Initial Disclosures

Initial Disclosures

Unless the proceeding is exempt or the parties or the court have provided otherwise, a party must give the other parties four kinds of information without awaiting a discovery request.¹ The first is the name of each individual likely to have discoverable information that the party may use to support its claims or defenses, with the address and telephone number if known and the subjects of the person's information, unless the use would be solely for impeachment.² The second is a copy, or a description by category and location, of the documents, electronically stored information, and tangible things in the party's possession, custody, or control that it may use to support its claims or defenses, again excluding use solely for impeachment.³ The third is a computation of each category of damages the party claims, together with the evidentiary material behind each computation.⁴ The fourth is any insurance agreement under which an insurance business may be liable to satisfy part or all of a possible judgment.⁵

The first two items share a limit that the last two lack. Witnesses and documents fall within the duty only when the party may use them, which ties the duty to the party's own plans for the case; damages and insurance follow triggers of their own, a claim for money and an insurer's possible liability for a judgment. Taken together, the items give an opponent the people who may testify, the records that may be offered, the size and basis of the money claim, and the coverage that may stand behind a judgment. Automatic disclosure aims at a more efficient and expeditious discovery process, and the drafters hoped that exchanging this core information early would also encourage settlement and make discovery less adversarial and less costly.⁶

The may-use standard comes first because it sets the boundary of the witness and document duties. The four categories follow, and the last subsection takes up the proceedings that owe no initial disclosure at all.

The May-Use Standard

The words may use decide what the witness and document items reach. A party must identify the individuals likely to have discoverable information, and must copy or describe the documents, electronically stored information, and tangible things in its possession, custody, or control, that it “may use to support its claims or defenses,” unless the use would be solely for impeachment.⁷

The test looks to the disclosing party's own plans. For each person or item, the question is whether the party may use it to support a claim or defense that the party itself is asserting, so information that would help only the opponent never enters the calculation, however relevant it may be. May looks forward and does not demand certainty: the party asks whether it might rely on the witness or document, not whether it has already decided to. The word use is read broadly to include pretrial conferences, discovery, and motions, not only trial, and parties are encouraged to think carefully at the outset about what they will use to support their claims or defenses.⁸ A former employee whose sworn declaration will support a summary-judgment motion is therefore a person the party may use, even if that employee never takes the stand.

The standard is narrower than the one that governs discovery on request, and the narrowing is deliberate. A manufacturer sued over a defective design may have hundreds, if not thousands, of employees, agents, and individual contractors with information relevant to the plaintiff's allegations, yet at the outset it must disclose only the witnesses and documents it intends to use to support its defenses, a far smaller universe.⁹ The earlier version of the rule drew vigorous and enduring criticism because it could require a party to volunteer harmful material without a discovery request. The 2000 amendments answered by limiting the duty to witnesses and documents the party may use and by removing the link to particularity in the pleadings.¹⁰ Adverse information does not vanish from the case; the opponent must ask for it.

The impeachment carve-out removes one further class of material. Impeachment means attacking a witness's credibility, usually an adverse witness's. Material used solely for that purpose supports no claim or defense of the party's own, and a party need not disclose information it plans to use solely to impeach adverse witnesses.¹¹ The word solely carries the exception: a person whose only value lies in undermining someone else's credibility need not be listed, but a person who will also testify to facts that support the party's side is someone the party may use for more.

Discovery on request runs on a different measure. A party who asks may obtain any nonprivileged matter that is relevant to any party's claim or defense and proportional to the needs of the case, and that includes the names and contact information of persons who know of discoverable matter.¹² Initial disclosure does not require disclosure of all evidence relevant to a claim or defense; its scope is narrower, confined to information the disclosing party may use.¹³ Required disclosure therefore asks what the party intends to rely on, while requested discovery asks what bears on any party's claim or defense and is proportional to the case.¹⁴ Adverse information marks the gap: a witness whose account hurts the disclosing party is relevant, and the opponent can reach that witness by asking, but the witness falls outside the may-use duty because the party will not use the testimony to support its side. The common error is to reason that a witness with relevant information must appear on the initial disclosures, or that leaving such a witness off the list breaches the rule.

Bar questions test the point by having one lawyer discover a witness the other side left off its list and asking whether the omission was a violation.

Quinn must be listed because Bashad may use Quinn's testimony to support his defense; Patrick need not be, because Bashad will not use Patrick.¹⁵ Bashad therefore did not violate the rule by omitting Patrick, and DiMento has no ground for sanctions. Without a discovery request, the duty to supplement reaches only disclosures the party was required to make, so Bashad has no later duty to add Patrick.¹⁶ DiMento's lawyer should turn to practical steps: seek an informal interview or a deposition of Patrick, and serve an interrogatory asking for the identity of everyone who knows of discoverable matter, which reaches Patrick whether or not Bashad would use him.¹⁷ The work-product doctrine, which protects a lawyer's trial-preparation materials, does not shield Patrick's name and contact information.¹⁸ Nor does it matter that DiMento found Patrick before trial; Bashad is protected because he never broke the rule.

Witnesses With Information

Rule 26(a)(1)(A)(i) asks for three things about each person a party may rely on: the person's name, the person's address and telephone number if known, and the subjects of the information the person is likely to have.²⁰ The may-use test decides who goes on the list, with the same exclusion for anyone whose only use would be impeachment, and this item decides what the party must say about each person who does.

The three components are not equally flexible. The qualifier if known excuses only the address and telephone number, and for good reason: a party may truly lack contact details for a witness who left its employ or moved away. The name is not excused, because a person the party may use is a person the party can identify. The subjects are not excused either, because the party chose the witness for what the witness knows, and the subjects tell the opponent what that is. The common mistake is to read if known as excusing the whole entry.

The subjects requirement turns a bare roster into a map. If the opponent learns that one listed witness knows how a braking system was designed and another knows a vehicle's maintenance history, the opponent can decide whom to interview or depose first and what to ask. Name and subjects together replace the opening interrogatory, the one asking who knows what, that would invariably have been served.²¹ The rule calls for the subjects of the information and not for a statement of what the witness will say, so the disclosure points the opponent toward the witness without scripting the testimony.

The defective-design suit against Toyota shows how identical wording yields lists of different length for different sides. Hundreds, if not thousands, of people may have information relevant to the design and manufacture of a car, yet Toyota lists only those it may use to support its defenses.²² One of them is Harvey Gannick, a retired automotive engineer who was on the team that designed the original Camry and who appears among the witnesses Toyota identified in its initial disclosures.²³ A retired engineer is also the kind of witness whose current address a party may not hold, which is where the if known qualifier does its work. The Toyota employees it will not use stay off its list, and the plaintiff, Peter Painter, must reach them by interrogatory or deposition. Painter's own list, built on the same wording, names the witnesses he would use to support his claims, so the two lists differ because the two parties' plans differ.²⁴

Documents and ESI

Rule 26(a)(1)(A)(ii) applies the may-use test to things instead of people. A party must provide a copy, or a description by category and location, of all documents, electronically stored information, and tangible things in its possession, custody, or control that it may use to support its claims or defenses, unless the use would be solely for impeachment.²⁶ The party measures what to disclose against its own plan for the case, the documents and things it plans to use.²⁷ Three kinds of material fall within the item: documents in the ordinary sense; electronically stored information, meaning data kept in electronic form, such as emails and files on a server; and tangible things, meaning physical objects, such as a component taken from the product in dispute. The impeachment exclusion carries over, so a memo the party would use only to contradict an adverse witness stays off the list, as an impeachment-only witness does.

Students often read this item as a duty to deliver documents, but the text lets the party supply copies or describe the material by category and location.²⁸ The option allows a party with voluminous records to identify them without reproducing them at the outset. A description tells the opponent what kinds of material the party expects to use and where that material is kept, and an opponent who then wants copies or the originals can aim a precise request.

The description in that example contains both halves the rule asks for. The category tells the opponent what the material is, and the location tells the opponent where it can be found; a description naming only a category would leave the opponent unsure where to aim a request, and one naming only a location would leave it unsure what is stored there.

The phrase possession, custody, or control marks which material the duty reaches, and it covers more than what sits in the party's own offices.²⁹ The same phrase governs requests for production, where a party cannot avoid its duty by transferring information to an agent, including the party's own lawyer.³⁰ The duty would be easy to evade if physical location controlled, because a party could place records with an agent and say it no longer has them. Measuring the duty by possession, custody, or control closes that route, and attorneys are now expected to learn where and how a client's records, including electronic records, are maintained before disclosures are made.³¹

A non-party's own files are reached by a different device. In the defective-design suit against Toyota, Painter's lawyer wants whatever Camry design documents the retired engineer Harvey Gannick still has, and obtains them by serving Gannick with a subpoena duces tecum that commands him to appear for a deposition and to bring the described documents.³² A disclosing party's duty concerns material it holds or can reach; an opponent who wants a stranger's records goes to the stranger.

The may-use measure still leaves questions of scope. Painter was injured in a Camry, but automakers often build several models on shared platforms or similar mechanical systems, so the parties must ask whether Toyota's disclosure of documents it may use in its defense reaches the Corolla, the Prius, or the Venza as well.³³ The answer turns on Toyota's own plan: if its defense will rely on material about a system the models share, those documents are things it may use, and if the defense stays within the Camry's own design, they are not. Because the answer is not obvious in advance, the rules require the parties to meet and confer, and the Rule 26(f) conference directs them to make or arrange for the initial disclosures and to discuss the preservation of discoverable information.³⁴

Damages Computation

Rule 26(a)(1)(A)(iii) imposes two duties on a party that claims damages. The first is a computation of each category of damages claimed. The second is access to the records behind it: the party must make available for inspection and copying, as under Rule 34, the documents or other evidentiary material on which each computation is based, including materials bearing on the nature and extent of the injuries suffered, unless privileged or protected from disclosure.³⁶

The damages item has a different trigger from the witness and document items. Those two are limited by the may-use phrase, which the 2000 amendments introduced in place of the earlier link to particularity in the pleadings.³⁷ The damages item has no such phrase. Its trigger is the party's own claim for money, so the duty belongs to whoever claims damages, and a party that claims none has nothing to compute. Students err in both directions, treating the computation as owed by every party or treating a bare dollar total as enough.

The duty serves a practical need. An opponent cannot value a case, plan a defense, or answer a settlement overture without knowing what is claimed and how the number was reached. The claimant knows its own losses and holds most of the records, so it is best placed to supply the computation. Making the supporting records available lets the opponent test the arithmetic at the outset. The carve-out for privileged or protected material keeps the duty from stripping ordinary protections, so a document does not become available merely because the claimant relied on it. The form of the supporting-material duty also differs from the document item taught above: there the party may choose between a copy and a description by category and location, while here the material must be made available for inspection and copying, as it would be in response to a Rule 34 request.³⁸

A disclosure that states a single total of $180,000 for all damages and promises detail at trial fails on the same facts. The rule calls for a computation of each category, and a lump sum leaves the defendant unable to value the medical claim apart from the wage claim or to test either against a record.³⁹

Like the other initial disclosures, the computation is due at or within 14 days after the Rule 26(f) conference unless a stipulation or court order sets a different time.⁴⁰ Parties sometimes defer it, together with the witness and document disclosures, by agreement, and the timing sections below take up how that works.

Insurance Agreements

Rule 26(a)(1)(A)(iv) requires a party to make available, for inspection and copying as under Rule 34, any insurance agreement under which an insurance business may be liable to satisfy all or part of a possible judgment in the action or to indemnify or reimburse for payments made to satisfy the judgment.⁴² The text covers two arrangements. Some liability policies call for the insurer to pay what the insured owes once judgment is entered, while others have the insured pay first and the insurer indemnify, that is, repay, the insured afterward. By naming both, the rule makes the disclosure turn on whether an insurer may end up bearing the loss, whichever way the money moves. The rule also speaks of a possible judgment, so the duty arises at the outset, long before anyone knows whether the claimant will win.

This item departs from the witness and document items in one respect that matters a great deal. Those items reach only what the disclosing party may use to support its claims or defenses, and the insurance item contains no such phrase.⁴³ The omission matters because a defendant seldom relies on its liability coverage to prove its defense, so a may-use limit would rarely reach a policy. The item asks instead about the policy's possible role in paying a judgment, which is the fact an opponent actually needs.

That fact drives the value of a case. A plaintiff weighing a settlement offer needs to know whether a $50,000 policy or a $5 million policy stands behind the defendant, because the coverage limit suggests a settlement range.⁴⁴ The defendant alone holds the policy, and nearly every plaintiff would ask for it, so the rules collect it automatically instead of making each plaintiff send the same request. That is the saving automatic disclosure promises throughout, since it targets information that would invariably be requested.⁴⁵

The form of the duty also differs from the document item. There the party may supply copies or describe the material by category and location, but here the text calls for the agreement itself to be available for inspection and copying as under Rule 34.⁴⁶ The opponent may therefore examine the policy and copy it, exactly as in response to a request for production, and a summary of the coverage does not discharge the duty. The damages item taught above works the same way for the records behind a computation.

The value that justifies early disclosure also explains why defendants resist it. Coverage limits reveal a settlement range, and policies may also be evidence of ownership or control of environmentally sensitive lands when the defendant is a copper company.⁴⁷ In Flores v. Southern Peru Copper Corp., the parties stipulated to postpone three categories of initial disclosure until the court ruled on the defendant's dispositive motion, but they disagreed about the insurance disclosure: the defendant wanted it postponed as well, and the plaintiffs did not.⁴⁸ The case shows that insurance is the category a defendant is most tempted to hold back. The rule offers two authorized ways to change the timing, a stipulation among the parties and, when they disagree, an order of the court.⁴⁹ Simply withholding the agreement while a motion is pending is not among them.⁵⁰

The usual error is to run this item through the may-use filter that governs witnesses and documents and to conclude that an unmentioned policy can stay out of the disclosures. The filter that applies here is whether an insurer may stand behind the judgment. That filter also marks the other pole, where the item produces nothing.

The rule requires a party to make available an agreement that exists; it does not require anyone to buy coverage or to create a document. A self-insured defendant, and a defendant whose only policies cover unrelated losses, have nothing to disclose under this item.

Exempt Proceedings

Rule 26(a)(1)(B) lists proceedings that are exempt from initial disclosure, so the witness, document, damages, and insurance duties never arise in a case on the list.⁵² The list contains eight categories of cases, each chosen on the theory that the case is one in which there is likely to be little or no discovery.⁵³ The principle that ties the categories together matters more than any single entry, so one illustration follows in place of a catalogue.

That expectation explains why the exemption exists. Initial disclosure stands in for the information a party would invariably request first, and its value depends on there being a first round to replace.⁵⁴ Where no one expects the parties to ask each other for much of anything, the duty has little to accomplish.

The list is exclusive, and no other general exemptions may be created by local rule.⁵⁵ History explains the limit. Under the 1993 version of the rule, a court could by local rule or order exempt types of cases, modify the information to be disclosed, or eliminate the disclosure requirements in a particular case. Many districts chose to opt out in whole or in part, and others adopted disclosure requirements of their own that often differed from those in Rule 26(a).⁵⁶ The 2000 amendments established a nationally uniform practice by removing the authority to opt out by local rule, and they invalidated the informal standing orders of an individual judge or court that purport to create exemptions from, or to limit or expand, the national disclosure requirements.⁵⁷ A lawyer can therefore tell whether a proceeding is exempt by reading Rule 26(a)(1)(B), without checking a district's local rules or a judge's standing orders. An ordinary tort or contract case does not become exempt because it looks modest; the expectation of little discovery shaped the choice of categories and does not operate case by case.

Case-specific orders are a different matter. The committee note to the 2000 amendments says that case-specific orders remain proper and are expressly required if a party objects that initial disclosure is not appropriate in the circumstances of the action.⁵⁸ The rule itself builds in two ways to delay initial disclosures: a stipulation by the parties if they agree, and an order of the court if they do not.⁵⁹ When a party objects during the conference and states the objection in the proposed discovery plan, the court must determine what disclosures, if any, are to be made and must set the time for them.⁶⁰ In Flores, the parties used the first of those two ways for three categories by stipulation, and the defendant asked the court to use the second for the fourth.⁶¹ A blanket exemption for a class of cases conflicts with the national rule, while a deferral shaped to the facts of one case uses authority the rule preserves.

Exemption from initial disclosure carries two further consequences, and later sections develop both. Rule 26(f) requires the parties to confer as soon as practicable, and at least 21 days before a scheduling conference is held or a scheduling order is due, except in a proceeding exempted from initial disclosure or when the court orders otherwise.⁶² The conference agenda helps explain the exception, because it includes arranging the initial disclosures and developing a proposed discovery plan, and an exempt proceeding has no initial disclosures to arrange.⁶³

The moratorium on early discovery follows the same line. Rule 26(d) forbids a party to seek discovery from any source before the parties have conferred under Rule 26(f), except in a proceeding exempted from initial disclosure or when the rules, a stipulation, or a court order authorize it.⁶⁴ The bar is pegged to the conference, and an exempt proceeding does not require one, so the exception lets the parties use the ordinary request-driven devices from the start. A student who concludes that an exempt proceeding is a case without discovery has misread the exemption: it removes the automatic duty and the waiting period and leaves interrogatories, requests for production, and depositions fully available.