The atlas is a collective dream book. It contains 29,345 dream reports gathered over sixty years by researchers at the University of California, the oldest of them from the 1940s, the newest still arriving.
It begins here with 106, and those of a lone Swiss dreamer.
The atlas grows with every dream recorded.
Begin recording →—
What each decade could not feel, it dreamed.
The corpus comes from DreamBank, an open research archive, and from a second body of work begun by Calvin Hall and Robert Van de Castle, who built the first systematic coding system for dream content in 1966. Their method lets a computer ask questions about a dream that a person could only ask across thousands: how often does water appear in a woman's dream in her twenties. How the frequency of chase dreams shifts after forty. Whether the dead speak more in winter.
On top of the original Hall/Van de Castle codes we layer a second reading, drawn from Carl Jung and Marie-Louise von Franz, archetypal images, compensatory movements, shadow material, anima and animus figures. This second layer is interpretive; we never present it as fact. Where a Jungian reading is offered, it is offered as one lens among many.
When you write a dream in Reveries and a similar image appears in the atlas, the tidal wave you dreamed last night, dreamt also by a stranger in Cincinnati in 1962, the atlas will tell you. Not to diagnose you. To place you in company.
The dataset and the coding methodology are open source. Both the raw corpus and the enrichment pipeline we write on top of it will remain publicly downloadable for as long as Reveries exists. Research built on dreams should not belong to one company.
