BIG BANGS

Here's a little philosophy. Hang with me, I think the topic is worth discussing even if you disagree.

When I was studying control theory at the USAF Academy (it was my focus as part of an Astronautical Engineering degree -- yes, you could say I am a rocket scientist ;-> to the extent that means anything), we learned the difference between dynamically stable and unstable systems. Dynamically stable systems are, by design, built with intrinsic dampening forces that return them to a steady state. For example, aircraft of the classic design like to fly. You actually have to work hard to get them to do something uncontrollable.

However, this stability costs you a measure of performance. To increase performance you have to ADD instability to the system's design. This means that a high performance aircraft with a large percentage of instability built into its design occasionally wants to careen into oblivion -- the feedback from the system's interaction with the environment can create uncontrollable loops that tend towards infinity (which of course means catastrophic failure). To compensate for this, dynamically unstable systems have computer augmented control systems that dampen these feedback loops. For example, a plane of this type of design has computers (with double back-up) that constantly compensate for instability by moving control surfaces (at a much faster rate than the pilot can). Without compensation, a plane like the F-16 will go catastrophic in 3 seconds. With some of the forward swept wing designs, the time to instability is measured in fractions of seconds.

If we look at today's global environment we see a relatively high performance system driven by real-time global markets and rapid technological progress. Its performance explains why it is spreading so quickly. However, it is also moderately unstable. In our drive towards higher levels of performance we pursued a path of rampant global interconnectivity that has quickly outpaced our ability to dampen excess. The old dampening functions of borders, distance, government, etc are quickly fading. The result is a system vulnerable to rogue feedback. Even a small amount of it can cause global reverberations. Worse, there are people actively working on ways to introduce this rogue feedback. Iraq is a great demonstration of our inability to dampen excess in the face of active opposition (notice how our goals have drifted from building an allied democracy to stopping civil war).

The long-term solution is to build more stability into the system. The best approach I can think of is a highly interconnected but fundamentally decentralized system (most of the benefits of interconnectivity but with lots of local control). Unfortunately, we are far from realizing that goal, since our current view of the world is based on old models.

For example, instead of building resilience into the system, we have embarked on a path of introducing more rogue feedback into the system (the invasion of Iraq seen as a "big bang" in the Middle East). This is based on the belief that Fukuyama's "End of History," where we all live in capitalist democracies, is inexorable. It's not. There isn't any guarantee that our current system is the inevitable result of history and that all dampening forces are driving us to this result. Also, the system isn't tolerant of these shocks as it used to be. As a result of this misunderstanding, the more likely short term outcome is more chaos (we are seeing the start of that right now). As the instability continues, we will continue to see small attacks, like the one on the Askariya shrine and the facility at Abqaiq put entire sections of the system on the brink. Over the longer term, the system will continue on its unpredictable path with ever more numerous bouts of excess until the weight of numerous fundamental changes to the system's design and operation are made that dampen this chaos. Where we end up, at the end of this process, may be with dynamically stable system that is far from the current political and economic status quo. You might not recognize what you see.

This conclusion also calls into question the efficacy of the idea that merely increasing connectivity is an answer to our problems. Increasing connectivity too fast, in a system without intrinsic dampening or control systems that work, will only accelerate the chaos (human nature doesn't change as fast as technology). If you need proof of that, spend some time reading Jihadi Web sites and pondering the rapid growth of transnational crime. Also, the complexity of this system puts the lie to the idea that we know how to actively dampen its behavior through centralized systems of control. We neither have the scale nor the collective intelligence to pull it off. The only real solution rests on redesigning the system itself, to enable it to become more tolerant of rogue feedback.

Comments

[Mark]

John,

I've enjoyed the essay. Let me know if I've got this right:

To succeed and persist (stability), one must have high performance.

High performance requires instability and fast dampening feedback loops

The problem: Fast feedback loops require fast communications, but we have spent too much time on communication, and not enough on feedback. Fast communication 'by itself' brings both dampening and exciting frequencies. It is not a natural 'good'. With too much excitation, the system can take a disasterous trajectory.

[Duncan Kinder]

During the past generation or so, the means by which the financial markets have maintained stability is through derivatives. There are now literally trillions of dollars invested in them.

The massive capital inflows that have sustained the United States' economy despite outsourcing and various deficits actually constitutes payment for the service the United States has provided in providing this stability. In large measure, the United States now depends upon these capital inflows much as 17th Century Spain depended upon the inflows of gold from the Americas.

Should this "derivative-derived" stability be upset, the underpinings of the United States' economy would be undone.

[Robert J. Chassell]

> ... [in] today's global environment we see a
> moderately unstable system. ... The long-term
> solution ... build more stability into the
> system (decentralization) ... create dynamic
> market-places for security ...

This is true. Very good essay.

Which actions of government are natural
monopolies and which can be decentralized? For
example, I can see a natural monopoly for
deciding the amount of fossil carbon released
into the atmosophere, but see decentralized
entities figure out how to handle their fossil
carbon releases.

What are other examples?

[Robert J. Chassell]

Mark said,

> To succeed and persist (stability), one
> must have high performance.

Actually, you only need _higher_ performance to
succeed and persist. So long as your
competitors suffer low performance, your system
can be dynamically stable.

But when your competitors gain high rates of
return on their investments against you, for
example, by employing global guerrilla tactics,
then you need to increase your performance to
compete with theirs.

To do this, one way is to remake your system
with dynamic instabilities that increase
performance. At the same time, as Robb says,
"The old dampening functions of borders,
distance, government, etc are quickly fading."

Robb is arguing that the global environment now
possesses moderate instability and therefore
needs rapid negative feedback to dampen excess.
He fears that today's leaders do not perceive
the issue.

[Robert J. Chassell]

Mark said,

> Fast communication 'by itself' brings both
> dampening and exciting frequencies.

That is right. And without more dampening, the
current US-led system may die.

[Cardenio]

This is almost exactly what Baudrillard said after 9/11

[quigleydoor]

Great essay, John. What do you mean by "The end result may be a dynamically stable system that is far from the current status quo (an outcome generated by a reverse of the 30-years war)." Are you referring to the 17th century war of that name?

[Will]

Good statement of the social-structural issues facing us.

Your blog, and others like it, are giving us a better and better understanding of the problem.

The tendency now is to point fingers, e.g., at Bush and others.

No matter how legitimate the finger pointing, we have a much greater need for constructive exploration of design alternatives for the societal future. You provided some direction with the idea of decentralization -- I would say, mutual adjustment -- and private security forces.

Who's thinking further???

[Valdis]

It also matters where the fast communication and feedback are happening and the ability to understand it, or even receive it.

Are you stuck in an echo chamber that is just getting louder? Where is your external feedback coming from? Is it "hear-able" above the internal din? Are faint signals even perceptible? How diverse is this external feedback? Do you trust it? Or is it easily dismissed. Are there effective bridges that span the divide?

Can internal people see the external environment accurately? No matter how much feedback is being sent... Do they have trusted others that bridge the internal and external?

Are the sensors only set to receive info on the old problem/paradigm? Alpha waves coming in loud and clear, but the air is increasingly full of Beta waves. Is the crazy fringe the only ones hearing Beta? People expert at X usually do not hear/invent Y.

Forecast: increasing chaos overall, with areas of intense storms.

[Hakim Hazim]

This is thought provoking... keep up the good work.

[Capt. Jean-Luc Pikachu]

When Fukuyama recanted recently, did he turn his back on the idea that America needs to spread capitalist democracy, or the idea that capitalist democracies everywhere are inevitable?

[daCascadian]

Outstanding insight, well written, on our current situation !

I agree we need to "evolve" our planetary system towards more stability which, I believe, requires both more decentralized sub systems, as you suggested, and coherent higher level system(s) (above/beyond the "nation-state"); NOT suggesting at all that the current UN organism is the uber level

For those interested in such behavior(s) and the associated mind set, I suggest an investigation of the work of A. Stafford Beer and in particular "Platform for Change" which is a multilevel work covering these issues built around his experience in Chile. It is hard to find but keep trying because he has some very important things to say. "Brain of the Firm" is also worthwhile. He also had a gob of practical experience in these matters.

To begin here is a web site -->www.staffordbeer.com

Great stuff John !

I sure hope someone with system leverage is paying attention (or maybe it is also up to "We the people..." to take action & "Make it so !")

"Lack of awareness of the basic unity of organism and environment is a serious and dangerous hallucination." - Alan Watts

[s]

Well said. Along the same lines, on "open Macro" systems see following link

http://www.morganstanley.com/GEFdata/digests/20060227-mon.html#anchor0

[Rikkert]

Great essay. Could this have some connection with Boyds theory of OODA loops? I.e. system failure occurs if stabilising and destabilising feedback loops are out of sync? With communications speeding up, are our stabilising systems lagging behind?

[John Robb]

Wow. Lots of great discussion items. It's going to take me a bit to catch up with all of this.

Rikkert, from what I understand, Boyd didn't like systems. Of course, he was from an earlier era.

s, sounds like "open macros" are similar. Interesting to see Roach on the same track.

dC, thanks for the pointers to other sources.

Pik, not sure, I am going to have to check on that. Hakim, thanks.

Valdis, as always, incisive food for thought.

Will, I have some very long term ideas but not sure I want to post them during this decade.

qd, yes. The war that laid the foundation for the creation of states. This could be the start of a process towards decentralization (the pendulum may be swinging again).

Cardenio, Baudrillard sounds like Nietzsche-lite.

RJC, thinking of more examples right now.

DK, derivatives handle risk but not uncertainty (I trained to trade them at one point in my life).

Mark, you can persist without instability. You just don't grow as fast nor are things as rapidly responsive.

[Cameron Riley]

In systems architecture dampening isnt an issue as damage (rogue feedback) is routed around. The problem is, as you mentioned, many of our institutions are still industrial and heavily centralised - government, education, energy, etc. The nation state will have to devolve some more before damage flows from centralisation can be ignored.

[Eric]

"Increasing connectivity too fast, in a system without intrinsic dampening or control systems that work, will only accelerate the chaos (human nature doesn't change as fast as technology)."

I think you can generalize "increase connectivity" to "change" and better connect the discussion to classic political models. Consider, for example, Kissinger on Metternich in A World Restored:

" If Metternich considered the quest for formal constitutions chimerical, he saw in revolutions a physical disaster. In a universe characterized by a balance between the forces of conservation and those of destruction, revolution was due to a disturbance of the equilibrium in favor of the latter. But since the equilibrium was the natural condition, a revolution could achieve no more than a dislocation straining towards a new integration. The disorders attendant on revolutions were therefore symptoms of a transitional period and their violence a reflection of the ignorance of their advocates...The difference between a conservative and a revolutionary order was not the fact of change, but its mode: A consideration the liberal spirit usually ignores… is the difference in the life of states, as of individuals, between progress by measured steps or by leaps. In the first case, conditions develop with the consequence of natural law; while the latter disrupts this connection. … Nature is development, the ordered succession of appearances; only such a course can eliminate the evil and foster the good. But leaping transitions wind up by requiring entirely new creations—and it is not given to man to create out of nothingness. Civilization, then, was the degree to which change could come about naturally, to which the tension between the forces of destruction and of conservation was submerged in a spontaneous pattern..."

Considering the global political system in terms of change, enables us to better connect contemporary analyses to classic political models. Further, "change" thinking provides cleaner analogs--better ports to connect to--for physical systems, such as planes, insofar as we model their performance in terms of derivatives, or change rates.

[Bone]

"The long-term solution is to build more stability into the system through decentralization. Unfortunately, we are far from realizing that goal, since our current view of the world is based on old models."

Not a bad arguement but you dont present any proof for your :| Solution

"through decentralization"

You aircraft is the posed object, which reacts in a certain way due to internal and external influences on "designs".

It is the central object.

It would not fly better if it had alieron over De Moin, and an engine over brooklyn.

One would ask how intelligent is the design?
One would also ask where are we going?

In theory, one could count every grain of sand on the planet.
In theory our reality is limited. Finite.

You imply that the machine is not capable of responding to human capabilities, and if the design were more fluent "we" could be all that we chose to be.

We can travel no further than the last grain of sand.

Using capitalism as an anology for your airplane. Its destined to fail because its based on continued (perpetual)expansion to realize profits.

Its goal is to realize profits. It profits by selling things" This is not possible in a finite system.

View the earth from space with a big green splotch on it somwhere around Bolivia. The splotch is a fungus that grows and grows until eventually it covers the earth. It has eaten eaverything. there is no more food.

So

It dies

Which means

We need to find aliens to trade with, (and I dont mean the French,)

Nice bit of writing ,btw. I didnt understand the macro thing but I got your gist.

Peace

[NuSapiens]

Great post. One great form of negative feedback for all types of social forces is simple verbal communication. I.e., talking.

A big problem with Al-Q and the US is the Arabic-English language barrier. How many Americans have read even one paragraph of one speech by Bin Laden? Have Bush or Rice done this?

Human reactions to violence tend to create positive feedback loops. Violence continually escalating into worse violence. Talk has the opposite effect: negotiation, mediation, moderation.

Someone needs to get smart representatives of these radical factions (the U.S. and Al-Q in particular) in a room together for a week: Maybe C. Rice, Bin Laden, support people (but no media, no press advisors, no b.s. artists in general) - and they need to listen to one another.

[jamie]

If the argument is essentially that more efficient terrorism is the price of market efficiencies brought about by greater connectivity, then it follows that terrorism could be made less efficient through reducing connectivity, at the price of less efficient market functioning. So are you arguing for a trade-off here, or is connectivity a given, around which societies have to organise themselves politically?

More generally, I think you've reached the limits here of how essentially political problems can be examined through technological discourse. Any global political re-organisation is going to make its concrete effects felt right down to the neighbourhood level. So who delivers the post? And who picks up the trash? And who decides these things?

[steve laudig]

Speed kills.

[enigma_foundry]

sorry for the short post;) but, project due so here goes:

1. Excellent to look outside of the subject matter for comparitive structures (I say comparitive structures because it is much more than a metaphor) The control system is good and could even be explored further, with to necessity for a Deterministic (or Realtime) Operating System having some application here as well. I'd suggest the use of a biological comparison though. In particular The Web of Life in which he makes a very similar description of a living organism as a dynamic system.

2. The nature of the resilent decentralization that is apropriate response to GG actions can start to be outlined. It would be worthwile to actually sketch some diagrams here. As I had aludded to earlier, I suspect the flow of information will be highly global, mainly through the internet.

[Ohadi Langis]

Dampening is only one of several ways out of coping with brittle, high performance system components. C. West Churchman in his book "The Design of Inquiring Systems," 1971, writes that system components should also be capable of being severed from the larger system based on their relative weakness and their ability to compromise the system when they fail. Performance of the system is mneasured by the influence of failure modes of system components.

The problem in social and economic systems is that design isn't apparent and discontinuous change agents abound. In plain English that means any yahoo can throw a monkey wrench into the gears, and the more critical the gears are to the "system," the more damage can be done by low tech interventions.

Sociologist Karl Manheim talked about this in the pre-WWII era when he said that social scientists needed to identify the levers of change to improve society. He neglected to examing the opposite side of the coin.

[Tuor]

brilliant

[Greg]

A link to a bit of relevant research. Enjoy. - Greg

http://arxiv.org/PS_cache/cond-mat/pdf/0404/0404574.pdf

Great brief John!

I agree that emergentistic enviroments attract opportunist for better or worse and the best counter this is to increase system complexity at the lowest levels of gravitation - the topic reminds me of a systems dynamics pioneer: Forrester, Jay W., 1961. Industrial dynamics, Waltham, MA: Pegasus Communications.

Keep up the great work John - you truely are a pioneer!

[Bill Seitz]

So, in a macro social system based on free markets and decentralized power...

Individual components/actors are each trying to improve performance. But since they're all doing their own thing, the overall system retains stability/resilience.

But if those components get "too" big, an entire sector of the system becomes a MonoCulture, and thus sensitive to disruption.

And if the macro system is overly dependent on that one sector, then the "entire" system will be disrupted by a jiggle of that sector.

I wonder whether such a state is inevitable.

I think Lind would say "yes", to the extent that a complex/highPerf society leaves itself open to disruption by interaction with potentially-hostile outsiders. Hence his semi-isolationist attitudes toward immigration, etc.

[LJR]

"Dampening" is just a hi-falutin word for lowering the overall system frequency response. In a country like the USA, enamored with centralized control, this just ain't gonna happen as a result of choice at the top. Deciders stay deciders until the bitter end. With the exception of Fidel Castro, of course, who had little choice in the face of an almost instantaneous end to oil subsidies.

Concept makes for an interesting read but really doesn't offer much in the way of pointing toward a better future.

Grow your own tomatoes! That's what I do.