## Friday, February 22, 2013

### Complexity Bubble

There have been lots of economic bubbles in my lifetime: the housing and financial bubble, the higher education bubble, the stock market bubble(s), the Internet bubble, the savings and loan bubble, etc.  But perhaps everything's a bubble and the problem is not bubbles per se, but rather when bubbles become too big or too fragile (or both) and pop.

I think civilization and the extended order is just one big bubble.  Well, actually more of a high order multidimensional froth - a whole bunch of tightly intertwined hyperbubbles which I'll call a hyperfroth.  Each of the hyperbubbles is a mathematical representation of an entity of civilization - individuals (tiny hyperbubbles), companies, institutions of commerce and law, other agencies of governance, etc..  Some hyperbubbles are partially or completely enclosed by other hyperbubbles.

The reason I envision the froth in more than 3 dimensions is that it enables more interaction of the surfaces of far, far more hyperbubbles.  For example, it you pack oranges as densely as possible in a plane (2 dimensions), each orange touches 6 others. In three dimensions each orange will touch 12 others.  In 24 dimensions, each orange will touch 196,560 other oranges when tightly packed (see the Kissing Number Problem for an excellent description).  These additional dimensions are necessary for allowing more interactions between more entities represented by the hyperbubbles if a model was going to be created that was representative of civilization (it would, of course, be impossible to create an accurate model, but it would be possible to create a model that exhibited some emergent behaviors similar to civilization).  It's perfectly fine, however, to visualize the hyperfroth as soap suds in 3 dimensions and from here on I'll generally leave the "hyper" prefix off of bubble, froth, etc.

In this model, there is an elastic hypermesh that surrounds and puts pressure on the froth.  The mesh is civilization's frontier. The elastic pull of the mesh represents entropy, the tendency towards randomness that's the inherent enemy of order and usually an enemy of complexity and therefore civilization.  The pressure due to entropy acting on the mesh is pushing inwards trying to compress (and collapse) the froth.  As long as there's enough energy to keep the bubbles inflated and as long as the bubbles remain sufficiently resilient, entropy is kept at bay and civilization continues and possibly even expands (i.e. becomes frothier).

The bubbles of civilization are going to deflate or pop from time to time and usually it's no big deal.  If you look at soap suds with lots of bubbles and a few of them pop, you hardly notice.  Likewise, as long as only an occasional bubble of the extended order deflates (occasional relative to the vast number of bubbles), and as long as that bubble is either not all that large or the rate of deflation is moderate, civilization isn't much impacted and perhaps doesn't even really notice.  For example, every time someone dies, that's a tiny bubble popping and civilization isn't much affected.

On the other hand, every time a bubble pops, in the area of that bubble, entropy accelerates the frontier mesh towards the center of the froth.  If the bubble that pops is big, the resulting mesh momentum towards the center can be large.  If the nearby bubbles (and in hyperspace, a lot more hyperbubbles are nearby than in 3 dimensions) are insufficiently resilient, then these will also pop or otherwise collapse, the inward momentum increases, further stressing and popping bubbles, gaining more and more momentum and greatly increasing the probability of an all out collapse in which a substantial percentage or perhaps even all of the bubbles collapse leaving only a bit of soap scum instead of the froth.

In the general case for systems like this, events like popping bubbles occur according to a power law distribution where small bubbles (or small groups of bubbles) will pop with far more frequency than an event where a very large bubble or large group of bubbles pop.  This means that for long periods of time, the froth will have sufficient resilience to handle the popping events.  Eventually though, the big event with lots of bursting bubbles will happen, creating the acceleration and momentum that causes the whole froth to rapidly implode.

You'll remember from Resilience and Collapse that according to Joseph Tainter, an expert who studies the collapse of civilizations, that the complexity of a civilization is "generally understood to refer to such things as the size of a society, the number and distinctiveness of its parts, the variety of specialized social roles that it incorporates, the number of distinct social personalities present, and the variety of mechanisms for organizing these into a coherent, functioning whole. Augmenting any of these dimensions increases the complexity of a society."

The current civilization of which we are part is orders of magnitude more complex than anything that came before it.  If you consider the extent and complexity of specialization that's developed to produce a simple pencil and that the pencil is one of the simplest of tens of millions of products produced by and then distributed to billions of people with innumerable governments, institutions, laws, agencies, etc., it's a complexity far, far beyond the civilizations of the past.

It's also orders of magnitude more complex than it was a mere century ago when 50% of the population of the United States worked in agricultural versus less than 3% today.  But the 3% remaining in agricultural form a sub-froth that's far more complex and specialized than the 50% that used to engage in farming.  The knowledge, techniques, management, planning, business practices, etc. are far more specialized, complex, and intertwined than they were.  The other 47% are now engaged in whole new sets of activities creating goods and services that weren't even imagined back then and the vast majority of these new industries have a sophistication well beyond what the original 50% could comprehend a century ago.

Every story has an ending as does every civilization.  This civilization, like all others before it, will one day collapse, where collapse, in Tainterian terms, is the rapid simplification of society.  However, there's been an increased perception lately, especially among libertarians and conservatives, that the time till the collapse is short, perhaps measured in decades or years or even less, rather than millennia or at least centuries that might otherwise be expected to mark the time remaining for this civilization.  While the power law distribution of events might not have changed at all, the ability of the froth to withstand smaller events might have been significantly reduced in recent times.  There are a number of possibilities within this hyperfroth model to explain new weaknesses that could lead to collapse.

First, it's possible that the froth of civilization is too puffed up, with too many bubbles stretched beyond their limit.  It might be that the irrational exuberance of the Internet bubble was actually just a mirror of general over-exuberance across all of the bubbles of civilization and the momentum caused by this exuberance pushed the mesh of civilization beyond a sustainable frontier.  In a organizational sense, humans may now be stretched well beyond what they are capable of sustaining.  Perhaps Deirdre McCloskey's Bourgeois Dignity put humanity as a whole into a mania that hyperventilated the hyperbubbles and as the manic phase wears down and leads to the depression of the psyche across most or all of humanity that inevitably follows the euphoria, the stretched out bubbles will pop and/or rapidly shrink and the froth will collapse to nothing.  This is rather like Icarus flying too close to the sun and when the wings melt it's a really long ways down.

A second possibility is that civilization would potentially have been perfectly sustainable, but Statism has changed the characteristics of the froth.  The bubbles that represent the institutions of commerce, law, and governance have ceased to allow energy to flow through the rest of the froth with the necessary efficiency.  These sclerotic institutions have become rigid and sharp (as opposed to smooth) and instead of adapting when other bubbles deflate and cushioning the impact of the deflation, they've instead started puncturing and otherwise hastening the demise of the other bubbles.

A related problem is that when it comes to the institutions of governance and commerce, lots of small, independent and therefore resilient bubbles have been replaced by a few colossal bubbles that take a great deal of energy to keep inflated.  Not only do these bubbles absorb energy that could be used to keep other bubbles inflated, these colossi are also rigid, heavy and sharp and burst the surrounding bubbles at ever increasing speeds.

It's of course impossible to know exactly how things will play out but given that the volume of the mesh of civilization far exceeds anything that came before it, if and when it collapses, the collapse may make the dark ages look like a shining golden age.  From such a lofty expansion comes the possibility of an implosion that wipes out humanity or at least takes the small percentage of humans remaining all the way back to being hunter-gatherers with no or limited technology.

When Rome collapsed, it did so fairly slowly.  More importantly, most people during that era knew how to do enough to survive (farm, hunt, fish, gather, etc.).  Compare that to the current situation or what the situation will be in a few decades when even the 3rd world becomes more reliant on technological sophistication to survive.  Earlier I pointed to the example of creating a simple pencil.  It requires a huge network and no single individual knows how to do it.  Without an unbelievable number of entities a silicon fab can't be built/maintained, without computers and other advanced technologies, energy acquisition (finding it, horizontal drilling, distribution, etc.) is hopeless, without energy and other advanced technology can 7 billion people grow and trade enough food to keep themselves alive?  No!  They cannot, so the population will be reduced to a small percentage of what it is currently and that reduction will make it impossible to support the specialization necessary to wield sophisticated technology.  Those few of us who are left will be reduced to trying to scratch out a living in the dirt.

And then the really long, hard, and slow cycle of creating and pumping up hyperbubbles and building the next civilization will begin.

Jeff Shattuck said...

Great post. Favorite bit is this one:

"when it comes to the institutions of governance and commerce, lots of small, independent and therefore resilient bubbles have been replaced by a few colossal bubbles that take a great deal of energy to keep inflated. Not only do these bubbles absorb energy that could be used to keep other bubbles inflated, these colossi are also rigid, heavy and sharp and burst the surrounding bubbles at ever increasing speeds."

If only Obama acolytes understood this better, I think they would be far more skeptical of his plans to take us all forward under the aegis of a more centralized and powerful government.

Your post reminded of a great one by Clay Shirky, which is here:

Perhaps he got some of his ideas from Tainter?

Bret said...

Hi Jeff,

Good to hear from you and I'm glad to see a little bit of activity over at cerebellum blues yesterday (I think it was yesterday?).

Thanks for the Shirky link. Since Shirky mentions Tainter and devotes many paragraphs to describing Tainter's research, I'd say yes, he got some of his ideas from Tainter. :-)

I think Shirky perhaps mistakes bureaucratic complexity for general complexity. For example, he writes, "A world where that ["Charlie bit my finger"] is the kind of thing that just happens from time to time is a world where complexity is neither an absolute requirement nor an automatic advantage." But what he misses is the incredible complexity that enabled an amateur to shoot that minute of video and get it up on the Internet and be viewed by 170+ million people. There's far, far more complexity involved in that than in what the major studios require to put out a polished bit of video. The studios just have bureaucratic complexity which is expensive, rigid, and non-resilient. "Charlie bit his finger" is an emergent phenomenon that came from far greater complexity.

It's pretty tough to compete against emergent phenomena.

Susan's Husband said...

This is a reason I look at talk about The Singularity with a skeptical eye - it is precisely the inability to cope with increasing complexity that puts the breaks on such an event. Like approaching light speed, more and more energy goes in to maintaining the bubbles and less in to making them bigger.

Bret said...

The singularity guys (Kurzweil in particular) have been looking at Moore's law and extrapolating where it will take computer in another couple of decades. I find it intriguing and possible that computers will be smarter than humans eventually.

But that doesn't mean that civilization will necessarily be able to sustain yet more complexity. It could even be that the singularity causes the collapse of civilization. It'll certainly have a big effect, like, well, a singularity!

Harry Eagar said...

Very clever and self-coherent, although I think of historical examples that tend to contradict it.

It may not be correct, for example, that most in the Roman Empire knew how to stay alive. Vast areas were depopulated.

World War II popped many bubbles but did not end in scum.

Some historians have claimed that before a (local) collapse of a civilization, it shows a remarkable cultural efflorescence. Venice in the late 18th c. is an example.

The early posts at RtO, back in early 2008, often speculated that a highly ramified economy would be more resistant, rather than less, to collapse. I'd say events tend to support that, so far.