Concept
Contact Through Others
Communications with represented persons
Contact Through Others
Rule 4.2 bars a lawyer who is representing a client from communicating about the subject of the representation with a person the lawyer knows to be represented in the matter, unless that person's lawyer consents or the law or a court order authorizes the contact.1 The earlier sections tested that sentence on a lawyer who speaks to the represented person directly. But the rule's verb is communicate, and a lawyer who wants what a represented person knows has other routes. The lawyer can send the client with questions, have a paralegal place the call, or ask an investigator to arrange a conversation and report back. Whether those routes are open depends on whether the rule tracks the lawyer's words or the lawyer's purpose.
The rule tracks purpose. It keeps a lawyer from interfering with another lawyer's relationship with a client, guards against inadvertent or uncounseled disclosure of confidential information by a layperson whose lawyer is absent, and prevents the lawyer from obtaining a statement that would be an admission under the evidence rules.2 Each of those harms occurs whoever carries the message, since an admission drawn out by an agent helps the lawyer's client exactly as much as one the lawyer drew out personally. Comment 4 accordingly warns that a lawyer may not make a prohibited communication through “the acts of another,”3 and Rule 8.4(a) separately provides that a lawyer may not violate the rules “through the acts of another.”4
The principle has a limit. Rule 4.2 restrains lawyers, not parties, and the same comment adds that parties to a matter may communicate directly with each other.5 The question in each setting is therefore whether the person making the contact acts for himself or serves as the lawyer's instrument. The client is the first test, and the investigator is the second.
Client-to-Client Contact
On its face, a conversation between two parties who each have counsel is not a lawyer's communication at all, and the rule leaves the clients free to talk. Comment 4 says so directly: “[p]arties to a matter may communicate directly with each other, and a lawyer is not prohibited from advising a client concerning a communication that the client is legally entitled to make.”6 The rule exists to protect the represented person from “the supposed imbalance of legal skill and acumen between the lawyer and the party litigant,”7 and two laypeople talking raise no such imbalance. The permission covers advice as well. A client ordinarily may speak with the opposing party about the merits and the wisdom of settlement, and may receive the lawyer's counsel on those subjects beforehand, even for the specific purpose of approaching the other side.8
The permission has an edge, because the same comment bars a lawyer from making a prohibited communication through “the acts of another.” Where that edge lies has been a subject of controversy.9 A limit must exist, because otherwise the rule would be empty: a lawyer barred from calling a represented opponent could hand the client a script and send the client in the lawyer's place. The disagreement is over how much lawyer involvement is too much.
The permissive view appears in ABA Formal Opinion 11-461. Under it, the lawyer may go beyond general advice and draft a document that the client asks the other party to sign. The lawyer must advise the client to encourage the other party to consult counsel before entering an obligation, and any draft agreement must carry conspicuous language warning the other party to consult a lawyer before signing.10
The stricter view draws the line earlier. Its holders insist that a lawyer crosses it by scripting the conversation that will unfold between the clients or by preparing legal documents for the client to present for signature.11 Responding directly to the ABA opinion, the Minnesota Office of Lawyers Professional Responsibility stated that a lawyer “may not script any such communication [between clients] or draft an agreement to be presented to the adverse person, even at the client's request.”12 A treatise author on Minnesota ethics argues that Opinion 11-461 “is so permissive that it would subvert the basic purpose of Rule 4.2, viz. protecting represented clients, especially vulnerable clients, from other lawyers' overreaching.”13 Both camps allow the clients to talk and the lawyer to advise. They part over whether a lawyer's authorship of the words or the documents carries the lawyer into the room.
The other party who starts a conversation violates no ethical constraint, because the contact is not otherwise improper.14 An exam stem that says the client acted on the lawyer's instructions therefore calls for two questions: how much the lawyer did (general advice, a script, or a drafted agreement) and which approach the jurisdiction follows. Where the stem names no jurisdiction, the best answer states the permissive and stricter positions side by side and does not present the ABA opinion as the only view.
Investigators and Agents
The setting where this question arises most often is the pre-filing investigation. A lawyer who has not yet sued wants to learn who caused the harm and sends an investigator to talk with witnesses, and one of those witnesses may be a person the rule protects. On their face the Model Rules apply only to lawyers, but two provisions deal with a lawyer's responsibility for what others do. Rule 5.3(c)(1) makes a lawyer responsible for the conduct of another person if the lawyer orders the conduct, and Rule 8.4(a) states that a lawyer may not violate the rules “through the acts of another.”17
Together the two provisions close the gap. Rule 5.3(c)(1) attributes the investigator's conduct to the lawyer who ordered it, and Rule 8.4(a) turns that attribution into a violation, because the lawyer has used another person's act to accomplish what Rule 4.2 forbids. The reason follows from the harms the no-contact rule guards against. An uncounseled disclosure or an admission arises whenever someone on the adverse side questions the person while counsel is absent, and the investigator's report hands the lawyer's client the same advantage a personal call would.18 If delegation cured the problem, the rule would bind only lawyers who insisted on making their own calls.
The delegation question decides only whose act the contact is, and every other element still runs against the lawyer. An exam stem that stresses that the investigator, and not the lawyer, dialed the number adds nothing when the lawyer gave the instruction. An investigator sent to interview a bystander who watched the collapse, and who has no lawyer and no tie to the subcontractor, raises no issue at all. The rule reaches only a person the lawyer knows to be represented.19
The rule also has an edge on the other side, because it governs lawyers. Rules 5.3(c)(1) and 8.4(a) reach a nonlawyer's conduct through a lawyer who orders it, so a nonlawyer acting with no lawyer behind the contact is not within them.20
A prosecutor's duties shift when the accused has no lawyer. Rule 3.8(b) and (c) require a prosecutor to take reasonable steps to facilitate the accused's access to counsel and forbid procuring a waiver of important pretrial rights from an unrepresented accused. Those provisions aim to prevent overreaching by trained advocates who carry the government's authority and to prevent interference with an actual or imminent attorney-client relationship.21 Rule 4.2 does not control that situation, because it protects only a person the lawyer knows to be represented.
The two investigation scenes differ in one fact, which is whether a lawyer stood behind the contact. The common exam error runs in both directions: concluding that a nonlawyer's contact always escapes the rule, and concluding that any contact by an investigator is automatically the lawyer's. The test is whether a lawyer ordered or directed it.