Non-Consentable Conflicts

Rule 1.7(b) lets a lawyer proceed despite a concurrent conflict only when four conditions are all met. Three of them cannot be satisfied by anything a client does:

the lawyer reasonably believes that the lawyer will be able to provide competent and diligent representation to each affected client;¹
the representation is not prohibited by law; and²
the representation does not involve the assertion of a claim by one client against another client represented by the lawyer in the same litigation or other proceeding before a tribunal.³

The fourth condition, informed consent confirmed in writing, is the client’s contribution and has its own section below. Check the three limits first. A conflict that fails any one of them is non-consentable, and the client’s willingness to sign a waiver does not change the result.

The limits follow from what a waiver is. A client may relax the loyalty a lawyer owes because the client defines the client’s own interests, but that justification needs two premises: the interest at risk must be the client’s to give up, and the lawyer must still be able to deliver real representation afterward. Each limit marks a place where one premise fails. Under paragraph (b)(1), the lawyer cannot honestly promise real service to each client. Under paragraphs (b)(2) and (b)(3), the interest at risk belongs to the law or to the tribunal, and a client cannot waive an interest that is not the client’s.

Two labels sort the results. A conflict is non-consentable by circumstance when the facts of the particular representation defeat the lawyer’s reasonable belief in competent and diligent service.⁴ It is non-consentable per se when the rule’s own text removes the consent option, as paragraphs (b)(2) and (b)(3) do.⁵ The consequence is the same either way: the lawyer must decline the representation or, if it has begun, withdraw, with the tribunal’s permission when the matter is before one.⁶

Reasonable Belief in Competence

Rule 1.7(b)(1) makes a concurrent conflict waivable only if the lawyer reasonably believes that the lawyer “will be able to provide competent and diligent representation to each affected client.”⁷ Of the four conditions, this is the one that looks only at the lawyer and the work the lawyer can actually do. The client’s willingness to proceed does not enter it.

Consent tells the client about a risk; it cannot shrink the risk. A waiver can be accepted despite a formal conflict, or despite conflicting influences that an outside observer would notice, but only if the lawyer who will handle the matter “remains sufficiently uncompromised by competing demands” that the client will still receive “effective and zealous advocacy.”⁸ The client decides whether to accept a known risk. The lawyer’s circumstances decide whether the risk is small enough to offer in the first place.

The standard has two halves. The lawyer must sincerely hold the belief, so a lawyer who privately doubts that both clients can be served well cannot proceed by pointing to a signed form. And the belief must be reasonable, which is judged from outside: would a disinterested lawyer, looking at the same facts, reach the same conclusion?

The outside measure is necessary because both people in the arrangement are poorly placed to judge it. The client cannot foresee how divided loyalties will play out in a negotiation or a trial, and a lawyer who wants the engagement has every incentive to persuade herself that all will be well. A purely subjective test would let the conflicted person certify her own freedom from the conflict. The disinterested-lawyer test denies her that room, so consent cannot be used to buy an impaired lawyer.

When the objective half fails, the conflict is non-consentable by circumstance. The facts that defeat the belief come in two groups. The clients’ interests may be too sharply opposed for one lawyer to advise each of them effectively, or the lawyer’s own situation may impair the work.⁹

The clients’ interests can also clash so sharply that no lawyer could serve both. The standard illustration is one lawyer acting for both buyer and seller in a complex commercial real estate transaction. The contract has many terms that allocate price and risk, and each term that protects one party leaves exposure with the other.¹⁰ A reasonable lawyer would doubt that one negotiator can obtain a deal serving both parties across that many contested terms, so the conflict is likely non-consentable even when both parties are eager to agree.¹¹ New Jersey goes further: in Baldasarre v. Butler, its supreme court treated simultaneous representation of buyers and sellers in commercial real estate transactions as a conflict that is per se non-consentable.¹²

At the other end of the range, the lawyer’s tie to the adverse party can be slight. A lawyer retained to pursue a claim against a social club of which she is a member, or against a corporation whose shares make up a minimal part of her investment portfolio, can reasonably conclude that these are inconsequential interests that will not influence her judgment, and after being informed of the risks the client may agree.¹³ Compare the tenant’s lawyer above. There the spouse owns the entire adverse company, so the outcome of the suit reaches the lawyer’s household directly. A sliver of a diversified portfolio gives the lawyer no comparable stake in how a case ends.

Exam stems often say that the lawyer personally believes all will be well and that the client agrees in writing, and then ask whether the lawyer may proceed. Those two facts answer other questions. The lawyer’s confidence speaks only to the sincerity half of the belief condition, and the client’s signature speaks only to the consent condition in Rule 1.7(b)(4). The mistake is to think the client’s blessing overcomes any conflict regardless of its effect on the quality of the representation. When the facts also show an impairment that a disinterested lawyer would see, the answer is no, however confident the lawyer and however eager the client.¹⁴

Sometimes the clients’ positions diverge so far that stepping aside serves everyone better than a waiver. In Williams v. Delaware, a lawyer who was appellate counsel for more than one client asked the Delaware Supreme Court for permission to withdraw from representing Williams, and the court granted the request: “Given his clients’ disparate legal arguments, [the lawyer’s] independent obligations to his clients may compromise the effectiveness of his assistance as appellate counsel for one or both clients, unless his motion to withdraw is granted.”¹⁵ The lesson is practical. When each client’s best argument undercuts the other’s, the lawyer cannot discharge both sets of duties at full strength, and withdrawal protects both clients.

Representation Prohibited by Law

Rule 1.7(b)(2) sets the second limit: a concurrent conflict is consentable only if “the representation is not prohibited by law.”¹⁷ The law in question lies outside the conduct rules, in statutes, regulations, and similar legal prohibitions. The standard example is a statutory or regulatory bar on former government lawyers who move into private practice and take clients.¹⁸ Because the condition asks only whether some other body of law forbids the work, it is a quick check, and when it applies it ends the analysis: nothing else in Rule 1.7(b) matters.

Consent is a private agreement, and a private agreement can reach only interests that belong to the people making it. A statute or regulation that bars a lawyer from an engagement protects someone else, usually the public or the integrity of a government function, so the client has no authority to give that protection up. If a client’s signature could dissolve such a bar, the conduct rules would become a way around every law restricting who may act as a lawyer in a matter. The condition closes that route.

Two illustrations show the condition at work. In some states, substantive law forbids one lawyer from representing more than one defendant in a capital murder case, so co-defendants who both want the same lawyer and both sign waivers still cannot have one.¹⁹ Federal law bars a former federal officer or employee, for two years, from representing a client on a matter within the person’s “official responsibility as such officer or employee,” and an agency that purported to consent could not change that, because administrators have no power to override a federal criminal statute.²⁰

The prohibition lasts only as long as the law reaches the work. Once the two years have run, or if the matter was never within the lawyer’s official responsibility, the statute no longer applies and Rule 1.7(b)(2) stops nothing, although the reasonable-belief, same-proceeding, and informed-consent conditions must still be met. An exam stem that piles up a signed client waiver, an agency’s approval, and the lawyer’s confidence is therefore asking one question: does some other law forbid the work? If it does, the rest is beside the point.

Claims in the Same Proceeding

Rule 1.7(b)(3) sets the third limit: the representation must not “involve the assertion of a claim by one client against another client represented by the lawyer in the same litigation or other proceeding before a tribunal.”²² The text has three working parts. A claim must actually be asserted, so the condition looks at what the lawyer would do in the proceeding, not at whether the clients’ interests happen to compete. The claimant and the target must both be clients of the same lawyer. And the claim must arise in a single litigation or other proceeding before a tribunal. When all three are present, the conflict is non-consentable per se: the rule itself removes the consent option, whatever arrangement the lawyer and the clients propose.²³

The reason differs from the reasons behind the other two limits. An adversarial proceeding works because someone loyal to each side develops that side’s position vigorously, and the tribunal depends on the contest to reach a sound result. Comment 17 to Rule 1.7 rests the limit on the institutional interest in “vigorous development of each client’s position when the clients are aligned directly against each other in the same litigation or other proceeding before a tribunal.”²⁴ A lawyer who tried to serve both sides would resemble a person playing both white and black in a chess game: the game could not function, and the player would begin to favor one color, even if only subconsciously.²⁵ The injury falls on the process as well as on the clients. Because the interest belongs to the system, the clients cannot give it up, which is the same logic that kept a client from waiving a statutory bar under Rule 1.7(b)(2).

Applying the condition requires attention to the shape of the proceeding. Comment 17 says that “[w]hether clients are aligned directly against each other within the meaning of this paragraph requires examination of the context of the proceeding.”²⁶ The plain case is one lawyer acting for both plaintiff and defendant in a single lawsuit, which is never permitted. The harder case involves parties who sit on the same side of the caption. When co-defendants direct claims against each other, such as cross-claims, the same lawyer may not represent both of them, at least with respect to those claims.²⁷ What decides the question is whether one client’s claim runs against the other, whatever the caption calls them.

An exam stem that places two defendants under one lawyer is therefore not automatically a Rule 1.7(b)(3) problem. It becomes one at the moment a claim runs from one client to the other, and until then the ordinary conditions of paragraph (b) govern.

Direct adversity and the same-proceeding bar overlap without being identical, and the gap between them is where students most often err. Suppose a lawyer gives general business advice to Earl’s Rooter, a company that clears tree roots from drain pipes, and a motorist injured by one of the company’s trucks asks the lawyer to take a personal-injury case against it. The representation is directly adverse to a current client under Rule 1.7(a)(1), and it makes no difference that the two matters are unrelated.²⁸ The conflict is nonetheless consentable, because the lawyer does not represent Earl’s Rooter in the personal-injury litigation, so no client is asserting a claim against another client of the lawyer in that proceeding.²⁹ It must still clear the remaining conditions: a reasonable belief in competent and diligent representation, and informed consent, confirmed in writing, from each affected client. Only if the lawyer also defended Earl’s Rooter in the injury suit would the bar apply. A negotiation between a buyer and a seller shows the inverse. No tribunal is involved, so paragraph (b)(3) is silent, yet the reasonable-belief condition may still make the conflict non-consentable.³⁰

Some states word the condition differently. A Florida court quoting that state’s Rule 4-1.7 in Furman v. Furman shows a bar on “the assertion of a position adverse to another client when the lawyer represents both clients in the same proceeding before a tribunal,” and its paragraph (b)(4) accepts informed consent that is either “confirmed in writing or clearly stated on the record at a hearing.”³¹ The Florida text speaks of an adverse position rather than a claim and adds an on-the-record route to consent. Those are variations to recognize if a stem signals that state’s law. Otherwise the Model Rule controls: the bar is triggered by the assertion of a claim, and consent must be confirmed in writing.³² A conflict that clears all three limits is consentable in principle, and what remains is the client’s informed consent, the subject of the next section.