When the Client Persists

Rule 1.2(d) governs the lawyer’s own conduct: a lawyer “shall not counsel a client to engage, or assist a client, in conduct that the lawyer knows is criminal or fraudulent.”¹ The earlier parts of this chapter showed what that sentence forbids and what it leaves protected. Following it is harder when the client rejects the lawyer’s answer. A client told that a plan is unlawful may drop it, but may also reword it, hand the lawyer a smaller piece of it, or proceed and expect the lawyer to keep working. This unit follows the lawyer from the moment the problem is plain and the client does not change course.

The lawyer’s duties depend on when the lawyer learns of the wrongdoing and on how the client answers an objection, and they fall into a sequence. A lawyer who knows at the outset that the client will use the lawyer’s services for a crime or fraud must refuse the engagement. A lawyer who learns of the problem mid-representation must first try to talk the client out of it, and must then avoid furthering the conduct. If the client persists, the lawyer may have to leave, and the grounds for leaving come in a mandatory form and a permissive form. Rule 1.16 supplies those grounds; because paragraph lettering varies among versions and states, this unit identifies each ground by what it says. The sequence ends with its best-known application, the client who threatens to lie under oath.

Each step is the least drastic measure that stops the lawyer’s contribution to the wrong. Refusing an engagement costs the client nothing the client has come to rely on. A conversation can end the problem without ending the relationship. Withdrawal waits its turn because a lawyer who leaves mid-matter can strand the client. Confidentiality runs through every step: the duty to protect what the client has disclosed continues while the lawyer declines to take part. The lawyer therefore cannot solve the problem by exposing the plan, and cannot solve it by staying silent and working on. The task is to stop contributing to the wrong while keeping the client’s confidences, and it begins at the door.

Declining the Engagement

Rule 1.16(a)(1) turns the ban on assisting into a ban on accepting: a lawyer shall not represent a client if “the representation will result in violation of the rules of professional conduct or other law.”² The chain is short. Rule 1.2(d) forbids assisting conduct the lawyer knows is criminal or fraudulent.³ A lawyer who accepts a client while knowing that the client means to use the lawyer’s services for that conduct will perform the assistance, so the representation “will result in” a violation of Rule 1.2(d). The obligation therefore arises at the door, before the lawyer does any work.

Refusal comes first because it is the one move that guarantees the lawyer contributes nothing. A lawyer who accepts with knowledge has no innocent first task; every piece of work serves a purpose the lawyer already understands, and counts as assistance from the first hour. The same rule covers a lawyer already at work, since it directs withdrawal “where representation has commenced.”⁴ At intake, though, nothing has started, no client has come to rely on the lawyer, and nothing needs unwinding.

Does a prospective client’s admission of something unlawful oblige the lawyer to turn the matter away? That depends on whether the wrong lies behind the client or ahead of the client.

The line falls between past and future because the two clients put the lawyer in different relations to the wrong. A completed fraud exists whether or not the lawyer ever answers the telephone. The accused needs someone to test the charge, explain the consequences, and negotiate, and a lawyer who does that work responds to the wrong without bringing it about. A planned fraud differs at every point. It has not happened, the client has come to the lawyer because the lawyer’s skill is part of how it will happen, and each document the lawyer prepares becomes a step toward carrying it out. The same contrast returns in the confidentiality rules and in the crime-fraud exception later in this chapter.

Calling a wrong “past” does not shield a request that reaches into the future. A client charged with defrauding investors last year may also ask the lawyer to move the remaining proceeds where prosecutors cannot find them. That request concerns a continuing effort, not the completed fraud. The test is what the lawyer’s services will be used for. The comment to Rule 1.2 states the idea for a course of action already under way: the lawyer “is required to avoid furthering the purpose, for example, by suggesting how it might be concealed.”⁵ Defense of the old charge stays proper; the concealment work does not.

Timing governs everything here. The same lawyer, client, and plan produce different duties depending on when the lawyer comes to know. Knowledge before the engagement calls for refusal; knowledge acquired later calls for the warning, avoidance, and withdrawal duties in the sections that follow. At intake, the inquiry runs in order.

Exam Analysis Steps

Because the duty to decline depends on what the work is for and on what the lawyer knows when deciding, the intake decision is best made by taking those two questions in sequence.

Ask what the prospective client wants the lawyer’s services to accomplish, and whether the wrong involved is already complete or still lies ahead.
If the wrong lies ahead, ask whether the lawyer actually knows, from what the client has said or shown, that the services would be used to commit it. Suspicion alone does not satisfy the knowledge requirement of Rule 1.2(d).⁶
If the lawyer does know, decline the engagement, because accepting it would make the representation result in a violation of the rules or other law.
If the matter concerns only a completed wrong, the lawyer may accept it, and must keep watching whether the client’s later requests drift toward a continuing or future wrong.

The fourth step keeps the second client in the hypothetical from becoming a trap. Taking her defense creates no problem, but a lawyer who later receives a request to hide the fruits of the old fraud has acquired knowledge during the representation, and that places the lawyer on the later part of the timeline.

Warning and Dissuading

When a lawyer knows that a client expects help the rules do not allow, Rule 1.4(a)(5) directs the lawyer to “consult with the client about any relevant limitation on the lawyer’s conduct when the lawyer knows that the client expects assistance not permitted by the Rules of Professional Conduct or other law.”⁸ The duty supplies the communication step that Rule 1.2(d) omits. Rule 1.2(d) says what the lawyer may not do; Rule 1.4(a)(5) requires the lawyer to say so. The trigger reaches beyond crime and fraud, because the words “not permitted by the Rules of Professional Conduct or other law” cover any assistance the lawyer may not give.

The duty exists because silence communicates. A client who hands over a plan and watches the lawyer start work reasonably reads that work as approval. Each document the lawyer produces then looks like confirmation that the plan is lawful, and the client may proceed more boldly because of it. A stated limit removes the false signal, and it gives the client a chance to change course before the lawyer is entangled, when changing course costs least. A lawyer who quietly performs only the lawful work, or declines the improper task without saying why, keeps the lawyer’s own hands clean but leaves the client guessing. A client who does not know the limits can rephrase the request, carry it to another lawyer, or conclude that the lawyer will eventually come around.

The two failures belong to different rules, and an exam answer should keep them apart. The drafting violates Rule 1.2(d). The silence is a separate lapse: a lawyer who quietly refuses to draft the deed without telling the client why has avoided the assistance but has not performed the consultation. Examinees who stop at “the lawyer must not assist” skip the step that tells the client the lawyer cannot be counted on.

Consultation tells the client where the lawyer stands; dissuasion tries to move the client. In Nix v. Whiteside, a case about a defendant who threatened to testify falsely, the Supreme Court described the profession’s baseline: “It is universally agreed that at a minimum the attorney’s first duty when confronted with a proposal for perjurious testimony is to attempt to dissuade the client from the unlawful course of conduct.”⁹ The Court spoke of perjury, which this unit treats separately below, but the word “first” fixes dissuasion as the opening move, and the reasoning carries to other unlawful proposals.

Dissuasion comes first because it does the most good at the least cost. A lawyer who persuades a client to drop a plan stops the wrong at its source and discloses nothing, so the client’s confidences stay protected; the lawyer may not reveal a client’s wrongdoing except where Rule 1.6 permits.¹⁰ The relationship also survives, which keeps the lawyer in a position to steer the client toward lawful ways of reaching the same goal. Withdrawal ends the relationship and may send the plan to another lawyer, and disclosure damages the confidentiality on which candid consultation depends. Both remain available in the circumstances later sections describe, but neither does what a successful conversation does.

The lawyer’s tool is the one the earlier safe harbor protects. Rule 1.2(d) expressly permits a lawyer to “discuss the legal consequences of any proposed course of conduct with a client,” which lets the lawyer lay out what the client faces if the plan goes forward.¹¹ The comment requires an honest opinion about the actual consequences that appear likely to result, so the account is complete and not comforting, and it adds that a client’s use of advice in a criminal or fraudulent course of action does not, of itself, make the lawyer a party to it.¹² A lawyer who warns honestly and is ignored has therefore not joined the plan. The account must remain analysis, though. A lawyer who adds a hint about how the client might avoid detection has moved from dissuading to the facilitation condemned earlier.

Some states write the dissuasion norm into a duty with a defined trigger. The Texas rules, as quoted in In re Enron Corp. Securities, Derivative & ERISA Litigation, provide that when a lawyer has confidential information clearly establishing that a client is likely to commit a criminal or fraudulent act likely to result in substantial injury to the financial interests or property of another, the lawyer “shall promptly make reasonable efforts under the circumstances to dissuade the client from committing the crime or fraud.”¹³ The court also described the Texas scheme as requiring disclosure adverse to the client only where the lawyer’s information clearly establishes a likelihood of death or substantial bodily harm, and as otherwise making the lawyer’s first obligation an attempt to dissuade.¹⁴

The Texas provision illustrates variation and is not a rule to memorize. Under the Model Rules, an answer rests on the consultation duty in Rule 1.4(a)(5) and on the baseline that the lawyer first tries to persuade. A problem that names Texas calls for the added duty and its effort standard: the lawyer must make reasonable efforts and need not succeed. When the effort fails and the client goes forward, the lawyer’s obligations shift to avoiding further assistance and, where required, withdrawing.

Avoiding Further Assistance

Rule 1.2(d) forbids a lawyer to “assist a client, in conduct that the lawyer knows is criminal or fraudulent,” and the prohibition does not lapse because the lawyer has seen the problem and the client has gone ahead anyway.¹⁶ The comment to Rule 1.2 addresses the hardest version, a wrong already moving: “When the client’s course of action has already begun and is continuing, the lawyer’s responsibility is especially delicate.”¹⁷ The delicacy comes from two duties that pull opposite ways. The lawyer “is not permitted to reveal the client’s wrongdoing, except where permitted by Rule 1.6,” yet “is required to avoid furthering the purpose, for example, by suggesting how it might be concealed.”¹⁸

Those two duties leave a narrower path than students first expect. Faced with a wrong already under way, a lawyer is tempted either to expose it, which confidentiality forbids except as Rule 1.6 allows, or to keep working and hope it resolves, which is assistance with knowledge. The comment’s path runs between them: the lawyer stops contributing to the wrong and keeps the client’s confidences. The path exists because the duty concerns the lawyer’s own conduct. The lawyer controls it by refraining, and refraining discloses nothing. The comment’s example, suggesting how the wrong might be concealed, shows the target. Once a wrong is in progress, concealment is the help a lawyer is best equipped to give, and the longer the wrong stays hidden, the longer it can continue.

The comment also reaches the lawyer who began without suspicion. Such a lawyer “may not continue assisting a client in conduct that the lawyer originally supposes is legally proper but then discovers is criminal or fraudulent.”¹⁹ The Texas counterpart, quoted in In re Enron, says the same and describes the lawyer’s duty as avoiding “furthering the client’s unlawful purpose.”²⁰ Knowledge is measured act by act. Work done in good faith before the discovery does not become assistance afterward, but every act after the discovery is done with knowledge. Earlier innocence protects what came before and gives no cover for what comes next. Both comments end on the same consequence: withdrawal “may be required.”

Avoiding further assistance is mostly a matter of what the lawyer stops doing, but silence can carry a wrong forward too, and Rule 4.1(b) addresses that case. Rule 4.1(a) commands honesty in the lawyer’s own affirmative statements; Rule 4.1(b) bars silence that has the effect of assisting. It requires a lawyer to “disclose a material fact when disclosure is necessary to avoid assisting a criminal or fraudulent act by a client, unless disclosure is prohibited by Rule 1.6.”²¹

The mechanism is the lawyer’s own earlier work. A document the lawyer prepared, or a negotiation the lawyer sat in on, may be what carries the client’s deception to a third party. Declining to do new work does not stop the earlier work from deceiving, and the lawyer’s continued silence lets it go on deceiving. Separating from the fraud may therefore mean neutralizing what the lawyer already put in motion, and frequently the only effective way is to reveal the material facts that remove or counter the false effect of the earlier statements.²² The comment accompanying the withdrawal rules points the same direction by allowing a lawyer who withdraws to “disaffirm any opinion, document, affirmation, or the like.”²³ The rule’s closing clause matters as much, because disclosure that Rule 1.6 prohibits is not required. The 2003 additions to Rule 1.6 have left fewer cases in which the two rules collide, and the confidentiality subsection below works through them.²⁴

Exam Analysis Steps

Because Rule 4.1(b) applies only when the lawyer’s own work has become part of the deception, the analysis is best run by asking, in order, whether the lawyer’s services are involved, whether the lawyer has learned of the wrong, and whether silence would now amount to assistance.

Ask whether the client has appropriated the lawyer’s services to advance the wrong: was the lawyer present when the client made a false statement, a participant in negotiations in which misleading representations were made, or the preparer of documents the client used to facilitate the fraud?²⁵
Ask whether the lawyer has actually learned of the misconduct, since a lawyer who never discovers it cannot act to avoid assisting it.²⁶
Ask whether the lawyer learned of it while the wrong can still be prevented, because disclosure that comes too late avoids no assistance.
Ask whether the lawyer’s silence would amount to assisting the wrong, and if so whether Rule 1.6 prohibits the disclosure that would avoid it.

A lawyer outside that circle owes no Rule 4.1(b) duty. A lawyer who stayed in the background as an advisor, without telling the client how to carry out the wrongdoing, or who was retained only after the fact to answer charges of misconduct, has nothing in the deception that the lawyer’s own work is carrying.²⁷ The oblivious lawyer stands in the same place for a different reason: a lawyer who never learns of the fraud lacks the scienter that would make the lawyer an accomplice, though a factfinder may decline to credit a disclaimer of knowledge that the circumstances contradict.²⁸

When the lawyer cannot avoid furthering the client’s purpose while remaining in the matter, withdrawal becomes the remedy. The next two sections set out when leaving is mandatory and when it is optional.

Required Withdrawal

Rule 1.16(a)(1) turns Rule 1.2(d) into a duty to leave: a lawyer “shall withdraw from the representation of a client” where “the representation will result in violation of the rules of professional conduct or other law.”³⁰ The paragraph that required a lawyer to decline an engagement also governs a lawyer already at work, which matters because knowledge so often arrives mid-matter. The comment ties the rules together: “If the lawyer’s services will be used by the client in materially furthering a course of criminal or fraudulent conduct, the lawyer must withdraw, as stated in Rule 1.16(a)(1).”³¹ A lawyer who has tried to dissuade the client and has stayed clear of concealment work may therefore reach a point at which refraining is not enough, because the lawyer’s continued presence in the matter is itself the contribution.

The duty is mandatory because the alternative is a continuing violation. Rule 1.2(d) forbids assisting conduct the lawyer knows is criminal or fraudulent.³² Once the lawyer knows, every further service that advances the wrong is assistance with knowledge, so a lawyer who stays in a matter whose purpose has become the wrong violates the rule with each task. A permissive “may” would invite the lawyer to weigh the client’s convenience, or the lawyer’s fee, against knowing participation in a crime or fraud, and the rule allows no such weighing. Nor can the client release the lawyer, because the people the prohibition protects are the wrong’s victims, and none of them has agreed to anything.

Two features of the trigger deserve attention. The phrase “will result in” looks forward: the question is what the lawyer’s continued work will do, and what the client has already done settles nothing by itself. The comment’s phrase “materially furthering” confines the duty to services that actually advance the wrong, which is why a lawyer defending a client against a charge for a completed fraud has no mandatory ground for leaving. The underlying violation also requires knowledge, so a lawyer who only suspects has not yet reached the mandatory ground.

An exam answer must also say when withdrawal is not required, because a lawyer who dislikes a client’s plan may feel compelled to leave even when no wrong is involved.

The rules list separately the grounds on which a lawyer may withdraw. Among them are that the client persists in a course of action involving the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent, that the client has used the lawyer’s services to perpetrate a crime or fraud, and that the client insists on action the lawyer considers repugnant or with which the lawyer has a fundamental disagreement.³³ Paragraph numbers differ among versions, so learn the grounds by substance. Each departs from the mandatory ground along a different axis. The first asks only for a reasonable belief, which is less than the knowledge Rule 1.2(d) requires. The second looks back at a wrong already accomplished, where continued representation need not extend it. The third involves no wrong at all. In each of those situations the lawyer decides whether to go; in the mandatory case the rule has already decided.

The mandatory duty has two limits. The first is written into its opening words: the paragraph applies “[e]xcept as stated in paragraph (c).”³⁴ Paragraph (c) provides that “[w]hen ordered to do so by a tribunal, a lawyer shall continue representation notwithstanding good cause for terminating the representation.”³⁵ A court controls its own proceedings, and a lawyer who could leave unilaterally at a critical stage might disrupt a case that the court and the other parties have organized around the lawyer’s appearance. The order leaves the lawyer in a squeeze, because the lawyer must stay while still being forbidden to assist. The comment resolves it by keeping the duty to avoid furthering the client’s purpose in force, so the order governs whether the lawyer remains counsel and the ban on assisting continues alongside it.³⁶

The second limit governs how the lawyer leaves. Rule 1.16(d) requires a lawyer, upon termination, to take steps “to the extent reasonably practicable” to protect the client’s interests, such as giving reasonable notice, allowing time to hire other counsel, and surrendering papers and property to which the client is entitled.³⁷ A client who loses a lawyer over the client’s own wrongdoing may still have lawful claims, defenses, and deadlines running. A lawyer who walked out at once, or held the file, would satisfy the withdrawal rule while harming a client who has had no chance to find replacement counsel. The qualifier “reasonably practicable” and the nature of the listed steps show that the duty concerns an orderly handoff, and it revives no obligation to finish the improper project.

Withdrawal ends the lawyer’s contribution going forward and leaves open what the lawyer may say about the exit and about earlier work. Even after withdrawal, the lawyer must refrain from disclosing the client’s confidences except as Rule 1.6 allows.³⁸ The confidentiality limits on a withdrawing lawyer are taken up below.

Key Exam Rule

Required withdrawal

A lawyer must decline a representation, or withdraw from one already under way, when continuing will result in a violation of the rules of professional conduct or other law, which includes using the lawyer’s services to materially further a client’s crime or fraud that the lawyer knows about. A tribunal’s order to continue must be obeyed notwithstanding good cause, and it does not permit the lawyer to assist the wrong. On any termination the lawyer must take reasonably practicable steps to protect the client, such as giving notice, allowing time to hire other counsel, and surrendering papers and property the client is entitled to. Grounds such as a reasonable belief of wrongdoing, a completed misuse of the lawyer’s services, or a lawful but repugnant client course allow withdrawal without requiring it.³⁹

Permitted Withdrawal

Rule 1.16(b) lists the grounds on which a lawyer may withdraw, and two respond directly to client crime or fraud: the client “persists in a course of action involving the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent,” or the client “has used the lawyer’s services to perpetrate a crime or fraud.”⁴⁰ Both sit beside the mandatory duty taught above, and both ask less of the lawyer. Required withdrawal arises when continuing will result in a violation, which for client wrongdoing means the lawyer knows the services will materially further it. The permissive grounds rest on weaker facts, and when one applies, the lawyer decides whether to go.

The first ground exists because the moment of knowledge is a poor time to be deciding whether to leave. Rule 1.2(d) turns on what the lawyer knows, so a lawyer who only suspects has violated nothing and has no mandatory duty to withdraw.⁴¹ Suspicion is nonetheless an unstable position. A lawyer who keeps working on a plan that looks fraudulent is one document or one admission away from knowing, and from that point each task is assistance. A ground that demanded certainty would force the lawyer to wait inside that zone until the evidence became undeniable. Resting the ground on reasonable belief lets the lawyer step out first.

The ground has three working parts. The client must persist, which presupposes that the lawyer has already raised the problem, as the duty to warn and dissuade contemplates, and that the client has gone ahead anyway. The course of action must involve the lawyer’s services, so wrongdoing that never touches the lawyer’s work does not supply the ground. And the lawyer’s belief must be reasonable: the facts must support it, though it need not reach the knowledge Rule 1.2(d) requires.

The second ground looks backward. A client who has used the lawyer’s services to perpetrate a crime or fraud has given the lawyer a reason to leave even though nothing in the lawyer’s present work is being misused.⁴² The reason lies in the relationship. A client who has treated the lawyer as an instrument once has shown how the client regards the lawyer, and the lawyer’s earlier work may still be in circulation carrying the deception. The ground lets the lawyer separate before a pattern forms in which the lawyer’s name keeps appearing on the client’s wrongs.

Hypothetical Illustration

A lawyer helps a client obtain a series of bank loans and is later engaged on the client’s corporate merger. Partway through the merger work, the lawyer learns that the loan applications he prepared carried falsified financial statements, so the client used his services to defraud the bank. Nothing suggests that the merger papers are false. The mandatory ground is absent, because the merger work will not result in a violation, but the client has used the lawyer’s services to perpetrate a fraud, and he may withdraw from the merger matter. A different lawyer is retained on a merger by a client who defrauded a lender years earlier with another firm’s help. That lawyer’s services were never used for any wrong, so the past-use ground does not reach this client, and any right to leave would have to come from a different ground.

Both grounds are permissive because neither shows that the representation will result in a violation. A reasonable belief falls short of the knowledge Rule 1.2(d) requires, and past misuse leaves the present work clean. The client also has a continuing interest in having counsel, since a lawyer who leaves mid-matter can strand the client at a bad moment, and the rule leaves the lawyer free to weigh that cost against the risk. A lawyer whose belief is weak, or whose remaining work can go forward without furthering anything, may stay. The grounds in Rule 1.16(b) are alternatives, so a lawyer who has one of the crime-fraud grounds need not also show that leaving can be accomplished without material adverse effect on the client, which is a separate ground.⁴³ The limits on mandatory withdrawal carry over. A tribunal’s order to continue controls notwithstanding good cause, and on any termination the lawyer must take reasonably practicable steps to protect the client, such as giving notice, allowing time to hire other counsel, and surrendering papers and property.⁴⁴

Exam Analysis Steps

Because the three tiers differ in what the lawyer must know and in what the client has done, the grounds are best worked from the strongest down to the weakest, and the first one that fits sets the lawyer’s position.

Ask whether the lawyer knows that the client will use the lawyer’s services to materially further a crime or fraud. If so, withdrawal is required and the lawyer has no discretion to stay.⁴⁵
If the lawyer lacks that knowledge, ask whether the client persists, after the lawyer’s objection, in a course involving the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent. If so, the lawyer may withdraw.
If neither applies, ask whether the client has used the lawyer’s own services to perpetrate a crime or fraud, even though the current work is untouched. If so, the lawyer may withdraw.
If none of these fits, the crime-fraud grounds are unavailable, and any decision to leave must rest on one of the other grounds in Rule 1.16(b).

The usual error is to answer “must withdraw” on the strength of a suspicion, which treats a permissive tier as a mandatory one. The reverse error treats a lawyer who actually knows as free to stay or go. The verb changes at the knowledge line, and the answer should say which tier the facts reach before it says anything about the lawyer’s choice.

State rules sometimes convert a permissive ground into a mandatory one. In In re Enron, the court described the Texas rule as requiring withdrawal if the client “persists in a course of action involving the lawyer’s services that the lawyer reasonably believes may be criminal or fraudulent” or has used the lawyer’s services to perpetrate a crime or fraud.⁴⁶ The Texas wording also lowers the threshold, from conduct the lawyer reasonably believes is criminal or fraudulent to conduct that “may be” so.⁴⁷ An exam answer should state the Model Rule position, in which these two grounds permit withdrawal, and shift to a mandatory duty only when the problem names a jurisdiction that has made that choice.

Confidentiality and Noisy Withdrawal

A lawyer who has reached the point of required or permitted withdrawal has already decided to stop contributing to the client’s conduct, which Rule 1.2(d) forbids the lawyer to assist.⁴⁹ What remains is how to leave without creating a second problem. The comment to Rule 1.2 answers in two sentences. After withdrawal the lawyer must “refrain from making disclosure of the client’s confidences, except as otherwise provided in Rule 1.6,” yet nothing in Rule 1.6, Rule 1.8(b), or Rule 1.16(d) prevents the lawyer from “giving notice of the fact of withdrawal,” and the lawyer “may also withdraw or disaffirm any opinion, document, affirmation, or the like.”⁵⁰ The withdrawal rules quoted in In re Enron carry the same limit: after withdrawal the lawyer must still refrain from disclosing the client’s confidences except as Rule 1.6 allows.⁵¹

Two interests meet at the exit, and the comment gives each a bounded space. The duty of confidentiality survives withdrawal because it exists to make candor possible. A client who expects the lawyer to reveal what was said once the relationship ends will stop telling the lawyer the facts, and a lawyer who does not know the facts cannot talk the client out of a bad plan. If leaving freed the lawyer to disclose, the prospect of withdrawal would chill the very conversations in which lawyers dissuade clients. Against that interest stands a lawyer whose earlier work is still in circulation. The lawyer has stopped doing new work, but a letter, an opinion, or a filing the lawyer wrote keeps carrying a deception, and the lawyer’s name lends it credibility. Permission to disaffirm that work resolves the tension, because the lawyer withdraws the lawyer’s own endorsement and discloses nothing the client said.

The lawyer’s letter is an example of what some call a “noisy withdrawal”: the lawyer withdraws, tells a third party of the fact of withdrawal, and disaffirms a previously prepared document, opinion, or affirmation.⁵² The withdrawal is noisy because the third party hears it, and the lawyer need not say why. Sometimes the lawyer must do more than leave quietly, and Rule 4.1(b) is the reason: a lawyer shall not “fail to disclose a material fact to a third person when disclosure is necessary to avoid assisting a criminal or fraudulent act by a client, unless disclosure is prohibited by Rule 1.6.”⁵³

The two scenes show where the line runs: between speaking about the lawyer’s own work and speaking about what the client revealed. A disavowal is a statement about the lawyer’s own opinion, document, or affirmation, and the comment lets the lawyer withdraw it. A statement that the client falsified the lists, or an account of what the client said about them, is information relating to the representation, and the lawyer may give it only if a Rule 1.6 exception covers it. The bank may well infer fraud from a lawyer who withdraws and disclaims her letters at the same moment, but that inference does not turn the disavowal into a disclosure. Noisy withdrawal is therefore the safe fallback where confidentiality is strictest, and it is the wrong answer to a problem asking whether the lawyer may reveal the client’s fraud, because the disaffirmance needs no exception and the revelation does.

How much more the lawyer may say turns on the Rule 1.6 exceptions. In 2003 the ABA added exceptions that allow disclosure to prevent or rectify economic harm when the client has used the lawyer’s services to further a criminal or fraudulent act, and that permit disclosure to comply with other law.⁵⁴ The revision brought the Model Rule closer to the prevailing rule in the vast majority of states.⁵⁵ Rule 1.6(b)(2) and (b)(3) permit, but do not require, disclosure to prevent or rectify substantial financial injury from a client’s crime or fraud, and read together they reach both a wrong still unfolding and harm already done. Each applies only where the client has used the lawyer’s services in furtherance of the wrong.

The two conditions do different work. The harm condition, substantial injury to financial interests or property, confines the permission to serious losses, so confidentiality does not yield to every suspicion of wrongdoing. The services condition does subtler work. A lawyer whose work product the client has put to use in a crime or fraud has become an instrument of the wrong, which exposes the lawyer to accusation and leaves the lawyer’s earlier work harming a third party. Disclosure is the lawyer’s means of preventing or undoing that harm, and a lawyer who may use it can separate fully from the client’s conduct. Without the condition, any lawyer who heard of any client fraud could reveal it, and confidentiality would shrink into a general power to report on clients.

State rules vary on both sides of the Model Rule. California, Washington, and the District of Columbia continue to prohibit a lawyer from disclosing to sources outside the corporation that a client is about to engage in illegal conduct, while New Jersey and Florida provide that a lawyer must disclose client confidences to prevent certain criminal acts.⁵⁶ Texas lets a lawyer reveal confidential information when the lawyer has reason to believe it necessary to prevent the client from committing a criminal or fraudulent act, and requires disclosure adverse to the client only where the lawyer’s information clearly establishes likely death or substantial bodily harm.⁵⁷ An answer should apply the jurisdiction the problem names and otherwise use the Model Rule.

Rule 4.1(b) interacts with these exceptions. Its closing clause makes confidentiality the limit on the duty, so the narrower the confidentiality exceptions, the more often the limit bites.⁵⁸ After the 2003 exceptions, a disclosure that Rule 1.6(b)(2) or (b)(3) permits is no longer one that Rule 1.6 prohibits, and fewer circumstances remain in which confidentiality constrains the Rule 4.1(b) duty.⁵⁹ Even where a lawyer could construct a formal argument against mandatory disclosure, the permissive exceptions let the lawyer share the information and so extricate fully from the client’s conduct, which lowers the risk of being charged, even mistakenly, with complicity.⁶⁰

The past-versus-future contrast that governed the decision to accept an engagement returns here in refined form. A lawyer who learns of a client’s completed crime remains bound by confidentiality, so the completed wrong alone releases nothing. What widens the lawyer’s room is whether the lawyer’s own services were used.

The usual exam error treats withdrawal as the event that frees the lawyer to speak. Withdrawal changes what the lawyer does going forward and leaves the duty of confidentiality where it was. The permission to disaffirm comes from the comment, and the permission to reveal client information comes from Rule 1.6. A problem about what a withdrawing lawyer may say is best run in order.

Exam Analysis Steps

Because the lawyer’s room to speak depends on whether the statement concerns the lawyer’s own work or the client’s information, the question is best worked by sorting the statement first and testing it against the exceptions second.

Identify whether the statement is notice of the fact of withdrawal or a disaffirmance of the lawyer’s own opinion, document, or affirmation. If so, the comment permits it without any Rule 1.6 exception.
If the statement reveals information relating to the representation, ask whether a Rule 1.6 exception covers it, including whether the harm is substantial financial injury and whether the client used the lawyer’s services in furtherance of the crime or fraud.
Ask whether disclosure of a material fact is necessary to avoid assisting the client’s crime or fraud, because Rule 4.1(b) then requires it unless Rule 1.6 prohibits it.
Check the jurisdiction the problem names, since some states forbid, and others require, disclosure on facts where the Model Rule merely permits it.

Threatened Client Perjury

Perjury, the crime of lying under oath, gives Rule 1.2(d) one of its plainest applications. A lawyer who knows that a client plans to testify falsely, and who helps the client deliver the testimony, assists conduct the lawyer knows is criminal.⁶² The sequence this unit has described, dissuade, avoid furthering, withdraw where required, and leave without breaching confidences, runs here under its heaviest pressure, because the setting is a trial in progress or about to begin and the court is watching. The leading decision is Nix v. Whiteside, which asked what the Constitution permits a lawyer to do when a defendant says that he will perjure himself.⁶³

Whiteside, on trial for murder, told his lawyer, Robinson, that he would testify falsely that he had seen a gun in the victim’s hand. Robinson tried to dissuade him, warning that he would seek to withdraw and disclose the plan if Whiteside lied. Whiteside did not lie, and Robinson divulged no client communications until Whiteside’s post-trial attack on his performance compelled him to respond.⁶⁴ The Court of Appeals reasoned that counsel had impermissibly compromised Whiteside’s right to testify by conditioning continued representation and confidentiality on restricted testimony. The Supreme Court disagreed.⁶⁵

The question was a constitutional one: whether Robinson’s conduct fell within the range of reasonable professional conduct that the Sixth Amendment accepts under Strickland. Nix was therefore a case about effective assistance, and the ethics rules entered the opinion as evidence of what the profession treats as acceptable. The Court found no failure to meet reasonable professional standards. Whether Robinson’s conduct was seen as a successful attempt to dissuade his client or as a threat to withdraw and disclose, it fell well within the accepted range.⁶⁶

The Court grounded that conclusion in the limits of loyalty. An attorney’s duty to advance a client’s interests is limited by an equally solemn duty to comply with the law and standards of professional conduct, and that limit specifically ensures that the client may not use false evidence.⁶⁷ The Court described the principle as one that expositors of professional norms have stated in unequivocal terms since the 1908 Canons.⁶⁸ Perjury is as much a crime as tampering with witnesses or jurors, and it undermines the administration of justice, so no promise of loyalty reaches it.⁶⁹ The Court also treated the maxim that “a lawyer must believe his client, not judge him” as no license to be a party to known perjury or to give it any aid.⁷⁰

The stakes for a lawyer who goes along are personal. A lawyer who knows the client will lie and then examines the client in a way designed to draw out the falsehood has done more than tolerate it. The lawyer has joined the crime, with exposure to discipline and to prosecution, including for subornation of perjury, which means procuring another to lie under oath.⁷¹

What the lawyer should do instead begins with persuasion, because the first duty on facing a proposal for perjurious testimony is to try to dissuade the client.⁷² Dissuasion comes first for the reason given earlier: it can stop the wrong at its source without disclosing anything or ending the relationship. The Court also noted that the rules expressly permit withdrawal when a client threatens perjury, while observing that withdrawing at trial raises hard questions, including possible mistrial and double jeopardy claims. And it observed that the rules of its day did more than authorize disclosure of client perjury; they required it.⁷³

The Court then rejected the Court of Appeals’ objections. Whiteside did testify, and counsel restrained him only from testifying falsely; the right to testify does not include a right to testify falsely. Nor did counsel breach a confidence, because he revealed nothing until Whiteside himself put counsel’s performance in issue after trial.⁷⁴

A state comment shows how the same sequence looks once a jurisdiction writes it down. The Florida comment on false evidence says that a lawyer may not assist the client or any witness in offering false testimony, and may not permit the client to testify falsely in narrative form unless the tribunal orders it.⁷⁵ It then sets out the lawyer’s response in order and uses mandatory language at each stage where Nix required only reasonableness.⁷⁶

The ordering reflects the economy found throughout this unit. Persuasion costs the client nothing the client is entitled to, and a threat of disclosure gives the client a last chance to change course before the lawyer takes a step that damages the relationship. Disclosure comes last because it is the only step that exposes the client’s confidence. The bar on narrative testimony closes the escape route of avoiding questions while the lie goes to the jury.

The difference between the two lawyers is the one taught in the discussion of the knowledge requirement. A threat of perjury that rests on the client’s own admission supplies the knowledge; a lawyer’s unease about a story does not. For the lawyer’s duty to the tribunal when evidence is false, including the point at which disclosure becomes mandatory, see Rule 3.3, which is taught in a separate reading.

The usual exam errors run in three directions. Some answers treat the right to testify, or the right to counsel, as obliging the lawyer to present whatever the client says, which Nix rejects because neither right extends to false testimony. Others treat disclosure as the lawyer’s first move, when the first duty is to dissuade. A third group reads Nix as a discipline case that decided what Rule 1.2(d) requires. The holding is limited to the Sixth Amendment: a lawyer who dissuades, or who threatens to withdraw and disclose, acts within the range of reasonable professional conduct, and the ethical rules were consulted to define that range.