Determined: A Science of Life without Free Will

Lorenz’s findings went largely unnoticed for years, as most researchers preferred reductionism. Sapolsky criticizes the term “chaos theory” as inaccurate because the theory “is about the opposite of nihilistic chaos and is instead about the patterns of structure hidden in seeming chaos” (144). Chaoticism grew in popularity in the 1980s, and other fields including education, economics, literary criticism, the entertainment industry, and behaviorism considered chaoticism’s implications.

Some free will supporters feel that unpredictability proves free will, and some believe chaoticism disproves determinism. Sapolsky counters that predictability is separate from determinism and reaffirms that chaoticism is both deterministic and unpredictable. Determinism deals with why events happened, and predictability deals with what will happen in the future. If someone used graph paper to copy down a chaotic graphic from the previous chapter using the same rules, they would get the same image time and again because the pattern is determined. As a biological example, Sapolsky suggests that in 1922, one could not predict which out of 100 individuals would become inappropriate and impulsive in the future; however, in 2022, the blood samples could be genotyped to find one individual with a gene mutation that will eventually result in dementia. In 1922, the person’s inappropriate behavior would have been considered the person’s fault, while in 2022, it would be attributed to their disease. In this situation, the person’s behavior “stops being free will,” and Sapolsky argues, “There is something wrong if an instance of free will exists only until there is a decrease in our ignorance” (150). Humans intuitively sense that they have free will because they cannot comprehend the infinite number of factors contributing to their behavior.

A second misrepresentation of chaoticism in relation to free will deals with convergence, seen in evolution as two unrelated species developing similar characteristics or in the concept of having two individuals press a lethal injection button so neither knows who killed the person. Some chaotic systems are too complex to analyze with reductionism; thus, an individual occurrence’s causes cannot be determined. Some use this to argue for indeterminism, which would then allow space for free will to exist. Sapolsky agrees that chaoticism challenges radical reductionism, but, he argues, it does not impact determinism. He concludes by reiterating that chaoticism is deterministic rather than random, as is often assumed. Humans ascribe arbitrary reasons to concepts they do not yet, or may never, understand.

To introduce emergent complexity, Sapolsky proposes that enough moving bricks that follow a certain set of rules would eventually settle to form the Palace of Versailles: “Put enough of the same simple elements together, and they spontaneously self-assemble into something flabbergastingly complex, ornate, adaptive, functional, and cool” (155). Unified marching bands do not demonstrate emergent complexity because they follow a plan; colonies of ants do demonstrate emergent complexity because, without a plan, they create a complex functional unit. Emergent complexity requires a large number of simple elements that follow simple rules to form a complex unit consistently and sometimes unpredictably without a master plan. Sapolsky uses bee and ant colonies to elaborate, introducing the concept of swarm intelligence, which occurs when swarms work out the most efficient ways to perform a task. For example, two bees go in search of flowers; the one that finds the better site performs a longer dance, which encourages more bees to go to the better site. The new bees also perform the longer dance when they arrive back at the hive, which recruits more bees to go to the better site. Thus, the bees have, without a master plan, followed simple rules to create an efficient unit. Ants demonstrate similar swarm intelligence by finding the shortest route to resources, which they accomplish by leaving pheromone trails—the shorter the route, the stronger the trail, and the more ants recruited to the most efficient site. Slime molds achieve a similar feat by expanding to fill the available area and then retracting, leaving only the most efficient connections to resources. Emergent complexity in the nervous system is exemplified through new neurons following radial glia, cells that promote development in the nervous system, and leaving behind chemical trails that attract more neurons.

Sapolsky explores two examples of emergent complexity—bifurcation and Pareto distribution. Bifurcation is when something grows for a while and then branches out, with each new division repeating the process. Sapolsky uses bifurcation to discuss the concept of infinitely large items fitting into infinitely small spaces, as seen in the human circulatory system, which, in an adult, contains 48,000 miles of capillaries. Bifurcation follows a set of rules rather than each bifurcation happening at random or being controlled by an individual gene—a segment grows to a certain ratio, then divides, then those division grow to a certain ratio and divide, ad infinitum. Bifurcation can also be seen in non-biological circumstances (rivers bifurcating into smaller streams in deltas), and it can be seen in cultural development. Sapolsky illustrates the concept by designing a hypothetical town; the buildings and resources must be efficiently placed to optimize the town. Neurons will arrange themselves similarly using emergent complexity and bifurcation processes. Sapolsky then turns to Pareto distribution, or power-law distribution, which states that 80% of the effects come from 20% of the causes. The ratio appears in numerous examples, including that 80% of earthquakes are in the lowest 20% of magnitude, and 80% of texts are sent to 20% of a person’s contacts. Power-law distribution also appears as an adaptive response, as is seen in brain development. Sapolsky shares two more examples of emergent complexity. First, if someone cuts their toenail at an angle, the nail will grow in such a way that evens out the nail rather than retaining the angled shape as it grows. Second, neurons put in a petri dish will organize themselves into a brain-like structure.

Sapolsky identifies three problems with arguments that use emergent complexity to “prove” free will. Some argue that the unpredictability often seen in emergent complexity is indeterminant, but Sapolsky counters that unpredictability is different from indeterminism. Christian List’s work suggests that some events are deterministic while others are indeterministic. He declared that he had proved emergent indeterminism when he compared two similar systems, having rounded the variables in the second example; he argued that the different results demonstrated indeterminism. Sapolsky counters Lorenz’s claim by pointing out that Lorenz entered a slightly different variable: “Two things that are similar are not identical, and you can’t decide that they are simple because that represents the conventions of thinking” (194). Changing the data, even slightly, skews the outcome and thus does not prove indeterminism. The second argument Sapolsky examines is the idea that the elements in an emergently complex system are free to do as they choose. He counters by reiterating that the elements are constrained and cannot do as they choose.

The final argument he discusses is the idea that emergent complexity can “reach down” to impact the elements that comprise the system. Theoretically, weak downward causality cannot change the elements, but strong downward causality can impact the elements, which, some argue, demonstrates a source of free will. Sapolsky agrees that a complex system can impact the individual elements, but it cannot free those elements from their original constraints. Even if strong downward causality impacts individual elements, it still creates a preceding state, again leaving no room for free will.

Sapolsky hesitates to write on quantum indeterminacy, saying,

Yes, the Laplacian determinism really does appear to fall apart down at the subatomic level; however, such eensy-weensy indeterminism is vastly unlikely to influence anything about behavior; even if it did, it’s even more unlikely that it would produce something resembling free will (204).

Sapolsky refers to particles in the air moving within a beam of sunlight. The movement is caused by photons hitting the particles, causing them to move and bump into each other randomly and unpredictably. Such movement, called Brownian movement, increases with heat and decreases as particle size or environmental viscosity rises. Brownian movement is documented in biological processes, including cellular division. The Levy walk is another form of biological Brownian movement that occurs when prey animals, such as smaller fish, move about randomly searching for food while evading predators or when white blood cells search the body for pathogens.

Sapolsky outlines the areas of quantum indeterminacy that are relevant to the free will conversation, starting with wave-particle duality, which states that quantum entities move as both waves and particles. This is demonstrated through the double slit experiment, which shows that a single electron fired at a wall with two slits will pass through both slits—a phenomenon called superposition. However, if the electron is recorded passing through the wall, it will go through one slit, and which slit it will go through is unpredictable. Scientists do not yet understand why this occurs. Further, it is impossible to simultaneously measure both the momentum and location of a particle, as measuring the particle fundamentally changes it. The next example of quantum indeterminacy is entanglement, which occurs when two particles become “entangled” and become subject to each other’s behavior, even when they are separated by great distances: “Suppose you have two entangled electrons a light-year apart; alter one of them and the other particle is altered at the same instant…a year ago” (212). Quantum tunneling, the final discussed form of quantum indeterminacy, shows that particles can tunnel through objects; if a beam of light is shined at a wall, some photons will go through the wall.

Sapolsky lists and criticizes the ways people have used quantum indeterminacy to support free will and other neurological and psychological phenomena, including depression and schizophrenia. Some believe that quantum effects can “bubble up” and impact biology. Although most researchers agree that quantum effects cannot impact neurology because their effects cancel out, some still purport that free will arises from quantum mechanics, such as neuroscientist Peter Tse, who argues that brains evolved to magnify quantum effects. Sapolsky counters that quantum effects are far too miniscule to impact neurological function: “You’d need to have a staggeringly large number of such random events occurring at the same time, place, and direction” (220).

Neurons work by sending an action potential down the axon to the thousands of axon terminals, which then release neurotransmitters that move to another neuron to trigger another action potential. Axon terminals can contain millions of neurotransmitter molecules, and packets of neurotransmitters are released at a time. Bernard Kats, a neuroscientist, found that sometimes axon terminals release a small “hiccup” of neurotransmitters spontaneously. Some have used this phenomenon to support free will, but Sapolsky warns that the releases have underlying causes and purposes and are not indeterministic, nor do they amplify to affect neurology. Neurons may exhibit spontaneous action potentials, but they have causes and purposes, as seen in muscle twitches, and the brain consistently runs a “default network,” which is believed to assist in creatively solving problems. These examples of spontaneous neural activity are deterministic and thus do not support the concept that free will emerges from quantum activity.

Sapolsky broadens the discussion, addressing those who feel that quantum indeterminacy shows that the world is indeterminate in general. He argues that if quantum indeterminacy was the foundation for behavior, behavior would be random: “You’d just be making gargly sounds because the muscles in your tongue would be doing all sorts of random things” (229). If amplified quantum indeterminacy did impact behavior, it still would not prove free will, as the behaviors would have been generated by quantum effects rather than free will.

Sapolsky explores Dennett’s idea that brains “harness” quantum events and examines two potential methods of doing so, “filtering” and “messing with.” The immune system offers an example of “filtering”; it uses randomness in its search for a pathogen and in its early attempts to destroy the pathogen, and then it filters for the most efficient way to eliminate those pathogens. Dennett suggests that decision-making is a similar process, where humans randomly generate potential solutions and then filter for the best option. Sapolsky argues that this concept is implausible; the brain does not provide random potential answers, and if it did, it would be impossibly time-consuming to process all the potential random options. Additionally, the process of choosing the best option would be dependent on the same factors discussed earlier—biology and experiences—and thus would not constitute free will. The “messing with” hypothesis works in reverse, with a high-level self using quantum effects to manifest free will. Sapolsky notes that the idea of controlling quantum indeterminacy is an oxymoron and that he does not understand what proponents of this view, namely Kane and Tse, mean when they discuss messing with quantum occurrences.

Sapolsky concludes by reiterating that not only is it unlikely that quantum indeterminacy impacts behaviors, but experts also believe it to be impossible. The idea of determined indeterminism as presented in the idea that humans can harness quantum effects is an oxymoron, and if behavior was impacted by quantum effects, it would result in random, nonsensical behaviors.