Free will vs. neuroscience – a different analysis

The argument sometimes characterized as free will vs. neuroscience is the most recent form of the mind-body and, more generally, free will vs. determinism argument, going back at least to the ancient Greeks. It is relatively common of late for neuroscientists to discover something about what happens in the brain under various conditions, and then for someone else (usually not the neuroscientists) to say, “Neuroscientists have discovered that so-and-so is really nothing more than neurological process X occurring in structure Y of the brain.” The neuroscience is fascinating, but the interpretations are misleading and destructive. They essentially claim that people are nothing more than “meat machines,” and many have objected to them, but the objections have had little impact on the argument because the mind-body, free-will vs. determinism, issue has never been resolved. It hasn’t been resolved because, stated in that form, it can’t be. Accepting the fundamental division of a person into mind and body dooms us to endless and fruitless argument. This posting presents a different way to proceed, one that allows real resolution, and leads to affirming and appreciating the insights of neuroscience while avoiding the disastrous error of treating persons as unusual configurations of meat.

The appearance of a conflict between free will and neuroscience is due to an elementary error by interpreters of neuroscience, namely confusing correlation with causality. There are three kinds of neuroscience findings: 1) correlational: what happens neurologically when behavior B (doing mental arithmetic, making financial decision, etc.) or when condition C obtains (the subject is tired, the subject is in love, the subject is looking at something disgusting, etc.); 2) constraining: subjects become unable to do behavior B because some portion of the neurological “machinery” required for the behavior is broken; and 3) state-inducing: chemical C induces behavioral state S, ethyl alcohol being the archetypal and universally known example. None of these findings are causal. The constraining ones can look that way, but that’s because of sloppy thinking. If I tear a knee ligament so that I can’t play golf, it doesn’t show that playing golf is nothing more than leg movements, nor that knee movement cause golf swings. It does show that knee movements are part of playing golf, but nothing more.

Neuroscientists know all this, and rarely make the overblown, erroneous interpretations that the popular press loves. It seems to be primarily certain philosophers of science, especially those focused on the intersection of philosophy and psychology, that do so. It’s philosophers, after all, who worry about such things as “models of the world.” And that group of philosophers of science worry a great deal about causality because they think science is about finding “underlying” causes. If you actually observe scientists and talk to them, you find out that they don’t worry about causality; they worry about what happens, and finding it out, in greater detail: events and processes. Nobel Laureate Richard Feynman observed that physics is not about why, but about how.

Perhaps because they think science is about cause, philosophers of science focus virtually exclusively on the relationship of cause and effect – one particular relationship between A and B. That intense and exclusive focus very often results in overlooking the fundamental logical fact that causality is only one of an enormous number (actually, an unlimited number) of relationships between states of affairs. Having three sides and three angles does not cause a figure to be a triangle; breakfast does not cause lunch; the neurological processes that occur when a person decides what stock to buy do not cause the purchase; there is little doubt that important neurological events and processes occur when I see my wife of 28 years, with whom I am very much in love, but it would be grotesque – and, worse, utterly unscientific – to claim that that relationship is caused by those processes.
Trying to derive conclusions that reduce choice and behavior to neuroscience results is committing what philosophers call a “category error”. Behavior is not physiology; the range of possible facts about behavior includes physiology (including neurophysiology), but is much wider. Psychologists have struggled for decades to say clearly what they mean by “behavior,” to little avail: there is not, in the field of psychology today, an agreed-on definition of that concept. The most commonly accepted ones are either that behavior is a set of movements or, in the cognitive psychology school, a set of movements in response to external physical stimuli with intervening “mental” or “cognitive” processes mediating the response. Neither of these definitions is adequate for the task of articulating the concept of behavior — the range of facts about behavior is simply too broad.

The category error and insistence on causal relationships results in philosophical positions and beliefs that are logically incoherent. Specifically, determinism, and more generally any account of behavior that equates behavior with a process in response to a physical input, fails the reflexivity test. For example, as I type these comments, I am doing so because I value clarity in issues related to human behavior, I see (whether accurately or not) an opportunity to show respected colleagues and other interested persons a different way to approach an interesting conceptual issue, and am attempting to bring about a particular state of affairs, namely that others see that there is another approach and that it is of value. By doing that, I am attempting to change the accepted articulation of the concept of behavior in the world at large. In short, I am acting on what I take to be the case, to bring about states of affairs I value. The words I am typing at this moment are the physical performance that I take to be the best way to achieve those valued states of affair. If we suppose that what is really happening is only that my fingers are making motions to produce characters in an electronic document, mediated by various processes in my brain in response to recent physical inputs (the characters found in some popularizations of scientific findings) and perhaps other “stored” inputs, then it follows that this document is not actually part of an intellectual and philosophical discussion about the relationship between the brain and the mind, but is simply a sequence of characters typed on a computer and sent via physical signals to biological receivers (the brains of readers), which will emit various signals, perhaps including characters in reply postings, etc. If behavior equals physiology, then there is nothing but physiology: we do not actually have a question of “what is the relationship between the mind and the body,” because there is no such thing as a question — just physical signals; I am not attempting to show you a different way of looking at things, and in fact we are not actually having a conversation right now. In short, if it’s all physiology, there is nothing to discuss and, further, no such thing as discussion. Worse, there’s no such thing as commenting that there’s no such thing as discussion.

  1. Here’s the basic approach I am advocating. First, change the question, from “what’s the relationship between mind and brain” to “what’s the relationship between person and brain.” Why? Because it avoids buying into the basic assumption that the division is between mind and brain. (In my view, that’s the fundamental error that has led Western civilization down the garden path for a very long time.) To address the person-brain question:
  2. Start with the observation that there are persons behaving in the real world. Then observe that, whether or not we have a definition of behavior, we share the concept of behavior, that is, we can distinguish between behavior and other things. We know that because we can argue about what is a good definition of behavior — it we did not already have the concept of behavior, we could not disagree about how to define it. In the same way, if we did not share the concept of table, we could not disagree about whether a formal definition of it was correct. The most we could do would be to misunderstand each other.
  3. Articulate that concept, by identifying the ways in which behaviors can differ (just as we might identify the ways in which tables differ):
  • Two behaviors that are otherwise alike but are engaged in in order to bring about different results are counted as different
  • Two behaviors that are otherwise alike but are not acting on the same distinction are counted as different
  • If someone succeeds at some behavior by accident, rather than via the exercise of competence, we count it as different.
  • Two behaviors that are otherwise alike but involve different physical, observable performances are different.  Example: if I capture a Bishop by a forking attack or by discovered check, it’s still capturing the Bishop, a recognizable behavior in chess
  • Two behaviors that are otherwise alike but are expressions of different personality characteristics are counted as different behaviors.  Example: “Oh, he wasn’t angry — that’s just his style.”
  • Two behaviors that are otherwise alike but have different outcomes (by chance or due to unknown other facts) are counted as different.
  • The same behavior that is a way of accomplishing two distinct other behaviors is counted as different.  If I buy my wife flowers (Behavior 1) as way to celebrate our anniversary (Behavior 2), it’s a different behavior than if I buy her flowers to apologize (Behavior 3).

Putting these aspects together, in a formal notation, someone’s behavior is specified by specifying values for the parameters:

  • W (want to accomplish)
  • K (know — distinctions acted on)
  • Kh (know-how — competencies)
  • P (performance — the visible episode)
  • A (the actual achievement)
  • PC (personality characteristics)
  • S (the significance — what the person was doing, by doing that)

And finally we get to choice: the paradigm (the unmistakable, if-ever-there-were-a-case-this-is-it) case of human action is deliberate action: the person knows what they are doing (the K parameter includes the behavior they are doing) and wants (W parameter) to do it, of all the actions he knows he could do in these circumstances.

“Mind” is a reification of certain facts about human behavior, namely the W and K and, sometimes, the PC parameters.

What is called “free will” is a reification of the fact of deliberate action.  (Which is why I never argue about “free will vs. determinism” in its stated form.)

So where does the brain fit in this?  In exactly the place it has been empirically discovered to fit: when a person engages in an action, that action paradigmatically involves a physical performance — moving one’s hand and arm, emitting the sounds comprising the word “Check,” etc. Those performance are processes, and we can (and people have) studied those processes to find out the steps or stages that comprise them.  Nerves send signals to the muscles, the muscles contract in a certain way, etc. How is the signal produced in the nerves? By processes in the brain.  We can then study, as neuroscientists do, what happens in the brain when various behaviors — deliberate actions — are engaged in.

The analysis of behavior above may seem to incorporate concepts sometimes called “folk psychology” or, by some philosophers, aspects of a (discredited) theory of mind.  They are neither. In fact they are, in every case, formally defined concepts, “formal” in exactly the sense that mathematical logic articulations are formal.  The formal articulation of all these concepts may be found in What Actually Happens, by P. G. Ossorio.

All of this analysis is a summary restatement of work by Ossorio, who articulated it at length and in detail in several books.  The most easily available is The Behavior of Persons, on

This entry was posted in Descriptive Psychology, Neuroscience, Uncategorized and tagged , . Bookmark the permalink.

4 Responses to Free will vs. neuroscience – a different analysis

  1. cj stone says:

    “In short, if it’s all physiology, there is nothing to discuss and, further, no such thing as discussion. Worse, there’s no such thing as commenting that there’s no such thing as discussion.”

    That’s the good stuff. Nice work, Joe.

  2. tonyputman says:

    From the New York Times, 12/5/11: “Brains are low-power, nimble computing mechanisms …”

    And the beat goes on …

  3. brucetrago says:

    Thanks Joe for a great review from a slightly different perspective.

  4. Greg Colvin says:

    Excellent review, Joe. Tony, I don’t have a big problem with “Brains are low-power, nimble computing mechanisms …” I don’t think it is a good to describe brains, but I’m not a neuroscientist. Where I have problems is when that hypothesis becomes “People are low-power, nimble computing mechanisms.”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s