Understanding complex dynamic systems

2011-04-12

in Geek stuff, Science, The environment

Complex dynamic systems are the most difficult things in the universe to understand because they are bundles of relationships that interact in complex ways. It’s easiest to explain what they are through an example. Think of the Earth’s climate. It has discrete elements like incoming sunlight and the physical properties of water. The elements interact in complex ways that vary with time. Water forms clouds and icesheets which affect the reflection of light. The amount of ice on Earth has an effect on the totality of life on Earth, which then interacts in complex ways with other elements of the climate system: the erosion of rock, the composition of the atmosphere, etc, etc, etc.

Understanding a complex dynamic system at all is challenging. For instance, there is the task of understanding all the interactions that are ongoing when something is in a steady state. The level of complexity jumps when you consider the totality of steady and unsteady states, and all the ways by which they can turn into one another.

It seems arguable that the main task of thinking entities in the universe is to better understand complex dynamic systems. That understanding is always partial – akin to the French concept of connaitre rather than the concept of savoir. You can write down the totality of a person’s phone number on a piece of paper, but you can only express a partial view of what ‘Paris’ or ‘German’ or ‘physics’ is. In addition, it seems that complex dynamic systems are nested and that if we want to be able to behave intelligently in the world, we need to have some kind of understanding of all of them:

  • The rules of the universe: gravitation, electromagnetism, the nature of matter, etc
  • The physical Earth: the composition of the planet, and the way physical elements interact
  • The totality of life on Earth: genetics, behaviour, the history of life, etc
  • The human body: cells, organs, genes, the endocrine system, the physical brain, etc
  • The human mind: cognition, politics, economics, creativity, etc

At some point in history, it may be necessary and useful to consider the physical and/or mental characteristics of life from places other than the Earth.

The better a particular being understands each of these complex dynamic systems, the more capable they are of acting effectively in the world (a concept that presumes the existence of intentions, which ties back to each of the dynamic systems under consideration). Understanding them all better is thus a strategy capable of advancing the achievement of any conceivable goal, with the possible exception of intentional laziness or the avoidance of mentally taxing work.

Report a typo or inaccuracy

{ 6 comments… read them below or add one }

Tristan April 13, 2011 at 5:32 pm

I’m glad you’ve come across this concept. I think it’s the most powerful idea I’ve been exposed to within materialist philosophy. One reason I love it is that it exposes what we might call “truth-biases” in scientific thinking. For instance:

“The better a particular being understands each of these complex dynamic systems, the more capable they are of acting effectively in the world…”

If by this you mean, “the more accurately a particular being models each of these complex systems, the more effective they will be at modifying the world according to their goals”, then are a few problems with that. Why is accurate modelling rather than modelling which simplifies and ignores much which is technically true but not relevant to the goal at hand likely to produce “more effective” action? In fact, we are incredibly weak computers, and our brains can only handle something between 4 and 7 units of information at once. We are so effective at acting through intuitive selective application of latent inhibition; we train ourselves to only see what is relevant so that we don’t have to accurately model and compete anything.

If, on the other hand, you mean something like “the more effective someone is at acting, the more accurate their understanding of complex systems is”, then is probably closer to the truth – because it would mean grounding truth in practice (Nietzsche: “Truth serves life”). Of course, it still isn’t necessarily true, because someone might be effective at acting for reasons that have nothing to do with their understanding at all (could have to do with how people respond to them, with their position in social dominance hierarchies, it could be pure luck, etc…)

The relation between practical usefulness and truth is one that plagues the Dawkins-atheist crowd. It actually reveals how atheism itself be a groundless religious assertion of some form of Platonism, where one assumes the world is intelligible and assumes that if one is acting effectively, one must be actually getting a grip on the real intelligible (knowable) structures in the real reality. From a strict interpretation of evolutionary theory, this is way off base. Alvin Plantinga, a theist, has argued that evolution suggests beliefs should be adaptive, not “true”, and truth can not be reduced to adaptivity for reasons such as the one I made above.

The key issue here is why do we adhere to a “realist” interpretation of reality, based on an ideal of a 3rd person (God?) observer who could, ideally, verify the claims we make as true or false. Since we are always to some extent stuck in our own perspectives, this drive towards absolute objectivity will, I believe, always produce contradictions and excuses for people to think religiously (which means to talk about things not only that they don’t understand, but that they don’t understand at all). To me, the sober-realism of a post-ideological world means recognizing that all ideas are ideological in the sense that they frame our experience in one way or another, that we should be honest with ourselves about the practical import of ideas, and that we should give up the idea that science will replace religion as the bearer of ultimate truth. Ultimate truth was just a bad idea.

Milan April 13, 2011 at 6:48 pm

I’ve been familiar with the concept since I read The Quark and the Jaguar in high school, but Steven Pinker’s book has reminded me of its importance.

Even with complex systems, science seems to be the approach that yields the most understanding. Experiments may be limited or impossible, but we can still use scientific methods like validation with empirical evidence and an open system of peer review.

Tristan April 14, 2011 at 5:48 pm

” science seems to be the approach that yields the most understanding.”

“Most understanding” according to what?

Tristan April 14, 2011 at 5:51 pm

” we can still use scientific methods like validation with empirical evidence and an open system of peer review.”

There are plenty of problems with current pressures in scientific institutions. Most obvious is the pressure to find something in the data, a pressure which grows with the amount of time you spent gathering the data, and the fact you won’t get your PhD or tenure promotion if you can’t use the data to prove something. So, there is a huge pressure to find results that are not there.

Science should operate on the same incentive models that modern corporations like Honda use – where individuals get the same bonus for failed projects as for ones that succeed. The pressure to succeed every time kills creativity and ensures that failing projects will be worked on far too hard to prevent them from becoming failures.

Milan April 14, 2011 at 8:26 pm

Institutional problems aside, we have made demonstrable progress in improving our understanding of complex dynamic systems. Predictions have been verified, experiments replicated, models improved, and technology developed.

Just look how much we have learned about genetics in the last fifty years. There are small organisms where we know the role of every gene in the whole process of development, producing bodies from a set plan with a specific number of cells, each with consistent characteristics from individual to individual.

. August 5, 2011 at 4:21 pm

Epigenetics is a type of gene regulation that can be passed from a cell to its daughters. The most common mechanism is methylation. This attaches methyl groups (a carbon atom and three hydrogens) to either adenine or cytosine, two of the four chemical bases that form the alphabet of DNA, depending on the gene involved. The consequence is to inactivate the gene being methylated.

Their team examined the methylation of the glucocorticoid-receptor gene in a group of children ranging in age from ten to 19 years, and in those children’s mothers. The researchers also used a psychological survey to try to determine which of the mothers had been physically or psychologically abused before, during or after pregnancy. They found that women abused during pregnancy were significantly more likely than others to have a child with methylated glucocorticoid-receptor genes. By contrast, abuse before or after pregnancy resulted in no such correlation. Nor was the mother’s own methylation affected by violence towards her. Taken together, these results suggest that glucocorticoid-receptor-gene methylation happens in the fetus in response to a mother’s stress, and persists into adolescence.

Leave a Comment

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Previous post:

Next post: