Who Counts as Represented

Rule 4.2 forbids a lawyer acting for a client to communicate about the subject of the representation with “a person the lawyer knows to be represented by another lawyer in the matter,” unless that lawyer consents or law or a court order authorizes the contact.¹ The earlier sections examined the contacting lawyer: the capacity in which the lawyer acts, the subject of the conversation, and what the lawyer knows. This section examines the person at the other end of the line. The rule says only “person.” It never says whether that person must be a litigant, an individual, or an adversary, so each answer comes from reading that one word against a particular setting.

Five settings recur. The first asks whether protection begins only when suit is filed. The second and third ask which people count when the represented person is an organization, and whether the answer changes once an employee has left. The fourth takes up witnesses and class members, who test the boundary from opposite directions. The fifth takes up representation that has ended or was narrow from the start. In every setting the same two questions control: does this person have a lawyer in this matter, and does the contacting lawyer know it?

Persons, Not Just Parties

The Model Rule protects “a person the lawyer knows to be represented by another lawyer in the matter.”² It does not ask whether the person is a plaintiff or a defendant, whether a complaint has been filed, or whether the person's interests run against the lawyer's client. The Alabama rule at issue in Gaylard v. Homemakers of Montgomery, Inc. was narrower: it barred communication with “a party the lawyer knows to be represented by another lawyer in the matter.”³ Alabama is the exception, because most states now follow the Model Rules in substituting “person” for “party.”⁴ That one word decided Gaylard.

The facts are short. Before suing, the plaintiff's lawyer telephoned Ms. Taylor, the caregiver who had attended the plaintiff, and recorded the call without telling her.⁵ The employer moved to exclude the recording under Rule 4.2, and the circuit court sustained the objection. The Alabama Supreme Court reversed, holding that the lawyer could not have violated the rule.⁶

The court gave two reasons. The first was timing. The rule reached only communications with a “party,” and the employer was not one when the call was made, because the plaintiff had not yet filed an action against it; the rules, the court said, “do not require an attorney to immediately file an action at law before communicating with the person with whom the attorney's client has a dispute.”⁷ The court acknowledged a committee comment that would reach an agent or employee of a corporate party when the corporation's liability will rest on that person's actions, yet it treated the employer's lack of party status as controlling.⁸ The second reason was knowledge. Apart from party status, nothing showed that the plaintiff's lawyer knew, or had reason to think, that the employer had retained counsel. Ms. Taylor had objected at first that she should not talk without speaking to her supervisor, but she never said or implied that her employer had a lawyer.⁹

Only the first reason depends on the word “party.” The knowledge ground survives any rewording of the rule, and earlier sections developed it. The timing ground is the one the Model Rule's wording removed.

The drafters changed “party” to “person” to make clear that the prohibition applies to contact before a lawsuit is filed. They went further. The rule is not limited to litigation and applies equally to represented parties to a deal, and it does not require that the persons' positions or interests be adverse, so it reaches a represented co-party or another person aligned in interest in a transactional matter.¹⁰ The purposes the rule serves explain why neither limit survives. The rule keeps one lawyer from interfering with another lawyer's relationship with a client, guards against inadvertent or uncounseled disclosure of confidences by a layperson whose lawyer is absent, and prevents a lawyer from obtaining a statement that would count as an admission.¹¹ Each danger arises as soon as a person has a lawyer in the matter, with or without a complaint, and an aligned co-party can blurt out a concession as readily as an opponent can.

Under the Model Rule, then, the absence of a lawsuit would not by itself have rescued the call in Gaylard. The outcome would turn on the other elements: the knowledge element, which the court found missing, and the question whether a caregiver is among the employees whose contact counts as contact with the organization, which the next subsection takes up.¹²

A stem that stresses the absence of any lawsuit invites the conclusion that the call is free. Under the Model Rule the filing date does not enter the analysis; what matters is whether the lawyer knows the person has counsel in the matter.

Set that call beside the one in Gaylard. In both, no complaint had been filed. The difference lies in the text applied: Alabama's rule reached only a party, so timing took the call outside it, while the Model Rule reaches a person, so timing carries no weight.

Organizational Employees

When the represented person is a corporation, a hospital, or an agency, the rule's text leaves the obvious next question open: which of the people who work there are protected? An organization acts only through human beings, so the question is which of them stand behind its lawyer. The Massachusetts Supreme Judicial Court confronted the gap in Messing, Rudavsky & Weliky, P.C. v. President and Fellows of Harvard College. It observed that the rule's language leaves open just which employees of an entity are out of bounds, that lawyers had little guidance, and that attorneys apprehensive about crossing ethical lines might refrain from contacting any employee of a represented organization.¹⁵

Both extreme readings defeat the rule. If only the organization's top officer were protected, an adverse lawyer could bypass counsel by speaking to the manager who actually runs the matter. If every employee were protected, the organization would be sealed off from informal fact-finding, since it acts only through employees and nearly everyone who saw the events works for it. The tests below try to shield the people who effectively are the organization on the matter and to leave the rest free to be interviewed.

Comment 7 to the Model Rule supplies one such test. It sorts an organization's constituents into three groups who are off limits without the consent of the organization's lawyer:¹⁶

Constituents who supervise, direct, or regularly consult with the organization's lawyer about the matter.
Constituents who can make decisions on behalf of the organization with respect to the matter.
Constituents whose acts or omissions in the matter can be imputed to the organization for purposes of vicarious liability.

Each group maps onto a danger the rule guards against. Constituents who consult with the organization's lawyer hold the confidences and strategy the rule protects. Constituents who decide for the organization are, in effect, the organization speaking, so a statement from them is a statement from the represented party. Constituents whose acts can be imputed have a different claim to protection: they may have no authority and no contact with counsel, but their conduct is the factual core of the claim, so their own account of it shapes the organization's exposure.

Hypothetical Illustration

A scaffold collapses at a building site and injures a worker, whose lawyer knows that the general contractor has retained counsel on the claim. The lawyer wants to interview two of the contractor's employees. The first is the vice president who supervised the project and has been meeting regularly with the contractor's lawyer about the claim. The second is a bookkeeper who was eating lunch in the yard and watched the collapse; she has no managerial duties, cannot decide anything for the company, and nothing she did is part of the claim. The vice president falls in the first category and probably the second, so a call to him without the contractor's lawyer's consent violates the rule. The bookkeeper falls in none of the categories, so the lawyer may interview her under the three-category approach and under narrower ones, although a court that forbids contact with practically all employees would treat her as off limits too.

Courts do not agree on how far the protected class extends. Their decisions fall along a range:

The narrow approach applies the rule only after litigation has begun and only to managerial employees or others high enough in the organization to speak for it, as in Weider Sports Equipment Co. v. Fitness First, Inc., and a District of Columbia ethics opinion likewise reaches only non-party employees with authority to bind the employer.¹⁷
The broad approach forbids contact with practically all employees of the represented organization, as in In re Air Crash Disaster near Roselawn, Indiana.¹⁸
The middle approach forbids contact only with defined categories of employees, which is the course the Massachusetts court took in Messing.¹⁹

The choice among these decides the case an exam is most likely to pose: a non-managerial employee whose own conduct is at issue. Under the middle approach that employee is protected through the third category. Under the narrow approach, an employee with no authority to speak for the organization is not.

Ms. Taylor, the caregiver in Gaylard, is that employee. The majority never reached her status, but a dissent stressed that the lawyer telephoned her knowing that her acts or omissions could be imputed to her employer and that her statements could be admissions. The dissent quoted Alabama's comment, which extended the ban to “any other person whose act or omission in connection with that matter may be imputed to the organization for purposes of civil or criminal liability or whose statement may constitute an admission on the part of the organization.”²⁰ Secondary commentary on the case concludes that once the lawyer knew of the company's counsel, contact with certain employees, “(probably) Ms. Taylor,” would require the company lawyer's permission.²¹

The “admission” language in that Alabama comment has a history. Older versions of the rule referred to employees whose statements might be “admissions” of the organization, and state supreme courts, among them Palmer v. Pioneer Inn and Messing, had to clarify what the phrase meant. The word is not part of the current Model Rule.²² A stem that uses the older wording is a signal to check which version the jurisdiction follows.

Because the categories turn on an employee's real role, the facts of the contact matter. In Weider the court noted that an earlier decision had allowed contact with low-level employees even after litigation began, and that its own case was further removed, because suit had not yet commenced when the agents contacted the employee, and the employee was not a party.²³

The breadth of the protected class creates a dilemma before suit. A lawyer who must investigate before filing a complaint may find that the only people who know what happened are the opposing organization's employees. If those employees are protected, the lawyer cannot call them without the organization's lawyer's permission, while the organization's lawyer, once the organization learns of the claim, remains free to call the claimant's family and friends.²⁴ The practical lesson is blunt: before interviewing employees of a prospective institutional party, check the applicable rules of professional conduct and the judicial gloss on them.²⁵

Exam Analysis Steps

Because the question is which person at the organization the lawyer wants to reach, the analysis runs employee by employee:

Confirm that the organization is represented in the matter and that the contacting lawyer knows it, since the earlier elements of the rule still have to be met.
Identify the employee's actual role in the matter, looking to what the person does and did rather than to a job title.
Ask whether the employee supervises or consults with the organization's lawyer on the matter, can make decisions for the organization on it, or performed acts or omissions in it that can be imputed to the organization.
Identify which approach the jurisdiction follows, and apply the three categories as the default when the facts do not say.
If the employee is protected, route the contact through the organization's lawyer unless that lawyer consents or the contact is otherwise authorized.

Candidates tend to err in opposite directions: treating every employee of a represented company as off limits, or treating only executives as protected and so overlooking the third category.

Former Employees

Most courts have held that Rule 4.2 does not prohibit interviews with former employees of a represented organization.²⁸ Comment 7 agrees, permitting ex parte contact with former constituents of an entity.²⁹ The topic divides into two questions that should be kept apart. The first is whether Rule 4.2 applies to the contact, and generally it does not. The second is whether some other rule still limits what the interviewing lawyer may do, and it does.

The first answer fits the rule's purposes. The employee categories sort current constituents by the role each plays in the matter: consulting with the organization's lawyer, deciding for the organization, or acting in a way the law attributes to it. A person who has left holds none of those roles. The practical stakes are real, because disgruntled ex-employees are often a fruitful source of information against their erstwhile employer, and a rule that put them off limits would let the organization wall off the people most willing to speak candidly.³⁰ Had Ms. Taylor already left her employer's service when the lawyer called her in Gaylard, the recording would have been admissible for a reason the court did not need to reach: she would have been a former employee.³¹ The permission has one limit that comes from the rule's own words. A former employee who has hired a personal lawyer in the matter is a person the contacting lawyer knows to be represented in it, so that contact must go through the personal lawyer.³²

The second question exists because freedom from Rule 4.2 does not free the lawyer from every other restraint. Comment 7 itself states that in communicating with a current or former constituent of an organization, a lawyer “must not use methods of obtaining evidence that violate the legal rights of the organization,” that is, privileged or other protected information.³³ Rule 4.4(a) supplies the general prohibition, and its comment counts “unwarranted intrusions into privileged relationships, such as the client-lawyer relationship” among the forbidden methods. The result is that a lawyer may not attempt to obtain from an entity's employees or agents any information protected by the attorney-client privilege or the work-product doctrine.³⁴

The protection those rules enforce is the organization's. A former employee's willingness to talk therefore changes nothing, because she has no right to give away. The same ethical line appears in the discovery setting, where a request crosses it when it plainly seeks irrelevant, privileged, or otherwise protected information.³⁵ An informal interview is no safer than a formal request: a question that plainly seeks privileged communications crosses the line whether it appears in a discovery request or comes up in conversation.

On the permitted side of that line stand questions about what the former employee saw, did, and heard from coworkers, which are ordinary fact-gathering. On the forbidden side stand questions that seek the substance of her conversations with the organization's lawyers, and any request for the organization's privileged documents.

The two halves of that interview fall under different rules, so the analysis runs in two stages.

The usual error is to stop after step one and conclude that a former employee is fair game for every question, which overlooks that the organization's rights continue to bind the interviewing lawyer.

Witnesses and Class Members

Rule 4.2 reaches only “a person the lawyer knows to be represented by another lawyer in the matter,” so protection depends on whether the person has counsel in the matter and not on the role the person played in the dispute.³⁸ Witnesses and class representatives test that principle from opposite directions. A person who merely saw the events usually has no lawyer in the matter and falls outside the rule. A class representative who has a lawyer falls inside it.

The witness is the simpler case. The rule exists to keep an adverse lawyer from reaching around counsel to a person who has hired one, whether to interfere in the relationship, to draw out uncounseled disclosures, or to obtain an admission.³⁹ A witness with no lawyer in the matter has no counsel to go around, and walling such a person off would block every lawyer from learning what happened, since the facts of most disputes live in the heads of people who are not parties. A witness's usefulness to the other side is therefore irrelevant. A witness whose account will devastate the opposing client is as free to be interviewed as one whose account is neutral, because the test asks about representation and takes no account of whom the testimony favors.

The label “mere witness” therefore describes a person with no role and no counsel. An employee whose own acts or omissions can be imputed to a represented organization is not a mere witness, as the previous subsection showed, and a witness who retains personal counsel in the matter becomes a represented person at that moment.

The class action presents the opposite pole, and the facts of Berkson v. Gogo show it clearly. Welsh's lawyer notified Gogo that Welsh intended to bring a putative class action. After that notice, Gogo allegedly communicated directly with Welsh and delivered a check to him.⁴⁰ The court held that the violation of the no-contact rule made the check irrelevant to the class litigation, so Welsh kept a live controversy that the payment had not mooted.⁴¹ The mechanism explains the result. A defendant who can settle individually with the named plaintiff, outside the presence of the plaintiff's lawyer, may end the class action before it begins, and the lawyer's presence is exactly what would expose the maneuver. The court said that picking off named plaintiffs with a payout before a class certification motion is not acceptable when it has the effect of preventing a viable class action from proceeding.⁴²

How far this protection reaches past the named plaintiff is less settled. The rule asks whether the person is represented “in the matter,” and its text does not say whether class counsel's representation of a class makes each absent member a represented person. A lawyer unsure of that status has a safe route, because the rule itself lists a court order as authorization,⁴³ and an order answers the question in advance.

An exam stem may emphasize that no class has been certified, or that the consumer is only the proposed lead plaintiff. Neither fact removes protection from a person who has a lawyer in the matter, and a direct settlement offer to that person after notice from the lawyer is the contact the Gogo court condemned. A stem that describes a bystander as a key witness for the other side tests the other pole, and there the lawyer may talk to the witness.

Ended or Limited Representation

Rule 4.2 protects a person the lawyer “knows to be represented,” a standard that looks to the facts when the lawyer speaks.⁴⁷ The communication bar is raised only when the lawyer has actual knowledge of the fact of the representation.⁴⁸ Knowledge may be inferred from the circumstances, however, and a lawyer cannot evade the requirement of obtaining the consent of counsel by closing eyes to the obvious.⁴⁹ Representation can change while a matter is pending. A client may discharge a lawyer, or may hire a lawyer for one defined task and handle everything else personally. Each change raises the same question: does the contacting lawyer still know the person to be represented?

Start with a person who announces that counsel no longer matters. Comment 3 states that the rule applies even though the represented person initiates or consents to the communication, and it warns lawyers that they must silence the represented person who calls, no matter how badly that person wants to talk. The represented person's lawyer is the sole source of any legitimate consent, and purported consent extracted by the communicating lawyer is void.⁵⁰ A person's own say-so that counsel is no longer needed therefore cannot end the protection. In Attorney Disciplinary Board v. Box, a lawyer defended direct dealings with a woman whose own lawyer had announced the representation by pointing to her insistence that she could speak to whomever she chose, and the Iowa Supreme Court rejected the defense.⁵¹ A person may in fact discharge a lawyer and proceed unrepresented. A lawyer who knows of earlier representation and sees no clear sign that it ended should confirm its status with the person's lawyer, ordinarily by a call, before discussing the matter. Rule 4.2 contains no separate verification clause; the practice follows from the knowledge standard and from the rule on consent.

The verification burden sits on the contacting lawyer for a reason visible in who can do what. If a casual remark could end the protection, the lawyer asking the questions could draw the remark out, and the rule would defeat itself. That is why consent extracted by the communicating lawyer is void.⁵² Verification costs the lawyer one telephone call, while the represented person is usually a layperson who cannot be expected to know the rule or the exact reach of the lawyer's engagement.

The dance-lesson sale case Syester v. Banta is no guide here. Its excerpted facts record that the defendants sold the plaintiff 3,222 hours of instruction for $21,020.50 by May 2, 1955,⁵³ and 4,057 hours in all, at a cost that the defendants' own exhibit puts at $29,174.30 after refunds and credits.⁵⁴ Those are facts about a sale and say nothing about when a representation ends.

The second situation is a representation that was narrow from the start. A lawyer may provide limited services to an otherwise self-represented person if the client gives informed consent and the scope of the engagement is clearly defined, as Rule 1.2(c) contemplates. A person who receives only limited representation is treated as unrepresented for the no-contact rule unless the opposing lawyer knows of the arrangement or has received a written notice of appearance or other written notice from the person's lawyer.⁵⁵ Some states say so expressly in their no-contact rules, so check the local wording.

Two mechanisms explain that rule's shape. The rule protects a person through that person's lawyer, and someone who has counsel for one task has no lawyer positioned to guard the rest of the matter. A private engagement letter, moreover, is invisible to the opposing lawyer, who cannot be said to go around anyone without knowing the arrangement exists. The notice requirement lines up the lawyer's knowledge with the protection, and it is the knowledge element taught earlier applied to a narrow engagement.

The hinge in that pair is the notice, which turns a private arrangement into something the contacting lawyer knows. A stem that calls a litigant self-represented but mentions that a lawyer helped with one filing is testing whether the opposing lawyer knew of, or was told about, that help.

Exam Analysis Steps

Because an ended or narrowed representation matters only through the contacting lawyer's knowledge, the analysis is a series of status questions:

Identify whether the contacting lawyer knows that the person has been represented in the matter, through a letter from counsel, a statement by the person, or the surrounding circumstances.
Ask whether anything clearly shows that the representation has ended, remembering that the person's own remark that counsel no longer matters does not count.
If the status is doubtful, confirm it with the person's lawyer, ordinarily by a telephone call, before discussing the matter.
If the person has counsel only for a defined task, ask whether the contacting lawyer knows of the arrangement or has received written notice of the lawyer's appearance, and treat the person as unrepresented for this rule only if neither is true.

Step two does the most work on exam facts, because stems often dress up a remark the contacting lawyer drew out as though it were a discharge.