How could we rationally suppose that we lack free will?

[The following is a guest post by Bob Lockie. — JS]

He who says that all things happen of necessity can hardly find fault with one who denies that all happens by necessity; for on his own theory this very argument is voiced by necessity (Epicurus 1964: XL).

Lockie, Robert. Free Will and Epistemology: A Defence of the Transcendental Argument for Freedom. Bloomsbury, 2018.

Epicurus’s famous fragment is one of the earliest examples of a peritrope or ‘reversal’ argument – what we, after Kant, would call a transcendental or ‘self-refutation’ argument. It is also the first deployment of this argument type in perhaps its most classic area of application: the free will debates. This argument has never gone away.

Free Will & Epistemology is a monograph concerned with normative epistemology, free will, and the relationship between these. The book’s summational aim is to defend a modern (i.e. not merely scholarly) version of the famous transcendental argument for free will: that we could not be justified in undermining a strong notion of free will, as a strong notion of free will is required for any such process of undermining to be itself epistemically justified. This is the first professional, in-depth defence of this argument for many decades, and the first to marshal the resources of modern epistemology in its development.

The incompatibilist freedom that this work argues for (‘self determinism’) is a freedom to be the source of one’s own actions, including mental actions. This idea of self determinism is separately defended with strongly emergentist arguments – arguments which make reliance on the neuroscientific and cognitive-psychological literature. It is these psychological and neuroscientific arguments that the readers of The Brains Blog may find most of interest in the book. Chapter Four, the longest chapter of the work, articulates an account of epistemic freedom – of freedom of thought – which stands in stark opposition to much philosophical commentary on the sciences of cognition: commentary which emphasises the involuntary, automatic, fast, parallel, triggered, old mind, lower-brain or more posterior cortical, mandatory operation, (often modular) nature of cognition. In contrast, this work emphasises the distinctively human powers of executive functioning – of slow, serial, largely conscious, controlled, new-mind, often language-mediated, high level, abstract, values-laden, rational and (at least potentially) justified cognition.

Chapter four, and the book as a whole, provides an exhaustive account of our powers of executive, controlled, agential thought, and an account of epistemic justification in terms of thinking how we ought. This chapter involves a fairly exhaustive review of the neuropsychological and psychological literature on executive functioning – generously conceived, to include cognate, overlapping areas in psychology more generally. The influence of Luria, Fuster, Shallice and the clinical neurology literature make their presence felt – alongside Baddeley, Miyake & Friedman, Badre & D’Esposito, Koechlin and the more experimental neuropsychological literature. Additionally, the clinical and educational self-regulation literatures and the extended and embodied cognition literature inform this chapter. A freedom of executive control is defended against charges of homuncularism or regress. It is argued that our high level thought involves mental agency, where that involves efferent as well as afferent cognition – what Fuster calls ‘perception-action cycles’. It is argued that the deflationary tendency in epistemology and the philosophy of the cognitive sciences has greatly over-emphasised the afferent, passive and automatic in human thought; to the exclusion of the efferent, agential and executive. Against this deflationary tendency, it is painstakingly argued with reference to a wide range of overlapping neuropsychological literatures that we have very considerable powers of freedom of thought and self-determination more generally: that we are cognitive agents.

These constructive, scientific arguments, coexist in the book with the more sweeping, aprioristic transcendental argumentation – that we have to presuppose freedom. Were determinism true, the argument goes, we would lack the control over our cognition needed to think how we ought – the levels of control needed for our thought to be epistemically justified. But then, the argument goes, we would not be epistemically justified in anything at all – including our deflationary arguments against freedom. Those interested in deciding whether to invest the time and effort to read the work as a whole can find a specimen chapter (Chapter 8: admittedly one of the most aprioristic) here. Commentary on the work and a 35% discount code for readers of the Brains Blog may be found here.


  1. Carl Gillett

    Hi Bob, many congratulations on the book which also sounds really interesting.

    I am interested to read the line of thought on a slow-track, executive function-based picture of one form of free-will.

    One quick question, since it is one of my obsessions do you have a quick outline of what you mean by a “strongly emergentist” view? It sounds as if you take everything to be done by us through our brain and certain of its properties, and hence for all of our properties (and activities) to be composed. I take one form of strong emergence to be compatible with such composition — just wanted to see if you endorse full composition of our properties and activities alongside your “strong emergence”? I hope you do.

    Anyway, I look forward to reading the book in the summer, very best, Carl

  2. A woman goes into a restaurant, sits down, peruses the menu for a minute, and then calls the waiter over to give him her order.

    We cannot see what is going on in her head. But, assuming she has a normal brain, we expect it to be similar to what happens when we do the same thing, read a menu, evaluate our options, and place an order.

    We call this empirical event “choosing”. Options are presented, evaluated by some comparative criteria, and a choice is output.

    So, we know what happened (a choice was made) and we know who did it (the woman in the restaurant). There are no “illusions” involved here. It is simply an empirical fact.

    As to whether she was “free” to decide for herself what she “will” do, that too seems to be a matter of empirical fact. We saw no one holding a gun to her head, literally or figuratively, and telling her what to order.

    So, her freedom, to make the choice herself, also seems to be a matter of empirical fact, and not any kind of “illusion”.

    But, wasn’t her choice causally inevitable? Wasn’t it a result of her own genetic dispositions and tastes, her own beliefs and values, her own life experiences, her own thoughts and feelings, her own neurology?

    Of course it was. But all of that stuff is her. None of those things were external to who and what she was at the time she made her choice.

    So, it appears that causal necessity and free will are not incompatible after all. (A) Whenever someone decides for themselves what they will do, according to their own purpose and their own reasons, then it is authentic free will. (B) Whenever someone decides for themselves what they will do, according to their own purpose and their own reasons, then it is authentic determinism.

  3. Sorry, but this just misses the interesting point raised by, say, Robert Kane and other indeterminists/incompatibilists. Suppose the woman in the restaurant has a medical condition, and her physician has imposed dietary restrictions. Today, she decides to run the risk of eating something that, medically speaking, she ‘shouldn’t.’

    Does it make any sense to say that, in some moral/ethical sense, she is dong the wrong thing? If you are an absolutist about bodily autonomy, up to and including suicide, you will presumably contend that no one should judge this. If you are a consequentialist, though, you might say that her poor choice of food could lead to an ER visit imposing various costs on other parties, other patients, the doctors, insurers, possibly taxpayers, etc., and you might say she was doing something wrong in a sense that goes beyond ‘wrong for her.’

    But … this is the James/Berlin/Kane point — before we even get to all of that we ought to have a sense of whether there was an alternative. Whether, once she has chosen meal X, it is even rational to suppose there was a possible other world in which she might have chosen meal Y. The authors I’ve named think it is rational to suppose that, and they are all rather happy about that conclusion, because it allows for moral judgments.

    Of course there are many philosophers who disagree, who say that if she chooses X, then the Y-meal and its world path are and have always been impossible. Some of them are rather happy about THAT because they are incompatibilists too, and say, “good riddance” to the whole body of moral judgments.

    Dismissing an issue is one thing. Addressing it another.

  4. “Were determinism true, the argument goes, we would lack the control over our cognition needed to think how we ought – the levels of control needed for our thought to be epistemically justified. But then, the argument goes, we would not be epistemically justified in anything at all – including our deflationary arguments against freedom.”

    I’m unclear on where the inconsistency lies.

    Were determinism true, whether I can do arithmetic is completely determined by my genetics, education, brain functioning, etc. How is this incompatible with me justifiably coming to believe that 2 + 2 = 4?

    If I’m determined to do math correctly, then there will be reasons to conclude that 2 + 2 = 4, and I will correctly discern those reasons. Those reasons cause me to believe that 2 + 2 = 4; they *determine* my belief. In this case how I’m constructed lines up with those reasons, just like a calculator is constructed lines up with mathematical truth.

    On the other hand, if I’m determined to do math incorrectly, I will incorrectly discern the sum 2 and 2. My reasoning is faulty; I’m like a calculator that has been programmed incorrectly. The output produced by the calculator does not line up with mathematical truth.

    2 + 2 = 4 is true independently of anything I think. I either discern that truth or I don’t. Why do I need contracausal freedom to discern it correctly? A calculator does not.

    I don’t see how my my correctly discerning that determinism is true (assuming that it is true) is incompatible with my discernment being determined. Is the argument that reasons cannot be causes–and if so, why think that?

Comments are closed.