Every large document review carries the same quiet risk: somewhere in the millions of emails, Slack threads, and contract drafts, a privileged document is going to slip through. The question isn't whether it will happen — it's whether producing it will cost you the privilege entirely. Federal Rule of Evidence 502 exists to answer that question, and 502(d) is the tool that makes the answer "no."
The Problem 502 Was Built to Solve
Before 2008, inadvertent production of a privileged document could trigger subject-matter waiver — not just for that document, but for an entire topic. Courts applied inconsistent multi-factor tests, and producing parties spent enormous sums on privilege review just to avoid a single slip. FRE 502 was enacted to reduce discovery costs and bring predictability to waiver analysis.
502(a) and 502(b): The Default Rules
- 502(a) limits subject-matter waiver to situations where the disclosure was intentional and the withheld material concerns the same subject matter, and fairness requires broader disclosure.
- 502(b) protects against waiver for inadvertent disclosures, but only if the holder took reasonable steps to prevent disclosure and promptly took reasonable steps to rectify the error, including under Federal Rule of Civil Procedure 26(b)(5)(B).
502(b) sounds protective, but it's a fact-intensive, litigated standard. Courts weigh the scope of review, the volume of documents, the review methodology, and how quickly counsel acted after discovery of the error. Litigating "reasonableness" after the fact is expensive and outcome-uncertain — exactly the problem the rule was meant to avoid.
502(d): Waiver Protection by Court Order
502(d) lets parties skip the reasonableness fight altogether. A federal court may order that privilege is not waived by disclosure in the pending litigation — and that order is enforceable against non-parties and in any other federal or state proceeding. Critically, a 502(d) order can protect even careless or negligent disclosures, not just inadvertent ones, depending on how the order is drafted.
A well-drafted 502(d) order converts privilege review from a source of existential risk into a manageable operational issue — but only if it is negotiated and entered before review begins, not after a mistake is discovered.
Why 502(d) Matters More With Modern ESI Volumes
Rule 26(f) already directs parties to discuss claims of privilege and procedures for asserting them after production, and Rule 26(b)(5)(B) governs clawback mechanics once a claim is made. But those provisions assume you're litigating waiver case by case. With Teams messages, Slack threads, shared drives, and TAR-assisted review pulling in exponentially more documents than a decade ago, the odds of a slip rise with volume — and so does the cost of manual privilege screening. A 502(d) order lets review teams use technology-assisted review, keyword culling, or reduced manual QC on privilege issues without betting the case on it.
Drafting a 502(d) Order That Actually Protects You
Not all clawback orders are equal, and the difference is rarely the caption at the top of the page — it is the operative language two paragraphs down. Two decisions a year apart show how much turns on the wording.
The Narrow Order: Reciting 502(d) Is Not Enough
In Absolute Activist Value Master Fund Ltd. v. Devine, 262 F. Supp. 3d 1312 (M.D. Fla. 2017), the parties had a stipulated protective order whose operative paragraph opened with the words "Pursuant to Federal Rule of Evidence 502(d)" and provided that a disclosure made without the express intent to waive would not be deemed a waiver. On its face, that is exactly the protection a producing party wants.
The problem was the next sentence. It set out a notice procedure and provided that, once notice of an inadvertent disclosure was given, "Federal Rule of Evidence 502 and Federal Rule of Civil Procedure 26(b)(5)(B) must apply." The court held that because the order incorporated Rule 502 in its entirety, with no limitation or exclusion, and because only one subsection of Rule 502 addresses inadvertent disclosure, subsection (b) supplied the governing standard. The producing party was back where it started — obliged to show it had taken reasonable steps to prevent the disclosure and to rectify it. An order captioned under 502(d) delivered a 502(b) fight.
The Broad Order: Displacing 502(b) on Purpose
Contrast In re Testosterone Replacement Therapy Products Liability Litigation, 301 F. Supp. 3d 917 (N.D. Ill. 2018). There the protective order stated that production, "whether inadvertent or intentional, is not a waiver of any privilege or protection," and that the provision "is, and shall be construed as, an Order under Rule 502(d) of the Federal Rules of Evidence." The court held that the protective order governed inadvertent disclosures and Rule 502(b) did not — and, having so held, declined to reach the reasonableness arguments at all.
The lesson is not that one court was right and the other wrong. Both applied the order in front of them. The lesson is that a 502(d) order has to say affirmatively that 502(b) does not apply. Silence is not read in your favor, and neither is a recital of the subsection number.
The "Irrespective of the Care Taken" Clause
The single most valuable sentence you can put in a 502(d) order is not your drafting at all — it comes from the rule's own Advisory Committee Note, which states that the court order contemplated by the rule "may provide for return of documents without waiver irrespective of the care taken by the disclosing party." That is the drafters saying, in terms, that a 502(d) order may protect a careless disclosure. The court in the Testosterone Replacement Therapy litigation quoted exactly that language in rejecting the argument that a producing party must first prove reasonable efforts before invoking a clawback provision.
Put the clause in the order, in those words. It tells the court, in the rule-drafters' own vocabulary, that what you are asking for is what the rule contemplates — which makes entry routine rather than a negotiation. And it forecloses the argument that your review methodology, your reviewer-to-document ratio, and your quality-control sampling are fair game in a waiver fight. Without it, every one of those operational decisions is in play the moment a privileged document slips.
502(d) Is an Order. 502(e) Is Only an Agreement.
Parties routinely negotiate a clawback provision, sign it as part of an ESI protocol, and never present it to the court. Rule 502(e) is explicit about what that buys: an agreement on the effect of disclosure in a federal proceeding is binding only on the parties to the agreement, unless it is incorporated into a court order.
Read that limitation as an adversary would. A co-defendant who did not sign is not bound. A plaintiff in a later, related case is not bound. A state-court adversary is not bound. A regulator is not bound. Every one of them is free to argue that your production waived the privilege, and your carefully negotiated agreement is, as to them, a piece of paper between two other people.
A 502(d) order is different in kind, not degree. The rule provides that when a federal court orders that privilege is not waived by a disclosure connected with the litigation pending before it, the disclosure is also not a waiver in any other federal or state proceeding. That reach — past the signatories, past the case, and into state court — is the entire reason the subsection exists, and it is why "we have a clawback agreement" and "we have a 502(d) order" are not two ways of saying the same thing.
The practical consequence is simple: get it entered. A non-waiver clause exchanged between counsel but never docketed as an order is a 502(e) agreement wearing a 502(d) label.
How a 502(d) Order Changes the Rule 26(b)(5)(B) Mechanics
Rule 26(b)(5)(B) is the procedural machinery of clawback. When a producing party notifies the receiving party of a claim of privilege over already-produced material, the receiving party must promptly return, sequester, or destroy the material and any copies; must not use or disclose it until the claim is resolved; must take reasonable steps to retrieve it if already disclosed to others; and may promptly present it to the court under seal for a determination of the claim.
What Rule 26(b)(5)(B) does not do is decide whether the privilege survived the production. It is a sequencing rule — it says what happens in what order, and it preserves the status quo while the dispute is briefed. The substantive question of waiver is answered by Rule 502. That seam is what a well-drafted 502(d) order exploits: keep the sequencing, remove the fight.
A 502(d) order should supersede the Rule 26(b)(5)(B) default in three ways, each worth negotiating explicitly rather than inheriting:
- It fixes the clock. "Promptly" is a litigable word. "Within five business days of written notice" is not, and the producing party benefits most from a number, because the alternative is arguing about its own diligence.
- It narrows what the receiving party may argue. Under the bare rule, a receiving party presenting material under seal argues both that the document was never privileged and that any privilege was waived by the manner of production. A 502(d) order that displaces 502(b) removes the second argument, leaving only the substantive privilege question — the one that should be litigated.
- It decouples the demand from the log. Some orders condition a clawback notice on service of a compliant privilege log. That is a trap: it converts an urgent notice into a document-drafting exercise and hands the receiving party an argument that the notice was defective. Provide for notice first, with the log to follow within a stated period.
Model 502(d) Language
The following is a starting point, not a form to file unread — local practice, a judge's standing orders, and the shape of the case all matter. Present it as a standalone order, or as a clearly delineated section of a protective order the court actually signs.
- The production or disclosure of any document, electronically stored information, or other material in this action, whether inadvertent or otherwise, does not constitute a waiver of the attorney-client privilege, the work-product doctrine, or any other applicable privilege, protection, or immunity, either as to the material produced or as to the same or related subject matter, in this action or in any other federal or state proceeding.
- This Order is entered under Federal Rule of Evidence 502(d). The protections it affords apply irrespective of the care taken by the disclosing party. Federal Rule of Evidence 502(b) does not apply to any disclosure governed by this Order, and no party bears any burden of demonstrating that it took reasonable steps to prevent, or to rectify, any such disclosure.
- A producing party may assert a claim of privilege, protection, or immunity over produced material at any time. Within five business days of written notice of such a claim, the receiving party must return, sequester, or destroy the specified material and all copies, summaries, and notes reflecting its contents; must not use or disclose it for any purpose; and must take reasonable steps to retrieve any copies it disclosed before receiving the notice.
- Written notice under Paragraph 3 need not be accompanied by a privilege log; the producing party must serve a log conforming to Federal Rule of Civil Procedure 26(b)(5)(A) within fourteen days of the notice. The receiving party may present the claim to the Court under Federal Rule of Civil Procedure 26(b)(5)(B), but may not rely on the fact of production, the circumstances of production, or the care taken by the producing party as a ground for contesting it.
- This Order applies to all disclosures made in this action, including any made before its entry, and to disclosures by or to any non-party who agrees in writing to be bound by it. A party's decision to forgo or limit a privilege review before production is not a basis for finding waiver.
Two drafting notes. Keep the 502(d) provision structurally distinct from the confidentiality provisions it usually travels with, so a later dispute about designations cannot be argued to have swallowed the non-waiver protection. And resist the urge to sprinkle "inadvertent" through the order: every use of the word invites litigation over whether a particular production was inadvertent, which is a reasonableness inquiry by another name — the exact fight the order exists to prevent.
What a 502(d) Order Does Not Protect Against
A 502(d) order is a precise instrument, and selling it internally as a cure-all sets up an unpleasant conversation later. Everything outside its scope remains a live problem:
- It does not make a document privileged. The order addresses waiver by disclosure, not the existence of the privilege. If the document was never privileged — a business communication copied to counsel, a compliance report prepared in the ordinary course — clawing it back fails on the merits.
- It does not excuse affirmative use. Rule 502(a) still governs intentional disclosure, and a party that puts privileged material into issue can still face subject-matter waiver on fairness grounds. The order protects a slip; it does not license privilege as both sword and shield.
- It does not un-ring the bell. Opposing counsel who read the document before returning it cannot unread it. Clawback restores the legal position, not the tactical one, and disqualification is available only in extreme cases.
- It does not reach foreign tribunals or arbitration. The rule's text runs to "any other federal or state proceeding." A foreign court, a foreign regulator, and a private arbitral tribunal are not bound by force of the rule.
- It is not a confidentiality order. Trade secrets, commercially sensitive terms, and source code are the province of a Rule 26(c) protective order. The two are usually filed together, which is exactly why teams assume one does the other's job.
- It does nothing about privacy exposure. Over-producing personal data because privilege review was relaxed creates obligations under GDPR, the CCPA, and their analogues that no evidentiary order cures.
- It does not touch non-privilege objections. A 502(d) order lowers the cost of privilege review; it does not lower the cost of reviewing the wrong population.
Read your order and ask one question: if we produced a privileged document tomorrow because the review was rushed, does anything in this document require us to prove the review was reasonable? If the answer is yes — or "probably not, but I would have to argue it" — the order is not doing its job.
Get the FRE 502(d) order on the record early
The 502(d) order belongs on the Rule 26(f) agenda, not in the motion practice that follows a mistake. Judges expect the topic there, and some maintain standing or model orders precisely because entry is uncontroversial. Asking early is also strategically cheap: before anyone has produced anything, neither side knows which of them will need the protection, so the negotiation is genuinely mutual. After a privileged document surfaces, the same request looks like an attempt to fix a specific problem, and the other side has every incentive to resist it.
Arriving with proposed language rather than an agenda item is what makes this work. A standing 502(d) provision maintained as part of your discovery playbook means counsel never drafts one under time pressure. If your clawback language is inherited from a form protective order nobody has read closely in years, it is worth an hour of scrutiny before your next large production. Check three things: whether it is an order or only an agreement, whether it disclaims 502(b) explicitly, and whether it says "irrespective of the care taken." Those three lines are most of the protection.