What if we tried to solve the wrong problem?

One of the things I've learned in computer science is that there are often cases where we think we know what the problem is and we launch off trying to solve it, only to find later that we didn't actually understand the problem in the first place.

This is the problem with problems. And this is why too often we might find solutions to what we thought was the problem only to discover we haven't solved the problem at all.

Does this apply to non-computer based problems? What about social problems? I suspect you already have a notion that this happens all the time. Think of the various 'war on...' attacks on poverty, drugs, cancer, and now terrorism. These are direct assaults on what our leaders perceive as the 'problems'. The solution is to focus money and people and cross your fingers. But as is becoming clear, the war metaphor doesn't really work when you aren't fighting an opposing, uniformed army.

The corollary to this is that sometimes the solution to the 'real' problem is staring you in the face but you don't recognize it because you haven't actually understood the problem. The danger here, of course, is that you expend energy and resources trying to solve what you think is the problem and end up losing the opportunity to solve the actual problem.

I suspect that is the case with global warming (and peak oil, over-population, and many more of our global challenges). Part of the problem with recognizing the real problem is that we think we already know what the solution will look like so we settle quickly on the problem description that seems to match. Here is the problem: cataclysmic climate change as a result of anthropogenic greenhouse gases causing an increased thermal forcing of the atmosphere and oceans. Solution: drastically cut the emissions of greenhouse gases. But then how do you do that since our entire economic structure is based on burning fossil fuels that emit carbon dioxide! At this point it looks like you either destroy the economy or destroy the earth! Great choice that.

So how do we escape this conundrum? Well clearly we have to replace our carbon-based energy production with something benign, something long-term sustainable. Solar and wind energy would work, maybe. So would nuclear, maybe. Maybe a combination of all of these. How much time do we have to convert? Whoops! Not much. OK, that calls for a war on energy, err, well maybe a Manhattan Project-like effort to build the infrastructure for alternative energy production. OK, where can we get that much dough? At this time of night? Whoops, we already leveraged ourselves with credit. The way we were borrowing assumed that energy production would always be there (money is really just a symbolic form of energy).

This tangle of sub-problems is beginning to reek of intractability, a sure sign you are trying to solve the wrong problem. That is, unless you really believe the problem IS intractable, in which case you should just lay down and wait to die! Well I'm not there so let's consider other options. If we are trying to solve the wrong problem, why?

It starts with the initial assumption that the problem is framed as:

1) We want to keep our way of living the way it is. We always imagined things would be the same in perpetuity. Our kids would have what we have or even more.
2) Climate change presents a threat to that way of living.
3) Therefore we need to do something about global warming in order to keep everything exactly as it is now.

The hitch in this logic is the premise. We, as a species, are so used to the idea that we can alter our local environment to suit our needs that we tend to think that will always be the case, even on a global scale. We refuse to recognize limits, both physical reality (like the Second Law of Thermodynamics as it applies in systems far from equilibrium — stop the flow of energy and see how quickly such systems return to equilibrium) and psychological realities. The thing about change is, we don't always get to choose. Where is it written in stone tablets that the way we live is a given and that it will always be so?

What if the real problem isn't how are we going to save the population and go on with our happy consumption? What if it isn't even how are we going to save the species after climate change causes a massive die-off? What if the real problem is how are we going to save the genus?!

Suppose it is already too late to stop global warming from irreversibly altering our more-or-less benign climate. In fact many climate scientists, including James Hansen and maverick James Lovelock think this is the case. Suppose it is too late to prevent catastrophic decimation of large numbers of the population. Quite a few scientists now think this is the case as well. Suppose, in fact, that we are headed for a worst-case scenario where we choose to keep on spewing greenhouse gases and a number of positive feedback mechanisms kick into play catapulting the climate into unknown territory. Do you really believe humans will respond gracefully? Do you really believe we humans will then suddenly start making wise choices?

Let's examine the hierarchy of problem characterizations and solution proposals. We need to characterize the requirements for a solution and we need to establish the constraints under which those solutions must be made. The big constraint in all cases is time. How rapidly do we need to act in order to produce a solution? Suppose that we consider the worst-case IPCC and ASPO models of global warming and peak oil respectively along with population growth projections from the UN. The combination of these three scenarios paints a very scary picture. Essentially mankind has about 20 years time before the complete collapse of industrial societies world-wide, with wide-spread violence, starvation, disease, thirst, and other horrors of living in a world falling apart. Twenty years. That is what we call a tight constraint. [Just a note about this time scale and what for most of you reading this will be a reaction of "No Way!" The most insidious aspect of a collapse is that it is precipitous and you are partying-on right to the moment that the bottom falls out. One minute everything looks hunky dory, the next catastrophe. You may not appreciate the math involved here but the concept of a precipitous catastrophe has a solid grounding, due to René Thom, in chaos theory.]

For starters lets consider solving 'the' problem as saving humanity and the ecosystems of earth AND saving our economic system so we can go on consuming in comfort — what has been characterized as business as usual. We want the living standards (whatever that means) of developed countries to remain what they are today and we want developing countries to attain similar (equitable) living standards. After all, isn't that what we mean by 'developing'. This scenario requires what? We literally have to stop putting out CO₂ today. We have to find some alternative source of energy today. In fact, we have to start taking CO₂ out of the atmosphere today. Oh yes, and we have to stop expanding our footprint in the ecology so as to not cause any further damage to the ecos or cause any more species to go extinct.

And the likelihood that any and all of that is going to happen is what????

In fact this is exactly the problem that most people think we have to solve, including Al Gore and most politicians who at least accept the reality of global warming. There are still many that don't even see any problems. For them business as usual is just that. Nothing wrong here, nothing to see, move on folks. But for those who do see the threats but believe we can solve the problem with technology and the good old market (with a little subsidization of alternative energy thrown in to sweeten the deal), I claim they are trying to solve the wrong problem. Furthermore, what they are trying to do is what we call intractable. It can't be solved because the constraints so overpower the requirements that there is NO feasible solution. Period. [Note: many of you will simply not accept this, but since this blog is called Question Everything I challenge you to ask yourself why? Why do you reject it? Is it because you have extensive special knowledge that negates my argument? Or is it because you don't want to believe such a thing? Be honest with yourself.]

So what would be the next characterization of a problem we could try to solve? Suppose we tackle this: save humanity and ecosystems but recognize that we will have to sacrifice comforts and consumption. Also the developing world will not get to an equitable level unless we in the developed world are willing to give up a great deal and share. By doing so we will reduce carbon output, reduce consumption of fuels and simply learn to adapt to climate change.

What is wrong with this picture? Aside from the fact that it might take catastrophe just to wake people up to the fact that sacrifice is needed (and then it may be too late). How quickly do you think people will be able to substantially reduce their use of fuels? I'm betting it will take a lot longer than 20 years. And then there is the part about adapting to climate change. I suspect it will actually take more energy, not less, to do the adapting. Adapting means building new infrastructure. It means moving cities inland from the coasts. It means reinforcing structures against higher winds and more frequent tornados. It means desalinizing sea water to drink (takes considerable energy to do that). In a world of diminishing fossil fuels and a slow ramp-up of alternatives, where will the energy come from?

And then there is still the growing population. Even if per capita consumption goes down, the increase in population could swamp the gains. Right now folks are counting on the so-called demographic effect in developed countries, where the birth rates decline due to reaching a level of wealth and educating women. Well, what happens when we revert to the land? The reason people in underdeveloped countries kick out so many kids is, in part, that they need the extra hands to care for the parents in old age. People all over the world and long before civilization have known where babies come from. They also know a bunch of tricks to keep the birth rate down naturally. So it isn't ignorance or lack of wealth per se that causes people to have more kids. If we all end up doing manual labor for a living, you can bet people will start having more kids.

Solving this problem would be really, really hard (perhaps really intractable also). I have great difficulty imagining people all over the world going along with this solution. And even if they did, would it really work? The constraints still far overpower the requirements.

Which leads to solving another problem: saving the species. As inhumane as it sounds we might actually be faced with exactly this problem. We could very well have gone past the point of no return that prevents us from solving either of the above. We may actually be on the road to extinction now.

Saving the species is a very different problem. We give up all hope of saving the majority of living humans against the kinds of calamities envisioned. We admit that it can't be done regretfully but honestly. Then what? Rather than give up entirely we set about as humanely as possible reducing the population at a drastic rate. If as many scientists believe, we are in over-shoot with respect to sustainable population numbers (estimates for sustainable populations range from 0.5 to 2 billion), then we would need to stop reproducing almost entirely. At the same time we need to outfit pockets of civilization with enough alternative and sustainable energy production as we can muster (a fine point here is that it takes a lot of energy to manufacture energy conversion equipment like solar cells and wind turbines). These pockets need to be situated in the most favorable locales with respect to rainfall and arable land.

Again it is hard to picture people doing this voluntarily. Unless of course the calamities have made it eminently clear that there is no other choice. My fear is that the timing constraints will have by then obviated the whole plan and it will be too late. Homo sapiens will go extinct because we would not be able to act fast enough to implement the plan relative to the rate of calamitous events.

It could be even worse than this. The problem we may be faced with solving could very well be: Save the genus so that some day far in the future evolution will have at least some raw material for the further evolution of sentience and, I hope, wisdom. It's hard to even imagine what the solution to this problem would involve. But the fact that it is unthinkable at present (because we have never had to face anything like it before) doesn't mean it can't be the problem we are faced with. Like the threat of nuclear winter — unthinkable in the 1940s and 50s — maybe we should at least consider the possibility.

If we do not recognize the real problem that we need to solve, we will never get a second chance. And how we go about solving the real problem will depend on recognizing it for what it is. You wouldn't treat stomach cancer with an aspirin. But you might if all you know about is the complaint of feeling pain.

So before we waste any more time treating climate change with aspirin, I would recommend that we make sure we know what the problem we are faced with really is, and then we can go about designing a solution.

Comments

[Trinifar]

George, your last paragraph sounds too much like "let's not do anything (at least on climate change) until we get some basic agreement just what all the fundamental problems of the world are and what kind of world we want." I'm sure that's not what you intend. Can you clarify?

[George Mobus]

Trinifar,

Thanks for the concern. You are right, that isn't what I meant. The key word in the paragraph is 'aspirin', meaning that to date our approaches to climate change have been wholly inadequate compared with the scale of the problem. To wit: the current mechanisms being discussed for reducing our carbon emissions are weak (cap-and-trade, even a carbon tax) compared with the sheer volume of carbon dioxide being vented, and the rate. I suspect the latter will so overpower the rate at which these economic forces begin to respond that the so-called tipping point will have come and gone before anyone realizes it. I covered this relative rate phenomenon in an earlier posting.

In a similar vein, on a Dot Earth comment I indicated that we will need to proceed along many paths simultaneously. There the solution being discussed involved the equivalent to radical surgery being performed on the patient before knowing what the disease actually is. So at one extreme you have aspirin at the other you have surgery. Both are options, but may or may not cure the patient.

My point is that we have done a lot of science to indicate that a problem exists. Now we need to do a lot more science to guide our choices in actions. Instead we are running off with wild proposals as if we actually did understand what to do. If we choose one of these (say the massive solar farm in Arizona covered in Scientific American) and launch into it without reasonably understanding the consequences (some assurance that we will in fact reduce our carbon emissions adequately to address global warming) we will have expended considerable resources on something that could very well not achieve the ends we thought.

Note that solutions like the SciAm article are developed in relative isolation from the other related problems that exist. This means there are certainly factors that have not been taken into account in formulating the solution. For example the solar farm proposal did not take into account risk analysis regarding extreme weather damage or terrorist attacks. If you concentrate a significant portion of your generating capacity in one location, you are surely asking for trouble. And that is just one example of the non-thought that went into that proposal.

Always I advocate a systems approach to these global issues. The interactions between global warming, peak oil (energy), population, and human psychology (especially under stress) all need to be taken into consideration.

The issue of feasibility, vis-a-vis telling us what the nature of the problem really is, also should be considered. Putting all of our efforts into solving an intractable problem has moral dimensions that should be explored. That was my point re: are we saving us and the way we live, are we saving our species, or are we going to try to save our genus? If we focus on the first and fail we may miss the opportunity for one of the others. Perhaps we will choose the former and do so for what we believe are justifiable moral reasons. In fact I suspect that is precisely the default position of everyone right now.

It just seems prudent to me to ask the questions.

[Trinifar]

Thanks for the clarification. And, naturally, I agree "Putting all of our efforts into solving an intractable problem has moral dimensions that should be explored." That's important yet tough to do (to accumlate the funding and take the time and effort in a quality way) when we're told the tipping point is nearby (80% reduction of GHG by 2020 or 2050 depending on who you listen to). We're left with taking the shotgun/scattershot approach of doing all sorts of things without enough information to know about unforseen consequences while simultaneously pursuing more knowledge.

As an engineer/business-developer I am caught up in what I think is the realistic outlook that the size of the problem is beyond our capabilities in terms of timeframe, scale of investment, change of capital allocation, pace of political change, etc. We just can't meet the deadlines under any realistic scenario. At the same time, I am constantly reminded by many people with a different point of view (i.e. not bound by engineering and real-world business/political constraints) of the chaotic nature of the system. Those people advocate doing almost anything feasible (raising CAFE standards, carbon taxes, LEED certification-based zoning, etc.) at least does no harm and contributes to mitigation -- and may change the playing surface enough to allow for a more thoughtful, deterministic approach.

Damn, it's a hard problem, NP-hard.