The Constraints of Life

Playing Conway’s Game of Life is an easy way to spend an afternoon. The game is based on the concept of cellular automatons; essentially, it consists of a grid of cells where each cell can be either on or off (also referred to as alive or dead), and at each time step each cell updates its state based on the other cells around it (the quickest way to grasp the nature of the game is by playing it. In Conway’s game, each cell’s next state depends on its eight neighbors (the square surrounding the cell) and itself, with update rules as follows:

1. If a cell is on, it stays on if it has two or three on neighbors; else, it turns off
2. If a cell is off, it turns on if it has exactly three neighbors on; else, it stays off

It took Conway some time to find this rule, as there are 512 (2^9) possible sets of rules for this type of game (and even more if you expand the concept of “neighbor”). For example, you could have an on cell stay on with two, three, or eight on neighbors, and an off cell turn on with zero or three neighbors (adding zero here changes things rather drastically!) Many of these alternative options generate interesting behavior, and people have spent a significant amount of time exploring them. If you’re interested, https://conwaylife.com/ provides good detail (though it’s mostly focused on innovations within Conway’s game), and A New Kind of Science by Stephen Wolfram (creator of Wolfram Alpha) offers an extremely deep dive into the behaviors of all the different rules (together with insights into how a cellular automaton manner of thinking could help advance science).

The Game of Life universe is so interesting because of its constraints. Constructing any type of artifact requires fitting within the universe’s very exact rules – but even with these limitations, there’s surprising freedom to create (see 1:54 and onwards). 

Though these large scale creations are possible, figuring out how to construct them is not easy. The pieces aren’t linear – each cell interacts with each other cell. There are few regularities to latch onto for engineering purposes. In this type of environment, generally the only way to understand what will happen is to actually run the simulation.  You can’t change one part of the environment without affecting the entire environment; only certain types of structures have “staying power” in the universe, with most disintegrating into either an empty world or a few small repeating patterns. Building something permanent (or even semi-permanent) in the Game of Life requires “fitting in” the design to the underlying substrate; building from the top down without considering this substrate leads to failure. A designer must respect the substrate.

I kept thinking back to the Game of Life during a conversation with some friends a couple nights ago. We had ventured onto the topic of climate change and were discussing possible solutions. A variety of proposals were submitted, from encouraging an attitude of sacrifice, to passing laws to limit corporations, to giving up and hoping intelligent life with a better nature evolves down the road (this one may have been proposed in jest). For the serious proposals, we all agreed on their efficacy, but disagreed strongly on their feasibility – particularly from the perspective of incentives. It seemed that all the ideas checked the box of reducing emissions, but were limited in their ability to actually come to fruition given human nature. For example, instilling an attitude of sacrifice (specifically towards energy usage) in the general populace would certainly lower emissions, but the hard (if possible) part is incentivizing people to actually have that attitude. There’s a certain amount of selfishness embedded deep in human nature – getting people to want to act counter to it is a tall order. This type of issue seemed common across all the proposals; they all sounded good in theory, but it was unclear whether they could exist in the world. This was the connection to the Game of Life; it seemed we were talking about configurations without respecting the substrate.  

The constraints of the real world are far less explicit than those of the Game of Life, but they limit possibilities in a similar way. At the most basic level, these constraints are the laws of physics, which in many ways are directly analogous to the Game of Life rules (in that they determine how each particle, or “cell”, moves and interacts). Due to the regular nature of these laws, however, we can understand the world with a similar degree of accuracy from “higher” levels (chemistry, biology, psychology, sociology, etc.) at which different constraints govern behavior. For example, in chemistry, the constraints take the form of limitations on the possible persistent forms subatomic particles can take (hence the periodic table), and in biology the constraints take the form of limitations on the possible phenotypes DNA sequences can give rise to (though this class is vast). At these levels, the analogy is fairly direct and the notion of “constraint” fairly mathematical. The extension continues further, though less directly; at the psychological and sociological levels, possibilities are constrained primarily by human nature, and secondarily by the pre-existing cultural and social “memes” already circulating through the minds of the populace. Put differently, the human mind can only “fit” into certain configurations (though the possible space is again vast), and this space is further limited as minds are molded by cultural and social pressures. For example, we have a powerful innate drive for self-preservation (generally – the suicidal mind represents a different possible configuration). Attempting to instill a state of mind opposite to that requires an uphill battle, and even then can only be accomplished in certain ways (appealing to honor, duty, family, etc.). The configuration of “happy and content, with no sense of self-preservation” is not one human minds can occupy (outside of the most contrived of circumstances). When considering more global ideas, like how to limit climate change, the ideas must not only solve the problem, but also “fit” the substrate; the idea must be able to exist given the constraints of human minds and societies. Like in the Game of Life, it’s difficult to reason about what type of ideas fit, because the process is again highly non-linear. However, we have at least one example to learn from: Capitalism.

In designing an economic system, Adam Smith started by examining the constraints of the substrate. Though he may have privately desired certain outcomes (e.g. a system which offered a base level of dignity, or which maximized human happiness), he recognized that he must work within the constraints given. In his view, the main constraint was human self-interest. The system he designed, Capitalism, plays directly on this constraint, offering an opportunity for (or, more pessimistically, forcing) each person to advance their own self-interest through economic production. As we’ve seen over the last few centuries, Smith’s analysis was in large part correct; Capitalism has made use of our self-interested tendencies to transform the world. We can look at Capitalism like the repeating structures of the Game of Life – both “fit” into the constraints of their world.

While Smith’s process offers an example of success, it tells us little about how to attack problems like climate change. There’s no clear (accurate) simplifying assumption on human nature from which a solution would easily arise. Additionally, even with a stable solution, you still need to “get to” that state. In the Game of Life, the “getting to” is easy; you simply select which cells you want on and off at the start. In the real world, however, “getting to” a solution can match “finding” a solution in difficulty; the shift to Capitalism took decades. It seems any path forward will be complex, requiring consideration of various aspects of human nature, culture, and social norms. The substrate is not an easy one to build from. To tie back to a favorite post of mine, the substrate is Moloch.

If you enjoyed this post, please consider subscribing to be notified of new posts 🙂

Name*

Email*