Navy Photo Removal and Records Purges Raise Transparency Concerns

The Navy's newest removal order is easy to misread in both directions. NAVADMIN 26170, reported on July 17, 2026, directs commands to take down most command-triad portraits and biographies from public-facing Navy websites for officers at O-9 and below, while flag-officer biographies remain public and internal, non-public hosting continues.[1][2] That is not, by itself, proof that a federal record has been destroyed. A public web page can disappear while the underlying photograph, biography, metadata, and publication history remain preserved somewhere lawful and searchable.

That distinction matters for any serious discussion of the Navy photo-removal controversy and its legal transparency concerns. Public removal is a content-management act. Records destruction is a legal event. The problem with the Navy order, at least from the public record now available two days after it surfaced, is that the reported rationale is clearer than the preservation trail: adversary data aggregation is named, but no reported procedure explains where the removed material goes, who preserves it, how long it remains retained, or how FOIA officers are supposed to locate it later.[1][2]

The order would be a narrower transparency issue if it stood alone. It does not. Since early 2025, the Defense Department has also been removing large volumes of imagery under a DEI-related directive, and the USS Iwo Jima boat-strike controversy has raised allegations that survivor photo and video documentation was ordered deleted after a deadly October 2025 strike.[3][4] Those three events do not share the same stated justification. They do not, on the available record, show coordination. What they do share is a missing public answer to the same records question: was the material merely taken out of public circulation, or was it preserved in a way that keeps statutory accountability alive?

Three Removal Regimes, Three Different Rationales

NAVADMIN 26170 has the cleanest stated security theory. Public biographies and portraits can be aggregated. A command-triad page that looks harmless when viewed by a sailor's family, a local reporter, or a congressional staffer can look different when scraped at scale, cross-linked with deployment cycles, family information, social-media accounts, and command histories. The Navy is not wrong to treat aggregation as a modern operational-security problem.

But OPSEC does not answer the records question. It answers why public access may be curtailed. It does not answer whether the removed biographies are retained as federal records, whether prior public versions are captured, whether the removal decision itself is documented, or whether an eventual FOIA search will find the material without depending on the memory of a web administrator who has rotated out of the command.

The Pentagon DEI image purge operates from a different rationale: compliance with a political and executive-order-driven directive. Associated Press reporting in March 2025 described more than 26,000 images flagged for removal, with one U.S. official estimating the final total could exceed 100,000.[3] The same reporting said the Pentagon acknowledged it had "incorrectly" removed some content, including photographs of the Enola Gay aircraft, apparently flagged because the word "gay" appears in its name.[3]

That episode is less useful as a culture-war exhibit than as a process exhibit. If a system can flag the Enola Gay as DEI content, the review architecture is too blunt for the records environment it is operating in. If one Marine Corps civilian was handling removals for the entire service, as AP reported, the staffing model was not built for careful distinctions among duplicate public copies, historically significant records, commemorative imagery, caption metadata, and legally responsive materials.[3]

The USS Iwo Jima allegation is more combustible. American Oversight said personnel were allegedly ordered to delete survivor photo and video documentation after the deadly October 2025 boat strike, sent a December 2025 letter demanding that the National Archives and Records Administration investigate under the Federal Records Act, and renewed that demand in May 2026 after the Navy acknowledged an "ongoing investigation."[4] That is not the same fact pattern as taking a biography off a public website. If survivor documentation was actually destroyed, and if it was a federal record or evidence relevant to an inquiry, the issue moves much closer to unlawful removal, spoliation concerns, and obstruction-adjacent territory. The important qualification is that the NARA and Navy processes described in the public materials remain unresolved.[4]

The Legal Baseline Is Preservation Before Messaging

Federal records law is not a public-relations norm. It is a preservation system. The Federal Records Act requires agencies to make and preserve adequate documentation of their functions, policies, decisions, procedures, and essential transactions, and it creates duties around unlawful removal or destruction of records. FOIA then supplies a separate access baseline: if a record exists and is not exempt, the public may seek it. The Administrative Procedure Act can also matter when agency action is alleged to be arbitrary, inadequately explained, or inconsistent with legal obligations.

Those laws do not require the Navy to publish every command portrait forever. They do not require the Pentagon to keep every image online. They do not require release of operational imagery that is properly classified, exempt, or tied to an active investigation. The legal question is narrower and more durable: did the government preserve records before it removed public access, and did it preserve them in a way that leaves later oversight practically possible?

That practical point is often where records disputes become real. A FOIA officer cannot search what was never migrated, indexed, or retained. An inspector general cannot reconstruct a deletion chain from an instruction that says only to remove content. Congressional staff cannot test an agency's explanation if the agency cannot identify the custodian, repository, retention schedule, or exception used. NARA lawyers cannot assess unlawful removal cleanly if the first serious inventory occurs months after the purge.

The visible difference between a lawful takedown and a legally dangerous purge is usually not the disappearance of a web page. It is the presence or absence of a contemporaneous paper trail: the records schedule consulted, the preservation repository selected, the metadata retained, the exemptions considered, the deletion authority invoked or rejected, and the person accountable for confirming that public removal did not become record loss.

Why the DEI Purge Matters to the Navy Order

The DEI image sweep is the bridge between ordinary website administration and institutional records risk. The volume alone matters: more than 26,000 images flagged, with a possible final count above 100,000 according to an official cited by AP.[3] But scale is not the whole problem. Scale plus vague criteria plus limited staffing is how public-removal projects become records-disposition projects without anyone formally admitting that the second thing is happening.

The Enola Gay error has become the memorable example because it is almost too obvious. Yet its legal significance is not that an embarrassing keyword match occurred. It is that a historically important military image could be swept into a removal process for reasons unrelated to its historical meaning, operational sensitivity, or records status.[3] If the system caught that error because people were watching, the correction process matters. If it caught that error only after public embarrassment, the public still does not know how many less famous images were treated the same way.

This is where NAVADMIN 26170 deserves scrutiny early, not after a lawsuit. The Navy's order is narrower than the DEI sweep. It is targeted at command-triad portraits and biographies, not an open-ended ideological image category. It has a facially plausible OPSEC rationale. It reportedly leaves internal non-public hosting in place.[1][2] Those facts reduce some risks. They do not eliminate the need for a records instruction attached to the takedown instruction.

A well-built version of the Navy process would not have to disclose sensitive details. It could still say, in general terms, that removed portraits and biographies must be retained under applicable records schedules, that commands may not delete source files absent authorized disposition, that metadata and publication dates must be preserved, and that FOIA search locations remain available to records personnel. The public reporting so far does not describe that kind of concurrent safeguard.[1][2]

The Iwo Jima Allegation Is Different in Kind

The USS Iwo Jima records dispute should not be flattened into the same category as portrait removal. Public biographies and command photos are institutional information products. Survivor photo and video documentation after a deadly strike, if it exists as described, is potential evidence, potential operational documentation, and potential historical record at the same time.

American Oversight's account is still an allegation-driven record, not a final government finding. Its December 2025 demand asked NARA to investigate possible Federal Records Act violations, and its May 2026 renewal cited the Navy's acknowledgement of an ongoing investigation.[4] That posture matters. It would be premature to state that the Navy unlawfully destroyed records. It would be equally unserious to treat the allegation as a routine website cleanup dispute.

The reason is evidentiary timing. A command can decide that a public biography creates aggregation risk before any particular controversy attaches to it. A DEI image purge can be defended, at least initially, as compliance activity gone overbroad. But an alleged order to delete survivor documentation after a deadly strike lands inside a zone where foreseeable investigations, oversight requests, and preservation duties are not abstractions. If deletion occurred, the agency's later explanation would have to do more than invoke sensitivity. It would have to account for preservation, chain of custody, and lawful disposition.

That is why the Iwo Jima matter sharpens the broader pattern. It shows the high-end risk of a culture that treats removal, deletion, and records control as administrative synonyms. They are not synonyms. A commander may have reasons to restrict access. An investigator may have reasons to sequester evidence. A records officer may have reasons to preserve material outside public systems. Those are different acts, with different legal consequences.

The Counterarguments Are Real

There are several reasons not to overstate the case. NAVADMIN 26170 is only two days old as of July 19, 2026. No reported FOIA lawsuit, congressional inquiry, or formal legal challenge has yet tested it. The full administrative message has been described through press reporting rather than litigated agency production.[1][2]

The DEI purge totals also remain uncertain. AP reported more than 26,000 flagged images and a possible final total above 100,000 based on an official's estimate, not a completed public accounting.[3] The Pentagon's acknowledged erroneous removals show process failure, but they do not by themselves prove unlawful destruction. A removed public image may still exist in Defense Visual Information Distribution Service archives, local command systems, records-management repositories, or other non-public stores. The public problem is that outsiders cannot assume preservation merely because preservation would have been legally prudent.

The Iwo Jima matter is also unresolved. American Oversight's demand and the Navy's acknowledged ongoing investigation establish that a serious records question exists; they do not establish the final answer.[4] An inquiry could find that files were preserved elsewhere, that the alleged deletion order was misunderstood, that copies survived in official systems, or that a lawful investigative hold displaced ordinary access. It could also find something worse. The current public record does not justify pretending either outcome has already happened.

There is also no direct evidence that the Navy portrait order, the Pentagon DEI image sweep, and the USS Iwo Jima alleged deletions were coordinated. The better claim is cumulative and structural: different offices, under different pressures, have repeatedly produced removal activity without a visible preservation architecture appearing at the same time.

What a Credible Preservation Trail Would Show

The recurring defect is not hard to describe. Before public-facing military material is removed at scale, the agency should be able to identify what category of record it is dealing with, which records schedule governs it, whether the material is duplicated elsewhere, whether metadata and captions are preserved with the file, and how future FOIA searches will locate it. If operational risk requires speed, the instruction can still include a temporary hold: remove from public view now, preserve source material and publication metadata until records counsel approves disposition.

For command biographies, that means the Navy should know whether the biography is merely a current public-affairs page or part of a command-history record. For photographs, it means source files and captions should not be treated as disposable just because a derivative web copy is being unpublished. For strike documentation, it means deletion should be almost unthinkable until investigators, records officers, and counsel have resolved preservation obligations.

• A takedown instruction should say whether removal is public-only or includes source-system deletion.
• A records instruction should identify the applicable retention schedule or require legal review before disposition.
• A FOIA instruction should tell search personnel where removed material remains accessible for lawful processing.
• An audit trail should record who ordered removal, who executed it, and what repository retained the material.
• An error-correction process should restore wrongly removed public material without erasing the fact that it was removed.

None of that requires the military to publish sensitive material. It requires the military to keep public removal from becoming silent legal disappearance. The distinction is especially important in defense institutions, where the same record can serve several audiences over time: a commander today, a FOIA officer next year, an inspector general after that, and a historian decades later.

The Pattern Is Procedural, Not Conspiratorial

The strongest transparency concern is not that every removal was unlawful, or that the three events were centrally coordinated. The available record does not prove either. The concern is that three distinct removal regimes, arising within roughly the same 18-month window from March 2025 through July 2026, all point toward the same institutional weak spot: removal orders are visible, preservation procedures are not.

That weakness has legal consequences even before a court rules. FOIA becomes less meaningful when requesters cannot know whether removed material is still searchable. Federal Records Act duties become harder to police when agencies do not say whether removal triggered records review. Administrative-law scrutiny becomes more plausible when large-scale agency action is justified in broad terms but implemented through opaque, under-documented processes.

Security can justify taking command biographies offline. Executive directives can require agencies to review public content. Operational sensitivity can justify restricting access to strike documentation. None of those rationales answers the preservation question by itself. The answer has to be built into the process while the removals occur, not reconstructed after reporters, watchdogs, or NARA ask what happened.

On the record now available, the qualified conclusion is enough: no single incident is clearly unlawful on its own, and there is no direct evidence tying the three initiatives together. But the repeated absence of concurrent, public-facing archival-preservation safeguards makes the combined pattern more serious than three isolated deletion stories. In military records practice, a missing web page is not automatically a destroyed record. A missing preservation trail is the warning sign.

References

1. Navy instructs commands to remove portraits, biographies from websites, Navy Times, July 17, 2026.
2. Navy removes commander bios, Task & Purpose.
3. Pentagon purge of DEI images removes — and then restores — historic military photos, AP News, March 2025.
4. American Oversight Calls on NARA to Investigate Boat Strike Records Destruction, American Oversight.