28,471 American Rifles and the Dead of Semanggi

By Desmond Okafor · September 10, 2026

Seventeen civilians were gunned down in Jakarta in 1998; a confidential DIA cable confirms the bullets came from US rifles provided as grant aid decades earlier.

Seventeen people died in the streets of Jakarta because they stood in the way of a special session of parliament. They were unarmed, and they were shot by the very forces meant to secure the state.

This was the cost of the first Semanggi incident on November 13, 1998. Semanggi, a place named for the aquatic ferns common to the region, became a killing field where the Indonesian police—the national civilian force responsible for law and order—and military special forces opened fire on protesters. This is the human wreckage that underlies a brief, clinical report from the Defense Intelligence Agency. I am writing this column from a curated archival description of the record rather than the original pages, but the summary of the cable is sufficient to map the trajectory of the ammunition.

The Forensic Link

The record is a DIA Combined Information Report dated December 4, 1998, marked Confidential. It does not offer a first-hand intelligence assessment or a forensic report authored by US technicians. Instead, it summarizes open-source reporting from Kompas, the national newspaper of record in Jakarta.

According to the record, Kompas detailed preliminary forensic testing of bullets recovered from the scene of the November 13 shootings. The tests reached a definitive conclusion: the "ammunition used in the killings came from US-provided or sold M-16 rifles used by the Indonesian special forces."

Those forces were identified as KOSTRAD, the Army Strategic Reserve Command. Despite a name that implies a backup force, KOSTRAD is the primary combat unit of the Indonesian Army, a corps-level command of up to 35,000 troops tasked with strategic security operations. In the streets of Semanggi, KOSTRAD did not act as a reserve; it acted as the primary instrument of state violence.

Document imagery from nsarchive.gwu.edu From the files: nsarchive.gwu.edu

A Twenty-Year Paper Trail

The DIA cable does more than mirror a newspaper article; it provides the provenance of the hardware. The record establishes that the United States provided 28,471 M-16 rifles as grant aid to the Indonesian special forces between 1972 and 1977. Beyond that massive grant, the US sold additional rifles under commercial auspices.

This is where the timeline reveals a systemic failure. The rifles used to kill protesters in 1998 were not recent acquisitions. They were the legacy of a geopolitical investment made two decades earlier. The US supplied the tools of the massacre during the mid-seventies, and by the late nineties, those same tools were being used to suppress a democratic impulse in Jakarta.

There is a cold irony in the designation of these weapons as "grant aid." In the lexicon of the State Department, aid is a tool of stability and partnership. In the streets of Semanggi, that aid manifested as high-velocity rounds in the chests of unarmed civilians. The connection is a straight line from a 1972 ledger in Washington to a forensic lab in Jakarta in 1998.

The Mirror and the Shield

The most telling part of this record is not what it says, but how it says it. The DIA is not reporting its own findings. It is reporting that Kompas found the findings.

The desk's reading is that the US government utilized open-source intelligence (OSINT) as a buffer to avoid creating an internal, official finding of complicity. By framing the forensic link as a summary of media reporting rather than an internal intelligence product, the DIA created a layer of plausible deniability. If the US government does not conduct its own forensic audit, it does not have to formally acknowledge that its own grant-aided rifles were the murder weapons. It can simply report that a local newspaper claims they were.

Furthermore, the classification of this cable as "Confidential" is an absurdity if viewed as a means of protecting a secret. The information—that M-16s were used and that they were American—was already public, printed in the pages of the most influential newspaper in Indonesia. The pattern suggests that the "Confidential" stamp was not intended to hide the forensic link itself, but to hide the US government's awareness of that link.

The Gap in the Record

There is a glaring silence in this file. The cable details the weapons, the death toll, and the source of the reporting, but it contains no mention of a policy response. There are no instructions for the embassy in Jakarta, no mentions of diplomatic protests, and no discussions of whether further arms sales should be frozen.

If this file is shaped the way it looks, it is a perception management tool. It was designed to alert the State Department to a public relations liability—the fact that the world could see American bullets in Indonesian bodies—rather than to initiate a human rights investigation. The focus was on the reporting of the event, not the event itself.

This suggests the existence of a parallel, more sensitive set of cables—files that likely detail the diplomatic damage control applied to the Indonesian government to ensure that the "grant aid" relationship remained intact despite the blood on the pavement. The absence of a moral or legal reaction in this report is a redaction by omission.

The Reading

The pattern of this record indicates a total abandonment of end-user monitoring. Between 1977 and 1998, the US government effectively lost track of nearly 30,000 rifles, or chose to stop caring where they were pointed. The grant aid program operated on a trust system that the Indonesian military violated with clinical efficiency. When the violation became public, the US intelligence community responded not with an inquiry into the chain of command, but by summarizing the news for its superiors.

The desk's reading is that this record represents a calculated decision to treat a massacre as a media problem. By relying on Kompas as the primary source, the DIA ensured that the US government remained a spectator to the forensic evidence rather than a participant in the investigation. The still-withheld pages of this era likely protect the names of the officials who decided that 17 dead civilians were an acceptable cost for maintaining a strategic partnership with KOSTRAD.

This is how the machinery of state aid works: the weapons are delivered as grants, the monitoring is forgotten, and when the bodies pile up, the intelligence reports are written as summaries of the local news. The people of Semanggi paid for this administrative convenience with their lives, and the record shows the US government watched it happen through the lens of a newspaper.