How remote viewing was supposed to work
Stargate was an umbrella name for a U.S. government effort to test whether people could describe a distant place or object without ordinary sensory access. In a typical remote-viewing session, a viewer received a target identifier while the target itself was kept hidden. The viewer recorded impressions—shapes, textures, or rough sketches—before anyone compared them with the target.
That sequence mattered: recording impressions in advance was meant to reduce the chance that a viewer would simply adapt a story to information revealed afterward. The hope was that a signal, even an imperfect one, might help intelligence officers learn about a site they could not readily observe by conventional means.
The gap between a vivid impression and useful intelligence
A description can feel striking when matched to a place after the fact. But intelligence work needs more than resemblance: it needs details specific enough to distinguish one target from another, and reliable enough to guide a decision. Broad impressions such as “a large structure” or “something metallic” can fit many locations. If analysts must choose which details count as hits, a seemingly impressive match may not tell them what they did not already know.
Remote viewing was also not the same as proving astral projection. The program tested claims about obtaining information at a distance; it did not establish that a person’s consciousness literally traveled outside the body. The practical test was whether the reports could provide dependable, actionable information.
Why the program ended
By 1995, independent reviewers assessed the work and found that its results did not translate into intelligence that agencies could use reliably. Even when some descriptions appeared to correspond with a target, their vagueness made them difficult to act on. That distinction—between an intriguing possible match and a dependable lead—was decisive: a method that cannot narrow choices or support a concrete next step has little value to an intelligence operation.
