What was eating at Sir Arthur G. Tansley? On the occasion of 76th anniversary of the ecosystem concept

Remembering the late Frank Golley who wrote about it.

This month is the 76th anniversary of the naming of the ecosystem concept. The paper in which the term first appeared is entitled, appropriately enough, “The use and abuse of vegetation concepts and terms”, (the ecosystem being offered, naturally enough, as a useful one) and appeared in Ecology, the flagship journal of the Ecological Society of America, in July 1935 . In it, the British ecologist Sir Arthur George Tansley takes a scythe to the conceptual crop that the infant discipline of ecology had produced since the early years of the 20th Century. He separated the terminological wheat from the chaff. The task was not, of course, simply to determine which were the more delicious ecological terms; ecologists have always loved their terminology: climax communities, subclimaxes, proclimax, seres, plagioseres, prisere, succession and so forth; little of it is euphonious. Rather, he determined the utility of terminology based upon his perception of the “naturalness” of the units of nature they entail. Terms are abused when they promote analogies which do not produce fruitful empirical research programs. Terms are also abused, as Tansley hinted, when they smuggle in a set of metaphysical assumptions (e.g. “an imagined future whole” to be realized in an ideal human society); assumptions Tansley believed had no business in the dispassionate contemplation of nature. Especially when ecology was struggling to assume a position among the other natural sciences it seemed not the time for the discipline to divert on philosophical tangents .

The definition of ecosystem given in the 1935 paper is remarkable for its expansiveness – Tansley’s ecosystem cannot be captured in a sound bite. Contrasting it with the “biome”, a term that signifies an aggregate of all the organisms of a region, and one which Tansley declared “unobjectionable”, he stated however that:

“the more fundamental concept is, as it seems to me, the whole system (in the sense of physics), including not only the organism-complex, but also the whole complex of physical factor forming what we call the environment of the biome – the habitat factors in the widest sense.”

Two things are especially noteworthy here: on the one hand, the ecosystem was proposed as a tractable unit of nature, that is, as a methodologically useful aggregate comprised of the biome compounded with its non-living environment. This unit is therefore simply one of the “multitudinous physical systems of the universe”, systems that start with the atom and agglomerate to the scale of the universe. The systems are continuous but it is useful to isolate them for the study of nature. Less obvious from the definition or, indeed, from the way in which contemporary ecology takes up the ecosystem concept, is the fact that the term was proposed as part of the framework in which detailed studies of the processes governing ecological change, so-called succession – can and should be undertaken. The persistent question of how we should understand vegetation change has been answered in a variety of ways in the history of ecology. The changes that we see as vegetation develops on nude land can be regarded as akin to the development of an organism – each stage before the adult emerges can be seen as regarded as larval stages inexorably progressing to that adult condition (a “complex organism”). On another view, changes in the complexity of vegetation can be seen as an outcome of biotic interactions constrained by climatic and/or edaphic factors without the changes being necessarily progressive or converging on a predetermined end point. From the view point of models of successional change the ecosystem is terminological useful in tagging a certain type of complexity, and in identifying a methodological unit, the investigation of which may elucidate the mechanisms of change.

The ecosystem as enunciated by Tansley is therefore conceptually situated at the intersection of two importance ecological themes: the identification of a ecologically relevant entities in the hierarchy of physical objects, and the mechanistic understanding of successional change. The two themes are mutually reinforcing in a manner that can get forgotten. It may be useful to examine ecosystem properties of relatively stable system states - in fact, in Tansley usage the ecosystem can be regarded as the biotic and abiotic conditions that exist at one particular time. Tansley regards the ecosystem unit as having stability (“a dynamic equilibrium”) less than the stability of the chemical elements but which can nevertheless be stable for thousands of years. That being said, the ecosystem is nevertheless also relevant for understanding the balance of stasis and change in the history of a system over relatively long periods of time. Biotic pressures can influence the development of an ecosystem; bison grazing influence the development of the short-grass prairie being an obvious example. It is not easy, therefore, to draw a bright line between the regulatory effects of some animals from the effects of man. An ecosystem perspective, therefore, is a potent one for contemplating the contemporary effects of people of ecosystems.

Perhaps all great scientific papers should open with an apology. Tansley’s does because he intended to be blunt and provocative with his interlocutors. The proximate object of his comments was a series of papers written by the South African ecologist John Phillips who related the ideas of Frederic Clements on successional vegetation patterns with the philosophical musings of Jan Christian Smuts. Though Phillips bore the brunt of Tansley’s critique (the former had made the unfortunate mistake of calling on the latter as a supportive authority for his speculations) the real object of the paper was to weed out notions of a complex organism from ecology. When he summarized the following terms he regarded are unobjectionable: succession, climax, ecosystem, quasi-organism (some aspects of the successional development can be regarded as developing as an organism). Rooted out and cast aside are: biotic community (animals and plants considered together), complex organism (a climax community seen as having the properties of an organism). Both of these are ones singled out for special attention by Phillips. These notions may have been philosophically unsavory to Tansley, but he largely ignores a philosophical treatment of them, preferring to not that any methodological value they may seem to possess is illusory.

The papers of John Phillips are largely forgotten, the successional theories of Clements, though influence are stripped of their super-organismic flavor and succession is largely discussed in mechanistic term and without the burdensome terminology that Clements created. The ecosystem concept remains strikingly influential and in a way that I think Sir Arthur Tansley would have appreciated, the arbitrary borders between questions about organism-complex, that is, the biological community and ecosystem processes are increasingly being broken down. A prevalent question in contemporary ecology is the degree to which the complexity of the community influences the functioning of ecosystems. Strikingly these are question posed in both human-dominated systems and in “natural” ones. 

1.Tansley, A G (1935) The Use and Abuse of Vegetational Concepts and Terms. Ecology, Vol. 16, No. 3. (Jul., 1935),

2. For an excellent account of the background to Tansley’s paper see: Frank B. Golley: A History of the Ecosystem Concept in Ecology. More than the Sum of its Parts. New Haven/London: Yale University Press 1994

3.  Tansley, 299