Project Stargate: The CIA's 20-Year Psychic Spy Program
From 1978 to 1995, the U.S. government funded research into remote viewing — psychic intelligence gathering. The results were inconsistent but not zero, which is why it ran for nearly two decades with military funding.
What It Was
Project Stargate is the umbrella name for a series of U.S. government programs that investigated remote viewing — the alleged ability to perceive distant or hidden targets using only the mind — for intelligence-gathering purposes.[1]
The program operated under various names from 1978 to 1995:
- SCANATE (1972–1973, preliminary work)
- GONDOLA WISH (1977–1978)
- GRILL FLAME (1978–1983)
- CENTER LANE (1983–1985)
- SUN STREAK (1985–1990)
- STAR GATE (1990–1995)
Total funding over the program’s life was approximately $20 million. The primary research was conducted at Stanford Research Institute (SRI) in Menlo Park, California, later at Science Applications International Corporation (SAIC).
The program was sponsored at various times by the CIA, Defense Intelligence Agency (DIA), Army Intelligence, and other DoD entities.
This is not fringe speculation. The CIA declassified the files. They are publicly available.
How It Worked
Remote viewing, as practised in Stargate, followed a structured protocol:
- A target — a geographic location, facility, or object — was selected by a handler. The viewer had no conventional access to information about the target.
- The viewer would enter a relaxed state and attempt to perceive and describe the target using impressions, sketches, and verbal descriptions.
- Descriptions were recorded, then compared against the actual target by independent judges.
The protocol was developed primarily by physicists Harold Puthoff and Russell Targ at SRI, and later refined into a methodology called Coordinate Remote Viewing (CRV) by Army officer Ingo Swann and others.[2]
Key personnel included:
- Ingo Swann — artist and claimed psychic, developed the CRV methodology
- Pat Price — claimed psychic who produced several notable results
- Joe McMoneagle — Army Chief Warrant Officer, the program’s longest-serving remote viewer (designated “Viewer 001”)
The Results
What the Data Showed
The most comprehensive evaluation of Stargate was commissioned by the CIA in 1995, conducted by the American Institutes for Research (AIR), with statistical analysis by Professor Jessica Utts (UC Davis) and sceptical review by Professor Ray Hyman (University of Oregon).[3]
Utts concluded:
“Using the standards applied to any other area of science, it is concluded that psychic functioning has been well established. The statistical results of the studies examined are far beyond what is expected by chance.”
She found a statistically significant effect across the laboratory experiments, with effect sizes that could not be attributed to methodological flaws.
Hyman concluded:
“I agree with Jessica Utts that the statistical departures from chance appear to be too large and consistent to be dismissed as statistical flukes.”
However, Hyman argued that the statistical significance did not necessarily indicate a psychic mechanism — potential methodological issues (sensory leakage, judging biases) could not be entirely ruled out.
Operational Use
The program was used in actual intelligence operations — not just laboratory experiments:
- Iranian hostage crisis (1979–1981) — remote viewers were tasked with locating the hostages. Some descriptions reportedly matched the actual location.
- General Dozier kidnapping (1981) — viewer Joe McMoneagle provided information during the search for the kidnapped NATO officer.[4]
- Various Soviet military facilities — viewers were asked to describe installations that satellite imagery could not fully resolve.
The operational results were mixed. Some descriptions were strikingly accurate. Others were vague or wrong. The signal-to-noise ratio was poor enough that the intelligence community could not rely on remote viewing as a primary source — but good enough that some analysts continued to find it useful as a supplementary tool.
Why It Ran for 20 Years
This is the key point. If the results were zero, the program would have been cancelled quickly. Defence budgets are scrutinised. A program with no results does not survive nearly two decades under multiple sponsoring agencies.
The answer appears to be that the results were inconsistent but not zero — enough above chance to keep various program managers convinced, but not reliable or controllable enough to scale into a standard intelligence tool.
Why It Was Shut Down
In 1995, the CIA commissioned the AIR evaluation specifically to determine whether to continue the program. Despite Utts’s statistical findings, the recommendation was to terminate:
- The intelligence value was judged insufficient — remote viewing could not be directed with enough precision to produce actionable intelligence on demand
- The mechanism was unknown, making it impossible to improve or standardise
- Operational reliability was too low for intelligence purposes
The program was declassified and shut down. The files — approximately 89,000 pages — were released to the public.
The Sceptical Position
Sceptics make several strong points:
Methodological concerns. Early SRI experiments had weak controls — possibilities for sensory leakage (unconscious cues from handlers), feedback loops, and subjective judging. Later experiments improved controls significantly, but sceptics argue the stronger protocols produced weaker results.
Publication bias. Positive results may have been selectively reported while failures went unrecorded. This is a legitimate concern in any research program, but the declassified files include both hits and misses.
The file-drawer problem. Across thousands of sessions, some impressive hits are expected by chance alone. Without a rigorous accounting of all sessions (not just the notable ones), the overall success rate is hard to establish.
No mechanism. Physics provides no known mechanism for perceiving distant locations without sensory input. Without a mechanism, the statistical results — however significant — may point to unidentified methodological flaws rather than a real phenomenon.
What the Files Actually Show
The declassified Stargate archive is publicly accessible through the CIA’s Electronic Reading Room. Anyone can read the session transcripts, target feedback, and internal evaluations.[1]
What the files show is neither the dramatic confirmation that believers claim nor the total failure that sceptics assert:
- Some sessions produced descriptions that were strikingly accurate — geographic features, building layouts, and target characteristics that matched the actual targets
- Many sessions produced vague, ambiguous, or incorrect descriptions
- The statistical aggregate across controlled experiments shows results above chance, with effect sizes that multiple independent statisticians have confirmed
- The operational utility was never sufficient to replace conventional intelligence methods
The Uncomfortable Middle
Stargate occupies a genuinely uncomfortable position:
| Strong Evidence For | Strong Evidence Against |
|---|---|
| 20 years of funding across multiple agencies | No physical mechanism identified |
| Statistically significant laboratory results | Inconsistent operational reliability |
| Independent statistician confirmation | Possible methodological artefacts |
| Declassified operational use in real crises | Results never sufficient for primary intelligence |
| Above-chance hit rates across thousands of sessions | Cannot be produced on demand |
The honest summary: the U.S. government spent $20 million over two decades studying psychic phenomena, found results that were statistically significant but operationally unreliable, and shut the program down not because the results were zero but because they were not useful enough.
Research Verdict
| Assessment | Documented — Declassified government program with ambiguous results |
| Confidence | High (on program existence), Low-Moderate (on psi reality) |
| Summary | Project Stargate is fully documented and declassified. It ran from 1978 to 1995 with military and intelligence funding. The statistical results were above chance by independent assessment. The mechanism is unknown, the operational reliability was poor, and the scientific establishment remains divided on interpretation. The program's existence and results are not in dispute; only their meaning is. |
Sources
- CIA Electronic Reading Room — Stargate Collection (89,000+ pages).
- Puthoff, H.E. (1996). “CIA-initiated remote viewing program at Stanford Research Institute.” Journal of Scientific Exploration, 10(1), 63–76.
- Utts, J. (1996). “An assessment of the evidence for psychic functioning.” Journal of Scientific Exploration, 10(1), 3–30.
- CIA declassified operational files — various Stargate tasking documents.