a fiduciary duty ... (2) that results in a detriment to the employer,” including one occasioned by an employee’s decision to “withhold material information, i.e., information that he had reason to believe would lead a reasonable employer to change its conduct.” Id., at 547. The Fifth Circuit did not address Skilling’s argument that the honest-services statute, if not interpreted to exclude his actions, should be invalidated as unconstitutionally vague. Brief for Defendant-Appellant Skilling in No. 06-20885 (CA5), p. 65, n. 21.

Arguing that the Fifth Circuit erred in its consideration of these claims, Skilling sought relief from this Court. We granted certiorari, 558 U.S. 945, 130 S. Ct. 393, 175 L. Ed. 2d 267 (2009), and now affirm in

[561 U.S. 377]

part, vacate in part, and remand for further proceedings.9 We consider first Skilling’s allegation of juror prejudice, and next, his honest-services argument.

II

Pointing to “the community passion aroused by Enron’s collapse and the vitriolic media treatment” aimed at him, Skilling argues that his trial “never should have proceeded in Houston.” Brief for Petitioner 20. And *641even if it had been possible to select impartial jurors in Houston, “[t]he truncated voir dire . . . did almost nothing to weed out prejudices,” he contends, so “[f]ar from rebutting the presumption of prejudice, the record below affirmatively confirmed it.” Id., at 21. Skilling’s fair-trial claim thus raises two distinct questions. First, did the District Court err by failing to move the trial to a different venue based on a presumption of prejudice? Second, did actual prejudice contaminate Skilling’s jury?10

A

1

The Sixth Amendment secures to criminal defendants the right to trial by an impartial jury. By constitutional design, that trial occurs “in the State where the . . . Crimes . . .

[561 U.S. 378]

have been committed.” Art. Ill, § 2, cl. 3. See also Amdt. 6 (right to trial by “jury of the State and district wherein the crime shall have been committed”). The Constitution’s place-of-trial prescriptions, however, do not impede transfer of the proceeding to a different district at the defendant’s request if extraordinary local prejudice will prevent a fair trial—a “basic requirement of due process,” In re Murchison, 349 U.S. 133, 136, 75 S. Ct. 623, 99 L. Ed. 942 (1955).11

2

“The theory of our [trial] system is that the conclusions to be reached in a case will be induced only by evidence and argument in open court, and not by any outside influence, whether of private talk or public print.” Patterson v. Colorado ex rel. Attorney General of Colo., 205 U.S. 454, 462, 27 S. Ct. 556, 51 L. Ed. 879 (1907) (opinion for the Court by Holmes, J.). When does the

[561 U.S. 379]

publicity attending conduct charged as criminal dim prospects that the trier can judge a case, as due process requires, impartially, unswayed by outside influence? Because most cases of consequence garner at least some pretrial publicity, courts have considered this question in diverse settings. We begin our discussion by addressing the pre*642sumption of prejudice from which the Fifth Circuit’s analysis in Skilling’s case proceeded. The foundation precedent is Rideau v. Louisiana, 373 U.S. 723, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1963).

Wilbert Rideau robbed a bank in a small Louisiana town, kidnaped three bank employees, and killed one of them. Police interrogated Rideau in jail without counsel present and obtained his confession. Without informing Rideau, no less seeking his consent, the police filmed the interrogation. On three separate occasions shortly before the trial, a local television station broadcast the film to audiences ranging from 24,000 to 53,000 individuals. Rideau moved for a change of venue, arguing that he could not receive a fair trial in the parish where the crime occurred, which had a population of approximately 150,000 people. The trial court denied the motion, and a jury eventually convicted Rideau. The Supreme Court of Louisiana upheld the conviction.

We reversed. “What the people [in the community] saw on their television sets,” we observed, “was Rideau, in jail, flanked by the sheriff and two state troopers, admitting in detail the commission of the robbery, kidnapping, and murder.” Id., at 725, 83 S. Ct. 1417, 10 L. Ed. 2d 663. “[T]o the tens of thousands of people who saw and heard it,” we explained, the interrogation “in a very real sense was Rideau’s trial—at which he pleaded guilty.” Id., at 726, 83 S. Ct. 1417, 10 L. Ed. 2d 663. We therefore “d[id] not hesitate to hold, without pausing to examine a particularized transcript of the voir dire,” that “[t]he kangaroo court proceedings” trailing the televised confession violated due process. Id., at 726-727, 83 S. Ct. 1417, 10 L. Ed. 2d 663.

We followed Rideau’s lead in two later cases in which media coverage manifestly tainted a criminal prosecution. In Estes v. Texas, 381 U.S. 532, 538, 85 S. Ct. 1628, 14 L. Ed. 2d 543 (1965), extensive publicity

[561 U.S. 380]

before trial swelled into excessive exposure during preliminary court proceedings as reporters and television crews overran the courtroom and “bombard [ed] . . . the community with the sights and sounds of’ the pretrial hearing. The media’s overzealous reporting efforts, we observed, “led to considerable disruption” and denied the “judicial serenity and calm to which [Billie Sol Estes] was entitled.” Id., at 536, 85 S. Ct. 1628, 14 L. Ed. 2d 543.

Similarly, in Sheppard v. Maxwell, 384 U.S. 333, 86 S. Ct. 1507, 16 L. Ed. 2d 600 (1966), news reporters extensively covered the story of Sam Sheppard, who was accused of bludgeoning his pregnant wife to death. “[B]edlam reigned at the courthouse during the trial and newsmen took over practically the entire courtroom,” thrusting jurors “into the role of celebrities.” Id., at 353, 355, 86 S. Ct. 1507, 16 L. Ed. 2d 600. Pretrial media coverage, which we characterized as “months [of] virulent publicity about Sheppard and the murder,” did not alone deny due process, we noted. Id., at 354, 86 S. Ct. 1507, 16 L. Ed. 2d 600. But Sheppard’s case involved more than heated reporting pretrial: We upset the murder conviction because a “carnival atmosphere” pervaded the trial, id., at 358, 86 S. Ct. 1507, 16 L. Ed. 2d 600.

In each of these cases, we overturned a “conviction obtained in a trial atmosphere that [was] utterly *643corrupted by press coverage”; our decisions, however, “cannot be made to stand for the proposition that juror exposure to . . . news accounts of the crime . . . alone presumptively deprives the defendant of due process.” Murphy v. Florida, 421 U.S. 794, 798-799, 95 S. Ct. 2031, 44 L. Ed. 2d 589 (1975).12 See also, e.g., Patton v. Yount, 467

[561 U.S. 381]

U.S. 1025, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (1984).13 Prominence does not necessarily produce prejudice, and juror impartiality, we have reiterated, does not require ignorance. Irvin v. Dowd, 366 U.S. 717, 722, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (1961) (Jurors are not required to be “totally ignorant of the facts and issues involved”; “scarcely any of those best qualified to serve as jurors will not have formed some impression or opinion as to the merits of the case.”); Reynolds v. United States, 98 U.S. 145, 155-156, 25 L. Ed. 244 (1879) (“[E]very case of public interest is almost, as a matter of necessity, brought to the attention of all the intelligent people in the vicinity, and scarcely any one can be found among those best fitted for jurors who has not read or heard of it, and who has not some impression or some opinion in respect to its merits.”). A presumption of prejudice, our decisions indicate, attends only the extreme case.

3

Relying on Rideau, Estes, and Sheppard, Skilling asserts that we need not pause to examine the screening questionnaires or the voir dire before declaring his jury’s verdict void. We are not persuaded. Important differences separate

[561 U.S. 382]

Skilling’s prosecution from those in which we have presumed juror prejudice.14

First, we have emphasized in prior decisions the size and characteristics of the community in which the crime occurred. In Rideau, for ex*644ample, we noted that the murder was committed in a parish of only 150,000 residents. Houston, in contrast, is the fourth most populous city in the Nation: At the time of Skilling’s trial, more than 4.5 million individuals eligible for jury duty resided in the Houston area. App. 627a. Given this large, diverse pool of potential jurors, the suggestion that 12 impartial individuals could not be empaneled is hard to sustain. See Mu’Min v. Virginia, 500 U.S. 415, 429, 111 S. Ct. 1899, 114 L. Ed. 2d 493 (1991) (potential for prejudice mitigated by the size of the “metropolitan Washington [D. C.] statistical area, which has a population of over 3 million, and in which, unfortunately, hundreds of murders are committed each year”); Gentile v. State Bar of Nev., 501 U.S. 1030, 1044, 111 S. Ct. 2720, 115 L. Ed. 2d 888 (1991) (plurality opinion) (reduced likelihood of prejudice where venire was drawn from a pool of over 600,000 individuals).15

Second, although news stories about Skilling were not kind, they contained no confession or other blatantly prejudicial information of the type readers or viewers could not reasonably be expected to shut from sight. Rideau’s dramatically

[561 U.S. 383]

staged admission of guilt, for instance, was likely imprinted indelibly in the mind of anyone who watched it. Cf. Parker v. Randolph, 442 U.S. 62, 72, 99 S. Ct. 2132, 60 L. Ed. 2d 713 (1979) (plurality opinion) (“[T]he defendant’s own confession [is] probably the most probative and damaging evidence that can be admitted against him.” (internal quotation marks omitted)). Pretrial publicity about Skilling was less memorable and prejudicial. No evidence of the smoking-gun variety invited prejudgment of his culpability. See United States v. Chagra, 669 F.2d 241, 251-252, n. 11 (CA5 1982) ( “A jury may have difficulty in disbelieving or forgetting a defendant’s opinion of his own guilt but have no difficulty in rejecting the opinions of others because they may not be well-founded.”).

Third, unlike cases in which trial swiftly followed a widely reported crime, e.g., Rideau, 373 U.S., at 724, 83 S. Ct. 1417, 10 L. Ed. 2d 663, over four years elapsed between Enron’s bankruptcy and Skilling’s trial. Although reporters covered Enron-related news throughout this period, the decibel level of media attention diminished somewhat in the years following Enron’s collapse. See App. 700a; id., at 785a; Yount, 467 U.S., at 1032, 1034, 104 S. Ct. 2885, 81 L. Ed. 2d 847.

Finally, and of prime significance, Skilling’s jury acquitted him of nine insider-trading counts. Similarly, earlier instituted Enron-related prosecutions yielded no overwhelming victory for the Government.16 In Rideau, Estes, and Sheppard, in marked contrast, the jury’s verdict did not undermine in any way the supposition of *645juror bias. It would be odd for an appellate court to presume prejudice in a case in which jurors’ actions run counter to that presumption. See, e.g., United States v. Arzola-Amaya, 867 F.2d 1504, 1514

[561 U.S. 384]

(CA5 1989) (“The jury’s ability to discern a failure of proof of guilt of some of the alleged crimes indicates a fair minded consideration of the issues and reinforces our belief and conclusion that the media coverage did not lead to the deprivation of [the] right to an impartial trial.”).

4

Skilling’s trial, in short, shares little in common with those in which we approved a presumption of juror prejudice. The Fifth Circuit reached the opposite conclusion based primarily on the magnitude and negative tone of media attention directed at Enron. But “pretrial publicity— even pervasive, adverse publicity— does not inevitably lead to an unfair trial.” Nebraska Press Assn. v. Stuart, 427 U.S. 539, 554, 96 S. Ct. 2791, 49 L. Ed. 2d 683 (1976). In this case, as just noted, news stories about Enron did not present the kind of vivid, unforgettable information we have recognized as particularly likely to produce prejudice, and Houston’s size and diversity diluted the media’s impact.17

Nor did Enron’s “sheer number of victims,” 554 F.3d, at 560, trigger a presumption of prejudice. Although the widespread community impact necessitated careful identification and inspection of prospective jurors’ connections to Enron, the extensive screening questionnaire and followup voir dire were well suited to that task. And hindsight shows the efficacy of these devices; as we discuss infra, at 389-390, 177 L. Ed. 2d, at 648-649, jurors’ links to Enron were either nonexistent or attenuated.

Finally, although Causey’s “well-publicized decision to plead guilty” shortly before trial created a danger of juror

[561 U.S. 385]

prejudice, 554 F.3d, at 559, the District Court took appropriate steps to reduce that risk. The court delayed the proceedings by two weeks, lessening the immediacy of that development. And during voir dire, the court asked about prospective jurors’ exposure to recent publicity, including news regarding Causey. Only two venire members recalled the plea; neither mentioned Causey by name, and neither ultimately served on Skilling’s jury. App. 888a, 993a. Although publicity about a codefen-dant’s guilty plea calls for inquiry to guard against actual prejudice, it does not ordinarily—and, we are satisfied, it did not here—warrant an automatic presumption of prejudice.

Persuaded that no presumption arose,18 we conclude that the District *646Court, in declining to order a venue change, did not exceed constitutional limitations.19

B

We next consider whether actual prejudice infected Skilling’s jury. Voir dire, Skilling asserts, did not adequately detect and defuse juror bias. “[T]he record . . . affirmatively confirm[s]” prejudice, he maintains, because several seated jurors “prejudged his guilt.” Brief for Petitioner 21. We disagree with Skilling’s characterization of the voir dire and the jurors selected through it.

[561 U.S. 386]

1

No hard-and-fast formula dictates the necessary depth or breadth of voir dire. See United States v. Wood, 299 U.S. 123, 145-146, 57 S. Ct. 177, 81 L. Ed. 78 (1936) (“Impartiality is not a technical conception. It is a state of mind. For the ascertainment of this mental attitude of appropriate indifference, the Constitution lays down no particular tests and procedure is not chained to any ancient and artificial formula.”). Jury selection, we have repeatedly emphasized, is “particularly within the province of the trial judge.” Ristaino v. Ross, 424 U.S. 589, 594-595, 96 S. Ct. 1017, 47 L. Ed. 2d 258 (1976) (internal quotation marks omitted); see, e.g., Mu’Min, 500 U.S., at 424, 111 S. Ct. 1899, 114 L. Ed. 2d 493; Yount, 467 U.S., at 1038, 104 S. Ct. 2885, 81 L. Ed. 2d 847; Rosales-Lopez v. United States, 451 U.S. 182, 188-189, 101 S. Ct. 1629, 68 L. Ed. 2d 22 (1981) (plurality opinion); Connors v. United States, 158 U.S. 408, 408-413, 15 S. Ct. 951, 39 L. Ed. 1033 (1895).

When pretrial publicity is at issue, “primary reliance on the judgment of the trial court makes [especially] good sense” because the judge “sits in the locale where the publicity is said to have had its effect” and may base her evaluation on her “own perception of the depth and extent of news stories that might influence a juror.” Mu’Min, 500 U.S., at 427, 111 S. Ct. 1899, 114 L. Ed. 2d 493. Appellate courts making after-the-fact assessments of the media’s impact on jurors should be mindful that their judgments lack the on-the-spot comprehension of the situation possessed by trial judges.

Reviewing courts are properly resistant to second-guessing the trial judge’s estimation of a juror’s impartiality, for that judge’s appraisal is ordinarily influenced by a host of factors impossible to capture fully in the record—among them, the prospective juror’s inflection, sincerity, demeanor, candor, body language, and apprehension of duty. See Reynolds, 98 U.S., at 156-157, 25 L. Ed. 244. In contrast to the cold transcript received by the appellate court, the in-the-moment voir dire affords the trial court a more intimate and immediate basis for assessing a venire member’s fitness for jury

[561 U.S. 387]

service. We consider the adequacy of jury selection in Skilling’s case, therefore, attentive to the respect due to district-court determina*647tions of juror impartiality and of the measures necessary to ensure that impartiality.20

2

Skilling deems the voir dire insufficient because, he argues, jury selection lasted “just five hours,” “[m]ost of the court’s questions were conclu-sory[,] high-level, and failed adequately to probe jurors’ true feelings,” and the court “consistently took prospective jurors at their word once they claimed they could be fair, no matter what other indications of bias were present.” Brief for Petitioner 10-11 (emphasis

[561 U.S. 388]

deleted). Our review of the record, however, yields a different appraisal.21

As noted, supra, at 370-372, 177 L. Ed. 2d, at 637-638, and n. 4, the District Court initially screened ve-nire members by eliciting their responses to a comprehensive questionnaire drafted in large part by Skilling. That survey helped to identify prospective jurors excusable for cause and served as a springboard for further questions put to remaining members of the array. Voir dire thus was, in the court’s words, the “culmination of a lengthy process.” App. 841a; see 554 F.3d, at 562, n. 51 (“We consider the . . . questionnaire in assessing the quality of voir dire as a whole.”).22 In other Enron-related prosecutions,

[561 U.S. 389]

we note, District Courts, after inspecting *648venire members’ responses to questionnaires, completed the jury-selection process within one day. See supra, at 374, n. 6, 177 L. Ed. 2d, at 639.23

The District Court conducted voir dire, moreover, aware of the greater-than-normal need, due to pretrial publicity, to ensure against jury bias. At Skilling’s urging, the court examined each prospective juror individually, thus preventing the spread of any prejudicial information to other venire members. See Mu’Min, 500 U.S., at 425, 111 S. Ct. 1899, 114 L. Ed. 2d 493. To encourage candor, the court repeatedly admonished that there were “no right and wrong answers to th[e] questions.” E.g., App. 843a. The court denied Skilling’s request for attorney-led voir dire because, in its experience, potential jurors were “more forthcoming” when the court, rather than counsel, asked the question. Record 11805. The parties, however, were accorded an opportunity to ask followup questions of every prospective juror brought to the bench for colloquy. Skilling’s counsel declined to ask anything of more than half of the venire members questioned individually, including eight eventually selected for the jury, because, he explained, “the Court and other counsel have covered” everything he wanted to know. App. 967a.

Inspection of the questionnaires and voir dire of the individuals who actually served as jurors satisfies us that, notwithstanding the flaws Skill-ing lists, the selection process successfully secured jurors who were largely untouched by Enron’s collapse.24 Eleven of the seated jurors and alternates

[561 U.S. 390]

reported no connection at all to Enron, while all other jurors reported at most an insubstantial link. See, e.g., Supp. App. lOlsa (Juror 63) (“I once met a guy who worked for Enron. I cannot remember his *649name.”).25 As for pretrial publicity, 14 jurors and alternates specifically stated that they had paid scant attention to Enron-related news. See, e.g., App. 859a-860a (Juror 13) (would “[b]asically” start out knowing nothing about the case because “I just. . . didn’t follow [it] a whole lot”); id., at 969a (Juror 78) (“[Enron] wasn’t anything that I was interested in reading [about] in detail. ... I don’t really know much about it.”).26 The remaining

[561 U.S. 391]

two jurors indicated that nothing in the news influenced their opinions about Skilling.27

The questionnaires confirmed that, whatever community prejudice existed in Houston generally, Skilling’s jurors were not under its sway.28 Although many expressed sympathy for victims of Enron’s bankruptcy and speculated that greed contributed to the corporation’s collapse, these sentiments did not translate into animus toward Skilling. When asked whether they “ha[d] an opinion about . . . Jeffrey Skilling,” none of the seated jurors and alternates checked the “yes” box.29 And in response to the question whether “any opinion [they] may have formed regarding Enron or [Skilling]

[561 U.S. 392]

[would] prevent” their impartial consideration of the evidence at trial, every juror—despite options to mark “yes” or “unsure”—instead checked “no.”

*650The District Court, Skilling asserts, should not have “accept[ed] at face value jurors’ promises of fairness.” Brief for Petitioner 37. In Irvin v. Dowd, 366 U.S., at 727-728, 81 S. Ct. 1639, 6 L. Ed. 2d 751, Skilling points out, we found actual prejudice despite jurors’ assurances that they could be impartial. Brief for Petitioner 26. Justice Sotomayor, in turn, repeatedly relies on Irvin, which she regards as closely analogous to this case. See post, at 448, 177 L. Ed. 2d, at 685 (dissent). See also, e.g., post, at 441-442, 458, 460, 464, 177 L. Ed. 2d, at 680-681, 690, 691-692, 694. We disagree with that characterization of Irvin.

The facts of Irvin are worlds apart from those presented here. Leslie Irvin stood accused of a brutal murder and robbery spree in a small rural community. 366 U.S., at 719, 81 S. Ct. 1639, 6 L. Ed. 2d 751. In the months before Irvin’s trial, “a barrage” of publicity was “unleashed against him,” including reports of his confessions to the slayings and robberies. Id., at 725-726, 81 S. Ct. 1639, 6 L. Ed. 2d 751. This Court’s description of the media coverage in Irvin reveals why the dissent’s “best case” is not an apt comparison:

“[S]tories revealed the details of [Irvin’s] background, including a reference to crimes committed when a juvenile, his convictions for arson almost 20 years previously, for burglary and by a court-martial on AWOL charges during the war. He was accused of being a parole violator. The headlines announced his police line-up identification, that he faced a he detector test, had been placed at the scene of the crime and that the six murders were solved but [he] refused to confess. Finally, they announced [Irvin’s] confession to the six murders and the fact of his indictment for four of them in Indiana. They reported [Irvin’s] offer to plead guilty if promised a 99-year sentence, but also the determination, on the other hand, of the prosecutor to secure the death penalty, and that [Irvin] had confessed to 24 burglaries (the modus operandi of these robberies was compared to that of the

[561 U.S. 393]

murders and the similarity noted). One story dramatically relayed the promise of a sheriff to devote his life to securing [Irvin’s] execution .... Another characterized [Irvin] as remorseless and without conscience but also as having been found sane by a court-appointed panel of doctors. In many of the stories [Irvin] was described as the ‘confessed slayer of six,’ a parole violator and fraudulent-check artist. [Irvin’s] court-appointed counsel was quoted as having received ‘much criticism over being Irvin’s counsel’ and it was pointed out, by way of excusing the attorney, that he would be subject to disbarment should he refuse to represent Irvin. On the day before the trial the newspapers carried the story that Irvin had orally admitted [to] the murder of [one victim] as well as ‘the robbery-murder of [a second individual] ; the murder of [a third individual], and the slaughter of three members of [a different family].”’ Ibid.

“[Njewspapers in which the[se] stories appeared were delivered regularly to approximately 95% of the dwellings in” the county where the trial occurred, which had a population of only 30,000; “radio and TV stations, which likewise blanketed that county, also carried extensive newscasts covering *651the same incidents.” Id., at 725, 81 S. Ct. 1639, 6 L. Ed. 2d 751.

Reviewing Irvin’s fair-trial claim, this Court noted that “the pattern of deep and bitter prejudice” in the community “was clearly reflected in the sum total of the voir dire“370 prospective jurors or almost 90% of those examined on the point. .. entertained some opinion as to guilt,” and “[8] out of the 12 [jurors] thought [Irvin] was guilty.” Id., at 727, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (internal quotation marks omitted). Although these jurors declared they could be impartial, we held that, “[w]ith his life at stake, it is not requiring too much that [Irvin] be tried in an atmosphere undisturbed by so huge a wave of public passion and by a jury other than one in which two-thirds of

[561 U.S. 394]

the members admit, before hearing any testimony, to possessing a belief in his guilt.” Id., at 728, 81 S. Ct. 1639, 6 L. Ed. 2d 751.

In this case, as noted supra, at 382-383, 177 L. Ed. 2d, at 644, news stories about Enron contained nothing resembling the horrifying information rife in reports about Irvin’s rampage of robberies and murders. Of key importance, Houston shares little in common with the rural community in which Irvin’s trial proceeded, and circulation figures for Houston media sources were far lower than the 95% saturation level recorded in Irvin, see App. to Brief for United States 15a (“The Houston Chronicle . . . reaches less than one-third of occupied households in Houston.” (internal quotation marks omitted)). Skilling’s seated jurors, moreover, exhibited nothing like the display of bias shown in Irvin. See supra, at 389-392, 177 L. Ed. 2d, at 648-650 (noting, inter alia, that none of Skilling’s jurors answered “yes” when asked if they “ha[d] an opinion about . . . Skilling”). See also post, at 444, 177 L. Ed. 2d, at 682 (dissent) (distinguishing Mu’Min from Irvin on similar bases: the “offense occurred in [a large] metropolitan . . . area,” media “coverage was not as pervasive as in Irvin and did not contain the same sort of damaging information,” and “the seated jurors uniformly disclaimed having ever formed an opinion about the case” (internal quotation marks omitted)). In light of these large differences, the District Court had far less reason than did the trial court in Irvin to discredit jurors’ promises of fairness.

The District Court, moreover, did not simply take venire members who proclaimed their impartiality at their word.30 As noted, all of Skilling’s jurors had already affirmed on their questionnaires that they would have no trouble basing

[561 U.S. 395]

a verdict only on the evidence at trial. Nevertheless, the court followed up with each individually to uncover concealed bias. This face-to-face opportunity to gauge demeanor and credibility, coupled with information from the questionnaires regarding jurors’ backgrounds, opinions, and sources of news, gave the court a sturdy foundation to assess fitness for jury service. See 554 F.3d, at 562 (The District Court made “thorough” credibility determinations that “required] more than just the [venire members’] statements that [they] could be fair.”). The jury’s not-guilty verdict on nine insider-trading *652counts after nearly five days of deliberation, meanwhile, suggests the court’s assessments were accurate. See United States v. Haldeman, 559 F.2d 31, 60, n. 28 (CADC 1976). Skill-ing, we conclude, failed to show that his voir dire fell short of constitutional requirements.31

3

Skilling also singles out several jurors in particular and contends they were openly biased. See United States v. Martinez-Salazar, 528 U.S. 304, 316, 120 S. Ct. 774, 145 L. Ed. 2d 792 (2000) ( “[T]he seating of any juror who should have been dismissed for cause . . .

[561 U.S. 396]

require[s] reversal.”). In reviewing claims of this type, the deference due to district courts is at its pinnacle: “A trial court’s findings of juror impartiality may be overturned only for manifest error.” Mu’Min, 500 U.S., at 428, 111 S. Ct. 1899, 114 L. Ed. 2d 493 (internal quotation marks omitted). Skilling, moreover, unsuccessfully challenged only one of the seated jurors for cause, “strong evidence that he was convinced the [other] jurors were not biased and had not formed any opinions as to his guilt.” Beck v. Washington, 369 U.S. 541, 557-558, 82 S. Ct. 955, 8 L. Ed. 2d 98 (1962). With these considerations in mind, we turn to Skilling’s specific allegations of juror partiality.

Skilling contends that Juror 11— the only seated juror he challenged for cause—“expressed the most obvious bias.” Brief for Petitioner 35. See also post, at 460-461, 177 L. Ed. 2d, at 692 (dissent). Juror 11 stated that “greed on Enron’s part” triggered the company’s bankruptcy and that corporate executives, driven by avarice, “walk a line that stretches sometimes the legality of something.” App. 854a-855a. But, as the Fifth Circuit accurately summarized, Juror 11

“had ‘no idea’ whether Skilling had ‘crossed that line,’ and he ‘didn’t say that’ every CEO is probably a crook. He also asserted that he could be fair and require the government to prove its case, that he did not believe everything he read in the paper, that he did not ‘get into the details’ of the Enron coverage, that he did not watch television, and that Enron was ‘old news.’ ” 554 F.3d, at 563-564.

Despite his criticism of greed, Juror 11 remarked that Skilling “earned [his] salar[y],” App. 857a, and said he would have “no problem” telling his co-worker, who had lost 401(k) funds due to Enron’s collapse, that the jury voted to acquit, if that scenario came to pass, id.., at 854a. The District Court, noting that it had “looked [Juror 11] in the eye and . . . heard all his [answers],” found his assertions of impartiality credible. Id., at 858a; cf. supra, at 394, n. 30, 177 L. Ed. 2d, at 651. We agree with the

[561 U.S. 397]

Court of Appeals that “[t]he express finding that *653Juror 11 was fair is not reversible error.” 554 F.3d, at 564.32

Skilling also objected at trial to the seating of six specific jurors whom, he said, he would have excluded had he not already exhausted his peremptory challenges. See supra, at 374-375, 177 L. Ed. 2d, at 639-640. Juror 20, he observes, “said she was ‘angry’ about Enron’s collapse and that she, too, had been ‘forced to forfeit [her] own 401(k) funds to survive layoffs.’ ” Reply Brief 13. But Juror 20 made clear during voir dire that she did not “personally blame” Skilling for the loss of her retirement account. App. 875a. Having not “pa[id] much attention” to Enron-related news, she “quite honestly” did not “have enough information to know” whether Skilling was probably guilty, id., at 873a, and she “th [ought] [she] could be” fair and impartial, id., at 875a. In light of these answers, the District Court did not commit manifest error in finding Juror 20 fit for jury service.

The same is true of Juror 63, who, Skilling points out, wrote on her questionnaire “that [Skilling] ‘probably knew [he] w[as] breaking the law.’ ” Reply Brief 13. During voir dire, however, Juror 63 insisted that she did not “really have an opinion [about Skilling’s guilt] either way,” App. 936a; she did not “know what [she] was thinking” when she completed the questionnaire, but she “absolutely” presumed Skilling innocent and confirmed her understanding that the Government would “have to prove” his guilt, id., at 937a. In response to followup questions from Skilling’s counsel, she again stated she would not presume that Skilling violated any laws and could “[a]bsolutely” give her word that she could be fair. Id., at 937a-938a. “Jurors,” we have recognized, “cannot be expected invariably to express themselves carefully or even consistently.” Yount, 467 U.S., at 1039, 104 S. Ct. 2885, 81 L. Ed. 2d 847. See also id., at 1040, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (“It is here that the federal [appellate]

[561 U.S. 398]

court’s deference must operate, for while the cold record arouses some concern, only the trial judge could tell which of these answers was said with the greatest comprehension and certainty.”). From where we sit, we cannot conclude that Juror 63 was biased.

The four remaining jurors Skilling said he would have excluded with extra peremptory strikes exhibited no sign of prejudice we can discern. See App. 891a-892a (Juror 38) (remembered no media coverage about Enron and said nothing in her experience would prevent her from being fair and impartial); Supp. App. 131sa-133sa, 136sa (Juror 67) (had no connection to Enron and no anger about its collapse); App. 969a (Juror 78) (did not “know much about” Enron); Supp. App. 165sa; App. 971a (Juror 84) (had not heard or read anything about Enron and said she did not “know enough to answer” the question whether she was angry about the company’s demise). Skilling’s counsel declined to ask followup questions of any of these jurors and, indeed, told Juror 84 he had nothing to ask because she “gave all the right answers.” Id., at 972a. Whatever Skill-ing’s reasons for wanting to strike these four individuals from his jury, *654he cannot credibly assert they displayed a disqualifying bias.33

In sum, Skilling failed to establish that a presumption of prejudice arose or that actual bias infected the jury that tried him. Jurors, the trial court correctly comprehended, need not enter the box with empty heads in order to determine the facts impartially. “It is sufficient if the juror[s] can lay aside [their] impression [s] or opinion [s] and render a verdict

[561 U.S. 399]

based on the evidence presented in court.” Irvin, 366 U.S., at 723, 81 S. Ct. 1639, 6 L. Ed. 2d 751. Taking account of the full record, rather than incomplete exchanges selectively culled from it, we find no cause to upset the lower courts’ judgment that Skilling’s jury met that measure. We therefore affirm the Fifth Circuit’s ruling that Skilling received a fair trial.34

Ill

We next consider whether Skilling’s conspiracy conviction was premised on an improper theory of honest-services wire fraud. The honest-services statute, § 1346, Skilling maintains, is unconstitutionally vague. Alternatively, he contends that his conduct does not fall within the statute’s compass.

A

To place Skilling’s constitutional challenge in context, we first review the origin and subsequent application of the honest-services doctrine.

1

Enacted in 1872, the original mail-fraud provision, the predecessor of the modern-day mail- and wire-fraud laws, proscribed, without further elaboration, use of the mails to advance “any scheme or artifice to defraud.” See McNally v. United States, 483 U.S. 350, 356, 107 S. Ct. 2875, 97 L. Ed. 2d 292 (1987) (internal quotation marks omitted). In 1909, Congress amended the statute to prohibit, as it does today, “any scheme or artifice to defraud, or for obtaining money or property by means of false or fraudulent pretenses, representations, or promises.” § 1341

[561 U.S. 400]

(emphasis added); see id., at 357-358, 107 S. Ct. 2875, 97 L. Ed. 2d 292. Emphasizing Congress’ disjunctive phrasing, the Courts of Appeals, one after the other, interpreted the term “scheme or artifice to defraud” to include deprivations not only of money or property, but also of intangible rights.

In an opinion credited with first presenting the intangible-rights theory, Shushan v. United States, 117 *655F.2d 110 (1941), the Fifth Circuit reviewed the mail-fraud prosecution of a public official who allegedly accepted bribes from entrepreneurs in exchange for urging city action beneficial to the bribe payers. “It is not true that because the [city] was to make and did make a saving by the operations there could not have been an intent to defraud,” the Court of Appeals maintained. Id., at 119. “A scheme to get a public contract on more favorable terms than would likely be got otherwise by bribing a public official,” the court observed, “would not only be a plan to commit the crime of bribery, but would also be a scheme to defraud the public.” Id., at 115.

The Fifth Circuit’s opinion in Shushan stimulated the development of an “honest-services” doctrine. Unlike fraud in which the victim’s loss of money or property supplied the defendant’s gain, with one the mirror image of the other, see, e.g., United States v. Starr, 816 F.2d 94, 101 (CA2 1987), the honest-services theory targeted corruption that lacked similar symmetry. While the offender profited, the betrayed party suffered no deprivation of money or property; instead, a third party, who had not been deceived, provided the enrichment. For example, if a city mayor (the offender) accepted a bribe from a third party in exchange for awarding that party a city contract, yet the contract terms were the same as any that could have been negotiated at arm’s length, the city (the betrayed party) would suffer no tangible loss. Cf. McNally, 483 U.S., at 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. Even if the scheme occasioned a money or property gain for the betrayed party, courts reasoned, actionable harm lay in the denial of that party’s right to the offender’s “honest services.” See, e.g., United States v. Dixon, 536 F.2d 1388, 1400 (CA2 1976).

[561 U.S. 401]

“Most often these cases ... involved bribery of public officials,” United States v. Bohonus, 628 F.2d 1167, 1171 (CA9 1980), but courts also recognized private-sector honest-services fraud. In perhaps the earliest application of the theory to private actors, a District Court, reviewing a bribery scheme, explained:

“When one tampers with [the employer-employee] relationship for the purpose of causing the employee to breach his duty [to his employer,] he in effect is defrauding the employer of a lawful right. The actual deception that is practised is in the continued representation of the employee to the employer that he is honest and loyal to the employer’s interests.” United States v. Procter & Gamble Co., 47 F. Supp. 676, 678 (Mass. 1942).

Over time, “[a]n increasing number of courts” recognized that “a recreant employee”—public or private— “c[ould] be prosecuted under [the mail-fraud statute] if he breache[d] his allegiance to his employer by accepting bribes or kickbacks in the course of his employment,” United States v. McNeive, 536 F.2d 1245, 1249 (CA8 1976); by 1982, all Courts of Appeals had embraced the honest-services theory of fraud, Hurson, Limiting the Federal Mail Fraud Statute—A Legislative Approach, 20 Am. Crim. L. Rev. 423, 456 (1983).35

*6562

In 1987, this Court, in McNally v. United States, stopped the development of the intangible-rights doctrine in its tracks. McNally involved a state officer who, in selecting Kentucky’s insurance agent, arranged to procure a share of the agent’s commissions via kickbacks paid to companies the

[561 U.S. 402]

official partially controlled. 483 U.S., at 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. The prosecutor did not charge that, “in the absence of the alleged scheme[,] the Commonwealth would have paid a lower premium or secured better insurance.” Ibid. Instead, the prosecutor maintained that the kickback scheme “defraud [ed] the citizens and government of Kentucky of their right to have the Commonwealth’s affairs conducted honestly.” Id., at 353, 107 S. Ct. 2875, 97 L. Ed. 2d 292.

We held that the scheme did not qualify as mail fraud. “Rather than constru[ing] the statute in a manner that leaves its outer boundaries ambiguous and involves the Federal Government in setting standards of disclosure and good government for local and state officials,” we read the statute “as limited in scope to the protection of property rights.” Id., at 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. “If Congress desires to go further,” we stated, “it must speak more clearly.” Ibid.

3

Congress responded swiftly. The following year, it enacted a new statute “specifically to cover one of the ‘intangible rights’ that lower courts had protected . . . prior to McNally: ‘the intangible right of honest services.’ ” Cleveland v. United States, 531 U.S. 12, 19-20, 121 S. Ct. 365, 148 L. Ed. 2d 221 (2000). In full, the honest-services statute stated:

“For the purposes of th[e] chapter [of the United States Code that prohibits, inter alia, mail fraud, § 1341, and wire fraud, § 1343], the term ‘scheme or artifice to defraud’ includes a scheme or artifice to deprive another of the intangible right of honest services.” § 1346.

B

Congress, Skilling charges, reacted quickly but not clearly: He asserts that § 1346 is unconstitutionally vague. To satisfy due process, “a penal statute [must] define the criminal offense [1] with sufficient definiteness that ordinary people can understand what conduct is prohibited and [2] in a manner that does not encourage arbitrary and discriminatory

[561 U.S. 403]

enforcement.” Kolender v. Lawson, 461 U.S. 352, 357, 103 S. Ct. 1855, 75 L. Ed. 2d 903 (1983). The void-for-vagueness doctrine embraces these requirements.

According to Skilling, § 1346 meets neither of the two due process essentials. First, the phrase “the intangible right of honest services,” he contends, does not adequately define what behavior it bars. Brief for Petitioner 38-39. Second, he alleges, § 1346’s “standardless sweep . . . allows policemen, prosecutors, and juries to pursue their personal predilections,” thereby “facilitat[ing] opportunistic and arbitrary prosecutions.” Id., at 44 (quoting Kolender, 461 U.S., at 358, 103 S. Ct. 1855, 75 L. Ed. 2d 903).

In urging invalidation of § 1346, Skilling swims against our case law’s current, which requires us, if we *657can, to construe, not condemn, Congress’ enactments. See, e.g., Civil Service Comm’n v. Letter Carriers, 413 U.S. 548, 571, 93 S. Ct. 2880, 37 L. Ed. 2d 796 (1973). See also United States v. National Dairy Products Corp., 372 U.S. 29, 32, 83 S. Ct. 594, 9 L. Ed. 2d 561 (1963) (stressing, in response to a vagueness challenge, “[t]he strong presumptive validity that attaches to an Act of Congress”). Alert to § 1346’s potential breadth, the Courts of Appeals have divided on how best to interpret the statute.36 Uniformly, however, they have declined to throw out the statute as irremediably vague.37

[561 U.S. 404]

We agree that § 1346 should be construed rather than invalidated. First, we look to the doctrine developed in pr e-McNally cases in an endeavor to ascertain the meaning of the phrase “the intangible right of honest services.” Second, to preserve what Congress certainly intended the statute to cover, we pare that body of precedent down to its core: In the main, the pr e-McNally cases involved fraudulent schemes to deprive another of honest services through bribes or kickbacks supplied by a third party who had not been deceived. Confined to these paramount applications, § 1346 presents no vagueness problem.

1

There is no doubt that Congress intended § 1346 to refer to and incorporate the honest-services doctrine recognized in Courts of Appeals’ decisions before McNally derailed the intangible-rights theory of fraud. See Brief for Petitioner 39; Brief for United States 37-38; post, at 416, 422, 177 L. Ed. 2d, at 665, 668 (Scalia, J., concurring in part and concurring in judgment). Congress enacted § 1346 on the heels of McNally and drafted the statute using that decision’s terminology. See 483 U.S., at 355, 107 S. Ct. 2875, 97 L. Ed. 2d 292 (“intangible righ[t]”); id., at 362, 107 S. Ct. 2875, 97 L. Ed. 2d 292 (Stevens, J., dissenting) (“right to . . . honest services”).38 As the Second Circuit observed in its leading analysis of § 1346:

“The definite article ‘the’ suggests that ‘intangible right of honest services’ had a specific meaning to Congress when it enacted the statute—Congress was recriminal-izing mail- and wire-fraud schemes *658to deprive others

[561 U.S. 405]

of that ‘intangible right of honest services,’ which had been protected before McNally, not all intangible rights of honest services whatever they might be thought to be.” United States v. Rybicki, 354 F.3d 124, 137-138 (2003) (en banc).39

2

Satisfied that Congress, by enacting § 1346, “meant to reinstate the body of pre-McNally honest-services law,” post, at 422, 177 L. Ed. 2d, at 668 (opinion of Scalia, J.), we have surveyed that case law. See infra, at 407-408, 410, 177 L. Ed. 2d, at 659-660, 661. In parsing the Courts of Appeals decisions, we acknowledge that Skilling’s vagueness challenge has force, for honest-services decisions preceding McNally were not models of clarity or consistency. See Brief for Petitioner 39-42 (describing divisions of opinions). See also post, at 417-420, 177 L. Ed. 2d, at 665-667 (opinion of Scalia, J.). While the honest-services cases preceding McNally dominantly and consistently applied the fraud statute to bribery and kickback schemes—schemes that were the basis of most honest-services prosecutions—there was considerable disarray over the statute’s application to conduct outside that core category. In light of this disarray, Skilling urges us, as he urged the Fifth Circuit, to invalidate the statute in toto. Brief for Petitioner 48 (Section 1346 “is intolerably and unconstitutionally vague.”); Brief Defendant-Appellant Skilling in No. 06-20885 (CA5), p. 65, n. 21 (“[S]ection 1346 should be invalidated as unlawfully vague on its face.”).

It has long been our practice, however, before striking a federal statute as impermissibly vague, to consider whether the prescription is amenable to a limiting construction. See,

[561 U.S. 406]

e.g., Hooper v. California, 155 U.S. 648, 657, 15 S. Ct. 207, 39 L. Ed. 297 (1895) (“The elementary rule is that every reasonable construction must be resorted to, in order to save a statute from unconstitutionality.” (emphasis added)). See also Boos v. Barry, 485 U.S. 312, 330-331, 108 S. Ct. 1157, 99 L. Ed. 2d 333 (1988); Schneider v. Smith, 390 U.S. 17, 26, 88 S. Ct. 682, 19 L. Ed. 2d 799 (1968).40 We have accordingly instructed “the federal courts ... to avoid constitutional difficulties by *659[adopting a limiting interpretation] if such a construction is fairly possible.” Boos, 485 U.S., at 331, 108 S. Ct. 1157, 99 L. Ed. 2d 333; see United States v. Harriss, 347 U.S. 612, 618, 74 S. Ct. 808, 98 L. Ed. 989 (1954) (“[I]f the general class of offenses to which the statute is directed is plainly within its terms, the statute will not be struck down as vague .... And if this general class of offenses can be made constitutionally definite by a reasonable construction of the statute, this Court is under a duty to give the statute that construction.”).

Arguing against any limiting construction, Skilling contends that it is impossible to identify a salvageable honest-services core; “the pre-McNally caselaw,” he asserts, “is a

[561 U.S. 407]

hodgepodge of oft-conflicting holdings” that are “hopelessly unclear.” Brief for Petitioner 39 (some capitalization and italics omitted). We have rejected an argument of the same tenor before. In Civil Service Comm’n v. Letter Carriers, federal employees challenged a provision of the Hatch Act that incorporated earlier decisions of the United States Civil Service Commission enforcing a similar law. “[T]he several thousand adjudications of the Civil Service Commission,” the employees maintained, were “an impenetrable jungle”—“undiscoverable, inconsistent, [and] incapable of yielding any meaningful rules to govern present or future conduct.” 413 U.S., at 571, 93 S. Ct. 2880, 37 L. Ed. 2d 796. Mindful that “our task [wa]s not to destroy the Act if we c[ould], but to construe it,” we held that “the rules that had evolved over the years from repeated adjudications were subject to sufficiently clear and summary statement.” Id., at 571-572, 93 S. Ct. 2880, 37 L. Ed. 2d 796.

A similar observation may be made here. Although some applications of the pre-McNally honest-services doctrine occasioned disagreement among the Courts of Appeals, these cases do not cloud the doctrine’s solid core: The “vast majority” of the honest-services cases involved offenders who, in violation of a fiduciary duty, participated in bribery or kickback schemes. United States v. Runnels, 833 F.2d 1183, 1187 (CA6 1987); see Brief for United States 42, and n. 4 (citing dozens of examples).41 Indeed, the McNally case itself, which spurred Congress to enact § 1346, presented a paradigmatic kickback fact pattern. 483 U.S., at 352-353,

[561 U.S. 408]

360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. Congress’ reversal of McNally and reinstatement of the honest-services *660doctrine, we conclude, can and should be salvaged by confining its scope to the core pre-McNally applications.

As already noted, supra, at 400-401, 177 L. Ed. 2d, at 654-655, the honest-services doctrine had its genesis in prosecutions involving bribery allegations. See Shushan, 117 F.2d, at 115 (public sector); Procter & Gamble Co., 47 F. Supp., at 678 (private sector). See also United States v. Orsburn, 525 F.3d 543, 546 (CA7 2008). Both before McNally and after § 1346’s enactment, Courts of Appeals described schemes involving bribes or kickbacks as “core . . . honest services fraud precedents,” United States v. Czubinski, 106 F.3d 1069, 1077 (CA1 1997); “paradigm case[s],” United States v. deVegter, 198 F.3d 1324, 1327-1328 (CA11 1999); “[t]he most obvious form of honest services fraud,” United States v. Carbo, 572 F.3d 112, 115 (CA3 2009); “core misconduct covered by the statute,” United States v. Urciuoli, 513 F.3d 290, 294 (CA1 2008); “most [of the] honest services cases,” United States v. Sorich, 523 F.3d 702, 707 (CA7 2008); “typical,” United States v. Brown, 540 F.2d 364, 374 (CA8 1976); “clear-cut,” United States v. Mandel, 591 F.2d 1347, 1363 (CA4 1979); and “uniformly . . . cover[ed],” United States v. Paradies, 98 F.3d 1266, 1283, n. 30 (CA11 1996). See also Tr. of Oral Arg. 43 (counsel for the Government) (“[T]he bulk of pre-McNally honest services cases” entailed bribes or kickbacks); Brief for Petitioner 49 (“Bribes and kickbacks were the paradigm [pre-McNally] cases,” constituting “[t]he overwhelming majority of prosecutions for honest-services fraud.”).

In view of this history, there is no doubt that Congress intended § 1346 to reach at least bribes and kickbacks. Reading the statute to proscribe a wider range of offensive conduct, we acknowledge, would raise the due process concerns underlying the vagueness doctrine.42 To preserve the

[561 U.S. 409]

statute without transgressing constitutional limitations, we now hold that § 1346 criminalizes only the bribe-and-kickback core of the pre-McNally case law.43

3

The Government urges us to go *661further by locating within § 1346’s compass another category of proscribed conduct: “undisclosed self-dealing by a public official or private employee—i.e., the taking of official action by the employee that furthers his own undisclosed financial interests while purporting to act in the interests of those to whom he owes a fiduciary duty.” Brief for United States 43-44. “[T]he

[561 U.S. 410]

theory of liability in McNally itself was nondisclosure of a conflicting financial interest,” the Government observes, and “Congress clearly intended to revive th[at] nondisclosure theory.” Id., at 44. Moreover, “[although not as numerous as the bribery and kickback cases,” the Government asserts, “the pre-McNally cases involving undisclosed self-dealing were abundant.” Ibid.

Neither of these contentions withstands close inspection. McNally, as we have already observed, supra, at 401-402, 407, 177 L. Ed. 2d, at 656, 659, involved a classic kickback scheme: A public official, in exchange for routing Kentucky’s insurance business through a middleman company, arranged for that company to share its commissions with entities in which the official held an interest. 483 U.S., at 352-353, 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. This was no mere failure to disclose a conflict of interest; rather, the official conspired with a third party so that both would profit from wealth generated by public contracts. See id., at 352-353, 107 S. Ct. 2875, 97 L. Ed. 2d 292. Reading § 1346 to proscribe bribes and kickbacks—and nothing more—satisfies Congress’ undoubted aim to reverse McNally on its facts.

Nor are we persuaded that the pre-McNally conflict-of-interest cases constitute core applications of the honest-services doctrine. Although the Courts of Appeals upheld honest-services convictions for “some schemes of non-disclosure and concealment of material information,” Mandel, 591 F.2d, at 1361, they reached no consensus on which schemes qualified. In light of the relative infrequency of conflict-of-interest prosecutions in comparison to bribery and kickback charges, and the inter-circuit inconsistencies they produced, we conclude that a reasonable limiting construction of § 1346 must exclude this amorphous category of cases.

Further dispelling doubt on this point is the familiar principle that “ambiguity concerning the ambit of criminal statutes should be resolved in favor of lenity.” Cleveland, 531 U.S., at 25, 121 S. Ct. 365, 148 L. Ed. 2d 221 (quoting Rewis v. United States, 401 U.S. 808,

[561 U.S. 411]

812, 91 S. Ct. 1056, 28 L. Ed. 2d 493 (1971)). “This interpretive guide is especially appropriate in construing [ § 1346] because . . . mail [and wire] fraud [are] predicate offense [s] under [the Racketeer Influenced and Corrupt Organizations Act], 18 U.S.C. § 1961(1) (1994 ed., Supp. IV), and the money laundering statute, § 1956(c)(7)(A).” Cleveland, 531 U.S., at 25, 121 S. Ct. 365, 148 L. Ed. 2d 221. Holding that honest-services fraud does not encom*662pass conduct more wide ranging than the paradigmatic cases of bribes and kickbacks, we resist the Government’s less constrained construction absent Congress’ clear instruction otherwise. E.g., United States v. Universal C. I. T. Credit Corp., 344 U.S. 218, 221-222, 73 S. Ct. 227, 97 L. Ed. 260 (1952).

In sum, our construction of § 1346 “establish [es] a uniform national standard, define[s] honest services with clarity, reach [es] only seriously culpable conduct, and accomplish [es] Congress’s goal of ‘overruling’ McNally.” Brief for Albert W. Alschuler as Amicus Curiae in Weyhrauch v. United States, O. T. 2009, No. 08-1196, pp. 28-29. “If Congress desires to go further,” we reiterate, “it must speak more clearly than it has.” McNally, 483 U.S., at 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292.44

[561 U.S. 412]

4

Interpreted to encompass only bribery and kickback schemes, § 1346 is not unconstitutionally vague. Recall that the void-for-vagueness doctrine addresses concerns about (1) fair notice and (2) arbitrary and discriminatory prosecutions. See Kolender, 461 U.S., at 357, 103 S. Ct. 1855, 75 L. Ed. 2d 903. A prohibition on fraudulently depriving another of one’s honest services by accepting bribes or kickbacks does not present a problem on either score.

As to fair notice, “whatever the school of thought concerning the scope and meaning of ” § 1346, it has always been “as plain as a pikestaff that” bribes and kickbacks constitute honest-services fraud, Williams v. United States, 341 U.S. 97, 101, 71 S. Ct. 576, 95 L. Ed. 774 (1951), and the statute’s mens rea requirement further blunts any notice concern, see, e.g., Screws v. United States, 325 U.S. 91, 101-104, 65 S. Ct. 1031, 89 L. Ed. 1495 (1945) (plurality opinion). See also Broadrick v. Oklahoma, 413 U.S. 601, 608, 93 S. Ct. 2908, 37 L. Ed. 2d 830 (1973) ( “[E]ven if the outermost boundaries of [a statute are] imprecise, any such uncertainty has little relevance . . . where appellants’ conduct falls squarely within the ‘hard core’ of the statute’s proscriptions.”). Today’s decision clarifies that no other misconduct falls within § 1346’s province. See United States v. Lanier, 520 U.S. 259, 266, 117 S. Ct. 1219, 137 L. Ed. 2d 432 (1997) ( “[C]larity at the requisite level may be supplied by judicial gloss on an otherwise uncertain statute.”).

As to arbitrary prosecutions, we perceive no significant risk that the honest-services statute, as we interpret it today, will be stretched out of shape. Its prohibition on bribes and kickbacks draws content not only from the pre-McNally case law, but also from federal statutes proscrib*663ing—and defining—similar crimes. See, e.g., 18 U.S.C. §§ 201(b), 666(a)(2); 41 U.S.C. § 52(2) (“The term ‘kickback’ means any money, fee, commission, credit, gift, gratuity, thing of value, or compensation of any kind which is provided, directly or indirectly, to [enumerated persons] for the purpose of improperly obtaining or rewarding favorable

[561 U.S. 413]

treatment in connection with [enumerated circumstances].”).45 See also, e.g., United States v. Ganim, 510 F.3d 134, 147-149 (CA2 2007) (Sotomayor, J.) (reviewing honest-services conviction involving bribery in light of elements of bribery under other federal statutes); United States v. Whitfield, 590 F.3d 325, 352-353 (CA5 2009); United States v. Kemp, 500 F.3d 257, 281-286 (CA3 2007). A criminal defendant who participated in a bribery or kickback scheme, in short, cannot tenably complain about prosecution under § 1346 on vagueness grounds.

C

It remains to determine whether Skilling’s conduct violated § 1346. Skilling’s honest-services prosecution, the Government concedes, was not “prototypical.” Brief for United States 49. The Government charged Skilling with conspiring to defraud Enron’s shareholders by misrepresenting the company’s fiscal health, thereby artificially inflating its stock price. It was the Government’s theory at trial that Skilling “profited from the fraudulent scheme . . . through the receipt of salary and bonuses, . . . and through the sale of approximately $200 million in Enron stock, which netted him $89 million.” Id., at 51.

The Government did not, at any time, allege that Skilling solicited or accepted side payments from a third party in exchange for making these misrepresentations. See Record 41328 (May 11, 2006 Letter from the Government to the District Court) (“[T]he indictment does not allege, and the government’s evidence did not show, that [Skilling] engaged in bribery.”). It is therefore clear that, as we read § 1346, Skilling did not commit honest-services fraud.

[561 U.S. 414]

Because the indictment alleged three objects of the conspiracy— honest-services wire fraud, money-or-property wire fraud, and securities fraud—Skilling’s conviction is flawed. See Yates v. United States, 354 U.S. 298, 77 S. Ct. 1064, 1 L. Ed. 2d 1356 (1957) ( constitutional error occurs when a jury is instructed on alternative theories of guilt and returns a general verdict that may rest on a legally invalid theory). This determination, however, does not necessarily require reversal of the conspiracy conviction; we recently confirmed, in Hedgpeth v. Pulido, 555 U.S. 57, 129 S. Ct. 530, 172 L. Ed. 2d 388 (2008) (per curiam), that errors of the Yates variety are subject to harmless-error analysis. The parties vigorously dispute whether the error was harmless. Compare Brief for United States 52 (“[A]ny juror who voted for conviction based on [the honest-services theory] also would have found [Skilling] guilty of conspiring to commit securities fraud.”) with Reply Brief 30 (The Government “cannot show that the conspiracy conviction rested only on the securities-fraud theory, rather than the distinct, legally-flawed *664honest-services theory.”). We leave this dispute for resolution on remand.46

Whether potential reversal on the conspiracy count touches any of Skill-ing’s other convictions is also an open question. All of his convictions, Skill-ing contends, hinged on the conspiracy count and, like dominoes, must fall if it falls. The District Court, deciding Skilling’s motion for bail pending appeal, found this argument dubious, App. 1141a-1142a, but the Fifth Circuit had no occasion to rule on it. That court may do so on remand.

[561 U.S. 415]

For the foregoing reasons, we affirm the Fifth Circuit’s ruling on Skilling’s fair-trial argument, vacate its ruling on his conspiracy conviction, and remand the case for proceedings consistent with this opinion.

It is so ordered.

SEPARATE OPINIONS

Justice Scalia,

with whom Justice Thomas joins, and with whom Justice Kennedy joins except as to Part III, concurring in part and concurring in the judgment.

I agree with the Court that petitioner Jeffrey Skilling’s challenge to the impartiality of his jury and to the District Court’s conduct of the voir dire fails. I therefore join Parts I and II of the Court’s opinion. I also agree that the decision upholding Skilling’s conviction for so-called “honest-services fraud” must be reversed, but for a different reason. In my view, the specification in 18 U.S.C. § 1346 (2006 ed.) that “scheme or artifice to defraud” in the mail-fraud and wire-fraud statutes, §§ 1341 and 1343 (2006 ed., Supp. II), includes “a scheme or artifice to deprive another of the intangible right of honest services” is vague, and therefore violates the Due Process Clause of the Fifth Amendment. The Court strikes a pose of judicial humility in proclaiming that our task is “not to destroy the Act . . . but to construe it,” ante, at 407, 177 L. Ed. 2d, at 659 (internal quotation marks omitted). But in transforming the prohibition of “honest-services fraud” into a prohibition of “bribery and kickbacks” it is wielding a power we long ago abjured: the power to define new federal crimes. See United States v. Hudson, 7 Cranch 32, 34, 3 L. Ed. 259 (1812).

I

A criminal statute must clearly define the conduct it proscribes, see Grayned v. City of Rockford, 408 U.S. 104, 108, 92 S. Ct. 2294, 33 L. Ed. 2d 222 (1972). A statute that is unconstitutionally vague cannot be saved by a more precise indictment, see Lanzetta v. New

[561 U.S. 416]

Jersey, 306 U.S. 451, 453, 59 S. Ct. 618, 83 L. Ed. 888 (1939), nor by judicial construction that writes in specific criteria that its text does not contain, see United States v. Reese, 92 U.S. 214, 219-221, 23 L. Ed. 563 (1876). Our cases have described *665vague statutes as failing “to provide a person of ordinary intelligence fair notice of what is prohibited, or [as being] so standardless that [they] authoriz[e] or encourag[e] seriously discriminatory enforcement.” United States v. Williams, 553 U.S. 285, 304, 128 S. Ct. 1830, 170 L. Ed. 2d 650 (2008). Here, Skilling argues that § 1346 fails to provide fair notice and encourages arbitrary enforcement because it provides no definition of the right of honest services whose deprivation it prohibits. Brief for Petitioner 38-39, 42-44. In my view Skilling is correct.

The Court maintains that “the intangible right of honest services” means the right not to have one’s fiduciaries accept “bribes or kickbacks.” Its first step in reaching that conclusion is the assertion that the phrase refers to “the doctrine developed” in cases decided by lower federal courts prior to our decision in McNally v. United States, 483 U.S. 350, 107 S. Ct. 2875, 97 L. Ed. 2d 292 (1987). Ante, at 404, 177 L. Ed. 2d, at 657. I do not contest that. I agree that Congress used the novel phrase to adopt the lower-court case law that had been disapproved by McNally— what the Court calls “the pre-McNally honest-services doctrine,” ante, at 407, 177 L. Ed. 2d, at 659. The problem is that that doctrine provides no “ascertainable standard of guilt,” United States v. L. Cohen Grocery Co., 255 U.S. 81, 89, 41 S. Ct. 298, 65 L. Ed. 516 (1921), and certainly is not limited to “bribes or kickbacks.”

Investigation into the meaning of “the pre-McNally honest-services doctrine” might logically begin with McNally itself, which rejected it. That case repudiated the many Court of Appeals holdings that had expanded the meaning of “fraud” in the mail-fraud and wire-fraud statutes beyond deceptive schemes to obtain property. 483 U.S., at 360, 107 S. Ct. 2875, 97 L. Ed. 2d 292. If the repudiated cases stood for a prohibition of “bribery and kickbacks,” one would have expected those words to appear in the opinion’s description of the cases. In fact,

[561 U.S. 417]

they do not. Not at all. Nor did McNally even provide a consistent definition of the pre-existing theory of fraud it rejected. It referred variously to a right of citizens “to have the [State]'s affairs conducted honestly,” id., at 353, 107 S. Ct. 2875, 97 L. Ed. 2d 292, to “honest and impartial government,” id., at 355, 107 S. Ct. 2875, 97 L. Ed. 2d 292, to “good government,” id., at 356, 107 S. Ct. 2875, 97 L. Ed. 2d 292, and “to have public officials perform their duties honestly,” id., at 358, 107 S. Ct. 2875, 97 L. Ed. 2d 292. It described prior case law as holding that “a public official owes a fiduciary duty to the public, and misuse of his office for private gain is a fraud,” id., at 355, 107 S. Ct. 2875, 97 L. Ed. 2d 292.

But the pre-McNally Court of Appeals opinions were not limited to fraud by public officials. Some courts had held that those fiduciaries subject to the “honest services” obligation included private individuals who merely participated in public decisions, see, e.g., United States v. Gray, 790 F.2d 1290, 1295-1296 (CA6 1986) (per curiam) (citing United States v. Margiotta, 688 F.2d 108, 122 (CA2 1982)), and even private employees who had no role in public decisions, see, e.g., United States v. Lemire, 720 F.2d 1327, 1335-1336 (CADC 1983); United States v. Von Barta, 635 F.2d 999, 1007 (CA2 1980). Moreover, “to say that a man is a fiduciary only begins [the] analysis; it gives direction to further inquiry. . . . What obligations does he *666owe as a fiduciary?” SEC v. Chenery Corp., 318 U.S. 80, 85-86, 63 S. Ct. 454, 87 L. Ed. 626 (1943). None of the “honest services” cases, neither those pertaining to public officials nor those pertaining to private employees, defined the nature and content of the fiduciary duty central to the “fraud” offense.

There was not even universal agreement concerning the source of the fiduciary obligation—whether it must be positive state or federal law, see, e.g., United States v. Rabbitt, 583 F.2d 1014, 1026 (CA8 1978), or merely general principles, such as the “obligations of loyalty and fidelity” that inhere in the “employment relationship,” Lemire, supra, at 1336. The decision McNally reversed had grounded the duty in general (not jurisdiction-specific) trust law, see Gray, supra,

[561 U.S. 418]

at 1294, a corpus juris festooned with various duties. See, e.g., Restatement (Second) of Trusts §§ 169-185 (1976). Another pre-McNally case referred to the general law of agency, United States v. Ballard, 663 F.2d 534, 543, n. 22 (CA5 1981), modified on other grounds by 680 F.2d 352 (1982), which imposes duties quite different from those of a trustee.1 See Restatement (Second) of Agency §§ 377-398 (1957).

This indeterminacy does not disappear if one assumes that the pre-McNally cases developed a federal, common-law fiduciary duty; the duty remained hopelessly undefined. Some courts described it in astoundingly broad language. Blachly v. United States, 380 F.2d 665 (CA5 1967), loftily declared that “[l]aw puts its imprimatur on the accepted moral standards and condemns conduct which fails to match the ‘reflection of moral uprightness, of fundamental honesty, fair play and right dealing in the general and business life of members of society.’” Id., at 671 (quoting Gregory v. United States, 253 F.2d 104, 109 (CA5 1958)). Other courts unhelpfully added that any scheme “contrary to public policy” was also condemned by the statute, United States v. Bohonus, 628 F.2d 1167, 1171 (CA9 1980). See also United States v. Mandel, 591 F.2d 1347, 1361 (CA4 1979) (any scheme that is “contrary to public policy and conflicts with accepted standards of moral uprightness, fundamental honesty, fair play and right dealing”). Even opinions that did not indulge in such grandiloquence did not specify the duty at issue beyond loyalty or honesty, see, e.g., Von Barta, supra, at 1005-1006. Moreover, the demands of the duty were said to be greater

[561 U.S. 419]

for public officials than for private employees, see, e.g., Lemire, supra, at 1337, n. 13; Ballard, supra, at 541, n. 17, but in what respects (or by how much) was never made clear.

The indefiniteness of the fiduciary duty is not all. Many courts held that some je-ne-sais-quoi beyond a mere breach of fiduciary duty was needed to establish honest-services fraud. See, e.g., Von Barta, supra, at 1006 (collecting cases); United States v. George, 477 F.2d 508, 512 (CA7 1973). There was, unsurprisingly, some dispute about that, at least in the context of acts by persons owing duties to *667the public. See United States v. Price, 788 F.2d 234, 237 (CA4 1986). And even among those courts that did require something additional where a public official was involved, there was disagreement as to what the addition should be. For example, in United States v. Bush, 522 F.2d 641 (1975), the Seventh Circuit held that material misrepresentations and active concealment were enough, id., at 647-648. But in Rabbitt, 583 F.2d 1014, the Eighth Circuit held that actual harm to the State was needed, id., at 1026.

Similar disagreements occurred with respect to private employees. Courts disputed whether the defendant must use his fiduciary position for his own gain. Compare Lemire, supra, at 1335 (yes), with United States v. Bronston, 658 F.2d 920, 926 (CA2 1981) (no). One opinion upheld a mail-fraud conviction on the ground that the defendant’s “failure to disclose his receipt of kickbacks and consulting fees from [his employer’s] suppliers resulted in a breach of his fiduciary duties depriving his employer of his loyal and honest services.” United States v. Bryza, 522 F.2d 414, 422 (CA7 1975). Another opinion, however, demanded more than an intentional failure to disclose: “There must be a failure to disclose something which in the knowledge or contemplation of the employee poses an independent business risk to the employer.” Lemire, supra, at 1337. Other courts required that the victim suffer some loss, see, e.g., Ballard, supra, at 541-542—a proposition that, of course, other courts

[561 U.S. 420]

rejected, see, e.g., United States v. Newman, 664 F.2d 12, 20 (CA2 1981); United States v. O’Malley, 535 F.2d 589, 592 (CA10 1976). The Court’s statement today that there was a deprivation of honest services even if “the scheme occasioned a money or property gain for the betrayed party,” ante, at 400, 177 L. Ed. 2d, at 655, is therefore true, except to the extent it is not.

In short, the first step in the Court’s analysis—holding that “the intangible right of honest services” refers to “the honest-services doctrine recognized in Courts of Appeals’ decisions before McNally,” ante, at 404, 177 L. Ed. 2d, at 657—is a step out of the frying pan into the fire. The pre-McNally cases provide no clear indication of what constitutes a denial of the right of honest services. The possibilities range from any action that is contrary to public policy or otherwise immoral, to only the disloyalty of a public official or employee to his principal, to only the secret use of a perpetrator’s position of trust in order to harm whomever he is beholden to. The duty probably did not have to be rooted in state law, but maybe it did. It might have been more demanding in the case of public officials, but perhaps not. At the time § 1346 was enacted there was no settled criterion for choosing among these options, for conclusively settling what was in and what was out.2

II

The Court is aware of all this. It knows that adopting by reference “the *668 pre-McNally honest-services doctrine,” ante, at 407, 177 L. Ed. 2d, at 659, is adopting by reference nothing more precise than

[561 U.S. 421]

the referring term itself (“the intangible right of honest services”). Hence the deus ex machina: “[W]e pare that body of precedent down to its core,” ante, at 404, 177 L. Ed. 2d, at 657. Since the honest-services doctrine “had its genesis” in bribery prosecutions, and since several cases and counsel for Skilling referred to bribery and kickback schemes as “core” or “paradigm” or “typical” examples, or “[t]he most obvious form,” of honest-services fraud, ante, at 408, 177 L. Ed. 2d, at 660 (internal quotation marks omitted), and since two cases and counsel for the Government say that they formed the “vast majority,” or “most” or at least “ [t]he bulk” of honest-services cases, ante, at 407-408, 177 L. Ed. 2d, at 659-660 (internal quotation marks omitted), THEREFORE it must be the case that they are all Congress meant by its reference to the honest-services doctrine.

Even if that conclusion followed from its premises, it would not suffice to eliminate the vagueness of the statute. It would solve (perhaps) the indeterminacy of what acts constitute a breach of the “honest services” obligation under the pre-McNally law. But it would not solve the most fundamental indeterminacy: the character of the “fiduciary capacity” to which the bribery and kickback restriction applies. Does it apply only to public officials? Or in addition to private individuals who contract with the public? Or to everyone, including the corporate officer here? The pre-McNally case law does not provide an answer. Thus, even with the bribery and kickback limitation the statute does not answer the question, “What is the criterion of guilt?”

But that is perhaps beside the point, because it is obvious that mere prohibition of bribery and kickbacks was not the intent of the statute. To say that bribery and kickbacks represented “the core” of the doctrine, or that most cases applying the doctrine involved those offenses, is not to say that they are the doctrine. All it proves is that the multifarious versions of the doctrine overlap with regard to those offenses. But the doctrine itself is much more. Among all the pre-McNally smorgasbord offerings of varieties of

[561 U.S. 422]

honest-services fraud, not one is limited to bribery and kickbacks. That is a dish the Court has cooked up all on its own.

Thus, the Court’s claim to “respec[t] the legislature,” ante, at 409, n. 43, 177 L. Ed. 2d, at 660-661 (emphasis deleted), is false. It is entirely clear (as the Court and I agree) that Congress meant to reinstate the body of pre-McNally honest-services law; and entirely clear that that prohibited much more (though precisely what more is uncertain) than bribery and kickbacks. Perhaps it is true that “Congress intended § 1346 to reach at least bribes and kickbacks,” ante, at 408, 177 L. Ed. 2d, at 660. That simply does not mean, as the Court now holds, that “§ 1346 criminalizes only” bribery and kickbacks, ante, at 409, 177 L. Ed. 2d, at 660.

Arriving at that conclusion requires not interpretation but invention. The Court replaces a vague criminal standard that Congress adopted with a more narrow one (included within the vague one) that can pass constitutional muster. I know of no precedent *669for such “paring down,”3 and it seems to me clearly beyond judicial power. This is not, as the Court claims, ante, at 406, 177 L. Ed. 2d, at 659, simply a matter of adopting a “limiting construction” in the face of potential unconstitutionality.

[561 U.S. 423]

To do that, our cases have been careful to note, the narrowing construction must be “fairly possible,” Boos v. Barry, 485 U.S. 312, 331, 108 S. Ct. 1157, 99 L. Ed. 2d 333 (1988), “reasonable,” Hooper v. California, 155 U.S. 648, 657, 15 S. Ct. 207, 39 L. Ed. 297 (1895), or not “plainly contrary to the intent of Congress,” Edward J. DeBartolo Corp. v. Florida Gulf Coast Building & Constr. Trades Council, 485 U.S. 568, 575, 108 S. Ct. 1392, 99 L. Ed. 2d 645 (1988). As we have seen (and the Court does not contest), no court before McNally concluded that the “deprivation of honest services” meant only the acceptance of bribes or kickbacks. If it were a “fairly possible” or “reasonable” construction, not “contrary to the intent of Congress,” one would think that some court would have adopted it. The Court does not even point to a post-McNally case that reads § 1346 to cover only bribery and kickbacks, and I am aware of none.

The canon of constitutional avoidance, on which the Court so heavily relies, see ante, at 405-406, 177 L. Ed. 2d, at 658-659, states that “when the constitutionality of a statute is assailed, if the statute be reasonably susceptible of two interpretations, by one of which it would be unconstitutional and by the other valid, it is our plain duty to adopt that construction which will save the statute from constitutional infirmity.” United States ex rel. Attorney General v. Delaware & Hudson Co., 213 U.S. 366, 407, 29 S. Ct. 527, 53 L. Ed. 836 (1909); see also United States v. Rumely, 345 U.S. 41, 45, 73 S. Ct. 543, 97 L. Ed. 770 (1953) (describing the canon as decisive “in the choice of fair alternatives”). Here there is no choice to be made between two “fair alternatives.” Until today, no one has thought (and there is no basis for thinking) that the honest-services statute prohibited only bribery and kickbacks.

I certainly agree with the Court that we must, “if we can,” uphold, rather than “condemn,” Congress’s enactments, ante, at 403, 177 L. Ed. 2d, at 656-657. But I do not believe we have the power, in order to uphold an enactment, to rewrite it. Congress enacted the entirety of the pre-McNally honest-services law, the content of which is (to put it mildly) unclear. In prior vagueness cases, we have resisted the temptation to make all things

[561 U.S. 424]

right with the stroke of our *670pen. See, e.g., Smith v. Goguen, 415 U.S. 566, 575, 94 S. Ct. 1242, 39 L. Ed. 2d 605 (1974). I would show the same restraint today, and reverse Skilling’s conviction on the basis that § 1346 provides no “ascertainable standard” for the conduct it condemns, L. Cohen, 255 U.S., at 89, 41 S. Ct. 298, 65 l. Ed. 2d 516. Instead, the Court today adds to our functions the prescription of criminal law.

Ill

A brief word about the appropriate remedy. As I noted supra, at 416, 177 L. Ed. 2d, at 665, Skilling has argued that § 1346 cannot be constitutionally applied to him because it affords no definition of the right whose deprivation it prohibits. Though this reasoning is categorical, it does not make Skilling’s challenge a “facial” one, in the sense that it seeks invalidation of the statute in all its applications, as opposed to preventing its enforcement against him. I continue to doubt whether “striking down” a statute is ever an appropriate exercise of our Article III power. See Chicago v. Morales, 527 U.S. 41, 77, 119 S. Ct. 1849, 144 L. Ed. 2d 67 (1999) (Scalia, J., dissenting). In the present case, the universality of the infirmity Skilling identifies in § 1346 may mean that if he wins, anyone else prosecuted under the statute will win as well, see Smith, supra, at 576-578, 94 S. Ct. 1242, 39 L. Ed. 2d 605. But Skilling only asks that his conviction be reversed, Brief for Petitioner 57-58, so the remedy he seeks is not facial invalidation.

I would therefore reverse Skilling’s conviction under § 1346 on the ground that it fails to define the conduct it prohibits. The fate of the statute in future prosecutions—obvious from my reasoning in the case—would be a matter for stare decisis.

It is hard to imagine a case that more clearly fits the description of what Chief Justice Waite said could not be done, in a colorful passage oft-cited in our vagueness opinions, United States v. Reese, 92 U.S., at 221, 23 L. Ed. 563:

[561 U.S. 425]

“The question, then, to be determined, is, whether we can introduce words of limitation into a penal statute so as to make it specific, when, as expressed, it is general only.

“It would certainly be dangerous if the legislature could set a net large enough to catch all possible offenders, and leave it to the courts to step inside and say who could be rightfully detained, and who should be set at large. This would, to some extent, substitute the judicial for the legislative department of the government. . . .

“To limit this statute in the manner now asked for would be to make a new law, not to enforce an old one. This is no part of our duty.”

Justice Alito,

concurring in part and concurring in the judgment.

I join the judgment of the Court and all but Part II of the Court’s opinion. I write separately to address petitioner’s jury-trial argument.

The Sixth Amendment guarantees criminal defendants a trial before “an impartial jury.” In my view, this requirement is satisfied so long as no biased juror is actually seated at trial. Of course, evidence of pretrial media attention and widespread community hostility may play a role in the bias *671inquiry. Such evidence may be important in assessing the adequacy of voir dire, see, e.g., Mu’Min v. Virginia, 500 U.S. 415, 428-432, 111 S. Ct. 1899, 114 L. Ed. 2d 493 (1991), or in reviewing the denial of requests to dismiss particular jurors for cause, see, e.g., Patton v. Yount, 467 U.S. 1025, 1036-1040, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (1984). There are occasions in which such evidence weighs heavily in favor of a change of venue. In the end, however, if no biased juror is actually seated, there is no violation of the defendant’s right to an impartial jury. See id., at 1031-1035, 1040, 104 S. Ct. 2885, 81 L. Ed 2d 847; Murphy v. Florida, 421 U.S. 794, 800-801, 803, 95 S. Ct. 2031, 44 L. Ed. 2d 589 (1975); see also Rivera v.

[561 U.S. 426]

Illinois, 556 U.S. 148, 157-159, 129 S. Ct. 1446, 173 L. Ed. 2d 320 (2009); United States v. Martinez-Salazar, 528 U.S. 304, 311, 316-317, 120 S. Ct. 774, 145 L. Ed. 2d 792 (2000); Smith v. Phillips, 455 U.S. 209, 215-218, 102 S. Ct. 940, 71 L. Ed. 2d 78 (1982).

Petitioner advances a very different understanding of the jury-trial right. Where there is extraordinary pretrial publicity and community hostility, he contends, a court must presume juror prejudice and thus grant a change of venue. Brief for Petitioner 25-34. I disagree. Careful voir dire can often ensure the selection of impartial jurors even where pretrial media coverage has generated much hostile community sentiment. Moreover, once a jury has been selected, there are measures that a trial judge may take to insulate jurors from media coverage during the course of the trial. What the Sixth Amendment requires is “an impartial jury.” If the jury that sits and returns a verdict is impartial, a defendant has received what the Sixth Amendment requires.

The rule that petitioner advances departs from the text of the Sixth Amendment and is difficult to apply. It requires a trial judge to determine whether the adverse pretrial media coverage and community hostility in a particular case have reached a certain level of severity, but there is no clear way of demarcating that level or of determining whether it has been met.

Petitioner relies chiefly on three cases from the 1960’s—Sheppard v. Maxwell, 384 U.S. 333, 86 S. Ct. 1507, 16 L. Ed. 2d 600 (1966), Estes v. Texas, 381 U.S. 532, 85 S. Ct. 1628, 14 L. Ed. 2d 543 (1965), and Rideau v. Louisiana, 373 U.S. 723, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1963). I do not read those cases as demanding petitioner’s suggested approach. As the Court notes, Sheppard and Estes primarily “involved media interference with courtroom proceedings during trial.” Ante, at 382, n. 14, 177 L. Ed. 2d, at 643; see also post, at 446, 177 L. Ed. 2d, at 683 (Sotomayor, J., concurring in part and dissenting in part). Rideau involved unique events in a small community.

I share some of Justice Sotomayor’s concerns about the adequacy of the voir dire in this case and the trial judge’s findings that certain jurors could be impartial. See post, at

[561 U.S. 427]

458-462, 177 L. Ed. 2d, at 691-693. But those highly fact-specific issues are not within the question presented. Pet. for Cert. i. I also do not understand the opinion of the Court as reaching any question regarding a change of venue under Federal Rule of Criminal Procedure 21.

Because petitioner, in my view, is not entitled to a reversal of the decision below on the jury-trial question that is before us, I join the judgment of the Court in full.

*672Justice Sotomayor,

with whom Justice Stevens and Justice Breyer join, concurring in part and dissenting in part.

I concur in the Court’s resolution of the honest-services fraud question and join Part III of its opinion. I respectfully dissent, however, from the Court’s conclusion that Jeffrey Skilling received a fair trial before an impartial jury. Under our relevant precedents, the more intense the public’s antipathy toward a defendant, the more careful a court must be to prevent that sentiment from tainting the jury. In this case, passions ran extremely high. The sudden collapse of Enron directly affected thousands of people in the Houston area and shocked the entire community. The accompanying barrage of local media coverage was massive in volume and often caustic in tone. As Enron’s onetime chief executive officer (CEO), Skilling was at the center of the storm. Even if these extraordinary circumstances did not constitutionally compel a change of venue, they required the District Court to conduct a thorough voir dire in which prospective jurors’ attitudes about the case were closely scrutinized. The District Court’s inquiry lacked the necessary thoroughness and left serious doubts about whether the jury empaneled to decide Skilling’s case was capable of rendering an impartial decision based solely on the evidence presented in the courtroom. Accordingly, I would grant Skilling relief on his fair-trial claim.

[561 U.S. 428]

I

The majority understates the breadth and depth of community hostility toward Skilling and overlooks significant deficiencies in the District Court’s jury selection process. The failure of Enron wounded Houston deeply. Virtually overnight, what had been the city’s “largest, most visible, and most prosperous company,” its “foremost social and charitable force,” and “a source of civic pride” was reduced to a “shattered shell.” App. ¶¶11, 13, pp. 649a-650a, 1152a. Thousands of the company’s employees lost their jobs and saw their retirement savings vanish. As the effects rippled through the local economy, thousands of additional jobs disappeared, businesses shuttered, and community groups that once benefited from Enron’s largesse felt the loss of millions of dollars in contributions. See, e.g., 3 Supp. Record 1229, 1267; see also 554 F.3d 529, 560 (CA5 2009) (“Accounting firms that serviced Enron’s books had less work, hotels had more open rooms, restaurants sold fewer meals, and so on”). Enron’s community ties were so extensive that the entire local U. S. Attorney’s Office was forced to recuse itself from the Government’s investigation into the company’s fall. See 3 Supp. Record 608 (official press release).

With Enron’s demise affecting the lives of so many Houstonians, local media coverage of the story saturated the community. According to a defense media expert, the Houston Chronicle—the area’s leading newspaper—assigned as many as 12 reporters to work on the Enron story full time. App. 568a-569a. The paper mentioned Enron in more than 4,000 articles during the 3-year period following the company’s December 2001 bankruptcy filing. Hundreds of these articles discussed Skilling by name. See 3 Supp. Record 2114. Skilling’s expert, a professional journalist and academic with 30 years’ experience, could not “recall another instance where a local paper dedicated as *673many resources to a single topic over such an extended period of time as the Houston Chronicle . . . dedicated to Enron.” App. ¶32, at 570a.

[561 U.S. 429]

Local television news coverage was similarly pervasive and, in terms of “editorial theme,” “largely followed the Chronicle’s lead.” Id., ¶ 11, at 559a; see also id.., at 717a. Between May 2002 and October 2004, local stations aired an estimated 19,000 news segments involving Enron, more than 1,600 of which mentioned Skilling. 3 Supp. Record 2116.

While many of the stories were straightforward news items, many others conveyed and amplified the community’s outrage at the top executives perceived to be responsible for the company’s bankruptcy. A Chronicle report on Skilling’s 2002 testimony before Congress is typical of the coverage. It began, “Across Houston, Enron employees watched former chief executive Jeffrey Skill-ing’s congressional testimony on television, turning incredulous, angry and then sarcastic by turns, as a man they knew as savvy and detail-oriented pleaded memory failure and ignorance about critical financial transactions at the now-collapsed energy giant.” App. 1218a. “ ‘He is lying; he knew everything,’ said [an employee], who said she had seen Skill-ing frequently over her 18 years with the firm, where Skilling was known for his intimate grasp of the inner doings at the company. ‘I am getting sicker by the minute.’ ” Id., at 1219a. A companion piece quoted a local attorney who called Skilling an “idiot” who was “in denial”; he added, “I’m glad [Skilling’s] not my client.” Id., at 592a-593a (internal quotation marks omitted).

Articles deriding Enron’s senior executives were juxtaposed with pieces expressing sympathy toward and solidarity with the company’s many victims. Skilling’s media expert counted nearly a hundred victim-related stories in the Chronicle, including a “multi-page layout entitled ‘The Faces of Enron,’ ” which poignantly described the gut-wrenching experiences of former employees who lost vast sums of money, faced eviction from their homes, could not afford Christmas gifts for their children, and felt “scared,” “hurt,” “humiliat[ed],” “helpless,” and “betrayed.” Id., ¶71, at

[561 U.S. 430]

585a-586a. The conventional wisdom that blame for Enron’s devastating implosion and the ensuing human tragedy ultimately rested with Skill-ing and former Enron Chairman Kenneth Lay became so deeply ingrained in the popular imagination that references to their involvement even turned up on the sports pages: “If you believe the story about [Coach Bill Parcells] not having anything to do with the end of Emmitt Smith’s Cowboys career, then you probably believe in other far-fetched concepts. Like Jeff Skilling having nothing to do with Enron’s collapse.” 3 Supp. Record 811.

When a federal grand jury indicted Skilling, Lay, and Richard Causey— Enron’s former chief accounting officer—in 2004 on charges of conspiracy to defraud, securities fraud, and other crimes, the media placed them directly in their crosshairs. In the words of one article, “there was one thing those whose lives were touched by the once-exalted company all seemed to agree upon: The indictment of former Enron CEO Jeff Skilling was overdue.” App. 1393a. Scoffing at Skill-ing’s attempts to paint himself as “a ‘victim’ of his subordinates,” id., at 1394a, the Chronicle derided “the doo-fus defense” that Lay and Skilling *674were expected to offer, id., at 1401a.1 The Chronicle referred to the coming Skilling/Lay trial as “the main event” and “The Big One,” which would

[561 U.S. 431]

finally bring “the true measure of justice in the Enron saga.” Record 40002; App. 1457a, 1460a.2 On the day the superseding indictment charging Lay was issued, “the Chronicle dedicated three-quarters of its front page, 2 other full pages, and substantial portions of 4 other pages, all in the front or business sections, to th[e] story.” Id., ¶57, at 580a-581a.

Citing the widely felt sense of vic-timhood among Houstonians and the voluminous adverse publicity, Skill-ing moved in November 2004 for a change of venue.3 The District Court denied the motion, characterizing the media coverage as largely “objective and unemotional.” App. to Brief for United States 11a. Voir dire, it concluded, would provide an effective means to “ferret out any bias” in the jury pool. Id., at 18a; see ante, at 370, 177 L. Ed. 2d, at 636.

To that end, the District Court began the jury selection process by mailing screening questionnaires to 400 prospective jurors in November 2005. The completed questionnaires of the 283 respondents not excused for hardship dramatically illustrated the widespread impact of Enron’s collapse on the Houston community and confirmed the intense animosity of Hous-tonians toward Skilling and his codefendants. More than one-third of the prospective jurors (approximately 99 of 283, by my count) indicated that they

[561 U.S. 432]

or persons they knew had lost money or jobs as a result of the Enron bankruptcy. Two-thirds of the jurors (about 188 of 283) expressed views about Enron or the defendants that suggested a potential predisposition to convict. In many instances, they did not mince words, describing Skilling as “smug,” “arrogant,” “brash,” “conceited,” “greedy,” “deceitful,” “totally unethical and criminal,” “a crook,” “the biggest liar on the face of the earth,” and “guilty as sin” (capitalization omitted).4 Only about 5 *675percent of the prospective jurors (15 of 283) did not read the Houston Chronicle, had not otherwise “heard or read about any of the Enron cases,” Record 13019, were not connected to Enron victims, and gave no answers suggesting possible antipathy toward the defendants.5 The parties jointly stipulated to the dismissal

[561 U.S. 433]

Of 119 members of the jury pool for cause, hardship, or disability, but numerous individuals who had made harsh comments about Skilling remained.6

On December 28, 2005, shortly after the questionnaires had been returned, Causey pleaded guilty. The plea was covered in lead newspaper and television stories. A front-page headline in the Chronicle proclaimed that “Causey’s plea wreaks havoc for Lay, Skilling.” Record 12049, n. 13; see also ibid, (quoting a former U. S. attorney who described the plea as “a serious blow to the defense”). A Chronicle editorial opined that “Causey’s admission of securities fraud . . . makes less plausible Lay’s claim that most of the guilty

[561 U.S. 434]

pleas were the result of prosecutorial pressure rather than actual wrongdoing.” Id., at 12391.

With the trial date quickly approaching, Skilling renewed his change-of-venue motion, arguing that *676both the questionnaire responses and the Causey guilty plea confirmed that he could not receive a fair trial in Houston. In the alternative, Skilling asserted that “defendants are entitled to a more thorough jury selection process than currently envisioned by the [c]ourt.” Id., at 12067. The court had announced its intention to question individual jurors at the bench with one attorney for each side present, and to complete the voir dire in a single day. See, e.g., id., at 11804-11805, 11808. Skilling proposed, inter alia, that defense counsel be afforded a greater role in questioning, id.., at 12074; that jurors be questioned privately in camera or in a closed courtroom where it would be easier for counsel to consult with their colleagues, clients, and jury consultants, id., at 12070-12072; and that the court “avoid leading questions,” which “tend to [e] licit affirmative responses from prospective jurors that may not reflect their actual views,” id., at 12072. At a minimum, Skilling asserted, the court should grant a continuance of at least 30 days and send a revised questionnaire to a new group of prospective jurors. Id., at 12074-12075.

The District Court denied Skilling’s motion without a hearing, stating in a brief order that it was “not persuaded that the evidence or arguments urged by defendants . . . establish that pretrial publicity and/or community prejudice raise a presumption of inherent jury prejudice.” Id., at 14115. According to the court, the “jury questionnaires sent to the remaining members of the jury panel and the court’s voir dire examination of the jury panel provide adequate safeguards to defendants and will result in the selection of a fair and impartial jury in this case.” Id., at 14115-14116. The court did agree to delay the trial by two weeks, until January 30, 2006.

[561 U.S. 435]

The coming trial featured prominently in local news outlets. A front-page, eve-of-trial story in the Chronicle described “the hurt and anger and resentment” that had been “churn [ing] inside” Houstonians since Enron’s collapse. Id., at 39946. Again criticizing Lay and Skilling for offering a “doofus defense” (“a plea of not guilty by reason of empty-headedness”), the paper stated that “Lay and Skilling took hundreds of millions in compensation yet now fail to accept the responsibility that went with it.” Ibid. The article allowed that the defendants’ guilt, “though perhaps widely assumed, remains even now an assertion. Ajury now takes up the task of deciding whether that assertion is valid.” Id., at 39947. The next paragraph, however, assured readers that “it’s normal for your skin to crawl when Lay or Skilling claim with doe-eyed innocence that they were unaware that something was amiss at Enron. The company’s utter failure belies the claim.” Ibid, (one paragraph break omitted); see also id., at 39904 (declaring that Lay and Skilling would “have to offer a convincing explanation for how executives once touted as corporate geniuses could be so much in the dark about the illegal activities and deceptive finances of their own company”).

It is against this backdrop of widespread community impact and pervasive pretrial publicity that jury selection in Skilling’s case unfolded. Approximately 160 prospective jurors appeared for voir dire at a federal courthouse located “about six blocks from Enron’s former headquarters.” 554 F.3d, at 561. Addressing them as a group, the District Court began by briefly describing the case and provid*677ing a standard admonition about the need to be fair and impartial and to decide the case based solely on the trial evidence and jury instructions. The court then asked whether anyone had “any reservations about your ability to conscientiously and fairly follow these very important rules.” App. 815a. Two individuals raised their hands and were called forward

[561 U.S. 436]

to the bench. One told the court that he thought Lay and Skilling “knew exactly what they were doing” and would have to prove their innocence. Id., at 818a-819a. The second juror, who had stated on his written questionnaire that he held no opinion that would preclude him from being impartial, declared that he “would dearly love to sit on this jury. I would love to claim responsibility, at least 1/12 of the responsibility, for putting these sons of bitches away for the rest of their lives.” Id., at 819a-820a. The court excused both jurors for cause.

The court proceeded to question individual jurors from the bench. As the majority recounts, ante, at 373-374, 177 L. Ed. 2d, at 638-639, the court asked them a few general yes/no questions about their exposure to Enron-related news, often variations of, “Do you recall any particular articles that stand out that you’ve read about the case?” App. 850a. The court also asked about questionnaire answers that suggested bias, focusing mainly on whether, notwithstanding seemingly partial comments, the prospective jurors believed they “could be fair” and “put the government to its proof.” Id., at 852a. Counsel were permitted to follow up on issues raised by the court. The court made clear, however, that its patience would be limited, see, e.g., id., at 879a, and questioning tended to be brief—generally less than five minutes per person. Even so, it exposed disqualifying biases among several prospective jurors who had earlier expressed no concerns about their ability to be fair.7

[561 U.S. 437]

Once it identified 38 qualified prospective jurors, the court allowed the defense and Government to exercise their allotted peremptory challenges. This left 12 jurors and 4 alternates, who were sworn in and instructed, for the first time, “not [to] read anything dealing with this case or listen to any discussion of the case on radio or television or access any Internet sites that may deal with the case” and to “inform your friends and family members that they should not discuss with you anything they may have read or heard about this case.” Id., at 1026a. Start to finish, the selection process took about five hours.

Skilling’s trial commenced the next day and lasted four months. After several days of deliberations, the jury found Skilling guilty of conspiracy, 12 counts of securities fraud, 5 counts of making false representations to auditors, and 1 count of insider trading; it acquitted on 9 insider trading counts. The jury found Lay guilty on all counts.

On appeal, Skilling asserted that he had been denied his constitutional right to a fair trial before an impartial jury. Addressing this claim, the Court *678of Appeals began by disavowing the District Court’s findings concerning “community hostility.” There was, the court concluded, “sufficient inflammatory pretrial material to require a finding of presumed prejudice, especially in light of the immense volume of coverage.” 554 F.3d, at 559. “[Prejudice was [also] inherent in an alleged co-conspirator’s well-publicized decision to plead guilty on the eve of trial.” Ibid. The Court of Appeals, moreover, faulted the District Court for failing to “consider the wider context.” Id., at 560. “[I]t was not enough for the court merely to assess the tone of the news reporting. The evaluation of the volume and nature of reporting is merely a proxy for the real inquiry: whether there could be a fair trial by an impartial jury that was not influenced by outside, irrelevant sources.” Ibid, (internal quotation marks and footnote omitted). According to the Court of Appeals, “[t]he district court seemed to overlook that the

[561 U.S. 438]

prejudice came from more than just pretrial media publicity, but also from the sheer number of victims.” Ibid.

Having determined that “Skilling was entitled to a presumption of prejudice,” the Court of Appeals proceeded to explain that “the presumption is rebuttable, . . . and the government may demonstrate from the voir dire that an impartial jury was actually impanelled.” Id., at 561 (internal quotation marks omitted). Describing the voir dire as “exemplary,” “searching,” and “proper and thorough,” id., at 562, the court concluded that “[t]he government [had] met its burden of showing that the actual jury that convicted Skilling was impartial,” id., at 564-565. On this basis, the Court of Appeals rejected Skilling’s claim and affirmed his convictions.

II

The Sixth Amendment right to an impartial jury and the due process right to a fundamentally fair trial guarantee to criminal defendants a trial in which jurors set aside preconceptions, disregard extrajudicial influences, and decide guilt or innocence “based on the evidence presented in court.” Irvin v. Dowd, 366 U.S. 717, 723, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (1961); see also Sheppard v. Maxwell, 384 U.S. 333, 362, 86 S. Ct. 1507, 16 L. Ed. 2d 600 (1966). Community passions, often inflamed by adverse pretrial publicity, can call the integrity of a trial into doubt. In some instances, this Court has observed, the hostility of the community becomes so severe as to give rise to a “presumption of [juror] prejudice.” Patton v. Yount, 467 U.S. 1025, 1031, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (1984).

The Court of Appeals incorporated the concept of presumptive prejudice into a burden-shifting framework: Once the defendant musters sufficient evidence of community hostility, the onus shifts to the Government to prove the impartiality of the jury. The majority similarly envisions a fixed point at which public passions become so intense that prejudice to a defendant’s fair-trial rights must be presumed. The majority declines, however, to decide whether the presumption is rebuttable, as the Court of Appeals held.

[561 U.S. 439]

This Court has never treated the notion of presumptive prejudice so formalistically. Our decisions instead merely convey the commonsense understanding that as the tide of public enmity rises, so too does the danger that the prejudices of the community will infiltrate the jury. The underlying *679question has always been this: Do we have confidence that the jury’s verdict was “induced only by evidence and argument in open court, and not by any outside influence, whether of private talk or public print”? Patterson v. Colorado ex rel. Attorney General of Colo., 205 U.S. 454, 462, 27 S. Ct. 556, 51 L. Ed. 879 (1907).

The inquiry is necessarily case specific. In selecting a jury, a trial court must take measures adapted to the intensity, pervasiveness, and character of the pretrial publicity and community animus. Reviewing courts, meanwhile, must assess whether the trial court’s procedures sufficed under the circumstances to keep the jury free from disqualifying bias. Cf. Murphy v. Florida, 421 U.S. 794, 799, 95 S. Ct. 2031, 44 L. Ed. 2d 589 (1975) (scrutinizing the record for “any indications in the totality of circumstances that petitioner’s trial was not fundamentally fair”). This Court’s precedents illustrate the sort of steps required in different situations to safeguard a defendant’s constitutional right to a fair trial before an impartial jury.

At one end of the spectrum, this Court has, on rare occasion, confronted such inherently prejudicial circumstances that it has reversed a defendant’s conviction “without pausing to examine . . . the voir dire examination of the members of the jury.” Rideau v. Louisiana, 373 U.S. 723, 727, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (1963). In Rideau, repeated television broadcasts of the defendant’s confession to murder, robbery, and kidnaping so thoroughly poisoned local sentiment as to raise doubts that even the most careful voir dire could have secured an impartial jury. A change of venue, the Court determined, was thus the only way to ensure a fair trial. Ibid.; see also 6 W. LaFave, J. Israel, N. King, & O. Kerr, Criminal Procedure § 23.2(a), p. 264 (3d ed. 2007) (hereinafter LaFave) (“The best reading

[561 U.S. 440]

of Rideau is that the Court there recognized that prejudicial publicity may be so inflammatory and so pervasive that the voir dire simply cannot be trusted to fully reveal the likely prejudice among prospective jurors”).

As the majority describes, ante, at 379-380, 177 L. Ed. 2d, at 642, this Court reached similar conclusions in Estes v. Texas, 381 U.S. 532, 85 S. Ct. 1628, 14 L. Ed. 2d 543 (1965), and Sheppard, 384 U.S. 333, 86 S. Ct. 1507, 16 L. Ed. 2d 600. These cases involved not only massive pretrial publicity but also media disruption of the trial process itself. Rejecting the argument that the defendants were not entitled to relief from their convictions because they “ha[d] established no isolatable prejudice,” the Court described the “untoward circumstances” as “inherently suspect.” Estes, 381 U.S., at 542, 544, 85 S. Ct. 1628, 14 L. Ed. 2d 543. It would have been difficult for the jurors not to have been swayed, at least subconsciously, by the “bedlam” that surrounded them. Sheppard, 384 U.S., at 355, 86 S. Ct. 1507, 16 L. Ed. 2d 600. Criticizing the trial courts’ failures “to protect the jury from outside influence,” id., at 358, 86 S. Ct. 1507, 16 L. Ed. 2d 600, the Court stressed that, “where there is a reasonable likelihood that prejudicial news prior to trial will prevent a fair trial, the judge should continue the case until the threat abates, or transfer it to another [venue] not so permeated with publicity.” Id., at 363, 86 S. Ct. 1507, 16 L. Ed. 2d 600. Estes and Sheppard thus applied Rideau’s insight that in particularly extreme circumstances even the most rigorous *680 voir dire cannot suffice to dispel the reasonable likelihood of jury bias.

Apart from these exceptional cases, this Court has declined to discount voir dire entirely and has instead examined the particulars of the jury selection process to determine whether it sufficed to produce a jury untainted by pretrial publicity and community animus. The Court has recognized that when antipathy toward a defendant pervades the community there is a high risk that biased jurors will find their way onto the panel. The danger is not merely that some prospective jurors will deliberately hide their prejudices, but also that, as “part of a community deeply hostile to the accused,” “they may unwittingly [be] influenced”

[561 U.S. 441]

by the fervor that surrounds them. Murphy, 421 U.S., at 803, 95 S. Ct. 2031, 44 L. Ed. 2d 589. To ensure an impartial jury in such adverse circumstances, a trial court must carefully consider the knowledge and attitudes of prospective jurors and then closely scrutinize the reliability of their assurances of fairness. Cf. Morgan v. Illinois, 504 U.S. 719, 729, 112 S. Ct. 2222, 119 L. Ed. 2d 492 (1992) (“[P]art of the guarantee of a defendant’s right to an impartial jury is an adequate voir dire to identify unqualified jurors”).

Irvin offers an example of a case in which the trial court’s voir dire did not suffice to counter the “wave of public passion” that had swept the community prior to the defendant’s trial. 366 U.S., at 728, 81 S. Ct. 1639, 6 L. Ed. 2d 751. The local news media had “extensively covered” the crimes (a murder spree), “arous[ing] great excitement and indignation.” Id., at 719, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (internal quotation marks omitted). Following Irvin’s arrest, the press “blanketed” the community with “a barrage of newspaper headlines, articles, cartoons and pictures” communicating numerous unfavorable details about Irvin, including that he had purportedly confessed. Id., at 725, 81 S. Ct. 1639, 6 L. Ed. 2d 751. Nearly 90 percent of the 430 prospective jurors examined during the trial court’s voir dire “entertained some opinion as to guilt—ranging in intensity from mere suspicion to absolute certainty.” Id., at 727, 81 S. Ct. 1639, 6 L. Ed. 2d 751. Of the 12 jurors seated, 8 “thought petitioner was guilty,” although “each indicated that notwithstanding his opinion he could render an impartial verdict.” Id., at 727, 724, 81 S. Ct. 1639, 6 L. Ed. 2d 751.

Despite the seated jurors’ assurances of impartiality, this Court invalidated Irvin’s conviction for want of due process. “It is not required,” this Court declared, “that the jurors be totally ignorant of the facts and issues involved. ... It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court.” Id., at 722-723, 81 S. Ct. 1639, 6 L. Ed. 2d 751. The Court emphasized, however, that a juror’s word on this matter is not decisive, particularly when “the build-up of prejudice [in the community] is clear

[561 U.S. 442]

and convincing.” Id., at 725, 81 S. Ct. 1639, 6 L. Ed. 2d 751. Many of Irvin’s jurors, the Court noted, had been influenced by “the pattern of deep and bitter prejudice shown to be present throughout the community.” Id., at 727, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (internal quotation marks omitted). The Court did not “doubt [that] each juror was sincere when he said that he would be fair and impartial to [Irvin], but . . . [w]here so many, so many times, ad*681mitted prejudice, such a statement of impartiality can be given little weight.” Id., at 728, 81 S. Ct. 1639, 6 L. Ed. 2d 751.

The media coverage and community animosity in Irvin were particularly intense. In three subsequent cases, this Court recognized that high-profile cases may generate substantial publicity without stirring similar public passions. The jury selection process in such cases, the Court clarified, generally need not be as exhaustive as in a case such as Irvin. So long as the trial court conducts a reasonable inquiry into extrajudicial influences and the ability of prospective jurors to presume innocence and render a verdict based solely on the trial evidence, we would generally have no reason to doubt the jury’s impartiality.8

The first of these cases, Murphy, 421 U.S. 794, 95 S. Ct. 2031, 44 L. Ed. 2d 589, involved a well-known defendant put on trial for a widely publicized Miami Beach robbery. The state trial court denied his motion for a change of venue and during voir dire excused 20 of the 78 prospective jurors for cause. Distinguishing Irvin, this Court saw no indication in the voir dire of “such hostility to [Murphy] by the jurors who served in his trial as to suggest a partiality that could not be laid aside.” 421 U.S., at 800, 95 S. Ct. 2031, 44 L. Ed. 2d 589. Although some jurors “had a vague recollection of the robbery with which [Murphy] was charged and each had

[561 U.S. 443]

some knowledge of [his] past crimes,” “none betrayed any belief in the relevance of [Murphy’s] past to the present case.” Ibid.; see also ibid., n. 4 (contrasting a juror’s “mere familiarity with [a defendant] or his past” with “an actual predisposition against him”). “[T]hese indicia of impartiality,” the Court suggested, “might be disregarded in a case where the general atmosphere in the community or courtroom is sufficiently inflammatory, but the circumstances surrounding [Murphy’s] trial [were] not at all of that variety.” Id., at 802, 95 S. Ct. 2031, 44 L. Ed. 2d 589.

In a second case, Yount, 467 U.S. 1025, 104 S. Ct. 2885, 81 L. Ed. 2d 847, the defendant was granted a new trial four years after being convicted of murder. He requested a change of venue, citing pretrial publicity and the widespread local knowledge that he had previously been convicted and had made confessions that would be inadmissible in court. The state trial court denied Yount’s motion and seated a jury following a 10-day voir dire of 292 prospective jurors. Nearly all of the prospective jurors had heard of the case, and 77 percent “admitted they would carry an opinion into the jury box.” Id., at 1029, 104 S. Ct. 2885, 81 L. Ed. 2d 847. Declining to grant relief on federal habeas review, this Court stressed the significant interval between Yount’s first trial—when “adverse publicity and the community’s sense of outrage were at their height”—and his second trial, which “did not occur until four years later, at a time when prejudicial publicity was greatly diminished and community sentiment had softened.” Id., at 1032, 104 S. Ct. 2885, 81 L. Ed. 2d 847. While 8 of the 14 seated jurors and *682alternates had “at some time . . . formed an opinion as to Yount’s guilt,” the “particularly extensive” voir dire confirmed that “time had weakened or eliminated any” bias they once may have harbored. Id., at 1029-1030, 1034, n. 10, 1033, 104 S. Ct. 2885, 81 L. Ed. 2d 847. Accordingly, this Court concluded, “the trial court did not commit manifest error in finding that the jury as a whole was impartial.” Id., at 1032, 104 S. Ct. 2885, 81 L. Ed. 2d 847.

This Court most recently wrestled with the issue of pretrial publicity in Mu’Min v. Virginia, 500 U.S. 415, 111 S. Ct. 1899, 114 L. Ed. 2d 493 (1991).

[561 U.S. 444]

Mu’Min stood accused of murdering a woman while out of prison on a work detail. Citing 47 newspaper articles about the crime, Mu’Min moved for a change of venue. The state trial court deferred its ruling and attempted to seat a jury. During group questioning, 16 of the 26 prospective jurors indicated that they had heard about the case from media or other sources. Dividing these prospective jurors into panels of four, the court asked further general questions about their ability to be fair given what they had heard or read. One juror answered equivocally and was dismissed for cause. The court refused Mu’Min’s request to ask more specific questions “relating to the content of news items that potential jurors might have read or seen.” Id., at 419, 111 S. Ct. 1899, 114 L. Ed. 2d 493. Of the 12 persons who served on the jury, “8 had at one time or another read or heard something about the case. None had indicated that he had formed an opinion about the case or would be biased in any way.” Id., at 421, 111 S. Ct. 1899, 114 L. Ed. 2d 493.

Rejecting Mu’Min’s attempt to analogize his case to Irvin, this Court observed that “the cases differ both in the kind of community in which the coverage took place and in extent of media coverage.” 500 U.S., at 429, 111 S. Ct. 1899, 114 L. Ed. 2d 493. Mu’Min’s offense occurred in the metropolitan Washington, D. C., area, “which has a population of over 3 million, and in which, unfortunately, hundreds of murders are committed each year.” Ibid. While the crime garnered “substantial” pretrial publicity, the coverage was not as pervasive as in Irvin and “did not contain the same sort of damaging information.” 500 U.S., at 429-430, 111 S. Ct. 1899, 114 L. Ed. 2d 493. Moreover, in contrast to Irvin, the seated jurors uniformly disclaimed having ever formed an opinion about the case. Given these circumstances, this Court rebuffed Mu’Min’s assertion that the trial court committed constitutional error by declining to “make precise inquiries about the contents of any news reports that potential jurors have read.” 500 U.S., at 424, 111 S. Ct. 1899, 114 L. Ed. 2d 493. The Court stressed, however, that its ruling was context specific: “Had the trial court in this case been confronted with the ‘wave of public passion’

[561 U.S. 445]

engendered by pretrial publicity that occurred in connection with Irvin’s trial, the Due Process Clause of the Fourteenth Amendment might well have required more extensive examination of potential jurors than it undertook here.” Id., at 429, 111 S. Ct. 1899, 114 L. Ed. 2d 493.

Ill

It is necessary to determine how this case compares to our existing fair-trial precedents. Were the circumstances so inherently prejudicial that, as in Rideau, even the most scrupulous voir dire would have been “but a hollow formality” incapable of *683reliably producing an impartial jury? 373 U.S., at 726, 83 S. Ct. 1417, 10 L. Ed. 2d 663. If the circumstances were not of this character, did the District Court conduct a jury selection process sufficiently adapted to the level of pretrial publicity and community animus to ensure the seating of jurors capable of presuming innocence and shutting out extrajudicial influences?

A

Though the question is close, I agree with the Court that the prospect of seating an unbiased jury in Houston was not so remote as to compel the conclusion that the District Court acted unconstitutionally in denying Skilling’s motion to change venue. Three considerations lead me to this conclusion. First, as the Court observes, ante, at 382, 177 L. Ed. 2d, at 643-644, the size and diversity of the Houston community make it probable that the jury pool contained a nontrivial number of persons who were unaffected by Enron’s collapse, neutral in their outlook, and unlikely to be swept up in the public furor. Second, media coverage of the case, while ubiquitous and often inflammatory, did not, as the Court points out, ante, at 382-383, 177 L. Ed. 2d, at 644, contain a confession by Skilling or similar “smoking-gun” evidence of specific criminal acts. For many prospective jurors, the guilty plea of co-defendant and alleged co-conspirator Causey, along with the pleas and convictions of other Enron executives, no doubt suggested guilt by association. But reasonable minds exposed to such information

[561 U.S. 446]

would not necessarily have formed an indelible impression that Skilling himself was guilty as charged. Cf. Rideau, 373 U.S., at 726, 83 S. Ct. 1417, 10 L. Ed. 2d 663 (a majority of the county’s residents were “exposed repeatedly and in depth to the spectacle of Rideau personally confessing in detail to the crimes with which he was later to be charged”). Third, there is no suggestion that the courtroom in this case became, as in Estes and Sheppard, a “carnival” in which the “calmness and solemnity” of the proceedings were compromised. Sheppard, 384 U.S., at 358, 350, 86 S. Ct. 1507, 16 L. Ed. 2d 600 (internal quotation marks omitted). It is thus appropriate to examine the voir dire and determine whether it instills confidence in the impartiality of the jury actually selected.9

*684B

In concluding that the voir dire “adequately detected] and defuse[d] juror bias,” ante, at 385, 177 L. Ed. 2d, at 646, the Court downplays the

[561 U.S. 447]

extent of the community’s antipathy toward Skilling and exaggerates the rigor of the jury selection process. The devastating impact of Enron’s collapse and the relentless media coverage demanded exceptional care on the part of the District Court to ensure the seating of an impartial jury. While the procedures employed by the District Court might have been adequate in the typical high-profile case, they did not suffice in the extraordinary circumstances of this case to safeguard Skilling’s constitutional right to a fair trial before an impartial jury.

In conducting this analysis, I am mindful of the “wide discretion” owed to trial courts when it comes to jury-related issues. Mu’Min, 500 U.S., at 427, 111 S. Ct. 1899, 114 L. Ed. 2d 493; cf. ante, at 386-387, 177 L. Ed. 2d, at 646-647. Trial courts are uniquely positioned to assess public sentiment and the credibility of prospective jurors. Proximity to events, however, is not always a virtue. Persons in the midst of a tumult often lack a panoramic view. “[AJppellate tribunals [thus] have the duty to make an independent evaluation of the circumstances.” Sheppard, 384 U.S., at 362, 86 S. Ct. 1507, 16 L. Ed. 2d 600. In particular, reviewing courts are well qualified to inquire into whether a trial court implemented procedures adequate to keep community prejudices from infecting the jury. If the jury selection process does not befit the circumstances of the case, the trial court’s rulings on impartiality are necessarily called into doubt. See Morgan, 504 U.S., at 729-730, 112 S. Ct. 2222, 119 L. Ed. 2d 492 (“ ‘Without an adequate voir dire the trial judge’s responsibility to remove prospective jurors who will not be able impartially to follow the court’s instructions and evaluate the evidence cannot be fulfilled’ ” (quoting Rosales-Lopez v. United States, 451 U.S. 182, 188, 101 S. Ct. 1629, 68 L. Ed. 2d 22 (1981) (plurality opinion))); see also Mu’Min, 500 U.S., at 451, 111 S. Ct. 1899, 114 L. Ed. 2d 493 (Kennedy, J., dissenting) (“Our willingness to accord substantial deference to a trial court’s finding of juror impartiality rests on our expectation that the trial court will conduct a sufficient voir dire to determine the credibility of a juror professing to be impartial” ).

[561 U.S. 448]

1

As the Court of Appeals apprehended, the District Court gave short shrift to the mountainous evidence of public hostility. For Houstonians, Enron’s collapse was an event of once-in-a-generation proportions. Not only was the volume of media coverage “immense” and frequently intemperate, but “the sheer number of victims” created a climate in which animosity toward Skilling ran deep and the desire for conviction was widely shared. 554 F.3d, at 559-560.

The level of public animus toward Skilling dwarfed that present in cases such as Murphy and Mu’Min. The pretrial publicity in those cases consisted of dozens of news reports, most of which were “largely factual in nature.” Murphy, 421 U.S., at 802, 95 S. Ct. 2031, 44 L. Ed. 2d 589. There was no indication that the relevant communities had been captivated by the cases or had adopted fixed views about the defendants. In contrast, the number of media reports in this case reached the tens of thousands, and *685full-throated denunciations of Skill-ing were common. The much closer analogy is thus to Irvin, which similarly featured a “barrage” of media coverage and a “huge . . . wave of public passion,” 366 U.S., at 725, 728, 81 S. Ct. 1639, 6 L. Ed. 2d 751, although even that case did not, as here, involve direct harm to entire segments of the community.10

Attempting to distinguish Irvin, the majority suggests that Skilling’s economic offenses were less incendiary than Irvin’s violent crime spree and that “news stories about Enron contained nothing resembling the horrifying information rife in reports about Irvin’s rampage of robberies and murders.” Ante, at 394, 177 L. Ed. 2d, at 651. Along similar lines, the District Court described “the facts of this case [as] neither heinous nor sensational.” App. to Brief for United States 10a. The majority also points to the four years that passed between

[561 U.S. 449]

Enron’s declaration of bankruptcy and the start of Skilling’s trial, asserting that “the decibel level of media attention diminished somewhat” over this time. Ante, at 383, 177 L. Ed. 2d, at 644. Neither of these arguments is persuasive.

First, while violent crimes may well provoke widespread community outrage more readily than crimes involving monetary loss, economic crimes are certainly capable of rousing public passions, particularly when thousands of unsuspecting people are robbed of their livelihoods and retirement savings. Indeed, the record in this case is replete with examples of visceral outrage toward Skilling and other Enron executives. See, e.g., Record 39946 (front-page, eve-of-trial story describing “the hurt and anger and resentment. . . churn [ing] inside” the people of Houston). Houstonians compared Skilling to, among other things, a rapist, an axe murderer, and an al Qaeda terrorist.11 As one commentator observed, “ [i]t’s a sign of how shocked Houstonians are about Enron’s ignominious demise that Sept. 11 can be invoked—and is frequently—to explain the shock of the company’s collapse.” 3 Supp. Record 544. The bad blood was so strong that Skilling and other top executives hired private security to protect themselves from persons inclined to take the law into their own hands. See, e.g., App. 1154a (“After taking the temperature of Enron’s victims, [a local lawyer] says the Enron executives are wise to take security precautions”).

[561 U.S. 450]

Second, the passage of time did little to soften community sentiment. Contrary to the Court’s suggestion, ante, at 383, 177 L. Ed. 2d, at 644, this case in no way resembles Yount, where, by the time of the defendant’s retrial, “prejudicial publicity [had] greatly diminished” and community animus had significantly waned. 467 *686U.S., at 1032, 104 S. Ct. 2885, 81 L. Ed. 2d 847; see also ibid, (in the months preceding the defendant’s retrial, newspaper reports about the case averaged “less than one article per month,” and public interest was “minimal”). The Enron story was a continuing saga, and “publicity remained intense throughout.” 554 F.3d, at 560. Not only did Enron’s downfall generate wall-to-wall news coverage, but so too did a succession of subsequent Enron-related events.12 Of particular note is the highly publicized guilty plea of codefendant Causey just weeks before Skilling’s trial. If anything, the time that elapsed between the bankruptcy and the trial made the task of seating an unbiased jury more difficult, not less. For many members

[561 U.S. 451]

of the jury pool, each highly publicized Enron-related guilty plea or conviction likely served to increase their certainty that Skill-ing too had engaged in—if not masterminded—criminal acts, particularly given that the media coverage reinforced this view. See supra, at 433-434, 177 L. Ed. 2d, at 675-676. The trial of Skilling and Lay was the culmination of all that had come before. See Record 40002 (noting that “prosecutors followed the classic pattern of working their way up through the ranks”). As the Chronicle put it in July 2005, shortly after the trial of several Enron Broadband Services executives ended without convictions: “The real trial, the true measure of justice in the Enron saga, begins in January. Let the small fry swim free if need be. We’ve got bigger fish in need of frying.” App. 1460a (paragraph breaks omitted); see also ibid. (“From the beginning, the Enron prosecution has had one true measure of success: Lay and Skilling in a cold steel cage”).

Any doubt that the prevailing mindset in the Houston community remained overwhelmingly negative was dispelled by prospective jurors’ responses to the written questionnaires. As previously indicated, supra, at 431-433, 177 L. Ed. 2d, at 674-675, more than one-third of the prospective jurors either knew victims of Enron’s collapse or were victims themselves, and two-thirds gave responses suggesting an antidefen-dant bias. In many instances their contempt for Skilling was palpable. See nn. 4, 6, supra. Only a small fraction of the prospective jurors raised no red flags in their responses. And this was before Causey’s guilty plea and the flurry of news reports that accompanied the approach of trial. One of Skilling’s experts, a po*687litical scientist who had studied pretrial publicity “for over 35 years” and consulted in more than 200 high-profile cases (in which he had recommended against venue changes more often than not), “c[a]me to the conclusion that the extent and depth of bias shown in these questionnaires is the highest or at least one of the very highest I have ever encountered.” App. ¶¶2, 7, at 783a, 785a (emphasis deleted).

[561 U.S. 452]

2

Given the extent of the antipathy evident both in the community at large and in the responses to the written questionnaire, it was critical for the District Court to take “strong measures” to ensure the selection of “an impartial jury free from outside influences.” Sheppard, 384 U.S., at 362, 86 S. Ct. 1507, 16 L. Ed. 2d 600. As this Court has recognized, “[i]n a community where most veniremen will admit to a disqualifying prejudice, the reliability of the others’ protestations may be drawn into question.” Murphy, 421 U.S., at 803, 95 S. Ct. 2031, 44 L. Ed. 2d 589; see also Groppi v. Wisconsin, 400 U.S. 505, 510, 91 S. Ct. 490, 27 L. Ed. 2d 571 (1971) (“‘[A]ny judge who has sat with juries knows that in spite of forms they are extremely likely to be impregnated by the environing atmosphere’ ” (quoting Frank v. Mangum, 237 U.S. 309, 349, 35 S. Ct. 582, 59 L. Ed. 969 (1915) (Holmes, J., dissenting))). Perhaps because it had underestimated the public’s antipathy toward Skilling, the District Court’s 5-hour voir dire was manifestly insufficient to identify and remove biased jurors.13

[561 U.S. 453]

As an initial matter, important lines of inquiry were not pursued at all. The majority accepts, for instance, that “publicity about a codefendant’s guilty plea calls for inquiry to guard against actual prejudice.” Ante, at 385, 177 L. Ed. 2d, at 645. Implying that the District Court undertook this inquiry, the majority states that “[o]nly two venire members recalled [Causey’s] plea.” Ibid. In fact, the court asked very few prospective jurors any questions directed to their knowledge of or feelings about that event.14 Considering how much news the plea generated, many more than *688two venire members were likely aware of it. The lack of questioning, however, makes the prejudicial impact of the plea on those jurors impossible to assess.

The court also rarely asked prospective jurors to describe personal interactions they may have had about the case, or to consider whether they might have difficulty avoiding discussion of the case with family, friends, or colleagues during the course of the lengthy trial. The tidbits of information that trickled out on these subjects provided cause for concern. In response to general media-related questions, several prospective jurors volunteered that they had spoken with others about the case. Juror 74, for example, indicated that her husband was the “news person,” that they had “talked about it,” that she had also heard things “from work,” and that what she heard was “all negative, of course.” App. 948a. The court, however, did not seek elaboration

[561 U.S. 454]

about the substance of these interactions. Surely many prospective jurors had similar conversations, particularly once they learned upon receiving the written questionnaire that they might end up on Skilling’s jury.

Prospective jurors’ personal interactions, moreover, may well have left them with the sense that the community was counting on a conviction. Yet this too was a subject the District Court did not adequately explore. On the few occasions when prospective jurors were asked whether they would feel pressure from the public to convict, they acknowledged that it might be difficult to return home after delivering a not-guilty verdict. Juror 75, for instance, told the court, “I think a lot of people feel that they’re guilty. And maybe they’re expecting something to come out of this trial.” Id., at 956a. It would be “tough,” she recognized, “to vote not guilty and go back into the community.” Id., at 957a; see also id., at 852a (Juror 10) (admitting “some hesitancy” about “telling people the government didn’t prove its case”).

With respect to potential nonmedia sources of bias, the District Court’s exchange with Juror 101 is particularly troubling.15 Although Juror 101 responded in the negative when asked whether she had “read anything in the newspaper that [stood] out in [her] mind,” she volunteered that she “just heard that, between the two of them, [Skilling and Lay] had $43 million to contribute for their case and that there was an insurance policy that they could collect on, also.” Id., at 998a. This information, she explained, “was just something I overheard today—other jurors talking.” Ibid. It seemed suspicious, she intimated, “to have an insurance policy ahead of time.” Id., at 999a. The court advised her that “most corporations provide insurance for their officers and directors.” Ibid. The court, however, did not investigate the matter further, even though it had earlier instructed prospective jurors not to talk to each other about the case. Id.,

[561 U.S. 455]

at 843a. It is thus not apparent whether other prospective jurors also overheard the information and whether they too believed that it reflected unfavorably on the defen*689dants; nor is it apparent what other outside information may have been shared among the venire members. At the very least, Juror 101’s statements indicate that the court’s questions were failing to bring to light the extent of jurors’ exposure to potentially prejudicial facts and that some prospective jurors were having difficulty following the court’s directives.

The topics that the District Court did cover were addressed in cursory fashion. Most prospective jurors were asked just a few yes/no questions about their general exposure to media coverage and a handful of additional questions concerning any responses to the written questionnaire that suggested bias. In many instances, their answers were unenlightening.16 Yet the court rarely sought to draw them out with open-ended questions about their impressions of Enron or Skilling and showed limited patience for counsel’s followup efforts. See, e.g., id., at 879a, 966a.17 When prospective

[561 U.S. 456]

jurors were more forthcoming, their responses tended to highlight the ubiquity and negative tone of the local news coverage, thus underscoring the need to press the more guarded members of the venire for further information.18 Juror 17, for example, mentioned hearing a radio program that very morning in which a former Enron employee compared persons who did not think Skilling was guilty to Holocaust deniers. See id., at 863a (“[H]e said he thought that he would find them guilty automatically if he was on the jury because he said that it would be worse than a German trying to say that they didn’t kill the Jews”).19 Other jurors may well have encountered, and been influenced by, similarly incendiary rhetoric.

These deficiencies in the form and content of the voir dire questions con*690tributed to a deeper problem: The District Court failed to make a sufficiently critical assessment of prospective jurors’ assurances of impartiality. Although the Court insists otherwise, ante, at 392, 177 L. Ed. 2d, at 650, the voir dire transcript indicates that the District Court essentially took jurors at

[561 U.S. 457]

their word when they promised to be fair. Indeed, the court declined to dismiss for cause any prospective juror who ultimately gave a clear assurance of impartiality, no matter how much equivocation preceded it. Juror 29, for instance, wrote on her questionnaire that Skilling was “not an honest man.” App. 881a. During questioning, she acknowledged having previously thought the defendants were guilty, and she disclosed that she lost $50,000-$60,000 in her 401(k) as a result of Enron’s collapse. Id., at 880a, 883a. But she ultimately agreed that she would be able to presume innocence. Id., at 881a, 884a. Noting that she “blame[d] Enron for the loss of her money” and appeared to have “unshakeable bias,” Skilling’s counsel challenged her for cause. Id., at 885a. The court, however, declined to remove her, stating that “she answered candidly she’s going to have an open mind now” and “agree [ing]” with the Government’s assertion that “we have to take her at her word .’’Id., at 885a-886a.20 As this Court has made plain, jurors’ assurances of impartiality simply are not entitled to this sort of talismanic significance. See, e.g., Murphy, 421 U.S., at 800, 95 S. Ct. 2031, 44 L. Ed. 2d 589 (“[T]he juror’s assurances

[561 U.S. 458]

that he is equal to th[e] task cannot be dispositive of the accused’s rights”); Irvin, 366 U.S., at 728, 81 S. Ct. 1639, 6 L. Ed. 2d 751 (“Where so many, so many times, admi[t] prejudice, ... a statement of impartiality can be given little weight”).

Worse still, the District Court on a number of occasions accepted declarations of impartiality that were equivocal on their face. Prospective jurors who “hope[d]” they could presume innocence and did “not necessarily” think Skilling was guilty were permitted to remain in the pool. App. 932a, 857a. Juror 61, for instance, wrote of Lay on her questionnaire, “Shame on him.” Id., at 931a. Asked by the court about this, she stated that, “innocent or guilty, he was at the helm” and “should have known what was going on at the company.” Ibid.; see also id., at 934a (Skilling is “probably” “in the same boat as” Lay). The court then asked, “[C]an you presume, as you start this trial, that Mr. Lay is innocent?” Id., at 932a. She responded, “I hope so, but you know. I don’t know. I can’t honestly answer that one way or the other.” Ibid.; see *691also id., at 933a (“I bring in my past history. I bring in my biases. I would like to think I could rise above those, but I’ve never been in this situation before. So I don’t know how I could honestly answer that question one way or the other. ... I do have some concerns”). Eventually, however, Juror 61 answered “Yes” when the court asked if she would be able to acquit if she had “a reasonable doubt that the defendants are guilty.” Id.., at 933a-934a. Challenging her for cause, defense counsel insisted that they had not received “a clear and unequivocal answer” about her ability to be fair. Ibid. The court denied the challenge, stating, “You know, she tried.” Ibid.

3

The majority takes solace in the fact that most of the persons actually seated as jurors and alternates “specifically stated that they had paid scant attention to Enron-related

[561 U.S. 459]

news.” Ante, at 390-391, 177 L. Ed. 2d, at 649, and n. 26.21 In context, however, these general declarations reveal little about the seated jurors’ actual knowledge or views or the possible pressure they might have felt to convict, and thus cannot instill confidence that the jurors “were not under [the] sway” of the prevailing community sentiment. Cf. ante, at 391, 177 L. Ed. 2d, at 649. Jurors who did not “get into details” of Enron’s complicated accounting schemes, App. 856a, nevertheless knew the outline of the oft-repeated story, including that Skilling and Lay had been cast as the leading villains. Juror 63, for instance, told the court that she “may have heard a little bit” about Enron-related litigation but had not “really pa[id] attention.” Id., at 935a. Yet she was clearly aware of some specifics. On her questionnaire, despite stating that she had not followed Enron-related news, she wrote about “whistleblowers and Arthur Andersen lying about Enron’s accounting,” and she expressed the view that Skilling and Lay “probably knew they were breaking the law.” Supp. App. 105sa-106sa. During questioning, which lasted barely four minutes, the District Court obtained no meaningful information about the actual extent of Juror 63’s familiarity with the case or the basis for her belief in Skilling’s guilt. Yet it nevertheless accepted her assurance that she could “absolutely” presume innocence. App. 937a.22

[561 U.S. 460]

Indeed, the District Court’s anemic questioning did little to dispel similar doubts about the impartiality of numerous other seated jurors and alternates. In my estimation, more than half of those seated made written and oral comments suggesting active antipathy toward the defendants. The majority thus misses the mark when it asserts that “Skilling’s seated ju*692rors . . . exhibited nothing like the display of bias shown in Irvin.” Ante, at 394, 177 L. Ed. 2d, at 651. Juror 10, for instance, reported on his written questionnaire that he knew several co-workers who owned Enron stock; that he personally may have owned Enron stock through a mutual fund; that he heard and read about the Enron cases from the “Houston Chronicle, all three Houston news channels, Fox news, talking with friends [and] co-workers, [and] Texas Lawyer Magazine”; that he believed Enron’s collapse “was due to greed and mismanagement”; that “[i]f [Lay] did not know what was going on in his company, he was really a poor manager/leader”; and that the defendants were “suspect.” Supp. App. 11sa-19sa. During questioning, he said he “th[ought]” he could presume innocence and “believe [d]” he could put the Government to its proof, but he also acknowledged that he might have “some hesitancy” “in telling people the government didn’t prove its case.” App. 851a-852a.

Juror 11 wrote that he “work[ed] with someone who worked at Enron”; that he got Enron-related news from the “Houston Chronicle, Channel 2 News, Channel 13 News, O’Reilly Factor, [and] talking with friends and co-workers”; that he regularly visited the Chronicle Web site; that “greed on Enron’s part” caused the company’s collapse; and that “[a] lot of people were hurt financially.” Supp. App. 26sa-30sa. During questioning, he stated that he would have “no

[561 U.S. 461]

problem” requiring the Government to prove its case, but he also told the court that he believed Lay was “greedy” and that corporate executives are often “stretching the legal limits .... I’m not going to say that they’re all crooks, but, you know.” App. 857a, 854a. Asked whether he would “star[t] the case with sort of an inkling that because [Lay is] greedy he must have done something illegal,” he offered an indeterminate “not necessarily.” Id., at 857.23

[561 U.S. 462]

While several seated jurors and alternates did not make specific com*693ments suggesting prejudice, their written and oral responses were so abbreviated as to make it virtually impossible for the District Court reliably to assess whether they harbored any latent biases. Juror 13, for instance, wrote on his questionnaire that he had heard about the Enron cases from the “[n]ews.” Supp. App. 42sa. The court questioned him for two minutes, during which time he confirmed that he had “heard what’s on the news, basically,” including “that the trial had moved from the 17th to the 31st.” He added that the story “was all over the news on every detail of Enron.” App. 858a-860a. No meaningful information about his knowledge or attitudes was obtained. Similarly, Juror 78 wrote that she had not followed Enron-related news but was aware that “[m]any people lost their jobs.” Supp. App. 151sa. The court questioned her for less than 90 seconds. During that time, she acknowledged that she had “caught glimpses” of the coverage and “kn[e]w generally, you know, that the company went bankrupt” and that there “were some employees that went off and did their own businesses.” App. 969a. Little more was learned.24

In assessing the likelihood that bias lurked in the minds of at least some of these seated jurors, I find telling the way in

[561 U.S. 463]

which voir dire played out. When the District Court asked the prospective jurors as a group whether they had any reservations about their ability to presume innocence and put the Government to its proof, only two answered in the affirmative, and both were excused for cause. Id., at 815a-820a. The District Court’s individual questioning, though truncated, exposed disqualifying prejudices among numerous additional prospective jurors who had earlier expressed no concerns about their impartiality. See n. 7, supra. It thus strikes me as highly likely that at least some of the seated jurors, despite stating that they could be fair, harbored similar biases that a more probing inquiry would likely have exposed. Cf. Yount, 467 U.S., at 1034, n. 10, 104 S. Ct. 2885, 81 L. Ed. 2d 847 (holding that the trial court’s “particularly extensive” 10-day voir dire ensured the jury’s impartiality).25

The majority suggests, ante, at 383-384, 395, 177 L. Ed. 2d, at 644-645, 651, that the jury’s decision to acquit Skilling on nine relatively *694minor insider trading charges confirms its impartiality. This argument, however, mistakes partiality with bad faith or blind vindictiveness. Jurors who act in good faith and sincerely believe in their own fairness may nevertheless harbor disqualifying prejudices. Such jurors may well acquit where evidence is wholly lacking, while subconsciously resolving closer calls against the defendant rather than giving him the benefit of the doubt. Cf. United States v. McVeigh, 918 F. Supp. 1467, 1472 (WD Okla. 1996) (Prejudice “may go unrecognized

[561 U.S. 464]

in those who are affected by it. The prejudice that may deny a fair trial is not limited to a bias or discriminatory attitude. It includes an impairment of the deliberative process of deductive reasoning from evi-dentiary facts resulting from an attribution to something not included in the evidence. That something has its most powerful effect if it generates strong emotional responses”). In this regard, it is significant that the Government placed relatively little emphasis on the nine insider trading counts during its closing argument, declining to explain its theory on those counts in any detail whatsoever. Record 37010. The acquittals on those counts thus provide scant basis for inferring a lack of prejudice.

In sum, I cannot accept the majority’s conclusion that voir dire gave the District Court “a sturdy foundation to assess fitness for jury service.” Cf. ante, at 395, 177 L. Ed. 2d, at 651. Taken together, the District Court’s failure to cover certain vital subjects, its superficial coverage of other topics, and its uncritical acceptance of assurances of impartiality leave me doubtful that Skilling’s jury was indeed free from the deep-seated animosity that pervaded the community at large. “[RJegardless of the heinousness of the crime charged, the apparent guilt of the offender[,] or the station in life which he occupies,” our system of justice demands trials that are fair in both appearance and fact. Irvin, 366 U.S., at 722, 81 S. Ct. 1639, 6 L. Ed. 2d 751. Because I do not believe Skilling’s trial met this standard, I would grant him relief.