Why the discourse on sustainable intensification needs to change

By Joern Fischer

Just a couple of days ago, we highlighted a new paper we published in Frontiers in Ecology & the Environment on the topic of “sustainable intensification”. By coincidence, two new papers on sustainable intensification landed on my desk today. One, a paper by Charles Godfray and Tara Garnett in Philosophical Transactions of the Royal Society B, and two, a report by the RISE Foundation called “The sustainable intensification of European Agriculture“. I guess it is fair to say that it is papers such as these two that motivated us in the first place to critically re-appraise sustainable intensification. For those who haven’t read our Frontiers paper: in a nutshell, we argue that intensification without addressing the issue of who benefits from it, and who is involved in the process, cannot legitimately claim to be “sustainable”. The latest two papers I flagged above are emblematic of an analytical frame that is satisfied to deal with production first, and sort the rest later. Adopting such an approach, while arguing that food systems are “complex”, makes no sense — in complex systems, you can’t meaningfully optimise one component in isolation of the others, but you need to deal with various components and their interactions from the outset. (That’s what our figure in the Frontiers paper, reproduced again below, is all about.)

Contrasting ways to conceptualize the role of intensification for food security. (a) Conventional view of several variables influencing food security, implying that variables are independent and additive (additional variables may be considered important by some authors). (b) Alternative view, highlighting interactions and conditionality, with increased production only increasing food security if it passes through distributive and procedural filters. According to this view, intensification can only be said to be sustainable if it successfully passes through filters of procedural and distributive justice.

Contrasting ways to conceptualize the role of intensification for food security. (a) Conventional view of several variables influencing food security, implying that variables are independent and additive (additional variables may be considered important by some authors). (b) Alternative view, highlighting interactions and conditionality, with increased production only increasing food security if it passes through distributive and procedural filters. According to this view, intensification can only be said to be sustainable if it successfully passes through filters of procedural and distributive justice.

In their new paper, Godfray and Garnett argue: “In this paper, we review one aspect of this food sustainability challenge: the goal of producing more food. … We argue strongly that SI and indeed any supply side policy should be developed within the broad context of food system policy including issues of diet, waste and governance. Such an approach is needed to address food security but is also important to allay concerns that SI is a part of a purely ‘productionist’ agenda“.

Looking at the report by the RISE Foundation highlights why this concern might in fact be justified. It argues that “… this report confines itself to issues of agricultural production. The reasons are that sustainable intensification refers to production not consumption, and the expertise and interest of the organisations and researchers involved concerns agriculture …“. Yet, this report, too, talks about intensification being “sustainable”.

Personally, I am very tired of a rhetoric that is holistic and interested in “sustainability”, but that ultimately devotes the vast majority of text to the question of how to increase production. Increasing food production, especially at an aggregate level, is not meaningful for sustainability (i.e. inter- and intragenerational justice), nor for food security (unless the question of who benefits is addressed, simply producing more won’t automatically feed people). Yet Godfray and Garnett argue, for example: “The goal of SI is to achieve higher yields at the aggregate level with fewer negative impacts on the environment“.

As we highlighted in our response to Garnett et al. in Science, we agree that a holistic strategy is needed to address food security; and indeed, a holistic response is needed to achieve sustainability. But once the need for holism is recognised, holistic analyses must follow — issues of equity and governance cannot be afterthoughts to be somehow tackled once enough is being produced. After all, enough is being produced to feed everyone, right now, but nearly one billion people are malnourished.

We can produce more, sure. But it turns out that, per capita, we have in fact produced more and more for many years, but yet, the number of malnourished people has remained constant (as shown by Barrett in Science). Why would we believe that producing more in the future will somehow be more successful in addressing food security — unless we deal with equity and governance issues from the outset, as the primary problem? That, after all, is where we have failed most seriously in the past, not on the production side.

More deeply, the dominant discourse on “sustainable” intensification is centred around supply-demand type thinking, primarily hoping for price mechanisms to improve food security (more food => lower price => more food security). Indeed, prices are part of the story, but recent work on food sovereignty has impressively argued that there are very different analytical frames one can take, which lead to very different answers (see Chappell’s review paper here) — the kind of thinking dominating the existing discourse is just one analytical frame, and arguably one that puts far too much faith into markets sorting out distribution issues further down the line (see our paper highlighting three different discourses on food: production, security, sovereignty).

The notion of sustainable intensification is on the verge of being meaningless, unless some level of “sustainability” thinking is actually put back into this discourse. Focusing on production on its own, with all other issues as an afterthought, is not going to solve problems in complex food systems. If intensification is a small part of the solution — as so many authors are eager to emphasise — then we need to do better at paying more than lip service to all the other issues that are quite possibly more important for both food security and sustainability.

16 thoughts on “Why the discourse on sustainable intensification needs to change

    • Thanks Toby … I am actually very sure some people in this business (not all!) mean very well, but there is a real danger in this discourse, I think, if it glosses over some of the key issues. So, straightening this out would be good — and yes, intensification can be part of the agenda in some settings, but it shouldn’t dominate time and time again as a general vision for improving global food security.

  1. What I find particularly frustrating is that these type of papers, of which there are many, is that they do not even have just a Ceteris paribus (with all other things being equal) assumption (unjustifiable as that would be) that increased food production will increase food security. Rather, they have a “with all other things changed in some completely unexplained way” assumption that increased food production will lead to increased food security.

    This is compounded be the, similarly ridiculous, implicit assumption that the “sustainable intensification” will not impact on the “other important things” like distribution. Here I think that Richard Norgaard’s paper “Ecosystem services: form eye opening metaphor to complexity blinder” is essential reading (link below). While Norgaard is talking about ecosystem services, his points are equally valid for the sustainable intensification debate. Essentially he make two key points, which I liberally interpret below:
    First that simplified policy narratives that engage with ideas from other domains and then ignore the nuanced and sophisticated ways of knowing and understanding those ideas are potentially dangerous because they propose solutions that are simple, elegant and wrong. This “complexity blinding” is clearly the case with the sustainable intensification (SI) debate with regards to conceptual understandings of sustainability.

    Second, that you cannot have a proposed policy intervention that is intended to change the world (as SI clearly is) and then only consider the changes in a case by case basis with the assumption that each and every case makes no changes to the large system in which they are based.

    However, the RISE report, is a whole other level of wrongness. The “made up figure” on the production possibility frontier (Figure 1, page 16) is a brilliant exercise of ideologically driven disinformation and economic illiteracy. I am tempted to write a post about this figure alone, but might have to wait until I am slightly less angry…

    Click to access 10_NORGAARD.pdf

    • Hi Dave,
      as have already written below: Norgaard, too, has pointed to very critical problems of the ecosystem services approach–at the conceptual level (so does, by the way, Clive Spash, a similar case). He does not, however, offer a clear recommendation as to how to apply the insights flowing from his analysis. The translation of such conceptual ideas to actual application is anything but simple (which I am realising within my PhD thesis–I don’t really know how to make the conceptual framework I have come up with “operationable”).

      I’m looking forward to your figure-post;-)

      • hi zielonygrzyb,

        I agree that in terms of ecological complexity Norgaard provides no answers, but he does aknowledge that ES creates incentives to promote certain ways of knowing and I think the same danger is present in the sustianable intensifcation debate. Norgaard does suggests that partial equlibrium models should not be used when addressing non-marginal.system wide changes and suggest that full equlibrium models are needed… that of course does not make them any easier to operationalize.

        For me one of the points of these conceptual modelling tussles comes down to George Box’s comment that “Effectively, all models are wrong, but some are useful”, I would take this further and state that some conceptual models are useful not because they provide insights into how the world is (i.e. scientifically useful), but because the provide a useful “scientifically legitimized” political narrative.The Environmental Kuznets curve is a good example of polically useful, but scientifically empty concpetual model as I think increasingly is the productionist approach of the Sustainable intensifcation debate.

  2. I completely agree with you that if one argues in favour of a holistic view of the issue one should also offer one. And in this specific context, to be sure, there is no way around holistic approaches. However, I can understand why people are doing that–focusing in the end on “sustainable” intensification. It is one thing to admit that the issue is of vast complexity. It is another to actually be able to provide an analysis that accounts for this complexity. Your own framework is rather conceptual, and at this level it works. But to make it more concrete, practical, to offer, say, specific policy recommendations, is extremely difficult when you want to account for complexity in a holistic way. That is, I think, one reason why people are sticking to one aspect (food production/supply) and do as if it could be treated “ceteris paribus”, delinked from all the other issues, be it distribution, empowerment, governance etc.

    This is an issue that I know all too well from my own discipline, economics–we might all be aware that mainstream economics is rubbish, its models too simplistic etc. But it is extremely difficult to offer a consistent alternative that would work better in actual applications. Having said that, I do not want to pledge for sticking to “sustainable” intensification or mainstream economics. But I can understand why such narrow approaches prevail.

    • Thanks for your comment. It’s really good to see commentary on this that is critical of our re-appraisal — it’s only through engaging with this from multiple angles that we can move forward!

      I think you are completely right that what can be done conceptually is quite different from “the real world”. But I think that sustainable intensification — as an agenda — just SEEMS more concrete than other steps. What we advocate in our new paper in Frontiers is that we should work out on a regional-scale basis what is a useful way forward. That is, in some places in sub-Saharan Africa, intensification is what is needed, and of course, we should make sure it is at minimal environmental cost. But there is simply no need — to my mind — for advocating “sustainable” intensification as a general vision, precisely because if you take it to this very not-concrete level, it is extremely likely that the benefits will not reach those in most need of more food.

      Ultimately, this boils down to the “normal” clash between changing the world within a given paradigm or political system, or challenging the paradigm and political system. Sustainable intensification is successful in policy circles because it does not rock the boat. Moreover, it uses rhetoric that is agreeable to everyone from the farming community to green groups.

      Again, don’t get me wrong: sometimes, intensification is needed. But it is equally, and often more, important to put our collective brainpower into those other, trickier things, rather than hope they will sort themselves. I think focusing on specific regions, it is possible to do this. For example, as a regional government within country X, I can think about if I want to focus on access mechanisms, or if I want to focus on modernising farming to get higher yields. Or I might need a mix of both, because just focusing on modernising is useless unless access mechanisms improve.

      I think focusing on regional-scale solutions is what is needed. Globally, the single biggest problem is not a lack of production, so pretending it is, seems wrong, and is basically a displacement strategy from dealing with known evils that require more than marginal adjustments.

      I’m not sure this turned out very coherent, but so yes, I understand why it’s practical, but I think a focus on intensification is simply not going to be particularly helpful, on its own.

  3. @Joern:
    I think, you hit the nail on the head when you emphasise the need for small-scale approaches. Indeed, this might be a general problem of policy and science that deal with global problems (such as sustainability in general, hunger, climate change etc.)–the assumption that since the problem is global, the (one-size-fits-all) solution should be as well. But as global problems have many different facets in different regions, there is also need for “tailored” solutions.

    That intensification (or: supply of food in general) is not decisive (although, as you rightly point out, it might be important in specific cases), should actually be clear since at least 1981, when Amartya Sen published Poverty and Famines

    I didn’t want to downplay Norgaard’s contribution–in fact, it is one of the most important ones in the ESS debate. He pointed out to very important dangers and weaknesses of the ecosystem services framework. I only wanted to emphasise that the step from recognition that the critique is justified to adoption of the models/paradigm is anything but simple.

    I also agree with your “extension” of Box’s comment on models. The only thing I don’t agree with is to call the environmental Kuznets curve a good thing;-) Even in terms of its political impact. It creates the impression that we in the wealthy countries of this world can lean back and focus on (paternalistically) helping the poor, as they are the ones who damage the global environment, while in fact it might be argued that we in the rich countries are responsible for the bulk of environmental destruction around the globe.

    P.S. As you called me zielonygrzyb, I guess you don’t realize that we worked in the same room last week in Lüneburg;-)

    • Bartosz! (I blame my aging eyes and the tiny size of the picture).

      I entirely agree it is easy to criticise, but often harder to do something better. I think in the context of the sustainable intensification debate one apporach would be to work out what the determinants of food security are in a specific casestudy and try and map out the interdepencies of those drivers (i.e. if you change the governace, but not the production what happens? if you change the production, but don’t address distribution what happens? etc..). This kind of “systems thinking” is pretty well established in sustainability science and its application to food security would help counter the “we will worry about increased production and someone else can deal with the other issues” mindset that dogs the current debate.

  4. I don’t know that I can add anything to the excellent points being made here. But I will try futilely nonetheless!

    Actually, I can’t do better than to re-emphasize Dave and Joern’s last points: the seeming complexity and intractability of more holistic approaches somewhat (SOMEwhat) falls out when applied to specific cases. In some ways, to lean too heavily on cliche, I view this area as a perfect storm of academia-applied science-reality clashes.

    The complexity and challenge of making a globally applicable model appeals to scientific zeal, pride, and reward systems — one can propose any number of models and model refinements that attempt to tackle the problem holistically or within particular domains, all the while with caveats that the model is of course, too simple, but may be useful. You can make progress nearly continuously without, necessarily, breaking through to a truly globally useful model.

    Meanwhile, you may come up with any number of high-quality case studies, qualitative examinations, or models that apply very well to a particular place and time. Each of these is ****far***** more likely, imho, to be useful for the decisions in any particular community or for any particular set of decisionmakers, but in being locally bound, contingent, and of limited generalizability–not to mention perhaps being “insufficiently”, if at all quantitative and not rigorously controlled–all mitigates against such a work being publishable in a top-impact journal.

    So the reward systems are set up to keep giving great profile to simplified, less-holistic work, or more complex, but less-specific models, while the likely very useful work of making very specific, holistic, but scope- and scale-limited models is seen as… well, I’m not sure what it’s seen as, but it doesn’t seem to be what scientists are most rewarded to do!

    All this is to say that I think Norgaard–and our–critiques can be well, if not quite “easily” dealt with by working in specific contexts and cases where one has a better chance of being able to specify complicated elements like governance, distribution, marginal prices vs. approximate lifecycle costs, qualitatitve and multiple ecosystem services, etc. — getting at some of these with participatory research and modelling of the type that is almost inherently un-scale-up-able. The seemingly intractable complexities are less so when one is not trying to require a model to be highly specified, holistic, and generalizable.

  5. Pingback: The Not-So-New Climate Economy Report | The Sceptical Economist

  6. Pingback: Communicating beyond the ivory tower (or why comics matter) | Ideas for Sustainability

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s