Elements of the Privilege

Elements of the Privilege

A court deciding whether a communication is privileged first identifies the governing law and then tests the communication against the privilege’s elements. In an entirely federal matter, Rule 501 sends the court to the common law, and the court looks principally to federal decisions applying it.¹ The Restatement (Third) of the Law Governing Lawyers § 68 states the elements in four parts: the privilege protects (1) communications (2) made between privileged persons (3) in confidence (4) for the purpose of obtaining or providing legal advice for the client.² Privileged persons are the client, the lawyer, and the agents who help either of them deliver legal services, and the sections below work out who belongs in each group. When a client conveys information to a lawyer in confidence to secure legal advice, the privilege protects that communication against compelled disclosure, and a communication that lacks any one of the four parts falls outside the protection. A court can therefore resolve most claims by asking the four questions in turn.

Each element restates the privilege’s purpose as a test. Privileges exist to promote full and frank communication in valued relationships,³ so this one attaches to the communication and not to the facts the communication describes. It covers the people needed to deliver legal services and stops at outsiders, because every outsider admitted to the conversation could later be compelled to repeat it. It requires confidence, since a client who meant the conversation to reach a third party was not relying on secrecy. And it requires a legal-advice purpose, because the privilege serves legal help, and a conversation with a lawyer about something else does not engage that function.

Older formulations list the same material under different labels. Wigmore’s eight-part statement, applied by the Seventh Circuit in United States v. Lawless and repeated in many other opinions, protects (1) legal advice of any kind sought (2) from a professional legal adviser in that capacity, (3) communications relating to that purpose, (4) made in confidence (5) by the client, (6) permanently protected at the client’s instance (7) from disclosure by the client or the adviser, (8) unless the protection is waived.⁴ The four-part formulation associated with United States v. United Shoe Machinery Corp. asks whether the holder is or sought to become a client, whether the recipient was acting as a lawyer, whether the communication was made without strangers present to secure a legal opinion or legal services and not to commit a crime or tort, and whether the privilege was claimed and not waived.⁵ The extra parts in both lists concern who may claim the privilege and whether it was waived, so the four elements above remain the test for whether a communication was protected in the first place.

The party asserting the privilege must establish every element. The D.C. Circuit put the rule directly in In re Lindsey: the claimant bears the burden of proving that the communications are protected, and a blanket assertion of the privilege will not suffice.⁶ The allocation tracks who knows the facts. The proponent knows who was present, why the conversation took place, and who later saw it, and the privilege withholds probative evidence, so the person who wants the evidence kept out must justify the request. Courts reinforce the point by confining the privilege within the narrowest possible limits consistent with its logic.⁷

Because the showing is made communication by communication, a single file can hold protected and unprotected items, and the elements work as a sorting tool for each piece. The first element to apply is the communication itself, and in particular the line between what a client said and what a client knows.

Communications, Not Facts

The first element draws a line that every later question about scope depends on: what a client said to a lawyer is protected, and what the client knows is not. Upjohn Co. v. United States quoted with approval a lower court’s statement of that line. A client “cannot be compelled to answer the question, ‘What did you say or write to the attorney?’ but may not refuse to disclose any relevant fact within his knowledge merely because he incorporated a statement of such fact into his communications to his attorney.”⁸ A fact and a communication about that fact are different things, and the privilege reaches only the communication.

Two reasons support the line, and each answers a different question. The first asks what the privilege needs in order to work. A client speaks candidly once sure that the conversation itself will stay secret, and protecting the communication gives that assurance completely. Protecting the underlying fact adds nothing, because the fact exists whether or not anyone consulted a lawyer, and applying the privilege to communications leaves the adversary in no worse position than if the conversation had never taken place.⁹ The second asks what would happen if the line moved. If a fact became privileged merely because the client told it to a lawyer, any client could put damaging knowledge beyond discovery by reciting it in the lawyer’s office. The first reason shows that protecting facts is unnecessary, and the second shows that it would do harm.

The Upjohn record shows both sides of the line at once. General counsel Thomas gathered information from employees through questionnaires and interviews, at the direction of corporate superiors, so that he could give the company legal advice about questionable payments, and the company treated the responses as confidential.¹⁰ The Court protected those communications, yet the Government stayed free to question the employees about the events themselves, and the IRS had already interviewed some 25 of them.¹¹ Subpoenaing the questionnaires and counsel’s notes would have been more convenient for the IRS, but convenience does not outweigh the policies the privilege serves.¹² An employee’s written answer to counsel was therefore protected, while the employee’s knowledge of the payments was not.

On an exam the distinction usually arrives in the wording of a question. A client who told a lawyer about knocking down a telephone pole with a red car cannot be asked what the client told the lawyer about the pole or the car, yet can be asked whether the client knocked down the pole and what color the car is.¹³ The first question calls for an answer that would reveal what was said in confidence, while the second calls for an answer that states a fact. The test is what the answer would reveal.

Documents and objects follow the same logic. Handing over a package usually does not count as a communication, and a client who gives a lawyer a diary has not made the diary privileged, because the diary was not prepared at the lawyer’s behest or to communicate with the lawyer.¹⁴ Pre-existing business records work the same way. A company that gives its financial statements to its lawyers has not made them privileged, because the statements were prepared for business purposes and not to obtain legal services, and a rule that protected documents on delivery would let any client put damaging records out of discovery’s reach by giving them to counsel. The privilege is generally inapplicable to a client’s effort to conceal physical evidence, since protecting that kind of material is only tenuously connected to the purpose of the privilege.¹⁵ A confidential letter in which the client explains the records and asks for advice is a communication and is protected. The records themselves are not.

What counts as a communication depends on the speaker’s intent. A communication is something the client meant to convey to the lawyer, much as the hearsay rules treat a statement as an intended assertion, so physical appearance and incidental behavior carry no protection. A client who arrives slurring speech from drinking has not communicated the slurring, and only the content of what the client said is covered.¹⁶ Intent can also turn conduct into communication. In City and County of San Francisco v. Superior Court, the California Supreme Court explained that the privilege embraces actions, signs, and other means of conveying information to the attorney, as well as oral or written statements.¹⁷ Almost any act done in the lawyer’s sight during the consultation may be done as a communication, and the only question is whether, in the circumstances, the client intended it as one. A client may make a handwriting specimen for the lawyer’s information, exhibit an identifying scar, or show a secret token. The scar illustrates the whole section: the act of showing it to the lawyer in a confidential consultation can be protected, while the scar itself remains a physical fact outside the privilege.

Physical evidence is the third setting, and here the lawyer’s own conduct matters as much as the client’s. A weapon or stolen money exists independent of anything a client told a lawyer, so placing it in a lawyer’s hands does not turn it into a communication. In re Ryder shows the point. The lawyer, Ryder, took stolen money and a shotgun from his client, Cook, and held them. The court held that Ryder, not his client, took the initiative in transferring the incriminating possession, and that Ryder’s conduct went far beyond receiving and retaining a confidential communication from his client. Ryder’s own counsel conceded that his acts were not within the privilege.¹⁸

People v. Meredith sits on the other side of that line. The defendant told his lawyer where the victim’s wallet could be found, the lawyer sent an investigator to look, and the investigator found it in a garbage can behind the defendant’s residence. The wallet was admissible, and the dispute concerned testimony about where it was found. The California Supreme Court recognized that the privilege may reach information a lawyer or investigator acquires as a direct result of a confidential communication, and it then asked how far that protection survives when the defense has moved the evidence beyond the prosecution’s reach.

The two decisions differ in what was sought. In Ryder, the lawyer’s own act of taking custody was conduct and not a client message, so nothing in it could be protected. In Meredith, the testimony traced back to a confidential communication about where an item could be found, so the privilege applied at the outset and the open question was how far the defense’s handling of the evidence cut the protection back. Waiver and the crime-fraud exception can strip protection from a communication that otherwise qualifies, but they raise a different question from the one asked here, which is whether the thing sought is a communication at all.

Client Identity and Fees

The line between communication and fact decides the objection a lawyer hears most often when asked about a client: that the client’s name, the fact that the client came in, and the fee are all secrets of the relationship. In Baird v. Koerner, the Ninth Circuit concluded that no federal body of law requires a client’s identity to be excluded from the privilege and that the question must be assessed case by case on the particular facts.²⁰ The court also stated the test for the other direction: if identifying the client would convey information that ordinarily would be conceded to be part of a privileged communication between lawyer and client, the privilege should extend to the identification, absent other factors.²¹ Baird was decided before the current Rule 501, so its analysis of identity, and not its choice of law, is what carries forward.

The default rule follows from what the privilege is for. A client’s name, the existence of a consultation, and the amount agreed as a fee are facts about the relationship, and a client who knows they can be asked about can still describe the legal problem to the lawyer with complete candor. Concealing identity is only tenuously connected to the full and frank communication the privilege exists to produce, so the privilege is generally inapplicable to it.²² The point repeats the lesson of the previous section: the privilege protects what was said and leaves what is known open to questioning.

Fee arrangements follow the same logic. In re Grand Jury Investigation, the Third Circuit stated that the attorney-client privilege does not ordinarily extend to fee arrangements. The government there was not seeking the substance of any confidential communication between lawyer and client. It sought the fact of consultation and the fee arrangements, and that information was not privileged.²³ The opinion allowed that exceptional circumstances might bring a fee arrangement within the privilege, but it found none in that case.²⁴ A fee figure or a consultation date tells the fact-finder that a relationship existed and reveals nothing about what the client confided.

Baird itself supplies the case in which identity was protected. Lawyers acting for unnamed clients sent the government money as voluntary payments toward income taxes for earlier years, and the government demanded the clients’ names even though no litigation was pending.²⁵ The court reasoned that the names were useful to the government for one purpose only, to learn which taxpayers believed they had been delinquent, and that belief was the reason the lawyers had been hired. Compelling the names would therefore let the government obtain by indirection the very admission the clients had brought to counsel.²⁶ The court treated the case as falling within a recognized exception under which a client’s name is privileged where it is material only to show an acknowledgment of guilt of the offenses for which the lawyer was employed.²⁷

Even where a name would reveal a confidence, Baird identified four factors that cut against extending the privilege to identity:

1. The lawyer has commenced litigation on the client’s behalf, so that the client has voluntarily submitted to the court’s jurisdiction.
2. The identification relates to the lawyer’s employment by a third person who is neither the client nor the client’s agent.
3. The lawyer was employed with respect to future criminal or fraudulent transactions.
4. The lawyer is personally a defendant in a criminal matter.²⁸

The first factor is the clearest. Once litigation is commenced, a party should not be made to struggle in the dark against unknown forces, and the court said that had the taxpayers sued the government, their lawyer would have had to reveal for whom the lawyer was acting.²⁹ The absence of any suit was the difference in Baird, and it shows how one fact can move identity from protected to unprotected. The third factor reflects a limit built into the privilege itself, since a lawyer consulted about a crime or fraud still to be committed is not giving the kind of help that candor between client and lawyer is meant to encourage.

Side by side, the two decisions show that the hinge is what the answer would reveal, not the form of the information. The Third Circuit subpoena sought the fact of consultation and the fee, which disclose only that a relationship existed. The Baird demand for names would have disclosed that the clients came to counsel because they believed they owed back taxes, which is the content of the confidence. A stem in which a grand jury asks for a retainer amount or the date of a first meeting tests the first situation, and one in which the client’s name alone would confess the matter the lawyer was hired about tests the second.

The general statement of the privilege is that, once it attaches, it is not subject to judicial override.³⁰ Baird’s weighing of factors is consistent with that statement because the weighing happens earlier, when the court decides whether identity falls within the privilege at all.

Confidentiality

Federal Rule of Evidence 502(g)(1) defines the attorney-client privilege as the protection that applicable law provides for confidential attorney-client communications, so confidentiality sits inside the definition itself.³² The Uniform Rules of Evidence say what the word means: a communication is confidential if it is not intended to be disclosed to third persons other than those to whom disclosure is made in furtherance of the rendition of professional legal services, or those reasonably necessary to transmit the communication.³³ Two ingredients follow and organize this section: the client’s intent that the communication stay within the privileged circle, and circumstances that make that intent reasonable. Both are needed because the privilege buys candor with a promise of secrecy. A client who meant the content to reach outsiders, or who spoke where strangers could plainly hear, was not relying on the promise, and shutting the conversation away from the fact-finder would cost evidence without protecting any candor that depended on it.

The first ingredient, intent, is met when the client expressly made the communication confidential or could reasonably assume that the lawyer would understand it to be meant in confidence. A showing that the words passed from client to lawyer is not enough, and the circumstances must indicate an intention of secrecy.³⁴ If the client made the same statements to third persons on other occasions, that is persuasive evidence that like statements to the lawyer were not meant to be confidential, and matters the client intended to make public plainly lack confidentiality.³⁵ A parallel formulation adds the channel: the client must intend confidence and use a means of communication that, so far as the client is aware, does not disclose the information to third persons who are not essential to the relationship.³⁶ Confidentiality is measured when the communication is made. A later disclosure by someone raises the separate question of waiver.

The sharpest application of the intent requirement involves information a client gives a lawyer so that it can be passed along. In United States v. Lawless, the question was whether the privilege protected two documents the client had transferred to a lawyer in connection with preparing a tax return. The district court said it did, reasoning that the documents were the type of information a client may furnish a lawyer and not have the lawyer disclose.³⁷ The Seventh Circuit found that reasoning inadequate, because information transmitted to a lawyer with the intent that it be passed to a third party, here on a tax return, is not confidential.³⁸ A client who sends information so that it can be used on the return expects it to reach the government, so no promise of secrecy shaped what the client said, and the lawyer served as a conduit. The court also noted that return preparation is primarily an accounting service, a separate objection about purpose that the section on legal-advice purpose takes up.³⁹

Preliminary conversations about a disclosure the client plans to make, and drafts of it, show the difference between asking for advice and handing over facts to forward. At least one court has held that privilege may attach to preliminary drafts, but only if they were prepared or circulated to give or obtain legal advice and contain information or provisions not included in the final version.⁴⁰ The protective view fits the privilege’s goal. A client and lawyer ought to be able to discuss the precise terms of a disclosure, including the best wording, without risking that matters ultimately left out would be unprivileged.⁴¹

The second ingredient asks who else was present, and the answer depends on whether each person belongs to the circle through which legal services are delivered. A stranger’s presence defeats confidentiality: the presence of a casual, disinterested third person within hearing, to the client’s knowledge, shows that the communication was not meant to be confidential.⁴² The circle is wider than two people, however. If the client needs an interpreter in order to consult the lawyer, the interpreter’s presence does not deprive the communication of its confidential and privileged character.⁴³ Where the client brings one of the client’s agents to the conference, or the lawyer calls in a clerk or confidential secretary, courts presume that these intermediaries do not undercut confidentiality, and the presumption does not depend on whether the person’s presence was reasonably necessary in the particular instance. It is the way business is generally done.⁴⁴ Relatives and friends are the unsettled middle, because the cases are not consistent. The fuller inquiry there asks both whether the client reasonably understood the conference to be confidential and whether the relative’s or friend’s presence was reasonably necessary to protect the client’s interests.⁴⁵ Staff and agents add no exposure beyond what the privilege already accepts as the cost of delivering legal services. A friend who comes along for company is an audience the client chose and the lawyer’s work did not require.

The cab driver sat openly within the client’s view. The harder case is the stranger who listens without the client’s knowledge. The common-law eavesdropper rule treated that listener harshly: a third party who overheard a lawyer-client conversation could testify to it even though the lawyer and client could not be made to. Most courts have abandoned that rule under the modern view, and the privilege applies to conversations held where it is reasonable to believe they will not be overheard or intercepted.⁴⁶ The old rule made protection depend on luck, since any unnoticed listener could strip away what the privilege promised, and a promise that fails whenever someone secretly listens does little to encourage candor. Under the modern view the client’s own precautions and knowledge decide the outcome, and the stranger’s success at listening does not.

Electronic channels put the reasonableness standard under pressure, because the risk of interception depends on the technology. Depending on the technology involved, some cell-phone conversations can be overheard, e-mail and other electronic communications raise similar problems, and law and practice on what steps are reasonable to prevent electronic eavesdropping continue to evolve.⁴⁷ An answer should state the standard, which is reasonable steps and a reasonable belief that the exchange will not be intercepted, and then look for any sign that the client knew the channel was watched. A jail inmate who knows that her calls are monitored sits at the clear end of the range: the call is not confidential, and the communication is not privileged.⁴⁸ A client who uses an ordinary channel with reasonable care sits near the other end, and the middle is where the evolving rules matter.

Joint clients present an audience problem of a different kind, because each client hears what the other tells the lawyer. The trend is to treat joint-client material as generally confidential but not confidential as between the clients. Neither should disclose it unless both agree to waive, and it is generally held unprivileged when the clients litigate against each other.⁴⁹

Against the lender, each client reasonably expected the lawyer to keep what was said from outsiders, so the privilege holds. Against each other, no such expectation existed, so the privilege gives way when the clients become adversaries.

Lawyer and Client

Rule 501 leaves the content of the privilege to the common law, so the questions of who counts as a lawyer and who counts as a client are answered by decided cases and by uniform formulations that restate them.⁵¹ The Uniform Rules of Evidence define a client as a person, including a public officer, corporation, association, or other public or private organization, for whom a lawyer renders professional legal services or who consults a lawyer with a view to obtaining those services.⁵² They define a lawyer as a person authorized, or reasonably believed by the client to be authorized, to practice law in any State or country.⁵³ The proposed privilege rules that Congress did not enact have nonetheless served as an influential standard in federal and state courts, although the scope of the privilege still differs from jurisdiction to jurisdiction, particularly among the states.⁵⁴ Two features of the definitions do most of the work. The clause covering a person who consults a lawyer in order to obtain services starts protection before any engagement exists, and the clause covering a person the client reasonably believes to be authorized measures the lawyer’s status from the client’s side of the desk. A person describing a legal problem cannot check bar records first or sign a retainer first, and a lawyer cannot decide whether to accept a matter without hearing the facts, so a privilege that waited for formal status would miss the conversations in which candor matters most.

The relationship must also be a professional one. The privilege depends on the client’s belief that she is consulting a lawyer in that capacity and on her manifested intention to seek professional legal advice.⁵⁵ Payment or agreement to pay a fee is not essential, but the person asserting the privilege carries the burden of showing that the consultation was professional.⁵⁶ Consulting a person who happens to be a lawyer as a friend, a business adviser or banker, a negotiator, an accountant, a mere scrivener, an attesting witness, an executor, or an agent does not create a professional consultation, and what is said is not privileged.⁵⁷ The limit is functional. The privilege buys candor so that a lawyer can give legal help, and if membership in the bar were enough, any client could shelter a business or personal conversation simply by choosing a lawyer as the listener.

In re Lindsey, a D.C. Circuit decision about a White House lawyer, applies the same limit to government counsel. Only communications that seek “legal advice” from “a professional legal adviser in his capacity as such” are protected, and a lawyer’s advice on political, strategic, or policy issues, however valuable, is not shielded by the privilege.⁵⁸ For the conversations at issue, the court also required the Office of the President to present the underlying facts demonstrating the existence of the privilege.⁵⁹

The lawyer definition turns on what the client could reasonably believe. It is sufficient if the client reasonably believes that the person consulted is a lawyer, though in fact the person is not.⁶⁰ A client has no practical way to detect a pretender, and the candor the privilege exists to encourage was already given in reliance on an expectation of confidence, so compelling the conversation would penalize the client for a deception practiced on the client. The reference to authorization in any State or country also means that a lawyer admitted only in another state or abroad can qualify, which departs from the traditional view that a continuing consultation with a lawyer outside the state of licensure was not privileged.⁶¹

Protection also begins before the engagement does. Communications in the course of preliminary discussion with a view to employing the lawyer are privileged even though the employment is in the end not accepted.⁶² A published illustration shows how far that reaches. Two dentists consider buying a retiring dentist’s practice and consult two small-business lawyers. The lawyers give the initial consultation to potential clients without charge and sign a formal fee agreement only if they are retained, and the dentists then abandon the plan.⁶³ The conversations remain within the privilege, because a client includes a person who consults a lawyer with a view to obtaining legal services. Protection depends neither on payment of fees nor on a formal contract, and it covers communications with prospective clients who meet with a lawyer to decide whether to retain the lawyer.⁶⁴

The same mechanism explains the rule on both sides. A would-be client who had to hire the lawyer before explaining the problem could not explain it, and a lawyer who had to commit before hearing the problem could not responsibly agree, so the first conversation, in which the facts come out and the lawyer decides whether to take the matter, would be the least protected of all. The professional-capacity requirement supplies the boundary. The consultation must be one made with a view to employing the lawyer, and the proponent must show it, so a remark to a lawyer at a social gathering does not become a consultation merely because one of the speakers practices law.

The privilege also reaches communications that travel from lawyer to client, though courts differ on how far. The privilege protects confidential communications from the lawyer to the client if disclosure would reveal the substance of a client communication. The reason is that an opponent could often deduce what the client said from the lawyer’s advice, which rarely springs from the lawyer’s head alone.⁶⁵ In SCM Corp. v. Xerox Corp., Xerox claimed privilege for the opinions of its patent attorneys on whether particular machines were covered by specified patents. The court framed the question as whether the privilege protects all advice from attorney to client or only advice that reveals, by adoption or implication, a fact the client communicated in confidence. The narrow position protects advice only in the second situation, and the broad position, which Wigmore supports and some decisions state, protects any legal advice from lawyer to client.⁶⁶

The Xerox court applied the narrow standard. It declined to treat Proposed Rule 503(b) as authoritative on a question the common law had left unsettled, while acknowledging that rejected proposed rules are some evidence of the common law, and it noted that advice risking disclosure of client confidences is protected even under the narrow standard.⁶⁷ Other federal courts use formulations that range from protecting communications that reveal confidential client communications to protecting those in which legal advice is the predominant element, and the decisions catalog that divergence.⁶⁸

In practice the dispute rarely changes an outcome, because legal advice usually responds to what the client disclosed. The standards part company only when a lawyer’s statement can be cleanly separated from anything the client communicated. A legal opinion that reveals nothing the client said is protected under the broad approach and unprotected under the narrow one, and information a lawyer gathers from a third party and passes to the client is not necessarily protected when disclosing it would reveal no client communication.⁶⁹ An answer on this point should state both standards, observe that advice built on the client’s disclosures is protected under either, and apply the standard the jurisdiction follows.

The privilege belongs to the client, so a lawyer who claims it acts only on the client’s behalf, and the personal representative of a deceased client may claim it.⁷⁰ The privilege can be claimed even after the client’s death.⁷¹

Agents and Consultants

Rule 501 leaves the content of the attorney-client privilege to the common law, and no enacted text lists the non-lawyers who may sit inside the privileged circle, so the answer comes from decided cases.⁷³ The anchor is United States v. Kovel, where the Second Circuit described two pulls in opposite directions. Courts resist expanding privileges because each one removes evidence from the fact-finder. Yet few modern lawyers could practice without secretaries, file clerks, telephone operators, messengers, clerks not yet admitted to the bar, and aides of other sorts, and a client who must speak to or through those aides needs the same assurance of secrecy that a private conversation with the lawyer would carry.⁷⁴ The court resolved the tension by holding that the privilege includes the persons who act as the lawyer’s agents, because their assistance is indispensable to the lawyer’s work and the client’s communications are often necessarily committed to them.⁷⁵

Kovel’s comparison of interpreter settings shows how far the reasoning runs. The most demanding setting is the last: the lawyer, ignorant of the client’s language, sends the client to a non-lawyer proficient in it, who interviews the client and gives the lawyer a summary so that the lawyer can give proper advice. The court saw no valid policy reason to treat that setting differently from one in which an interpreter merely translates, and it added that laymen consulting lawyers should not be expected to anticipate niceties perceptible only to judges.⁷⁶ Kovel itself involved an accountant. The client’s communications to an accountant employed by his lawyer were privileged because they were made so that the lawyer could understand the client’s situation and give legal advice.⁷⁷ The accountant did for the client’s finances what the interviewer did for the client’s language, converting a subject the lawyer could not read into one the lawyer could use.

The same reasoning marks the boundary. A client who first tells the story to his own accountant has no privilege for those communications, even if he later consults a lawyer on the same matter, while a client who goes first to a lawyer who retains an accountant as a listening post, or who consults the lawyer with his own accountant present, is covered.⁷⁸ Judge Friendly acknowledged that this draws what may seem a rather arbitrary line, and called it the inevitable consequence of reconciling the absence of a privilege for accountants with the effective operation of the lawyer-client privilege when the lawyer needs outside help.⁷⁹ The line follows function. Nothing in the policy of the privilege lets lawyers cloak a client’s communications to accountants, scientists, or investigators merely by putting them on the payroll, because the law has not extended a privilege to those people when they operate under their own steam.⁸⁰

Two facts separate the conversations in that scene: who retained the helper, and what the conversation was for. The privilege covers the first because a lawyer who needs help understanding the client’s world cannot get it unless the client can speak freely to the helper. The second scene lacks the lawyer’s engagement, so the bookkeeper has no privilege of his own and nothing connects the talk to the delivery of legal services. Kovel also left open whether a client’s communications to his own accountant could be protected when the accountant acts as the client’s agent to relay them to the lawyer, noting only that communications by a client’s agent to the attorney are privileged.⁸¹

In re Grand Jury Subpoenas carries the Kovel logic into public relations. Lawyers for the target of a federal grand jury investigation hired a public relations firm to help them deal with the media. The court accepted that the lawyers, who were not skilled at public relations, needed outside help, so the question became whether the problem on which they needed help related to legal advice.⁸² The firm was not retained to help the lawyers understand technical matters so they could state the law, as in Kovel, but the privilege already extends to consultants who help lawyers perform tasks beyond advising on the law. The government conceded that consultants advising on change of venue, juror attitudes, and how a client should behave while testifying fall within it.⁸³ The court held that the privilege protects communications that meet five conditions:

1. The communications are confidential.
2. They are between lawyers and public relations consultants.
3. The consultants were hired by the lawyers to assist them in dealing with the media in cases such as this one.
4. The communications are made for the purpose of giving or receiving advice.
5. The advice is directed at handling the client’s legal problems.⁸⁴

The reasons the court gave show why the extension is narrow. Lawyers cannot judge whether a public statement serves the client’s legal interest without skilled advice on how the press would report it, and the consultants cannot give that advice unless the lawyers may tell them non-public facts and defense strategy without fear that the consultants could be forced to repeat the conversation.⁸⁵ The usual purpose of promoting observance of law did little work, since the controversy concerned the consequences of past conduct and not an effort to conform present and future actions to the law’s requirements. The court rested instead on the proposition that encouraging frank communication among client, lawyers, and public relations consultants enhances the administration of justice.⁸⁶

The lawyer’s engagement is necessary but does not carry the privilege by itself. For a non-lawyer’s communications to be protected, the representative must be communicating with the lawyer or the client in order to assist the lawyer in delivering legal services,⁸⁷ and the specialist must be necessary to the legal representation.⁸⁸ Communications with a jury consultant can fall within the privilege, but communications with a public relations specialist are unlikely to be protected, because promoting the client’s public image is ordinarily not necessary to the legal representation.⁸⁹ The five conditions reconcile the two statements. They protect consultants hired by the lawyers to help deal with the media, for advice aimed at the client’s legal problem, and a specialist whom the client hires directly to polish a reputation meets none of them. An exam stem that mentions a public relations firm has therefore decided nothing yet. The facts that matter are who retained the firm and whether the lawyers needed its help to advise or defend the client.

Ordinary office help raises no such questions. A clerk or confidential secretary whom the lawyer calls in, an interpreter the client needs, and an agent the client brings to the conference are presumed not to defeat confidentiality, and the presumption does not depend on proving that each was necessary on the particular occasion.⁹⁰ Investigators fit the same pattern. In In re Kellogg Brown & Root, Inc., the district court tried to distinguish Upjohn because many of the company’s internal investigation interviews were conducted by non-attorneys. The D.C. Circuit disagreed, because the investigation ran at the direction of the attorneys in the company’s Law Department, and communications made by and to non-attorneys serving as agents of attorneys in internal investigations are routinely protected.⁹¹

The helper need not sit in the lawyer’s office or draw the lawyer’s pay. In City and County of San Francisco v. Superior Court, a physician who examined the client was held to be an intermediate agent for communication between the client and the attorneys, so the client could invoke the statutory attorney-client privilege. The physician had no privilege of his own, and the case involved no physician-patient privilege, which shows that the protection belonged to the client and passed through the physician as a channel.⁹²

Organizations and Government

An organization can speak only through its people, so the corporate client raises a question an individual client never does: when an employee talks to the company’s lawyer, whose communication is it? No constitutional provision, statute, or Supreme Court rule defines the privilege, so Rule 501 leaves the answer to the federal common law as United States courts interpret it in light of reason and experience.⁹⁴ Upjohn supplied the answer for corporations. The Court observed that it had assumed the privilege applies when the client is a corporation, and the Government did not contest the point, so what remained in dispute was whose words count as the corporation’s own.⁹⁵

The court of appeals had refused to extend the privilege beyond a control group, for fear that a broader rule would burden discovery and create a zone of silence over corporate affairs.⁹⁶ The Supreme Court rejected that limit because it overlooks half of what the privilege does. The privilege protects not only the giving of professional advice to those who can act on it but also the giving of information to the lawyer so that the lawyer can give sound and informed advice.⁹⁷ The second function matters inside a company because of where knowledge sits. Legal advice begins with ascertaining the facts, and the facts a lawyer needs about a questionable payment or a compliance failure are usually held by the managers and line employees who handled the matter, not by the senior officers who will receive the advice. In Upjohn, the information needed as a basis for advice on securities, tax, and foreign-law questions was not available from upper-echelon management.⁹⁸ A control-group rule would have let counsel hear candidly only from the people who knew least. As for the zone of silence, applying the privilege to these communications puts the adversary in no worse position than if the communications had never taken place.⁹⁹

The Upjohn record shows what the Court protected in place of the control-group test. After consulting the chairman of the board and outside counsel, general counsel Thomas investigated questionable payments by sending questionnaires to managers of the company’s foreign affiliates and interviewing officers and employees, so that he could give the company legal advice about them.¹⁰⁰ The communications were privileged for several connected reasons. Employees spoke to counsel acting as such, at the direction of corporate superiors, to secure legal advice. The subjects were within the scope of the employees’ corporate duties. The employees knew the questioning was meant to let the corporation obtain legal advice, because the questionnaire identified Thomas as the company’s General Counsel and referred to the possible illegality of the payments.¹⁰¹

Chief Justice Burger, concurring, would have turned that reasoning into a general rule so that corporations, counsel, and courts could predict what is protected. He proposed that a communication is privileged at least when an employee or former employee speaks, at management’s direction, with an attorney authorized by management to inquire, about conduct within the scope of employment, and the attorney seeks information to evaluate whether the conduct binds the corporation, to assess its legal consequences, or to formulate responses to what others may do about it.¹⁰² The limits built into that formulation explain why a broad rule does not let a company shelter its records by routing them through the legal department. A routine report that counsel merely receives was not made for legal advice, and an employee who is only a fortuitous witness is outside the corporate duties the privilege covers.

The last step of the checklist repeats the limit that runs through this chapter: the privilege shields the communication to counsel and leaves the underlying facts open. In Upjohn, the privilege protected disclosure of communications but not disclosure of the underlying facts by those who communicated with the attorney.¹⁰³ The opposing party therefore loses access only to the questionnaire answers and interview notes themselves, and the events they describe remain available from every other source.

The same reasoning disposes of the argument that a company’s own lawyers offer less protection than outside counsel. In re Kellogg Brown & Root, Inc., the district court tried to distinguish Upjohn on the ground that the investigation there had involved outside counsel, while this one was run by the company’s in-house Law Department. The D.C. Circuit answered that Upjohn does not hold or imply that outside counsel’s involvement is a necessary predicate for the privilege, and that under the general rule a lawyer’s status as in-house counsel “does not dilute the privilege.”¹⁰⁴ The rule follows from what gives a communication its legal character, which is the function the lawyer performs and the capacity in which the lawyer acts, and not the identity of the lawyer’s employer. A contrary rule would make the lawyers who know the company best the least safe confidants and would push companies to hire outside firms for every sensitive inquiry. In-house status therefore neither strengthens nor weakens the claim, which still depends on the communication having been made to obtain or give legal advice, the requirement taken up in the next section.

Government clients receive a more guarded version of the same protection. The Supreme Court’s clearest statement comes from civil litigation. In United States v. Jicarilla Apache Nation, a tribe sought documents about the government’s administration of tribal trusts and argued that a fiduciary exception overrode the government’s privilege. The Court reasoned that the objectives of the privilege apply to governmental clients, because the privilege helps government entities and employees obtain legal advice founded on a complete and accurate factual picture. It stated that, unless applicable law provides otherwise, the Government may invoke the privilege in civil litigation to protect confidential communications between Government officials and Government attorneys, and it reversed the order requiring disclosure to the tribe.¹⁰⁵

That civil default narrows when the demand comes from a grand jury investigating possible crimes. In re Lindsey, the Independent Counsel sought the testimony of Lindsey, a White House lawyer, before a federal grand jury about his communications with the President, and the Office of the President asserted the government attorney-client privilege.¹⁰⁶ The D.C. Circuit treated the governmental body as a potential client, noting that the proposed privilege rule’s definition of client includes governmental bodies.¹⁰⁷ It nevertheless held that when an executive branch attorney is called before a federal grand jury to give evidence about alleged crimes within the executive branch, “reason and experience, duty, and tradition dictate that the attorney shall provide that evidence.”¹⁰⁸ The court reasoned that government lawyers stand in a far different position from private lawyers in this setting, because their duty is not to defend clients against criminal charges or to protect wrongdoers from public exposure, and because the constitutional responsibility of the President and all members of the Executive Branch is to take care that the laws be faithfully executed.¹⁰⁹ The private privilege encourages a client to confide past misconduct to a lawyer who will help defend against its consequences. Applying it to a government lawyer in this setting would let the executive branch withhold evidence of its own officials’ crimes from a grand jury, and that conflict shifts the balance.

General statements of the law reflect the same pattern: a government attorney’s lawyer-client privilege has a narrower scope when information is sought by a grand jury investigating possible wrongdoing.¹¹⁰ Jicarilla and Lindsey therefore mark the two poles. Jicarilla concerned civil litigation over documents, with no suspicion of crime by officials, and the government’s privilege held. Lindsey concerned a grand jury looking into suspected crimes within the executive branch, and the privilege gave way. The hinge is what the proceeding is investigating, because the government is the client in both. An answer that denies a government client any privilege is wrong in civil litigation, and an answer that gives a government client the full private privilege against a grand jury inquiry into official wrongdoing overlooks the line these decisions draw.

Legal Advice Purpose

The point that an in-house lawyer’s employment status neither strengthens nor weakens a claim of privilege leaves the purpose of the communication as the question that decides it. Because Rule 501 leaves the content of the privilege to the federal common law, the requirement that a communication seek or give legal advice comes from decided cases.¹¹² The most developed modern statement comes from In re Kellogg Brown & Root, Inc., where the D.C. Circuit held that the privilege applies if obtaining or providing legal advice was one of the significant purposes of the communication.¹¹³ That formulation matters because many communications with a lawyer serve some other end as well, and a court must decide how much legal purpose is enough. The discussion below takes the basic requirement first, then the mixed-purpose problem the significant-purpose test solves, and then the communications that fall on the unprotected side of the line.

The basic requirement is that the client seek legal advice from a professional legal adviser acting in that capacity. As In re Lindsey states the standard formulation, only communications that seek “legal advice” from “a professional legal adviser in his capacity as such” are protected, and the party claiming the privilege bears the burden of proving that its communications fit that description.¹¹⁴ The requirement keeps the privilege tied to its reason. The privilege buys candor so that a lawyer can give legal help, and a lawyer consulted as a friend, a business adviser, a banker, a negotiator, an accountant, or a mere scrivener is supplying something else, so the consultation is not professional and the statements are not privileged.¹¹⁵ If a lawyer’s license alone were enough, a client could shelter any business conversation by choosing a lawyer as the listener, and the privilege would withhold evidence while protecting no candor that legal judgment needed. The inquiry therefore looks at what the communication was for, and the recipient’s bar membership does not settle it. Lindsey makes the same point for a government client: advice on political, strategic, or policy issues, however valuable, is not shielded by the privilege.¹¹⁶

United States v. Lawless shows the same limit in private practice. A lawyer who prepared a tax return asserted the privilege over two documents the client had transmitted to him, and the Seventh Circuit observed that several courts have treated tax-return preparation as primarily an accounting service, even though it calls for some knowledge of the law.¹¹⁷ The court added that, even assuming the lawyer gave legal advice in addition to preparing the return, nothing in the record showed that the two documents had been given to him for any purpose other than preparing it.¹¹⁸ That second point reaches beyond tax. A lawyer’s dual role does not carry the privilege from one task to the other, so protection is assessed communication by communication, and the proponent must connect the particular communication to the legal task.

Both letters went to the same lawyer in the same week, and the only difference is the function requested. A rule that protected the first letter because a lawyer happened to open the folder would let any client shelter ordinary paperwork by choosing a lawyer to handle it. The second letter asks for exactly what the privilege exists to encourage clients to seek candidly.

Real communications seldom serve a single end, and the difficulty begins when the legal purpose shares the communication with another one. A company asks counsel whether a transaction is lawful and also whether it makes sense, a regulator requires a review the executives would want anyway, and legal advice often arrives wrapped in commercial context. Kellogg Brown & Root is the leading answer. The contractor’s Law Department directed an internal investigation of alleged misconduct, and the investigation was conducted under a company compliance program required by regulation.¹¹⁹ The district court said that the primary purpose of a communication is legal advice only if the communication would not have been made “but for” the fact that legal advice was sought, found that the investigation was undertaken to comply with regulatory requirements and corporate policy, and denied the privilege.¹²⁰

The D.C. Circuit called that analysis a false dichotomy and rejected the but-for test as inconsistent with Upjohn and longstanding privilege law. So long as obtaining or providing legal advice was one of the significant purposes of the investigation, the privilege applies, even if there were also other purposes and even if regulation, and not company discretion, mandated the investigation.¹²¹ The consequences explain the rejection. Businesses would be less likely to disclose facts to their attorneys and to seek legal advice, which would limit the valuable efforts of corporate counsel to ensure their client’s compliance with the law, the work Upjohn sought to protect.¹²² A rule that made protection depend on the absence of any other motive would penalize the most heavily regulated companies, because the law itself supplies the second motive for the very investigation that the law calls for.

The replacement test asks whether obtaining or providing legal advice was a primary purpose of the communication, meaning one of the significant purposes.¹²³ The court explained why the older primary-purpose phrasing needed that gloss. Many courts have used it to resolve disputes over communications with both legal and business purposes, yet trying to find the one primary purpose of a communication driven by overlapping purposes is often an impossible task.¹²⁴ The court took its formulation from the Restatement’s reporter’s note, which reports general agreement among American decisions that the privilege applies if one of the significant purposes of the client in communicating with a lawyer is obtaining legal assistance.¹²⁵ For an internal investigation the test works out this way: if one of the significant purposes was to obtain or provide legal advice, the privilege applies, whether the investigation followed a program required by statute or regulation or was simply company policy. The court found no serious dispute that the contractor’s investigation met that standard.¹²⁶

The significant-purpose standard avoids two errors that run in opposite directions. A sole-purpose rule fails the compliance work just described. A rule that protected any communication with a trace of legal content would turn the privilege into a label companies could attach to ordinary business discussion, and the evidence it withheld would no longer be tied to any candor the privilege was meant to produce. Asking whether the legal purpose was a significant one protects integrated advice and leaves the incidental mention of counsel outside the privilege. Courts phrase the standard differently, so an answer should state the test the jurisdiction follows and then apply it to the particular communication.

Where lawyers direct the work, courts also lean toward protection. In Diversified Industries, Inc. v. Meredith, the Eighth Circuit treated a matter committed to a professional legal adviser as prima facie committed for the sake of legal advice, and therefore within the privilege absent a clear showing to the contrary. The law firm there was authorized to interview any employee, from the President to the nonmanagerial staff, to analyze the accounting data, to evaluate the propriety of past actions, and to recommend future courses, which required an independent analysis that accountants or lay investigators lacked the training to provide.¹²⁷ An investigation that feeds legal advice and requires that kind of analysis counts as legal work.

The significant-purpose test protects integrated legal advice. It does not convert the business reasons around that advice into privileged material. SCM Corp. v. Xerox Corp. drew the line in a deposition setting: protecting the business components of a decision would distort the privilege, which was not intended, and is not needed, to encourage businesspeople to discuss the business reasons for a particular course of action.¹²⁸ A client who acts on legal advice does not thereby expose the advice, but the business reasons executives weigh among themselves are like any other business evaluations, and they do not become privileged because counsel alluded to them.