Do Trees Exist?


Fig. 1. Encephalartos ferox

by Justin E. H. Smith


There are two very different essays I’ve been meaning to write, both of which equally merit the title of the present one.

The one would address the special meaning of ‘existence’ as distinct from ‘being’ in the Heideggerian tradition, and would ask whether that special meaning applies to trees. Heidegger had dismissed even animals as merely being rather than existing, to the extent that they are ‘poor-in-world’ and therefore do not have that special and rare power human beings supposedly have of standing outside of being, ex-isting as Heidegger the etymologist emphasises, and beholding being rather than simply being whisked along by it as everything else is. A fortiori, it is usually assumed, the world of plants must be even poorer, and thus trees are taken as exemplary instances of being without existing.

Looking at the hanging branches of Louisiana oaks though, as I am at present, reaching down to the bayou in search of something they want, I wonder if this is true, and it seems to me, at least fleetingly, that I can imagine what it is like to be a tree, that there is something it is like, and that it is weltreich indeed. If it is difficult to grasp this, this may only be because we are limited in our sympathy by the radical difference of time scale that separates their experience from ours. I am fairly certain that if this difference were removed, we would see that trees too have projects, and experience feelings of accomplishment, defeat, joy, pensiveness, and wistfulness.

But that’s not what I’m writing about here.


The other essay that may be called, “Do Trees Exist?”, i.e., the present one, concerns not phenomenology, but taxonomy, and the inexhaustible debate in the philosophy of science about social construction.

It is my opening contention that trees exist somewhat in the way, say, long-legged animals do. That is, there are such beasts, but together they do not constitute a natural kind. There is no taxon, based on any meaningful criteria, that would group together giraffes and Japanese spider crabs, but not giraffes and hyraxes. There is moreover no way of fixing a meaning of “long” that is not based on pure convention.

A tree is a perennial plant, as are many bushes, shrubs, subshrubs, lianas, and so on. Its specific differentium is supposed to be its trunk. But what is a trunk? Well, it’s a long stem. And what is it that all long-stemmed perennial plants have in common, other than the length of their stem, that separates them from non-long-stemmed plants? And, again, what counts as long? The answers are, respectively: “Nothing”; and, “No one knows”.

Phylogenetically, a pine tree is far closer to a cycad (see fig. 1), than it is to a pin oak (see fig. 2). Both of the former are gymnosperms, a relatively more ancient sort of plant that offers its seeds up to the wind through a loose and exposed structure such as a cone. Oaks are angiosperms, whose seed is enclosed in a fruiting structure, and usually requires the activity of animals to be broken out and transferred to the soil where it might grow to adulthood. This is a difference in the basic mechanics of reproduction that is at least as great as that between egg-laying and live birth in animals, and it is not surprising that there should be such a difference, given that gymnosperms evolved and colonised much of the earth’s terrestrial surface around 280 million years ago in the Permian era, while angiosperms did not appear until the Cretaceous, roughly 140 million years later.

Angiosperm reproduction is highly specialised, and, if you pause to think about it, no less strange than the multi-species reproductive strategies we observe in the animal kingdom, for example when female Sacculina barnacles insert themselves during the larval stage in the bodies of adult green crabs, and use their hosts to attract male mates. Flowering plants required, in order to evolve, pre-existing animals –insects, mostly– that would be attracted to the sight and smell of them; flowers and fruits are an adaptation that makes no functional sense considered in isolation. Insects existed for hundreds of millions of years before plants came up with this way to take advantage of them. Indeed flowering plants are so young in evolutionary time that there was even a period, not so long ago, when not just insects, but mammals as well already existed, while flowers did not.

Gymnosperms are in this sense the surviving prehistoric cousins of the new, young, and dynamic species that have colonised much of the world since the Cretaceous, and must appear to an apple or cherry tree in magnificent bloom something like a primitive coelacanth appears to us. Neither in their evolutionary histories nor in their perceptible adaptive functions are the pine and the oak more like each other than are the pine and the cycad. But the bare morphology of the pine and the oak –the fact that they both have trunks– overwhelms any other consideration and forces the folk-category of tree before our minds, blocking from view the stout and ground-hugging cycad.

Fig. 2. Quercus palustris

All this should matter for anyone concerned with the way folk-taxonomy and science match up. It is noteworthy that such a solid member of our folk-taxonomy as the tree —that is, a kind that is commonly and reflexively adduced whenever we wish to enumerate a short list of examples of the various things there are in the world— should so obviously turn out to have no secure place in any viable scientific taxonomy. We are, it seems, compelled to speak of a broad spectrum of plant species as if they formed a real kind, in view of our natural attunement to morphology and habitat. But we have long been insistent, to the point of superciliousness, on the irrelevance of morphology and habitat for the classification of many other biological kinds: observe for example the strong reactions you’ll get if you claim today that whales are fish, and note that it is no help at all, in making the case for this claim, to point out that whales and fish have the same general body type and live in the same environment. Why is there no parallel pressure to adhere to standards of botanical correctness? It seems that trunks are just too much a part of our social reality to treat the life forms that have them otherwise than as belonging to a real class of beings. Morphology is trumped by phylogenetics in the history of marine mammals, so that ‘cetacean fish’ is now a contradiction, while a phylogenetic argument against the existence of trees appears, for now, as unnecessary trouble-making.


We take the possession of a trunk to be not just the certain marker of membership in an ontologically secure kind, “tree”, but also to be the certain marker of individuality. Yet a bit of digging often reveals that every trunk in an entire grove, such as the colony of quaking aspen known as “Pando”, is genetically identical to every other, and connected to every other by an underground network of roots. Even when they are not connected, genetic identity  among trees is common. Every weeping blue Atlas cedar (Fig. 3) in Europe descends from a specimen at the Vallée-aux-Loups arboretum just south of Paris; or, to put this another way, there is a weeping blue Atlas cedar south of Paris, parts of which have now spread throughout Europe.

Fig. 3. Cedrus atlantica

At the outer limits of the folk-taxon in question (“tree”), we find far more problem cases than we do when asked to close our eyes and “think of a tree”. Our imagination pulls up a paradigm instance, a species with a large, solid, and singular shaft rising like a pole: a proud sequoia or spruce (though I suppose that adjective starts to bring us back to the other essay, the one I didn’t write). But nature also gives us the Quercus virginianus or Southern live oak, whose trunk often, though not always begins to branch in several directions often as soon as it emerges from the ground, some growing out diagonally, some curving back down towards a body of water, and some jutting out perfectly parallel to the earth. If it were smaller, the live oak would plainly be counted a shrub. This is very different from the Quercus ilex of the Mediterranean, with its much more paradigmatic and simple trunk structure, even though, obviously, any two Quercus species are phylogenetically much closer to one another than, say, the Quercus virginianus is to a rhododendron.

Quercus is among the genera notable for the frequent appearance of “knees” or pneumatophores, roots that grow up out of the ground, often forming colonies of lumps that, if one did not investigate, might easily be taken to be independent of the larger trunk standing somewhere in their vicinity. As their scientific name and their exposure to the atmosphere indicate, their function is breathing, or something akin to that (Figs. 4 and 5).

Fig. 4. Quercus virginianus

Fig. 5. Quercus virginianus (pneumatophores)

Early modern aesthetics, as, notably, the work of Shaftesbury, was preoccupied with such things as the unfathomable complexity of a bed of moss. This preoccupation is expressed systematically in Leibniz’s analysis of organic bodies as having structure within structure in infinitum, and endures even in Kant’s treatment in the 1790 Critique of the Faculty of Judgment of those features of organised beings (the example he chooses in his discussion is, interestingly, a tree) that distinguish them from artefacts and aggregates. Yet in European romantic landscape painting in the following century, the dynamic sublime –the apprehension of the power of nature as manifested by precipices, the ocean, skies black with clouds, and so on–, as depicted most famously in the work of Caspar David Friedrich, would take precedence over an earlier preoccupation with instances of the mathematical sublime, where imagination strains to take in what reason tells us must advance to infinity and must therefore surpass it.

At the same moment in history, forestry was emerging as a rigorous science that involved, among other things, censusing the trees in a given region of Europe as if they were its citizens, treating trees, politically and administratively, as paradigmatic individuals (see, on the history of forestry and its place in the emergence of the modern administrative state, James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed, 1999). Yet when we turn to the colonial experience, we see that the dynamic sublime is still very much at the centre of painterly preoccupations, and that, while colonial administrators were trying hard to impose order on new landscapes for which Europeans often lacked an adequate vocabulary, artists were undermining this project simply by rendering in honest strokes what they perceived. Thus in Henry Chapman Ford’s “Water Lilies and Spanish Moss” (Fig. 6) of 1874, an oil painting capturing a typical scene of the Louisiana bayou, we discern a world that manifestly refuses to let us tally up the number of individual vegetal beings within it:

Fig. 6. Henry Chapman Ford, “Water Lilies and Spanish Moss” (1874)

It is not just the ubiquitous Spanish moss that makes the counting task impossible, but also the features of the oaks themselves that the moss enshrouds. Thus in a detail from the lower right corner, we see Ford’s careful attention to the colony of “knees” next to an oak:

Fig. 7. “Water Lilies and Spanish Moss” (detail)

Ford seems to have understood that the pneumatophores are at least as worthy of representation, and of an attempt at tallying, as the trunks of the oaks of which they form a vital part. It is difficult indeed to observe the totality of strange excrescences and protrusions that the bayou yields up, and to continue to believe that any segment of it is more worthy of representation than any other.

In the 19th century European critics sometimes said of American landscape painters’ depictions of the bayou that these should not count as landscape painting at all, since it is the very essence of the bayou to evade easy determination of where the land leaves off and the water begins. Government surveyors were hard at work trying to create passably accurate maps of the region, delineating the boundaries of land and water, while in parallel painters, some of whom were also employed by the government as surveyors, consistently undermined this effort simply by showing the world as they observed it. In this respect some American landscape painting, depending on the landscape in question, represents an alternative trajectory that had been abandoned in the singular focus of the European romantic painters on the dynamic sublime.


The fact that we commonly take, as handy go-to examples of what there is, entities that are so easily shown by science to be “more complicated than that”, and the fact that we take to be countable entities par excellence the very same entities that a certain tradition of aesthetic reflection takes to be most resistant to our will to tally things up, should tell us something about how both folk-categories and scientific categories work. There are no such things as trees, yet many political, industrial, colonial, and other social considerations compel us to speak and act as though there are. Many other considerations, notably of aesthetic reflection and artistic creation, compel us to acknowledge that there are not.

John Stuart Mill did as good a job as any ever could when he proposed that a natural kind is one in which we discover ever new properties beyond the initial property that motivated us to group a cluster of things together under a single name. By this reasoning, flowering plants (some of which are folk-categorically held to be trees) constitute a natural kind, while trees do not. Yet trees are not going to go away any time soon, as we will continually find ourselves in social situations in which it makes sense to speak of them (e.g., when an acquaintance slams his car into one of them, or when we sit in their shade or protest attempts to cut them down), while it is hard to imagine a comparable social need for any single unifying term that unites all long-legged animals. An encounter with a Japanese spider crab is just too different from an encounter with a giraffe to be placed under the same heading.

Social salience is, in short, really important. No amount of botanical reeducation can train out of us the sense that trees ought to make it onto the list of what there is. Here then we have a case where folk knowledge is plainly preferable to its scientific alternative. So, I take back what I said earlier. Trees exist.

Or maybe they don’t. I don’t know. It depends on what you mean. Plainly, anyhow, whichever side we take on this issue, we will never generate the sort of heated controversy we see when certain other examples (e.g., “man”, “woman”) are chosen. This casuistic difference strongly compels the conclusion that, in general, side-taking in debates about social construction has little to do with theoretical conviction, and much more to do with specific attachment to a particular socially salient kind, and with a sort of faith that this attachment, in order to be worthy of us, must be rooted as it were in deepest ontology.


Lafayette, Louisiana, December 29 2019. Essay crossposted with