An alternative view (Holling 2000; Gunderson and Holling 2001) suggests that the complexity of living systems of people and nature emerges not from a random association of a large number of interacting factors rather from a smaller number of controlling processes. These systems are self-organized, and a small set of critical processes create and maintain this self-organization. . .

These processes establish a persistent template upon which a host of other variables exercise their influence. Such “subsidiary” variables or factors can be interesting, relevant, and important, but they exist at the whim of the critical controlling factors or variables. If sustainability means anything, it has to do with the small set of critical self-organized variables and the transformations that can occur in them during the evolutionary process of societal development. . .

The view presented here argues that there is a requisite level of simplicity behind the complexity that, if identified, can lead to an understanding that is rigorously developed but can be communicated lucidly. It holds that if you cannot explain or describe the issue of concern using at least a handful of causes, then your understanding is too simple. If you require many more than a handful of causes, then your understanding is unnecessarily complex. . .

The first requirement is to begin to integrate the essence of ecological, economic, and social science theory and to do so with the goal of being, in Einstein’s words, “as simple as possible but no simpler.” The purpose of this paper is to summarize a theoretical framework and process for understanding complex systems. . .

To be useful, such a framework and process must satisfy the following criteria:

• Be “as simple as possible but no simpler” than is required for understanding and communication.
• Be dynamic and prescriptive, not static and descriptive. Monitoring of the present and past is static unless it connects to policies and actions and to the evaluation of different futures.
• Embrace uncertainty and unpredictability. Surprise and structural change are inevitable in systemsof people and nature.

I doubt it. The idea of the Panarchy, nested levels that incite transformations from fast cycle change from below, moderated by slower change from above, is useful. But I doubt that it can be usefully modelled as described. The sticking point is the idea that a handful of factors is sufficient to predict future behaviors. It works often, but not always, and that is a recipe for disaster.

Eliezer Yudkowsky posted today about hindsight - learning from experience - and being sufficiently rational to cut your loses when things turn against you rather than clinging to remote hopes. He used an example of Long-Term Capital Management.

While LTCM raked in giant profits over its first three years, in 1998 the inefficiencies that LTCM were exploiting had started to vanish - other people knew about the trick, so it stopped working.

LTCM refused to lose hope. Addicted to 40% annual returns, they borrowed more and more leverage to exploit tinier and tinier margins. When everything started to go wrong for LTCM, they had equity of $4.72 billion, leverage of $124.5 billion, and derivative positions of $1.25 trillion.

I've used LTCM to make a different point, one that relates to Panarchy. Yudkowsky's post reminded me of it. LTCM was weakened by the things Yudkowsky notes, but it was wrecked by a "black swan".

The scheme finally unraveled in August and September 1998 when the Russian government defaulted on their government bonds (GKOs). Panicked investors sold Japanese and European bonds to buy U.S. treasury bonds. The profits that were supposed to occur as the value of these bonds converged became huge losses as the value of the bonds diverged. By the end of August the fund had lost $1.85 billion in capital.

As Peter Fisher put it:

If a random bolt of lightning hits you when you're standing in the middle of the field, that feels like a random event. But if your business is to stand in random fields during lightning storms, then you should anticipate, perhaps a little more robustly, the risks you're taking on.

LTCM was in the risk business. Though it was a hedge fund seeking to limit risk exposure, its size and duration made it ever more probable that lightning would strike. Nassim Nicholas Taleb sees the problem as "Hindsight Bias":

The equivalent of Google, where someone just takes over everything, would have been impossible to witness in the Pleistocene. These are more and more prevalent in a world where the bulk of the random variables are socio-informational with low physical limitations. That type of randomness is close to impossible to model since a single observation of large impact, what I called a Black Swan, can destroy the entire inference.

This is not just a logical statement: these happens routinely. In spite of all the mathematical sophistication, we're not getting anything tangible except the knowledge that we do not know much about these "tail" properties. And since our world is more and more dominated by these large deviations that are not possible to model properly, we understand less and less of what's going on.

While I find merit in the Panarchy concept, the idea that systems can usefully be modelled and used to determine policy seems doomed by hindsight bias; there is an insufficiently robust appreciation for black swan events, though they are growing ever more probable as time passes and socio-informational variables become more dominant. Such models can serve as thought experiments, but their prescriptions must be viewed as vague hopes which we view with suspicion, especially when the prescription calls for change, and so incurs risk.

It's much like the old, failed systems thinking of Meadows et. al. in their Limits to Growth / Club Of Rome phase, but is tarted up a bit with the vocabulary of complex adaptive systems. It goes too far, seeking to subvert the insights of CAS in service of the hope of controlled intervention. As Meadows belatedly recognized:

People who are raised in the industrial world and who get enthused about systems thinking are likely to make a terrible mistake. They are likely to assume that here, in systems analysis, in interconnection and complication, in the power of the computer, here at last, is the key to prediction and control. This mistake is likely because the mindset of the industrial world assumes that there is a key to prediction and control. . .

But self-organizing, nonlinear, feedback systems are inherently unpredictable. They are not controllable. They are understandable only in the most general way.

I've said much the same for a long time, and a challenge often raised to this view is that it seems to offer no hope at all, that we are doomed to being eternal victims of the whims of complexity, and that nothing we can do can be expected to be better than passively drifting in the flow. It's the "we" problem again, the assumption that there exists some entity that can act usefully. Usually it means a collective of some sort, a government for example, or any group acting as a single entity. There is no "we", it's an illusion of language. The way to intervene in complex systems is to diligently work to gain information, and then communicate that information to the "social mind". You have no way of predicting what will result, but feeding the mind with good information gives it the best shot at rational behavior. Anything less - or more - any spin, deception, omission or attempt at control will degrade the mind.

Tyler points to Nassim Nicholas Taleb's book, The Black Swan: The Impact of the Highly Improbable. The book description may make some of the previous comment more intelligible.

A black swan is a highly improbable event with three principal characteristics: It is unpredictable; it carries a massive impact; and, after the fact, we concoct an explanation that makes it appear less random, and more predictable, than it was. The astonishing success of Google was a black swan; so was 9/11. For Nassim Nicholas Taleb, black swans underlie almost everything about our world, from the rise of religions to events in our own personal lives.

Why do we not acknowledge the phenomenon of black swans until after they occur? Part of the answer, according to Taleb, is that humans are hardwired to learn specifics when they should be focused on generalities. We concentrate on things we already know and time and time again fail to take into consideration what we don’t know. We are, therefore, unable to truly estimate opportunities, too vulnerable to the impulse to simplify, narrate, and categorize, and not open enough to rewarding those who can imagine the “impossible.”