On free will-10: Ethical and legal implications of free will as simply a veto power

(For previous posts in this series, see here.)

The idea suggested by Benjamin Libet that what we call free will is not the popularly assumed ability to decide all our seemingly deliberate (as opposed to instinctive) actions but consists of the more limited ability to either let the predetermined action be completed or to veto it may be unsatisfying to some but its implications are worth exploring in case it turns out to be true. What this model says is that I have no control over what I decide to do in any given situation but I do have control over whether that decision is actually carried out. In other words, I cannot control my thoughts and decisions but I can control (within a limited range) my actions.

Libet suggests that if, as he believes, our decisions to act are involuntary but the decision on whether to allow that act to be carried out or to veto it is freely arrived at, that should influence how our justice and moral systems should work.

How do our findings relate to the questions of when one may be regarded as guilty or sinful, in various religious and philosophical systems? If one experiences a conscious wish or urge to perform a socially unacceptable act, should that be regarded as a sinful event even if the urge has been vetoed and no act has occurred? Some religious systems answer ‘yes’. President Jimmy Carter admitted to having had urges to perform a lustful act. Although he did not act, he apparently still felt sinful for having experienced a lustful urge. But any such urges would be initiated and developed in the brain unconsciously, according to our findings. The mere appearance of an intention to act could not be controlled consciously; only its final consummation in a motor act could be consciously controlled. Therefore, a religious system that castigates an individual for simply having a mental intention or impulse to do something unacceptable, even when this is not acted out, would create a physiologically insurmountable moral and psychological difficulty.

Religious systems like Christianity punish people for even thought crimes, because god is apparently monitoring everyone’s thoughts all the time to check for any transgressions. But this makes no sense if we have no control over our thoughts (granting for the sake of argument that the idea of a god who can read everyone’s thoughts makes any sense at all). I am sure that if this understanding of the brain ever becomes firmly established, the ever amenable and highly flexible theologians will come up with new interpretations of their holy books to say that how they interpreted them earlier was wrong and that the correct interpretation is that thoughts alone are not sinful. This kind of theological flexibility to accommodate the latest science has been the pattern so far.

Libet argues that secular ethical systems will also have to adjust, though not as much.

Ethical systems deal with moral codes or conventions that govern how one behaves toward or interacts with other individuals; they are presumably dealing with actions, not simply with urges or intentions. Only a motor act by one person can directly impinge on the welfare of another. Since it is the performance of an act that can be consciously controlled, it should be legitimate to hold individuals guilty of and responsible for their acts.

The idea that we should only punish people for their actions and not their thoughts has implications for so-called ‘hate crimes’ legislation whereby people are punished more harshly for the same act if their actions are deemed to arise from intent to harm someone because of animosity towards their victim’s race or ethnicity or gender or sexual orientation. If what people think is beyond their control, it does not make sense to factor it in when determining punishment. (Since I already oppose hate crime legislation for other reasons, I would not be sorry to have a scientific reason to eliminate them.)

Libet makes a final plaintive plea for retaining the idea of free will by suggesting that according to the evidence, at least at the time he wrote it in 1999, believing in free will is at least as good an option as determinism, and so we should hold on to it until more conclusive evidence against it turns up.

My conclusion about free will, one genuinely free in the non-determined sense, is then that its existence is at least as good, if not a better, scientific option than is its denial by determinist theory. Given the speculative nature of both determinist and non-determinist theories, why not adopt the view that we do have free will (until some real contradictory evidence may appear, if it ever does). Such a view would at least allow us to proceed in a way that accepts and accommodates our own deep feeling that we do have free will. We would not need to view ourselves as machines that act in a manner completely controlled by the known physical laws.

Benjamin Libet died in 2007 and thus did not have to confront the results of experiments carried out the very next year that seem to provide just the kind of evidence contradicting free will that he clearly hoped would not materialize. I will discuss these in the next post in this series.

Next: Recent fMRI studies of the time sequence of decisions and actions.

On free will-9: Attempts to salvage free will

(For previous posts in this series, see here.)

People who are determined to keep the Ghost in the Machine alive still have a few options. Ironically, although it was Libet’s early experiments that cast doubt on the idea that we have free will, he himself was disturbed by that implication and has sought to find ways to salvage it. In his many publications, he repeats his belief that his experiments did not rule out free will and suggests ways in which it could still operate.

The source of his belief in free will is similar to the reasons that we all cling on to, that we strongly feel that we act freely and that it is not desirable to abandon belief in it. As he says, (Do we have free will?, Benjamin Libet, Journal of Consciousness Studies, vol. 6, No. 8–9, 1999, pp. 47–57):

However, we must recognize that the almost universal experience that we can act with a free, independent choice provides a kind of prima facie evidence that conscious mental processes can causatively control some brain processes (Libet, 1994). As an experimental scientist, this creates more difficulty for a determinist than for a non-determinist option. The phenomenal fact is that most of us feel that we do have free will, at least for some of our actions and within certain limits that may be imposed by our brain’s status and by our environment. The intuitive feelings about the phenomenon of free will form a fundamental basis for views of our human nature, and great care should be taken not to believe allegedly scientific conclusions about them which actually depend upon hidden ad hoc assumptions. A theory that simply interprets the phenomenon of free will as illusory and denies the validity of this phenomenal fact is less attractive than a theory that accepts or accommodates the phenomenal fact.

Of course, the idea that we should retain belief in free will simply because everyone believes in it and the alternative is unpalatable is hardly persuasive as a scientific argument. So in a later paper published in 2002 (Consciousness and Cognition, vol. 11, 291–299, 2002), Libet goes further and suggests that we are not entirely the slaves of our subconscious decisions to act because we retain the power to veto a decision at the last minute. He argues that veto decisions may not follow the same pattern of arising from prior unconscious neural activity because the decision to veto an act may arise from a different source than the decision to take a specific act and thus not follow the pattern of precursor behavior that he observed about the decision itself. It is thus possible that a veto decision could be freely made. He says (referring to himself in the third person):

However, Libet noted that the conscious function still had enough time to affect the outcome of the process; that is, it could allow the volitional initiative to go to completion, it could provide a necessary trigger for the completion, or it could block or veto the process and prevent the act’s appearance. There is no doubt that a veto function can occur. The argument has been made that the conscious veto process would itself require preceding developmental processes, just like a conscious sensory awareness. But Libet (1999) argued that the conscious veto in a control function, different from awareness per se, need not be a direct product of the preceding processes, as is the case for simple awareness.

Libet suggests that although our decisions to carry out an act may be involuntary, the veto is different, and thus provides a new understanding of the role of free will.

I propose, instead, that the conscious veto may not require or be the direct result of preceding unconscious processes. The conscious veto is a control function, different from simply becoming aware of the wish to act. There is no logical imperative in any mind–brain theory, even identity theory, that requires specific neural activity to precede and determine the nature of a conscious control function. And, there is no experimental evidence against the possibility that the control process may appear without development by prior unconscious processes.

The role of conscious free will would be, then, not to initiate a voluntary act, but rather to control whether the act takes place.

Free will that can only act to veto an involuntary decision seems a rather weak substitute for the real thing. Those who want a bigger role for free will can, if they wish, can go beyond Libet’s limited concept and postulate that a ghostly free will does decide and is the cause of the early unconscious neural activity but that for some reason it does not produce any brain activity and also hides its early decisions from us, perhaps by wiping out that initial memory and creating our conscious thoughts only much later. In other words, when we become consciously aware of making a decision, that awareness is merely an echo of the same decision that was made earlier prior to the unconscious neural activity, but of which we are not aware. Daniel Dennett’s theory of consciousness (Consciousness Explained, 1991) says the brain is constantly writing and rewriting our personal narratives of what we experience in order to create a coherent narrative so this kind of mental revisionism could explain why we are not conscious of our early conscious decisions.

But why would the brain bother to go to this level of subterfuge? It is hard to think of an evolutionary advantage that is conferred to the organism by the brain covering its tracks in this way. These kinds of explanations for free will soon start to look suspiciously like the pseudo-explanations religious believers give for why we cannot see any evidence for their peripatetic god even though he is supposedly always busy doing stuff.

But even allowing for the possibility that our brain somehow hides the existence of free will, this would not seem to offer much consolation. What would be the point of decisions that are made before we are conscious of making them? The idea of free will gets its power from the fact that we consciously make the decisions that trigger motor activity.

Next: Ethical and legal implications of free will as a veto power

On free will-8: The 1983 and later experiments of Benjamin Libet

(For previous posts in this series, see here.)

In 1983, Benjamin Libet and his associates did some experiments that were similar to the 1963 Grey Walter experiment but with the added feature that the patients could observe the equivalent of a clock and thus note when they made the decision to act. This enabled a more objective determination of the time when they first had the conscious thought to carry out the action and not depend upon a possibly misleading feeling of surprise to infer the ordering of events.

One of the key original papers was published in the journal Brain (Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential): The unconscious initiation of a freely voluntary act, vol.106, p. 623-642, 1983) which does not seem to be available online but you can read online a later review published by Libet in 1999 (Do we have free will?, Journal of Consciousness Studies, vol. 6, No. 8–9, 1999, pp. 47–57) where he summarizes his findings and its implications for free will.

Using electrophysiological measurements of something called the readiness potential (RP) in the brain to detect unconscious brain activity, Libet and his co-workers asked people to move a finger (M) and also to indicate when they made a conscious decision to want to move the finger (W) by observing a clocklike device. The part of the brain where the readiness potential originates is called the supplementary motor area (SMA) and is the part involved in motor preparation, i.e., prior to taking an action. According to the free will model (D), there should be a definite temporal sequence in which an act of will (which cannot be detected experimentally) should occur first, followed by the conscious thought to do so (W, determined by the clock reading as noted by the patient), then unconscious brain activity (RP, measured using an EEG device), and behavior (M) last.

What they found was that while the RP time did precede M by an average of about 550 milliseconds (i.e., a little more than half a second), it also preceded W by about 350 milliseconds. The brain seemed to have made the unconscious decision to move the finger before the subject was aware of having made the decision to do so, suggesting that the actual temporal sequence of events was unconscious neural activity, followed by conscious decision to take an action, followed by the action. This was the same result as the Grey Walter experiment except for the crucial additional features that the Libet experiment was able to quantify the time intervals involved, had a more objective measure of when the conscious decision was made, and was able to locate the part of the brain where the precursor activity was occurring.

In other words, what we think of as our will (as manifested by our conscious thoughts) may be just an afterthought. What may be happening is that our unconscious neural activity makes a decision and then sends two signals out, one to create a conscious thought that we have decided to take an action and the other to actually take the action. Rather than the thought being the cause of our actions, our conscious thoughts are merely a passive recognition, after the fact, of decisions made unconsciously without a deliberate act of will.

In his 1999 paper, Libet spelled out what he thought was at stake in the question of whether we have free will or not.

The question of free will goes to the root of our views about human nature and how we relate to the universe and to natural laws. Are we completely defined by the deterministic nature of physical laws? Theologically imposed fateful destiny ironically produces a similar end-effect. In either case, we would be essentially sophisticated automatons, with our conscious feelings and intentions tacked on as epiphenomena with no causal power. Or, do we have some independence in making choices and actions, not completely determined by the known physical laws?

For example, actions by a person during a psychomotor epileptic seizure, or by one with Tourette’s syndrome, etc., are not regarded as actions of free will. Why then should an act unconsciously developed by a normal individual, a process over which he also has no conscious control, be regarded as an act of free will?

As one might expect with such a controversial result, others attempted to replicate Libet’s results and while there seemed to be a general consensus that there was nothing faulty about his data or his methods, their inferences were challenged. Judy Trevena and Jeff Miller were the chief skeptics who have had an ongoing back-and-forth with Libet. In their recent 2010 paper, Trevena and Miller do not dispute the finding that brain activity occurs before an awareness of the decision to move the finger. (Brain preparation before a voluntary action: Evidence against unconscious movement initiation, Judy Trevena and Jeff Miller, Consciousness and Cognition, vol.19, Issue 1, March 2010, Pages 447-456.)

What they find is that the strength of the brain activity is independent of whether a decision is made to move the finger or to not move the finger. They argue that hence one cannot take the RP signal as an unconscious decision to move the finger but that it must signify something else. To quote their own words:

We tested that assumption by comparing the electrophysiological signs before a decision to move with signs present before a decision not to move. There was no evidence of stronger electrophysiological signs before a decision to move than before a decision not to move, so these signs clearly are not specific to movement preparation. We conclude that Libet’s results do not provide evidence that voluntary movements are initiated unconsciously.

Furthermore, they say,

As with the movement-preceding negativity, we found no evidence that prevailing conditions in the brain just before a spontaneous decision can predict the outcome of that decision—namely, the spontaneously selected response hand. Thus, our results appear to contradict the idea that our spontaneous conscious decisions merely consist of “going along” with whatever our brains were going to do anyway.

But Trevena and Miller did not give believers in free will much to cheer about. They were careful to say that, “nothing in our results suggests that conscious decisions are produced by anything other than neural activity”, thus throwing cold water on the idea that there is an entity called the will that exists independently of the physical brain and makes the decisions. All they are saying is that the RP signal experiments of Libet only provide evidence of unspecific neural activity prior to an action and are not predictive of the actual action, and are hence not evidence of a decision.

Given the results of his experiments on free will, one might reasonably conclude that Libet is not a believer in it. What is interesting, as I will discuss in the next post in this series, is that it is Libet himself who, despite the evidence of his own experiments, defends the idea of free will and tries to find ways to retain it in the face of his own data.

Next: Trying to salvage free will

On free will-7: How reliable a historian is the brain?

(For previous posts in this series, see here.)

In post #6 in this series, I discussed the 1963 Grey Walter experiment in which patients who had electrodes implanted in their brain’s motor cortex that could send a signal to advance a slide were surprised that the projector seemed to anticipate their decision to advance the slide. Does this mean that their unconscious neural activity had decided to advance the slide before telling the conscious brain that it had decided to do so? If so, it seriously undermines the idea of free will. In his book Consciousness Explained (1991, p. 167) which discusses the experiment, Daniel Dennett warns that it is premature to accept this conclusion because it is based on the articulated sense of surprise reported by the patients, and the brain is not the most reliable of historians.

In the ‘multiple drafts’ model that Dennett suggests of how our brain works, the brain is not a recording and playback device that faithfully captures all that is going on around us. Instead, it is constantly creating different narratives to make sense of our experiences. These narratives are rapidly thrown into our consciousness and then disappear, to be replaced by new ones. The version that ‘sticks’ in our consciousness and become retained as the ‘true’ memory of what happened may not necessarily be the one that is the most faithful to the events as they actually happened.

The freedom to construct stories is not absolute, however, since our brain imposes some rules that restrict which narrative is selected as the ‘final’ one. For example, major sensory events have to be retained. If you definitely saw your friend Joan entered the room before your friend George, then the narratives will retain that order. But if you did not carefully note Joan’s facial expression as she entered, your narratives might vary with her looking happy or sad or angry depending on the need of the narrative.

The narratives that our brain constructs will also conform to our strong expectations (based on our prior experiences and learning) of how the world works and will not produce narratives in which, for example, people fly around because that would be going against what we strongly believe. If you stubbed your toe and felt pain, the narrative will not replace the pain with pleasure. However when we sleep, the brain enforcer also relaxes and so our dreams can be much wilder, though even there our senses impose some constraints. We have all experienced times when we were asleep and some sound that happened in our surroundings (a siren or telephone or door bell) seemed to be seamlessly woven into the narrative of our dreams. This is because of our brain’s ability to rapidly produce a narrative that incorporates any external reality.

We can think of the brain not as a scrupulous historian but as more like a highly imaginative, quick, and prolific writer of historical fiction. As with such authors, certain rules apply. There are certain anchoring events that cannot be changed (World War II must come after World War I, for example) but there are many things that are open for speculation because they are not firmly anchored by our senses or records or our memories or our understanding of how the world works, and the writer has great freedom to invent narratives that are plausible that connect the events that we are sure about. The version that is finally chosen to be published could well be based on idiosyncratic factors that have nothing to do with how accurate the story is about all these unanchored details.

This model of the brain also explains why our memories are so unreliable about some things and why we have the phenomenon of ‘false memory’. Witnesses to a crime turn out to be extraordinarily unreliable when reporting from memory on things that happened right before their eyes, which is why it is desirable to jot down notes at once for any thing that we want an accurate record of later.

I think all of us have experienced instances when something that we were sure happened in the past is challenged by someone who was also there (a relative or an old friend) who has a different recollection. What likely happened in such cases is that when we recounted that event, our brain picked out from all the narratives one that filled in all kinds of details that may not have happened but which made the story make sense or more interesting, and the repeated verbal recounting of these false events then became anchored so that all future narrative constructions treated these as incontrovertible facts, cementing them even further. Robert and Tamar Krulwich tell Ira Glass of This American Life an amusing story that perfectly illustrates how false memories can arise.



In the Grey Walter experiment, the brain expects to receive visual feedback (the slide advancing) on the successful execution of an act (pushing the button). The patients experience surprise when the feedback arrives earlier than expected. But does the feeling of surprise necessarily arise from the fact that (as the patients reported) the conscious thought to advance the slide came after the slide actually advanced?

The multiple drafts model gives us an alternative possibility. Because of our prior experiences, our unconscious brain is conditioned to know that effects do not occur instantaneously after causes. There is always a small time lapse involved in signals being sent and received though this may be so small that our conscious brain may see things as instantaneous. Because of this the unconscious brain expects feedback to occur after a certain minimal short time interval (say 300 ms) has elapsed after the decision to act. Any feedback that arrives before that time interval has elapsed may cause the unconscious brain to react in surprise, even though the cause may still precede the effect.

So it is quite possible that even if the slide advanced after the patient pushed the button and there was no seeming denial of free will or violation of causality, the unconscious brain still reacted with surprise because the time interval between the push and the slide advancing was less than expected. It is possible that the brain, always constructing narratives to make sense of things, might be ‘explaining’ its own sense of surprise by creating a fictional storyline in which the patient is made to think that the slide advanced before the decision was made, when in actual fact it is possible that the conscious thought did precede the act (thus maintaining the idea of free will) but by a time interval less than the expected one (thus causing the sense of surprise).

Grey Walter did not carry out follow up studies to investigate this question and it had to await the 1983 studies of Benjamin Libet.

Next: The famous and controversial Libet studies.

On free will-6: The 1963 Grey Walter experiment

(For previous posts in this series, see here.)

In the previous post, I provided a schematic description of two models of how the brain works, one with free will and the other without it. The traditional brain model with free will is given by

(D)                                    GES
                                        ↓
will → conscious thoughts → unconscious neural activity → action

Our genes (G), environment (E), and the inherent randomness in the laws of nature (S) all contribute right up to the present instant to the brain’s structure and unconscious neural activity. But in this model, there is a separate branch in which our (uncaused) free will makes decisions first which manifests itself as a conscious thought. In this model there should be a definite temporal sequence in which the act of will occurs first, followed by conscious thoughts, then unconscious brain activity caused by that conscious thought, and finally the action.

The model without free will is given by

(G)                                GES
                                    ↓
conscious thoughts/will ↔ unconscious neural activity → action

In this model, since conscious thoughts and what we think of as free will are not prior to unconscious neural activity but instead are the products of it, they need not temporally precede it.

The way that researchers investigate whether the idea of free will is tenable is by looking at the time sequence of events. One of the earliest experiments that threw the traditional model of (D) into doubt was done by neurosurgeon W. Grey Walter in 1963. Daniel Dennett, (Consciousness Explained, 1991, p. 167) describes the experiment.

Grey Walter performed his experiment with patients in whose motor cortex he had implanted electrodes. He wanted to test the hypothesis that certain bursts of recorded activity were the initiators of intentional actions. So he arranged for each patient to look at slides from a carousel projector. The patient could advance the carousel at will, by pressing the button on the controller. (Note the similarity to Libet’s experiment: This was a “free” decision, timed only by an endogenous rise in boredom, or curiosity about the next slide, or distraction, or whatever.) Unbeknownst to the patient, however, the controller button was a dummy, not attached to the slide projector at all! What actually advanced the slides was the amplified signal from the electrode implanted in the patient’s motor cortex. (My italics)

As far as the patient was concerned, and according to the model (D) that has free will, the temporal sequence the patients expect should be conscious thought → button push → slide advance. But the direct measurement of motor cortex brain activity introduces a new time step that is unknown to the patient but can be measured by the researchers. As a result, if free will exists, the patient should first become aware of making a decision, then send a command to the motor cortex, which produces both the amplified signal (which causes the slide to advance) and sends a signal to the finger to push the button. If the slide advanced after the patient was conscious of making a decision to push the button but before the button was actually pushed, that would definitely puzzle the patients because they were under the impression that it was their pushing of the button that advanced the slide. But all it would really imply to the researchers is that the speed with which the motor neuron activity sends an electrical signal to the slide projector is greater than the speed with which the motor neuron sends the push signal to the finger.

So what happened? Dennett continues the story:

One might suppose that the patients would notice nothing out of the ordinary, but in fact they were startled by the effect, because it seemed to them as if the slide projector was anticipating their decisions. They reported that just as they were “about to” push the button, but before they had actually decided to do so, the projector would advance the slide – and they would find themselves pressing the button with the worry that it was going to advance the slide twice! (My italics)

In other words, the motor cortex activity that triggered the slide advance seemed to occur not only before the finger pushed the button but even before the patients said they were conscious of making the decision to push the button. This experiment could be interpreted as an early indication that there was a spike in brain activity about half a second before the person was conscious of making a decision to carry out an action.

At first glance, this experimental result might seem to be a devastating blow to the idea of free will. If the brain’s unconscious neural activity makes and executes a decision before a person is conscious of making that same decision, then that refutes the expected temporal sequence that is at the heart of the model (D) that has free will in it.

But we have to be careful of jumping to that conclusion. There is a danger of over-interpreting these results because the experimenter is dependent on the patients’ reporting of when they had the conscious thought and Dennett argues that the brain is not a reliable source of information about its own workings, for reasons to be outlined in the next post in this series.

Next: How reliable a historian is the brain?

On free will-5: Models of how the brain works

(For previous posts in this series, see here.)

It is time to look at specific models of how the brain works.

In the previous post, I pointed to a paper by biologist Anthony Cashmore which argues that our brains are the product of genes (G), environment (E), and stochastic (i.e., random) processes (S). This GES combination influences the unconscious neural activity in our brains, which in turn gives instructions to the motor neurons that control our actions. So the causal and completely physiological chain goes like (A):

(A) GES → unconscious neural activity → action

The directions of the arrows signify the causal relationships. Our bodies are in a state of constant activity, with hearts beating, blood flowing, digesting food, breathing, secreting chemicals, producing new cells and disposing of old one, and so on, all of which take place without us being aware of it. I think everyone (except those religious people who can’t bear to see god not taking part in every single activity) will accept that our brains control and moderate all this unconscious behavior. What is in dispute is what gets added on to this basic model.

The prevailing assumption is that in addition our unconscious neural activity, there is another part of us where we have conscious thoughts, that gives instructions to our unconscious brains to produce specific actions. So when I choose to pick up my pen, this conscious decision is transmitted to my unconscious neural network activity which somehow, in ways that are opaque to my conscious mind, tells the motor neurons what to do in order to execute the order. This leads to model (B):

(B) conscious thoughts → unconscious neural activity → action

These conscious thoughts are obviously products of the brain too. After all, when we die, our bodies cease to have conscious thoughts. The question is what causes these conscious thoughts to arise? Believers in free will assume the existence of yet another entity called the ‘will’ that acts on our conscious thoughts driving it in the directions that ‘we’ (i.e., our will) want it to go. Thus another element is added to the causal chain (B), with the will driving behavior via conscious thoughts and unconscious neural activity, resulting in the causal chain (C):

(C) will → conscious thoughts → unconscious neural activity → action

Note that chain (A), in which GES created the brain where the unconscious neural activity takes place, is still operational so (A) and (C) are both acting simultaneously and can be combined to give (D):

(D)                                    GES
                                        ↓
       will → conscious thoughts → unconscious neural activity → action

This is the schematic model of human behavior that has free will embedded in it and to which many subscribe, or would like to subscribe. The ‘will’ is the Ghost in the Machine. It is the ‘I’ that we like to believe represents the ‘real’ us, that makes decisions and is responsible for our actions. Model (D) represents Cartesian dualism is a schematic form. The will acts in the Cartesian Theater.

But now we have to deal with the problem of what caused or created the will and what it is made of. One has the problem of infinite regress unless one arbitrarily asserts that the will is not a product of the brain but somehow magically came into being at some point in our existence (like the ‘soul’), somehow has the ability to direct our conscious thoughts, and is subject to no further causal explanation. This is remarkably similar to the way that religious people think of god and try to elide the question of who created god, what god is made of, and how he interacts with the world, which likely explains why believers in god are also strong believers in free will. Once you are willing to ignore all the difficulties and believe in the unlikely existence of god, believing in free will becomes not only necessary but easy.

It is possible to eliminate this causality problem by saying that what we think of as the will is not a free-standing entity but is also simply a creation of the brain, another product of unconscious neural activity. If so, we would now have a closed causal loop described by (E), that begins and ends with the brain and thus has a purely material basis.

(E) unconscious neural activity → will → conscious thoughts → unconscious neural activity → action

In (E), will and conscious thoughts are both products of the brain and operationally indistinguishable and thus can be combined into one entity. Thus unconscious neural activity gives rise to conscious thoughts/will that in turn transmits decisions back to the unconscious neural activity, thus eliminating the need for the problematic autonomous, non-material, self-creating, independently existing will. In this model, conscious thoughts/will is an emergent property of the brain, so that (E) can be simplified to (F) where,

(F) conscious thoughts/will ↔ unconscious neural activity → action

Note the causal arrow going in both directions. The first sequence (A) of GES → unconscious neural activity → behavior continues to be still present, so that the final model then becomes a completely causal one, consisting of (A) and (F) combined to give (G):

(G)                                GES
                                    ↓
conscious thoughts/will ↔ unconscious neural activity → action

According to model (G), our behavior and all our thoughts and consciousness and perceptions of free will are all caused by our unconscious neural networks that are the products of our genes, environment, and the random (stochastic) events that constitute our personal history.

In this model, there is no need to insert a mysterious non-material entity at any point. The Ghost in the Machine has been exorcised.

But while models are all well and good, we need experiments and evidence to distinguish between the ones that are close to reality and those that are wishful thinking.

Next: What does the evidence tell us?

On free will-4: The implications of modern physics for determinism

(For previous posts in this series, see here.)

The possibility of the existence of Lucretian random swerves that destroy determinism received a boost in the early twentieth century with the advent of quantum mechanics and its associated uncertainty principle that eliminated strict classical determinism.

Believers in free will seized on the inherent randomness built into these newly discovered laws of nature to argue that free will could exist and manifest itself at the quantum level. However, as our understanding of quantum mechanics has increased, few scientists seriously accept this possibility anymore because of the many problems such a model has. After all, random processes are, well, random, meaning that they are not subject to being controlled. If indeterminancy at the quantum level is what undermines determinism, what we would have is not free will but what we might call ‘random will’, in the sense that we would be acting according to the random outcomes of quantum level phenomena over which we have no control. Furthermore, while individual quantum events may be completely indeterminate, they do obey laws that enable us to accurately predict statistical outcomes, so these events cannot be truly free. Free will as popularly conceived does not consist of random or statistically predictable behavior but of the ability to deliberate and determine specific outcomes. No mechanism has been proposed to suggest how that might occur.

Another feature of modern physics that has been floated as an escape route for free will is chaos theory. But chaos theory is strictly deterministic. What it says is that certain systems are so sensitive to the specification of their initial state, that the initial state can never be specified with sufficient accuracy to enable the prediction of final outcomes. So chaotic system are deterministic (hence do not allow for free will) but unpredictable. Furthermore, not all complicated systems are chaotic. Systems that are chaotic have to obey certain types of laws and it is not clear that the brain is a chaotic system.

The idea that free will can manifest itself by taking advantage of quantum uncertainty or chaos theory is an argument phrased in a vague form that currently only theologians and religious apologists take seriously, a desperate clasping at straws.

Biologist Anthony Cashmore summarizes the modern view that each one of us is a product of our genes (G), our environment (E), and a stochastic (random) component (S). (The Lucretian swerve: The biological basis of human behavior and the criminal justice system, Proceedings of the National Academy of Sciences, March 9, 2010, vol. 107, no. 10, 4499-4504.) The stochastic part comes from the randomness of events at the quantum level as well as those involved in things like genetic recombination, synapse formation, and so on. In other words, even identical twins (i.e., having the same G) reared in identical environments (E) will still have neuronal networks that differ due to this unpredictable stochastic contribution (S).

Cashmore points to the failure of advocates of free will to provide a mechanism to substantiate their belief.

Whereas much is written claiming to provide an explanation for free will, such writings are invariably lacking any hint of molecular details concerning mechanisms. Also, it is often suggested that individuals are free to choose and modify their environment and that, in this respect, they control their destiny. This argument misses the simple but crucial point that any action, as “free” as it may appear, simply reflects the genetics of the organism and the environmental history, right up to some fraction of a microsecond before any action.

To understand more deeply the issues involved, we need to understand the structure of the brain and how it works. An infant begins life with about 100 billion neurons (these are specialized nerve cells) in the brain, each with about 1000 links (called synapses) connecting it to other neurons, creating a complex interlocking web of neurons. The initial state of the brain is largely the product of our genes, though the environment in the form of the conditions in the womb undoubtedly plays a formative role as well. As the child grows, its life experiences result in a pruning of the number of neurons, the elimination of some synapses and the creation of new ones, resulting in each one of us having brains that have a unique neural network, shaped by genes and environment and the inherent randomness of the laws of nature. Even allowing for some attrition on the way to adulthood, the number of neurons and synapses we are left with is enormous, resulting in brains of immense power and complexity that dwarf even the most sophisticated computers of our age. It should not be surprising that the workings of the brain can be so subtle that it can create the illusion that it has powers in the form of a ghostly mind.

Given our modern understanding of the brain as a purely material entity, the widespread persistence of the strong belief in the existence of free will demands an explanation. It cannot be due entirely to social reasons such as needing the concept in order to assign responsibility for people’s actions. Like almost any feature that is ubiquitous among diverse populations (like the desire to believe in god), it is likely that susceptibility to belief in free will originated early in our evolutionary history and is hardwired in our brains because it has considerable survival value that has resulted in it being strongly selected for by natural selection. People who believe they have free will (even if this belief is false) are more likely to feel a sense of responsibility for their own actions and therefore less likely to do foolish and dangerous things, and thus more likely to survive and reproduce.

But while our powerful brains may be pre-disposed to believe in free will, it is also powerful enough to turn its analytical capacity to study its own workings. And as it does so, and our awareness of the power and complexity of the brain rises, it has started to undermine the notion of free will. But there is strong resistance to this trend, even among non-religious scientists. Cashmore says that while scientists are fairly open about their disbelief in a god, they tend to hedge their bets concerning free will, or at least are less reluctant to speak openly about the growing evidence that it likely does not exist. It is thought to be too explosive a topic, one that the general public might not be able to handle. Darwin himself seemed to feel that while there was likely no such thing as free will, it was better to keep this knowledge within the province of highly educated people who could deal with all its implications and not panic and go berserk.

Resistance to the idea that we do not have free will is likely to be far greater than the resistance Darwin encountered to his idea that human beings are, like any other species, just one of the products of evolution, nothing more. The idea of free will is not going to be given up except in the face of overwhelming evidence.

Next: Models of free will and the brain.

On free will-3: Free will and determinism

(For previous posts in this series, see here.)

Defining what is meant by free will is not easy. In a loose sense it implies a denial of strict determinism, in which all our actions are completely determined by the past and the immediate environment we find ourselves in. The philosopher John Searle describes free will as the belief “that we could often have done otherwise than we in fact did.” In other words, although I am currently sitting at my desk typing, I think I could just as easily stand up and sing or hop around the room or do any other seemingly spontaneous act. My decision to not do so and continue typing seems like a conscious, freely chosen decision that is not entirely pre-ordained. The catch is that it is hard to reject the alternative hypothesis that all the options I considered were already determined by my history and the external stimuli of the moment, as was also my decision as to which option to choose.

Biologist Anthony Cashmore, in a recent paper (The Lucretian swerve: The biological basis of human behavior and the criminal justice system, Proceedings of the National Academy of Sciences, March 9, 2010, vol. 107, no. 10, 4499-4504) that provided the Searle definition above, suggests a better definition of free will and is what I will use. He says that, “free will is better defined as a belief that there is a component to biological behavior that is something more than the unavoidable consequences of the genetic and environmental history of the individual and the possible stochastic laws of nature.” The reason I like this better is because it focuses on the crucial question of the actual mechanism by which free will acts, rather than on our subjective perceptions about the inevitability or otherwise of our actions.

The idea that we may not have free will in the classical sense, of being able to make decisions that are not entirely determined by our personal history and external factors, is very difficult to accept. Even biologists, who would have little trouble agreeing with the statement that all biological systems are governed by the laws of physics and chemistry, and indeed depend upon that belief for their research, tend to resist the necessary conclusion that this likely eliminates the possibility of an autonomous, independent non-biological free will.

Rejecting free will is harder than rejecting the idea of god because the idea that we are free and autonomous agents is so deeply ingrained into our psyche. Apart from the emotional impact that abandoning this idea entails, there are those who worry about the consequences of that realization. Would the rejection of free will mean that we can have no morality? If our decisions are not freely made, then how can we speak of right and wrong decisions? How can we assign responsibility for people’s actions? How can society punish people for breaking laws if their actions are not freely chosen? Would it not be like punishing someone who was forced to do something because of a gun pointed at his head?

My position is that in the long run it is better to know the truth than believe in fictions. If there is no free will, it is better to face up to it and devise social and legal systems that deal with the consequences than pretend otherwise. It is the same reasoning that causes me to reject the arguments that even if there is no god, it is better to maintain the fiction of god in order to frighten people into behaving better.

We need to look at what evidence there is for the existence of free will and also at how to deal with the consequences if we can show conclusively that it is a fiction. This is not a hypothetical philosophical exercise. There is already considerable evidence that free will as we know it does not exist and I think it is only a matter of time before it is conclusively shown to be the case. This realization will first occur in the scientific community as they are the ones more familiar with the evidence, and it will take longer for the general public to come to terms with it.

The early Greek philosophers were troubled by the implications for free will of the atomistic and mechanistic ideas that were current at that time. If everything in nature consisted (as they believed) of atoms in motion obeying unchanging laws, then everything that happened was just the playing out of pre-determined events. Cashmore quotes philosopher Daniel Dennett on what troubled the Epicureans:

If all movement is always interconnected, the new arising from the old in a determinate order—if the atoms never swerve so as to originate some new movement that will snap the bonds of fate, the everlasting sequence of cause and effect—what is the source of the free will possessed by living things throughout the earth?

Note the assumption (by the Epicureans and perhaps also Dennett) that free will must obviously exist. The problem for the Greeks was how to reconcile this with their atomistic model of nature in which all actions were determined by past events. To overcome this strict determinism, Lucretius proposed that what might happen is that occasionally atoms might execute ‘random swerves’, caused by the gods, and that was what broke the deterministic pattern of events.

To this day, most people who have an understanding of the science of the brain and appreciate the strong evidence that everything has a materialist basis, look for the modern equivalent of the ‘Lucretian swerve’ in order to salvage the notion of free will. Some still assign the cause of the swerve to the gods, others (as I will discuss in the next post in the series) to causes that have some kind of scientific veneer. But both seek some way of holding on to the idea of an entity that controls my material body on the basis of decisions that are freely made, and that this entity is the real me.

Next: Can physics rescue free will?

On free will-2: The Ghost in the Machine

(For previous posts in this series, see here.)

The philosopher Gilbert Ryle (1900-1976) didn’t think much of Rene Descartes’ idea of a disembodied mind, using its free will, acting as some kind of captain of the body, and coined the derogatory term ‘the Ghost in the Machine’ for it.

There is a doctrine about the nature and place of minds which is so prevalent among theorists and even among laymen that it deserves to be described as the official theory… The official doctrine, which hails chiefly from Descartes, is something like this. With the doubtful exception of idiots and infants in arms every human being has both a body and a mind. Some would prefer to say that every human being is both a body and a mind. His body and his mind are ordinarily harnessed together, but after the death of the body his mind may continue to exist and function. Human bodies are in space and are subject to mechanical laws which govern all other bodies in space… But minds are not in space, nor are their operations subject to mechanical laws…

…Such in outline is the official theory. I shall often speak of it, with deliberate abusiveness, as “the dogma of the Ghost in the Machine.” (quoted by Stephen Pinker, The Blank Slate, p. 9)

Ryle’s skepticism about the Cartesian dualistic model is being validated as our understanding of the brain increases. The idea of some centralized information gathering and decision-making location in the brain has become increasingly untenable. We now know that the brain is a material entity, consisting of specialized cells (called neurons) connected to each other (by things called synapses), with the entire system serviced by blood vessels and glands that secrete chemicals. While there are some localized regions dealing with specific functions (such as sight and smell), the brain activity that we are aware of, such as thoughts and feelings, consist of patterns of complex neutral activity that are spread over large parts of the brain, with individual neurons firing in response to stimuli received from other neurons and the external environment. There does not seem to be any evidence of some kind of command center where decisions are made.

And yet we have a strong sense of the existence of things like the ‘will’ and ‘consciousness’, which are the attributes we assign to the Ghost in the Machine. The challenge is to understand them in terms of the workings of the material substrate of the brain.

In the first post in this series, I said that belief in an independently existing mind shares many similarities with the belief in god. One important difference between the mind and god is that each one of us is assumed to possess our own mind with its own identity, while god is a single universal entity. A second difference is that our personal entity can only interact with our own neurons and thus control our own thinking and actions and not those of others while god supposedly can control everyone’s thinking and even move physical objects like mountains if he wants to. (I am going to ignore the weak evidentiary claims for the existence of telekinesis, extra-sensory perception, mind control, and the like)

But those differences are not the ones that are most relevant to the problem of free will. The key problem is the implausibility of the idea that there exists a dualism, a mind-body distinction in which an independently existing non-material mind can influence the body.

I find the idea of a personal, independently existing, conscious, non-material, entity that I can call my ‘mind’ as hard to accept as the existence of god, and for all the same reasons. Assuming that such a thing exists causes far more problems than it solves. It seems so much more likely that what I call my mind or my will, rather than controlling my body, is actually the product of my body, caused by the firing of the neutrons in my brain, and the firing of any one of those neurons is in response to the stimuli it receives either from other neurons or, if it is a neuron that is directly connected to a sensory organ such as the eye or nose, the sensations my body receives from the external world.

All of my brain’s workings arise from my personal life history that has made my body what it is and created my brain and its neural networks. In other words, ‘I’ am a unitary system, not a dualistic one, made up of material objects obeying the laws of nature. This process, acting over the duration of my personal life history as well as the longer term evolutionary history, has created a brain that in turn makes decisions that influence my body, leading to new experiences that further shape my brain, and so on. It is a self-contained and closed system that does not require some external non-material entity that interacts mysteriously with it.

But the implication of this unitary view of the body is that any decision I make is completely determined by the facts of my personal history and the external stimuli that I experience at any moment. Although I may think that there is another ‘I’ within my body making decisions freely using my ‘free will’, this perception is an illusion and the reality is that the decisions are the consequences of the laws of nature simply working their way through my material body.

Next: To what extent are we strictly deterministic animals?

On free will-1: Cartesian dualism and the Cartesian Theater

It’s been awhile since I inflicted on this blog’s readers a long multi-part series of posts but I have decided to look at the question of free will, something that I have not addressed before, and this is such a weighty and controversial subject that it requires a somewhat lengthy discussion.

It used to be thought that what distinguished living things from inanimate matter was the presence of some mysterious life force, an élan vital. Modern biology has dispelled that myth of a vital essence, replacing it with the understanding that biological systems are nothing more than the working out of the laws of physics and chemistry on atoms and molecules. But there are some forms of vitalistic thinking that are still extant because people tend to want to cling on to the idea that there is something special about living things, especially human beings.

Religious people give the name the ‘soul’ to the feature that supposedly distinguishes them from non-human animals. They believe it enters and becomes part of the body at some point in its development from its start as a fertilized egg. The pope says that god inserts the soul at the moment of conception but others may allow for the soul to make its mysterious appearance later, as an emergent property that arises along with other things like consciousness once the neural system has reached a suitably advanced state. But, whenever and however it makes its appearance, people seem to believe that it has an autonomous existence, independent of the body. The soul supposedly lives on after the body has dispersed into its constituent atoms.

Belief in god and belief in free will have an obvious connection. Religious people need to believe in free will because there is no virtue in having faith in god if we have no control over our decisions. Of course, religious people seem to think that there is no problem with god coercing people to believe, on pain of eternal damnation and torments in hell. You would think that such belief would be just as tainted and useless as (say) confessions obtained under torture, which we rightly exclude from evidence in legal trials, or at least used to until this horror known as the ‘war on terror’ was unleashed. But religious people determinedly cling to the idea that their faith is freely given.

While the idea of a soul and its associated concept of god still continue to exert a powerful attraction for religious people, most scientists have given up on those ideas. But that does not mean that vitalistic thinking has entirely disappeared among the non-religious. Where it still lurks is in beliefs about the existence of free will.

There is an important similarity between god and free will. For most people, god is some independently existing conscious entity that is non-material and not subject to the laws of nature and yet is capable of interacting with the world, to the extent of knowing everything that happens everywhere at every moment and influencing events whenever he wants to. To achieve this requires this non-material entity to be able to interact with matter (in order to move objects) and to trigger the firing of neurons in brains (in order to influence people’s thinking). Religious people rarely probe their beliefs to this level of detail and if they do, the mechanisms by which those things occur is never elaborated on. It is all part of god’s mysterious ways.

The idea of free will is quite similar. To say I have free will implies that I can first freely make a decision (to pick up a pen, say) and then impose that decision on my body, forcing it to carry out that action. But who is this ‘I’ that is making this decision and how does it do it? At the most straightforward level, it implies a dualism that is similar in many respects to what we think of as god. It implies that there is an independently existing conscious entity that is non-material that we refer to as the ‘mind’ (or soul if you are religious) and which can make decisions and make my body comply with them.

The scientist, mathematician, and philosopher Rene Descartes (1596-1650) is credited with being the founder of such a dualistic model. He started out by questioning the nature of this thing we call ‘I’, arriving at the conclusion that there could not be any doubt about the existence of such an independent entity, writing, “But what, then am I? And what is that? A thing which doubts, understands, affirms, denies, wills, does not will, which also imagines and senses… For the fact that it is I who doubt, who understand, who will, is so obvious that there is nothing which could make it more evident.” (Rene Descartes, Second Meditation, Translated by Desmond M. Clarke, Penguin Classics 1998, p. 26)

He went on to say that it was fairly obvious to show that we must have such an independent mind:

I perceive that there is a big difference between the mind and the body insofar as the body, by its nature, is always divisible whereas the mind is evidently indivisible. When I reflect on the mind (or on myself insofar as I am simply a thinking thing), I certainly cannot distinguish any parts in myself; instead I understand myself to be a completely unified and integral thing. And even though the whole mind seems to be united with the whole body, if however a foot, an arm, or any other part of the body is cut off, I know that nothing is thereby taken away from the mind. Nor can the faculties of willing, sensing, understanding, etc., be said to be parts of the mind, because it is one and the same mind that wills, senses and understands. In contrast, I cannot think of any physical or extended body that I cannot divide easily in my thought; for that reason alone, I understand that it is divisible. That would be enough to teach me that the mind is completely different from the body if I did not already know it adequately from other considerations. (Sixth Meditation, p. 67)

We are all deeply influenced by this Cartesian dualistic model, tending to think that somewhere inside our brain is some kind of sophisticated control room where information streams in from everywhere via our senses and from the prior knowledge stored in our brains, and that in this room is some disembodied entity, the real ‘I’, who is a kind of commander-in-chief that views all this data, makes judgments, decides what to do, and then sends out commands that are executed by the body. This image of a command center, which Daniel Dennett calls the Cartesian Theater, is very powerful and hard to shake off.

As this series develops, we will see this traditional idea of dualism and its associated concept of free will take quite a beating. This does not mean that people will abandon the idea. We know in the case of religion that the desire to believe in god is so strong that people cling tenaciously to that idea, creating all manner of convoluted theories to explain the absence of any evidence in its favor. The idea of free will is even more deeply engrained in us and thus harder to let go, so one should expect similar attempts at countering any evidence that free will may well turn out to be an illusion.

Next: The problems with the Cartesian model.