Phrenology

Our next allegedly clear example of a pseudoscience is phrenology. Crudely put, phrenology is the attempt to determine a person's character by examining the bumps and pits on his or her skull. Although many of us might not like to admit it, we do often judge someone primarily by his or her appearance, despite the fact that we ought to know better. In studies done by social psychologists, subjects have been found to rely primarily on the incorrect equation "intelligence = beauty." In some ways, phrenology is simply a more precise form of the common underlying prejudice that you can judge the contents by the package. But there is much more to phrenology than that. To see what more there is to it, we need to take a brief trip through the history of neuroscience. It will turn out that, for all its errors, phrenology had important hypotheses to contribute to modern neuroscience. It is recognized that no history of the development of modern neural science is complete without a discussion of phrenology. Of course, there are two sorts of reasons why it might deserve such discussion: phrenology may have had no correct and important content, but profound effects; or it may have had effects because of the importance of its content.

One nineteenth-century writer offered the following comparison:

This is a fair description of the relationship between astrology and astronomy. Let's see if it holds for phrenology, or if there is more to phrenology than this comparison suggests.

History of Neuroscience: The Brain Hypothesis

The belief that mental function is located primarily in the brain has not always been held and is certainly not obvious. (Suppose you were challenged to prove it, from scratch, with no textbooks in which to look up the answers. It's not immediately obvious how to proceed, is it?) From about 500 B.C. to A.D. 1500, there was debate over whether the heart or the brain was the location of the mind. The ancient Greek scientist and logician Aristotle argued (around 350 B.C.) that the brain served merely as a radiator to cool the blood; he gave pride of place to the heart, which was a detectably warm and active organ. Although he knew that human brains were larger in proportion to body size than other animals' brains, he thought it was because humans had richer and hotter blood and so needed the better radiator. He also knew that most of the waste heat that people lose is exhausted through their heads (so parents' advice to kids to "keep your hat on - it's cold outside" actually has a basis in fact, unlike much of what worried parents say). There were other ancients who held the brain hypothesis instead, and who sought to support it with observation. Galen (A. D. 129 - 199) was an early practitioner of sports medicine, as doctor to the Roman gladiators. He saw plenty of brain injuries, noted their effects, and argued for the brain hypothesis. Although the cardiac hypothesis is no longer taken seriously, it has left its mark on our language for describing our emotions: love is a heart pierced by an arrow; people are said to be heartbroken; the apathetic don't put their hearts into it; angry people sometimes say, "That makes my blood boil."

The modern debate about the nature of mental function begins with the great seventeenth-century mathematician, scientist, philosopher, and mercenary soldier René Descartes (1596 - 1650). In part out of concern to be consistent with religious doctrine, he presented a dualistic view of human mental activity. We will examine this view more fully later in this book. For now, we can describe Cartesian Dualism as the view according to which people have a dual, or two-part, nature: a purely physical part, the body, and a purely mental part, the nonphysical soul, or mind. Descartes held that mind and body interact at a precise spot within the brain (the pineal gland). But he was quite clear on one crucial matter: mental function does not take place in the brain; it takes place instead in the soul or mind and transmits its effects through the brain. On this view, the brain is nothing more than a sophisticated organic transceiver for the mind. Mind, Descartes held, is unitary: you don't feel love with one part of it, process visual sensations with another part of it, and make decisions with a third. Mind is not a complex physical object and has no parts; whatever it does, it does as a whole, as one undivided thing. Descartes thus took a negative view of what might be called the "hypothesis of localization of mental function", and he took a stand against holding the brain to be the locus of mental activity.

In modern times, the contrary anti-dualistic position was first argued for by the pioneering phrenologists, the Viennese neuroanatomists Franz Josef Gall (1758 - 1828) and Johann Casper Spurzheim (1776 - 1832), Gall's disciple. They opposed Descartes on two scores: (i) they believed that the brain is the organ of mental function, and no mere transceiver; and (ii) they believed that certain parts of the brain were responsible for particular mental functions, including higher mental functions, like love or religious sentiment. They were thus early champions of the hypothesis of localization of mental function. Both Gall and Spurzheim were accomplished neuroanatomists, with a number of important discoveries to their credit. When they embraced phrenology, they left their anatomical researches far behind.

Phrenology seems to have begun with Gall's hypothesis that an enlarged memory center in the brain, which center he took to be behind the eyes, could cause the eyes to protrude. Generalizing on this rather modest initial hypothesis, Gall and Spurzheim identified 27 regions on the skull associated with particular mental functions. A bump in the region signified increased capacity for that function; a depression, or pit, signified a decreased capacity. For example, the associated the region just below and behind the ears with Amativeness - tendency towards romantic love. But with the regions of Destructiveness and Combativeness so nearby, an expert's eye may be needed for any judgments about personality. (Every time I talk about this subject, about 20% of the people in the audience begin checking behind their ears. What on earth do they expect to discover? Does a bump in all three regions excuse their appetites for kinky sex?) Further refinements of phrenology led to a larger number of regional subdivisions, over 80 at one time. Gall and Spurzheim sought to gather data to support their hypotheses and examined hundreds of skulls of both human and animals, never once finding any refutation of their views.

Phrenology had plenty of opponents among other scientists of the era; the most effective among them was Pierre Flourens (1794 - 1867), himself a Cartesian. But phrenology also had plenty of adherents among medical researchers and the educated public. At the height of its popularity in this country, during the mid-1800s, it was a widely accepted theory of how the mind worked. It had all the social trappings of science: the respect of scientists and of educated lay people; training programs; application in the care of the diseased and the deviant; a significant economic role; even its own learned journals, thick with technical jargon. People were hired and fired on the basis of phrenological examinations. Some job ads specified prior phrenological certification (much as some employers now require polygraphing). Just as we are nowadays given the results of the President's medical exam, politicians then would make public the results of phrenological examinations to help show their suitability for public office. Since insanity could be seen as a result of innate malfunction rather than divine curse or moral depravity, attitudes toward the insane took a turn for the better and conditions at some asylums improved; similar benefits accrued to the prison inmates of the day, with great rejoicing when Sing Sing became a "phrenologically conducted institution," Among phrenology's more eloquent adherents were Horace Mann, Walt Whitman, and Edgar Allan Poe.

It's also interesting that Flourens's supposed experimental refutation of Gall's and Spurzheim's guiding hypothesis of localization of brain function was badly flawed. How would you show that the human brain is unnecessary for mental activity? Using human subjects was not feasible, so Flourens settled for what he regarded as a close second best: he cut the brains of chickens by about 90% and observed them running about ("like chickens with their heads cut off"). (This, by the way, is the source of the oft-repeated remark that we human beings use only about 10% of our brain capacity. Unless you think that you're a chicken-head, you have no reason to accept that remark.)

It took many more decades of research to put the hypothesis on firmer ground. It is still the subject of active research and is a guiding influence in hypothesis formation in neuroscience.

Is the comparison with which we began a just one? Were the effects of phrenology on the development of science incidental and largely independent of its content? This brief history makes it clear that the answer is "no." By opposing the Cartesian views in some detail, phrenology helped to advance a science of the mind.

Is Phrenology Pseudoscience?

Some writers describe phrenology as prescientific, or as false science. The suggestion is that, like astrology, it may have given science a needed push in the right direction, but without itself being science. On what principled basis might phrenology be dismissed as pseudoscience?

Although the phrenologists did not pretend to know how the various regions of the brain worked to produce their hypothesized capacities, the absence of any knowledge of the physical mechanism was given little or no notice until the hypothesis of localization of function was championed. This third point is especially important because it reminds us that the search for new information won't even begin until there is a hypothesis to prompt and guide it. Science does not proceed by blindly going out and looking for new facts, by randomly doing experiments. The Cartesian conception of mental function made research into many mental functions virtually pointless. When phrenology recommended a new and detailed conception, particular experiments were then called for to either confirm or refute the hypotheses suggested by the new conception. So a key role for any scientific theory is to highlight ignorance.

It's been a long time since astrology has been taken seriously by good minds. But phrenology was admired by some of our grandparents' grandparents - not so long ago. It provides a useful test case for two popular gatekeeping criteria.

We can all agree that the various branches of science have extremely complex and interesting social roles in contemporary society. In addition to all the rites and rituals required of those who seek undergraduate and graduate degrees, postdoctoral fellowships, and other sorts of advancement in academia or industry, there are many professional societies, editorial boards for book and periodicals publishing with a system of peer review, advisory boards convened by government and business... The web of power relationships is important and intricate. One of the main purposes of graduate school in science is to teach the student how to get grant money - and if you think that's a joke, try managing a career in science without the requisite political skills and knowledge.

Some sociologists have suggested that science not only has, but is defined by, its distinctive social role. (One influential argument for this view is based on the view that science is "socially constructed," in the sense that it is, and is not just defined by, a social role.) If this approach is to classify phrenology as its critics believe it should be, then

But this is both unfair to phrenology and a poor way to distinguish science from pseudoscience. In the early nineteenth century, phrenology's social role was more like that of twentieth-century physics than was the social role of early nineteenth-century physics. Social roles take time to grow, and any relatively new branch of science will likely not have had sufficient time to establish the web of social and political relationships that we associate with entrenched science.

The main problem with the social-role definition of science is that it has no connection to the sorts of cognitive standards that ought to be in the forefront of debates about pseudoscience. We can readily imagine a crackpot multibillionaire spending part of his fortune to secure for his lunatic views a social role like that of established branches of science. Nor is money necessary. All that's really needed is access to political power. Lysenko managed to persuade Stalin that evolutionary theory was anti-Marxist and anti-Leninist. Lysenkoism held that acquired characteristics could be inherited by the offspring of those who acquired them, so Soviet citizens 'improved' under Stalin's regime would be 'scientifically' assured that their children would inherit those improvements. Those biologists who disagreed with Lysenko were dealt with very effectively, by being put to death, for example, and Soviet biology was set back many decades relative to the rest of biology. So much for the social-role definition.

It's clear from our discussion of criticisms of astrology that being based somehow on observation, or being testable, is not enough to make a theory scientific; the theory must have other characteristics, as well. Just as a good person is known by her virtues, so is a good theory thus known. While there isn't complete agreement on what the theory-virtues are, there is substantial agreement. I'll mention five characteristics that everyone agrees belong on the list.

The most obvious virtue is internal consistency. A theory is no good if it says, for example, about one and the same event that it will and that it will not occur at a given time. No acceptable theory contradicts itself. If a theory is sufficiently complex, then assuring its internal consistency can require great skill.

Most of us have some beliefs that are not well-supported by evidence or reasoned out with great care. Fortunately, most of us have some beliefs that are well-supported by evidence carefully reasoned out, and these beliefs are worth conserving. If a theory comes along that conflicts with such worthwhile beliefs, then, all other things being equal, the theory's in trouble. So conservatism, understood as consistency with previously held, well-justified beliefs, is another virtue for theories.

One way of taking the measure of a theory's explanatory power is to find out how general it is in its intended domain. Before Newtonian mechanics came along, there was serious doubt about whether the same set of natural laws governed both Earthly and celestial objects. Newton was able to show how one set of laws could explain both, and this theory thus enjoyed a great triumph of generality. On a less grand scale, we would rather have a general theory of conductivity of all materials under all circumstances than a theory that covered just long, thin pieces of copper wire at room temperature - useful though the latter would be. Other things being equal, the more generality a theory has, the better.

It's not always wise to maximize conservatism, however. By giving up on previously well-justified beliefs, we can sometimes gain a great deal in generality. If this happens in a big way, we call it a 'scientific revolution'. When Einstein proposed special relativity, it was immediately accepted by physicists despite its inconsistency with Newtonian mechanics because relativity theory was so much more general than the latter. They knew they were getting a great deal, so they readily accepted the trade-off between conservatism and generality.

Precision is a widely recognized virtue of theories. Colleges and universities do a lot to train students in assessing the numerical or quantitative precision of theories, and that's very important. But there's another kind of precision that's at least as important: definitional or qualitative precision. The fundamental concepts that a theory offers us as tools for explanation should have well-defined boundaries and clear conditions of application. (See "The Meaning of 'Life'" for more on analytical definition.) The more precise a theory is, the more strongly it will tend to be confirmed or disconfirmed by the evidence.

Many add simplicity to this list of virtues. I'm not sure what simplicity is. Is it a feature of the theories themselves? Some theories that seem very complex to me seem quite simple to the relevant experts. Is simplicity then relative to educational background and brains? I do know that some very simple characterizations of simplicity are too simple to work. For example:

Despite its vagueness, we can see that this can't be right. Suppose a Theory of Everything 1 said that all matter in the world was made out of just four different kinds of particles, while a Theory of Everything 2 said everything was made out of (just) eight different kinds of particles. Is Theory 1 necessarily simpler than Theory 2? Of course not. The laws and relationships that Theory 1 postulates may be far more complex than those postulated by Theory 2, and the net result may be greater overall simplicity for the latter. We need a general account of the simplicity of laws or relations before we can even get this approach to simplicity working. I conclude that simplicity is very complex, and I'll leave it at that.

Just to have a short expression for all this, let's give the label "systematic" to the combination of these virtues of theories. So, we can now say, the more systematic a theory is, the better. And a theory that is utterly nonsystematic is worthless.

I gave you a 'rule of thumb' test for assessing progress in science (in "Astrology"). I'll give you a similarly rough test for systematicity. To determine how systematic a theory or part of science is, gather up all the relevant texts and weigh them - the lighter they are, the better. This test works pretty well because highly unified, systematic theories can usually be boiled down to a few basic principles that can be clearly and briefly stated (often using mathematical formulae), and the rest of the text can be devoted to illustrating their applications via instructive examples. Even (or especially) an advanced text on classical (Newtonian) mechanics in physics will be relatively short and won't weigh much. Things are different in medicine, which deals with the most complex systems in the known universe; there isn't as much in the way of systematic theories in medicine, so medical texts tend to provide a multitude of different ways to categorize huge numbers of facts for different medical purposes, and this makes them very large and heavy. There is, for example, one textbook on eyes that's fifteen thick, oversized volumes long. (Ironically, it's called a "system.")

If we're to deploy these ideas against phrenology, we'd have to argue that

But phrenology was highly systematic, in fact, and also quite specific in its predictions. The basic text of phrenology weighs just a few pounds. It may be better off in this respect than some of the cookbook-systems accepted by modern doctors. While phrenology was nonconservative in its inconsistency with dominant Cartesian views, it offered in return a more general account of nervous system functioning, which seems a fair trade-off between conservatism and generality.

The main problem for us is that our picture of science does not allow us to use systematicity in assessing theories because while it makes perfectly good sense to say that one theory is more (or less) systematic than another, there is no rational basis, with this sort of standard, for drawing a sharp line between starters and nonstarters in the race for truth. In the Introduction, however, we agreed provisionally to adopt the common view that a theory is sharply defined as either a starter or a nonstarter, where there's no middle ground. Given the inherent plausibility of assessing theories by systematicity, this is our first clear signal that, perhaps, something is seriously wrong with our provisional picture of science. We will soon find that this is so.

In some ways, then, phrenology seems clearly superior to astrology. One might say that, at worst, it consists of a tangled weave of science and pseudoscience. It is apparently quite difficult to separate the threads.

Additional Sources

2000 David F. Austin

Go to next section on Parapsychology