Redaction is most commonly used to either withhold specific sensitive information as required by court rules, such as Social Security or telephone numbers, or to obscure privileged information when producing documents that contain both privileged and nonprivileged information. However, some practitioners have used redaction as a tool to withhold nonresponsive information from otherwise responsive documents.

The question of whether redacting for nonresponsiveness is permissible has been raised with increasing frequency given the demands of modern discovery. In particular, broad document requests have led to serious concerns on the parts of corporate  litigants who are loath to reveal sensitive information to their opponents or even to their co-defendants, many of which are direct competitors. The Federal Rules of Civil Procedure are silent regarding this issue, and the case law arising out of New York federal district courts provides no definitive guidance.

Perhaps the largest divide rests in the chasm between decisions that implicitly endorse the practice and those that prohibit it. This conundrum is demonstrated by cases such as Bank Brussels Lambert v. Chase Manhattan Bank (S.D.N.Y. Sept. 10, 1997), in which the court permitted redaction of documents without explanation where “[t]he information is non- responsive,” as opposed to John Wiley and Sons Inc. v. Book Dog Books LLC, 298 F.R.D. 184 (S.D.N.Y. 2014), in which the same court noted that “redactions of portions of a document are normally impermissible unless the redactions are based on a legal privilege.” See also Howell v. City of New York (E.D.N.Y. Sept. 25, 2007) (“It is not the practice of  this court to permit parties to selectively excise from otherwise discoverable documents those portions that they deem not to be relevant.”).

Further, while in In re State Street Bank and Trust Co. Fixed Income Funds Investment Litigation (S.D.N.Y. April 8, 2009) the Southern District Court of New York stated that redacting for nonresponsiveness was “generally unwise” as it “breed[s] suspicions, and ... may deprive the reader of context,” the Southern District of New York allowed redacted documents to remain so because the redacted portions were irrelevant in the case Strategic Growth International v. RemoteMDX Inc. (S.D.N.Y. Nov. 9, 2007).[1]

Courts generally appear more willing to permit redaction in order to protect fundamental privacy considerations. But even in cases where fundamental privacy rights are at issue, these considerations rarely become an explicit element of the courts’ analyses. To  illustrate, the reasoning behind the court’s approval of redacting for relevancy from meeting minutes in Schiller v. City of New York, (S.D.N.Y. Dec. 7, 2006), which appears to stem from the fact that the discovery dispute primarily falls within First Amendment considerations, is still not expressly stated. The court merely states: “Suggestions that were considered and rejected by the organization have no bearing on whether the plaintiffs intended to engage in civil disobedience at the World Trade Center site.”

Schiller echoes Davis v. City of New York (April 28, 1988), one of the few earlier cases on point, in which the court found that the state’s interest “in protecting the privacy interest of certain categories of government employees clearly outweighs any federal interest,” noting that “the threshold issue must, of course, be relevance of the discovery sought.” Whether this logic will be extended to include more ordinary privacy and confidentiality considerations that arise in standard commercial litigation, such as documents produced in run-of-the-mill breach of contract, fiduciary duty, tortious interference cases, remains to be seen.

Along those same lines, the only guidance provided by the Second Circuit occurs in the dicta of a case involving the intersection of a reporter’s telephone records and the First Amendment. The New York Times Co. v. Gonzales, 459 F.3d 160, 170 (2d Cir. 2006). More specifically, The New York Times sought to repel a grand jury subpoena on the basis that the sought records were privileged. Id. In rejecting arguments related to the records containing information regarding “other newsworthy matters,” the court noted that “[r] edactions of documents are commonplace where sensitive and irrelevant materials are mixed with highly relevant information.” Id. While the Southern District of New York cited The New York Times in rejecting one party’s argument that relevance is an illegitimate basis for redaction in Autohop, in State Street Bank, the same court cited The New York Times in distinguishing case law that allow redactions based on relevance as concerning fundamental rights.

What all of these cases lack is any sort of governing principle. The closest is the Eastern District of New York’s finding in Howell, in which the court stated that under Rule 26(c), allowing a party to redact nonresponsive information from otherwise discoverable documents would “require a finding of ‘good cause’ based on a need ‘to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense.” However, the court neither elaborated on this analysis nor elucidated the basis for requiring a protective order.

Silent rules of procedure and unclear case law place practitioners in a tenuous position. The issue is less about which practice is advisable and more about the uncertainty of courts’ positions. With some courts increasingly expressing somewhat of a distaste for the practice without clarity as to their bases for disfavoring it, redacting for nonresponsiveness necessarily becomes more risky, leaving practitioners with a cost-benefit analysis between following the safest and most expensive course, seeking a protective order, and riskier options, such as following the redact now, ask permission later practice.

Originally appeared on Law360 on July 31, 2015.