THE GAUSSIAN VS. THE PARETIAN STRATEGIST

There are two types of strategists, the Gaussian and the Paretian (this is a simplification, but it is very explanatory). The difference between the two is in how they view the probability of human events. The Gaussian strategist views the distribution of events along a normal curve (the cure to the left in the diagram from Albert-Laszlo Barabasi's, "Linked"). The Paretian strategist views the distribution via the lens of the Pareto curve (the curve to the right). Here's a good explanation of the difference from business strategist John Hagel:

These are two very different ways of viewing the world, with some events following a Gaussian distribution (classic example: the heights of individual human beings) and other events following a Pareto distribution (classic examples: frequency of word use, size of human settlements, distribution of Internet traffic and intensity of earthquakes).

Historically, Gaussian expectations for most events derived from human systems were usually correct. In that world, dampening factors dominated within relatively sparse and simple systems, driving events towards the mean. Over the last decades, however, systems have shifted towards towards ever greater levels of complexity and information density. The result has been a shift towards Paretian outcomes, particularly within any event that contains a high percentage of informational content. Another way to see this is through the lens of what Nassim Taleb calls mediocristan (Gauss) and extremistan (Pareto) in his excellent book, The Black Swan. Here's a great comparison of attributes from these two worlds (paraphrased from that excellent book):

Our collective problem is that Gaussian strategists dominate public policy and Paretian strategists are usually relegated to the status of pariahs (this is much less true in business, where the mechanism of entrepreneurship harnesses Paretian outlooks). In contrast, the mechanisms and processes of our opposition is increasingly focused on creating Pareto outcomes (with ever greater levels of success). Essentially this means the intentional creation of small events that use global system dynamics to create outsized outcomes (to potentially change the world). For example, a bad Paretian event is 9/11 and on the good side is weblogs/RSS (where I played a major role in generating events that took the technology from rough ideas to a major upgrade of the WWW). As Paretian outcomes continue on their path to dominating events, that gap is going to loom large unless we can find a way to bridge it. From my personal experience with the nearly complete lack of interest within big government bureaucracies for Paretian thinking that is far more explanatory, actionable, and predictive than what they currently produce, I don't think we will unless we develop it outside the traditional public organizations. In that sense, we will all need to become global guerrillas.

Comments

[georgelarson]

Do you think the extremistan situation is what Vice President Cheney was trying to explain when he allegedly said we could no longer ignore terrorist plots with unlikely outcomes? Or am I giving him too much credit?

[John Robb]

Actually, that was fairly perceptive of him. His solutions, on the other hand, disregarded a similar approach.

[Duncan Kinder]

I all candor, I'm not sure I follow how shifting a system toward greater levels of complexity and information density gives rise to Pareto rather than to Gaussian distributions.

Nevertheless, managerial techniques dealing with Pareto distribution are well known, having been developed by the quality control expert Joseph Juran.

see: http://en.wikipedia.org/wiki/Joseph_M._Juran

Pareto analysis was part of the "reinventing government" movement spearheaded by Vice President Al Gore during the Clinton administration. See, eg.

http://www.kfmaas.de/q_tqm.html

or simply Google ""reinventing government" Pareto"

So Pareto concepts consistent with modern government actually are well known and - indeed - would likely be a prominent feature of a Hillary Clinton administration.

[Russell120]

--The result has been a shift toward Paretian outcomes, particularly within any event that contains a high percentage of informational content.--

Possibly, however I suspect that the continuing shift away from the agrarian economy is the driving force behind the change. People involved in basic life sustenance activities do not have the time or resources to explore alternative avenues of anything (business, cultural change, etc.). The anarchists and early nationalists seem to have managed some pretty successful black swan events in their day, and their rise seems to pretty well correspond with the rise of the urban middle class.

IMO Warfare has always had a tendency toward Pareto events because by definition it is an activity that involves active opposition to whatever plans the participants have in mind. Thus all plans have an active element of failure within them. In the very long run (per Paul Kennedy anyway) large economies on strategic flanks have generally been able to get favorable results in a rather Gaussian manner. The rise of the British and then American Empires would be classic examples. But being able to run through history only once, issues of sample size and survivor bias are both a problem.

Presumably the globalization reduces the "flank" advantage of the United States. I would concede that it is relevant to issues of 4th generation warfare, but it still strikes me as very relevant to issues of intra-state war. So in a rambling sort of way, I am questioning whether the US can use rely on its strategic distance, economic strength and cultural...(inertia?) to muddle it way through.

As a side note, there is only a very slight delay in getting your book from Amazon (1 week). However, whether this result lies within a Gaussian distribution (7 days versus average of 6 days) or Peretian (7 days versus the average of 6 months) is unknown.

[joep]

Interesting you bring up the difference between a gaussian and a pareto approach to interpreting facts. Mathematically interpreted assumptions -incidental or structural- make such a difference in the scope of assessment. Excess or promise?

At what point should incidents be regarded structural? And, could mathematical models assuming an answer to this question, provide with an answer to this question?

[Jeffery]

John: I have not read The Black Swan yet but I have visited Nick Taleb's blog and in one of his audio interviews he said that most of us live in mediocristan, if so why will we all need to become global guerrillas?

[John Robb]

Duncan, I've built systems that employ TQM/Juran/6 sigma approaches into them. From my understanding these methods are for process improvement (to reduce randomness and tighten gaussian distributions of outcomes). This is quite a different thing entirely. Pareto has a big body of work.

Joep. This could be structural. Not sure how a mathematical model would be built. However, you can build an index to measure your success or failure.

Russell, you are exactly right. However, this is much more so due to the presence of global systems that can amplify events. Looks like a Gaussian distribution to me ;->

Jeffery, working on how to be one right now with a proposal for a new book.

[Jeffery]

OOPS, I apologise for not writing more clearly.
My question wasn't how we become global guerrillas, but why become global guerrillas at all.
According to Nick Taleb most of us are residents of mediocristan so we live in the land of the bell shaped curve, not the power curve. According to Dr.Nick. Right?

[Russell120]

Jeffery:

Taleb had some of the chapters of the Black Swan up on his web site prior to publishing. Although he describes it as "mediocristan", he is not necessarily derisive toward those who occupy that zone.

He notes that Dentists, Doctors, Electricians, etc. all occupy mediocristan. There activities are frequently performed, and highly skill dependent. You cannot be a bad, but lucky dentist for long. However, artists , politicians, business executives, in his mind, live in extrimistan. Although, skill is certainly important, being in the right place and time at a few key moments can get you a long way.

I think the point made is that if you want to make a global impact as an individual, most people would need to think in the extrimistan ways to maximize the impact. An obvious starting point would be basing activities not only around their success or failure, but on their media impact. With the risk of descending too much into "buzz-speak", you should look for the tipping points.

I am almost finished with Richard Bookstaber's A Demon of Our Own Design, and have new copies of Brave New War, and The Black Swan in my greedy little hands. Which should I read first if I wish to enter extrimistan most quickly?

[John Robb]

Jeffrey, I suspect that since we all increasingly live in a world dominated by extreme events (extremistan) then we need to find ways to operate in that environment. It goes beyond your day job.

Russell, read BNW first. It's a much faster read. The BS takes some time and should probably be read in parts to better digest the philisophical contents.

[georgelarson]

http://www.fooledbyrandomness.com/nyt2.htm
Taleb's thoughts on terrorism 2005

http://www.fooledbyrandomness.com/nyt.htm Taleb's thoughts on terrorism 2004