Concept
Sources of Privilege Law
Governing law and scope of the attorney-client privilege
Sources of Privilege Law
Before a court can ask whether a communication is privileged, it must decide which body of law supplies the answer. In federal court, Federal Rule of Evidence 501 makes that decision. Its first sentence sends a claim of privilege to the common law, as United States courts interpret it in the light of reason and experience, unless the United States Constitution, a federal statute, or rules prescribed by the Supreme Court provide otherwise. Its second sentence creates an exception: in a civil case, state law governs privilege regarding a claim or defense for which state law supplies the rule of decision.1 Rule 501 therefore contains no list of privileges and no definition of the attorney-client privilege. It directs courts to one of two destinations, federal common law or state law.2
The two-track design makes sense once you see what it replaced. The Supreme Court had proposed a series of rules that would have codified nine specific privileges, the attorney-client privilege among them, and Congress rejected them and enacted Rule 501 in their place.3 The proposed rules defined the privileges the Advisory Committee thought should be recognized and expressly refused to honor broader state privileges, an intrusion on state policy that became a major source of controversy.4 Rejecting them did not disapprove the privileges they described, for Congress chose to let privilege law evolve by judicial decision.5 Congress also took privilege rulemaking away from the Court: a rule “creating, abolishing, or modifying an evidentiary privilege” has “no force or effect” unless Congress enacts it.6
Each sentence of Rule 501 answers one of those problems. The first leaves privilege with the federal courts, which refine it case by case. The second protects state policy where a state created the right being litigated, because the committee behind the state-law proviso saw no federal interest strong enough to justify departing from state policy when a civil claim or defense does not rest on a federal question.7
The second sentence speaks of “a claim or defense,” so a court makes the choice claim by claim and issue by issue. Labeling the whole lawsuit “federal” or “state” skips a step the rule requires, because the question is what supplies the rule of decision on the issue to which the communication relates.8 The four subsections below follow that analysis in order: the federal default, the state-law exception, the case in which one communication bears on both a federal and a state claim, and the choice among states once state law governs.
Federal Common Law Default
The first sentence of Rule 501 sets the default. Federal courts decide a claim of privilege under the common law, meaning the principles judges develop through decided cases, as United States courts interpret it in the light of reason and experience. The Constitution, a federal statute, or a Supreme Court rule displaces that law only when one of them actually speaks to the claim.9
The default governs matters in which federal law supplies the rule of decision, which generally means criminal cases, civil actions brought by the United States, and private federal-question cases.10 It reaches grand jury proceedings too. Rule 1101(d)(2) makes the other Rules of Evidence inapplicable to grand jury proceedings “except for those on privileges,” so a federal grand jury witness's privileges are determined by federal law.11
Federal law governs these matters because states should not decide when a federal court must refrain from hearing useful testimony in a matter involving federal law.12 Federal criminal practice followed this approach before the Evidence Rules existed. Former Criminal Rule 26 governed the privileges of witnesses, absent a federal statute, by “the principles of the common law as they may be interpreted by the courts of the United States in the light of reason and experience,” and the courts made almost no use of state evidence rules in criminal cases.13 The first sentence of Rule 501 repeats that formula.
The three sources that can displace the common law rarely matter to an attorney-client claim. Apart from a constitutional privilege, such as the one against self-incrimination, and a rare federal statutory privilege, the focus in a federal proceeding is on the common law.14 The attorney-client privilege belongs to that common-law category, so a court deciding such a claim in a federal-law matter works from decided cases instead of an enacted definition. An answer that begins by searching for a federal statute defining the privilege begins in the wrong place.
“Reason and experience” does not freeze the privilege at the date Rule 501 took effect. The rule was not intended to freeze privilege law, and Congress gave the courts flexibility to develop privileges case by case.15 Because the common law grows through interpretation, it can add privileges in situations that did not exist in the past, as it did when courts recognized a federal psychotherapist-patient privilege.16 In Upjohn Co. v. United States, the Supreme Court applied the same method to the attorney-client privilege. It accepted that a case-by-case approach may undermine certainty at the privilege's boundaries to some slight extent, found the approach consistent with the spirit of the Rules, and held that the narrow “control group test” could not, consistent with the common law interpreted in the light of reason and experience, govern the development of the law in this area.17 No enacted text therefore fixes the edges of the privilege. Decided cases do.
Decided cases also hold the doctrine's details. The Supreme Court has assumed, for example, that the privilege applies when the client is a corporation, even though a corporation is an artificial creature of the law and complicates the privilege's application.18 The cases draw a firm line for accountants as well. A client who consults her own accountant first has no privilege for those communications, even if she later consults a lawyer on the same matter. A client whose lawyer retains an accountant as a “listening post,” or who consults the lawyer with her own accountant present, can keep the privilege. The line may seem arbitrary, but it is “the inevitable consequence of having to reconcile the absence of a privilege for accountants and the effective operation of the privilege of a client and lawyer under conditions where the lawyer needs outside help.”19 Later sections return to both topics.
Courts use that flexibility cautiously. The Supreme Court has said it is “disinclined to exercise this authority expansively.”20 The caution looks different when a party asks a court to narrow a privilege instead of creating one. In Swidler & Berlin v. United States, which asked whether the attorney-client privilege survives the client's death, the Independent Counsel cited United States v. Nixon and Branzburg v. Hayes to argue that privileges should be strictly construed because they conflict with truth-seeking. The Court answered that both cases dealt with creating privileges the common law had not recognized, whereas the attorney-client privilege is one of the oldest in the law, and the request was not simply to construe the privilege but to narrow it against the weight of existing case law.21 The Court also described the privilege as intended to encourage full and frank communication between attorneys and their clients, and it decided the case by interpreting the common law under Rule 501 in the light of reason and experience.22
Cases from other federal matters show the default at work. In re Grand Jury Subpoenas, the court found the matter “entirely federal in nature,” held that Rule 501 governed the privilege's scope, and looked “principally to decisions applying the federal common law of attorney-client privilege.”23 In In re Kellogg Brown & Root, Inc., the D.C. Circuit called the company's privilege claim “materially indistinguishable” from Upjohn's. As in Upjohn, the company started an internal investigation to gather facts and ensure compliance with the law after learning of potential misconduct, and its in-house legal department, acting in its legal capacity, conducted the investigation.24 The court added that Upjohn does not make outside counsel a necessary predicate for the privilege, because a lawyer's status as in-house counsel “does not dilute the privilege.”25
The same lawyer-client conversation would face a different body of law in a state-law diversity suit. What changes the governing law is the source of the rule of decision for the claim, which is why Rule 501 cannot be applied by looking only at the basis for the court's jurisdiction.26
Proposed Rule 503, the Supreme Court's detailed lawyer-client rule, never became law, yet courts consult it constantly. Congress made clear that leaving privilege law to judicial development did not disapprove the proposed privileges, and Proposed Rule 503 has served as an influential standard in federal and state courts that has encouraged greater uniformity across jurisdictions.27 Courts treat the unenacted standard as a useful guide that, with a few minor exceptions, restates the common-law lawyer-client privilege.28 In the Eighth Circuit, precedent describes Rule 503 as “an accurate definition of the federal common law of attorney-client privilege.”29 The proposed rule is thus a persuasive reference for the privilege's contours. It does not bind a court on a question its text leaves open.
State Law as Rule of Decision
The second sentence of Rule 501 reverses the default for one category of civil cases. In a civil case, state law governs privilege regarding a claim or defense for which state law supplies the rule of decision.31 The rule of decision is the substantive law that determines who wins the claim or defense, so the sentence asks which sovereign's law created the right being litigated. When state law created it, state privilege law controls, as usually will be true in diversity cases. In other civil cases and in criminal cases, Rule 501 calls for the federal law of privileges.32 Erie supplies the background: under Erie Railroad Co. v. Tompkins, federal courts apply state substantive law in diversity cases, and privilege is generally viewed as substantive.33
Diversity is the usual example, but it cannot be the test, because the rule speaks of a particular claim or defense. State law is looked to in some federal-question cases, and federal law controls some issues in diversity cases, so the court asks which body of law supplies the rule of decision on the issue at hand instead of reading the answer from the basis of jurisdiction.34
The clean case is the one in which the labels agree. A Virginia plaintiff sues a Delaware defendant in federal court over an automobile collision in Virginia, and jurisdiction rests on diversity. Virginia substantive law applies, so the federal judge also applies Virginia's law of privilege, including any Virginia rule that defers to the privilege law of another state, such as Delaware's.35
The Virginia suit and the Texas suit look alike in the one respect that might tempt a shortcut, because the parties in each are citizens of different states. They differ in where the right being enforced comes from, and that difference moves the privilege question from one track to the other.
Three reasons support making privilege follow the claim. The first is the nature of a privilege. Most evidence rules aim, at least in part, to improve the accuracy of factfinding, and they are generally treated as procedural, coming from forum law alone. Privileges aim at nothing of the kind: they impair accuracy, and they exist not because of their effect on factfinding but in spite of it.36 For claims and defenses governed by state law, Rule 501 therefore treats privileges the way the second sentence of Rule 601 treats competence: as a matter of substance rather than procedure.37 And where a civil claim or defense is not grounded on a federal question, the committee behind the proviso saw no federal interest strong enough to justify departing from state policy.38
The second reason is reliance. A state's effort to encourage confidential relationships is hampered if the secrets of the attorney's office can be kept in state court but disclosed because, by the accident of diversity, suit is brought in a federal court a block away.39 A client decides what to tell a lawyer before knowing where any suit will be filed, and protection that shifts with the forum cannot encourage the full and frank communication the privilege exists to promote.40
The third reason is forum shopping. The committee observed that the Court's proposed Article V would have promoted forum shopping in some civil actions, depending on differences in the privilege law the state and federal courts applied. Binding federal courts to state privilege law in actions founded on a state-created right or defense removes the incentive to shop.41 Congress used the same logic for witness competence: the second sentence of Rule 601 ensures that state dead man statutes operate in any federal case governed by state substantive law.42
The alternatives show what the second sentence accomplishes. The Advisory Committee's rules would have refused to honor broader state privileges even where state law supplied the rule of decision.43 Before the Evidence Rules, former Civil Rule 43(a) left admissibility, outside its first two branches, to the rules of evidence of the state in which the federal court sat.44 Rule 501's second sentence replaces that general borrowing with a rule tied to the claim: state law governs privilege where state law supplies the rule of decision.
Baird v. Koerner, decided long before Rule 501, shows the intuition behind the state-law sentence. The court reasoned that state law creates and controls the attorney-client relationship, so the privilege growing out of it required federal courts to follow state law, and it applied the law of the forum state in a proceeding to enforce an Internal Revenue Service summons.45 Rule 501 now divides the field by the rule of decision, and it does not defer to state privilege law in federal-question cases or federal criminal prosecutions. In United States v. Schenheinz, an IRS summons required a stenographer to testify and produce records on the employer's tax matters even though a state employer-stenographer privilege covered the information.46 The reasoning that state law creates the lawyer-client relationship no longer decides the question in federal-law matters. The source of the rule of decision does.
Once state law governs, the federal court's first task is to find that state's rule, and much of it is codified. Most states with codes modeled on the Federal Rules of Evidence have codified their privilege law, and their provisions vary considerably.47 Hawaii Rule of Evidence 503(b), for example, gives a client “a privilege to refuse to disclose and to prevent any other person from disclosing confidential communications made for the purpose of facilitating the rendition of professional legal services to the client.”48 State courts construe such provisions in their own decisions; Fellerman v. Bradley, a New Jersey Supreme Court decision, is one example. A federal court in a diversity case begins from the state's statute and the state decisions applying it.
Federal and State Claims Together
Rule 501 assigns a privilege question to a body of law one claim or defense at a time, and that wording works smoothly when each claim has its own evidence. It creates a puzzle when a federal claim and a state-law claim are tried together and one lawyer-client communication is relevant to both. A federal court can hear the state claim under its supplemental jurisdiction if the events behind it are the same as, or closely related to, the events behind the federal claim. That overlap makes it likely that some witnesses will give testimony relevant to both claims.50 Rule 501's text then points to federal common law for one claim and to state law for the other, and it does not say which prevails when a single communication sits under both.
When one body of privilege law must apply to both claims, federal courts generally apply federal privilege law. A contrary rule would have the federal court defer too far to state law, because it would apply state privilege law to a federal claim.51
Pearson v. Miller applies the rule. The action joined federal and state claims, and the discovery dispute concerned material relevant to both, so the Third Circuit held that Rule 501 directs the application of federal privilege law, citing its 1982 decision in Wm. T. Thompson Co. v. General Nutrition Corp.52 The appellants conceded that the district court erred in applying state law directly. They argued instead that federal law should recognize an evidentiary privilege reflecting the concerns behind the Commonwealth's confidentiality statutes.53
Applying federal law has a practical consequence for a party who relies on a state statute. The statute no longer binds the court, so the party must persuade it to recognize the same protection as federal common law. In Pearson, the court found no decision of its own, and none from another circuit, recognizing privileges of the kind the appellants sought, so the appellants were effectively asking for quite novel privileges.54 A federal court may look to state privilege law for guidance without being bound by it.55 Pearson involved statutory confidentiality rules rather than the lawyer-client privilege, but the choice-of-law reasoning does not depend on which privilege is asserted. For a lawyer-client claim, the usual effect is that the federal formulation of the privilege decides the objection.
Two separate reasons support the result. The first explains why a single law must govern: the testimony bears on both claims, so one body of privilege law has to decide whether the fact-finder may hear it.56 The second explains why federal law is the law that governs. A state privilege applied to a federal claim would let the state say when a federal court must refrain from hearing useful testimony in a matter involving federal law.57 No available choice honors both sovereigns at once, and the federal court should not defer to state law on a claim that federal law created.
A diversity suit that pleads only state-law claims presents no such conflict. No federal claim shares the evidence, state law supplies the rule of decision for everything the jury will decide, and the second sentence of Rule 501 sends privilege to state law. The tie-breaker therefore applies only where the same communication bears on a federal claim and a state claim. When an exam fact pattern sets a state privilege against a federal claim, ask whether the communication is relevant to the federal claim. If it is, federal common law governs. If the suit is a state-law case that merely sits in federal court, the second sentence controls.58
Choosing Among States
The second sentence of Rule 501 sends a state-law claim to state privilege law without saying which state's law. The silence costs nothing when only one state is involved, as in the Virginia collision case. It matters when the forum state and the state whose substantive law governs the claim are different states with different privilege rules on the point in dispute. Such differences exist, because the scope of the attorney-client privilege varies among jurisdictions, and especially among the states.60
In KL Group v. Case, Kay & Lynch, the Ninth Circuit left the choice open but noted that commentators have suggested several methods for resolving it:61
- Assume that the state that supplies the rule of decision also supplies the privilege law.
- Apply the privilege rules of the state in which the federal court sits.
- Apply the conflict-of-laws doctrine of the state in which the federal court sits.
When one state plays every role, the three methods agree. When a federal court in one state hears a claim governed by another state's substantive law, they part. The first points to the claim's state, the second to the forum, and the third to whichever state the forum's conflict-of-laws doctrine selects. The third method also accommodates a forum whose own privilege law defers elsewhere, as Virginia's might defer to Delaware's.62 The second method treats privilege like an ordinary forum evidence rule. The Restatement generally assigns evidence rules to forum law,63 and some federal decisions have applied forum-state privilege law without stating a principle.64 That treatment sits uneasily with Rule 501's view that privilege is substance rather than procedure.65
The difficulty is old. Garner v. Wolfinbarger concluded that the choice of law “cannot be settled by reference to any simple talisman” and can be reached only after considering the state and federal interests that are inseparable from the factors bearing on the privilege's availability.66
KL Group avoided the choice. A letter to Lawyers Title was produced inadvertently during discovery, and the question was whether the production waived the privilege. The court did not decide whether Hawaii or California law applied, because the result was the same under both: in each state the client holds the privilege, and under either state's law Lawyers Title did not waive it. The court cited a California decision holding that a lawyer's production of files in response to a subpoena duces tecum did not waive the privilege because the production was not voluntary.67 A choice among laws matters only when the laws would produce different results. Where they agree, ranking the three methods changes nothing.
When the states' rules conflict, Rule 501 supplies no formula, and the sources describe three competing methods rather than a settled one. The outcome then depends on the method the court adopts, so a careful answer names the method and says why it applies instead of assuming one.