New York, Times Square, an eighteen-storey building on the corner of 46th Street and 7th Avenue. Fourth floor, the laboratory of a lie detection school. It is seven in the morning on 2 February 1966, a man has worked all night, his doberman Pete is asleep at the back of the room. On a table, two potted plants his secretary bought for about two dollars each at the ground floor shop that was clearing its stock, a rubber plant and a dracaena. Cleve Backster has never owned a plant in his life. On his coffee break he waters them, and out of curiosity he wonders whether he could measure how fast the water rises from the root to the leaf.
He takes the GSR module of a polygraph, clamps a leaf between the two skin electrodes, and holds the whole thing together with a rubber band. The leaf’s resistance happens to fall within the range of the device, around 250,000 ohms. Instead of the drop expected as the water arrives, the trace rises. At about a minute, a brief change of contour looks like something he knows by heart in humans, the pattern of someone who briefly fears being found out. So he applies to the plant the rules of his trade. He dips the tip of a neighbouring leaf into his cup of scalding coffee, nothing. Then, at fourteen minutes, another idea comes to him, the ultimate threat, to take a match and burn the leaf that is wired up. At exactly 13 minutes 55 seconds, the trace runs off the top of the chart. He is four and a half metres from the plant, the device one and a half metres away. Not a word spoken, no contact, no match struck. Nothing but the intention.
That is the first chapter. The whole book lies in what this man made of the thirty-six years that followed.
What is in it
Primary Perception came out in 2003 from White Rose Millennium Press, 168 illustrated pages, written by a man of seventy-eight who says himself that he hates writing, and who took four and a half years to finish the volume. The French translation, the one I read, is called L’intelligence émotionnelle des plantes, published by Guy Trédaniel, and runs to 265 pages. A good part of the difference comes from the editorial apparatus added by the French publisher.
Nine chronological chapters, two appendices, an index. There is one thing to know before opening it: chapters 4 and 8, more than seventy pages between them, are not about the research but about its reception. Conferences, visits, press articles, proper names in bursts. That is almost half the book. The experimental core is chapters 1, 2, 3, 5, 6, 7 and 9, and it reads in a few hours.
The chain of accidents
Backster’s method fits in one sentence from his second chapter: “If you watch the phenomenon too closely, you may already be affecting it.” Everything else follows from it, including the shape of the book.
Almost every discovery comes by domestic accident, and he does not hide it, he makes it his way of working. He empties the scalding remains of a kettle into the sink, the plant being monitored jumps the instant the water hits the drain. Months later he takes a sample from the bottom of the trap and looks at it under the microscope, a jungle of life he compares to the cantina in Star Wars. The laboratory adjoins the toilets on that floor, and during demonstrations for visitors the trace runs to the stop at the exact moment of each flush. He breaks an egg for the dog’s bowl, and the galvanometer needle goes wild while he is monitoring a philodendron five metres away. He stirs the strawberry jam into a yoghurt, the tone generator goes off in the next room. He did not know there were living bacteria in a yoghurt, he buys a manual of dairy bacteriology to find out.
Another observation from this period, rarely quoted and yet a fine one: for almost any harm done to a non-human life form, the plant adapts after three or four repetitions and stops reacting, not only on the day itself but on the days that follow. Backster sees a memory in this. The exception is the destruction of human cells, which always keeps the same intensity.
The brine shrimp protocol, which is the real subject of the book
The experiment published in 1968 in the International Journal of Parapsychology deserves to be described piece by piece, because it is not an experiment on plants, it is an experiment on the experimenter.
Pairs of brine shrimp in the act of mating, a guarantee that at least one of the two is in good health. A mechanical randomiser that picks one moment out of six twenty-five-second slots, without anyone knowing. A delay switch armed at the start, with a minimum of ten minutes’ delay. A tipping device that flips the dish into the boiling water in one go, and Backster explains why the flip has to be clean, a single drop falling from a dish left tilted could contain a shrimp and trigger a reaction outside the slot. A thermostat-controlled heating element, a stirring paddle, a thermometer. A control polygraph wired to a fixed resistor, whose trace must stay a straight line. A home-made programmer, twenty notched aluminium discs, four pens, eight channels, with switches bought in a model train shop.
And above all, this. The plants are bought by a third party and left in a room where neither Backster nor his associate Bob Henson touches them. They fetch them just beforehand, arm everything on their way out, leave the building and move at least a whole avenue away. The same plants are never used twice, because of habituation. That is Backster’s invention, far more than the wired plant: taking the observer’s consciousness out of the setup. Later, in San Diego, he goes as far as putting an opaque screen in front of the recorder during sessions, and wiring push buttons in every room to time-stamp from a distance without ever looking at the trace.
The egg, and the cells
Two results stand out. On an unfertilised egg, fitted with two sponges boiled in salt water and held by a rubber band, he records small cyclical serrations, about 157 cycles a minute, which matches the documented heart rate of a chick embryo at three days of incubation. The biologist Charles Granger witnesses the phenomenon, takes the egg to Cornell, and confirms that it was not fertilised. Backster plainly names the interpretation he draws from it, the etheric field, and he presents it for what it is, an interpretation.
On 30 June 1980, Steve White takes a sample of his own oral leucocytes following a protocol that a dental researcher, James Klinkhammer, had shown Backster in Houston. The cells are in a screened cage fifteen feet away, two cameras are filming, one over White’s shoulder, the other the trace. White leafs through a Playboy to find an interview with William Shockley, comes upon the centrefold, and his cells run to the top and bottom stops for two minutes, even as he states out loud that he does not find the model to his taste. Magazine closed, they settle. A minute later, when he reaches out to open it again, they take off once more.
What has been said against it, and what that teaches us
It has to be said, and Backster says it himself, in a note to his chapter 4: the experiment has not been replicated. On 8 August 1975, Horowitz, Lewis and Gasteiger publish in Science a repeat with a Faraday cage and two statistical methods, with no result at all. In 1978, John Kmetz publishes a refutation carried out in consultation with Backster himself, with 63 plants against 7, and 378 shrimp drops against 21. P = 0.70 and P = 0.35, nothing.
The interesting point in Kmetz is not the verdict, it is the mechanics: he finds that the electrode setup produces GSR fluctuations even when there is no leaf at all. That says something precise to anyone building an instrument today, you need a control channel with no plant. Here is a criticism that teaches something, and it is worth reading.
Backster, for his part, always gives the same answer, those who fail have not automated the experimenter’s consciousness, they believed a wall and closed-circuit television were enough. He accepts the logical consequence all the way, and writes it in black and white: “Extraordinary proof that requires repeatability rules is incompatible with the spontaneous nature of this phenomenon.” His lighter phrase for it has stuck: “Mother Nature doesn’t appear to want to jump through a hoop ten times in a row merely because someone wants her to.”
The language, and the place of the book
The tone surprises anyone expecting a crank. Backster is modest and often funny, he wonders whether his dracaena is developing an ego problem from being shown at conferences, he notes that his brine shrimp were already carrying heavy karma, and he tells how after 2 February he had no wish to go down into Times Square shouting that his plant had just read his mind. The precision is real on the setups, the ohms, the millilitres, the material of the electrodes, the gauge of the tubing. It is weak on the results, and that is the book’s real failing: one reads “a statistically significant number of occasions” with never an n, never a p, never a table. The illustrations, on the other hand, are worth a look, they are for the most part traces annotated in felt tip in his own hand, with date, chart speed and gain.
He explicitly sets himself apart from Bose, and the distinction is a fair one. Bose measures what happens to the plant. Backster claims to measure, in the plant, what happens elsewhere. It is not the same object of study. His last chapter, finally, is a specification, a portable EEG-type device, battery-powered, cheap, to be put in the hands of every curious person, funded in part by his royalties. With a single rule of use, always the same since 1966, never look at the needle during the session, record, and correlate only afterwards.
What I have kept from it for my own work fits in one line, and it is not primary perception. It is the discipline of the setup: the one who listens is inside the circuit, and you have to build the instrument knowing it.
Read in March 2016. Note written in 2026.
