Attack of the Brain Atoms!

Once more a public thinker has set forth to prove in public that he has no mind.  The brain atoms inhabiting the vehicle called "Sam Harris" have moved Mr. Harris's fingers to type a blog post yclept: Free Will (And Why You Still Don't Have It) contending that he did not intend to type a blog post, but was caused to do so by a congeries of external and internal stimuli. 

The post is an excerpt and/or digest of a section of the book which the brain atoms, responding to information from the external world, internal states of the body, and the ennoiasphere, entitled The Moral Landscape.  Like Condorcet before him, Sam Harris expects that he can find morality inductively by means of physical science, as if by knowing the mass of an electron he would tell us whether we should off granny for the inheritance.  In the course of this, his brain output replicated the following external stimuli:

Maundering Becomes Elektra

"The physiologist Benjamin Libet famously demonstrated that activity in the brain's motor regions can be detected some 300 milliseconds before a person feels that he has decided to move.  Another lab recently used fMRI data to show that some "conscious" decisions can be predicted up to 10 seconds before they enter awareness (long before the preparatory motor activity detected by Libet). Clearly, findings of this kind are difficult to reconcile with the sense that one is the conscious source of one's actions." 

Actually, findings of this kind are difficult to reconcile with the idea that they are scientific experiments, since they invariably over-interpret the actual empirical observations.  Identifying the "response potential" with the "moment of decision" is to beg the question. 

In 1983, neuroscientist Benjamin Libet  asked volunteers wearing scalp electrodes to flex a finger or wrist. When they did, the movements were preceded by a dip in the signals being recorded, called the "readiness potential". Libet interpreted this RP as the brain preparing for movement.  Crucially, the RP came a few tenths of a second before the volunteers said they had decided to move.  Harris cites this as evidence that free will is an illusion, even though there is no proof the RP represents a decision to move. 

Now, how did Lisbet know when his subjects made a decision?  The subjects self reported where a dot was on a clockface when they "felt" they had decided.  But visual processing is sluggish.  At the moment of decision, the subject would be "seeing" an earlier time than the present but which had only then been processed by the visual system.  Curiously, no one contends that this proves that vision is an illusion.  Besides, if I were to insert Elektra into your brain (see picture to right) you would be unlikely to think straight either. 

Lisbet was actually aware of the problem of viscosity in response and tried to correct for it by a separate experiment estimating the time lag.  He did this by applying a mild electrical shock to the back of the hand and noting when the subject reported feeling the shock.  However, the time lag for tactile sensations is shorter than for visual (or auditory) sensations, thus Lisbet's correction was not enough.  The 300 milliseconds appears to be an artifact of this effect. 

Two other researchers, Miller and Trevena, also used scalp electrodes, but asked the subjects to wait for an audio tone before deciding whether or not to tap a key. An RP dutifully appeared but the signal was the same whether or not they elected to tap the key. Clearly, the RP is not an actual decision to move.  Miller concluded that it might merely signal that the brain is paying attention. 

Harris's brain atoms did not output this particular external stimulus into his essay. 

The second experiment mentioned by Harris' brain atoms is that of Chun Siong Soon (et al.), described here.  Subjects' brains were scanned by fMRI as they decided to press a button with their right or left index fingers, thus making it really-truly scientificalistic. To peg the moment-of-decision, subjects referred to a stream of letters on a screen. Patterns of brain activity in two areas "correlated" with the left/right decision appeared "up to" ten seconds before participants reported making their conscious decision. 

Soon's study also predicted whether the subject would use his left or right hand well before they supposedly knew which choice they'd made.  None of the accounts I've seen mentioned whether Soon had taken handedness into account.  (I can predict pretty near 100% which hand my wife will use to write a note.)  Also, in one account, I saw that this prediction was made correctly only about 65% of the time.  This is significant only if Ho = 50% and people choose their hands at random.  But it is well known that people cannot choose at random, and Soon's study seems to be another, more expensive confirmation of this.  In my stat classes I have asked people to randomly choose one of the numbers 1 2 3 4, and about 50% will choose 3 when random chance would predict 25%.  Almost no one chooses 1 or 4. 

That someone might rev up his motor response region in anticipation of an immanent decision seems not to have crossed anyone's mind.  Perhaps it was not one of the external stimuli in the "sphere of meaning" in which the Harris brain atoms float.   

Mind That Brain!

It is ludicrous to suppose that such tests are measuring "free will" rather than some congeries of unspoken assumptions.  In addition to the belief that the null hypothesis is 50%, we have the assumption that the subconscious does not participate in acts of the will.  Furthermore, when participants are asked to "choose one at random," they try to put themselves in a state where they let go of their will and "go with the flow."  Is there any circumstance better designed to rermove the will from the process and capture a simple flow-through of stimulus to response?  Soon's "up to" 10-second pre-decision is bizarre inasmuch as that is more than enough time for the subject to pull out the electrodes and stalk indignately from the room!  And, yes, I did notice that all the accounts I read reported only an "up to" number, a max, with no average or standard deviation.  That sounds suspiciously like reporting an outlier as if it were the normal response. 

All this is due to a profound misunderstanding of free will and a confusion of free will with free choice.  (Choice is an act of the will; it is not the will itself.) 

The brain atoms of Mr. Harris output the following response to various external stimuli: 

"The human brain must respond to information coming from several domains: from the external world, from internal states of the body, and, increasingly, from a sphere of meaning--which includes spoken and written language, social cues, cultural norms, rituals of interaction, assumptions about the rationality of others, judgments of taste and style, etc. Generally, these streams of information seem unified in our experience." 

Mr. Harris' brain atoms are to be congratulated for coming up with a nugget of Aristotelian-Thomistic psychology.  (Although they seem to refer to "brain" and "mind" as if they were the same thing.  This is circular, assuming that which ought to be proved.)    I have given the name ennoiasphere to the "sphere of meaning."  This provides a suitable pseudo-scientific patina to it.  Harris' brain atoms seem to thing this ennoiasphere is something new that "increasingly" provides something called "information."  But that "spoken and written language, social cues, cultural norms, rituals of interaction, assumptions about the rationality of others, judgments of taste and style, etc." was well known to Thomas Aquinas, who included it under the term "habits."  Habits could be cultural, personal, or (as we now know) genetic.  But habit is not a defeater for free will.  It was known to Aquinas, and to Aristotle.  One suspects even Plato knew of it. 

Most of life must be lived on automatic pilot.  That's why in addition to genetically-induced habits, we deliberately memorize alphabets and multiplication tables (or used to) as well as virtuous habits.  Humans are rational animals and a rational animal is of course an animal.  I once walked home from the dry cleaners down the block on autopilot while I mulled over some statistical issue, and did not come to until, inserting the key, I missed the lock and had to call on my conscious mind to finish the task.  During that brief walk, I experienced something of what it was like to be a non-rational animal. 

The Aristo-Thomist model of the mind (or "soul") is shown in the figure below.  I posted it once before, but it is worth reposting, since Harris and his disciples appear never to have heard of it. 

Let us first take the animal channel.

1. Sensation.  A flood of meaningless particles cascade against the senses: photons onto the retina, molecules up the nose, sound waves hammer the ear, and so on.  Harris's brain atoms think these are "information," but they are not.  Such particles do not become information until the mind gives them meaning.  (This is so routine in the human mind that it is actually very difficult to convince a materialist to act like a materialist.)  The Harris brain atoms are close, but as yet win no cigar. 

2. Perception.  The inner sense called "common sense" unifies all the sensory inputs into a single phantasm or ymago.  As Harris's brain atoms put it: "Generally, these streams of information seem unified in our experience."  In addition, the inner senses include memory (which stores the ymago) and the imagination (which manipulates the ymago).  This enables us to "see" an elephant even when it is not physically present (so we are already outside the realm of the physical), and to consider pink elephants and polka-dotted zebras.  Sometimes this whole process is called the imagination.  Imagination can accomplish wondrous things, like finding the cheese in a maze, or a bear riding a unicycle or a dog learning to fetch.  One of the most frequent sources of confusion in the Late Modern world is the confusion of imagination with intellect.  (This is confounded with such vague terms "intelligence" or "sentience.") 

So much for the stimulus side.  Now for the response side. 
3. Emotion.  These are the sensitive appetites for products of perception.  They prepare the animal for motion; hence "e-motion" (move out, agitate).  Keep in mind that it may be a disposition toward or away from the percept or ymago. 

4. Motion.  This is any change of form.  Motion of location is only one sort of motion. 

Note that the traditional model has a "short circuit."  The autonomous nervous system generates motion directly from perception, as for example when we touch a hot stove or are struck under the kneecap with a mallet. 

None of this need involve free will.  In fact, there is no will at all involved.  An animal becomes hungry, and it will seek out food, sometimes calling on memory to locate it.  No animal goes on a diet or fasts for world peace.  The mistake the Harris atoms make is to suppose that because some human acts are of this type, all human acts are of this type. 

As to the rational part of the mind/soul:
5. Intellection.  The intellect reflects upon the percepts and abstracts concepts.  From the perception of Fido and Lassie and Rover, it abstracts the concept of "dog."  It strips "dog" of all particulars of this dog or that dog, and presents to the will an indeterminate and immaterial abstracted object. 

6. Volition.  This is the intellective appetite  for products of intellection, a disposition toward or away from the concept. 
Because that which is higher governs that which is lower, the will modulates the emotions.  Thus, rational animals are capable of overruling the sensitive appetites and exercise courage, temperance, prudence, etc.  We can diet or fast even when we are hungry. 

Immediately, we see how the Lispet and Soon experiments missed the point.  They are testing mere physical motion, which is not especially the object of the will.  It confuses free will with free choice. 

No Matter What You Think

The Harris brain atoms write on and claim:

"The truth seems inescapable: I, as the subject of my experience, cannot know what I will next think or do until a thought or intention arises; and thoughts and intentions are caused by physical events and mental stirrings of which I am not aware." 

Now the first part is uncontroversial.  It is like saying I cannot take a step without making a footprint.  That I do not know what I will think until I think it is a tautology.  I cannot taste food until I eat it.  But the second part is simply to assert what the brain atoms ought to be proving.  The Harris vehicle's admission that he has written an entire essay without being aware of the "mental stirrings" is indeed tragic and we ought to entertain the possibility that this is actually what Harris' mind is like. 

However, it is as logically impossible for brain states to be thoughts as it is for printed text to be thoughts.  The letters or patterns are themselves devoid of meaning.  The lines that combine to form the shape H do not necessarily give rise to the knowledge of the sound "en."  It does so only when a mind uses the Cyrillic alphabet to represent sounds.  If erosion scratched into a rock parallel lines and cross-lines that took on the shape НЕТ, it would not be Nature telling us "NO."  If we observe a stone in free fall, it does not necessarily give rise to the idea of gravity.  It might give rise to the idea of "Run! Avalanche!"  Materialistically, we can only ever observe a stone in free fall.  Whatever meaning it has depends on the POV from which we view it. 

Consider the following, due to Deely and explained by Chastek:

M. No observable thing is such that it necessarily gives rise to the knowledge of something other than itself
m. Ideas or concepts are such that they necessarily give rise to the knowledge of something other than themselves.
C: Therefore, no  idea or concept is an observable thing.

The major premise (M) states that if there is some sign, there is no necessity compelling the observer it to understand it as a sign, still less as a sign for this or that particular meaning. 
The minor premise (m) is just the definition of a formal sign, or an idea. Ideas make things known other than themselves by necessity. 
In particular, since brain states and neural patterns are observables, an idea cannot be "nothing but" a brain state or a neural pattern.  This does not mean there is no association between ideas and brain states; but the relationship is not one of identity.  Which is the foot and which the footprint?  After all, I cannot know if you have an idea unless you tell me ("I have to stop and pick up some milk!") but that does not mean that the idea just is the words of that sentence even though your idea does not exist for me unless it is instantiated in those words.  Similarly, the brain states are simply another kind of instantiation.  That an idea might be induced by forcing the brain state to occur is not different in kind from inducing that idea by speaking it into your ear.  Or writing it on this screen. 

To confuse mind with brain states is like confusing a drive in the countryside with the workings of the pistons in the automobile engine.  The drive is not possible without the piston strokes, but there is more to it than that.  Consider the case of the woman in Oregon who after dental surgery acquired a British accent.  The fact that a British accent was somehow induced by the anesthesia/surgery does not mean that people in Britain have been uniformly subjected to that surgery in order to give them funny accents.  

Where There's a Way, There's a Will

We have already seen that one illusion of the Harris vehicle is that a freely willed choice is random.  It is not.  Any experiment that compares actual choice frequencies to 50% null hypothesis is testing the wrong thing.  There is also the false assumption that a freely-willed choice is unpredictable.  Why should it be unpredictable?  My wife can nearly always predict what I will order at the restaurant, and do so at better than chance.  This does not mean my wife causes my choices.

Recall that the proper object of the will is a concept (not switch flipping or finger-twitching) stripped by the intellect of all the particulars of a percept.  This means that the intellect presents to the will a concept that is undetermined to this or that particular "dog."  Since the object presented by the intellect is not known perfectly, the will is free to accept or reject it, or to choose this or that means to attain/avoid it.  The Harris atoms come close to this by saying that they "cannot know what I [sic] will next think or do until a thought or intention arises," but he seems to think the indeterminacy of the intellect is an argument against the freedom of the will rather than the basis for it! 

When the intellect presents to the will a concept such as 2+2=4, the will is perfectly determined toward it and cannot withhold consent.  (We assume that the signs have been learned and understood in the normal way.)  But when the intellect presents, say, the idea of helping the poor, the concept is not fully grasped.  Who are the "poor"?  What does it mean to "help" them?  A particular means -- say, this program -- does not command assent since it may be ineffective, counterproductive, or less effective than that program.  The will is therefore not determined to this or that and may give or withhold assent freely. 

Think of "free" as being like "play" in engineering.  The will is determined always to the good, but the intellect does not always know perfectly what is good, and a particular object may not be good from every perspective.  If it were good from every perspective, the will could not freely withhold consent. But then it would be free in the sense that nothing now holds it back from attaining its natural end.  Think of a stone in free fall, moving always toward the minimal gravitational potential.  In the same manner, the will when unencumbered by ignorance would move toward the perfect good.  Makes you wish we had a name for the Perfect Good, hey? 

"Free" does not mean that the will assents randomly or unpredictably; nor does it mean that it assents without reason or motive or in the absence of external stimuli from the ennoiasphere.  Nor even that it might not be hobbled by ignorance, habit, brain injury, and what have you.  We do not freely will in a vacuum.

But then the Harris brain atoms go on to spoil the perfect reasoning:

"None of this, however, renders the choices we make in life any less important. ... the fact that our choices depend on prior causes does not mean that they do not matter. If I had not decided to write my last book, it wouldn't have written itself. My choice to write it was unquestionably the primary cause of its coming into being." 

Choices?  What choices?  There is only the wind of causation blowing through the neural trees of the Harris brain. Decided?  But that he decided to write the book implies that he could have decided not to write it.  (Or that he could have written it differently.)  But by his prior account, the book did indeed write itself, because it was only a response to a set of causes passing through the forest of neurons.  On what grounds do we privilege those neurons as the cause of the book and not the sundry stimuli beyond them?  How do we justify any particular cut-off point in what must be a chain of causation stretching back to the Big Bang and say here is where the choice begins? 

"Decisions, intentions, efforts, goals, willpower, etc., are causal states of the brain, leading to specific behaviors, and behaviors lead to outcomes in the world. Human choice, therefore, is as important as fanciers of free will believe." 

The Harris vehicle cannot resist flinching at the last and surrendering its materialism.  How can there be willpower without a will?  How can choice be important when the will is not free?  (For that matter, if there is no volition, how can intellect act?)  This is utterly incoherent.  No one doubts that decisions, intentions, et al. are associated with brain states, but just as footprints don't cause walking, the brain states might not "cause" thinking. 

Heck, there are interesting cases of people leading normal lives with virtually no brain at all.  A young boy lacking a cerebellum should not be able to walk; but he does.  A student lacking nearly all his cortex should not get A's in math, but he does. What happens is that in some cases the mind recruits other regions of the brain to carry out functions usually performed by the missing or damaged regions.  How this could happen would be an intriguing research topic.

The Harris atoms conclude that "the illusion of free will is itself an illusion" but they do not inform us as to what it is that is suffering the illusion. 

Raymond Tallis takes a different view in a dinner talk at the Manifesto Club in London (13 September 2007)

"Take my coming here tonight to talk.  This action required, among other things, my getting to this restaurant, which in turn depends upon the making and unmaking of cross-bridges in my leg muscles as I walk. Those unconscious physiological events are not things that I do: they are mere mechanisms.  As we ascend from the material details of an action to its larger purpose, several things happen. Firstly, intention and consciousness becomes more important. Secondly, the action draws on more and more of what I would regard as distinctly myself." 

As usual when we get to the mountain top we find Thomas Aquinas there with a lemonade stand.  Thomas distinguished between "human acts" and "acts of a man."  The acts of a man were precisely those mechanical acts that Tallis speaks of and which the Harris brain-states output as being the whole ball of wax.  The human acts are precisely the rational acts, the ones that call upon intellect and will. 

In computer architecture, this is called subsumption architecture.  and is used in such things as mechanical cockroaches that simulate actual cockroach behavior.  Reflex behaviors go on the lowest layers, like 'run along edges’ or ‘avooid obstacles.’  The more abstract behaviors (such as they might be for cockroaches) are layered incrementally atop the simpler ones, and they control the direction to be taken to achieve an overall task.  Thomas' model of the human mind is thus more supple and actually more in accord with modern thought than the Harris atoms' 19th century mechanistic model. 


______________________________________
The scope of Thomas' discussion of the Will can be found in the titles of the Questions regarding the will:

  6. Of the Voluntary and the Involuntary
  7. Of the Circumstances of Human Acts
  8. Of the Will, in Regard to What It Wills
  9. Of That Which Moves the Will
  10. Of the Manner in Which the Will Is Moved
  11. Of Enjoyment, Which Is an Act of the Will
  12. Of Intention
  13. Of Choice, Which Is an Act of the Will with Regard to the Means
  14. Of Counsel, Which Precedes Choice
  15. Of Consent, Which Is an Act of the Will in Regard to the Means
  16. Of Use, Which Is an Act of the Will in Regard to the Means
  17. Of the Acts Commanded by the Will

Each Question is considered in a series of Articles, for example Question 6 is considered in eight articles.
    (1) Whether there is anything voluntary in human acts?
    (2) Whether in irrational animals?
    (3) Whether there can be voluntariness without any action?
    (4) Whether violence can be done to the will?
    (5) Whether violence causes involuntariness?
    (6) Whether fear causes involuntariness?
    (7) Whether concupiscence causes involuntariness?
    (8) Whether ignorance causes involuntariness?

References

Free Will (And Why You Still Don't Have It) 
Free will and free choice
Free will and electrodes 
Ramble on free will 
Free will as free choice. 
Karen Butler, Oregon Woman, Develops Foreign Accent After Surgery 
Deely’s objection to physicalism/ naturalism
Free will is not an illusion
Nonlinear Brain Dynamics and Intention According to Aquinas