Project Megiddo
Federal Bureau of Investigation, Domestic Terrorism/Counterterrorism Planning Section · Released 20 October 1999; this public PDF produced 31 January 2000
A report everybody characterises and almost nobody quotes. It is short — 32 pages — and the FBI removed it from its own website years ago, so the versions in general circulation are third-party retypings on advocacy and religious-studies sites. This is the Bureau’s own PDF, preserved by GPO. Read directly, its executive summary contains a concession that neither its defenders nor its critics tend to mention.
Open the PDF — 32 pages, 0.1 MB
Hosted here in full. Nothing has been re-compressed, re-OCR’d or altered.
What is on which page
p.4 (Executive Summary)
The concession, in the FBI’s own summary. “The Project Megiddo intelligence initiative has identified very few indications of specific threats to domestic security. Given the present nature of domestic extremism, this is to be expected. However, this is a function of the limitations of the group-oriented model of counterterrorism analysis and should not be taken necessarily as reflective of a minor or trivial domestic threat.” The report says plainly that it found very few specific threats, and then argues that this absence is not evidence of absence.
p.4
The actual prediction. “In the final analysis, while making specific predictions is extremely difficult, acts of violence in commemoration of the millennium are just as likely to occur as not.” A coin flip, stated as such. Retrospective accounts on both sides tend to describe the report as more confident than this sentence is.
p.4
Why the report is structured around movements rather than plots. “[T]he overwhelming majority of extremist groups in the United States have adopted a fragmented, leaderless structure where individuals or small groups act with autonomy… the worst act of domestic terrorism in United States history was perpetrated by merely two individuals: Timothy McVeigh and Terry Nichols.” The FBI names the limitation of its own analytical model on the same page it presents the model’s results.
p.2 (Table of Contents)
The taxonomy that drew the criticism. The report’s substantive sections are Christian Identity (p.14), White Supremacy (p.18), Militias (p.21), Black Hebrew Israelites (p.23), Apocalyptic Cults (p.26) and The Significance of Jerusalem (p.30). Four of the six organise around a belief system rather than around conduct — the structural fact underneath the religious-liberty objection.
p.23
The most-criticised section, as written. “As the millennium approaches, radical fringe members of the Black Hebrew Israelite (BHI) movement may pose a challenge for law enforcement… current intelligence from a variety of sources indicates that extreme factions of BHI groups are preparing for a race war to close the millennium.” The section repeatedly qualifies its subject as “radical fringe,” “extreme factions” and “violent BHI followers” — qualifiers that did not survive into much of the coverage, in either direction.
p.3
Where the name comes from. “For over four thousand years, MEGIDDO, a hill in northern Israel, has been the site of many battles… The Hebrew word ‘Armageddon’ means ‘hill of Megiddo.’” (The etymology is loose — “Armageddon” is Greek, transliterating a Hebrew phrase — but this is the report’s own account of its title.)
p.32 (Conclusion)
What the report actually asks law enforcement to do. Its closing recommendations are three watch items, all officer-safety framed: “1) plans to initiate conflict with law enforcement; 2) the potential increase in the number of extremists willing to become martyrs; and 3) the potential for a quicker escalation of conflict during routine law enforcement activities (e.g. traffic stops, issuance of warrants, etc.).” It ends with no watchlist, no named targets and no recommended action against any group.
Note. This is the revised public version. The PDF’s embedded metadata records that it was produced by Acrobat PDFWriter 3.02 from a WordPerfect file named public megiddo revised.wpd — sitting on drive A:\, a floppy disk — on 31 January 2000 — three months after the 20 October 1999 release. The filename indicates a version prepared for public distribution and subsequently revised; what changed between the October release and this January file is not stated in the document, and we have not found a copy of the original to compare against. We flag this rather than pass the file off as the October text.
The FBI no longer hosts the report. The copy here comes from GPO’s Permanent Access programme, which archived the original fbi.gov/library/megiddo/ page — which is why the URL preserves that path inside it.
Provenance and licence
Released by: Federal Bureau of Investigation, Domestic Terrorism/Counterterrorism Planning Section.
Original copy: U.S. Government Publishing Office, Permanent Access archive (preserving the original fbi.gov/library/megiddo/ page). We did not obtain this document; we are hosting a copy so that the citation in the case file resolves to something durable.
Licence: Public domain — work of the U.S. federal government.
How this copy was checked: Downloaded from GPO Permanent Access; 112,950 bytes, SHA-256 914998893df5921a6fd0abb8b489c1aca7eac7330787bf80df6959d1fcfa96ae, byte-identical to the hosted copy. Opened and confirmed 32 pages, all 32 carrying a text layer. Every passage quoted above was located by page in that text layer. The cover note on p.1 reads “The attached analysis, entitled PROJECT MEGIDDO, is an FBI strategic assessment of the potential for domestic terrorism in the United States undertaken in anticipation of or response to the arrival of the new millennium.”
32 pages · 0.1 MB · PDF