Concept
Recognizing New Privileges
Other privileges
Recognizing New Privileges
Rule 501 lists no privileges. It tells federal courts to apply the common law, “as interpreted by United States courts in the light of reason and experience,” unless the Constitution, a federal statute, or rules prescribed by the Supreme Court provide otherwise.1 That wording does two jobs. It authorizes courts to recognize a privilege the common law has not yet established, and it supplies the only test for doing so, because reason and experience is what a court consults when no text answers the question.
The authority is costly, as the chapter’s opening section explained. Every privilege removes relevant evidence from a trial in exchange for a benefit that is hard to measure. If courts read the open-ended language of Rule 501 as an invitation, every confidential relationship could claim protection, and the search for truth would give way across a widening range of settings. They have read it instead as a standard that a proposed privilege must meet before it joins federal common law.
The psychotherapist-patient privilege in Jaffee v. Redmond serves as the running example, with accountant-client and parent-child communications as foils. Many people regard those relationships as confidential, and federal courts have been asked to protect them. The subsections below state the balancing standard, show what counts as evidence that reason and experience favor protection, explain why a recognized privilege must be predictable, and identify the privileges federal courts have declined.
The Balancing Standard
Rule 501 authorizes federal courts to recognize privileges the common law has not established, but it names no privilege and states no test.2 “Reason and experience” is a phrase, so the working standard comes from the Supreme Court’s privilege decisions, which lower courts restate in nearly identical terms. A court recognizes a new privilege only when it “promotes sufficiently important interests to outweigh the need for probative evidence,” meaning evidence that tends to prove a fact the factfinder must decide.3 The sentence describes a scale, with the claimant’s interest on one side and the lost evidence on the other.
The scale does not start level. The public has a right to every person’s evidence, so the general duty to testify is the baseline and every exemption departs from it.4 The party seeking evidence has that baseline duty on its side, so a party who asks for a new privilege must justify the departure. The justification must be strong, because each privilege costs accuracy: the factfinder decides the case without evidence that may be reliable and highly probative. Courts therefore call privileges exceptions that are “not lightly created nor expansively construed,” since each is “in derogation of the search for truth.”5 Trammel v. United States gave the same caution as a rule of method: privileges are strictly construed and accepted only to the very limited extent that a public good justifies them.6
Congress did not intend to freeze the law of privilege. It meant to give courts flexibility to develop rules case by case and to leave the door open to change.7 The Supreme Court has nonetheless been “disinclined to exercise this authority expansively,” and it is especially reluctant to recognize a privilege where Congress has considered the competing concerns and has not provided the privilege itself.8 The reluctance has a reason. A court that creates a privilege Congress weighed and left out overrides a legislative balance instead of filling a gap.
On the privilege’s side of the scale, the claimant must show more than a wish for secrecy. The interest must amount to a “public good transcending the normally predominant principle of utilizing all rational means for ascertaining the truth,” it should be shown “with a high degree of clarity and certainty,” and courts give considerable weight to federal and state precedent in judging whether it has been shown.9 A private interest in confidentiality, however strongly felt, cannot meet that description alone. An asserted privilege must also serve public ends, so its benefit has to reach past the two people in the conversation to the community.10
Jaffee shows the weighing come out in favor of a new privilege. Redmond, a police officer, had shot a man, and afterward she attended about 50 counseling sessions with Karen Beyer, a clinical social worker licensed by Illinois. The plaintiff sought Beyer’s notes to cross-examine Redmond, and the defendants vigorously resisted.11 Because the notes were not produced, the trial judge told the jury it could presume their contents would have been unfavorable to the respondents, and the jury awarded $45,000 on the federal claim and $500,000 on the state-law claim.12 The Supreme Court held that confidential communications between a licensed psychotherapist and her patient in the course of diagnosis or treatment are protected from compelled disclosure under Rule 501.13
The Court’s reasoning follows the parts of the standard. The first part is the private interest: effective psychotherapy depends on an atmosphere of confidence and trust in which the patient is willing to make a frank and complete disclosure, and the mere possibility of disclosure may impede that willingness.14 The second is the public end. The privilege serves the public interest by facilitating treatment, and the mental health of the citizenry is a public good of transcendent importance.15 The third is the evidentiary cost, which the Court found modest. Without a privilege the conversations would be chilled, and the admissions litigants hope to obtain would probably never be spoken.16 In the Court’s words, that unspoken “evidence” serves no greater truth-seeking function than if it had been spoken and then privileged.17 The third step shows why a privilege sometimes costs less than it appears to. Where the promise of secrecy is what brings the communication into existence, withholding it takes nothing from the factfinder that it would otherwise have had.
The same weighing yields the opposite result when the proponent cannot make the showing. In University of Pennsylvania v. EEOC the Court declined to recognize a privilege for confidential academic peer review materials that the EEOC sought, and it described itself as disinclined to use its Rule 501 authority expansively.18 The two decisions differ in what each proponent could show. Redmond could point to a relationship whose usefulness depends on confidentiality, a public benefit from the treatment it produces, and a privilege that all fifty states and the District of Columbia had enacted in some form.19 The university’s request ran into the Court’s reluctance to recognize privileges where Congress has weighed the competing concerns and has not provided one.20
Because the standard is a balance, an exam answer on a new privilege is best built as a sequence of questions that puts both sides of the scale on the page before it reaches a conclusion.
Evidence of State Consensus
A court deciding whether reason and experience support a new privilege needs something concrete behind those words,22 and one concrete thing is what the states have done. In Jaffee the Court treated “a consistent body of policy determinations by state legislatures” as reflecting both reason and experience, and lower federal courts now count state enactments when they weigh a proposed privilege.23 The consensus fills the side of the scale that is hardest to fill. The benefit of a privilege is easy to assert and hard to measure, since no one can say how much more candidly clients, patients, or spouses speak because of it. A uniform judgment by legislatures that have watched a privilege operate in their own courts is the nearest thing to a measurement the law has.
The Court gave two reasons why state legislation carries this weight, and the reasons work differently. The first is informational. State legislatures are fully aware of the need to protect the integrity of fact-finding in their own courts, so when they independently conclude that a relationship deserves protection, the agreement indicates that the privilege is worth the evidence it costs.24 The second is practical. A patient speaks freely only if the patient believes the conversation will stay private, and a state’s promise of confidentiality would have little value if the patient knew a federal court would not honor the privilege. Refusing the federal privilege would therefore frustrate the purposes of the state legislation.25 The first reason explains why the consensus persuades. The second explains why a federal court has its own reason to care even where state law does not control, because a federal rule that cut across a uniform state policy would undo the confidence the states set out to create.
The record in Jaffee was as complete as such a record can be. All fifty states and the District of Columbia had enacted some form of psychotherapist privilege, and the Court said the uniformity confirmed what its analysis of the therapeutic relationship already suggested.26 A second body of evidence pointed the same way. The Advisory Committee’s proposed rules, the specific privilege rules that Congress replaced with the open-ended Rule 501, had included a psychotherapist-patient privilege, and the Senate Judiciary Committee said that the substitution “should not be understood as disapproving any recognition of a psychiatrist-patient ... privilege contained in the [proposed] rules.”27 The unenacted draft thus supplies consensus evidence of a different kind, the considered judgment of the drafters instead of the states, and a privilege that appears in it begins with support that a privilege absent from it lacks.28
A privilege for confidential communications between parents and children shows the same evidence running the other way. A federal court of appeals asked to create that privilege read Jaffee as an instruction to look to the experience of the states, and it counted only four states that protect such communications in any manner. New York’s courts recognized a limited privilege; Idaho and Minnesota enacted limited statutory privileges for a minor’s confidences to a parent; and Massachusetts disqualifies minor children from testifying against their parents in criminal proceedings.29 Four narrow determinations did not amount to a “consistent body of policy determinations by state[s].”30 The court went further. Because the overwhelming majority of states had chosen not to create the privilege, it reasoned, “reason and experience” dictate that federal courts “should refuse to recognize a privilege rejected by the vast majority of jurisdictions.”31 The claimants had nothing else to offer, because they had identified no common-law principle examined in light of reason and experience.32
Set the two records side by side. Fifty-one jurisdictions and an Advisory Committee draft stood behind the psychotherapist privilege, while four jurisdictions acting narrowly stood behind the parent-child privilege. With state policy pointing in opposite directions, the same balancing standard produced opposite results. Consensus is the variable that tips an otherwise close balance toward recognition or away from it.
Consensus is nonetheless one weight on the scale, and it does not settle the balance alone. Courts give considerable weight to federal and state precedent, but the proponent must still show a public good that transcends the ordinary demand for every person’s evidence, and must show it with a high degree of clarity and certainty.33 Jaffee itself described the state laws as confirming a conclusion already supported by the private interest in confidential treatment, the public interest in mental health, and the modest evidentiary cost of protection.34 Federal courts have also refused several privileges that states often recognize, including the accountant-client, physician-patient, and parent-child privileges, which shows that a state enactment does not create a federal privilege by its own force.35 The last subsection returns to those refusals.
An essay asking whether a federal court should recognize a new privilege invites a count, and the count has to be of the right thing.
The lone statute is not worthless, because federal courts give weight to state precedent. That weight is slight when no other state, no draft rule, and no federal decision stands beside it.36
Certainty Once Recognized
Rule 501 lets federal courts recognize privileges one decision at a time, by pointing them to the common law as interpreted in light of reason and experience.38 That method carries a risk the earlier subsections did not address. A privilege recognized in outline but open to adjustment in every later dispute would give its holder nothing to rely on in advance. Jaffee v. Redmond confronted the risk in two holdings. The Court extended the psychotherapist privilege to confidential communications with licensed social workers in the course of psychotherapy, and it rejected the balancing component the court of appeals had attached, under which a trial judge would later weigh the patient’s privacy interest against the litigant’s evidentiary need.39
The reason follows from how a privilege produces its benefit. The benefit is candor, and candor is decided at the moment of speaking, long before any judge sees the dispute. A patient choosing how much to tell a therapist can rely only on the rule as it stands that day. A rule that protects the conversation unless a court later finds the opponent’s need greater than the patient’s privacy gives the patient no answer to the one question that matters: is it safe to speak? The Court concluded that making confidentiality contingent on a trial judge’s later weighing of those interests “would eviscerate the effectiveness of the privilege.”40 It relied on Upjohn’s warning that participants in a confidential conversation “must be able to predict with some degree of certainty whether particular discussions will be protected,” and that an uncertain privilege, or one that “results in widely varying applications by the courts,” is “little better than no privilege at all.”41
Balancing therefore brings back the chilling effect the privilege exists to remove, and it does so most severely where the stakes are highest. The conversations a litigant needs most are the ones most likely to hold damaging admissions, so a rule that unlocks conversations in proportion to need would strip protection from the patients with the most reason to stay silent. Holding the privilege firm costs the factfinder little. Without a privilege, much of the evidence a litigant hopes to reach, such as a party’s admissions against interest, would never come into being, and unspoken evidence serves no greater truth-seeking function than evidence that was spoken and then privileged.42
The extension to social workers shows the same attention to how the privilege will be used. Social workers provide a significant amount of mental health treatment, and their clients often include the poor and those of modest means who could not afford a psychiatrist or psychologist but whose counseling serves the same public goals. The Court therefore agreed with the court of appeals that distinguishing costly psychotherapists from more accessible social workers “serves no discernible public purpose.”43 The Advisory Committee’s proposed privilege had defined psychotherapists as psychologists and medical doctors, but the Court explained that this 1972 limit did not counsel against including social workers, because only 12 states regulated social workers in 1972 and all 50 do today.44 The line the Court drew is also one a patient can apply before speaking. A patient can usually tell whether a counselor is licensed and is providing treatment, but could not tell which counselors a later court would consider enough like a psychiatrist to count. In Jaffee itself, the conversations between Redmond and Beyer, a licensed clinical social worker, and the notes of their sessions were protected from compelled disclosure under Rule 501, with no inquiry into how much the plaintiff needed those notes for cross-examination.45
The same principle explains Upjohn Co. v. United States, the source of the language Jaffee quoted. The court of appeals there had adopted a control-group test, which protected a corporate employee’s communication to counsel only if the employee was in a position to control or take a substantial part in a decision about action the corporation might take on counsel’s advice.46 The Supreme Court rejected that test because the privilege protects the giving of information to the lawyer, on which sound advice depends, as well as advice to those who can act on it.47 Predictability supplies a second reason against the test. Whether an employee falls within the control group depends on rank and influence, facts an employee answering counsel’s questions may be unable to assess, so neither the employee nor the lawyer can know at the time of the conversation whether it will be protected. A subject-matter approach asks instead whether the employee spoke so the corporation could obtain legal advice, at a superior’s direction, about matters within the employee’s duties, and whether the communication stayed within the circle of people who need to know it. Every one of those criteria is visible to the employee at the time of speaking.48
Upjohn also shows how little the certainty principle demands of courts at the outset. Chief Justice Burger, concurring, agreed that the control-group test had to go and that lawyer and client must be able to predict, with some degree of certainty, whether a discussion will be protected, and for that reason he wanted the Court to announce a general standard for corporate communications.49 The majority declined to draft such rules. It conceded that case-by-case development may “to some slight extent undermine desirable certainty in the boundaries of the attorney-client privilege” but held that the approach obeys the spirit of the Rules.50 Jaffee took the same position about its own holding. Because it was the first case recognizing a psychotherapist privilege, the Court said, it was neither necessary nor feasible to delineate the full contours of the privilege.51
Those passages mark a distinction that keeps the principle from being misread. Case-by-case development of where a privilege’s boundaries lie remains the method under Rule 501. What certainty rules out is a privilege whose protection in a given conversation is decided afterward by weighing the litigant’s need against the holder’s privacy, because the criteria that decide protection must be ones the holder can apply before speaking. The balance described in the previous subsection happens once, when a court decides whether reason and experience support the privilege at all, and it does not recur each time someone asks for the protected communications.
The Court did not claim that the privilege has no limits. In a footnote it said it did not doubt that there are situations in which the privilege must give way, for example if a serious threat of harm to the patient or to others can be averted only by a disclosure by the therapist, and it left the details of such situations to later cases.52 That example is defined by a danger that disclosure can avert, and it asks nothing about how badly a litigant wants the evidence. An exam answer can mention it as the Court’s acknowledged reservation without treating it as permission to balance need against privacy.
An exam problem on this point will often make the litigant’s need unusually strong and ask whether that changes the answer.
A trial judge who reasons that a litigant’s need for the notes outweighs the patient’s privacy interest has applied the balancing component the Supreme Court rejected. A judge who asks only whether the communications were confidential, made to a licensed psychotherapist, and made in the course of diagnosis or treatment has applied the rule the Court adopted. The corporate setting poses the same choice. A court that asks whether the employee’s rank placed the communication within the privilege has revived the control-group test, and a court that asks about the circumstances of the communication has followed Upjohn.
Privileges Federal Courts Decline
Rule 501 makes reason and experience the test for what federal common law protects, and that test decides which privileges a federal court adds to the short list it already honors.54 Applying it, federal courts have declined a group of privileges that many states provide. The group is a recurring source of exam traps, because the claimant can point to genuine state support and still lose. The pattern follows from the character of the privileges involved. Federal courts have accepted the most basic and traditional privileges as federal common law, but they have been far less receptive to more modern ones, which typically arose in the states by statute.55
A state legislature that creates a privilege decides, for its own courts, that a relationship is worth the evidence it costs. That judgment does not bind a federal court applying federal law, because Rule 501 sends the court to the common law and the proponent must show a public good large enough to outweigh the need for probative evidence. State enactments count as evidence of reason and experience, as the preceding subsection showed, but their weight depends on how many states agree. The federal interest supplies a second reason for caution. A federal court enforcing a grand jury subpoena or an IRS summons would lose access to evidence whenever a single state happened to protect the relationship.
Five examples recur, and in each the claimant could point to a valued relationship and some state support but could not make the federal showing.
- Communications to accountants are privileged in perhaps a third of the states, yet federal courts have refused to recognize an accountant-client privilege.56
- Federal courts have likewise not recognized a physician-patient privilege, although states that recognize one generally protect confidential communications with a physician, including information gathered in a physical examination, and treat the protection as waived when the patient puts a physical condition in issue in the litigation.57
- A court of appeals refused to create a parent-child privilege after counting only four states with any protection for such communications, each of them limited, and after finding that the claimants had identified no common-law principle examined in light of reason and experience.58
- The Supreme Court declined to recognize a federal privilege for the confidential peer review materials of academic institutions in University of Pennsylvania v. EEOC.59 The same discovery pressure arises in malpractice litigation, where plaintiffs frequently seek the proceedings of hospital quality assurance committees.60
- The Ninth Circuit declined to recognize an employer-stenographer privilege in United States v. Schoenheinz, where a stenographer had to testify and produce records about an employer’s tax matters in response to an IRS summons even though a state employer-stenographer privilege covered the information.61
The usual exam error is to treat state recognition as if it settled the federal question, and the accountant-client privilege shows why that instinct loses points.
The statute does not reach the subpoena. In a federal criminal investigation, Rule 501 sends the privilege question to federal common law, and the state-law exception covers only a civil case in which state law supplies the rule of decision.62 A privilege that federal courts do recognize produces the opposite result on otherwise similar facts. In Jaffee v. Redmond a police officer who had shot a man attended counseling with a licensed clinical social worker, and the plaintiff sought the session notes in a federal civil action.63 The notes were protected, because federal courts recognize a psychotherapist-patient privilege that reaches licensed clinical social workers.64 Both scenes involve a confidential conversation with a licensed professional, and the difference lies in what stands behind the claim. Every state had enacted some form of psychotherapist privilege, while accountant communications are protected in perhaps a third of the states and parent-child communications in four.65 The same comparison defeats the argument that a physician is a health professional like a therapist, since federal courts that recognize the one have not recognized the other.66 The privileges they do recognize include the attorney-client, clergy-penitent, and both marital privileges, along with governmental privileges such as the informer’s identity privilege.67
A declined privilege is not unavailable everywhere. Federal courts develop the common law of privilege except in diversity cases where state law supplies the rule of decision, and in those cases the state’s law of privilege controls.68 The accountant-client communication that a federal grand jury may compel can therefore stay protected in a diversity action governed by state law, if the state recognizes the privilege, because the civil proviso of Rule 501 hands the question to the state.69 The stenographer in Schoenheinz lost the state privilege for the opposite reason: the IRS summons was a federal matter. The first step in any such problem is to identify which law governs the claim, and only then to ask what that law provides.