Governance in the 21st Century
While many agree that institutions are in a crisis there is less consensus about what happens afterward, if and when the crisis is resolved. How will the brave new world look? To answer this question is necessary to advance a hypothesis about the causes of our present troubles.
One theory advanced in an AEI paper by Leo Linbeck is that burgeoning complexity has overloaded existing structures into collapse. Intrinsically human beings can only operate within a range of complexity described by the Dunbar Number. "We can maintain stable social relationships with around 150 people—Dunbar’s number, named after British anthropologist Robin Dunbar ... the size of [an infantry] company ... the upper limit on the number of shareholders of a family business... Congress when political parties first emerged ... Dunbar showed that even if you had a million Facebook “friends,” you really have about only 150 real friends ... For better or worse, we live ... in Dunbar-sized tribes."
But companies need some way to become battalions, regiments and armies. Going beyond Dunbar tribes requires the familiar formal, hierarchical systems -- law and government for example. Yet even these are now falling down around us. The explanation is that the systems have gotten so complicated they are folding under their own weight.
When the United States Constitution was written in 1789, the US population was about four million people. By ... by 2010 [it] was more than 300 million. Along the way, the number of laws and regulations has exploded ... The first Congress passed laws that totaled 225 pages. ... In 2015, the Federal Register had grown to 81,402 pages. ... When Linus Torvalds published ... Linux ... in 1991, it had 10,239 lines of code. By ... June 2015, [it] had more than 19.5 million lines of code and 14,000 contributors. It is safe to say that no human being has read every page of the Federal Register or every line of Linux code. Both of these systems are way beyond the scale of a single person.
We simply can't deal with this. Linbeck argues that "this complexity must be managed by employing a number of strategies" besides augmenting them with subsystems since "we do not reduce complexity if we create additional subsystems but still control everything from the center. In fact, it makes the complexity problem worse." A way must be found to get intelligence and initiative to the outer nodes away from the center.
To solve the problem he advocates a solution from computer science: abstraction. "In programming, interaction between components is managed through an interface. Higher-order components, for instance, cannot directly access and modify the properties of lower-order components—they must access those properties through the interface of that lower-order component. This rule—which, perhaps counterintuitively, limits the power of the higher-order component—is a way to keep complexity under control."
Linbeck calls the gradual replacement of unwieldy complexity by local, componentized action 'greenfield refactoring', a method of finding a way to return America to the Constitutional spec way without being weighed down by legacy bureaucracy. "There is a problem with making these big transitions from older systems: They often break special features that users have added to their system. As a result, software companies rarely if ever force users to upgrade their system. Instead, they announce the availability of the upgrade and then allow users to choose when to make the switch and allow application developers to update and upgrade their systems to adapt to the new code base."