Criminal Justice Section  


Criminal Justice Magazine
Winter 2002
Volume 18 Issue 1


Peter A. Joy and Kevin C. McMunigal

Peter A. Joy is professor of law and director of the Criminal Justice Clinic at Washington University School of Law in St. Louis, Missouri. Kevin C. McMunigal is the Judge Ben C. Green Professor of Law at Case Western Reserve University School of Law in Cleveland, Ohio. Both are contributing editors to Criminal Justice magazine.

Anti-Contact Rule in Criminal Investigations

Confidentiality issues are magnets for attention, as evidenced by news coverage this past summer of the Ethics 2000 Commission’s proposal to permit lawyers to disclose future client financial crime or fraud. Conflict of interest, the most heavily litigated subject area in legal ethics, also regularly draws interest from the public and lawyers. In comparison, the anti-contact rule has a much lower public profile. Nor does it register brightly on the ethical radar screens of most lawyers. Indeed, many lawyers seem unaware that the rule exists.

Nonetheless, the anti-contact rule has generated considerable controversy in the past decade, especially its application to criminal investigations. That controversy continued last fall in the wake of the September 11th terrorist attacks on the World Trade Center and the Pentagon with the introduction of Senate Bill 1435 aimed at freeing federal prosecutors from the restrictions of state anti-contact rules out of fear they may hinder undercover investigations.

In this column we examine the anti-contact rule in the context of criminal investigations, its underlying policies, its application both to prosecutors and defense lawyers, and the debate that continues to surround the rule.

The rule

Today either the Model Code or the Model Rules version of the anti-contact rule has been adopted in each of the 50 states. The Model Rules current version, Rule 4.2, provides:

In representing a client, a lawyer shall not communicate about the subject of the representation with a person the lawyer knows to be represented by another lawyer in the matter, unless the lawyer has the consent of the other lawyer or is authorized by law to do so.

The rule forbids a lawyer who is representing a client from communicating with the client of another lawyer regarding the representation. Its prohibition can be represented graphically as follows:

Communication is allowed along the sides of the square. One lawyer may communicate with the opposing lawyer, one client may communicate with the opposing client, and each lawyer may communicate with his or her own client. But communications between lawyer and opposing client diagonally across the square are generally forbidden.

The rule has a number of key features. The lawyer must know that another lawyer represents the person. Communication is barred only if it relates to "the subject matter of the representation." And only the opposing lawyer, not the client, may waive the protection of the rule.

One limiting feature worth noting is that the rule applies only if the lawyer is representing a client. Consider, for example, a situation in which a client represented by Lawyer A in a criminal matter is dissatisfied with Lawyer A and wishes to communicate with Lawyer B about replacing Lawyer A. May Lawyer B speak with Lawyer A’s client? Since B is not "representing a client" at the time of the communication, the anti-contact rule’s prohibitions do not prevent Lawyer B from talking to A’s client.

The rule also has a vicarious aspect. The Model Code’s anti-contact rule, DR 7-104(A)(1), specifies that a lawyer may not "cause another to communicate" with another lawyer’s client under the same circumstances in which the attorney would himself or herself be barred from so communicating. The text of Model Rule 4.2 does not contain a similar explicit prohibition on communicating through another person. But the Model Rules achieve the same effect through Model Rules 5.3 and 8.4(a).

Model Rule 5.3 imposes responsibilities on lawyers for nonlawyers "employed, retained, or associated with a lawyer." Rule 5.3(b) requires a lawyer to "make reasonable efforts to ensure" that the conduct of such nonlawyers "is compatible with the professional obligations of the lawyer." And Rule 5.3(c) imposes responsibility on the lawyer for conduct of such nonlawyers "which would be a violation . . . if engaged in by the lawyer" if the lawyer orders or ratifies the offending conduct or, under certain circumstances, "fails to take reasonable remedial action." Model Rule 8.4(a) provides that "[i]t is professional misconduct for a lawyer to . . . violate . . . the Rules of Professional Conduct . . . through the acts of another."

This vicarious dimension of the anti-contact rule has been the source of much of the controversy in applying the anti-contact rule to prosecutors based on the actions of police and informants working with those prosecutors during criminal investigations, a subject we examine below.

Model Code DR 7–104(A)(1) is virtually identical in content to the current version of Model Rule 4.2 with one exception. The Model Code bars communication with a represented "party" rather than "person." Model Rule 4.2 used the word "party" until 1995, when "person" was substituted. Some had argued that use of the word "party" implied that the rule’s protections attached only after the filing of an action, since a client does not become a party until an action is filed. The Model Rule’s change of "party" to "person" helps clarify the rule’s applicability prior to the filing of an action as well as in non-litigation contexts. Many states retain the word "party" in their anti-contact rules, but courts have nonetheless found them applicable in the pre-filing context.

Considerable attention has recently been focused on the question of the anti-contact rule’s application to corporate clients. The rule has uniformly been held to apply to corporate clients, though there is wide variation as to which employees of the corporation an opposing lawyer is barred from contacting under the rule. In an effort to bring consistency to identifying which employees are subject to the anti-contact rule, the ABA Ethics 2000 Commission’s most recent proposed solution prohibits communication with "a constituent" of the corporation who "supervises, directs, or regularly consults with the organization’s lawyer concerning the matter or has authority to obligate the organization with respect to the matter, or whose act or omission in the matter may be imputed to the organization for purposes of civil or criminal liability."

Potential sanctions for violating the anti-contact rule include professional discipline, disqualification, and suppression. The use of suppression as a sanction has been widely accepted by courts. This acceptance has given courts increased occasion to address application of the anti-contact rule in criminal investigations and fueled the growth in the last decade of case law on the subject.

Underlying policies

Three primary rationales have traditionally been advanced in support of the anti-contact rule. First, it protects a lawyer’s client from "overreaching and deception" by an opposing lawyer. Second, the rule shields the relationship between a client and lawyer from interference by an opposing lawyer. Third, it helps guarantee that an opposing lawyer will not breach the confidentiality of a client’s communications with his or her lawyer. (Restatement of the Law Governing Lawyers
§ 158, cmt. b; ABA Comm. on Ethics and Prof’l Responsibility, Formal Op. 95–396 (1995).)

Courts have recognized that the anti-contact rule also protects a "societal interest" that laypersons not make decisions with legal implications without the advice of their own lawyers, United States v. Batchelor, 484 F. Supp. 812 (E.D. Pa. 1980), and fosters "the internal integrity of and public confidence in the judicial system." ( United States v. Talao, 222 F.3d 1133 (9th Cir. 2000).)

Not as frequently mentioned are the potential costs of the rule. It can hinder access to information by the client’s adversary, especially when the adversary is a corporation. It may also hinder the prevention or remedying of client wrongdoing. Finally, it may hinder the prevention or remedying of unethical conduct by attorneys, such as conflict of interest or failing to keep a client informed.


Police commonly use undercover agents or informants to gather information, including damaging admissions, from the subject of an investigation. What if the prosecutor knows the subject of the investigation is represented by a lawyer in regard to the matter under investigation? Can the prosecutor nonetheless authorize the undercover agent or informant to communicate with the represented person?

Two powerful policy concerns militate against applying the rule to bar such undercover investigative activity. First, it would effectively insulate wealthy and sophisticated clients with retained counsel, such as corporate and white collar criminals and those engaged in organized crime, from investigative techniques that are effective at unmasking their misdeeds. Second, since a prosecutor’s involvement triggers the anti-contact rule, it would discourage prosecutorial supervision of police investigations.

There are, however, cogent arguments in favor of applying the anti-contact rule to police undercover contact with persons represented by lawyers. The idea of exempting prosecutors from a generally applicable legal ethics standard strikes many as unwise. It also seems unfair if defense counsel are not also exempted. Finally, it seems contrary to the plain language of the rule.

"Authorized by law" exception. The existence of these competing views helps explain the troubled history of the most well-known case to grapple with these issues, United States v. Hammad. In United States v. Hammad, 846 F.2d 854 (2d Cir. 1988), the Second Circuit originally held against any such prosecutorial exemption. It reviewed the actions of a federal prosecutor who arranged for the undercover video and audio recording of an informant meeting with a represented person under criminal investigation. The court held that the prosecutor violated the anti-contact rule by sending the informant to talk with the represented person. Concerned federal prosecutors in the Second Circuit petitioned for rehearing, which was denied. While a petition for hearing en banc was pending, the original panel issued another opinion on the matter, United States v. Hammad, 858 F.2d 834 (2d Cir. 1988), cert. denied, 498 U.S. 871 (1990).

In this final Hammad opinion, the Second Circuit held that the anti-contact rule does not restrict the government from using informants against represented persons in a pre-indictment, noncustodial situation. The Second Circuit reasoned that "a prosecutor is ‘authorized by law’ to employ legitimate investigative techniques in conducting or supervising criminal investigations, and the use of informants to gather evidence against a suspect will frequently fall within the ambit of such authorization." (858 F.2d at 840.) The court, however, found that the prosecutor’s use of a sham subpoena issued to the informant to help the informant surreptitiously gain the confidence of the represented suspect constituted misconduct.

Since Hammad, there has been a running battle among the Department of Justice (DOJ), the Conference of Chief Justices representing all 50 state supreme courts, the ABA, disciplinary bodies, and others. The DOJ has continued to push for a rule change or separate rules to permit ex parte contacts with represented persons by government lawyers. Congress entered the fray by passing the Citizen Protection Act, known as the McDade Amendment, in 1998. The McDade Amendment provides:

An attorney for the Government shall be subject to State laws and rules, and local Federal court rules, governing attorneys in each State where such attorney engages in that attorney’s duties, to the same extent and in the same manner as other attorneys in that State.

Most recently, the U.S. Senate tried to repeal the McDade Amendment as part of the anti-terrorism package passed in the aftermath of the September 11th attacks, but failed when the House of Representatives refused to include it in the Patriot Act of 2001. (28 U.S.C. § 530B(a).)

Current law. Some courts have espoused a categorical approach finding that the anti-contact rule does not apply in the investigative phase of a criminal case. It is triggered only with the commencement of adversarial proceedings. The Tenth Circuit, for example, in United States v. Ryans, 903 F.2d 731, 740 (1990), held that the anti-contact rule’s "proscriptions do not attach during the investigative process before the initiation of criminal

Following the lead of Hammad, though, most courts have rejected this categorical view and approached the application of the anti-contact rule in criminal investigations on a case-by-case basis. Uniformly these courts have held that prosecutorial use of an undercover agent or informant to communicate with someone known to be represented by counsel in the matter does not by itself violate the anti-contact rule.

Outside the undercover agent/informant context, the meaning of the rule continues to evolve as courts respond to the nuances of particular factual contexts. Two recent cases are illustrative. Both find the anti-contact rule applicable in the investigative context. But they differ on whether the rule was violated.

In State v. Miller, 600 N.W.2d 457(Minn.1999), the Minnesota Supreme Court dealt with the anti-contact rule in the context of a 40-minute police interview of the defendant in the case, a corporate employee, during the execution of a search warrant at the corporation’s place of business. The corporation’s lawyer, having been notified that a search warrant was being executed, contacted the police by telephone, asked that no statements be taken from employees, and specifically asked to speak to the defendant. The police, upon advice of the prosecutor, refused to stop the interview or notify the defendant that the lawyer wanted to speak with him.

Despite the Minnesota anti-contact rule’s use of the word "party" rather than "person," the court found that the rule applies prior to the filing of a formal charge and explicitly adopted a case-by-case approach to determining violations. The court found a violation and suppressed all statements taken after the lawyer asked the police to terminate the interview. The court interpreted the "authorized by law" exception to include "legitimate investigative processes" but not "a systematic isolation of the client from his attorney." ( Id. at 468.)

United States v. Talao, 222 F.3d 1133 (9th Cir. 2000), applied California’s version of the anti-contact rule, as in Miller, in the context of an interview of a corporate employee. After receiving a subpoena to testify before a grand jury investigating her employer for false statements regarding employee wages and hours, the corporation’s bookkeeper approached the prosecutor and said she was being pressured to give false testimony and did not want to be represented by the corporation’s lawyer. The prosecutor informed the bookkeeper of her right to be represented by an attorney, but she declined. She then provided the prosecutor and case agent with information that she later recounted to the grand jury, including information about the possible destruction of corporate documents. Again, as in Miller, the corporation’s lawyer became aware of the interview and demanded to speak with the bookkeeper. The prosecutor in Talao, however, unlike the prosecutor in Miller, informed the bookkeeper of the lawyer’s presence and desire to speak with her, but the bookkeeper declined to speak with the lawyer.

After an indictment was returned against the corporation, it moved for dismissal on the basis of an anti-contact rule violation. The trial judge denied the motion, but found that the prosecutor had violated the anti-contact rule and stated that he would inform the jurors of the prosecutor’s misconduct and instruct them to take it into account in assessing the bookkeeper’s credibility.

The Ninth Circuit reversed. Endorsing the Hammad case-by-case approach, the court stated that the anti-contact rule governed under circumstances "involving fully defined adversarial roles, impending grand jury proceedings, and awareness on the part of the responsible government actors" that the corporation was represented by counsel. ( Id. at 1139.) Analogizing to the crime-fraud exception to the attorney-client privilege, the court stated that "it would be a perversion of the [anti-contact rule] to extend it to protect corporate officers who would suborn perjury by their employees." ( Id. at 1140.) The court emphasized the employee’s initiation of the communication and the prosecutor’s advising the employee of her right to counsel in finding no violation of the anti-contact rule. ( Id.)

Defense counsel

Because recent controversy concerning the anti-contact rule has focused on prosecutors, some might conclude that the anti-contact rule has significance in criminal investigations only for prosecutors. Defense lawyers, though, should give thought to how the rule may affect them. Nothing in the text or comments to Rule 4.2 suggests that criminal defense lawyers are exempt from the rule. Indeed, ABA Formal Opinion 95–396 (1995) specifically states that "[d]efense counsel in criminal cases of course are . . . subject to the provisions of Rule 4.2."

Use of the word "party" in some versions of the anti-contact rule might suggest that criminal defense lawyers are free to contact another lawyer’s client if he or she has not been charged and thereby made a party to a criminal case. But courts, such as those in Miller and Talao, have rejected such a limitation when applying the rule to prosecutors and have construed the term "party" to encompass represented persons who have not been charged. And some jurisdictions, following the lead of the Model Rules as noted above, have replaced the word party with person in their anti-contact rule.

A victim or complainant in a criminal case is normally not protected by the anti-contact rule based on the relationship with the prosecutor because the prosecutor represents the government and not the victim or complainant. A Michigan ethics opinion, for example, approves of a defense lawyer contacting a complaining witness not represented by his or her own lawyer provided the lawyer makes it clear that the lawyer represents the defendant, does not give advice to the witness, and does not "request the witness to testify falsely or to refrain from giving information to the prosecutor or offer any inducement that is prohibited by law." (Michigan Bar Comm. on Prof’l & Judicial Ethics, RI–302 (1997).)

Due to the scarcity of reported cases and ethics opinions involving defense attorneys contacting uncharged represented persons during defense investigations means that the precise contours of the anti-contact rule as applied to criminal defense lawyers have yet to be determined.


Although the anti-contact rule has a relatively low public profile, consequences for violating it can be severe. Failure to abide by the anti-contact rule may lead to disqualification, suppression, possible dismissal, or create grounds for appeal based on prosecutorial misconduct. Prosecutors and defense counsel may also face professional discipline for violating the anti-contact rule.

Prosecutors and defense counsel must know how the anti-contact rule in their jurisdiction is interpreted and applied. When in doubt, a prosecutor or defense lawyer may seek a clarifying court order, for example, to determine which employees of a represented corporation may be contacted. ( See In re Criminal Investigation of John Doe, Inc., et al., 194 F.R.D. 375 (D. Mass. 2000).) As with most ethics rules, proceeding cautiously when contacting or directing others to contact represented persons is usually the best policy.

Return to Table of Contents - Winterl 2002

Return to Criminal Justice magazine home page