Ecosystems are Not Smart, We Are – Applications on the Farm

A display of multiple variety of seeds.
17 species cover crop blend. Photo: A. McGuire.

In a recent post, I argued that we should cast aside the ideas of “balance of nature” and “nature knows best” in designing farming systems. If nature has not been optimized by any process that we know of, and therefore consists of mostly random mixes of species dictated primarily by natural disturbances, then there is no reason to “follow nature’s lead.”  But if we don’t, what are we left with?



We are left with an agriculture based on human ingenuity, consisting of:

  • Crop rotations; or better yet, dynamic crop sequences;
  • Residue management and no-till planting to keep the soil covered and control erosion;
  • Careful use of synthetic fertilizers in conjunction with organic fertilizers;
  • Cover crops and green manures, including cover crop cocktails; this is where we can study unused and underused species to take advantage of “nature’s wisdom.”  Precision crop planting in sequence with cover crops could potentially improve cover cropping benefits by allowing crop roots to advantageously colonize the root channels of the dead cover crops (i.e., sequential root channel colonization).
  • Integrated pest management including the use of improved pesticides.
  • GMO crops, including cover crops.

All these practices could be more widely used and more effectively applied.

How about those practices promoted because they mimic nature? Denison, in Darwinian Agriculture1, evaluates four of these nature-based practices: perennial grain crops, reliance on only local sources of nutrients, intercropping, and reliance on diversity to control pests. He then goes on to survey the evidence for each and gives his assessment. I’ve summarized Denison’s assessment here.

  • Perennial grain crops – Denison concludes that lower-yield perennials have their place, especially in feeding livestock, but “given the tradeoff between perenniality and seed production, emphasis on grain production may be misplaced.”
  • Reliance on only local sources of nutrients – With regard to following nature Denison asserts, “local sourcing of nutrients in natural ecosystems is a constraint imposed by lack of external inputs, not an example of ‘nature’s wisdom’.” In other words, although it may be advised to use local sources of nutrients as much as possible, it should not be a constraint to us, just because it is a constraint of natural lands.
  • Intercropping – Denison points out the errors commonly found in intercropping experiments, chiefly the failure to find or use the optimum density for the monocropped plots, which favors the intercropped plots. The design of these types of experiments is complex, but even in those carefully designed, Denison finds in his survey of research results, that “most intercrops yield more than the average of the two [or more] crops, but less than the best crop alone.” In this case, farmers will tend to grow the best crop. He adds, “Diversity may be there for a reason…but that does not mean that diversity is there for a purpose.”
  • Reliance on diversity to control pests – Here Denison compares diversity in space, the “balance of nature” that inspired intercropping, with diversity in time, or crop rotation, which has a long history of success and is not commonly found in nature. With intercropping, he points out, you expend the advantages of diversity (here, for pest control) the first year the intercrop is planted. What do you do for diversity the second year, he asks? On the other hand, with crop rotation, the whole system changes each year. Denison suggests that the latter would work better than the former in the long run, but he is not aware of any research that has addressed this question. Nevertheless, there is no lasting solution to pests; “ongoing evolution [of pests] will tend to undermine all of our pest control measures, not just those based on toxic chemicals.” We cannot get off the pest control treadmill.

Although I think we would be better off without the “mimic nature” baggage, I am not saying that biodiversity is not important. We should incorporate more diversity in our cropping systems, not because nature is diverse, but so we can better use the properties and processes in individual plants. If it exists anywhere, the “genius” of nature is in individual species and not in the ecosystems. Right now, other than corn, soybeans, rice and wheat, we have not thoroughly explored the capabilities of many plants, domesticated or not.

Neither am I saying that interactions between species are not important. We should study intentional combinations of species, mixtures not found in nature, searching for simple, direct interactions between species that will give us our desired results.  As Denison points out, we may gain more knowledge of individual species when they are studied in communities with other species – a job for agroecologists! However, if the idea that the sustainability of natural ecosystems depends on complexity is an illusion, agricultural systems should only be as complex as needed for our purposes, not more so.

Another benefit gained by casting aside “the romantic notions of a stable Eden”2 is that it should make us less susceptible to “silver bullet” solutions, wishful thinking and other such nonsense. In my experience, this is most needed in soil and pest management. There are no quick, easy, and cheap methods to improve soils. It takes bulk organic materials, either grown on-site (less expensive) or imported (more expensive). In the long-term, the nutrients that are harvested in the crop must be replaced; they cannot be produced by “better biology.” For insects, weeds, and disease, no amount of tweaking the system will make them go away.

There are those who will find this whole notion yet another example of arrogant man trying to control nature, and there are plenty of examples of where we have done a poor job at managing the Earth. However, we must realize that farming is controlling nature for our own purposes. We still need nature, and “wild” places, but unless critics can point to a mechanism by which natural ecosystems were consistently improved, we should not use them as blueprints for agriculture, nor should we assume that we cannot improve on them. There is no utopian state of nature, so we can stop trying to restore, recover, or regain any such state in agriculture. There is no way back, but there is a way forward.


1 Denison, R. F. (2012). Darwinian Agriculture: How Understanding Evolution Can Improve Agriculture. Princeton University Press. (Denison’s other main thesis in this book is that natural selection has left us few tradeoff-free opportunities for genetic improvement through genetic engineering. Full Disclosure – Denison served as my major professor in graduate school at UC-Davis.). His book-related blog.

2 Marris, E. (2013). Rambunctious Garden: Saving Nature in a Post-Wild World. Bloomsbury Publishing USA.


6 comments on "Ecosystems are Not Smart, We Are – Applications on the Farm"
  1. Andy, I’ll write a proper response to these two fascinating posts of yours on my blog when I have the time. Until then, I offer seven thoughts on why I’m not persuaded by your overall argument (even though I agree with some of the details):

    1. If research studies fail to identify an emergent order in ecosystems, then describing them as stochastic and not ‘balanced’ is no doubt a reasonable shorthand description for the empirical findings. But if you go on to infer from that that farmers shouldn’t mimic nature on the farm, but improve it, you’re going out of the realm of science (can you formulate that statement as some kind of testable hypothesis?), and into the realm of a myth which references various implicit philosophical and sociological commitments.

    2. But maybe that’s OK: for me, ‘myth’ isn’t some pejorative term for fictions that can be banished by scientific rationality. Myths are organising ideas for how we think and live – worldviews or ideologies if you will (forgive me, I was trained as a social anthropologist). Our everyday thought is inextricably mythological, and so is scientific thought, though in the latter case there are some nice ways of preventing the myths from getting too out of hand. But myths do still get out of hand, particularly when they interact with other myths. I can foresee your ‘improving nature’ myth interacting with another presently dominant myth, namely that the pursuit of private material interest maximises social welfare, to produce outcomes such as the destruction of tropical rainforest in favour of soya or palm oil and an attempt on humanity’s part to assume the burden of the ecosystem services previously provided by the forests. I’m not convinced that’s a great idea, and therefore I’m inclined to think that your myth is potentially quite a dangerous one.

    3. Talking of ecosystem services, one problem with the ‘improving nature’ myth is that it effaces the question of trade-offs between ecosystem services and other goals such as productivity, in a way that the balance of nature myth doesn’t – this is possibly similar to the point made by Ford Denison on his blog about your post. Can you simultaneously improve on nature in delivering all the outcomes people need from the agroecosystem? I’m not so sure – where’s the evidence?

    4. Even if it’s true that there’s no emergent order in natural ecosystems, it doesn’t follow that the concept of ‘balance’ in nature has no utility.

    5. When you talk of ‘romantic notions of a stable Eden’, I think you may be reading the Bible rather too literally – a common problem in the USA, if I might make so bold. It strikes me that the Eden story in Genesis was written by somebody or somebodies who knew a lot about farming, and who wanted to make the point that humans have gained immense transformative power over the environment, that our actions often have unforeseen and unintended negative consequences, and that we have no choice but to make the best of this we can (I published an article on this a few years ago in the ‘Journal for the Study of Religion, Nature and Culture, should you be interested). I’d say that their thesis stands up pretty well to historical scrutiny…

    6. …so though I’ve learned a lot from the critique of a ‘natural balance ecology’ I’m inclined to agree with Daniel Goodman, despite his condescension, when he wrote “The diversity-stability hypothesis…will recede to a revered position in the popular environmental ethic, where it doubtless will do much good”. More good, I suspect, than an ‘improving nature’ myth at any rate.

    7. Finally, though I agree with Dave Wood’s comment under your last post that it’s possible to design agroecological systems badly (eg. cereals under forest cover) is anyone actually doing this? A better design to optimise productivity and ecosystem services would be alley cropping of cereals in between woody windbreaks, which people are doing. Is such a design based on a model of ‘improving nature’ or on ‘the balance of nature’? I’d say both, possibly. In which case, maybe we shouldn’t get too hung up on the models.

    1. 1: First, why shouldn’t we infer that farmers shouldn’t mimic nature if it has no purpose or direction or order other than that determined by natural disturbances? What would be the reason to mimic nature, other than making use of what individual species have to offer, which I commend? I never meant “we can improve nature” as an experimental hypothesis, but how about this, “If we ate only what nature by itself produced, people would starve.” However, we have improved on nature in agriculture – as I stated in the previous blog, it’s what we’ve been doing all the time – and so starvation is not a production problem. That is a fact, not a myth.

    2. 2: In regards to what you are talking about here, perhaps principle is a better word than myth? I don’t think many people would agree that destruction of tropical rainforest is “improving nature,” but if people have to eat, and the forest is gone, and we have choices to make about how to conduct the agriculture that took its place, then copying nature should not be our model – how many people would survive eating what nature produced out of the rainforest? We can improve on nature in relation to the purposes important for agriculture, and by doing so, we just might be able to save more of that rainforest.

      3: Again, “improving nature” is not a myth in regards to agriculture, it is a fact. Otherwise, our population would still be very low, limited to what nature produces. And my assertion is that, by more widespread and effective use of the practices that I mentioned, we can produce and obtain the ecosystem services we need. After all, undirected nature is not arranged to provide ecosystem services, why should we think we cannot do as well or better with intelligence?

      4: I disagree. If you keep the concept, how will you go about copying a balance that does not exist?

      5: This is a quote from Rambunctious Garden, and I do not know if the author was reading anything into the Bible, or, as I thought, was simply using the general reference to Eden as a metaphor for nature in its pristine, balanced, state. I highly recommend this book, although I am not sure if the author would agree with my views here.

      6, 7: I disagree. Look at all the practices, derived from nature , that Denison evaluates in his book, and which I summarized in my post. Do you think that all the time and resources used trying to mimic nature through these practices were well spent, given that they failed to produce what we wanted from them? Furthermore, I believe that this idea, combined with the “appeal to nature” argument, has crept into our thinking in other areas, with similar disappointing results; anti-vaccine movement, anti-GMO movement, fad diets, alternative medicine, cancer treatment, etc. We need to make a clean break from this misleading notion and take full responsibility for what we do in agriculture.

      Chris, I look forward to reading your blog post on this.

  2. OK, well I’ll try to write something on this soon. My feeling though is that you’re running off a bit too much with the anti-mimicry theme. True, any type of farming is necessarily an effort to ‘improve’ on nature. But it still involves ‘mimicry’ of nature. Denison criticises ‘misguided mimicry of nature’, which seems reasonable to me; presumably he’s open to the possibility of ‘well guided mimicry of nature’. Perhaps part of our disagreement is about the context of terms: ‘nature’ is not the same as ‘ecosystem assembly’. I don’t have too much of a problem with the examples of misguided mimicry you give, or your examples of ‘agriculture based on human ingenuity’ (though all of the latter, I’d argue, also involve ‘nature mimicry’). But I think the wider inference you’re making from it is a bit problematic – I’ll have a think about it more and write something on it when I get the chance between ploughing, fertilising etc… Thanks for a thought-provoking post anyway.

  3. In this discussion, Andy McGuire and Chris Smaje answer different questions.
    Andy asks: Does the balance of nature metaphor describe how the natural world works? He summarizes considerable evidence that does is not.
    Chris Smaje asks: Does the balance of nature metaphor provide ethical guidance? He implies that it will “do … [m]ore good . . . than an ‘improving nature’ myth”.
    I suggest that as a guide to ethical behavior, the record of the balance of nature metaphor is not stellar, at least when viewed through contemporary understandings of human rights. Yes, this view can foster humility toward the natural world. But, it has also served and still serves as a justification for removing people from their homelands in order to create “pristine” reserves where we like to believe that nature, divorced from human impacts, is in balance. (See Cronon, W., ed., 1996, Uncommon Ground: Rethinking the Human Place in Nature, 2nd edn, New York, Norton; Jacoby, K., 2001, Crimes Against Nature: Squatters, Poachers, Thieves, and the Hidden History of American Conservation, Berkeley, University of California Press; Agrawal, A. and K. Redford, 2009, Conservation and Displacement: An Overview, Conservation and Society 7:1-10; Dowie, M., 2009, Conservation Refuges: The Hundred-Year Conflict Between Global Conservation and Native Peoples, Cambridge, USA, MIT Press.)

  4. Re Bertie Weddell’s point, I agree that the ‘balance of nature’ myth has been used inappropriately to dispossess people, but then again so has the ‘improving nature’ myth: there are countless examples of hunter-gatherers, pastoralists and peasants throughout the world being sacrificed to the idea of improved productivity ever since John Locke’s ‘Second Treatise of Government’. I don’t think there’s any one single metaphor of nature that uniquely safeguards human rights.

    My argument isn’t principally that we should stick to a ‘balance of nature’ script even though it’s wrong because it’s ethically preferable – it’s more that lack of evidence for a ‘balance of nature’ does not logically support the case for human ‘improvement’ of nature. I’ll post a more detailed analysis of this soon on

Leave a Reply

Your email address will not be published. Required fields are marked *