Search This Blog

Thursday, September 30, 2004

What is Bad Psychology?

Bad psychology is a branch of bad science, and like bad science, it should be easy to separate the good from the boring, bland, or really awful. Problem is, unlike the physical and biological sciences, in the social sciences our language too often gets in the way. Science of course begins and ends with description, and description is merely an aspect of language, and is no more than a bunch of logically related symbols which correspond to or map the sum of our experience. Scientific inquiry starts with a symbolic accounting of these facts, which can mount up with the unending detail of a detective’s casebook. And truly, a scientist is a detective. In the popular imagination, he’s like a character from ‘Dragnet’, a Joe Friday lookalike with a white coat and an observant eye who looks upon mother nature with a mixture of dispassion and awe, and proceeds to ask her for ‘just the facts, mam’.

Of course, in modern life the facts often just won’t do, since their sheer profusion can easily do us in, so we shorten or parse the facts of life by explaining how we act and feel in terms of similar situations that we can analogously relate to. Like you know what I mean? We explain the world with analogy or metaphor, because more detailed explanations would be too long, too confusing, and are in general pretty unnecessary.

Take for example, a child’s description of throwing a ball. "Billy threw the ball real hard to get way high to fall out yonder with a big thud." We can easily understand Billy’s behavior and fill in all of the blanks in this description such as the facts of his throwing motion, his stance, and the arch of his throw. We ‘know’ these things because we’ve been there, done that, or at the very least have seen it being done. What we can’t know and usually don’t need to know is the force applied to the ball, its inertia, speed, exact angle, level of wind resistance, and the other physical factors which allow us to exactly reproduce Billy’s behavior and predict where the ball will go. These facts constitute the subject matter of the science of physics, which uses a special language of its own, the calculus, to figure out exactly where Billy’s ball is going. The calculus also has the side benefit of being applicable to all sorts of moving objects in space, such as cannonballs, pistons, and orbiting satellites, as all calculus inflicted high school students surely know.

We use Billy’s language to describe physical objects when we generally want to communicate the facts of observation and experience. However, if we sought to apply such thinking to bridge building, locomotive engines, and rockets, our subsequent career path would become suddenly confined to becoming chief engineer for some yet undiscovered Indian tribe in New Guinea. That’s because a single metaphorical description does not make for a complete description of that event, as science represents the attempt to achieve complete descriptions from many different perspectives and methodologies. This brings us now to our definition of bad science. Bad science is merely the supposition that a single metaphorical description of an event represents a complete description of an event. Billy’s description of the mechanics of ball throwing is Ok as a way of generally communicating what he is doing, but its bad science if he thinks it’s a complete description of his behavior.

Bad science occurs when you think you’re being scientific, when actually you’re not. That is, when you think you’ve got it, and completely understand a particular subject, you actually don’t. The delusion of completeness that permeates bad science has the insidious quality of short circuiting the analytical faculties in one’s brain since, after all, nothing more can be added. Thus, if Billy thinks the physics of throwing a ball is completely described by saying he threw it real hard, then he has no more need to think about the mechanics of throwing balls. This is bad news if Billy has aspirations of becoming a rocket scientist. But even though most of us don’t aspire to scientific careers, we still know that a statement such as the "the sun rises" represents a mere metaphor. That is, the sun acts as if it were rising, as we implicitly know that the sun’s actual movements are only accessible by levels of description that aren’t in the province of common speech.

Analogies are half truths, and like a dot to dot image, we obligingly yet unconsciously fill in the gaps. But how do we implicitly know that there are higher truths out there that describe Billy’s behavior a heck of a lot better? Clearly, we know that the statement that Billy threw the ball hard is not really a complete description of his behavior, but merely a cipher or placeholder that ultimately implies what the true facts are, even if we never consciously derive them. For the physics of everyday life, we guard against the conclusion that common sense descriptions are complete descriptions because we know that Billy’s throwing behavior is exhaustively described by the pre-established laws of physics. That is, we can jump from partial (but easy and speedy) descriptions as provided by common language to complete (but slow and difficult) descriptions that are provided by physics. Knowing even the rudiments of physics allows us to describe throwing behavior more realistically, improves our chances of controlling the behavior of the ball, and keeps us from hypothesizing weird and unlikely intermediary causes (e.g. , planetary conjunctions, or that taco I had for lunch today) that don’t bestow any predictive power at all.

Now, if Billy was stubborn, and a bit stupid, he may figure that physics books may have nothing much to teach him, since his own common sense language pretty accurately describes the necessary facts of throwing balls. Thus, throwing the ball hard means that it will go very high, and throwing the ball really hard means it will go very, very high. If the facts of throwing can only be generally described, then Billy cannot describe the specific details, but at least he can establish the reliability of his observation. To do this, Billy interviewed a few dozen or so of his buddies, and found to his great satisfaction that they too reported that throwing the ball real hard resulted in the ball going a lot farther than if they said they threw it softly. But to his amazement, he also discovered that some of them said that the ball would go a bit further if they put a little body english on the ball, or put a spin on it, or after they rubbed a lucky rabbit’s foot. By noting the correlations of all of these "constructs" with the movement of the ball, Billy could make all sorts of correlations between high balls, low balls, top spins, and rabbit feet. Although he didn’t know it at the time, Billy had discovered the perfect companion to bad science, namely, bad research! Indeed, when Billy grew up and entered college, he applied advance statistics to his observations to demonstrate that throwing a ball far is as easy as a real hard throw and a rabbit’s foot or two.

What is Bad Research?

Bad research confirms bad science, and proves once again the old computer maxim, garbage in, garbage out. Bad research has the virtue of reaffirming that you don’t know what you’re talking about, but this time with authority. Bad research merely verifies what you have poorly described, and in the large, makes your bad description consistent with everybody else’s. So if everyone else confidently predicts the flight of balls through the use of common sense language, then by virtue of your consistency with everybody’s opinion, your description of the mechanics of throwing a ball must be complete. Demonstrating the reliability of an observation does not bring you any closer to validating or describing the processes which underlie that observation, but with so many people agreeing with you, how can you be wrong? Besides, often the processes which cause an event to occur can never be known, so its safe to ignore them and concentrate on the correlations which we can put to use in our daily lives.

This type of thinking is Ok for common folk, who may never know or care to know much about the workings of the more complex facts of life. It’s enough to know merely that A causes B, and behave accordingly. For example, we know smoking causes cancer, but we are perfectly content with the easy analogy that smoking in some metaphorical way causes cancer. The correlation between smoking and cancer that was discovered from comparing the medical histories of millions of smokers and non-smokers shortened the long series of biological causes and effects that occur for each individual to a neat and easily remembered cause-effect link. Of course, this common sense view does not forestall the scientific endeavor to discover actual physical linkages between the components of cigarette smoke and the cell damage that ultimately results in cancer. Indeed, few well informed people would assume that the link was so simple. The original research which established a correlation between cancer and smoking would only represent bad research if the results were interpreted to represent the final say on the matter. That is, A causes B, and we need not look at anything in between because it’s impractical, unnecessary, or because intermediary events are simply not there.

We accept the rules of science when it comes to the behavior of amoebae, frogs, balls, and rockets, and understand and appreciate the ever refined descriptions of biological and physical events that provide ways of looking at the processes which animate our world. From DNA to Quantum Theory, these new descriptions allow us to fill out the common sense world of cause and effect, and permit the generation of new procedures that underlie the practical disciplines that range from genetic engineering to computer science.

Unfortunately, the same hard headed approach to science that characterizes the physical and biological sciences is scarcely evident in the social sciences, which for the most part uses the lingua franca of common speech and common sense as the best measure of man, and his behavior. Because we use common speech in our day to day lives, it’s easy to accept explanations which are based on those terms. Thus, if I need a pencil, we normally wouldn’t think twice if some learned psychologist postulated some inner psychological need for pencils. Likewise, the easy observation that Billy throws balls well (lets say he’s playing for the Yankees now) because he has a need to achieve, has a lot of character and courage, and loves God and country can easily translate into a roster of psychological needs, traits, and drives whose endless correlations can fill rows of journals and racks of pop psychology books.

What makes much of psychology so bad is its presumptive arrogance, the assumption that explanations for behavior can be valid simply because they refer to the accepted authority of common speech and the consensus of common opinions. Unfortunately, the motives which drive Billy as well as his own account of how he throws a ball both serve to explain his behavior, but they still don’t quite fully describe his behavior. When no higher terminology is available that, like the laws of physics, can accurately describe the facts of behavior, then common sense terms are used instead. However, common sense can only partially describe behavior. Ultimately, you need a better language to describe behavior, but how do you derive such a language?

What is good science, or scientific thinking?

Scientific thinking is, well, elementary. Like Sherlock Holmes, you start out with some obvious facts, connect them with a bridge of hypothetical events, and proceed by a logical process of elimination to a final explanation that, regardless of how unlikely, is true. Now most of the workings of the world are hidden to even the most astute intellects, and we often come across gaps in our description of events that require the creation of new entities that can fill in those gaps, and make the resulting description neat and tidy. This is called saving the appearances. For example, to medieval scientists, concepts like fire and the transmission of light could only be partly described. To save the appearances and give the semblance that they knew what they were talking about, they hypothesized mysterious entities called phlogiston and ether that were respectively the agent of combustion and the medium that transmitted light. Of course these new concepts added nothing to the understanding or prediction of how fires start and how light travels through space. But it made these explanations at least sound more erudite, and the pretensions of being wise, then as now, counted for public respect, funding, and most importantly, tenure! Of course, phlogiston and ether have long since been discarded by contemporary physicists. Whether they exist or not is anybody’s guess, as they, like leprechauns, centaurs, and men from mars, can never be totally disproved. What is true is that they proved to be totally useless. Since we don’t need phlogiston to build or explain fires, and we don’t need ether to explain the speed or composition of light, scientists have rightly discarded them as unnecessary verbal baggage.

Now saving the appearances is generally not a bad thing, and indeed is the essence of the scientific enterprise. It is what scientists do, and gives them a reason for, well, being scientists. Saving the appearances is called deduction, or deriving particular facts (e.g. The butler did it.) from general observations (e.g. Mr. White, in the pantry, with the nylons). However, to play the game of science, you must not just make a deduction, for that’s child’s play, but justify it and prove it to be true. That is, the rules of scientific deduction not only mean that you must have something to actually deduce, but that there must be some logical reason why you are making that deduction, and some logical way you can later demonstrate that your deduction represents a real fact. For example, to say to a traffic cop that demons made you go through a stop sign is not exactly something one would logically derive from the circumstances, and it is certainly not something that one can demonstrably prove. The observation that you’re ‘hell on wheels’ doesn’t provide the logical basis for the deduction that the devil made you do it.

To fully describe anything, you must make and logically justify a deduction, employ the means to prove it true, and then go right back to making the next deduction. This can result in an infinite regress, as the description of the properties of matter may be decomposed into elements, atoms, quarks, and quantum effects. What sets bad science apart from the rigorous deductive standards of good science is that logic doesn’t count for much as a tool for describing and validating events. Like Humpty Dumpty, things are the way they are because that’s the way we say they are. What makes us accept bad science so uncritically is that seeming is believing, since that’s the way we generally describe the ways of the world.

There’s a lot of high irony in this, since a little logic can go a long way in demonstrating that the most profound sounding theories are actually bone headed opinions that would seem stupid and foolish to even a six year old. Science is full of such precedents, from the off kilter theories of the great ancient physician Galen, who thought that the purpose of the brain was to warm the blood, to Ptolemy’s sensible observation that the sun revolved around the earth, to the infallible pronouncements of Aristotle, whose crack brained physics inspired Galileo to timeless ridicule. Who are our new Galens, Ptolemys and Aristotles? In the pages that follow we will show that they have found a secure and loving home in the field of psychology. We will note with amazement and awe the over arching profundity and infallibility of so much of modern psychological thinking, timeless stuff that would have made Aristotle proud.

And our answer to all of this? Following the traditions of Galileo and Swift, I will simply show the irony of it all.