Circular reporting

A single source — a citizen informant, a paid asset, a single defector, a single police report — makes a claim. The claim enters Agency A's database. Agency A shares it laterally with Agency B and Agency C, per the post-9/11 information-sharing doctrine that broke down the inter-agency walls and made lateral sharing the default. Each agency's records now show they have the claim. A senior analyst at Agency D pulls a multi-source report on the subject; the report shows Agency A, Agency B, and Agency C all carrying the claim. The senior analyst writes the report up as triangulated from three independent sources. The decision-maker reads the report and acts on what looks like a robust, multiply-confirmed intelligence picture.

There is only one source. The claim has never been independently verified. The same uncorroborated statement is doing all the work, three times, with the source-attribution stripped off at each handoff.

This is circular reporting — the canonical intelligence-community term for the failure mode. It is also, in plainer language, corroboration laundering — the same process by which money launderers strip the provenance off cash by passing it through enough hands that nobody asks where it came from. The mechanism is structural. It does not require malice from any individual analyst. It requires only that the institution incentive is to have the claim rather than to audit its provenance.

The pipeline companion to citizen-informant deployment.

Citizen-informant deployment puts the original report into the system: a participant in a fusion-center reporting network files a Suspicious Activity Report on a target. The target is now in the database. Circular reporting is what happens next. The fusion center shares with the FBI Joint Terrorism Task Force. The JTTF shares with the regional federal-state working group. The working group shares back with state DPS. State DPS pushes it into the National Counterterrorism Center database. At every step, the receiving agency's records now show the claim. The original single source — the citizen informant making the unverified observation — has been laundered through five institutional intermediaries, and any agency querying the system now sees the claim apparently confirmed by five different sources.

The two tactics are halves of one pipeline. Naming both separately matters because the defenses are different at each end: stopping the input is one problem (recruitment, accountability, audit on filing); stopping the multiplication is another (provenance metadata, source-collision detection, re-verification mandates on lateral imports).

The Curveball case.

The canonical national-scale instance is the Iraq WMD intelligence picture. A single defector — Rafid Ahmed Alwan al-Janabi, codename Curveball — was the source for the claim that Iraq had mobile biological weapons laboratories. Curveball was a single source, judged by his immediate handlers at the German BND to be unreliable. His claims were nonetheless passed to the U.S. Defense Intelligence Agency, then laterally to the CIA, the State Department's Bureau of Intelligence and Research, and ultimately into the National Intelligence Estimate and Secretary of State Colin Powell's UN address.

By the time the claim reached Powell's speech, the language treated the mobile-lab finding as a multiply-corroborated fact — Powell himself referred to multiple sources. The Senate Intelligence Committee's post-war report identified circular reporting as the load-bearing failure mode that produced the apparent multiplicity: every agency in the chain was citing the same underlying Curveball reporting, each treating the other agencies' presence of the claim as confirmation rather than as a derivative of the same upstream source. The war that followed was launched off a single uncorroborated claim that had been laundered through enough institutional intermediaries to appear corroborated.

This is the same mechanism, scaled up. What happened to a country can happen to a person.

Why it persists.

The structural reasons circular reporting is hard to root out apply equally at the national-intelligence level and the local-fusion-center level:

• Institutional incentive favors having. An analyst whose records show a claim is in a stronger position than one whose records show only an absence. Having the claim is "intelligence coverage." Re-verifying the claim is overhead.
• Provenance metadata degrades. As a claim moves through systems, the original source attribution gets compressed, summarized, or dropped. By the third or fourth handoff, the claim is often labeled with the transmitting agency rather than the originating source. The downstream reader sees the agency, not the underlying defector or informant.
• Lateral-sharing infrastructure was designed for breadth, not collision detection. The fusion-center, JTTF, and ISE architectures were built specifically to maximize information flow across institutional boundaries after 9/11. They were not built to track source-collision or to flag when a "multi-source" picture is actually a single-source echo.
• The bureaucratic cost of re-verification is real. Each agency would have to track the claim back to the originating source, contact the originating source's handler, and re-validate — every time the claim crosses a boundary. That is expensive, slow, and rarely done.

Aggregate signatures the defender can sometimes spot.

• Same wording across multiple reports. When the supposedly-independent confirmations use identical phrasing, the upstream source is shared. Honest independent corroboration produces different language describing the same underlying fact.
• Same time-window for the underlying observation. Multiple sources confirming the same observation should have observed it at slightly different times from slightly different vantages. Same date + same time + same place is a strong tell that it's one source echoing.
• The "everyone has this" feeling without anyone having had to verify. If the response to who confirmed this is it's in the system, the underlying source-collision is likely.
• Asymmetric impact on the subject. The subject of a circular-reporting picture is the only party affected enough to dig — and the only party structurally barred from auditing the lateral-sharing trail.

What this means for the individual target.

A single false claim about a target, once it enters the lateral-sharing infrastructure, will appear corroborated by the system within months — without any of the receiving agencies ever having re-verified it. The target has no way to know how many "sources" the system thinks it has on them, no way to audit the provenance, and no procedural mechanism to surface the source-collision. The legal-discovery process can occasionally do it during litigation, but the target has to first be charged or sued to get there.

The defensive implications are mostly upstream: stop the original false claim from entering the system in the first place (narrative discipline at the social layer; pattern documentation at the evidentiary layer; counter-record-building so a future investigator who pulls the file finds a contradiction sitting next to the laundered claim). The downstream multiplication is mostly out of the individual's hands. What can be done is to make sure that when the file is eventually examined, the counter-record is dated, sourced, and impossible to claim was assembled after the fact.

Why this tactic deserves its own name.

Naming the pattern is the first step in defending against it. Citizen-informant deployment names the input; circular reporting names the multiplier. Without naming the multiplier, the input-side defense ("don't let false claims get filed in the first place") looks like the whole defense. With the multiplier named, the additional work becomes visible: even a single successful upstream false claim becomes a system-wide picture, and the only way to counter the system-wide picture is to put a system-wide counter-record on the public timestamped table before the lateral sharing finishes its work.

The Iraq war was launched off the system-wide picture being unchallenged in time. The defender's job is to make sure the equivalent picture about them cannot be unchallenged in time.