United States v. Aguilar

994 F.2d 609, 1993 WL 151376

• Court: Court of Appeals for the Ninth Circuit
• Decided: September 2, 1993
• Docket: Nos. 90-10597, 91-10024
• Status: Published

Opinions

ORDER

On June 13, 1989, a federal grand jury returned an eight count indictment against United States District Judge Robert P. Aguilar. The indictment charged Judge Aguilar with conspiring to defraud the United States in violation of 18 U.S.C. § 371, disclosing the existence of a wiretap application in violation of 18 U.S.C. § 2232(c) (counts four and six), endeavoring to obstruct justice in violation of 18 U.S.C. § 1503 (counts seven and eight), and racketeering in violation of 18 U.S.C. § 1962(c). In August 1990, Judge Aguilar went to trial on five of the counts and, on August 22, a petit jury found him guilty on one count of illegally disclosing a wiretap in violation of 18 U.S.C. § 2232(c) (count six) and one count of endeavoring to obstruct a grand jury investigation in violation of 18 U.S.C. § 1503 (count eight). It acquitted Judge Aguilar on the other three counts. On November 1, 1990, the district court sentenced Judge Aguilar to two six month terms of imprisonment, to be served concurrently, and fined him $2,000. Aguilar appeals his conviction on both counts and the government appeals his sentence.

The separate opinions of Judges Hug and O’Scannlain are filed contemporaneously with this order and opinion. The conviction for wiretap disclosure, count six of the indictment, is AFFIRMED for the reasons expressed in this opinion, as to which Judge O’Scannlain concurs in part. The conviction for obstruction of justice, count eight of the indictment, is REVERSED for the reasons expressed in the dissenting opinion of Judge Hug, as to which Judge O’Scannlain concurs in part. The sentence of six months’ imprisonment and fine of $1,000 under count six is VACATED for the reasons expressed in the opinion of Judge O’Scannlain, as to which Judge Hug concurs in part, and REMANDED for reconsideration in light of United States v. Lira-Barraza, 941 F.2d 745 (9th Cir.1991) (en banc).

OPINION

CYNTHIA HOLCOMB HALL, Circuit Judge:

I

BACKGROUND¹

The chain of events leading to Judge Aguilar’s indictment began with the 1980 conviction of Michael Rudy Tham for embezzling $2,000 from Local 856 of the Freight Checkers, Clerical Employees and Helpers Union, an affiliate of the International Brotherhood of Teamsters. Tham was Secretary-Treasurer of Local 856 and a member of several local and national Teamster committees. In July 1987, Tham filed a motion under 28 U.S.C. § 2255 to have his conviction set aside. In an effort to gain favorable treatment from District Judge Stanley Weigel, Tham sought the assistance of attorney Edward Solomon, an acquaintance of Judge Aguilar, and Abraham Chalupowitz, alias “Trigger Abe” Chapman, a putative mobster distantly related to Judge Aguilar by marriage. Chapman arranged a September 12, 1987, meeting between Solomon and Judge Aguilar to discuss Tham’s case and how Judge Aguilar could assist Tham’s petition. Over the next several months, Judge Aguilar discussed the case regularly with Chapman and Solomon, reporting to them his conversations with Judge Weigel about the matter.

Throughout this time, the San Francisco FBI was investigating local labor racketeering as part of a nationwide probe into health care provider fraud. Tham and Chapman were among the subjects of the investigation. On April 20, 1987, the FBI applied to then-Chief United States District Judge Robert Peckham for authorization to conduct electronic surveillance of Tham’s business telephones and named Chapman as an intercep-tee of the wiretap. On May 20, the wiretap expired and, on July 15, Judge Peckham signed the first of a series of orders maintaining the secrecy of the wiretap while the investigation was ongoing. The last of the orders was signed on January 25, 1989.

As the investigation progressed, the FBI discovered that Chapman had a relationship with Judge Aguilar. On July 9, state agents observed Chapman leaving the Federal Building in San Jose with Judge Aguilar. The state agents informed the FBI and the FBI informed Judge Peckham. Judge Peck-ham expressed his concern, as well as his belief that Judge Aguilar must have been ignorant of Chapman’s criminal activities.

A month later at an ABA reception in the Mark Hopkins Hotel in San Francisco, in an effort to warn Judge Aguilar about the character of the company he was keeping and help him avoid the appearance of impropriety, Judge Peckham mentioned to Judge Aguilar that Chapman’s name had come up in connection with a wiretap application. He remarked that years earlier, as an Assistant United States Attorney, he had prosecuted Chapman on a drug charge and expressed his surprise that Chapman was still involved in criminal activity. Judge Aguilar indicated that he knew Chapman, but disclosed nothing else about their relationship.

Several months later, on February 6, 1988, Chapman paid a visit to Judge Aguilar at his home. As Chapman was leaving, Aguilar noticed that they were being observed from across the street by a man whom Judge Aguilar suspected was an FBI agent. Immediately Judge Aguilar telephoned his nephew Steve Aguilar, Chapman’s grandson, and asked him to come to Aguilar’s house. When Steve arrived, Judge Aguilar told him that an FBI agent was following Chapman and that he had heard “at work” that Chapman was being wiretapped. He instructed Steve to relay the information to Chapman, which he apparently did.

On April 26, 1988, a grand jury in the Northern District of California began to hear evidence regarding a conspiracy by Judge Aguilar, Tham, Chapman and Edward Solomon to influence the outcome of Tham’s case. In May, Solomon agreed to cooperate with the FBI. On May 17, Solomon wore a wire to a luncheon meeting with Judge Aguilar. Over the course of their conversation, Judge Aguilar revealed that he had discussed the Tham matter with Judge Weigel and had warned Chapman about the wiretap. He also instructed Solomon to lie to the FBI about Aguilar’s involvement in Tham’s case. On May 26, Solomon recorded a second conversation in which he informed Aguilar that a grand jury was investigating their dealings in the Tham case. The two men also discussed how Solomon should lie to the grand jury in order to explain the records of Solomon’s telephone calls to Judge Aguilar regarding the Tham matter.

Armed with this evidence, two FBI agents met with Judge Aguilar on June 22, 1988, and questioned him about his involvement in the Tham ease. Before Judge Aguilar would respond to any questions, he wanted to know if he was a “target” of a grand jury investigation. Agent Carlon hesitated in answering the question, but when Aguilar said, “I assume from this that I very likely am [a target] if I’ve done anything wrong,” Carlon responded affirmatively. Judge Aguilar then proceeded to lie to the agents about his relationship with Solomon, his involvement in the Tham case, and his knowledge of the wiretap. In response to Aguilar’s further questions about whether or not he was a target, Carlon conceded that a grand jury was meeting, but would say nothing about the subject of the investigation or Aguilar’s status. The interview continued and Aguilar lied further about his dealings with Chapman. At trial, Judge Aguilar testified that, by the end of the interview, he had concluded that his statements would be presented to the grand jury.

Based on this evidence, the jury convicted Judge Aguilar on one count of disclosing the existence of a wiretap in violation of 18 U.S.C. § 2232(c) and one count of endeavoring to obstruct a grand jury investigation in violation of 18 U.S.C. § 1503. Aguilar appeals each conviction on several grounds. He argues that the government’s evidence was insufficient to prove that he satisfied the “knowledge” requirement of either section 2232(c) or section 1503; that the jury instructions -misstated the “knowledge” elements of the statutes; and that the instructions imper-missibly shifted to him the burden of proving that he lacked the knowledge required by these statutes. He also argues that the application of section 2232(c) to his disclosure of the wiretap violates his First Amendment right to free speech.

The government appeals Judge Aguilar’s sentence, arguing that the district judge had no authority to depart downward six levels from the sentence level prescribed in the Sentencing Guidelines.

II

DISCLOSURE OF THE WIRETAP

The wiretap disclosure statute, 18 U.S.C. § 2232(c),² makes it a crime for any person with knowledge that a wiretap authorization has been applied for or granted to divulge that knowledge with the purpose of impeding the interception of communications. Accordingly, the trial court instructed the jury that the government was required to prove that “defendant had knowledge that authorizations to conduct electronic surveillance of Abe Chapman had been applied for.” The court continued:

Knowledge of a fact, members of the jury, means that you’re satisfied from the evidence that he knew it[.] Or knowledge of the existence of a particular circumstance may be satisfied by proof that the defendant was aware of a high probability of the existence of that circumstance[,\ unless you find from, the evidence that the defendant actually believed that the circumstance did not exist.

(emphasis added). Aguilar raises four challenges to his conviction under section 2232(c). He argues that: (1) the evidence was insufficient to demonstrate that he had “knowledge” that the FBI had applied to conduct a wiretap; (2) application of section 2232(c) to Aguilar’s disclosure of the wiretap violated his First Amendment right to free speech; (3) the trial court’s jury instruction on the “knowledge” requirement of section 2232(c) misstated the element; and (4) the “knowledge” instruction unconstitutionally shifted the burden of proof on “knowledge” onto Aguilar.

A. Sufficiency of the Evidence

Aguilar argues that the evidence was insufficient to show that he had knowledge of a wiretap application. In reviewing the sufficiency of the evidence, this Court must determine whether, “after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 2789, 61 L.Ed.2d 560 (1979).

A defendant may not be convicted under section 2232(c) for disclosing a wiretap unless there is sufficient evidence to show that he knew that permission to conduct a wiretap had been applied for or granted. The evidence in this case was sufficient to convince a rational trier of fact that Judge Aguilar had such knowledge. The jury heard the tape of the May 17 conversation between Judge Aguilar and Edward Solomon, in which Judge Aguilar stated that he “knew they were wiretapping [Chapman],” and described learning about the wiretap from Judge Peckham. A criminal defendant’s own statements about his knowledge are as strong evidence of that knowledge as a jury is ever likely to hear. This evidence was reinforced by Judge Peckham’s testimony that he told Judge Aguilar about the wiretap application and Steven Aguilar’s testimony that his uncle told him that he had learned at work , that Chapman’s telephone had been tapped.³

Aguilar argues that “the fact that both the application and the wiretap had long since expired when he first heard about the subject from Judge Peckham strongly suggests that he lacked the actual knowledge required by the statute.” Aguilar’s argument ignores the plain language of section 2232(e). The statute imposes liability on anyone “having knowledge that a Federal investigative or law enforcement officer has been authorized or has applied for authorization” to conduct a wiretap. 18 U.S.C. § 2232(c) (1988) (emphasis added). This language makes it unlawful to reveal a wiretap with the wrongful purpose required by section 2232(c) although the authorization is no longer current.

The dissent, agreeing with Aguilar, invites us to read “has been authorized” as “is currently authorized.” Although we find the statutory language standing alone sufficiently clear to rule out this construction, we note that the legislative history accompanying the enactment of section (c) also refutes this interpretation: the report describing the newly-created offense stated that the statute prohibits “giving notice of the possible interception to any person who was or is the target of the interception.” S.Rep. No. 541, 99th Cong., 1st Sess. 1, 34 (1986) (emphasis added), reprinted in 1986 U.S.C.C.A.N. 3555, 3588.

Congress has set a limit upon the time during which a wiretap may remain secret. See 18 U.S.C. § 2518(8)(d) (1988). It has also provided for extensions of that time upon a showing of good cause. See id. We are not here confronted with a prosecution for disclosure of a wiretap after the government had already served notice of the wiretap upon the interceptee, which might raise more troubling questions of the nature posed by the dissent. In this case, although the specific wiretap of which Judge Aguilar had learned expired on May 20, 1987, the district court ordered its secrecy maintained through January, 1989. Moreover, other wiretaps were authorized in the interim; wiretaps for which Chapman was an interceptee were in place from September 11 to October 12,1987, and from October 21, 1987, through May 8, 1988. Thus, at the time Aguilar warned Chapman of the surveillance, it was still lawfully secret.

The dissent reads the statute to require that a defendant must intend to impede the specific unexpired wiretap of which he has knowledge before he can violate the law. See post at 628, 629. This construction permits the disclosure of any expired wiretap authorization, regardless whether the wiretap is still secret and regardless whether the disclosure may impede ongoing undercover investigations. To interpret section 2232(c) to permit disclosure by third parties while the government has obtained judicial approval to maintain secrecy would defeat the purpose of section 2518(8)(d). It is unlikely Congress intended to permit judicially-authorized surveillance efforts to be frustrated by such a narrow reading of the statute. As such a reading is incompatible with the statutory language, we reject it.

The dissent’s emphasis on the existence of some interception with which disclosure of the particular wiretap could potentially interfere is unduly concerned with the possibility of actual interference. We think that the statute’s primary concern is with the giving of notice with specific intent to interfere with surveillance. That such notice may not actually, nor even possibly, impede the particular surveillance episode of which Aguilar learned from Judge Peckham is less important.⁴ Apart from the statutory language itself, this is evidenced by the fact that the statute also prohibits attempts to engage in such conduct.⁵

We hold that the fact that the wiretap authorization of which Aguilar learned from Judge Peckham had expired at the time Judge Aguilar disclosed its existence to Chapman does not render his “knowledge” insufficient to support a conviction under section 2232(c).

B. Freedom of Speech

Aguilar also argues that section 2232(c), by punishing him for speaking about the wiretap, violates his First Amendment right to free speech. We review constitutional challenges to statutes de novo. United States v. Scampini, 911 F.2d 350, 351 (9th Cir.1990).

Aguilar does not dispute that if he had a duty to maintain the confidentiality of the wiretap application, he had no First Amendment right to disclose the application. See Snepp v. United States, 444 U.S. 507, 509 n. 3, 100 S.Ct. 763, 765 n. 3, 62 L.Ed.2d 704 (1980) (per curiam). Rather, he argues that section 2232(c) violates his First Amendment rights because he had no such duty. He contends, therefore, that the district court erred by refusing to instruct the jury that “ ‘knowledge [for] the purposes of 2232(c) must come from confidential information, information derived from the judge[’]s employment.” Failure to give this instruction, he argues, deprived him “of the right to have the factual dispute about whether the disclosure involved a breach of duty resolved by the jury.” Aguilar maintains that had the jury been given the opportunity to consider this question, it would have found that he had no duty. The government could not have demonstrated the confidentiality of the information Aguilar received from Judge Peck-ham because it could not “articulate any conceivable relationship between the disclosure by Judge Peekham and Judge Aguilar’s official duties.”

The problem with Aguilar’s argument is that it posits limits on his responsibilities as a federal judge that do not exist. It simply does not matter that Judge Peekham informed him of the wiretap at a cocktail party rather than in chambers, or some other “official” setting. Judge Aguilar had a duty not to disclose the information because it came from Judge Peekham and related to official judicial business. The duty stemmed not from the place or time at which Judge Aguilar heard the information, but from the nature and source of the information.

In Snepp the Supreme Court considered whether a former CIA agent had a First Amendment right to publish a book on CIA activities in South Vietnam, based on information learned during the course of his employment. The Court found that an agreement Snepp had signed with the CIA prohibited him from publishing without CIA approval. But the Court noted that “apart from the plain language of the agreement, the nature of Snepp’s duties and his conceded access to confidential sources and materials could establish a trust relationship.” 444 U.S. at 511 n. 6, 100 S.Ct. at 766 n. 6. The same can be said of federal judges. Every day, we discuss confidential court business with our colleagues and our staff. It is well understood we are not free to discuss these matters with the attorneys that appear before us, with our families, or with our friends outside the court. Although it is difficult to find authorities discussing the existence of a duty on the part of federal judges to guard the confidentiality of their communications with their colleagues, that is undoubtedly because “its existence and validity has been so universally recognized. Its source is rooted in history and gains added force from the constitutional separation of powers of the three departments of government.” Nixon v. Sirica, 487 F.2d 700, 740 (D.C.Cir.1973) (MacKinnon, J., dissenting) (discussing existence of a privilege protecting disclosure of communications among judges).

Our conclusion about the nature of the judicial duty of confidentiality is backed by recognition that were it not for the existence of such a duty, Judge Aguilar would never have learned of the wiretap. Judge Peck-ham testified that he believed Judge Aguilar would keep the matter confidential. Surely he felt free to relate information about a wiretap application to Judge Aguilar only because he regarded the conversation as confidential. Judge Aguilar’s access to the information regarding the wiretap was a consequence of his judicial position and he therefore had a duty not to disclose the information.

Several years ago, a special federal appeals panel held that communications among judges “relating to official judicial business” are protected by a qualified privilege. See In re Certain Complaints Under Investigation by an Investigating Committee of the Judicial Council of the Eleventh Circuit, 783 F.2d 1488, 1520 (11th Cir.), cert. denied, 477 U.S. 904, 106 S.Ct. 3273, 91 L.Ed.2d 563 (1986). The court did not restrict the privilege to information a judge discloses to his staff or colleagues during the course of his official duties. It held that the privilege protected material “relating to official judicial business.” We see no reason to circumscribe the corresponding official duty of confidentiality more narrowly.

If we were to accept Aguilar’s logic, we would have to conclude that a judge never has a duty to keep confidential the information he or she learns from other judges in a social setting. There is a range of internal matters that judges and courts jealously guard from the public eye. Aguilar’s “official duties” limitation on his duty of confidentiality would permit judges to freely disgorge these matters if they were to learn about them over lunch with a colleague, or at a cocktail party at another judge’s home. The spectre of such disclosure alone is enough to stifle discussion among the members of the federal judiciary and undermine the integrity of the Article III branch.

Aguilar argues, however, that his acquittal on count four, which charged him with disclosing the wiretap in August 1987, after he spoke with Judge Peckham but before he saw the FBI agent observing Chapman, demonstrates that the jury concluded that he gained the requisite knowledge from his observation of the agent, not his conversation with Judge Peckham. He argues that even if he had a duty not to disclose what he learned from Judge Peckham, he had no such duty not to disclose what he learned from his own observation. We need not consider whether a federal judge’s duty of confidentiality is limited in this manner, because there can be no doubt that Judge Aguilar’s knowledge of the wiretap was based on what he learned from Judge Peckham. He himself said it.

Aguilar’s nephew testified that when Aguilar informed him of the wiretap he said he had learned of the wiretap “at work.” Indeed, the only sensible explanation for Judge Aguilar’s knowledge is that when he saw the FBI agent trailing Chapman, he recalled his conversation with Judge Peckham and put two and two together; it is difficult to imagine how his observation of the FBI agent alone could have led him to conclude that Chapman’s telephone had been tapped. If Judge Aguilar’s knowledge were based even partially on what Judge Peckham told him, then he had a duty not to disclose it. The jury could have acquitted Judge Aguilar of count four for a number of reasons; there is no reason for us to conclude that it acquitted him because he lacked “knowledge.” We decline Aguilar’s invitation to draw conclusions about the jury’s reasoning based on its decision to acquit him of count four.

Because Judge Aguilar had a duty not to disclose the wiretap, nothing in the First Amendment would prohibit the jury from applying section 2232(c) to his conduct.

C. Misstatement of the Knowledge Element

Aguilar argues that the court’s definition of knowledge as the awareness of a “high probability” of the existence of a fact was improper. He contends that this definition describes an objective state of mind akin to “negligence.” The definition, he suggests, conflicts with the “basic and widely understood” principle that knowledge is a subjective state of mind requiring “actual awareness” of a fact. It is this “settled interpretation”-of knowledge that Congress must have had in mind when it drafted section 2232(c), according to Aguilar. The definition used by the trial court, which Aguilar refers to alternatively as a “diluted” or “objective” standard of knowledge, is only appropriate when there is evidence that a defendant has consciously avoided learning an incriminating fact. If there is no such evidence, an instruction employing this definition creates the risk that the jury will convict the defendant for what he should have known, he argues. See United States v. Garzon, 688 F.2d 607, 609 (9th Cir.1982); United States v. Murrieta-Bejarano, 552 F.2d 1323, 1325 (9th Cir.1977). Because there was no evidence of “deliberate indifference” in this case, Aguilar maintains the government was required to prove that he possessed “actual knowledge.” Whether a jury instruction misstates an element of an offense is a question of law that this court reviews de novo. United States v. Spillone, 879 F.2d 514, 525 (9th Cir.1989), cert. denied, 498 U.S. 878, 111 S.Ct. 210, 112 L.Ed.2d 170 (1990).

The Electronic Communications Privacy Act of 1986, Pub.L. No. 99-508, § 109, 100 Stat. 1858 (1986) (“the Act”), of which section 2232(c) is a part, is silent on the meaning of the statute’s mens rea requirement. See 18 U.S.C. § 2510 (1988). Left without guidance from the statutory text, the trial judge turned to the legislative history and concluded that Congress had imported the Model Penal Code definition of “knowledge”⁶ into the Act. Although the legislative history is of little help in discerning Congress’s intent,⁷ I’ conclude that the trial judge’s definition was perfectly appropriate. The Model Penal Code view that one has knowledge of an attendant fact when one is “aware of a high probability of its existence” is as accurate a description of the subjective state of mind I call “knowledge” as any I can imagine.

I would reject the premise that Model Penal Code section 2.02(7) does not comport with traditional understandings of “knowledge.” We have “actual” or “positive” knowledge of very few things we claim to “know.” If we need a carton of milk, we go to the grocery store because we “know” there will be milk on the shelves. We slow down before we reach a yellow traffic light because we “know” that by the time we reach the light it will have turned red. Willie Sutton robbed banks because, to paraphrase the legendary thief, he “knew” the money was there. One could conceive of countless similar examples of facts we “know,” not because we have actual knowledge of their correctness' — knowledge that comes only when we arrive at the dairy section, when the light turns red, or when we enter the bank safe — but because we are “aware of a high probability” that the fact exists. And despite Aguilar’s contentions, this state of mind differs from “actual knowledge” only in degree, not in kind. It is a subjective belief based on fact.

Courts have long recognized that “actual awareness” is only a subset of what we commonly think of as “knowledge.” “ Absolute knowledge can be had of very few things,’ ... and the philosopher might add ‘if any.’ For most practical purposes ‘knowledge’ is not confined to what we have personally observed or to what we have evolved by our own cognitive faculties.” Rollin M. Perkins & Ronald N. Boyce, Criminal Law 865 (3d ed. 1982) (quoting Story v. Buffum, 90 Mass. 35, 38 (1864) and State v. Ransberger, 106 Mo. 135, 17 S.W. 290, 292 (1891)). Indeed, courts have often employed a definition of “knowledge” less demanding than that of the Model Penal Code. Courts routinely hold that a defendant may satisfy the “knowledge” requirement for the crime of receipt of stolen goods if he merely has a “guilty belief’ that the goods are stolen. See id. at 865-75.

Professor Perkins notes that the law will view the requirement of “knowledge” differently in different circumstances. He gives the example of a person who has been told that a certain bill of exchange is a forgery and believes the statement to be true. “Obviously [the person does not have ‘knowledge’ of the forgery] if the purpose of the inquiry is to determine whether he is qualified to take the witness stand and swear that the instrument is false; but if he passes the bill as genuine he will be uttering a forged instrument with ‘knowledge’ of the forgery if his belief is correct.” Id. Likewise, it would make no sense for courts to require that violators of section 2232(c) have the same degree of knowledge that they would require of witnesses. Judge Aguilar would not be qualified to testify that a wiretap application had been submitted unless he had witnessed the application. But if that degree of “knowledge” was required for prosecution under section 2232(c), then only the FBI agent and United States Attorney who applied for the wiretap and the judge who granted it could be subject to prosecution, and that surely is not what Congress intended to be the coverage of the statute.

Despite all this, Aguilar argues that the drafters of the Model Penal Code conceived of section 2.02(7) not as a definition of knowledge, but as an exception to the traditional requirement of “actual” knowledge applicable only when there is evidence that a defendant has remained deliberately ignorant of a fact. He suggests that courts have ratified this limitation. Nothing in the text of section 2.02(7) even implies the limitation Aguilar proposes. Aguilar, however, fastens onto the commentary to section 2.02, which states that “[sjubsection (7) deals with the situation that British commentators have denominated as ‘willful blindness.’ ” Model Penal Code § 2.02 comment 9 (Proposed Official Draft 1962). A more thorough reading of the comment reveals that the drafters did not conceive of subsection (7) as an exception to the knowledge requirement applicable only when “willful blindness” is an issue, but rather as a comprehensive definition of knowledge designed to account for “willful blindness.” “The inference of ‘knowledge’ of an existing fact is usually drawn from proof of notice of high probability of its existence, unless the defendant establishes an honest, contrary belief. Subsection (7) solidifies this tisual result and clarifies the terms in which the issue is submitted to the jury.” Id. (emphasis added). Read in its entirety, the meaning of comment 9 is clear: section 2.02(7) was designed to incorporate the concept of “willful blindness” into a general definition of knowledge that would limit the equation of “willful blindness” to “knowledge” to circumstances in which the defendant really did have knowledge. See Wayne R. LaFave & Austin W. Scott, Jr., Substantive Criminal Law § 3.5(b), at 307-08 (1986) (“The Model Penal Code requirement of an awareness of a high probability of the existence of the fact serves to ensure that the purpose to avoid learning the truth is culpable.”).

This circuit first adopted the Model Penal Code definition in United States v. Jewell, 532 F.2d 697 (9th Cir.) (en banc), cert. denied, 426 U.S. 951, 96 S.Ct. 3173, 49 L.Ed.2d 1188 (1976). There we recognized that section 2.02(7) purported to be a general definition of knowledge and we approved of it as such. “[IJn common understanding one ‘knows’ facts of which he is less than absolutely certain. To act ‘knowingly,’ therefore, is not necessarily to act only with positive knowledge, but also to act with an awareness of the high probability of the existence of the fact in question.” Id. at 700. Judge (now Justice) Kennedy, writing for the dissent, agreed that section 2.02(7) is “a definition of knowledge, not a substitute for it.” Id. at 707 (Kennedy, J., dissenting); see also United States v. Yermian, 708 F.2d 365, 371-72 (9th Cir.1983) (recognizing the Model Penal Code definition as the “ ‘subjective standard of criminal knowledge required by this circuit”), rev’d on other grounds, 468 U.S. 63, 104 S.Ct. 2936, 82 L.Ed.2d 53 (1984); 2 Edward J. Devitt et al., Federal Jury Practice and Instructions § 54.15, at 923 (1990) (citing Jeivell for proposition that in drug possession crimes, “[a] defendant does not have to act with positive knowledge in order to be culpable. It is sufficient if the defendant acts with an awareness of the high probability of the fact in question.”).

The Supreme Court’s approval of section 2.02(7) as a general definition of knowledge is clear from its use of the definition in cases in which “willful blindness” was not an issue. In Leary v. United States, 395 U.S. 6, 89 S.Ct. 1532, 23 L.Ed.2d 57 (1969), the Court considered the constitutionality of a presumption in 21 U.S.C. § 176a that evidence sufficient to show that a defendant possessed marijuana is also sufficient to show that the defendant knew the marijuana was illegally imported. The Court employed section 2.02(7) to interpret the meaning of “knowing” in the statute. See id. at 46 n. 93, 89 S.Ct. at 1553; see also Barnes v. United States, 412 U.S. 837, 845, 93 S.Ct. 2357, 2363, 37 L.Ed.2d 380 (1973) (considering constitutionality of inference of knowledge from possession of stolen goods and upholding defendant’s conviction for possession of stolen Treasury checks on ground that the evidence could support finding that defendant “must have known or been aware of a high probability that the checks were stolen”); Turner v. United States, 396 U.S. 398, 416, 90 S.Ct. 642, 652, 24 L.Ed.2d 610 (1970) (relying on section 2.02(7) definition to interpret knowledge presumption in federal drug possession statute, 21 U.S.C. § 174); United States v. Hester, 880 F.2d 799, 803 n. 4 (4th Cir.1989) (“[T]he Court has indicated general acceptance of the proposition that awareness of ‘a high probability’ that a fact exists may properly be equated with ‘knowledge’ in the interpretation of criminal statutes.”) (citing Leary, 395 U.S. at 46 n. 93, 89 S.Ct. at 1553 n. 93).

Aguilar nevertheless argues that if applied where there is no evidence of “willful blindness,” the Model Penal Code definition by itself implies a duty on the part of a defendant to learn a fact and therefore creates the risk that the defendant will be convicted for what he should have known. Aguilar relies on a line of cases holding that because a “willful blindness” instruction — which includes but is not limited to the Model Penal Code definition — carries with it the risk that the jury will evaluate the defendant’s behavior under a negligence standard and convict him for what he should have known, it should only be given in the rare case in which there is evidence that the defendant deliberately avoided learning the truth. See e.g., United States v. Sanchez-Robles, 927 F.2d 1070, 1073-75 (9th Cir.1991); Garzon, 688 F.2d at 609; Murrieta-Bejarano, 552 F.2d at 1325.

The risk identified by Murrieta-Bejarano and its progeny was never present in this case. That risk is created when an instruction that a defendant’s conscious effort to avoid the truth is the equivalent of knowledge is given in a case where there is no evidence that the defendant did indeed consciously avoid learning the truth. In such a case, there is a risk that the jury will conclude that the defendant had some duty to inquire about the existence of the fact, if a reasonable person would have done so. See Garzon, 688 F.2d at 609. It is the instruction on conscious avoidance, not the Model Penal Code definition of knowledge which must accompany it, that implies the existence of a duty. In this case the district court did not instruct the jury on “conscious avoidance.”

The Model Penal Code definition of knowledge standing alone implies no duty to discover facts. Contrary to Aguilar’s argument, the definition describes a subjective state of mind, and therefore cannot subject a defendant to liability for negligence.⁸ See Yermian, 708 F.2d at 372 (subsection (7) as the “ ‘subjective’ standard of criminal knowledge required by this circuit”); Ira P. Robbins, The Ostrich Instruction: Deliberate Indifference as a Criminal Mens Rea, 81 J.Crim.L. & Criminology 191, 227 (1990) (“[T]he Model Penal Code formulation protect[s] the defendant from conviction for merely negligent behavior.”). Indeed, the “balancing clause” of the definition, which instructs the jury that a defendant does not have knowledge if he “actually believes [the fact] does not exist,” ensures that the defendant is not convicted for negligence. United States v. Esquer-Gamez, 550 F.2d 1231, 1235-36 (9th Cir.1977).

I would hold that the trial court did not err by instructing the jury that a person has knowledge if he is “aware of a high probability that a fact exists, unless he actually believes it does not exist.” The definition describes a subjective state of mind and creates no risk that the defendant will be convicted for what he “should have known.” Indeed it perfectly describes the state of mind that lay people typically regard as knowledge, that best fits the purpose of section 2232(e), and that this circuit and the Supreme Court have approved as a general definition of “knowledge.” We conclude as well that the evidence of Judge Aguilar’s “knowledge” was so substantial, that had the jury received no instruction, it would still have found that Judge Aguilar had knowledge of the wiretap.⁹ We need not be concerned that the instruction might have allowed the jury to convict Aguilar “even if he was uncertain about whether there was a wiretap,” as the dissent fears, since the evidence reflected no uncertainty: Judge Aguilar himself admitted in tape recorded conversations with Solomon that he “knew.” If the instruction were error, it was harmless beyond a reasonable doubt in the face of Aguilar’s admissions of knowledge.¹⁰

D. Burden of Proof

Aguilar next argues that the jury instructions on knowledge unconstitutionally shifted the burden of proof onto him in violation of due process. Aguilar first raised his constitutional objection to the court’s “knowledge” instruction in his motion for a new trial. When there is no objection to a jury instruction at trial, this Court reviews the instruction for plain error. United States v. Bustillo, 789 F.2d 1364, 1367 (9th Cir.1986). “A plain error is a highly prejudicial error affecting substantial rights.” Id.

In Sandstrom v. Montana, 442 U.S. 510, 99 S.Ct. 2450, 61 L.Ed.2d 39 (1979), the Supreme Court held that a jury instruction stating that “ ‘the law presumes that a person intends the ordinary consequences of his voluntary acts,’ violates the Fourteenth Amendment’s requirement that the State prove every element of a criminal offense beyond a reasonable doubt.” Id. at 512, 99 S.Ct. at 2453. A jury could interpret such an instruction as shifting the burden of proof to the defendant. Id. at 524, 99 S.Ct. at 2459.

Aguilar argues that the last clause of the trial court’s instruction on “knowledge” had the same effect as the instruction in Sand-strom. The district court instructed the jury that

[kjnowledge of the evidence of a particular circumstance may be satisfied by proof that the defendant was aware of a high probability of the existence of that circumstance!),] unless you find from the evidence that the defendant actually believed that the circumstance did not exist.

(emphasis added).

In Esquer-Gamez, we held that the “balancing clause” of the Model Penal Code definition is necessary to protect the defendant against the risk that he will be convicted for what he “should have known.” See 550 F.2d at 1235-36. In United States v. Hogg, 670 F.2d 1358 (4th Cir.1982), however, the Fourth Circuit held that a nearly identical instruction violated Sandstrom. That court concluded that a reasonable jury could interpret the instruction to mean that “once the government shows that the defendants knew [the circumstance] probably existed, the defendants must introduce more than some evidence that they had no actual knowledge to avoid a finding against them on this element.” Id. at 1363-64. It held that Sand-strom “prohibits such a persuasion-shifting instruction.” Id. at 1364.

I disagree that the reasoning of Sand-strom supports the holding of Hogg. In Sandstrom, the Court found that an instruction that the law presumes a person to intend the consequences of his acts could have shifted the burden of proof on the element of intent to the defendant: The jury “could have concluded that upon proof by the State of the slaying, and of additional facts not themselves establishing the element of intent, the burden was shifted to the defendant to prove that he lacked the requisite mental state.” 442 U.S. at 524, 99 S.Ct. at 2459.

The instruction given in this case could have no such effect because it did not create a presumption that the defendant had to overcome. The function of the balancing clause is to ensure that the jury does not hold the defendant responsible for what a reasonable person would know, if in fact the defendant actually believed the fact did not exist. It directs the jury to focus on the defendant’s actual belief, nothing more. The clause does not imply anything about whose burden it is to prove the defendant’s state of mind. Here, the district court stated unequivocally that the government had to prove that Aguilar was aware of a high probability that the wiretap application existed. The balancing clause in no way lightened that burden.

Furthermore, even if the instruction itself were faulty, which I believe is not so, we do not consider jury instructions in isolation, but rather in the context of the other instructions given. Maddox v. City of Los Angeles, 792 F.2d 1408, 1412 (9th Cir.1986). We have held that reminders that the government has the burden of proving every element of the crime beyond a reasonable doubt are sufficient to cure instructions that might otherwise shift the burden of proof. See United States v. Wyatt, 807 F.2d 1480, 1481-82 (9th Cir.), cert. denied, 484 U.S. 858, 108 S.Ct. 170, 98 L.Ed.2d 124 (1987). Considered in context, the knowledge instruction hardly could have led a reasonable juror to conclude that Judge Aguilar had the burden of proving his lack of knowledge. The jury was reminded repeatedly that the government was required to prove every element of the offense beyond a reasonable doubt. And, as we have said, the evidence of Judge Aguilar’s knowledge, coming as it did from his own mouth, was overwhelming. Because the jury was cautioned about the government’s burden, and because the evidence was so weighty, the instruction could not have been “highly prejudicial” and therefore was not plain error.

For the foregoing reasons, I would reject each of Judge Aguilar’s objections to his conviction under section 2232(c); moreover, we conclude that any possible error was harmless beyond a reasonable doubt. See post at 641 (O’Scannlain, J., concurring).

Ill

OBSTRUCTION OF JUSTICE

The judicial obstruction statute, 18 U.S.C. § 1503,¹¹ also contains a “knowledge” component. Though the text of the statute prescribes no scienter requirement, this circuit has held that knowledge of a judicial proceeding is an element of a section 1503 offense. See United States v. Washington Water Power Co., 793 F.2d 1079, 1084 (9th Cir.1986); United States v. Rasheed, 663 F.2d 843, 852 (9th Cir.1981), cert. denied, 454 U.S. 1157, 102 S.Ct. 1031, 71 L.Ed.2d 315 (1982). The trial court, therefore, instructed the jury that “there must be knowledge of the existence of a judicial proceeding before a defendant can be convicted of this offense.” It then defined knowledge in precisely the same terms it had used in the wiretap instruction.

Aguilar challenges his conviction under section 1503 on the ground that the evidence was insufficient to show that he had knowledge that the FBI agents to whom he lied would be witnesses before a grand jury. He also raises the same objections to the court’s instruction on “knowledge” that he raised with respect to section 2232(c). For the same reasons that I would hold that the instruction on the “knowledge” element of section 2232(c) properly defined knowledge and did not violate due process, I would hold that the instruction on the knowledge element of section 1503 was also proper. A majority of this panel having concluded otherwise, a decision from which I respectfully dissent, this conviction will be reversed and we do not reach the question of the sufficiency of the evidence. Were it necessary to do so, I would also hold there was sufficient evidence for a rational jury to find Aguilar knew that FBI agent Carlon was expected to be a witness in .front of the grand jury. Judge Aguilar clearly knew that a grand jury was meeting, he repeatedly questioned the agents about whether he was a target, and he testified about how grand jury investigations operate, which demonstrated to the jury his ability to correctly determine both his own status and the purpose of the interview. A reasonable jury could have found that a federal judge, sophisticated on the subject of grand jury investigations, who knew that a grand jury was in session and who “was aware of a high probability” that he was a target of its investigation, also knew that Agent Carlon would be expected to testify before the grand jury about Aguilar’s statements to him. I therefore would uphold Aguilar’s conviction for violating section 1503.

IV

DEPARTURE FROM THE SENTENCING GUIDELINES

We agree that at a minimum, the district court failed to provide a reasoned explanation for its six-level downward departure, and I concur in Judge O’Scannlain’s opinion to that extent. I most vigorously dissent, however, from the conclusion that the district court had the legal authority to depart downward on the basis it did.

The mitigating circumstance upon which the district court based its downward departure was the “additional punishment” it anticipated Judge Aguilar would suffer during the course of potential disbarment and impeachment hearings. The plurality acknowledges that the Guidelines account for impeachment, but concludes that the Guidelines do not account for “the likelihood that Judge Aguilar will suffer additional punishment throughout the course of several disciplinary hearings.” Post at 643-44 (O’Scannlain, J., concurring). The plurality then concludes that the district court had legal authority to depart on this basis, a conclusion I find insupportable.

First, and most fundamentally, departures prompted by the substantial pain and humiliation or financial hardship those in high places suffer when they break the law is inconsistent with the Guidelines’ policy that disparity in sentencing should never be occasioned by socio-economic status. See 28 U.S.C. § 994(d) (1988); U.S.S.G. Ch. 1 Pt. A 4(b) & § 5H1.10.¹² The district court emphasized that its willingness to depart “is limited only to those features that I believe are so different and unique with this defendant.” It concluded that these features would be “so extensive and so enduring” that they formed an appropriate basis for departure. But there is no escaping the conclusion that the court’s departure was grounded in the notion that the consequences that accrue when someone who holds a position of high esteem and importance violates the law are uniquely significant and harsh. Though the court recognized that collateral consequences arising from a criminal defendant’s social and professional stature are inappropriate bases for departure, it did not seem to realize that its departure was grounded in precisely such considerations. With respect, I must say that a plurality of this panel now makes the same error.

Second, the circumstance upon which the district court based its departure is, regretfully, not all that unique. The Sentencing Guidelines’ policy statement on departures states that courts should depart only in the “atypical” case. See U.S.S.G. Ch. 1, Pt. A 4(b); cf. United States v. Anders, 956 F.2d 907, 912 (9th Cir.1992) (factors “not ordinarily relevant” may be considered only in “extraordinary circumstances”), cert. denied, — U.S. -, 113 S.Ct. 1592, 123 L.Ed.2d 158 (1993). The kind of additional humiliation and suffering Judge Aguilar will suffer, while not common, is not “atypical.” The district court’s reasoning would apply to many well-known figures in the worlds of government and finance who have been prosecuted over the last twenty years. I cannot accept the notion that the humiliation of disbarment or impeachment proceedings is different in kind or degree from the humiliation, shame and stigma borne by numerous other respected community members convicted of serious crimes.¹³

No one disputes that Judge Aguilar’s job accounts for the collateral consequences he now faces in the form of “long, humiliating, and burdensome” proceedings of disbarment and impeachment, as well as loss of pension benefits and preclusion from public service. See post at 643-44.¹⁴ The plurality finds this permissible, however, because a departure on the basis of Judge Aguilar’s job is not based on the judge’s “status in society.” Id. The plurality’s own support refutes this notion: “the phrase ‘socio-economic status’ refers to an individual’s status in society as determined by objective criteria such as education, income, and employment. ” United States v. Lopez, 938 F.2d 1293, 1297 (D.C.Cir.1991) (emphasis added). In our society, rightly or wrongly, one’s job is often equivalent to one’s status in society. What is more, the plurality buttresses its conclusion that departure on the basis of Aguilar’s job is permissible on the additional basis .that Judge Aguilar stands to lose his pension too. See post at 644-45. How one could assert that financial ramifications of conviction is not a socioeconomic factor is beyond comprehension.

A criminal conviction may mean a loss of current position and foreclosure of future professions or positions for many defendants. Professionals in the securities industry, once convicted of violating the securities laws, lose their brokers’ licenses in quasi-judicial proceedings not unlike disbarment, and thereafter are barred from holding certain positions in the securities industry. See 15 U.S.C. § 78o(a), (b)(4) (1988). Likewise, persons convicted of certain offenses are barred from holding positions within a labor union or pension plan, see 29 U.S.C. §§ 504(a), 1111 (1988) — ironically, the genesis of Rudy Tham’s solicitation of Judge Aguilar’s influence and this case. I doubt any court would support a downward departure for a securities broker convicted of insider trading violations on the ground that she faced a burdensome civil liability proceeding, stood to lose her broker’s license, would be barred from registering as a securities broker or dealer in the future, had to disgorge her profits, and suffered extreme public humiliation as she was drummed out of Wall Street. I cannot see that Judge Aguilar’s position is so different.

The Guidelines’ policy is that “persons who abuse their positions' of trust ... generally are viewed as more culpable.” U.S.S.G. § 3B1.3 comment, (backg’d). We must assume that the Sentencing Commission has adequately considered the special circumstances of defendants who hold high office, and rejected any notion that such persons should receive more lenient treatment. The district court’s departure on the basis of consequences flowing from Judge Aguilar’s breach of the public trust flies in the face of the Guidelines’ policy. Collateral consequences of conviction arising from or affecting one’s job simply are not a permissible basis for departure. See United States v. Rutana, 932 F.2d 1155, 1158-59 (6th Cir.), cert. denied, — U.S. -, 112 S.Ct. 300, 116 L.Ed.2d 243 (1991).

“A long course of adversarial proceedings” awaits many a defendant following criminal conviction — be they forfeiture proceedings, civil suits seeking tort damages, or administrative proceedings to revoke professional licenses. These suits are burdensome in themselves, and their consequences can be exceedingly onerous. Defendants can lose their homes, investments, businesses, and entire estates in forfeiture proceedings following conviction of drug or racketeering offenses, or can be hit with staggering civil damage awards for many white collar crimes. Judge Aguilar’s potential forfeiture of his pension rights is no different. Personal financial hardship is not a permissible basis for downward departure. U.S.S.G. § 5K2.12.

The plurality’s statement that the ‘extra punishment’ Judge Aguilar faces “can be distinguished” from other collateral effects arising from conviction, see post at 644 n. 3, is not convincing. First, the plurality addresses only deportation; the collateral effects more comparable to what Judge Aguilar faces are personal financial hardship and foreclosure of career opportunities. I submit that despite the fact that deportation “is not considered punitive,” its “undeniable impact” is far more punitive than mere loss of position. For that matter, disbarment is not considered punitive either:

It must be noted that although the word ‘punishment’ is frequently used, the discipline of an attorney is not punitive in character.... The purpose of such a proceeding is to determine the fitness of an officer of the court to continue in that capacity, and it has been said the disbarment of attorneys is not intended for the punishment of the individual, but for the protection of the courts and the legal profession.

Marsh v. State Bar, 2 Cal.2d 75, 39 P.2d 403, 405 (1934); accord In re Lavine, 126 F.Supp. 39, 48 (S.D.Cal.1954), rev’d on other grounds, 217 F.2d 190 (9th Cir.1954); United States ex rel. Fitzgerald v. Stump, 112 F.Supp. 236, 237 (D.Alaska 1953).

Second, because deportation is a normal collateral consequence of conviction for a drug offense, the plurality assumes the Sentencing Commission adequately considered it in determining the appropriate offense level for drug crimes. Apart from the fact that this logic is fallacious — the offense levels for drug convictions were set for aliens and citizens alike, so the consequence of deportation could not possibly be reflected in the offense level for drug crimes — the same reasoning would apply here: disbarment as a consequence of a serious criminal conviction is not rare, and indeed ought to follow as a matter of course. In re Quimby, 359 F.2d 257, 258 (D.C.Cir.1966) (per curiam). Accordingly, by the plurality’s logic, we should assume the Sentencing Commission adequately allowed for that circumstance in determining offense levels.

Of course, we need not rely on fictional assumptions since the Commission expressly considered the factors upon which the district court seeks to depart — and pronounced each of them either irrelevant, see U.S.S.G. §§ 5H1.10 (socio-economic status), 5H1.11 (public service), or a basis for harsher treatment, cf. id. §§ 3B1.3 (court must adjust upward for abuse of position of trust); 5E1.4 (court must impose forfeiture upon convicted defendant as provided by law).

In short, I see no meaningful distinction, either qualitative or quantitative, between the additional “punishment” Judge Aguilar faces and that suffered by hundreds of convicted criminals. The plurality states that Judge Aguilar’s case “does not appear to fall within the heartland of cases for which the Guidelines were designed.” Post at 645. Judge Aguilar’s case — his conduct and offense of conviction — is not at issue here. What the plurality of course means is that Judge Aguilar’s personal circumstances are not those of the average defendant, and because of those circumstances, the Guideline sentence for Judge Aguilar’s offense appears unduly harsh. That assessment is equally true for hundreds of decent and sympathetic defendants now serving time in federal prisons. The response to this severity cannot take the form of special justice simply because Judge Aguilar is “the first convicted federal , judge” upon whom it will be visited.

When a high-ranking member of the judiciary, especially a respected and popular judge, is arrested for a serious crime, the entire judicial system is placed on trial. The world waits to see if judges can judge other judges fairly and equally.

Alan M. Dershowitz, Justice on Trial, N.Y. Times, Nov. 18, 1992, at A27. No judge can be treated more leniently than would a nonjudicial defendant in a comparable case. To accord Judge Aguilar special consideration on the basis of his position is to violate the fundamental premise not only of the Guidelines, but of this nation: equal justice for all.

The circumstance upon which the court based its departure was inconsistent with the rules and policies of the Sentencing Guidelines and therefore in violation of 18 U.S.C. § 3553(b) as interpreted by this Court in Lira-Barraza.15 I would hold that the dis-triet court, on remand, may not depart downward on the basis of consequences stemming from Aguilar’s judicial position.¹⁶

Conclusion

In sum, we hold that the evidence was sufficient for a rational jury 'to find that Judge Aguilar had knowledge of the wiretap, and that the use of section 2232(c) to punish Judge Aguilar for disclosing the wiretap did not violate his rights under the First Amendment. Judge Aguilar’s conviction for violating 18 U.S.C. § 2232(c) is therefore AFFIRMED. I would also hold that the jury insti’uction did not misstate the “knowledge” element of section 2232(c) and that the instruction did not unconstitutionally shift the burden of proof.

I would also hold that the evidence was sufficient for a rational jury to find Judge Aguilar had knowledge that Agent Carlon would be a witness before the grand jury, that the jury instruction did not misstate the “knowledge” element of section 1503, and that the instruction did not unconstitutionally shift the burden of proof. I therefore respectfully dissent from the panel’s decision that Judge Aguilar’s conviction for violating 18 U.S.C. § 1503 must be REVERSED.

On the government’s cross appeal, we hold that the district court failed to include a reasoned explanation of the extent of its departure as required by Lira-Barraza. Judge Aguilar’s sentence is therefore VACATED and the case is REMANDED to the district court for resentencing. I would also hold that the district court lacked authority to depart downward from the offense level prescribed by the Sentencing Guidelines based on the additional “punishment” awaiting Aguilar. I most vigorously dissent from the panel’s contrary decision.

See p. 626 for dissent by Judge Hug and special concurrence by Judge O’Scannlain.

2 . Whoever, having knowledge that a Federal investigative or law enforcement officer has been authorized or has applied for authorization under chapter 119 to intercept a wire, oral, or electronic communication, in order to obstruct, impede or prevent such interception, gives notice or attempts to give notice of the possible interception to any person shall be fined under this title or imprisoned not more than five years, or both.

18 U.S.C. § 2232(c) (1988).

3 . Contrary to the dissent’s characterization, we do not conclude that Judge Aguilar had knowledge of the new wiretap on Chapman’s telephone in effect during February 1988. See post at 636 (Hug, J., dissenting). Rather, we conclude that the evidence of his knowledge of the May 1987 wiretap on Tham’s telephone is sufficient to support his conviction for the February 6 disclosure of that wiretap to Chapman, an interceptee of that wiretap.

4 . Wc note that impossibility is generally not a defense to a crime. See, e.g., United States v. Quijada, 588 F.2d 1253, 1255 (9th Cir.1978) ("generally, a defendant should be treated in accordance with the facts as he supposed them to be."). The fact that the particular wiretap had expired should not insulate Aguilar from his attempt to frustrate the surveillance effort, particularly where, as here, subsequent wiretaps were in place such that he accomplished his objective.

5 . We fail to see the significance of the fact that Congress chose to forbid attempted disclosures in the same statute in which it forbade disclosure. See post at 636 (Hug, J., dissenting). We also note that "erroneous belief” as to the existence of a wiretap only permits a conviction under section 2232(c) when accompanied by the specific intent to impede an interception. See 18 U.S.C. § 2232(c). This "guilty belief" is equivalent to "knowledge” in the sense of subjective certainty; it may be objectively incorrect, but it is nonetheless culpable when coupled with the proscribed conduct of disclosure.

6 . "(7) Requirement of Knowledge Satisfied by Knowledge of High Probability. When knowledge of the existence of a particular fact is an element of an offense, such knowledge is established if a person is aware of a high probability of its existence, unless he actually believes it does not exist.” Model Penal Code § 2.02(7) (Proposed Official Draft 1962).

7 . A passage in the House Report on the Act states that section 2232(c) requires “that the defendant have knowledge that the federal law enforcement or investigative officer has been authorized to or has applied for an interception order.” H.R.Rep. No. 647, 99th Cong., 2d Sess. 1, 60 (1986). The word "knowledge” is footnoted "[s]ee House Report 96-1396, Criminal Code Revision Act of 1980, at 32-36.” Id. at 60 n. 91. The cited pages describe the mens rea require--ments of the Model Penal Code, which were to have been adopted as part of the 1980 reform proposal. See H.R.Rep. No. 1396, 96th Cong., 2d Sess. 1, 32-36 (1980). The Report does not resolve the question at issue here — whether Congress believed that definition should apply in cases where "deliberate ignorance” is not an issue. See id. at 35-36.

8 . Even the dissent recognizes that the knowledge instruction imposed a subjective standard. See post at 634-35 (Hug, J., dissenting).

9 . It seems to be the typical practice of judges on this circuit not to give any definition of "knowledge.” The Ninth Circuit Model Jury Instructions do not include an instruction defining knowledge. And we have previously held that it is not error for a court not to define "knowingly." "The word is a common word which an average juror can understand and which the average juror could have applied to the facts of this case without difficulty.” United States v. Chambers, 918 F.2d 1455, 1460 (9th Cir.1990). As I have already indicated, I believe the average juror's understanding of "knowledge" comports with the instruction given in this case.

10 . The dissent finds a serious factual question as to what Aguilar knew and what he merely suspected. Post at 635. It is impossible for a jury to know the strength of certainty with which a belief is held, which may run the continuum from doubt to suspicion to belief to certainty, except by what is manifested objectively. I do not believe the jury was required to disbelieve what Aguilar said he knew and find that he could not actually have known it. Judge Aguilar’s contention that Aguilar could not have known a fact that did not exist, see post at 635-36, suggests that his concern is not with the knowledge instruction; rather, it is with the notion that Aguilar's crime is a factual impossibility. When disclosure of actual surveillance efforts occurs, the specific intent to frustrate surveillance efforts exists, and there is subjective belief in the existence of attendant circumstances that makes the conduct and intent criminal, any error as to that belief does not relieve the act of its culpability. Thus the drug dealer who sells soap powder mistakenly "knowing” it to be heroin is guilty. See Quijada, 588 F.2d at 1255 (citing United States v. Roman, 356 F.Supp. 434, 438 (S.D.N.Y.), aff'd, 484 F.2d 1271 (2d Cir.1973), cert. denied, 415 U.S. 978, 94 S.Ct. 1565, 39 L.Ed.2d 874 (1974)).

11 . Whoever ... corruptly or by threats or force, or by any threatening letter or communication ... endeavors to influence, obstruct, or impede, the due administration of justice, shall be fined not more than $5,000 or imprisoned not more than five years, or both.

18 U.S.C. § 1503 (1988).

12 . Where most defendant characteristics are deemed "not ordinarily relevant," socio-economic status is flatly "not relevant.” Thus, although some offender characteristics may be taken into account where not adequately considered by the Guidelines, socio-economic status is not one of these. The plurality’s suggestion to the contrary is troubling. See post at 627 n. 2. To the extent that United States v. Valdez-Gonzalez, 957 F.2d 643 (9th Cir.1992), stands for the proposition that a court may consider socio-economic "factors,” as opposed to "status,” it should be noted that (1) the "factors” considered therein were "the socioeconomics and the internal politics of the drug trade along the Mexican border,” 957 F.2d at 649, not factors unique to the defendants' personal situations; (2) the court looked to socioeconomic factors not as a basis for departure, but in evaluating an otherwise permissible factor (role in the offense); and (3) the departures were at least consistent with the Guidelines, paralleling downward adjustments the defendants would have received under section 3B1.2 had the other participants in their crimes been indicted. See 957 F.2d at 648. Valdez-Gonzalez is surely no support for the result in this case.

13 . Nor can I accept the plurality’s implication that Judge Aguilar’s suffering is rendered unique by his status as an "Article III" judge. Whether Aguilar is a federal, state, municipal, or traffic court judge is irrelevant and does not render his circumstances "atypical.”

14 . The plurality characterizes this as "extra punishment,” post at 644, and perhaps rightfully so. But it is a far cry from solitary confinement in prison, the only instance of extra punishment held to warrant downward departure. See United States v. Lara, 905 F.2d 599 (2d Cir.1990).

15 . Even if I were to conclude that it were permissible for the court to depart, I doubt that a six level downward departure was justifiable. A district court must consider the validity, of each level of departure. See Lira-Barraza, 941 F.2d at 748. The court must also provide a statement explaining and justifying the departure. See id. at 751. The district court jumped six levels without explanation. It is difficult to imagine how such an extensive departure could be justified by reference to the structure of the Guidelines.

16 . I do not find it troubling that a more severe sentence should be imposed on remand despite the reversal- of one count of conviction. This results from the district court’s failure to comply with the Guidelines in the first instance.