Jun 23 2009
When first introduced in 1796, phrenology was the latest advancement in the field of neurology. It was widely accepted, even welcomed, by many practicing neurologists as a powerful diagnostic tool. Phrenologists were even on the winning side of an important scientific debate concerning a central concept of brain anatomy and function. As more scientific methods began to take hold within medicine, however, and the secrets of the brain began to yield to more careful investigation, phrenology became increasingly marginalized. By the early 20th century the last vestiges of phrenology were gone from scientific medicine and mainstream neurology, but not gone completely. Phrenology survives to this day as a classic pseudoscience, with dedicated adherents convinced of its efficacy.
The history of phrenology, and the story of its modern believers, is a classic one in the history of pseudoscience. To contemporary skeptics, the claims of phrenology sound no different than any wacky belief system. Believers claim to be able to read an individual’s personality, their strengths and weaknesses, hopes and desires, by examining the pattern of bumps on their skull. At first the idea sounds no different, and no less ridiculous, than treating liver disease by rubbing the foot, or diagnosing heart disease by the pattern of colors in the iris. It isn’t, but phrenology has a very different origin than reflexology or iridology.
To understand phrenology, we must begin with the central debate of neurological scientists of the 18th and 19th centuries. The question concerned the organization of the brain. One school believed that the brain was relatively homogenous, the entire brain worked together as a whole to produce all mental and motor functions. One particular function, therefore, such as humor, aggression, the ability to control the right hand, or recognize the scent of a rose, could not be localized to any piece of the brain. Trying to identify what any one part of the brain did, therefore, was useless.
The other school believed the exact opposite, that the brain was exquisitely compartmentalized. Every function that can be attributed to the brain could, they argued, be localized to a particular part of the brain, which was dedicated to that single function alone. To this latter school, understanding the brain would come through identifying which pieces were responsible for which functions. This hypothesis was initially proposed and championed in 1796 by Austrian physician Franz Joseph Gall (1758-1828). (Gall, 1796)
Gall’s notions were first explored further by neurophysiologist Pierre Flourens, who in the 1800s followed up on Gall’s work with a series of experiments on rabbits and pigeons. He found that coordination localized to the cerebellum, while basic functions like respirations localized to the brainstem. However, he could not localize higher cortical functions, like memory, and so concluded that while primitive neurological functions did localize to one part of the brain, the higher cognitive functions were diffuse in the cerebral hemispheres.
Flourens was wrong, and this probably has something to do with the fact that he was studying rabbits and pigeons, making it difficult to assess cognitive function.
Pierre Paul Broca then took this research further, using clinical evaluations of humans to see if specific mental functions could be localized to specific brain tissue, and he was very successful in his research. One part of the cortex in the dominant frontal lobe is now called Broca’s area in his honor – this brain structure serves the specific function of translating language into the motor function of speech. Damage to this area results in what is called a Broca’s aphasia.
After Broca’s contributions to neuroscience the scientific community had largely settled on the compartmentalist view of the brain. The phrenologists belonged to this school from the beginning – they were compartmentalists. But what does this have to do with bumps on the skull? To get there we have to add another principle.
Phrenologists argued that the parts of the brain which corresponded to functions that an individual used a great deal would hypertrophy (grow larger), while those functions which were neglected would atrophy (shrink). The brain, they argued, was like muscle – those muscles which are exercised frequently bulked up, while those that are not used remain small and even shrink. Their vision of the brain, therefore, was that it had a lumpy and bulbous surface, with a landscape unique to each individual based upon their particular set of intellectual and neurological strengths and weaknesses. They further argued that the skull overlying the lumpy parts of the brain would bulge out to accommodate the hypertrophied brain tissue underneath. Therefore, by measuring those bumps, we can infer which parts of the brain are enlarged and therefore which characteristics are dominant.
By the middle of the 19th century, phrenology machines were in widespread use. The automated machines were composed of numerous spring-loaded probes. The device was placed over the head while the probes would extend to gently touch the scalp, thereby providing a measurement of the topography of the skull. The machine would then calculate the characteristics of the subject based upon this topography and produce an automated reading.
As it turns out, Gall and the phrenologists were correct when it came to the central debate of neurology of the time. The brain is compartmentalized, with each piece serving a specific function. The modern map of the brain, however, does not correlate to the classic map used by phrenologists. Theirs was more personality based, while the modern map is based on fundamental functions, such as the ability to perform mathematical calculations or interpret language.
At least this simple model of a compartmentalized brain is true when it comes to basic cognitive functions – such as language, motor control, vision, etc. However, for the higher cognitive functions, such as reality testing, decision making, inhibition of impulses, and similar functions – there is still a significant debate about the big picture of brain organization.
Today the debate is between modules and networks – do specific anatomical brain modules produce specific cognitive functions, or are these function dependent upon a network of connections among various brain regions with any individual region participating in multiple networks simultaneously. I go into more detail about this debate here, but the bottom line is that the answer is likely to be a combination of the two models – brain modules that are wired for specific types of processing, but participating in different networks for specific functions.
So the phrenologists got the compartmentalism thing mostly right, but all of the other assumptions of phrenology are mostly false. The brain is not, after all, a muscle. It does not hypertrophy or atrophy depending on use. The brain does change with use (a property called plasticity), but the changes occur on a microscopic level, and have to do with the strength and density of neural pathways.
It is true that the amount of gray matter that is dedicated to a task will increase with use of that task and is larger, as a proportion of brain tissue, in people with talent and skill in a specific area. Musicians have more gray matter dedicated to music, mathematical relationships, and motor control of the hands, than do non-musicians. This effect is only seen when we look at the internal structure of the brain with functional imaging – how much of the brain lights up when we perform a task. It does not affect the overall size and shape of the brain, and bits of the brain do not bulge out with use.
Also, the brain is very jelly-like in consistency. The soft brain conforms to the shape of the skull, even in severe cases of cranial deformity due to disease or the bindings which are practiced in some cultures. The skull does not conform itself to the brain. The notion of bone changing shape in reaction to underlying pressure is actually reasonable – bone is plastic and will reform to accommodate pressure. Brain tissue, however, is simply not capable of producing such pressure. Some types of brain tumor, on the other hand, are and often changes in the skull seen on X-ray can be a clue to an underlying tumor.
The scientific debate underlying phrenology was addressed over 100 years ago, and answered definitively. While some of the underlying concepts are true or at least reasonable, to a degree, the ultimate conclusion of phrenologists – that the bumps on the skull could be used to divine personality – turned out to be false. Modern neuroscience now far surpasses the preliminary knowledge on which the hypothesis of phrenology was based, and our contemporary perspective allows us to conclude with the highest degree of scientific certitude that two of the key assumptions of phrenology are incorrect, and in fact phrenology does not work. How then is it possible for belief in phrenology to persist? Frequent readers of this blog will likely suspect the answer.
Phrenology and Cold Reading
In the final analysis, practicing phrenologists were and still are using a method known as cold reading. Briefly, cold reading is the technique of making general statements about a target subject, statements which are likely to be somewhat true about almost any human being. In a dynamic cold reading, such as a psychic reading, subconscious feedback from the client is then used to make more and more specific statements, by pursuing the more accurate statements, the hits, and ignoring the misses. The results can seem very impressive, but the technique is actually quite simple once it is understood.
Cold reading can also be done, however, in a static fashion. In such cold readings, a limited set of pre-written statements concerning the subject are chosen according to some method. This can be done by making an astrological chart, handwriting analysis, or undergoing phrenological analysis. The pre-written statements, as with the opening statements of a dynamic cold reading, are designed to be vague and universal, so that anyone could see themselves to some degree in the statements. Readings such as “You like to be admired by others,” “At times you do not pay close enough attention to details,” or “You tend to feel more affection than you express to others,” are likely to strike a cord of recognition in all of us.
The most advanced expression of phrenology was the automated phrenology machine, which automatically measured the bumps on the subject’s skull, then produced a set of statements which were selected from 160 different possibilities, printed on small pieces of ticker tape and produced from the machine. I call this the “fortune cookie” style of cold reading. Phrenologists today also perform a more dynamic form of cold reading, by directly reading the bumps on a client’s skull while actively interpreting the results, allowing for the process of feedback and refinement.
It is interesting to see how such elaborate lists of correlations (palm creases with personality traits, and iris flecks with diseases) come into being. Some appear to have been made up out of whole cloth. Others are based upon a small set of uncontrolled observations, which are then presented as scientific evidence. This was the case for phrenology. Gall, in fact, developed his hypothesis initially after measuring the contours of the skulls of several family members and friends. He believed he detected certain patterns of bumps and personality traits in these individuals, and developed his theory of phrenology from this preliminary data. Gall then went on to develop the elaborate phrenological chart from this information.
Gall made a classic mistake in his methods – functionally he engaged in data mining and then assumed that the patterns he saw reflected underlying reality rather than random chance. He did not appreciate the fact that even if phrenology were completely wrong, he would still find correlations between the skulls of his subjects and their personality. He did not validate the apparent patterns he found with fresh data – by using the original observations to make predictions and then testing those predictions. He was not adequately skeptical of his own conclusions.
For a time, Gall’s work enjoyed prestige among the intellectual elite of Europe, who were just embracing the ideals of science and rationalism. His apparent systematic and scientific approach to the topic of human personality appealed to the rationalists of his time. It is likely that confirmation bias (a tendency to find confirming data and ignore or dismiss disconfirming data) reinforced his belief and confidence in his new system of phrenology.
As early as 1808, however, phrenology was already coming under scientific criticism. The Institute of France assembled a committee of savants, led by Cuvier, to investigate phrenology, and they concluded that it had no scientific basis. (Sabbatini, 1997) Over the ensuing years, scientists were unable to duplicate under more rigorous scientific conditions the phrenological charts made by Gall, and phrenology had failed a key test of true science, reproducibility. If the precepts of phrenology were correct, then any scientist in any lab could reproduce Galls chart through objective analysis of skull bumps and personality. It turns out, they couldn’t.
How, then, do believers defend their belief in phrenology, astrology, palmistry, iridology, or other such pseudoscience? Often the answer is simply, “I have seen it work.” They believe they have seen the method they employ work, and therefore the underlying principles must be true, no matter how much they appear to contradict established science. What true believers in such pseudosciences fail to appreciate is the basic skeptical principle that people are easily deceived, especially by themselves. The illusion of accuracy produced by cold reading often fools not only the client, but the reader as well, reinforced by each client who gushes over how accurate the readings were.
This principle is convincingly demonstrated by a classic tale told by psychologist and CSI fellow, Ray Hyman. As a youth he took up the practice of palmistry to make some extra money. He did not really believe in palm reading, but was amazed at the apparent accuracy of his readings, simply by following, cook book style, the formula in a standard handbook of palmistry. Before long he was convinced of palmistry’s fantastic power to penetrate the deepest secrets of his clients.
But Ray Hyman also had a scientific curiosity, an apparent rarity among palm readers. He decided to conduct a simple experiment to test the efficacy of palm reading. He began giving his clients the exact opposite reading as dictated by the handbook. To his surprise, his clients still claimed that his readings were amazingly accurate and were pleased with the results, to no less degree than when the “correct” readings were given. Dr. Hyman had discovered the power of cold reading and confirmation bias which lay behind palmistry.
Amazingly, phrenology persists to this day as a pseudoscientific form of divination. It is marginalized, and has no presence in mainstream science, but it persists none-the-less. There is, of course, a website dedicated to phrenology (phrenology.org) which proclaims:
Phrenology is a true science, which is there to benefit humanity.
Nearly two hundred years after Gall founded his theories, we can look back and assess the real value and impact of Phrenology. Today, much of the criticism against Phrenology can be easily dismissed. However, we should be careful not to repeat the errors that were done in the past and that, through shameless commercial exploitation or through the pursuit of evil, have disgraced the science and given it a bad reputation.
You might be curious how the site’s author deals with modern criticism of phrenology. He doesn’t. Rather, he addresses the 200 year old criticism that was mounted against Gall – as if it is still relevant today – and then ignores any of the points that are found in this summary. The site simply asserts that phrenology has been scientifically demonstrate through observation (confusing observation for science) but provides no references. Rather we are treated with positive quotations about phrenology from the likes of Thomas Edison.
Phrenology is certainly an interesting notion from a historical point of view. It emerged out of a largely correct view of neuroanatomy, but then became rooted in clinical practice even while continued progress in neuroscience was rendering it obsolete. Its originator – Gall – combined legitimate basic science with hopelessly naive clinical claims. He can hardly be blamed, as clinical science had not yet been developed and accepted.
Phrenology persisted long past the time that it had been rejected by the scientific mainstream, as practitioners were reluctant to abandon a lucrative procedure, but eventually science won out. By that time it had evolved into just another manifestation of divination through cold reading, and so it is no surprise that it survives as just that.
1) Gall, FJ, The Anatomy and Physiology of the Nervous System in General, and of the Brain in Particular, 1796.
2) Sabbatini, RME, Phrenology: the History of Brain Localization,
26 Responses to “Phrenology – History of a Science and Pseudoscience”
Leave a Reply
You must be logged in to post a comment.