In 2008, Monica Gagliano is a postdoctoral researcher in climate change ecology at James Cook University, and she spends her days in the water tracking the breeding of wild pairs of Pomacentrus amboinensis, a reef damselfish. After a week, the animals come to nestle in her open hand, her fingers closing gently around their scaly bodies then opening again. “They knew me personally. I knew them, one by one.” On the last day, she goes into the water to say goodbye to them, before coming back in the afternoon to capture and kill them all, as the protocol requires. That morning, no one comes. “In that moment, I knew they knew.” Then comes the hardest sentence in Thus Spoke the Plant: “Frozen and not knowing what to do, I did what I knew. That afternoon, I went back in the water with nets and catch bags and killed them all.” And the consequence: “And I never killed again.“
That is where the book begins, and you need to know it to understand everything that follows. A researcher decides that no scientific question is worth killing a living being, finds herself practically unable to continue her profession, and turns to plants. Her line on the world she is leaving is biting: “in the temple of modern science, a blood sacrifice to the old gods of the Enlightenment is still, for the most part, the required ceremonial procedure.“
A book whose structure is already the message
Eight chapters, designated by a letter and not by a number. Placed end to end, these letters spell ORYNGHAM. Gagliano explains it right in the first chapter, oryngham means “thank you for listening to me” in the language of plants, a word received during a vision under ayahuasca. The book is therefore built as the thank you it relates. I do not know whether the French edition keeps this acrostic, and the answer changes the reading of the whole volume published by Marabout. Foreword by Suzanne Simard, illustrations by Patrizia Gajaschi. Each chapter opens with an unattributed epigraph in poetic prose, which is a word spoken by a plant, and with a short maxim, “True leadership gives space“, “Appreciation and gratitude create beauty.” A delicious and cruel detail, the Library of Congress filed the work under the categories “Human-plant relationships, Anecdotes”. Yet it is the endnotes that close the volume, chapter by chapter, with the full references of the articles.
The mimosa that remembers for twenty-eight days
University of Florence, collaboration grant obtained in 2011. The Mimosa pudica, three inches tall, arrive at the end of September. Laboratory note quoted verbatim: “September 27. All plants are looking healthy and happy.” The climate chamber is divided into three compartments by black plastic curtains, humidity and temperature identical everywhere, two side compartments set up in low light (LL) and high light (HL), the central one serving as the main stage. The word is Gagliano’s, who sustains the theatrical metaphor from beginning to end, with “to close, or not to close, that was the question“, and who describes herself as stage manager, lighting designer and voice of the mimosa.
The setup is a graduated vertical steel rail mounted on a foam base, with a sliding holder that drops fifteen centimetres, fast enough to frighten the plant, not enough to damage it. One drop every five seconds, sixty drops per session.
Act 1, the morning of 3 October 2011. Sixteen naive plants, a single drop in the morning, a single one eight hours later. All of them fold their leaves completely, both times. Act 2, fifty-six new plants, six sessions in the day, that is three hundred and sixty drops each. Four to six drops are enough for them to reopen their leaves, and they keep them open while Gagliano goes on dropping them. The counter-intuitive result, the one that gives the article its title, is that the low light plants learn faster and reopen more completely. When the resource is scarce, a mistake is costly, so you learn fast.
Act 3, the obvious objection. Has the plant learned, or is it simply exhausted? Each plant goes onto a laboratory vibrating plate, shaken gently for five seconds. Mechanical disturbance of the same nature, but absolutely new. Without exception, all the leaves fold. Put back into the drop system, they stay wide open. Act 4, memory. After three days of quiet, twenty-eight plants come back, and “nothing happened. That was the most perfect and exciting “nothing” that I could have hoped for.” Three days later, the second half, at six days, remembers too. Then Gagliano swaps the environments, LL to HL and the other way round, and retests after twenty-eight days. The leaves stay open.
Twenty-eight days of memory, in an organism with no brain, no neurons and no synapses. The article appears in January 2014 in Oecologia, co-signed with Michael Renton, Martial Depczynski and Stefano Mancuso. Before that, the manuscript was rejected by more than a dozen major journals, several without even sending it out for review, sometimes without having read it.
The pea that disobeys its own instinct
The Pavlovian protocol, Gagliano says she received it by dictation. In an Amazonian dieta with the Ayahuma, the cannonball tree, under the guidance of a shaman called Don J, dry fasting, isolation under a mosquito net, no washing for a week. “she made sure I could hear her voice loud and clear when she started prescribing the complete set of instructions for testing Pavlovian learning in plants. I transcribed as she dictated.” The diagrams fill the travel notebook, and it is that notebook that comes back to Australia.
At the University of Western Australia, a room kept in complete darkness, switches taped over, builder’s plastic over the entrance, a sign on the door: “Plant learning in progress. Do not enter.” To her colleagues, she answers “I am trying to teach pea seedlings some tricks“. Y-mazes made of PVC pipe, one pea per maze, a fan as the conditioned stimulus, blue LEDs as the reward. Three days of training, three daily sessions of two hours spaced an hour apart, the fan running for ninety minutes and preceding the light by sixty, with thirty minutes of overlap. Two groups, fan and light on the same side or on opposite sides, plus a control. On the fourth day, fan alone.
For two weeks, she watches, and reads 50:50. She concludes it is a failure. One sunny Sunday morning, she comes in to dismantle everything, switches on the LEDs that glow like fireflies, looks at one maze, and blurts out: “Holy shit, they are doing it after all!” What she had missed is the most rigorous passage in the book. The 50:50 was not the right baseline for chance, because an untrained pea grows towards the last perceived light one hundred per cent of the time. What had to be measured was not a coin toss, it was the capacity to contradict a reflex. Between sixty and seventy per cent of the peas had made that rebellious choice. Scientific Reports, 2016, with Vladyslav Vyazovskiy and Alexander Borbély, two sleep specialists, an unusual methodological backing in botany.
Her conclusion holds well beyond peas: “as I tested their learning ability within the Pavlovian conditioning paradigm, the seedlings had tested mine“. And the last sentence of the chapter, “We don’t need to find our way out of the maze. We realize there never was a maze.“
A field investigation into her own artefacts
Chapter H tells the story of roots locating water by sound, and its real worth lies in its failures. First setup, a sealed pipe where water circulates without being accessible, against a real moisture gradient. The peas find the source equally well in both cases, and when both are offered, they go without hesitation to the real water. Second setup, playback, with a vibrating speaker and an MP3 player attached to one leg of the maze. Disconcerting result, most of the seedlings flee the equipped side, even when silence is played. Gagliano suspects the magnets, writes to the BioMagnetics group of her school of physics on 2 April, and receives this: “Hi Monica, to be honest, if I got your email a day earlier I would have just thought it was a April Fools joke.” The colleague promises a measurement, never takes it, she waits four months then buys a gaussmeter herself. Verdict, the equipment emits a magnetic field measurable within a radius of five centimetres, which corresponds almost exactly to the dimensions of the black pot. Insignificant at our scale, decisive at the scale of roots. Third setup, a speaker on each side to standardise the artefact, an image of Ulysses between Scylla and Charybdis. The sound of water clearly wins over silence, less clearly over white noise.
The nested boxes of chapter R, her “matryoshka boxes“, are chronologically the starting point. Chilli, fennel, basil, every known channel blocked one by one. Young chilli plants speed up their germination when fennel is present but its signals are cut off, and the beneficial effect of basil persists even with signals blocked. “with the guidance of the plants, I had uncovered a previously undocumented communication channel.” And she adds, to her credit, that she has no mechanistic explanation for it.
The ideas, one of which targets my own work
The central argument is epistemological. The standard objection to her results always takes the form “plants have neither brain nor synapses, so how could they remember?”. But since long-term memory is defined by synaptic connectivity, the question rules out its own answer in advance. “it is the formulation of the question itself that has been failing us all along.” An ecologist by training, Gagliano prefers to ask why a thing happens, and willingly leaves the mechanisms to physiologists. She proposes to see memory not as a fixed trait of the organism but as a property of the relationship, and the voice as something that exists in the space between self and other.
Then comes the ethical part, subtler than an ordinary environmentalist plea. She distinguishes ownership, which crushes the subjectivity of plants in order to abuse them, guardianship, which appears benevolent but “unwittingly infantilises them by treating them with a kindness that betrays our sense of superiority”, and the non-hierarchical co-participation she argues for. And there, a sentence that hit me directly, about those who “overwrite plants with the sounds that are familiar to us by ventriloquizing them through the approximation of human speech or the transposition of musical scales and various instruments onto them“. Gagliano places the sonification of the plant world on the side of condescending guardianship. It is an objection I do not share, and I give it in full because it is serious.
The two registers, and the gap she leaves open
The book presents itself as a phytobiography, written with plants and in their name, and Gagliano calls her style plant-writing. She explicitly refuses ventriloquism, the human is neither interpreter nor spokesperson, but “a listener who filters out personal noise to hear plants speak” and a co-author who carries the conversation onto the page.
The shamanic register is as detailed as the scientific one, and presented at the same level of reality. The ceremony in the maloka with Don M and his double dose. The ritual bath in the pulp of the Ayahuma, facing east then west. Osha, the bear spirit met at the drum, who buries her like a seed, and the twist when her friend Sa tells her afterwards that osha root, Ligusticum porteri, is called bear root. The Lakota vision quest in California, guided by an elder named Mato Ta Pejuta Wakan Najin, where the trees and the grasses cry out to her in unison “Tell our stories!“
The chain always runs in the same direction. An instruction received in a non-ordinary state, transcribed in a notebook, carried out to the letter in the laboratory with witnesses and statistics, published in a peer-reviewed journal. The Oecologia and Scientific Reports articles obviously make no mention of the tree. The book is the place where she tells the genesis that the publication cannot contain. And she does not close the gap, she frames the crudest objection against herself, “Did I seriously believe that a tree in the Amazon (…) had given me instructions“, and does not answer it. She notes that the protocol worked.
The book is not balanced, and it is better to say so. Chapters O, Y and A are narrative and spiritual, N, G and H are experimental, R is the hinge. You can read only N, G, H and the second half of R, you will have all the protocols, and you will have lost the reason why they exist.
What the controversy teaches, and what it does not
In 2020, Kasey Markel published in eLife a replication of the pea conditioning that does not reproduce the effect. Gagliano and her co-authors answered that the setup differed, in particular that the unconditioned stimulus used did not systematically trigger a response. Markel replied that these differences do not explain the failure. The whole exchange is open access, it should be read in full rather than reduced to one camp, and that is exactly what a science that works looks like. What is instructive is that this is not where the book is decided. The real complaint against her is the second half, judged disqualifying before any examination. Simard says it in her foreword: “We owe her a debt of gratitude for swimming against the tide.“
I take away three things. The discipline of admission, that way of laying out the statistical reference error and the speaker artefact instead of keeping quiet about them. The idea that our expectations literally limit what we are capable of seeing, verified by her on herself over two weeks. And that formula, learning to wait without waiting, which describes rather well what you do when you plug in a plant and listen.
Read in June 2021. Note written in 2026.
