Monday, July 13, 2015

Wicked problems and wicked solutions: the case of the world's food supply


I am back from two days of full immersion in a meeting on something rather new for me: the world's food supply. I am still reeling from the impact. Whenever you go in some depth into anything; you see how immensely more complex things are in comparison to the pale shadow of the world that you perceive in the glittering screen of your TV. Everything is complex, and everything complex becomes wicked once you start seeing it as a problem. And wicked problems usually generate wicked solutions. (image from Wikipedia)



Can you think of something worse than a wicked problem? Yes, it is perfectly possible: it is a wicked solution. That is, a solution that not only does nothing to solve the problem, but, actually, worsens it. Unfortunately, if you work in system dynamics, you soon learn that most complex systems are not only wicked, but suffer from wicked solutions (see, e.g. here).

This said, let's get to one of the most wicked problems I can think of: that of the world's food supply. I'll try to report here at least a little of what I learned at the recent conference on this subject, jointly held by FAO and the Italian Chapter of the System Dynamics Society. Two days of discussions held in Rome during a monster heat wave that put under heavy strain the air conditioning system of the conference room and made walking from there to one's hotel a task comparable to walking on an alien planet: it brought the distinct feeling that you needed a refrigerated space suit. But it was worth being there.

First of all, should we say that the world's food supply is a "problem"? Yes, if you note that about half of the world's human population is undernourished; if not really starving. And of the remaining half, a large fraction is not nourished right, because obesity and type II diabetes are rampant diseases - they said at the conference that if the trend continues, half of the world's population is going to suffer from diabetes.

So, if we have a problem, is it really "wicked"? Yes, it is, in the sense that finding a good solution is extremely difficult and the results are often the opposite than those intended at the beginning. The food supply system is a devilishly complex system and it involves a series of cross linked subsystems interacting with each other. Food production is one thing, but food supply is a completely different story, involving transportation, distribution, storage, refrigeration, financial factors, cultural factors and is affected by climate change, soil conservation, population, cultural factors...... and more, including the fact that people don't just eat "calories", they need to eat food; that is a balanced mix of nutrients. In such a system, everything you touch reverberates on everything else. It is a classic case of the concept known in biology as "you can't do just one thing."

Once you obtain even a vague glimpse of the complexity of the food supply system - as you can do in two days of full immersion in a conference - then you can also understand how poor and disingenuous often are the efforts to "solve the problem". The basic mistake that almost everyone does here (and not just in the case of the food supply system) is trying to linearize the system.

Linearizing a complex system means that you act on a single element of it, hoping that all the rest won't change as a consequence. It is the "look, it is simple" approach: favored by politicians (*). It goes like this, "look, it is simple: we just do this and the problem will be solved". What is meant with "this" varies with the situation; with the food system, it often involves some technological trick to raise the agricultural yields. In some quarters that involves the loud cry "let's go GMOs!" (genetically modified organisms).

Unfortunately, even assuming that agricultural yields can be increased in terms of calories produced using GMOs (possible, but only in industrialized agricultural systems), then the result is a cascade of effects which reverberate in the whole system; typically transforming a resilient rural production system into a fragile, partly industrialized, production system - to say nothing about the fact that these technologies often worsen the food's nutritional quality. And, assuming that it is possible to increase yields, how do you find the financial resources to build up the infrastructure needed to manage the increased agricultural yield? You need trucks, refrigerators, storage facilities, and more. Even if you can manage to upgrade all that, very often, the result is simply to make the system more vulnerable to external shocks such as increases in the cost of supplies such as fuels and fertilizers.

There are other egregious examples of how deeply flawed is the "'look, it is simple" strategy. One is the idea that we can solve the problem by getting rid of food waste. Great, but how exactly can you do that and how much would that cost? (**) And who would pay for the necessary upgrade of the whole distribution infrastructure? Another "look, it is simple" approach is 'if we all went vegetarian, there would be plenty of food for everyone'. In part, it is true, but it is not so simple, either. Again, there is a question of distribution and transportation, and the fact that rich westerners buy "green food" in their supermarkets has little impact on the situation of the poor in the rest of the world. And then, some kinds of "green" food are bulky and hence difficult to transport; also they spoil easily, and so you need refrigeration, and so on. Something similar holds for the "let's go local" strategy. How do you deal with the unavoidable fluctuations in local production? Once upon a time, these fluctuations were the cause of periodic famines which were accepted as a fact of life. Going back to that is not exactly a way to "solve the food supply problem."

A different way to tackle the problem is focussed on reducing the human population. But, also here, we often make the "look, it is simple" mistake. What do we know exactly on the mechanisms that generate overpopulation, and how do we intervene on them? Sometimes, proposers of this approach seem to think that all what we need to do is to drop condoms on poor countries (at least it is better than dropping bombs on them). But suppose that you can reduce population in non traumatic ways, then you intervene into a system where "population" means a complex mix of different social and economic niches: you have urban, peri-urban, and rural population; a population reduction may mean shifting people from one sector to the other, it may involve losing producing capabilities in the rural areas, or, on the contrary, reduced capabilities of financing production if you could lower population in urban areas. Again, population reduction, alone, is a linear approach that won't work as it is supposed to do, even if it could be implemented.

Facing the complexity of the system, listening to the experts discussing it, you get a chilling sensation that it is a system truly too difficult for human beings to grasp. You would have to be at the same time an expert in agriculture, in logistics, in nutrition, in finance, in population dynamics, and much more. One thing I noticed, as a modest expert in energy and fossil fuels, is how food experts normally don't realize that the availability of fossil fuels must necessarily go down in the near future. That will have enormous effects on agriculture: think of fertilizers, mechanization, transportation, refrigeration, and more. But I didn't see these effects taken into account in most models presented. Several researchers showed diagrams extrapolating current trends into the future as if oil production were to keep increasing for the rest of the century and more.

The same is true for climate change: I didn't see at the conference much being said about the extreme effects that rapid climate change could have on agriculture. It is understandable: we have good models telling us how temperatures will rise, and how that will affect some of the planet's subsystems (e.g. sea levels), but no models that could tell us how the agricultural system will react to shifting weather patterns, different temperatures, droughts or floods. Just think of how deeply agricultural yields in India are linked to the yearly monsoon pattern and you can only shiver at the thought of what might happen if climate change would affect that.

So, the impression I got from the conference is that nobody is really grasping the complexity of the problem; neither at the level of single persons, nor at the level of organizations. For instance, I never heard a crucial term used in world dynamics, which is "overshoot". That is, it is true that right now we can produce roughly enough food - measured in calories - for the current population. But for how long will we be able to do that? In several cases I could describe the approaches I have seen as trying to fix a mechanical watch using a hammer. Or to steer a transatlantic liner using a toothpick stuck into the propeller.

But there are also positive elements coming from the Rome conference. One is that the FAO, although a large, and sometimes clumsy, organization understands how system dynamics is a tool that could help a lot policy makers to do better in managing the food supply system. And, possibly, helping them device better ideas to "solve the food problem". That's more difficult than it seems: system dynamics is not for everyone and teaching it to bureaucrats is like teaching dogs to solve equations: it takes a lot of work and it doesn't work so well. Then, system dynamics practitioners are often victim of the "spaghetti diagram" syndrome, which consists in drawing complex models full of little arrows going from somewhere to somewhere else, and then watching the mess they created and nodding in a show of internal satisfaction. But it is also true that, at the conference, I saw a lot of good will among the various actors in the field to find a common language. This is a good thing, difficult, but promising.

In the end, what is the solution to the "food supply problem"? If you ask me, I would try to propose a concept: "in a complex system, there are neither problems, nor solutions. There is only change and adaptation." As a corollary, I could say that you can solve a problem (or try to) but you can't solve a change (not even try to). You can only adapt to change, hopefully in a non traumatic manner.

Seen in this sense, the best way to tackle the present food supply situation, is not to seek for impossible (wicked) solutions (e.g. GMOs) but to increase the resilience of the system. That involves working at the local level and interacting with all the actors working in the food supply system. It is a sensible approach. FAO is already following it and it can insure a reasonable supply even in the presence of the unavoidable shocks that are going to arrive as the result of climate change and energy supply problems. Can system dynamics help? Probably yes. Of course, there is a lot of work to do, but the Rome conference was a good start.


H/t: Stefano Armenia, Vanessa Armendariz, Olivio Argenti and all the organizers of the joint Sydic/FAO conference in Rome

Notes.

* Once you tackle the food problem, you can't ignore the "third world" situation. As a consequence, the conference was not just among Westerners and the debate took a wider aspect that also involved different ways of seeing the world. One particularly interesting discussion I had was with a Mexican researcher. According to her opinion, "linearizing" complex problems is a typical (and rather wicked) characteristic of the Western way of thinking. She countered this linear vision with the "circular" approach that, according to her, is typical of ancient Meso-American cultures, such as the Maya and others. That approach, she said, could help a lot the world to tackle wicked problems without worsening them. I just report this opinion; personally I don't have sufficient knowledge to judge it. However, it seems true to me that there is something wicked in the way Western thought tends to mold everything and everyone on its own image.

** In the food system, the idea that "look, it is simple: just let's get rid of waste" is exactly parallel to the "zero waste" approach for urban and industrial waste. I have some experience in this field, and I can tell you that, the way it is often proposed, the "zero waste" idea simply can't work. It involves high costs and it just makes the system more and more fragile and vulnerable to shocks. That doesn't mean that waste is unavoidable; not at all. If you can't build up a "zero waste" industrial system, you can build up subsystems that will process and eliminate that waste. These subsystems, however, cannot work using the same logic of the standard industrial system; they have to be tailored to operate on low yield resources. In practice, it is the "participatory management" approach, (see, e.g., the work of Prof. Gutberlet). It can be done with urban waste, but also with food waste and it is another way to increase the resilience of the system.






Who

Ugo Bardi is a member of the Club of Rome, faculty member of the University of Florence, and the author of "Extracted" (Chelsea Green 2014), "The Seneca Effect" (Springer 2017), and Before the Collapse (Springer 2019)