ESI Protocol Should Define “Documents” and Address Redaction Based on Irrelevancy

E-Discovery LLC, ESI Protocol Should Define “Documents” and Address Redaction Based on Irrelevancy by Michael Berman
Image: Holley Robinson, EDRM with AI.

[EDRM Editor’s Note: The opinions and positions are those of Michael Berman.]


We the Protestors, Inc. v. Sinyangwe, 348 F.R.D. 175 (S.D.N.Y. Dec. 18, 2024), makes several important points about the relationship between ESI Protocols and redaction of produced documents.

The decision begins:

Before the Court is a discovery dispute that underscores the importance of counsel fashioning clear and comprehensive agreements when navigating the perils and pitfalls of electronic discovery.

We the Protestors, Inc. v. Sinyangwe, 348 F.R.D. 175 (S.D.N.Y. Dec. 18, 2024).

When it comes to discovery of text messages, Fed.R.Civ.P. 26(f)(3)(B) provides an under-used tool.  It states that a “discovery plan” must state that parties views and proposals on “whether discovery should be conducted in phases….” Text message discovery may benefit from a phased discovery agreement in an ESI Protocol or Discovery Plan.1

While some suggest that an ESI Protocol should be bare bones, there seems to be a consensus that That an ESI Protocol Must be Specific (Apr. 2, 2025).

The primary issue in We the Protestors was whether a producing party could unilaterally redact parts of produced text message strings.  Implicit in that issue is the question of, in the digital realm, “what is a document?” The terms of the ESI Protocol provided the answers.

The underlying case, involving claims and counter-claims, arose “from an acrimonious breakup involving the founders of a nonprofit organization, Campaign Zero, formed to educate the public about police violence through data-driven analysis and to advocate for police reform.”

The court wrote that:

The two sides had agreed to collect and review all text messages in the same chain on the same day whenever there was a text message within the chain that hit on one of the agreed-upon search term[s]. [Emphasis added].

Id.

However, the parties interpreted the agreement differently:

  • “Unlike Defendants, Plaintiffs understood that they only needed to produce messages from the same-day period that were responsive or provided context for a responsive text message.”
  • “Upon reviewing Plaintiffs’ production, Defendants objected and claimed that Plaintiffs’ unilateral redaction of text messages within a same-day text chain was improper.”

Defendants moved to compel production of “unredacted copies of all text messages in the same chain that were sent or received within the same day.”  Plaintiffs objected and, alternatively, sought a protective order. 

This blog focuses on redaction.  Unilateral redaction was deemed to be improper under the parties’ incomplete agreement.  The protective order was granted in part subject to a meet-and-confer mandate and is not discussed in this blog.

The court began its analysis as follows:

Text messages are an increasingly common source of relevant and often critical evidence in twenty-first century litigation. They do not, however, fit neatly into the paradigms for document discovery embodied by Rule 34 of the Federal Rules of Civil Procedure, which was crafted with different modes of communication in mind.

Id.

The court contrasted email, writing that:  “Email at least retains a resemblance to conventional documents in that each email or email chain can be viewed as a single, identifiable ‘document.’”

In my opinion that statement is correct only as a general matter.  Hyperlinks or “modern attachments,” for example, also present the issue of “what is the document?”  See, e.g., Uber Technologies – Another Hyperlink Decision (Mar. 6, 2025); ESI Protocol Dispute – “Modern Attachments” and the “Humpty Dumpty Issue” (Nov. 13, 2023).  Like texts, an email with a “modern attachment” – – a pointer hyperlink and a target document – – does not fit neatly into the court’s description of email as a “single, identifiable document.”

However, having characterized email as a single document, the We the Protestors court next distinguished text messages from email:

With text messages this is not so clear. For discovery purposes, should each text message be viewed as its own stand-alone “document” or item of ESI? Or is the relevant “document” the entire chain of text messages between the custodian and the other individual or individuals on the chain—which could embrace hundreds or thousands of messages going back for years? Should the producing party be allowed to redact non-responsive texts and, if so, to what extent? Litigants, and courts, are still in the process of figuring out how to answer these questions.

Id.

Noting that plaintiffs’ counsel described email as a “stream of consciousness,” the We the Protestors court explained that:

“Federal courts have adopted different approaches with respect to text messages.” Lubrizol Corp. v. IBM Corp., No. 1:21-CV-870-DAR, 2023 WL 3453643, at *3 (N.D. Ohio May 15, 2023). Some courts have suggested that “a party must produce the entirety of a text message conversation that contains at least some responsive messages.” Id. (citing cases). This is the approach taken in the leading case on the issue in this District, Al Thani v. Hanke, No. 20 Civ. 4765 (JPC), 2022 WL 1684271 (S.D.N.Y. May 26, 2022).

Id.

In Al Thani, the court also restricted redacting.2

The We the Protestors court explained the opposing views:

By contrast, some courts in other jurisdictions have held that “the producing party can unilaterally withhold portions of a text message chain that are not relevant to the case.” Lubrizol, 2023 WL 3453643, at *4 (citing cases).

Id.

It added:

“Still other courts have taken a middle ground.” Id. (citing cases). In Lubrizol, which involved a dispute concerning Slack instant messages (which the court analogized to text messages), the court approved plaintiff’s proposal that, for any Slack conversation containing more than 20 total messages, the defendant be required to produce the ten text messages immediately preceding, and the ten messages immediately following, any responsive text message. Id. at *1, *4.

Id.

The We the Protestors court strongly suggested that the best course of action is an agreement between the parties:

Litigants are free to—and are well-advised to—mitigate the risk of this uncertain legal regime by coming to their own agreement about how to address text messages in discovery. Rule 29(b) specifically affords parties the flexibility to design their own, mutually agreed upon protocols for handling discovery. See Fed. R. Civ. P. 29(b)…. [Emphasis added].

Id.

The court suggested that negotiation was possible because of what, during the Cold War, was called “mutually-assured destruction”:

In the context of text messages, an agreed-upon protocol is particularly sensible where, as here, both sides are seeking and will be producing such information. A party may think twice about insisting on the most burdensome and costly method of reviewing and producing text messages for its adversary if it knows it will be subject to the same burden and cost. Similarly, when it comes to redactions, while a party may be happy to receive unredacted messages from the other side, it may forgo that potential benefit if that allows it to redact information from its own production. In general, the parties are better positioned than the court to customize a discovery protocol that suits the needs of the case given their greater familiarity with the facts, the likely significance of text message evidence, and the anticipated volume and costs of the discovery.

Id.

In We the Protestors, the parties reached an agreement, but disagreed over its application.

Counsel for Plaintiffs ultimately sent an email stating: “We are amenable to reviewing all texts in the same chain sent or received on the same day as any text that hits on any of the search terms. Please confirm whether [Defendants] are prepared to do the same.” … Counsel for Defendants replied: “Confirmed.”

Id.

Based on their view of the agreement, plaintiffs redacted 1,015 text chains.

The We the Protestors court wrote that it was resolving a dispute based on the parties’ agreement.  It emphasized that it was not seeking the “right answer” to the redaction question.  “The Court’s task on this motion is thus more akin to filling a gap in the parties’ incomplete agreement.”

What the parties’ agreement did not explicitly address is whether, in producing those same-day text chains, texts deemed irrelevant and non-responsive would be redacted or, instead, the chains needed to be produced in their entirety.

Id.

Viewing the agreement as a contract, the We the Protestors court then applied Al Thani’s holding that “’parties may not unilaterally redact otherwise discoverable’ information from text messages for reasons other than privilege.”  The court wrote that:

It should have been clear to Plaintiffs that Al Thani operated as a default rule forbidding redactions in text message chains absent a judicial decision, or an agreement by Defendants, permitting redactions…. [I]f Plaintiffs wanted to make redactions without Defendants’ agreement, Plaintiffs needed to seek court permission to do so….  Plaintiffs were not free to decide on their own that redactions to Plaintiffs’ production were appropriate. [Emphasis added].

Id.

The court added:

It is not as if the parties had agreed that redactions were permissible with respect to other document discovery. To the contrary, Plaintiffs’ counsel acknowledged that Plaintiffs produced emails without redactions, even though the parties followed the “same principle” for emails as they did for text messages….

Id.

The We the Protestors court wrote that “the Court construes the absence of a provision in the parties’ agreement allowing redaction of text messages to preclude Plaintiffs from unilaterally making such redactions.”

[T]he Court construes the absence of a provision in the parties’ agreement allowing redaction of text messages to preclude Plaintiffs from unilaterally making such redactions.

We the Protestors, Inc. v. Sinyangwe, 348 F.R.D. 175 (S.D.N.Y. Dec. 18, 2024).

Redactions of purportedly irrelevant information in a responsive document have generated a substantial amount of case law and I suggest that We the Protestors follows the majority approach:

One argument in favor of redaction of irrelevant parts of a responsive document is that the scope of discovery governs and the redacted material is outside the scope.  Therefore, it is not discoverable.

Arguments against unilateral redaction based on irrelevance include the need for context and that a discoverable document should be discoverable in its entirety:  “It is a rare document that contains only relevant information; and irrelevant information within an otherwise relevant document may provide context necessary to understand the relevant information.” Athletics Investment Grp., LLC v. Schnitzer Steel Industries, Inc., 2024 WL 2191007 (N.D. Cal. May 14, 2024); accord Al Thani v. Hanke, 2022 WL 1684271, at *2 (S.D.N.Y. May 26, 2022)(citation omitted). Courts have also said that redaction breeds suspicion.  And, some courts have agreed that non responsiveness redaction is not permitted unless there is an articulated harm supporting it.3

Another question presented in We the Protestors is “what is the document?”  The parties used a arbitrary temporal definition and that may be reasonable.  Others suggest an arbitrary yardstick of X messages before and X messages after the text message that is relevant. 

All of this is an effort to provide context to a communication method – text messaging – that commingles content.  A broken clock is correct twice a day.  Any arbitrary yardstick delineating the cutoff point for text messages is likely to capture irrelevant information at least some of the time and may also “miss” messages that are needed for context, but which are outside of the defined window.

Back in the days of paper-based litigation, it was rare to argue over the definition of a “document.”  What is a “Document?” (Aug. 17, 2021).  However, that view was moved to the rear-view mirror by decisions such as Sandoz v. Un. Therapeutics Corp., 2021 WL 2453142 (D.N.J. Jun. 16, 2021)(texts), and, Nichols v. Noom, Inc., 2021 WL 948646 (S.D.N.Y. Mar. 11, 2021)(hyperlinks).

I have raised a similar issue in the context of spreadsheets, Excel workbooks, and other multi-content electronic files, such as a PDF Portfolio. What is a Document? (Part II) (Aug. 28, 2021); What is a Document? (Part III) (Apr. 4, 2022). 

A parallel issue is presented by hyperlinked documents in email, “Modern Attachments” or “Pointers”- What is a Document? (Part IV) (Aug. 12, 2022); ESI Protocol Dispute – “Modern Attachments” and the “Humpty Dumpty Issue” (Apr. 29, 2024); More on “Modern Attachments,” “Pointers,” or Hyperlinked Documents – Humpty Dumpty and “Usability”

The Sedona Conference has raised the same question regarding collaboration platforms.  Sedona Conference Commentary on Discovery of Collaboration Platforms – What is a Document? (Apr. 10, 2025), referring to The Sedona Conference®  “Commentary on Discovery of Collaboration Platforms Data, Public Comment Version” (Apr. 2025).  

In We the Protestors, the defintional window was defined as one day.  One may ask, “what was the document?”  Was a full day of text messages a “document”?  What would happen if a key message was sent at the end of the day and the response sent the next day did not “hit” on a keyword?  Would the parties have to “live with” their agreement, thereby missing relevant, proportional, discoverable information?

I have suggested in several blogs that an arbitrary yardstick such as that used in We the Protestors, is likely reasonable and appropriate for discovery.  I have no criticism of it and it may be perfectly appropriate in an ESI Protocol or Discovery Plan.  Since the earliest days of e-discovery, parties have been advised to discuss the form or forms of production.  The Sedona Commentary emphasizes this – – and much more – – in the context of collaboration platforms.

However, because discovery is not an end in itself, evidentiary issues may arise when a “document” is truncated during the unitization and discovery process. ESI Protocol Dispute – “Modern Attachments” and the “Humpty Dumpty Issue” (Nov. 13, 2023);  What is a “Document?”  (Aug. 17, 2021). 

And, of course, the need for “context” was addressed in decisions such as Sandoz v. Un. Therapeutics Corp., 2021 WL 2453142 (D.N.J. Jun. 16, 2021)(text bubbles).

Let’s assume a civil action for battery.  A text that “Joe hit me” is relevant.  Assume it was followed by 30 texts about the following night’s dinner.  One litigant may deem them irrelevant while the other may contend that they demonstrate a lack of pain and suffering.

Let’s assume a civil action for breach of contract.  A text states “I ordered 50 copies and only receive 10.”  Assume it is followed by 30 texts about the next night’s dinner.  They are likely irrelevant by any standard.

There is  no one size fits all approach and, when parties are negotiating an agreement, at least one of them likely has little or no knowledge about the other’s text messages.

In my opinion, one factor that should be considered in these discussions is how the produced communications can be used in depositions, motions, and at trial.  See, e.g., What is a “Document?” (Aug. 17, 2021).  “Half” of a document, or a partial document, may be objectionable.

Fed.R.Evid. 106 states:

If a party introduces all or part of a statement, an adverse party may require the introduction, at that time, of any other part–or any other statement–that in fairness ought to be considered at the same time. The adverse party may do so over a hearsay objection.

Fed.R.Evid. 106.

The Advisory Committee Note provides an example: “[A]ssume the defendant in a murder case admits that he owned the murder weapon, but also simultaneously states that he sold it months before the murder. In this circumstance, admitting only the statement of ownership creates a misimpression….” 

A similar issue may be presented by complex electronic communications.  For a more detailed example of the evidentiary issue, please see What is a “Document?” (Aug. 17, 2021).

Agreements regarding ESI should not become traps.  Be Careful What You Agree To, But the Duty to Cooperate May Provide an Escape Hatch (Part II) (Nov. 11, 2022). 

Courts should not unreasonably require compliance with an agreement that is no longer reasonable and proportional, while also not casually disregarding such agreements.  Sanctions Denied, But Be Careful What You Agree To (Part I) (Oct. 2, 2022); The ESI Protocol: Your Word is Your Bond… Or, Is It? (Jan. 11, 2024); StubHub: Modification of ESI Protocol and Denial of Sanctions – Performance Was Impossible (May 23, 2024); ESI Protocol Dispute – “Modern Attachments” and the “Humpty Dumpty Issue”  (Apr. 29, 2024).

When it comes to texts, context should be the key determinant of the scope of the duty to produce. “[A] single text message, standing alone, is oftentimes meaningless without other messages in the text chain to provide context.”  Al Thani, 2022 WL 1684271, at *2; Sandoz, 2021 WL 2453142, at *2 (directing production of “context-related text messages”).

If parties make a reasonable agreement for discovery purposes based on what is known at the time; however, later production and review demonstrates a need for modification to provide appropriate contextual information, that changed circumstance should constitute cause for an application to the court to modify the agreement.

Assume, for example, that the parties in We the Protestors had agreed:

Text message discovery will be conducted in phases. 

In Phase 1, a producing party shall collect and review all text messages in the same chain on the same day whenever there was a text message within the chain that hit on one of the agreed-upon search terms.4 Absent good cause shown, and with the exception of privileged material, the entire day’s message chain shall be produced.  If the producing party has good cause, it may redact material that is beyond the scope of discovery under Fed.R.Civ.P. 26(b)(1) or subject to protection under Fed.R.Civ.P. 26(c).  Redactions shall be clearly marked with the specific reason for the redaction. 

In Phase 2, if the discovering party believes that necessary context is lacking, or that Fed.R.Evid. 106 requires production of redacted information, it shall notify the producing party and, after meeting and conferring, the discovering party may move to compel production of redacted text that it alleges is contextual or required by Rule 106. 

The substantive standard for resolution of any such motion shall be the scope of discovery under Fed.R.Civ.P. 26(b)(1), the rule of completeness under Fed.R.Evid. 106, the need for protection under Fed.R.Civ.P. 26(c), and, the need for the redacted material to provide context to the produced material.  The burden shall be on the redacting party to show that redaction is proper by a preponderance of the evidence.

Nothing contained herein modifies a party’s right to redact privileged information or attorney work product under the appropriate other provisions of this Discovery Plan.

One might think that, before producing hundreds of unredacted text chains in their entirety (in the case of Defendants), or before going to the trouble and expense of redacting tens of thousands of messages from more than a thousand text chains (in the case of Plaintiffs), counsel would have contacted its adversary to confirm that the other side was handling their production in the same manner.

We the Protestors, Inc. v. Sinyangwe, 348 F.R.D. 175 (S.D.N.Y. Dec. 18, 2024).

While not mentioning phased discovery, the We the Protestors court wrote:  “One might think that, before producing hundreds of unredacted text chains in their entirety (in the case of Defendants), or before going to the trouble and expense of redacting tens of thousands of messages from more than a thousand text chains (in the case of Plaintiffs), counsel would have contacted its adversary to confirm that the other side was handling their production in the same manner. Such an inquiry would have flushed out the parties’ disparate understandings and led to additional negotiations and, perhaps, an agreement on the redaction issue. But neither counsel called or emailed (or texted) its adversary to engage in that discussion.”


Notes

  1. For the differences between the two terms, please see “ESI Protocol” v. “Discovery Plan” (Jan. 2, 2024). ↩︎
  2. In Al Thani, the plaintiff self-redacted 259 communications based on her view of relevance, without involvement of her counsel.  2022 WL 1684271 at *1.  The court reviewed them in camera and ordered production without redactions. The court wrote that “[a] court may, under certain circumstances, permit redactions where there is a finding of good cause based on a need to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense.”  Id. at *2 (cleaned up).  The court relied in part on Fed.R.Civ.P. 26(c). ↩︎
  3. Of course, redaction for privilege is permitted.  Redaction of material protected by statute, personal identifiable information, and information governed by agreements, may also be permissible.  8 Lessons Learned – Part III – Redaction Provision of ESI Protocol. ↩︎
  4. This is the form used in We the Protestors.  Other iterations are possible, such as X messages before and X messages after the message containing the search term.  ↩︎

Assisted by GAI and LLM Technologies per EDRM GAI and LLM Policy.

Author

  • Miichael Berman's headshot

    Mike is the owner of E-Discovery, LLC, and of counsel at Rifkin Weiner Livingston LLC, in Baltimore. He concentrates on commercial litigation and offers mediation services. He was the primary editor of Electronically Stored Information in Maryland Courts (Md. State Bar Ass’n. 2020), and he co-edited M. Berman, C. Barton, and P. Grimm, eds., Managing E-Discovery and ESI: From Pre-Litigation Through Trial (ABA 2011), and J. Baron, R. Losey, and M. Berman, eds., Perspectives on Predictive Coding (ABA 2016). Mike has litigated a number of cases in the trial and appellate courts in Maryland. He is an Adjunct Professor at the University of Baltimore School of Law where he co-teaches a three-credit discovery workshop that focuses on e-discovery. He has lectured at the Maryland Judicial College and he chaired the Bar committee that drafted the proposed ESI Principles for the District of Maryland. He is a past: co-chair of the Federal District Court Committee of the Maryland State and Federal Bar Associations; chair of the Litigation Section Council, Maryland State Bar Association; and, co-chair of the American Bar Association Litigation Section Book Publishing Board. He graduated from the University of Maryland School of Law and is also an Army veteran. He is admitted to the Maryland bar. The opinions expressed in this blog are not necessarily those of Rifkin Weiner Livingston LLC.

    View all posts