not downvoted, but also not upvoted - this wasn't really useful for the LW audience - nothing particularly new, and no added clarity on any of the questions.
I would be very happy to see, and almost certainly upvote even if I didn't agree, with a strong attempt at an operational definition of "power". I think it's a missing element in a LOT of discussions about bargaining, AI takeover and influence, voluntary vs involuntary actions, etc.
Maybe LW is the wrong audience. This piece was written to be thought-provoking for an educated general audience and for social impact. It was not meant to be rigorous or to handle concepts at a higher technical level. Nevertheless, feedback from the LW community is valuable.
Disclaimer: I did point out in the beginning that this article will abstract only what I view as the essential components.
Hazem (the author)
It is probably not an exaggeration to state that superconcentrations of power are the cause of most of the ills of the world.
A solution, if it exists, requires a good and impartial understanding of the underlying dynamics, preferably not colored by political feelings, particular situations, and current events. The problem is complex in practice, but here I offer a view that abstracts only the essential components. I am not the first to point out these ideas in one way or another.
I've long thought that the tendency towards concentration of power is a feature not just of human nature and human society, but a manifestation of a phenomenon exhibited by a large class of systems in general, given certain conditions. I define power here as control of resources.
In fact, it's a mathematical thing: When the rate of growth is proportional to the amount already accumulated, in the absence of sufficient constraints.
This phenomenon is known as Exponential Growth.
As a mathematical model, exponential growth can describe many systems that satisfy its conditions. These systems may be social, physical, mathematical, etc. The behavior can even be simulated in software (see for example Agent-Based Modeling and Systems Dynamics)
When we study concentrations of power, it is necessary to pay attention to the nonlinear nature of the process, and also to the presence, absence, and nature of constraints.
Constraints can be internal or external, passive or reactive, or can even be other competing concentrations of power.
Resources can be many things, and can even be other agents.
Of course, no exponential growth process can continue forever, because in the end, it will run into the ultimate constraint of limited resources available.
This implies that a runaway process will eventually stabilize, or, more likely, reverse.
In social and historical evolutionary contexts this can explain the semi-cyclical rise and decline of states and empires, as well as corporations, tribes, and other power structures.
What it means practically is that we will eventually be free of any one particular tyranny, although, who knows after how much damage is already done.
This thesis also highlights the massive role technology plays in human history, since technology (more generally: know-how) essentially makes more resources extractable from the same pre-existing environment, thereby removing constraints on growth, for a while.
So yes, power tends to grow exponentially until checked by constraints ("The rich get richer" in popular wisdom, Class Differentiation in Marxism), and that seems to be a natural tendency explained by exponential growth laws. The question is: What constraints can we put on that process to eventually force a more cooperative rather than competitive social dynamic, where we can have growth without high concentrations of power, for as long as possible?
I submit: Rapid feedback mechanisms, knowledge and secure communication, and as much baseline self-reliance as possible.
More on that later ...