Wednesday, January 06, 2021

Solving Molyneux's Problem: The Innate Faculty for Leaning to Perceive Forms by Touch and Sight

If someone blind from birth were suddenly able to see, would he be able to recognize by sight the shape of an object he previously knew only by touch?  For example, if he had previously known a cube and a globe by touch, would he now be able to tell which was which just by looking and not touching?

Ever since William Molyneux posed that question to John Locke in 1693, philosophers and scientists have argued about it.  Answers to the question have been offered by George Berkeley, Goffried Leibniz, Voltaire, Denis Diderot, La Mettrie, Adam Smith, Hermann von Helmholtz, and William James. 

There's a profound question here about human understanding of the world--about whether we derive our knowledge from innate ideas or from learned experience or from some mixture of both (Degenaar 1996; Degenaar and Lokhorst 2017).

It has been common for modern philosophers to treat a question like this as a thought-experiment that can be answered only by purely speculative reasoning.  But in principle this question is open to empirical study and testing, and indeed some recent research by neuroscientists studying sensory perception does suggest an answer.

Like Locke, Molyneux was a polymath.  A Protestant Irishman living in Dublin, he was a philosopher, a scientist, and a politician.  He founded the Dublin Philosophical Society in 1683.  He was elected a Fellow of the Royal Society of London in 1686.  He wrote a book on the science of optics--Dioptrica Nova (1692)--in which he praised Locke's Essay Concerning Human Understanding.  He sent a copy of this book to Locke in 1692, which began a philosophical friendship between them sustained through correspondence, although they never met.  Molyneux became a member of the Irish Parliament representing Dublin University in 1692.  In 1698, he published The Case of Ireland, in which he used Locke's arguments in the Second Treatise for government by the consent of the people to support the claim that the Irish people were not subject to the laws of the British Parliament.  This was the first time that Locke had been identified in print as the author of the Two Treatises of Government.  It was only after Locke's death that he acknowledged his authorship in his will.  The Case of Ireland became a foundational text for 18th century Irish nationalism.  In 1774, John Dickenson referred to this book as indicating how Locke's theory of consent supported not only Irish independence from Great Britain but also American independence.

In a letter of March 2, 1693, Molyneux posed his "Jocose Problem" to Locke as something for him to take up in the second edition of the Essay Concerning Human Understanding.

"I will conclude my tedious lines with a Jocose Problem, that, upon Discourse with several concerning your your Book and Notions, I have proposed to Diverse very Ingenious Men, and could hardly ever Meet with One that at first dash would give me the Answer to it, which I think true; till by hearing My Reasons they were Convinced, tis this.  Suppose a Man born blind, and now adult, and taught by his Touch to Distinguish between a Cube and a Sphere (Suppose) of Ivory, nighly of the same Bigness, so as to tell, when he felt One and tother, Which is the Cube which the Sphere.  Suppose then, the Cube and Sphere placed on a Table, and the Blind man to be made to see.  Ouaere [the question is] whether by his sight, before he touched them, he could now Distinguish and tell which is the Globe which the Cube.  I answer, Not; for tho he has obtained the Experience of How a Globe, how a Cube affects his Touch.  Yet he has not yet attained the Experience, that what affects my Touch so or so, must affect my Sight so or so; or that a Protuberant Angle in the Cube that pressed his hand unequally, shall appear to his Eye as it does in the Cube.  But of this enough; perhaps you may find some place in your Essay, wherein you may not think it amis, to say something of this Problem."

Locke responded to this by adding a new passage to the second edition of the Essay in his chapter on the "Faculty of Perception" (II.9.8).  It's in a section where he explains how our sensations are changed by our judgment.  So, for example, when we look at a round globe of some uniform color such as gold, the visual sensation coming from our eyes and imprinted on our mind is of a flat circle shadowed with several degrees of light and brightness.  But we are accustomed to perceive the appearances of convex surfaces with reflections of light and shadows of color and to construct in our minds a globe with a uniform golden color; and once this becomes habitual for us, we do it automatically, and we forget how we had to learn by experience to do this.  We do this just as we look at the flat plane of a painting with perspective, shading, giving relief, and coloring, and we learn how to reconstruct a three-dimensional scene in our mind's eye.

Locke then adds:

"To which purpose I shall here insert a problem of that very ingenious and studious promoter of real knowledge, the learned and worthy Mr. Molineux, which he was pleased to send me in a letter some months since; and it is this:--'Suppose a man born blind, and now adult, and taught by his touch to distinguish between a cube and a sphere of the same metal, and nighly of the same bigness, so as to tell, when he felt one and the other, which is the cube, which the sphere.  Suppose then the same cube and sphere placed on a table, and the blind man be made to see: quere, whether by his sight, before he touched them, he could now distinguish and tell which is the globe, which the cube?'  To which the acute and judicious proposer answers, 'Not.'  For, 'though he has obtained the experience of how a globe, how a cube affects his touch, yet he has not yet obtained the experience, that what affects his touch so or so, must affect his sight so or so; or that a protuberant angle in the cube, that pressed his hand unequally, shall appear to his eye as it does in the cube.'--I agree with this thinking gentleman, whom I am proud to call my friend, in his answer to this problem; and am of opinion that the blind man, at first sight, would not be able with certainty to say which was the globe, which the cube, whilst he only saw them; though he could unerringly name them by his touch, and certainly distinguish them by the difference of their figures felt.  This I have set down, and leave with my reader, as an occasion for him to consider how much he may be beholden to experience, improvement, and acquired notions, where he thinks he had not the least use of, of help from them."

Here we see Locke's famous teaching that the mind is a tabula rasa, or "white paper," or "empty cabinet," with no innate principles, so that all of the content of the mind must be acquired through experience or learning.(I.1.15, I.2.22, II.1.2).  Those who believe this will give a negative answer to Molyneux's question, because they are inclined to believe that the connection between touch and sight in the perception of form--distinguishing a globe from a cube--must be learned by experience.  By contrast, a positive answer to the question suggests that there is some innate concept of space and figure common to both touch and sight that allows for the perception of forms such as globes and cubes.

To my mind, Locke and Molynieux are mistaken insofar as their rejection of innate ideas assumes a false dichotomy of instinct versus learning, which ignores the possibility of instincts for learning.  After all, the mind's learning by experience would be impossible if the mind did not have any inborn capacities for learning, or latent propensities that are elicited or evoked by experience.  Locke implicitly recognizes this when he speaks of the mind as having faculties for perception, retention, and discerning (II.9-11).  He also speaks of the mind as having "powers" or "capacities."  If the mind did not have these natural, inborn powers for learning by experience, it could not learn anything.  Moreover, Locke sometimes explicitly recognizes the mind's innate principles: "Nature, I confess, has put into man a desire of happiness and an aversion to misery: these indeed are innate practical principles" (I.2.3).

Recently, some neuroscientists studying how the brain learns to see have performed some experiments designed to answer Molyneux's question; and while their results seem to confirm Molyneux and Locke's answer, their experiments also seem to show that learning the perception of forms by touch and sight does require some innate faculties for such learning.

In India, children in poor families born blind because of cataracts don't have access to the surgery to correct their condition, and consequently many of them will grow to adulthood without ever having vision.  A few years ago, Pawan Sinha, a neuroscientist at MIT, led a team of doctors and scientists who set up clinics in India for studying and treating these congenitally blind individuals.  This created an opportunity to experimentally answer Molyneux's question: Would a blind person, on regaining sight, be able to immediately visually recognize an object previously known only by touch (Held et al. 2011; Sinha 2013; Sinha et al. 2014)?

For one study, five young people were recruited, aged from 8 to 17 years.  Their congenital blindness was so severe that they could only detect differences of light and dark.  They could not see objects.  Four of the individuals had cataract removal surgery and an intraocular lens implant.  One individual was provided with a corneal transplant.  After this, all five could see for the first time in their lives.

Within 48 hours of their sight-restoring surgery, all five were tested for their ability to distinguish shapes by touch and by sight.  They were presented with 20 pairs of Lego-like three-dimensional forms that had different shapes.  They would feel one object in the pair, without looking at it, because it would be hidden under a bed sheet.  They were then allowed to feel both objects in the pair, and they could correctly pick out the first object they had touched.  Then they were allowed to look at both objects and asked: Which one looks like the first object you touched?  They failed to answer this correctly.  They could not transfer their knowledge from touch to vision.  They could not integrate their knowledge from these two different sensory modalities.

After a few weeks, they were tested again, and most of these individuals could transfer their knowledge of shape by touch to their knowledge by vision.  A few weeks of experience in the world combining touch and vision in the perception of forms was enough to integrate the two sensory modalities.

So in these experiments, the answer to Molyneux's question is no, but it's a qualified no.  Someone blind from birth who suddenly is able to see will not immediately recognize an object by sight alone that he has previously known by touch.  Molyneux and Locke were right in believing that understanding the connection between touch and vision in the perception of forms must be learned by experience.  But they were wrong in suggesting that there was nothing innate in this process of learning.  The fact that these congenitally blind individuals learned this so quickly--within a few weeks of gaining vision--suggests "a latent ability for rapid learning" (Sinha 2013, 54).

Moreover, Sinha and his colleagues have seen evidence that the brain uses some instinctive rules for learning how to see the intelligible order of the world by organizing fragmented visual inputs into whole objects.  Initially, newly sighted individuals see only a chaotic flux of visual data of confusing shadows and shading and overlapping patterns with no meaningful structure.  In this, they are like newborn infants who see a "blooming buzzing confusion" around them.  Within a few weeks of birth, a baby begins to discern patterns of whole objects--such as mother's face.  The newly sighted Indians who had grown up in blindness took longer to do this.  Only after a year or more could they see distinct patterns of whole objects.  For example, if they saw a picture of a square overlapping a circle, they saw many interlocking pieces, and they could not identify the square as different from the circle.  But then if the square was moving around while the circle was motionless, they could identify the square as a separate figure.  

Thus, the brain uses motion as a visual cue for learning to see whole objects.  The brain seems to have the rule "things that move together belong together."  Infants employ this same rule in organizing their sensory experience.  Evolution by natural selection has favored brains that employ this rule for the perception of sensible forms.

Human beings are born with an innate capacity in the brain and nervous system for learning how to see.  The first few weeks and months of life are a critical learning period in which infants with normal functioning visual systems learn quickly how to perceive visible forms.  People who have been congenitally blind through most of their childhood are beyond this critical learning period.  But Sinha and his colleagues have shown that the plasticity of the brain is sufficient to allow people gaining sight later in life to slowly evoke the latent instinctive capacity for learning how to see.

The evolved nature of the human brain for understanding ourselves and our world gives us knowledge not from instinct alone nor from learning alone but from our evolved instincts for learning.

                                                   Pawa Sinha on How Brains Learn to See


Degenaar, Marjolein. 1996. Molyneux's Problem: Three Centuries of Discussion on the Perception of Forms. Dordrecht: Kluwer Academic.

Degenaar, Marjolein, and Gert-Jan Lokhorst. 2017. "Molyneux's Problem." Stanford Encyclopedia of Philosophy.

Held, Richard, et al. 2011. "The Newly Sighted Fail to Match Seen with Felt." Nature Neuroscience 14: 551-553.

Sinha, Pawan. 2013. "Once Blind and Now They See: Surgery in Blind Children from India Allows Them to See For the First Time and Reveals How Vision Works in the Brain." Scientific American, July.

Sinha, Pawan, Jonas Wolff, and Richard Held. 2014. "Establishing Cross-Modal Mappings: Empirical and Computational Investigation." In Senssory Integration and the Unity of Consciousness, ed. David Bennett and Christopher Hill, pp. 171-191. Cambridge: MIT Press.


Roger Sweeny said...

In 1774, John Dickenson referred to this book as indicating how Locke's theory of consent supported not only American independence from Great Britain but also American independence.

Do you mean "... but also Irish independence"?

Larry Arnhart said...

I've corrected it. Thanks.