Plants Can Learn: It’s a No-Brainer!

Mimosa pudica. Photograph by M. Martin Vicente.

by Monica Gagliano and Michael Marder

MARCH 18, 2014


When we consider the propensity to learn, plants are probably the last living beings that come to mind. They seem to be so passive, immobile, and completely determined by their attachment to the place, in which they are rooted, that any of the so-called “cognitive abilities” appear to be inapplicable to them. And this is not even to mention the holy grail of cognition—the central nervous system with extensive neural networks—that plants do not have.

At the same time, the suspicion that plants are capable of learning is not entirely new. While Sir J. C. Bose (1858-1937) proposed the idea just over a century ago, it has truly entered the arena of scientific enquiry only in recent years. The reason for the delay is obvious: one widespread assumption in the twentieth and even the twenty-first centuries has been that learning is reliant either on neuronal processes or, in the case of machines, on artificial neural networks modeled on their biological counterparts. With this assumption in the background, conducting experiments on or contemplating plant learning, memory, and decision-making have been interpreted as acts of sheer madness. But, in all fairness, although we have pondered human and animal learning since antiquity, we are still asking the basic questions about how learning really works because finding the answers is not an easy task even in these kinds of organisms.

So, how would we ask those same questions in plants? Do plants have motivations and preferences? Do they have expectations and, if so, how can we assess them? Do they make choices and if so, what are the implications for how we regard them? The creative challenge here is to develop suitable experimental and theoretical approaches that enable plants to show us what and how they learn, while avoiding the temptation to turn human or animal abilities into the standard template for these investigations.

If we consider the issue from the ground up, we realize that everything we know about the world, we learn through repetition. Learning from the things that we experience again and again is in fact one of the most effective ways of acquiring new behaviors, or of adjusting and refining old ones, in order to survive and flourish in a range of ecological and social settings. Why would plants be excluded from such a useful evolutionary process? Wouldn’t being barred from learning harm their chances for survival? Wouldn’t it be utterly wasteful, not to say dangerous, for them to register stimuli from their environments each time anew, without the accumulated memory of past experience that would enable them to respond appropriately in the future?

Like humans and many animal species, the sensitive plant Mimosa pudica, for example, also learns through repeated practice. Usually, when subjected to a disturbance for the first time, it instinctively closes its leaves, a mechanism designed to defend it against predators. But, as recent experiments have shown, Mimosa quickly learns that to continue closing its leaves when a repeated disturbance proves to have no negative consequences is a waste of energy. By considering the experience no longer threatening, this plant stops closing its leaves, a behavioral change motivated by the energetic reward that keeping its leaves open brings. After all, the opportunity for ‘feeding on sunlight’ drops considerably every time Mimosa keeps its leaves closed, which is something that can cost it dearly. That is where the learning processes steps in, helping the plant optimize its behavior.

Interestingly, the extent to which Mimosa is willing to keep its leaves open despite the disturbance depends on the environmental context. Experimental data clearly show that this plant does not simply react to the immediate stimuli available from the environment. Instead, it assesses a given situation and preferentially engages in behaviors that pay off from its perspective. The tendency of individual Mimosa plants to take the specific action of keeping the leaves open in response to a known disturbance is greater in limited light environments where the consequences of leaf closure can be dire. And, remarkably, the strength of Mimosa’s motivation to keep the leaves open does not loosen, as one may expect, when light conditions improve. These plants do not ‘relax’ when moved from a light-limited environment to one where light is abundant. They remain highly responsive and in a state of alertness, as if anticipating that the environment is likely to deteriorate again.

At every moment, plants, like humans and other animals, perceive a variety of things simultaneously, but they learn to focus their attention on whatever they need to perceive and exhibit a behavioral tendency to approach or avoid situations. In this process, repetition is the learning platform upon which motivation galvanizes and steers behavior toward specific goals and expected rewards. It is not by chance that, on the basis of plant learning, we comfortably use words such as motivation, tendencies, preferences, expectations and choices, which in everyday speech we often equate to feelings, desires, and conscious tendencies. As we learn about plant learning, the questions may not be about whether plants have motivations and preferences, expectations or choices, but what it means for us to know that they do.

In (Philosophical) Defense of Trees

Photograph by Michael Marder.

 By Michael Marder

MARCH 4, 2015


With deforestation claiming seventeen percent of the Amazonian forest in the last fifty years alone, who would have thought that trees would need to be defended from… a philosophical onslaught? And yet, the attack came from the most unlikely of corners, launched as it was by trendy representatives of French post-structuralism.

The writings of Gilles Deleuze and Félix Guattari have given trees a bad reputation. To put it succinctly, for them, arborescent logic and imagery connote hierarchy, verticality, and the movement of transcendence, whereas rhizomatic assemblages betoken a certain kind of equality, horizontality and immanence. Rhizomes are modified horizontal subterranean stems that, much like seeds, can develop roots and shoots. Those of ginger, turmeric, and lotus are perhaps the most familiar (because edible) examples. Grass, too, fares better in the eyes of the authors of Anti-Oedipus than trees. It is, after all, quite democratic and egalitarian; growing close to ground-level, its blades do not entertain any dreams of penetrating the depths of the earth and of soaring toward the expanse of the sky with its branches. Inside us, humans, there is also grass, or at least the grass-like structuration of the brain: “Many people have a tree growing in their head, but the brain itself is much more a grass than a tree.”[1]

It is, at the same time, crucial for Deleuze and Guattari not to fall back into the vertical system of valuation, which they criticize, in associating the rhizome, or grass, with something unequivocally “good” and the tree with the “bad.” As they note in A Thousand Plateaus: “The important point is that the root-tree and canal-rhizome are not two opposed models: the first operates as a transcendent model and tracing, even if it engenders its own escapes; the second operates as an immanent process that overturns the model and outlines a map, even if it constitutes its own hierarchies […].”[2] There is a little bit of a rhizome in a tree, just as there is a modicum of a tree in the rhizome; immanence is shot through with the possibilities for transcendence; a transcendent model can devolve into immanence through its own escape routes; equality can reconstitute a hierarchy; the horizontal and the vertical dimensions are entangled with one another.

This disclaimer aside, Deleuze and Guattari got “arborescence” all wrong. The physical verticality of trees does not mean that they are vertical in the way they live or grow. Trees can branch out in quite unpredictable ways; they can accommodate the grafts of other species; they can give rise to shoots that would survive independently of them; they can change their sexes or become hermaphrodites for a part or for the rest of their lives; and the list goes on. To put it in Heideggerese, trees are ontically vertical and ontologically horizontal. Although they tower in measurable height over and above the grass, they are as egalitarian as the most humble of plants. In fact, given how some tree species share their root system, they can be thought of as overgrown grass. For instance, Pando, which is a grove of quaking aspens in Fishlake National Forest (Utah) has the largest root system in the world: over 106 acres, what sprouts above the ground are genetically identical trees. So, what would be the advantage of a rhizome over the roots found in Pando?

Strictly speaking, the most ontologically vertical notion in biology is that of an organism. In an organism, there is a rigid hierarchy between the different organs—some of them vitally important, others less so—and the predominance of the whole over the parts. Plants, in their turn, do not follow the organismic model of development, and trees are not an exception. Even if they might be, or appear to be, more individuated than other vegetal beings, such as grass, their parts (for instance, branches) are not really organs limited in terms of their number, position in a body, and so forth. It may well be that all plants, including trees, are the most faithful instantiations of “bodies without organs,” which Deleuze and Guattari so admired.

Clearly, the penchant for the rhizome over the root derives from an objection to the metaphysical obsession with depth, often associated with the hidden and radical nature of the underlying source of visible appearances. But the root of a tree is far from being its origin; as in all other plants, it is but one extreme in the polarization of a shoot, a seed, or an acorn that grows up and down simultaneously when planted in the ground. Whatever resembles the source in a plant is always a variation on, of, and from the middle—the extension of the middle in every direction, both vertically and laterally. “Rhizome” is a part of the vegetal world that symbolizes the whole: trees, too, are rhizomes, proliferating between roots and shoots.

I could add to this brief defense of trees the observation that they have been beneficial for a venerable tradition of “philosophies of immanence,” going back to Plotinus, the tradition, to which Deleuze and Guattari willy-nilly belong. For Plotinus, the universe is a “great tree,” on which all living and even inorganic entities, are branches, leaves, and buds. The tree is a fold—the same, varied with regard to itself. Deleuze himself suggests that Spinozan “attributes” can be understood through the example of “a seed which ‘expresses’ the tree as a whole.”[3] For Bergson, each tree is a society, rather than an individual; an articulation of multiplicities, rather than a living unit. The tree and the root are essentially superficial, regardless of the height and the depth they have come to represent. It is both philosophically and ethically irresponsible to turn them into villains, especially compared to grass that grows in many recently deforestated areas, serving as feed for the cattle raised there. Instead of symbolically charged preferences, which Deleuze and Guattari clothe in sloppy philosophical justifications, philosophers of immanence would do well to cultivate all plants, both outside us and in us—in our daily living and thinking.



[1] Gilles Deleuze & Félix Guattari, A Thousand Plateaus (New York & London: Continuum, 2004), p. 17.

[2] Deleuze & Guattari, A Thousand Plateaus, p. 22.

[3] Gilles Deleuze, Expressionism in Philosophy: Spinoza, translated by Martin Joughin (New York: Zone Books, 1992), p. 80.