Concept
Inadvertent Disclosure
Attorney-client privilege
Inadvertent Disclosure
Federal Rule of Evidence 502(b) answers what happens to a privilege when a party hands over protected material by mistake. A disclosure made in a federal proceeding or to a federal office or agency is not a waiver, in a federal or a state proceeding, if three conditions are met.1 First, the disclosure was inadvertent. Second, the holder of the privilege or protection took reasonable steps to prevent it. Third, the holder promptly took reasonable steps to rectify the error, including, where it applies, following Civil Rule 26(b)(5)(B).2 The same test covers the attorney-client privilege and work-product protection.3 The conditions are cumulative, so a holder who satisfies two and fails the third gets no protection from the rule.
The rule exists because screening a modern production is expensive and still imperfect. Electronic discovery multiplies the material a party must examine, and most of the rising cost of e-discovery is attributed to making sure that no email, spreadsheet, or metadata contains privileged information.4 Some common law courts found waiver 'no matter how careful the privilege review,' which put the careful party and the careless party in the same position.5
A lawyer's lapse can cost the client the privilege, which surprises students who have just learned that an intentional but unauthorized disclosure by the lawyer does not. Negligent disclosure sometimes does waive. The lawyer holds the client's authority to respond to discovery requests, so the lawyer's negligence is in effect the client's, and the client answers for it.6
Before 2008, no rule settled the question, and the case law split three ways. Some courts held that the privilege is automatically lost. Some held that it persists, either because the disclosure was not intentional or because the disclosing lawyer lacked the client's authority to waive. Some made the answer depend on the diligence of the party claiming privilege in screening beforehand, and on the diligence and speed with which it caught the accident and notified the other side.7 Each pole has a visible cost. Automatic loss forces the document-by-document screening that made electronic discovery prohibitive.8 Automatic protection would give a producing party no reason to screen at all. Congress codified the middle position in 2008: a party cannot be careless, but it need not undertake herculean efforts to prevent every mistake.9
The rule's reach follows from the reason Congress had to act. Private agreements between litigants to ignore mistaken disclosures bind only the parties and generally cannot bind outsiders who later seek the material.10 Rule 502(b) therefore speaks to waiver itself, and it provides that a qualifying disclosure is no waiver in a federal or a state proceeding.11 Its waiver provisions also apply in a diversity action even if state law is to the contrary, which displaces the deference to state privilege law that Rule 501 would otherwise require.12 Even when a mistaken disclosure in discovery does waive the privilege, the worst consequence is loss as to the documents actually disclosed, and never a subject-matter waiver.13 The three conditions take up the rest of this section in order, beginning with inadvertence.
Disclosure Was Inadvertent
The first condition is that the disclosure was inadvertent.14 In this setting inadvertent means unintended: the holder never meant to give the material to the other side. The word sorts every disclosure into one of two groups before the other conditions come into play. A disclosure the holder meant to make is a choice, and the voluntary-disclosure principles covered earlier govern it. A disclosure the holder did not mean to make is a mistake, and only a mistake can earn protection under this rule. Reasonable precautions and prompt rectification matter only for a disclosure that clears this first gate.
Two kinds of mistakes account for most inadvertent disclosures. In the first, a privileged document is swept into a larger set of non-privileged material sent to a third party in the course of negotiation or litigation. In the second, a privileged communication is misaddressed and reaches someone outside the small circle of people with whom privileged information may be shared.15 Modern communications technology raises the risk of erroneous transmission to unintended recipients, and electronic discovery raises the odds that a mistake will slip into a production.16 The dispute in Sheet Metal Workers' National Pension Fund v. Palladium Equity Partners, LLC shows the setting. The defendants argued that Rule 502 was meant for exactly this case, in which vast numbers of documents are exchanged and privileged material is easily overlooked. They acknowledged that 116 recalled documents had been returned, but said the plaintiffs still held at least 18 more.17
Inadvertence turns on what the holder intended at the moment of disclosure. How carefully the holder worked belongs to the second condition, so a clerical slip that a better screening system would have caught is still an unintended disclosure and still satisfies this one. Keeping the two questions apart stops a court from rejecting a claim at the threshold merely because the mistake looks avoidable in hindsight. The usual error runs in two directions: treating a sloppy review as proof that the disclosure was intentional, and treating the holder's later regret as proof that it was a mistake.
The rule protects mistakes and gives no help to strategy, for reasons that follow from the two sides of the line. No party can eliminate every mistake from a modern production at a sensible price, and Rule 502 was adopted in response to complaints that the cost of guarding against waiver had become prohibitive, because even an innocent or minimal disclosure threatened waiver of far more.18 A client who chose nothing should not forfeit a confidence over an isolated slip. A deliberate disclosure is different, because the holder weighed secrecy against the advantage of disclosing and picked the advantage. If a party could take back a deliberate disclosure once it proved unwise, the party could use favorable privileged material as a sword and retreat behind the privilege when the material became inconvenient, the same unfairness that drives implied waiver. A deliberate disclosure also falls outside Rule 502(b) and can cost the holder more than one document, because Rule 502(a) applies a subject-matter standard to partial disclosures, as the discussion of subject-matter waiver below explains.19
In both scenes a privileged email from outside counsel ended up in the opposing party's hands, and the recipient's knowledge of its contents is equally complete. The one fact that separates them is whether anyone meant to send it. Exam stems often tuck that fact into a phrase such as 'a clerical error' on one side or 'to strengthen its position' on the other, and the first step in the analysis is to find it.
Reasonable Precautions
The second condition asks whether the holder of the privilege or protection took reasonable steps to prevent disclosure.22 The first condition looked at what the holder intended as the material went out, and the third will look at what the holder did once the error came to light. This one looks at the time before disclosure and asks what the holder had done to keep privileged material from leaving. A party can satisfy the first condition and fail this one, because an unintended disclosure that a sensible screening process would have caught is the mistake the rule expects the holder to prevent.
The standard is essentially negligence.23 A party cannot be careless, but it need not undertake herculean efforts to protect against mistaken disclosures in discovery.24 The line falls there because of what review costs. Palladium quoted the explanatory note's account of why the rule was adopted: litigation costs to protect against waiver had become prohibitive, especially in electronic discovery, where record-by-record review would impose costs out of proportion to what is at stake.25 A perfection standard would bring that burden back, because the only way to guarantee that nothing privileged escapes is to examine every record by hand. The opposite rule would reward indifference. Reasonable care sits between the two: it demands a working effort and forgives the occasional error that a real effort lets through. The ethics rules draw the same line. Model Rule 1.6(c) requires reasonable efforts to prevent inadvertent disclosure of client information, and prudent use of ordinary communications technology does not call for extraordinary precautions beyond the usual office procedures that limit access.26
Congress wrote no test into the rule. In Rhoads Industries v. Building Materials Corp., the court quoted the Advisory Committee's list of factors: the reasonableness of the precautions taken, the time taken to rectify the error, the scope of discovery, the extent of disclosure, and the overriding issue of fairness. None is dispositive, and the Committee left them out of the text because they are non-determinative guidelines that vary from case to case.27 The court also pointed to Judge Joyner's opinion in Fidelity & Deposit Co. of Md. v. McCulloch as a widely cited standard, and the first factor it identifies is the reasonableness of the precautions in view of the extent of the document production.28 The Committee Notes add that the factors' importance varies from case to case, and judges also differ on what counts as extreme carelessness and how to weigh it against the difficulty of undoing a disclosure from an extensive body of material or the sensitivity of what was disclosed.29 The upshot is that Rule 502(b) requires inquiry and balancing case by case.30 Only the first factor speaks directly to this condition. The time taken to rectify belongs to the third condition, and fairness sits over the whole analysis.
Courts give the first factor concrete content by comparing the number of mistaken disclosures with the volume reviewed. A party that let only a small proportion of the reviewed documents escape is more likely to be found reasonable than one that disclosed a large number of privileged documents out of a relatively limited set.31 The Committee also points to the number of documents to be reviewed and the time constraints for production.32 Each measure asks what a careful party could have done. No review reaches zero errors, so the useful question is how many escaped relative to how many were examined. The review a party can perform also depends on the time it had, and the rule was adopted for the setting in which vast numbers of documents are exchanged and privileged material is easily overlooked.33
An exam stem will usually supply the numbers: how large the production was, what review preceded it, how much time the party had, and how many privileged documents got out. The question is whether the steps taken were reasonable for that production.
The number of leaked documents did not decide either scene. Two stray emails from millions of screened records can reflect reasonable care, while dozens of advice emails from a few hundred unreviewed pages reflect none, because the precautions are measured against the extent of the production.34 Two errors recur on exams. The first treats any leak as proof of carelessness, which would make the condition impossible to satisfy once a single document escaped. The second treats a large production as a license to skip review, when volume changes which steps are reasonable and does not excuse a party from taking any. A party that clears this condition must still show that it acted promptly to rectify the error.
Prompt Rectification
The third condition looks at what the holder did once the mistake came to light: the holder must have promptly taken reasonable steps to rectify the error, including, where it applies, following Federal Rule of Civil Procedure 26(b)(5)(B).37 The three conditions examine three different moments. Inadvertence concerns what the holder intended when the material left, reasonable precautions concern what the holder did before it left, and this condition concerns the period after the holder discovered that privileged material had reached the other side.38 Because the conditions are cumulative, a holder whose disclosure was a true accident and whose screening was careful still forfeits the rule's protection by leaving the error uncorrected.
The text makes two demands, and each answers a different problem. Promptness answers the problem that harm grows with time. A recipient who has read a privileged document already knows what it says, so what remains in question is whether the recipient, and others who learn of the document, may use it or disclose it further, and every day of silence moves the material deeper into the other side's preparation.39 Reasonable steps answers a separate problem, which is that private regret changes nothing. Only an act that puts the receiving party on notice sets the protective machinery moving, and the form that machinery usually takes is the one the rule names.
Civil Rule 26(b)(5)(B) supplies the standard mechanism for civil discovery. The party that made the mistaken disclosure informs the receiving party that specified information was inadvertently disclosed privileged material, and that notice allows it to impose obligations on the receiving party to preclude further disclosure.40 After notification, the receiving party must promptly return, sequester, or destroy the specified information and any copies, and may not use or disclose it until the claim is resolved by agreement or by the court.41 The receiving party's duty arises after notification, which is why the burden of starting the process falls on the holder: a receiving party who has been told nothing has been given nothing to return. In practice the notice is a letter that identifies the document, says its production was a mistake, asserts the privilege, invokes the Civil Rule, and asks that the material be returned or sequestered and left unused.
Palladium shows how contested the aftermath of a recall can become. The defendants acknowledged that 116 recalled documents had been returned but said the plaintiffs still held at least 18, and they contended that the plaintiffs acted unethically by sitting on information they knew to be privileged.42 The receiving party's compliance and the holder's promptness are separate inquiries. The first asks whether the recipient obeyed after being told, while the third condition of Rule 502(b) asks whether the holder acted promptly and reasonably to rectify the error.
The receiving lawyer also carries an ethical duty. Since 2002, ABA Model Rule 4.4(b) has required a lawyer who receives an inadvertent disclosure of confidential information to notify the sending lawyer.43 A holder therefore cannot rest on the hope that the other side will speak up. The receiving lawyer's notice may be the way the holder learns of the mistake, and learning of it starts the holder's own clock, but that duty falls on the receiving lawyer and does not replace the steps Rule 502(b) requires of the holder.
How long is too long? The rule sets no number of days. The Advisory Committee's list of factors includes the time taken to rectify the error alongside the precautions taken, the scope of discovery, the extent of disclosure, and overriding fairness, and the factors are non-dispositive guidelines that vary from case to case.44 Speed and adequacy are therefore measured together. A letter sent the day after discovery that never asks for the material back accomplishes little, and a thorough demand sent months late arrives after the damage is done. The test looks to the holder's steps once it discovered the mistake, so the clock does not run before the holder knew.45 The words 'if applicable' show that the Civil Rule is the standard route and not the only one, since that provision governs discovery in civil litigation.46 Where it does not apply, the holder must still take whatever steps are reasonable to get the material back.
Both counsel found the same mistake, so conduct after discovery separated the outcomes. The other side's use of the memo during the months of silence does not itself breach the Civil Rule, whose return-and-sequester duties begin only with notification, so the cost of the silence falls on the holder. The usual error is to stop the analysis at inadvertence, or to treat a late letter as the equal of a prompt one. Inadvertence describes the moment of production and does not carry forward: a holder who knows of the mistake and does nothing has made a choice, and the rule treats it as one.