In the natural world around us, countless complex adaptive systems (CAS) thrive, such as an ant colony, a school of fish, and our very society . The benefits of applying an organization of CAS-like nature to tasks that humans encounter, regardless of field, is what led not only biologists but computer scientists, economists, and government officials alike to study and understand them. For any observed phenomenon, it needs to be properly defined before it can be studied. So, what is a CAS?
Defining Traits of a CAS
In brief, a CAS is a non-linear system that restructures systematically according to the environment . Generally, for a system to be considered a CAS, it will exhibit six traits :
- Relationality – the system is defined by the relationship between individual agents that make up the system
- Adaptability – the system must have space for adaptation
- Non-linearity – the magnitude of a system’s outputs cannot be measured in direct proportionality to the magnitude of its causes
- Openness – the structure of a CAS must allow a flow of information from the outside environment to most, if not all, individual agents in the system
- Environmental Dependency – it is expected to have different behaviors when relocated into a new environment
- Novel Emergence – individual agents cannot change the system alone, systematic changes are the result of complex interactions between agents within the system
Relationality
Within a CAS, the most fundamental significant components are not the independent agents that make up the system but rather the relationships and interactions between those agents. For example, consider two different football teams. One is an average team with average players, while the other is an average team with a newly-acquired star player. In their strategy discussions, for the first team, their coach emphasizes a playstyle that highlights cooperation between players. The coach cares more about how the players play together as a whole than the individuality of each player because they are all average players. For the second team, their coach tries their best to capitalize on their most recent acquisition, the star player. Every player is to focus on creating openings for the star player to score. What the star player can do is heavily emphasized over the interaction between all players in the team. Between these two teams, the first team resembles a CAS system better.
Adaptability
Any CAS must allow space for adaptation. This is often characterized by the decentralization control of the system. In addition, a high level of autonomy is also presented for its independent agents, enhancing the system’s capabilities to react to changes.
Non-linearity
This can be simply understood as “the magnitude of a system’s outputs cannot be measured in direct proportionality to the magnitude of its causes.” We can visualize this through the ripple effect. Imagine a person throwing three small rocks into a still lake. Had they thrown only one rock at a time, each rock would only create a family of circular ripples that shared the same center. However, because all three rocks were thrown at the same time, the family of ripples caused by each rock will interact with other ripples, forming a complex pattern that cannot be easily predicted. The complexity will increase exponentially with each addition of a rock.
Openness
For a CAS, there has to be a way for information from the environment to be passed into the system as input that will affect the internal structure of the system in some way. Due to the nonlinear nature of CAS, a seemingly insignificant environmental input can result in an abrupt systematic change. On the other hand, a supposedly significant environmental input can result in initial changes that eventually die out within the system.
Environmental Dependency
Suppose we have two identical CASs within the same environment. Then, we extract both CASs from their original environment and put each of them into a different environment. These CASs will then adapt to their new environments. Eventually, they will grow so different from each other that it will be difficult to tell that they were initially identical. A perfect illustration of this trait would be identical twins that grow up completely different from each other.
Novel Emergence
A CAS cannot be understood nor predicted on the sole basis of information relating to its individual parts . In other words, an independent agent within a CAS is helpless to change the system alone. Any systematic change is always a result of connected and complex causal reactions between multiple agents. As an example, imagine a potential failure of the banking system. Suppose, the day before the system failed, a billionaire withdrew all cash from all of his bank accounts; the system didn’t fail that day because he was only a single agent. However, the next day, a humble citizen comes to a small branch of his bank to withdraw some cash to buy a cow, but the small branch just happens to not have enough cash at hand. The man is asked to come back tomorrow. Missing an opportunity to purchase a cow in time, the citizen vents his frustration to his friends in his small town. Many of his friends, then, become worried that the bank is not keeping their money safe, come to the branch and demand their money altogether. The branch manager refuses everyone’s request and asks them to come back tomorrow, but the seed of distrust is already sown. The event is reported in the local newspaper, then the story explodes nationwide. Suddenly, everyone everywhere all at once flocks to their banks and demands their money back. The banking system has fallen because of a cow (and a billionaire).
The Free Market as a CAS
In the book Complex Adaptive Systems: An Introduction to Computational Models of Social Life, the authors claim that Adam Smith’s The Wealth of Nations is one of the earliest studies of a CAS, the free market . How does the free market we know today fit into a CAS?
- Relationality exists within the market. Suppose, a brilliant start-up invents a new device, but they want to sell the device at the price of $10 million each. To them, this is justified given the years of hard work developing it. The public does recognize the brilliance of their design, and most people would like to have one. However, to most people, the price outweighs the potential utility, hence, the demand for the device is almost non-existent. Eventually, the start-up is forced to bring down the price tag. This demonstrates that what matters within our CAS, the free market, is not the ambitious individual agents but rather the interactions between multiple independent agents together.
- Adaptability exists within the market. Companies are used to paying people an unfair wage, especially non-skilled workers, to maximize their profit. Minimum-wage workers find themselves so small and insignificant in the face of multi-million corporations. However, at one point, they start to unionize and demand higher wages and win. CEOs and stockholders all think that the economy will shrink because now they need to pay more money to their employees, but the economy keeps growing and growing, adapting to new living standards in our society.
- Non-linearity exists within the market. Wall Street is used to manipulate the market with its hedge funds, making the average trader irrelevant in its path. One day, some average trader on Reddit rallies people to buy GameStop and beat Wall Street at their own game. They deal a swift blow to those hedge fund managers. The accumulated interactions and agreements among the small traders change the market, showing non-linearity.
- The market is always open to environmental factors. Following the introduction of the credit card, to offset the interchange fee charged by banks, retailers raised prices across the board. Surprisingly, nothing really changes because the changes balance themselves out. While the higher prices discourage consumers from spending, diminishing demands, credit cards encourage consumers to spend more, raising the demands back up. A major change is passed as an input that is fully recognized and is felt by every independent agent in the system, but the changes eventually die out.
- The free market (or almost free) is environmentally dependent. This is demonstrated in how sale tax (VAT in other countries) is handled in the U. S. compared to, pretty much, the rest of the world. First, however, we need to point out the principle of elasticity and the burden of tax in economics. In short, economic principles dictate that the producers and consumers both share the burden of tax, determined by the relative relationship between demand and supply. If the demand for a certain good is relatively higher than the supply, then the consumers shall bear a larger percentage of the tax than the producer, and vice versa. This is mostly true in every other country where VAT is included in the price of most goods, but in the U. S., the consumers are usually to bear 100% of the sale taxes because they are not included in the price tag. What led to this distinction is the difference in the environment of the free markets. In most other countries, there is only one jurisdiction that determines the VAT, while there are multiple jurisdictions that determine the sale taxes on the same product in the U. S. (The state and the federal government are two of them in most cases.) Environmental dependencies also appear in other aspects of the economy as well, such as tipping, a practice that stemmed from Europe but was quickly altered fundamentally when it reached the American shore.
- For novel emergence, the banking system failure example above is an example within the market that demonstrates this trait of a CAS.
Complex Adaptive System or Complicated System?
A strictly complicated system is one in which, by identifying the change that occurs to an individual within the system, the observer can infer the effect on the entire organization eventually. However, in a CAS, it is impossible to predict the resulting systematic change by knowing an individual change. In addition, a complicated system might output different results when relocated to a new environment, but the underlying behaviors of the system are unchanged. A CAS, when relocated to a new environment, will restructure itself dynamically from the countless individual changes that result in a systematic change; it adapts. To put things in perspective, if you stand in front of a mountain range covered in snow and scream your heart out, causing an avalanche, then you’re witnessing a complex system: particles follow the same laws of physics. It can always be predicted whether the avalanche will happen or not based on your voice, your distance from the range, the temperature, etc. However, for a CAS, the resulting action is not as predictable. Imagine yourself doing the same scream but to a group of people drinking boba tea by the foot of said mountain range. They can choose to do anything from ignoring you, freaking out that you might cause an avalanche, joining your scream, going to vote, etc. We will never know for sure what the reaction of that group of people will be. A group member’s reaction might trigger different behaviors from other members subsequently. Even though few in number, the group of people is a CAS.