Series:
Essay #6:
Synopsis:
Causation
Randomness
There's randomness in all causation, and it’s the joker in the deck
In causation, randomness is the joker in the deck. All systems have random noise in the signal. Let’s pretend our quantum-god computer is tracking mechanical causation from step to step in a complex system... it still can’t account for randomness. The key to mechanical causation is that, if we have all the data about causes, we can predict all effects, which would be true except that 1) we don’t have all the data, and 2) random sh-- happens and messes up our deterministic causal connections. Per Melanie Mitchell in her book, Complexity, A Guided Tour, “Seemingly random behavior can emerge from deterministic systems, with no external source of randomness.”
Randomness is everywhere. We saw random noise in my essay Probability when jumping water droplets forced Schrodinger to look at aggregate behavior and to base causation on probabilities. We see it in evolution, where randomness isn’t a bug, it’s a feature, because random variations are fuel for the natural selection machine.
Remember when we started mapping the human genome? Folks said it was only a few years and we’d locate the defects that cause our hereditary diseases, then we’d alter DNA to fix the defects. Translation: DNA causes the hereditary disease, so fix the DNA and cure the disease. Wow, what a whopper of a deterministic causal theory! Recently we’re backing off of DNA as the determinator of traits. We’re seeing too many intermediary forces at work between DNA and trait. I’ll give an example of a newly found intermediary factor: random variation that occurs during gestation, after conception but before birth. It’s like taking a gallon container of Breyer’s ice cream and adding a layer of strawberry between the vanilla and chocolate. Now there are three flavors: nature, nurture and randomness in between.
It works like this. During gestation, as you divide cells from 1 to 2 to 4 to 8, every single division is a fork in the road permitting alternate paths. Mutations, genetic defects, environmental factors and other random processes cause the cell to take one path rather than another path. Each path taken, or foregone, determines future paths, that is, future paths all depend on prior alternatives chosen. It’s like an ant crawling out on the branches of a tree, from big branches to small; at each fork it moves farther away from the road not taken.
Imagine you have one cell that you intend to clone 1,000 times, then gestate them all to birth. Will the 1,000 clones be identical? No. Each newborn clone will be different from the others due to random input during gestation before birth, which gestational input causes the fetuses each to develop down different paths. The differences will be unpredictable because they come from random events. If we watch the baby-clone grow up, we’ll see further differentiation as the outside environment affects the differences that arose in gestation. For example, if a baby-clone is born with an affinity for music, that child will more likely grow up musically as that affinity expresses itself in the outside environment. The Matthew Principle is real: the rich get richer.
There’s a great example of this phenomenon at large in Europe. I read about it in the article, Nature Versus Nurture? Add Noise to the Debate, by Jordana Cepelewicz in Quanta Magazine (March 2020). Europe is besieged by mutant marbled crayfish, in their millions. They’re all clones of one mutant who arose around 1995. The crayfish are female and they reproduce asexually, making clones of themselves. Each clone’s DNA is identical to her sisters and her mother.
The interesting part is that each clone is different from her sisters and her mother although their DNA is identical. The crayfish are different from one another in color, size, behavior, longevity and other traits. Crayfish siblings raised in a lab with a fixed environment turn out sufficiently different that they can form a social hierarchy. All this diversity starts with random processes in gestation.
Schrodinger’s old friend, Brownian Movement, even makes an appearance in cellular development. From the article in Quanta Magazine: “The processes that make proteins and turn genes on and off are subject to this molecular jitter in the system… which leads to some degree of randomness in how many protein molecules are made, how they assemble and fold, and how they fulfill their function and help cells make decisions.” Random Brownian Movement is a factor in the gestating fetus’s choices at the various forks in the road, which is fascinating. If the random movement of molecules or even atoms can cause effects at higher, more abstract levels of organization (like traits), complexity becomes a lot more complex than we imagined.
Armadillos are another example from the same article. Nine-banded armadillos have litters of four genetically identical babies. A biologist, Jesse Gillis, found that identical embryos will differentiate either as male or female at the 25-cell stage. The mechanism is utterly random, like flipping a coin, but it affects all future development downstream that’s related to sex. Given that an armadillo eventually will have a trillion or so cells, a choice made at the 25-cell stage will have a complex ripple effect throughout the gestational development of the armadillo. The author says, “the random noise that establishes a slightly different pattern of gene expression in each armadillo embryo is amplified through its influence on other developmental processes and eventually yields differences in traits.” This is a great example of sensitive dependence on initial conditions from my essay, Complexity.
To sum up, in complex systems, causation doesn’t roll forward in a linear and determined path. We see this with crayfish and armadillos where DNA doesn’t determine a trait; instead there’s an attenuated relationship between DNA and traits over the development of the fetus that includes random events plus the amplification of prior randomness. Small changes start a cascade that leads to big changes, or maybe not, we can’t predict it. It’s random, which means what? It means we don’t know what the hell is going on here.