[home] [research] [getting started] [law] [personal stories] [secrets] [news] [about Trancenet]

Markovsky's 10 Problems with Common Sense

1. Common Sense Explanations Can Be Circular

   In a circular explanation, that which is to be explained appears as part of the explanation, often in a disguised from. A simple example appears in the following exchange:

   Person A: Do you believe in God?

   Person B: Yes.

   Person A: Why do you believe this?

   Person B: Because the Bible says He exists.

   Person A: Why do you believe what the Bible says?

   Person B: Because it's the word of God, of course!

   Saying that the Bible is "the word of God" already presumes a belief in God's existence, and it was that belief that was supposed to have been justified. Usually, circular explanations are much more difficult than this to identify. They can be hidden deep within layers of discussion, spread thinly over hundreds of pages, or obscured by changes in the words that are involved. Importantly, however, the conclusion of a circular argument is not automatically wrong. Circularity in the god-and-bible argument doesn't preclude the existence of God. In general, circular arguments are just irrelevant to whatever one is trying to explain. They have no bearing one way or the other.

   Scientific theories use several methods to help guard against circularities. These include adopting a formal language involving logic or mathematics to organize and check their statements, using as few terms as possible and making them as unambiguous as possible, and having others critically evaluate the theory. A formal language guards against circularities by making them immediately evident and defining them as problematic. Minimizing the number of terms used to express a theory simplifies its analysis and, again, helps to make circular statements more evident. Having the theory critically evaluated by others is a powerful method for locating problem areas and often leads to suggestions for remedies.

2. Common Sense Explanations Can Be Ad Hoc

   An ad hoc explanation is one that is concocted for a specific observation. Often it is a very general explanation, even though only based on a specific case. Because it is based on a specific case, the ad hoc explanation tends to ignore more general factors that may be relevant to the observation. For example, a sexual assault may be "explained" by blaming the victim: "Women are attacked because they sometimes dress too provocatively." This ad hoc reasoning ignores many factors that may not be apparent to the casual observer: The location of the assault was not a frequent site of similar violent crimes; the attacker had been stalking the victim for days; the victim was beaten and robbed and conservatively dressed; the attacker has a history of violent criminal activity. The ad hoc explanation ignores larger patterns of which the particular event is but a part, and ignores particulars that may not be consistent with the explanation.

   Ad hoc explanations are attractive as tools of common sense because they give us a feeling of security and control -- the feeling that we understand life's complexities and so can choose to seek pleasant outcomes and avoid the harmful ones. Common sense says that one may avoid sexual assault by dressing conservatively. That's a myth.

   In science, ad hoc explanations are frequently offered as provisional explanations for anomalous findings -- that is, unexpected observations. They are also widely recognized as practically useless. They may eventually prove fruitful by motivating the development of alternative theories and new directions of research. But by themselves, ad hoc explanations take us no further than the observation for which they were thought up.

3. Common Sense Explanations Can Be Particularistic

   The scope of common sense is often limited to particular times, places, and things. This is not usually a problem. Obviously it's very useful to be able to anticipate a friend's responses to your words and actions, or to know how your car behaves on wet roads. Most of us also have deep interests in certain worldly events, and so we turn to journalists who, as part of their jobs, obtain as much information as they can pertaining to specific occurrences. But a particularistic approach is self-defeating when the goal is to gain knowledge that can apply to other times, places and things. For instance, no amount of detailed information about the workings of the 1991 board of directors of General Motors will by itself increase our theoretical knowledge about group processes.

   Let's more carefully distinguish two types of approaches to gaining knowledge: the particularizing strategy and the generalizing strategy. If we accept that the construction of general theories is a goal of science, then particularizing approaches do not provide much benefit to science. In sociology, there is quite a lot of use of the particularizing approach. It usually takes the form of a collection of relatively unorganized, relatively concrete, and relatively common sense propositions about social behavior. This problem boils down to just this: The greater the depth of knowledge acquired with regard to any particular phenomenon or entity, the less the degree to which that knowledge will apply to anything else. This runs counter to the goal of theoretical generality.

   In contrast to a particularistic focus, theories are designed to be generalizable. Abstractness is the quality that allows this. Common sense often tells us that abstractness is a bad quality because it implies detachment from reality. Of course, a theory with no connection to reality is not going to be of any use in explaining real events. However, total concreteness would prevent any generalization beyond the case at hand. So good theories are themselves abstract, but they are designed to allow connections to be made with real phenomena. This lets one theory explain an unlimited number of different kinds of events.

4. Common Sense Explanations Can Be Post Hoc

   This may be the most serious problem with common sense explanations of paranormal claims, but one of the easiest to overcome using standard scientific procedures. The post hoc fallacy is the claim that event A caused event B because B followed A. Humans and numerous other species are physiologically wired to learn by recognizing temporal contiguities. These are events that follow one another relatively closely in time. We recognize the temporal contiguity and infer from it that the earlier event caused the later one. This inference of causality is oftentimes valid: Lightening does cause thunder, and a wound does cause pain. Sometimes, however, our programs backfire and we make false inferences. For example, millions of Americans believe all sorts of phenomena, from crime waves to hospital emergency ward admissions to losing car keys, are caused by full moons. Study after study has found no such effect, but it still seems very real to those who believe in it.

   The logic of scientific theorizing and testing actually cannot prove that a causal relation exists between events A and B. However, theories can generate testable hypotheses that should hold up under testing if the events are causally related. Statistical techniques are then applied to determine whether those observed relationships are unlikely to have been due to chance alone. To use the example I just mentioned, many people believe that increases in crimes, natural disasters, and other phenomena are associated with the full moon. To support this claim, believers in the moon effect tend to ignore the lack of theoretical or statistical basis for such relationships. Instead they may base their belief on what others have said, or perhaps can even recall specific events that occurred around the time of the full moon.

5. Common Sense Often Deals With Surface Features

   A good example of this is the formation of first impressions. We all quite easily formulate hunches about the deeper aspects of acquaintances' personalities, often based on very limited information. Even though we may feel confident about these impressions, they often turn out to be ill-informed and we rarely check their accuracy.

   Another example is our tendency to attach a great deal of weight to the first explanation that we hear for something that we don't understand. People often accept any scenario that appears to account for the surface features of a complex phenomenon. A related problem is that sometimes we take the lack of a known conventional explanation to support an unorthodox explanation. For example, a newspaper in upstate New York reported that citizens and scientists alike were at a loss to explain colored lights in the evening sky over a certain small city. Members of a state UFO-enthusiast network, questioned by the press, announced that this inability to identify the lights provided further evidence that the earth is under surveillance by alien life forms. The failure to account for the unidentified objects, however, does not increase the likelihood that they are the spacecraft of alien visitors. As I've said before, "unidentified" means unidentified. In general, there is no scientific shame in having insufficient data to draw conclusions. The only shame is promoting speculations whose only merit is an ability to weave surface features of phenomena into a whimsical fabric.

6. Common Sense Allows Contradictions

   A contradiction amounts to saying that something is both true and untrue. When friends are separated, we may want to predict whether their relationship will strengthen or weaken. Two famous sayings generate contradictory predictions: "Absence makes the heart grow fonder," and "Out of sight, out of mind." Each statement seems common sensical, but both cannot be simultaneously true without some type of reinterpretation. And yet both bits of folk wisdom remain in our storehouse of common sense, telling us in essence that separated friends will either forget about each other or grow closer. In science, if we wish to know whether or not y will follow x, it is altogether unsatisfying to be told that "if x occurs then y will follow, and it will not." Also, note how if somebody did a study to see what happens to friends or lovers who have been separated, whatever the result, someone is sure to come along and point out that common sense predicted it all along. What this critic will fail to mention is that common sense also predicted the opposite to hold.

   It is worth noting that if we allow contradiction in our explanation, then any false statement that you can imagine may be logically derived and proven true. Contradictions lead to absurdities, and that's why we have to take care to exclude them from our explanations. Theories guard against contradictions by adopting logical frameworks to help locate and eliminate them.

7. Common Sense is Unconditional

   A condition is a phrase that places constraints on a statement. For example, "grass is green" is an unconstrained, unconditional statement. To it we could add, "in the spring" as a limiting condition. When common sense does try to reach beyond particulars in search of generality -- such as with "absence makes the heart grow fonder," it is often unconditionalized. That is, the conditions under which the claim applies are either not considered or left unstated. Before my first science class I learned that water freezes at 32ºF -- an unconditional statement. What a rude complication it was for me to learn that water in the pond near my childhood home froze before water in the river that fed it, and that ocean water 25 miles away could be much colder than "freezing" without freezing! The actual freezing point is conditional on such factors as the purity of the water, its rate of flow, and the barometric pressure. Notice that just because the statement is conditional does not prevent if from being general. The statement applies any time, any place, to any vessel of water, if the appropriate conditions are satisfied. That amounts to a potentially infinite number of times and places.

   Conditionalization adds precision to theoretical claims and prevents their application to phenomena outside of the intended purview of the theory. Another benefit of conditionalization is its capacity to resolve apparent contradictions. For example, it may be reasonable to claim that for two people who have regular face-to-face contact and positive evaluations of one another, "physical separation increases emotional attachment" under the condition that the relationship has progressed beyond the level of an acquaintanceship when the separation takes place. The effect of separation may be claimed to be the opposite under the condition that the relationship has not progressed beyond this critical stage.

8. Common Sense is Not Systematically Tested

   It is easier and more psychologically satisfying to simply believe that we are correct about something than it is to carefully test our beliefs and risk being wrong. This is not to say that we have to test every opinion we hold, every observation we make, or every statement we utter. Much of what we believe, observe and say is based on time-tested practical knowledge, and little would be gained -- and much time and energy lost -- by subjecting this knowledge to systematic examination. It does say, however, that we should not be surprised to find that common sense messes up from time to time. If we are honestly concerned with the truthfulness of particular explanations and beliefs, and also with the possible consequences of acting on their behalf, then testing is the best way we know of verifying those beliefs, eliminating false alternative explanations, and resolving differences of opinions.

   Systematic testing is part of a method for improving upon untested common sense. This is because we may legitimately attach more confidence to claims or beliefs that have survived direct attempts to disconfirm than those that have not been so tested.

9. Common Sense Allows Subjective Validation

   Subjective validation is a method that we all use to verify our beliefs. It is a form of non-systematic testing that involves accepting evidence that supports a belief, and ignoring neutral or unfavorable evidence. While post-hoc inferences and other common sense methods may plant the seed of a belief, subjective validation helps it to flourish.

   In the example of belief in the "moon effect," once the possibility is allowed that moon phases affect people, attention is drawn to those events that can be interpreted as supporting the belief. Quite often this involves granting a good deal of latitude in those interpretations. A rise in ER admissions a few days prior or subsequent to the full moon may suggest an effect, as might reading about a moderate earthquake in Europe, a hefty drop in the Dow Jones Industrial Average, or a riot involving ardent fans at a London soccer match. During other lunar phases, the believer simply doesn't notice how often these events occur. During or close to the full moon, virtually any non-ordinary event may be interpreted as caused by the moon. Overlooked is the fact that such events actually occur very frequently and uniformly across all lunar phases.

   Many people can recount a time that they saw something in a dream which later came true, and so believe that dreams are capable of foretelling the future. For example, some have reported dreams in which a certain loved one dies, followed soon thereafter by his or her actual death. In fact, such dreams turn out to be fairly common -- especially when the loved one is elderly or ailing. We also tend to forget that the contents of the vast majority of our dreams are not played out in reality, and that we are prone to fill in many of a dream's details long after awakening. Still, the coincidence of dream and reality can produce such an emotional impact that the subjective validation of a dream's ability to foretell the future almost inevitably follows. In this example, the belief is quite harmless. Unfortunately, however, subjective validation is also at the root of many harmful beliefs and perceptions. The same mechanism that allows us to validate unfounded beliefs about dreams or moon phases also permits us to hold harmful and false beliefs such as racial and sexual stereotypes. It also leads to a misplaced faith in our abilities to make accurate judgments under circumstances that are complex or ambiguous.

   Scientists are humans too, and they often engage in a process of subjectively validating their theories. This alone, however, is not (and should not be) sufficient to convince others of the truthfulness of a theory's claims. To be generally accepted in a field, a theory's tests must survive the collective scrutiny of skeptics. These skeptics are bound and determined to point out when a test is too weak to rule out alternative explanations. Subjective validation fails as a theory-testing method precisely because it systematically ignores information ... that would suggest alternative explanations for observed phenomena.

10. Common Sense is Unorganized and Vague

    Common sense is unorganized because it lacks a system for keeping track of what is known already, the amount of evidence backing that knowledge, and how existing information can be used to garner further knowledge. The difference between the ways that common sense and theory are organized is something like the difference between a long list of independent sentences and a book with sentences organized in paragraphs, section, and chapters. Long, unstructured lists are extremely inefficient storehouses for knowledge. In constructing new theories, a great deal of attention is paid to what theories already exist, the kinds of evidence offered for them, and how they can be improved by making them simpler, more compact, and still as accurate and comprehensive as possible.

    Common sense is vague in the sense that its terms lack unambiguous definition. What do we really mean when we say that someone has a good personality? A strong will? A mean streak? What are we actually talking about when using the ideas of "power," "status," "love," or "justice" in everyday discourse? If you and nine of your friends wrote down definitions for each of these terms, chances are we would wind up with ten different meanings for each. This is a problem because explanations -- scientific or common sense -- can only be of use if we first agree on the meanings of their terms.

    In theories, it is perfectly reasonable to assume that there is no absolute, "true" definition for a particular idea or concept used in the theory. Instead, the theorist will try to insure accurate communication by stating exactly the meaning that he or she intends to communicate. This is done by defining the concept. Then it is possible to have the concept of "injustice," for example, defined in one theory as the "violation of objective allocation standards," and in another theory as "the emotional response to the violation of objective allocation standards," without implying that at least one of the definitions must somehow be wrong.

[home] [research] [getting started] [law] [personal stories] [secrets] [news] [about Trancenet]