Ecological Rationality and the Probabilistic Revoltuion: Simple Heuristics that can make Us Smart

(Photo: Solar CITIES team leader Hana Fathy and colleague Adham Fawzy, a teacher at the Zabaleen school, sit in front of their recently completed "solar-hot-water-animal-shelter" system on Adham's roof. Behind them to the right is the previous goat shelter.)

(Photo: Solar CITIES team leader Hana Fathy, seeking shade beneath the animal shelter's solar roof, smiles as he tests the hot water from the SOLAR CITIES hybrid water/shelter system, which provides rooftop cold and hot water storage for giving water to the animals and for cleaning the enclosures of the goats, chickens, rabbits and ducks that live on this urban agriculture space, provides essential cool shading for the animals so they don't overheat or dehydrate, as well as providing domestic hot water for the household to bathe and shower and wash clothes in, and cold water storage to get through days when the city water supply is cut).

In "Simple Heuristics that Make Us Smart", Gerd Gigerenzer and Peter M. Todd talk about "the demise of the dream of certainty and the rise of a calculus of uncertainty" that precipated "the probabilistic revolution". Their book goes on to address a second revolution that can be considered an outgrowth or consequence of probability theory -- the notion of "bounded rationality" (which they call "ecological rationality -- rationality that is defined by its fit with reality") The proposition their book asks us to contemplate is that we replace "the image of an omniscient mind computing intricate probabilities and utilities with that of a bounded mind reaching into an adaptive toolbox filled with fast and frugal heuristics."

As I read the first chapter of their book I am reminded of my recent experience in the slums and informal areas of Cairo clambering onto rooftops with my Solar CITIES team and local stakeholders to see whether and where we can build solar hot water systems. I noticed that many people I work with are quickly overwhelmed by the number of variables the non-uniform environments in the "incremental housing sector" present, and that we need to develop decision makers who, to quote Gigerenzer and Todd, "are not paralyzed by the need for vast amounts of knowledge or extensive computational power".

Such decision makers, I believe, are made, not born. I believe this because in the emergent field of "guerilla solar" (informal installations of renewable energy systems by "non-experts") I have made myself into such a decision maker. Where others look at the complex roof topography of inner city Cairo's urban poor and get stymied, I see a landscape of unique opportunities. Often my small team will call me in to help them make the critical decision for them and will marvel at the confidence with which I make decisions for which there is no precedent. I often marvel at myself! When they ask me, "are you sure?" I will say, "absolutely not, but I've got a good hunch that this will work, and a better hunch that the risk is low, so I say let's give it a go!". Then I step back and wonder how I arrived at such a conclusion and am able to state my awareness of its UNCERTAINTY with a confidence and aplomb that makes others decide to go along with my suggestion!

I am realizing now that my broad experience and years of trial and error have given me my own sets of "fast and frugal heuristics" and a neat adaptive toolbox.

Of course we expect that of so-called "experts", and there is an assumption that my pending Ph.D. title and my educational history (Harvard, UCLA) give me leverage over uncertainty (the assumption being that I have considered many of the options in some kind of exhaustive search for competency and truth). But inside my head, exploring the mechanics of my heuristic toolkit, I am convinced that the way I make decisions can be just as well taught to neophytes so that they can overcome their paralysis is decision making for development initiatives that are unfamiliar.

I think we need such heuristics because there are very few true "experts" in the field of sustainable development, and those who do exist usually cost far too much for the local NGOs and poor communities who need them most. Even expert "volunteers" usually demand to be housed and fed in the more expensive areas outside the poverty stricken areas that need the most help, and cannot stay the amount of time that is truly necessary to get uncommon initiatives, such as community led (as opposed to institution led) renewable energy projects and integrated industrial ecology projects off the ground.

My experiences working on sustainable development projects in Indonesia, Latin America, Inner City Los Angeles and the Middle East/North Africa over the past 20 years tell me that we are going to have to resign ourselves to the fact that most of the people who will need to lead and run such projects will have to be people with very little experience and most likely little confidence in the outcome of what many in the community see as "exogenous" or "exotic" enterprises.

But should we be able to identify and codify some of the nuances of what Gigerenzer and Todd call (and title their book) "Simple Heuristics That Make Us Smart" we might be able to give enthusiastic local stakeholders a much better chance at making changes that stick and sustain.

An example is warranted here: We had promised Adham Fawzy, one of the Zabaleen recycling school's teachers, that we would build a solar hot water system on his roof to compensate him for all the effort and time he volunteered when we were developing our "do it yourself domestic hot water systems". The problem we faced when we got up on his roof to determine its suitability, was that the neighbor's wall cast a shadow on the lower half of the panels at the only location that received continuous sunlight during the required hours of 10 and 4. The question then became "is it worth the extra labor and investment to raise the panels and thus both the hot and cold water tanks an extra meter to get them into continuous sunlight, or should we abandon this roof and seek a better candidate?"

We could have asked Adham if his family's marginal willingness to pay for the convenience of solar hot water was equal to the cost of the extra materials and labor, but we had already determined that average WTP was less that 100 LE (20 dollars) for the community as a whole, and this would never cover what we needed.

An economic heuristic (the one most people tend to apply) would have led to abandonment of the project. In fact most development initiatives, for all the sophistication of their models, basically revolve around a simple economic heuristic -- let us say a "simplistic" economic heuristic -- that ignores long term costs and externalities. People simply figure "nobody is willing to pay for solar hot water in these communities, so the project is unsustainable and thus it will fail."

Another issue that plagued Adham's family's roof and made it an unlikely place for a solar installation was that it was filled with animals. Adham pointed to all the rabbits, ducks, chickens, and goats and said "the goats will undoubtedly climb up and jump on the solar panels, breaking the glass. So the system won't last."

Untroubled (for reason's I can't quite pin down) I suggested a different heuristic. I asked Adham's family what they used the roof for, that is, what the roof's real value was to the family. Adham's father came up and said "we get extra income from the animals and they are like an insurance policy for us" (this is a truth of many poor families in cities who migrated from the rural countryside, and hearkens back to Chayanov's theories of the peasant economy).

So I asked, "would it be an advantage to have a supply of running water on the roof, instead of having to carry water up from the ground floor, so you could nourish the animals conveniently, particularly in the hot summers?" He said, "yes, we worry the animals can die of thirst in the hot summer." I asked, "does the city often cut your water supply in the summer?" He replied, "yes, and it makes things very bad for us as well as the animals."
"So would a roof top cold water storage tank as well as supply be helpful to you"
The answer was affirmative.
Then the father smiled as if a light had gone on in his head. He said, pointing to the two 180x80cm solar hot water panels we had been trying to find a placement for, "I wonder ("ya rait") if we couldn't use these as the roof of an animal shelter. We could raise them up on a brick wall a meter off the ground so they could get full sun, and that wall would prevent the goats from jumping on them. At the same time, we could build a brick box to hold the water tanks and that could be both the back wall of the goat shelter and the inside could be a rabbit hutch."
Once he had articulated this integrated vision of "solar-hot-water-system-as-animal-shelter-and-watering-device" the rest of the assembled friends and family members began to offer ideas of support:
"In the summer, the solar collectors would not only keep the sun off the animals underneath, but would probably cool them too, because the heat would be transferred into the water instead of to the air beneath the collectors, right? Usually, when we make a tin roof over the animals it gives them shade but the whole area heats up. But if the heat is moving with the water up into the storage tank..."

The frugal heuristic that liberated the community and led to acceptance of the solar project (and willingness to invest in it and maintain/sustain it) was, "if there is an alternative use to which any new system can be put that is within the domain of experience of the local stakeholders and satisfied a culturally established need, make that link and then see how the new system can add value to what is already part of the incremental housing plan of the local people."

Emboldened by this and other successes in the Zabaleen community, we took our idea to a meeting of the local environmental NGOs and community members at the Aga Khan Trust for Culture in Darb El Ahmar, where the discussion centered on why the rooftop gardening projects had all failed. After two hours of very fruitful discussion, we started applying this simpler heuristic ("don't dwell on the multiple factors that led to failure, and what doesn't work; figure out instead how rooftop gardens can integrate with the real and perceived needs of the stakeholders, and how they might satisfy multiple needs if integrated with a larger ecological system". This is the "ecological rationality" approach championed by Gigerenzer and Todd -- "rationality that is defined by its fit with reality".

(Photo: Samiya and Sou'ad, members of the Darb Shuglan/Darb El Ahmar local environmental committee, Samiya's children and Solar CITIES coordinator Mahmoud Dardir observe the integrated solar hot water/rooftop gardening system -- a combined initiative between Solar CITIES and the AKTC environmental NGO that provides rooftop stored hot and cold water for domestic household use and urban agriculture at the same time.)



What most poor people don't want is some pie in the sky idea that appeals to would-be do-gooders trying to "fight global warming" and "beautify the slum" according to the needs and standards of the well-to-do. The reality we must fit into is that of the daily struggle to make ends meet in the ghetto, and to live with uncertainty every day. Hell, the probabilistic revolution doesn't need to take place among those who live with uncertainty -- they already have the fast and frugal heuristics they need to eke out a living in their marginal environment. If those heuristics lead them to reject pretty but economically unfeasible things such as solar panels and roof top gardens, then we must accept the compromise solutions their heuristic has provided, regardless of what our "sophisticated" calculations of the long term rationality of their practices might be. We can't judge people's behavior by our calculations of outcomes that might occur if they lived in OUR MODEL of the world. As Bardhan and Udry state in "Development Microeconomics" (p. 5)

"In view of the empirical anomalies and the internal contradictions of the model of the hyper-rational economic man, many economists follow Simon (1957) in assuming bounded rationality with full recognition of the costs of observation, communication, and computation. Others go beyond this and admit that economic agents' behaviour is sometimes difficult or impossible to rationalize in terms of any well-defined deductive model; they fall back upon inductive means of reasoning and learning. People are supposed to have working hypotheses about the problems they are dealing with, and it is assumed that in their learning process they constantly update and adapt. Whether this process ultimately converges to rationality depends, even in the favourable case of the same situation repeated sufficiently often, on the characteristics of the decision problem or the game... Development economics if full of examples of how apparently irrational behaviour may be successfully explained as an outcome of more complex exercises in rationality, particularly with deeper probes into the nature of the feasibility constraints or the preference patterns."

The authors also point out (p. 4) that when dealing with "choice under uncertainty" people "are more averse to losses than they are attracted to same-size gains". Such "loss-aversion" explains a lot of the reluctance of the poor to invest in "new" ideas.

Keeping these ideas in mind, what we concluded in our multi-stake-holder meeting was that if we were to COMBINE roof top gardening with roof top water storage and "simultaneous-solar-heating-and-sun-shelter" installations (provides cold and hot water all year round; keeps the roof cooler in the summer) and see it all as part of a gradual move toward energy/food/economic independence (combined with roof top biogas production following the Indian ARTI model for utilizing and creating compost with urban waste , with small scale wind energy and photovoltaic technology for disaster preparedness and insurance against rising energy and food costs), we could generate more ideas and greater acceptance and stakeholder commitment. We weren't trying any more to introduce rooftop gardens or solar energy. No. We were trying to help a community get on its feet and get sustainable "by any means necessary".

Where my confidence for making decisions these days in the face of uncertainty in Cairo comes from is that I DO NOT see myself as a proselytizer of solar energy or of earthy crunchy environmental stuff. I DO see myself as an overall solutions provider, and as somebody who no longer sees things as they are labelled, but as they can be utilized.

And this is an expertise that the Zabaleen and the local craftspeople of Cairo have in abundance. This is a trait that exists in most communities who are faced with survival under conditions of poverty. There is usually an abundance of "satisficers" and "optimizers" who do the best they can with what they have. They are already using fast and frugal simple heuristics that make them smart. What they often lack is merely the link. And since most development experts look at the technologies we bring in as "pre-purposed objects" instead of mere pieces to an evolving puzzle, I believe that many development initiatives fail.

Right now, according to my preliminary dissertation survey results, very few people in urban poverty in Cairo are going to demand "solar hot water" per se. They are adept at heating their water using the stove and have adjusted to the inconvenience and are unwilling to pay more than they are paying now (their own labor is in greater supply than their income, and is discounted on a daily level). While they almost all recognize the long term advantages of SHW, they cannot justify the expense. So an attempt to introduce unsubsidized solar energy to the slums WILL fail if that is what we say we want to deliver.

But if we are delivering a concept of integrated industrial ecology -- if we are introducing flexible, modular, changeable solution sets to a multiplicity of local concerns ("all this and world war too?!" to quote the album title) -- then we can build confidence in and familiarity with each new innovative idea, and impart this heuristic -- again, already available at the local level -- to the task of conceiving environmental technology endeavors.

The brilliant Mauro Cherubini at the EPFL in Switzerland challenged me to come up with ways of conducting lasting and impactful education for sustainable development (much in the way that the Cooper Hewitt Museum challenged designers to "design for the other 90%") and I haven't answered him yet because I'm still kicking the idea around. But this is one possible salvo that comes from reading the literature and combining it with my recent personal experiences in Cairo:

Design an educational program that forces us to conceive of multiple uses for every object, and to look for novel ways to put them together to solve other problems. Get garbage recyclers, survivors of poverty, craftspeople and artists to be participants in every development training team to share their unique perspectives. Have academics and students play video games like "Return to Mysterious Island" where the whole idea of the game is figuring out ways to combine and uncombine the objects you find on the island to solve problems. Most importantly for THIS GENERATION OF LEARNERS: Design a game in which you have to do the same sort of object-deconstruction-reconstruction-problem-solving exercises but in REAL (though virtually represented!) locations in the poverty stricken areas of the world using simulations of real objects and technologies and materials (see our "SOLAR CITIES ARG" for an example) and use digital earth technology to make such simulations available to a global connected group of "players"/"problem solvers"/"engineers"/"policy-makers"/"stakeholders" so they can combine their collective intelligence and get real solutions out there, even if they can't physically visit the locations or volunteer their time.

It may be that the simple heuristics we need to make us smart about sustainable development can be rapidly learned in such contexts of mutual respect and playful collaboration, and can lead to more rapid deployment of solution sets. The probabilistic revolution makes it foolish to say anything with any dream of certainty. But such an approach may, at the very least, turn out to be more probable than our previous approaches to development!