Concept
Elements of the Privilege
Attorney-client privilege
Protecting Lawyer-Client Communications
Elements of the Privilege
Federal Rule of Evidence 501 sends federal courts to the common law of privilege, as federal courts interpret it “in light of reason and experience.”1 Applying that standard, Upjohn Co. v. United States called the attorney-client privilege the oldest of the privileges for confidential communications known to the common law.2 The Court also named the purpose that organizes everything below. The privilege encourages full and frank communication between attorneys and their clients, and so promotes broader public interests in the observance of law and the administration of justice, because sound legal advice or advocacy depends on a lawyer who has been fully informed by the client.3 Each element in this section earns its place by serving that purpose, and each limit marks where the purpose stops.
The client, not the lawyer, owns the privilege. The client may refuse to disclose protected communications and may prevent anyone, the lawyer included, from disclosing them; the client also holds the ultimate authority to waive.4 A lawyer who asserts or waives the privilege acts as the client’s agent and not under any authority of the lawyer’s own.5 The question for this section is therefore narrow: what must be true of a conversation before it belongs to the client in this way?
The privilege also has a price. It can keep potentially critical evidence from the factfinder, and the D.C. Circuit has acknowledged that this may be the practical result in a given case; the legal system tolerates the cost because the privilege encourages full and frank communication between lawyers and clients.6 Two justifications explain why the trade is worth making. The instrumental justification holds that the privilege makes possible the very communications it protects, so the truth-seeking process loses nothing, since those communications would never be made without it, while the legal system gains the superior advocacy that follows a full exchange of information.7 The humanistic justification stresses the fairness of a zone of privacy for lawyer and client and holds it profoundly wrong, even immoral, to force a trusted counselor to betray a client’s confidences.8 Because a privilege withholds evidence, a claim of privilege remains open to challenge. The person asserting it must show that the consultation was professional,9 and if a trial judge wrongly upholds a claim and excludes unprivileged evidence, the disadvantaged party may object whoever holds the privilege, and an appellate court can correct the error by holding the claimed privilege inapplicable.10
Put together, the privilege protects a confidential communication between a lawyer and a client made to obtain or provide legal advice. The Restatement breaks that sentence into four parts: (1) communications (2) made between privileged persons (3) in confidence (4) for the purpose of obtaining or providing legal advice for the client.11 Another common statement lists a communication, in confidence, between lawyer and client, in the course of providing legal services, and warns that some of these elements overlap.12 The four subsections below take them in turn: communications, confidentiality, lawyer and client, and legal advice purpose. At the poles the cases are easy. A confidential communication from a corporate CEO to the company’s attorney, made to get legal advice for the corporation, is protected; a routine statement that a lower-level employee prepares in the normal course of business is not, even if someone later sends it to corporate counsel.13 The difficulties lie between those poles, especially in corporations, where statements may pass through many hands before reaching counsel and the line between legal and business advice often blurs.14
Communications, Not Facts
The privilege protects what a client says to a lawyer and leaves alone what the client knows. Upjohn drew that line by quoting an earlier decision: “A fact is one thing and a communication concerning that fact is an entirely different thing.” The 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 included that fact in a communication to his attorney.15 A question about what was said targets the communication, and the privilege blocks it. A question about the underlying fact targets something within the client’s own knowledge, and the privilege does not reach it.
Two reasons support the line. First, limiting the privilege to communications puts the adversary in no worse position than if the communications had never taken place: the privilege bars disclosure of the communications but not of the underlying facts, even when those facts are sought from the people who spoke with the lawyer.16 Second, the opposite rule would invite abuse. If telling a lawyer a fact made the fact privileged, any client could put damaging information out of reach by describing it to counsel, and the privilege would protect evidence instead of communications. The error to avoid on an exam is treating a fact as privileged because the client mentioned it to a lawyer; the fact keeps whatever status it had before the conversation.
The facts of Upjohn show the line in operation. The company’s lawyers had gathered information from employees through questionnaires, and the Government sought the questionnaires and the attorneys’ notes. The Court found that the employees’ communications were made to counsel for the company, acting as such, at the direction of corporate superiors, to secure legal advice.17 The responses were therefore privileged, and the Government’s convenience in subpoenaing the questionnaires, instead of pursuing the facts another way, did not overcome the policies the privilege serves.18 The employees remained available as sources of the facts they knew.19 How employees come to count as the corporate client is the subject of the next section; here the only point is what the privilege leaves open to the adversary.
The same logic governs documents. A routine statement that a lower-level employee prepares in the normal course of business does not become protected because someone later sends it to corporate counsel.20 If delivery to a lawyer were enough, any client could shield an unwelcome record by handing it over. New communications can form around the document, however. A confidential letter asking the lawyer how the records affect the client’s position is itself a communication made to obtain legal advice, and the lawyer’s confidential advice about the records is protected too. The records themselves remain discoverable.
Confidentiality
A standard definition builds confidentiality into the privilege itself: a communication is confidential if it was 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 to the client or those reasonably necessary for the transmission of the communication.23 The cases add two requirements that sit beside that definition: the client must have had a reasonable expectation of privacy in the communication and must have preserved its confidentiality afterward.24 A communication that fails any of these is unprotected, however sensitive its content and however plainly the client sought legal advice.
Confidentiality is both the premise of the privilege and its limit. On the instrumental rationale, the privilege makes possible the very communications it protects,25 so a client who is indifferent to who hears has little need of its promise. The circle of permitted listeners cannot shrink to lawyer and client alone, however, because legal work needs helpers. The definition therefore draws the circle by function and not by affection.
Persons present in furtherance of the legal services, or reasonably necessary to transmit the communication, stay inside the circle; everyone else is a third person.26 A friend who comes along for moral support and plays no part in the legal work is outside it, so what the client says in the friend’s presence is not confidential. A secretary taking notes or an interpreter conveying the client’s words is inside. Judge Friendly’s opinion in United States v. Kovel saw no significant difference among four settings: the lawyer sends a client who speaks a foreign language to an interpreter for a literal translation; the lawyer, who knows a little of the language, keeps a more knowledgeable non-lawyer employee in the room; the client brings along someone to perform the same function; and the lawyer, ignorant of the language, sends the client to a non-lawyer proficient in it, who interviews the client on the lawyer’s behalf and summarizes the situation so that the lawyer can give proper advice.27 In each setting the helper carries the client’s words to a lawyer who needs them to advise, and the court did not treat it as important who brought the helper.
Necessity decides who belongs. The interpreter is the means by which a lawyer who does not speak the client’s language receives the communication at all; the taxi driver contributes nothing to the representation.
Outside specialists test the line. One court held that (1) confidential communications (2) between lawyers and public relations consultants (3) hired by the lawyers to assist them in dealing with the media in cases such as the one before it (4) made for the purpose of giving or receiving advice (5) directed at handling the client’s legal problems are protected by the attorney-client privilege.28 Its reasoning was practical: the lawyers could not hold meaningful discussions with the consultants unless they could share some non-public facts and their defense strategies and tactics without fear that the consultants could be forced to disclose those discussions.29 The court conceded a certain artificiality in letting the privilege turn on who does the hiring, and it justified the result by pointing to Judge Friendly’s Kovel opinion, which reconciles the absence of a privilege for accountants with the effective operation of the lawyer-client privilege where the lawyer needs outside help.30 The holding is narrow, and its five parts are the limits to remember.
The same reasoning reaches an accountant. If the lawyer has directed the client to tell her story first to an accountant the lawyer engaged, who will interpret it so that the lawyer can better give legal advice, the client’s communications reasonably related to that purpose fall within the privilege. What is vital is that the communication be made in confidence for the purpose of obtaining legal advice from the lawyer.31
The objective side of confidentiality also governs eavesdroppers. The old rule let a third party testify to a communication in circumstances where the lawyer and client themselves need not, and most courts have now abolished it.32 Under the modern view, if the client took reasonable measures to preserve confidentiality, the material stays protected even when an unauthorized person obtains it.33 What reasonableness requires is still being worked out in litigation, and it may differ for a lawyer and for an unsophisticated client; sophistication, expense proportioned to the size of the case, and the resources of the parties all play a role.34 Law and practice are likewise evolving on which steps to prevent electronic eavesdropping are reasonable.35
The inmate loses the privilege because his expectation of secrecy was unreasonable, not merely because someone listened. A client whose private call is secretly intercepted keeps the privilege, because her expectation was reasonable even though events disappointed it.
Joint clients present the last variation. The trend is to treat joint-client material as generally confidential but not confidential as between the joint clients themselves, so it should not be disclosed to outsiders unless the two clients agree to waive or are litigating against each other, in which case it is generally held unprivileged.36 The reason follows from the premise of the privilege: joint clients chose to speak in each other’s presence and to share the same advice, so none of them was promised secrecy from the others.
Lawyer and Client
A classic formulation, quoted in federal opinions, asks whether legal advice of any kind was sought from a professional legal adviser in that capacity, and whether the communications relating to that purpose were made in confidence by the client and are, at the client’s instance, permanently protected from disclosure by the client or the legal adviser unless the protection is waived.39 Two roles carry that sentence, the lawyer and the client. This subsection asks who counts as each, which direction the protection runs, and when a person with a law license is not acting as a lawyer at all.
The lawyer is defined from the client’s side of the desk. The privilege hinges on the client’s belief that she is consulting a lawyer in that capacity and on her manifested intention to seek professional legal advice, and it is enough that she reasonably believes the person consulted is a lawyer, though in fact the person is not.40 The cases also speak of a genuine lawyer-client relationship,41 but under this formulation the relationship turns on what the client reasonably believes she is doing, not on whether the adviser was ever admitted to practice. The rule fits the privilege’s purpose. A client cannot check the bar rolls before each conversation, and a promise of secrecy that depends on facts the client cannot verify would induce little candor. The word reasonably supplies the limit: a client who knows the adviser is not a lawyer has no belief for the privilege to protect. Courts have disagreed about whether the privilege covers communications to a non-lawyer who practices before an administrative agency.42
The lawyer’s role reaches the people through whom the lawyer works. Lawyers collect evidence personally or through assistants such as clerks and private investigators.43 Communications made by and to non-attorneys serving as agents of attorneys in internal investigations are routinely protected,44 and the interpreters in Kovel follow the same logic.45 The helper must be working for the lawyer in rendering legal services to the client. A client’s own accountant, whom the lawyer did not engage, does not qualify, because there is no privilege for accountants as such; the privilege reaches an accountant only where the lawyer needs outside help to give legal advice.46
The client is a wider category than the people who have signed an engagement letter. A client is a person or entity who is rendered professional legal services by a lawyer or who consults a lawyer with a view to obtaining them, and the category includes partnerships, associations, corporations, and governmental officers and agencies as well as individuals.47 The privilege does not depend on payment of legal fees or on a formal contract between client and attorney.48 Communications in preliminary discussion with a view to employing the lawyer are privileged although the employment is never accepted,49 and the privilege applies to an initial consultation even if the client or the attorney declines the representation.50 The rule follows the order of events. A person deciding whether to hire a lawyer must describe the problem first, and a lawyer deciding whether to accept the case must hear it first, so the first meeting is often the most revealing one. If protection began only at the retainer, the client with the most to hide would have the strongest reason to stay vague.
Protection runs in both directions. A lawyer’s advice almost invariably responds to the client’s requests and explanations, so revealing the advice necessarily discloses the client’s thoughts as conveyed to the lawyer, and the privilege is properly recognized as adhering to both sides of the conversation.51 The modern understanding is that the privilege includes both upstream and downstream communications.52 Without that rule, an opponent could subpoena the lawyer’s responses and infer from them the general content of what the client confided.
Because the client holds the privilege, courtroom practice follows the client’s lead. If the client is present when arguably privileged communications are offered into evidence and neither the client nor the lawyer objects, the privilege is deemed waived; if the client is absent, the lawyer has a duty to object on the client’s behalf.53
A law license does not turn every conversation with the licensee into a privileged one. The privilege attaches when the client consults the person as a lawyer, and it does not attach when the person is consulted as a friend, business adviser, banker, negotiator, accountant, mere scrivener, attesting witness, executor, or agent.54 The topic of the communication must also be connected with obtaining legal counsel,55 and attorneys engaged in a corporation’s business as opposed to professional legal activities frequently find the resulting communications held not privileged.56 A lawyer’s license is not the test. The test is the capacity in which the client consulted the person.
Legal Advice Purpose
In re Kellogg Brown & Root, Inc., the D.C. Circuit stated the purpose element simply: the privilege applies to a confidential communication between attorney and client if the communication was made for the purpose of obtaining or providing legal advice to the client.59 Earlier subsections asked whether the person consulted was a lawyer acting as a lawyer. Even between a genuine lawyer and a genuine client, a particular communication may serve no legal end, and then the privilege does not reach it. The difficulty is that most communications involving lawyers serve several ends at once, so the working question is how much legal purpose is enough.
Most clients bring a lawyer a problem that has business stakes beside the legal ones, and the lawyer needs the business facts to advise well. A rule that withdrew the privilege whenever any non-legal purpose appeared would fail in the settings where advice is most needed. A rule that protected anything sent to a lawyer would let a client shield ordinary records by copying counsel. Courts sit between the two and ask whether the legal purpose was a sincere and meaningful element of the overall exchange.60
Kellogg Brown & Root put that question in a regulated setting. KBR, a defense contractor, conducted an internal investigation of allegations of wrongdoing at the direction of the attorneys in its Law Department,61 and Department of Defense regulations require contractors like KBR to maintain compliance programs and to investigate such allegations.62 The district court ordered production of the investigation documents. It reasoned that KBR had not shown that the communication would not have been made “but for” the fact that legal advice was sought,63 and it concluded that the purpose of the investigation was to comply with regulatory requirements rather than to obtain or provide legal advice.64
The D.C. Circuit granted mandamus and vacated the production order, holding the district court’s decision irreconcilable with Upjohn.65 It called the district court’s contrast between compliance and legal advice a false dichotomy: so long as obtaining or providing legal advice was one of the significant purposes of the investigation, the privilege applies, even if there were other purposes and even if regulation mandated the investigation.66 It then restated the test as a single question: was obtaining or providing legal advice a primary purpose of the communication, meaning one of the significant purposes of the communication? The Restatement’s Reporter’s Note agrees that the privilege applies if one of the significant purposes of a client in communicating with a lawyer is that of obtaining legal assistance.67
The court’s reasoning is practical. A communication motivated by two overlapping purposes, one legal and one business for example, often cannot be sorted into a single primary one, and it is often not useful or even feasible to ask whether the purpose was A or B when it was A and B. A court should therefore not presume that a communication can have only one primary purpose.68 In an organization’s internal investigation, the same test applies: if one of the significant purposes was to obtain or provide legal advice, the privilege applies whether the investigation followed a compliance program required by statute or regulation or was otherwise conducted under company policy.69 The usual exam error is concluding that a regulatory or commercial motive defeats the privilege, when the only question is whether legal advice was also a significant purpose.
The district court also pointed out that in Upjohn attorneys conducted the interviews, while many of KBR’s interviews were conducted by non-attorneys. The D.C. Circuit answered that the investigation here proceeded at the direction of the attorneys in KBR’s Law Department, and that communications made by and to non-attorneys serving as agents of attorneys in internal investigations are routinely protected.70
Other courts reach similar results by asking about dominant intent or primary purpose, but they generally apply a qualitative approach and do not count how many paragraphs concern law and how many concern money. One federal court of appeals upheld protection over corporate committee meeting minutes because the matter was “infused with legal concerns.” If a later observer could strip the privilege by deciding that non-legal factors outweighed legal ones on a numerical scale, laypeople would be discouraged from seeking legal advice on exactly the mixed matters that need it.71 Where the dominant intent of the client is to obtain legal advice, the privilege is not lost because advice is also sought on related business, financial, or other not strictly legal matters.72 Return to the two dentists. The structure of their combined offices and the general terms of a sale contract were intimately associated with rendering legal services, and a side question about financing would not change the result.73
The privilege fails when the legal element is absent. Statements by clients seeking advice about how to commit crimes or frauds are not privileged, and attorneys engaged in activities of a business as opposed to a professional legal nature frequently find that the communications made in connection with them are not privileged.74
The two documents differ in why they exist. A report generated in the ordinary course of business exists whether or not a lawyer ever sees it, so copying counsel adds nothing to its purpose. The memo exists because someone wants legal advice. The contrast with Kellogg Brown & Root is therefore not that one document was required and the other voluntary, since the investigation there was mandated by regulation, but that legal advice was a significant reason for the investigation and played no part in why the safety report was written.