Determinism vs. Free Will: A Middle Ground

Dan Davis

Philosophical disagreements over free will and determinism date back as far as (approximately) 300 BCE, when the Greek philosopher Epicurus suggested that chance is a source of human freedom and the Stoics took him to task. Over the centuries, scientists and eventually even psychologists were drawn into the debate. The battle still rages on.

The dispute over causality and chance has spilled into cosmology, quantum mechanics, biology, and behaviorism. It tends to focus on an either-or scenario: either every choice and event is inevitably predetermined or everything must occur at random. This oddly limited discussion precludes a middle ground: that randomness and (relative) determinism can coexist, with free will included in the mix. Such an approach is proposed below.

Absolute (“hard”) determinism asserts that every choice and event is inevitable, because each is predetermined by antecedent causes and conditions that in turn were inevitable. Advocates of this interpretation believe today’s choices and outcomes represent the latest links in chains of causality going back to the big bang. The necessary corollary is that free will does not exist.

By the early twentieth century, scientists were approaching a consensus that the operations of the universe were predetermined—until the discovery of quantum mechanics muddied the waters. Albert Einstein, a determinist, despised the implications of quantum probability and famously asserted that “God does not play dice with the universe.” He was at least half right: since supernatural beings don’t exist, they probably don’t play dice with the universe.

However, there’s credible evidence that the universe plays dice with the universe. In fact, a multitude of universes may exist, all of which could be playing dice with themselves (or each other). These are only a couple of the scenarios hypothesized by physicists within their various and often conflicting interpretations of quantum mechanics.

Particle activities at quantum levels seem to follow principles of probability rather than reflecting causal determination, a behavior that conflicts with many prevalent ideas about how the laws of science are supposed to work. Even physicists specializing in quantum mechanics can’t come to a consensus. When a survey addressing the issue was presented at three separate conferences (each attended primarily by physicists, along with a smattering of philosophers and mathematicians), the responses varied widely from conference to conference.

Since each conference had its own particular theme, each set of participants was likely to hold viewpoints consistent with those of their conference sponsors but not necessarily similar to those prevalent within the other two groups. Due to this probable attendee bias (and the fact that the surveys were not, and did not purport to be, scientific in any case), the outcomes cannot be regarded as representative of anything but the opinions of the people who participated. Nonetheless, the responses to the following two questions provide some interesting insights.

  1. What is your opinion about the randomness of individual quantum events? (Some participants provided more than one response to this question.)
    Group 1 Group 2 Group 3
    The randomness is only apparent. 9% 36% 17%
    There is a hidden determinism. 0% 33% 28%
    The randomness is irreducible. 48% 26% 39%
    Randomness is a fundamental concept in nature. 64% 24% 39%
  2. With respect to quantum information, how much is the choice of interpretation a matter of personal philosophical prejudice?
    Group 1 Group 2 Group 3
    A lot. 58% 40% 78%
    A little. 27% 34% 17%
    Not at all. 15% 15% 6%

The disparity among the “determinism v. randomness” responses reflects the same disagreement that’s been going on since Epicurus v. the Stoics. The “personal philosophical prejudice” responses are downright amusing. The overwhelming majorities in all three groups felt that their (scientific?!) interpretations of quantum information were directed by personal philosophical prejudices to some extent, and most participants believed they were influenced by prejudices “a lot.”

Some determinists argue that even if indeterminacy is a “real” component of quantum mechanics, the uncertainty doesn’t appear outside the quantum microcosm and thus isn’t relevant beyond that realm. However, Martin Heisenberg, a German neurobiologist and geneticist, has found evidence of random behavior that precedes and informs “lawful” behavior in animal life-forms as simple as bacteria.

Heisenberg is professor emeritus in the department of biology at the University of Wurzburg in Germany. Up until his retirement in 2008, he held the chair for genetics and neurobiology at the university’s Bio Center. He is a son of Werner Heisenberg, the physicist who formulated the uncertainty principle.

In an essay in the May 14, 2009, Nature, Heisenberg noted that although the debate on free will has focused on humans and “conscious” free will, even animal behavior is not altogether involuntary. He referenced his observations of bacteria whose flagella can make them tumble in a random manner (by rotating clockwise) or move forward in a directed manner (by rotating counterclockwise and wrapping together). The bacteria do this in a two-step sequence, first by initiating random tumbles that generate alternate possibilities and then by moving forward while sensing the gradients of temperature, nutrients, and other relevant conditions along their bodies. If the environment seems favorable, the bacteria continue to move forward. If not, they try for a better outcome by tumbling again.

According to Heisenberg,

As with a bacterium’s locomotion, the activation of behavioral modules is based on the interplay between chance and lawfulness in the brain. Insufficiently equipped, insufficiently informed and short of time, animals have to find a module that is adaptive. Their brains, in a kind of random walk, continuously preactivate, discard and reconfigure their options, and evaluate their possible short-term and long-term consequences. The physiology of how this happens has been little investigated. But there is plenty of evidence that an animal’s behavior cannot be reduced to responses. For example, my lab has demonstrated that fruit flies, in situations they have never encountered, can modify their expectations about the consequences of their actions. They can solve problems that no individual fly in the evolutionary history of the species has solved before. Our experiments show that they actively initiate behavior.

Robert O. Doyle is an associate of Harvard University’s astronomy department. In conformity with Heisenberg’s findings, Doyle has articulated a two-stage model for decision making, particularly in higher animals and humans. In the first stage, the mind randomly generates alternative possibilities; in the second stage, the mind evaluates the possibilities and selects one. The two stages can overlap and go back and forth; when none of the available alternatives seems acceptable, the decision stage may switch back to the random generation stage in an effort to identify further options.

Doyle maintains that randomness plays a central role in both evolution and biological life strategies. He suggests that the two-stage model is the engine that powers creativity and that it has evolved from the two-step process observed by Heisenberg among bacteria and fruit flies. The progression from behavioral freedom in lower animals to free will in humans has largely resulted from the evolution of an increasing complexity in the second (decision) stage.

Evolution has resulted in the following levels of selection, in ascending order:

  • Natural selection proceeding from reproductive success;
  • Instinctive selection proceeding from genetically transmitted selection criteria;
  • Learned selection proceeding from criteria acquired from the environment, including parents and peers;
  • Predictive selection involving the use of imagination and foresight to evaluate potential outcomes; and
  • Normative selection, in which deliberation about cultural values may affect the choice of behaviors.

An organism that can make selections at a given level can concurrently bring the earlier levels into the process. For a human, every one of the selection levels may be included in making a single decision. These selection processes incorporate both a conscious and an unconscious generation and evaluation of ideas, concepts, and criteria.

The first stage of the two-stage model (random generation of ideas) might be influenced by any or all of the following factors:

  • Individual genetic makeup, particularly with respect to brain physiology;
  • Preexisting habits and inclinations determined by education and prior experiences;
  • Free-flowing association and synthesis of old and new concepts and ideas, which can continue right up to the microsecond before a decision is made; and
  • External variables, such as random events or the ideas of other people.

The influence of the first two factors is not random, but it’s not entirely deterministic either. In most cases, these variables would create tendencies to favor particular outcomes. The third and fourth factors are subject to random variability, which in turn permits the operation of free will.

The choice itself (second stage) is semi-deterministic, because physiological and experiential influences are apt to carry much more weight than they did during the first stage. (Doyle refers to this lawlike selection of a response as “adequate determinism.”) The choice also may be influenced by the individual’s ethical code. As noted above, the mind’s random generation of ideas (stage 1) can overlap into the second stage, and switchbacks from one stage to the other are not uncommon. Accordingly, outcomes for similar circumstances may vary in the instant of decision making, and no particular choice can be guaranteed.

The theory that only two interpretations are possible in the determinism-versus-free will debate (either absolute predetermination or a totally random universe) is integral to the arguments for determinism presented in a recent article by Jonathan M. S. Pearce, “Is Society Accepting That Free Will Is an Illusion?” (Free Inquiry, June/July 2015). According to Mr. Pearce, “In simple terms, something is either caused or random, and neither situation seems to easily allow for moral responsibility as convention understands it.”

As presented, the “random” horn of this dilemma posits a situation lacking any causal influence, which would be unlikely to occur in the real world. After setting fire to this highly flammable straw-man, Pearce concludes that the only remaining alternative is hard determinism. He goes on to cite what he describes as “a whole plethora of scientific evidence.”

His examples correctly illustrate that many behaviors, capabilities, and belief systems can be predicted to statistically significant extents based on the genetics, genders, and/or experiential histories of the members of a given study group. But as noted above, these characteristics only create tendencies toward certain behaviors and choices. The probability that one or more of these tendencies will result in an expected behavior may be significantly greater than chance, but rarely, if ever, will it approach certainty (a probability of 100 percent).

Even more important, the probabilities only apply at a group level and are not predictive for individuals. According to statistics, as a male I am 88.2 percent more likely to commit a violent crime than I would be if I were female. Yet like the vast majority of human males, I have managed to avoid committing even one violent crime.

To the extent that free will exists, it must be enabled by the same kinds of electrochemical processes within the brain that generally govern our thoughts and perceptions. Accordingly, its underlying mechanism can be inhibited, altered, or destroyed by injury, disease, or chemical changes. But in a “normal” brain some ability to choose seems apparent, even though internal and experiential variables may increase the probability of one choice being made over another. This ability would encompass Doyle’s levels of selection and the two-stage decision model, all of which must be physiologically enabled. The fact that free will must emanate from physical processes does not change the likelihood of its existence.

As far as we can determine, the universe operates within a framework of laws governing the properties and interactions of mass and energy. There is no reason that these universal laws cannot encompass both deterministic and random elements. Neither extreme of the determinism-versus-free-will debate is supported by a preponderance of the evidence. Although I am neither scientist nor philosopher, you may take these assertions as an official expression of my scientific personal philosophical prejudice.

Further Reading

  • Doyle, Robert O. “The Two-Stage Solution to the Problem of Free Will.” Available at www.ucl.ac.uk/~uctytho/DoyleOnDoyle1.pdf.
  • Heisenberg, Martin. 2009. “Is Free Will an Illusion?” Nature 459.
  • Norsen, T., and S. Nelson. 2013. “Yet Another Snapshot of Foundational Attitudes Toward Quantum Mechanics.” arXiv:1306.4646.
  • Schlosshauer, M., J. Kofler, and A. Zeilinger. 2013. “A Snapshot of Foundational Attitudes Toward Quantum Mechanics.” Studies in History and Philosophy of Modern Physics: 222–30.
  • Sommer, C. 2013. “Another Survey of Foundational Attitudes Toward Quantum Mechanics.” arXiv:1303.2719.

Dan Davis

Dan Davis is a writer and a semiretired sales and use tax-consultant with a tendency to overthink things.


Determinism or randomness: Are these the only options? In the debate over free will, perhaps there is a middle way.

This article is available to subscribers only.
Subscribe now or log in to read this article.