How We Know What Isn't So (19 page)

The reader should also understand that this section deals with verbal and written communication rather broadly defined—from face-to-face conversation to the dissemination of information through print and broadcast media. One consequence of this focus is that certain words are used very broadly. For example, the word “speaker” is intended to refer to any of a host of different “transmitters,” such as writers, broadcasters, or the person doing the talking in face-to-face conversation. Similarly, the words “communication,” “conversation,” and “interaction” are sometimes used interchangeably, as are the words “listener” and “audience.”

*
Estimates of the transmissibility of the AIDS virus vary enormously, but the estimate provided by the most visible proponent of the “myth” of heterosexual AIDS, Michael Fumento, is that there is a 1 in 500 chance of an infected person giving the virus to an uninfected partner through a single episode of penile-vaginal intercourse. The odds are considered to be somewhat less likely for penile-oral intercourse, and enormously higher for penile-anal intercourse.

*
The various distortions in the case of Little Albert are mentioned here
not
because they represent particularly egregious examples of either calculated or subconscious distortion. They do not. Rather, they may represent fairly typical examples of the tendency to do a little sharpening here, a little leveling there, in order to make a better story. Indeed, I must confess that I found it difficult to avoid sharpening a few points myself in order to make my own point more clearly. Telling a succinct, coherent story demands that one sharpen and level, and even when one tries to tell the story perfectly straight, doing so can be difficult.

*
The divorce rate is generally calculated by dividing the number of divorces by the number of marriages in a given year. For several years now, there has been one divorce for every two marriages, hence the phrase, “half of all marriages end in divorce.” Note, however, that this calculation includes multiple divorces, and thus exaggerates the likelihood of divorce for the average person. How much it does so is difficult to assess because of another problematic feature of this statistic: The overall divorce rate is the sum of the various age-specific divorce rates in a given year—the divorce rate of people in their 20s, 30s, 40s, etc. However, the cohort of people who are currently in their 50s and 60s may be less inclined to get divorced than today’s younger cohort may be when they reach that age. As a result, estimates of the chances that a young married couple will get divorced at some time during their lives are very uncertain and should be interpreted with caution.

W
hat we believe is heavily influenced by what we think others believe. We favor or oppose experimentation with sex, drugs, and various other “lifestyle” practices in part because of what we think other people think, or do, about these matters. We consider a theater production to be worthy or unworthy of our attendance partly by the number of people who line up to see it. When asked at the office to donate money for a “going-away” gift for someone, we usually try to find out how much others have given and then decide our own contribution accordingly.

Within limits, this tendency to let the beliefs of others influence our own beliefs is perfectly justified. What other people think and how other people behave are important sources of information about what is correct, valid, or appropriate. Other things being equal, the greater the number of people who believe something, the more likely it is to be true; the more people who do something, the more we are well-advised to do the same.

Unfortunately, our ability to utilize effectively the opinions of others as an important source of indirect information about the wisdom of our actions, or the validity of our beliefs, is compromised by a systematic defect in our ability to estimate the beliefs and attitudes of others. We often exaggerate the extent to which other people hold the same beliefs that we do. Because our beliefs appear to enjoy more social support than is actually the case, they are more resistant to change than they would be otherwise. Thus, our difficulty in accurately estimating what other people think represents an important determinant of the maintenance of erroneous beliefs.
*

The idea that we project onto others our own beliefs, attitudes, and predispositions has a long history. Perhaps the most widely known treatment of this notion is Freud’s analysis of the defense mechanism of projection.
¹
Freud, of course, was concerned with the special case of people detecting characteristics in others that, because of their threatening nature, they are
unaware
of possessing themselves. A man who has yet to come to grips with his dissatisfaction with his wife might see evidence of marital discord in numerous relationships. Since Freud’s time, however, there has also developed an extensive literature on the tendency of individuals to attribute to others characteristics that they know they themselves possess.
²
People who like loud music, fast cars, and late nights—and who are willing and able to say so—also tend to project their affinities onto others.

Most of the recent research on this topic has focused on what has come to be known as the “false consensus effect.”
³
The false consensus effect refers to the tendency for people’s own beliefs, values, and habits to bias their estimates of how widely such views and habits are shared by others. Francophiles think that more people are fans of French culture and cuisine than do Francophobes; drinkers believe that more people like to imbibe than do teetotalers. The most widely-cited demonstration of this phenomenon is one in which university students were asked, as part of an experiment, whether they would be willing to walk around campus wearing a large sandwich-board sign bearing the message “REPENT.” A substantial percentage agreed to wear the sign, and a substantial percentage refused. After agreeing or declining to wear the sign, the students were asked to estimate the percentage of their peers who would agree or decline. The students’ estimates were slanted in the direction of their own choices: Those who agreed to wear the sign thought that 60% would do so, whereas those who refused thought that only 27% would agree to wear it.
⁴

It is important to emphasize at the outset the
relative
nature of the false consensus effect. People do not always think that their own beliefs are shared by a majority of other people. Rather, the false consensus effect refers to a tendency for people’s estimates of the commonness of a given belief to be positively correlated with their own beliefs. Religious fundamentalists do not necessarily believe that most people have a similar orientation, although their estimates of the percentage of religious fundamentalists in the general population can be counted on to exceed similar estimates made by their more secular peers.

Most of the recent research on the false consensus effect has been devoted to understanding why people unknowingly exaggerate the extent to which others share their beliefs. The authors of the seminal paper on the subject had argued that there is probably no single cause. They claimed that the false consensus effect was most likely a multiply-determined phenomenon, and they described a number of specific mechanisms that might be responsible for it.
⁵
Subsequent research has largely confirmed their initial speculations by documenting the mediating role of a host of cognitive and motivational variables.

There is evidence, for example, that the false consensus effect is partly a motivational phenomenon that stems from our desire to maintain a positive assessment of our own judgment—a desire that is bolstered by thinking that our beliefs lie in the mainstream. Consistent with this idea, people have been shown to be particularly likely to exaggerate the amount of perceived social support for their beliefs when they have an emotional investment in the belief,
⁶
and when their sense of self-esteem has been threatened by a previous failure experience.
⁷
Also consistent with this explanation are results indicating that people are particularly likely to exaggerate the extent to which attractive, respected, and well-liked people have beliefs similar to their own.
⁸

Other explanations of the false consensus effect focus on the information to which we are generally exposed and the way we process that information. It is a fact of social life that we are selectively exposed to information that tends to support our beliefs.
⁹
Conservatives read conservative periodicals and thus receive support for a conservative political agenda; religious fundamentalists tend to read “creationist” literature rather than contemporary evolutionary biology and thus buttress their conviction that evolution is a mere theory, not a historical fact. Because we so often encounter arguments and evidence in support of our beliefs while generally staying clear of information that contradicts them, our beliefs appear to be more sensible and warranted—and therefore common—than they would if we were exposed to a less biased body of information. Furthermore, in addition to being exposed to a biased set of
arguments
relevant to a given belief, we are also exposed to a biased sample of
people
and their opinions. Liberals associate with fellow liberals; exercise enthusiasts affiliate with other athletes. Indeed, similarity of beliefs, values, and habits is one of the primary determinants of those with whom we associate. As a result, when trying to estimate the percentage of people who hold a particular belief, examples of people who believe as we do come to mind more readily than examples of people who believe differently. Our own beliefs thus appear to be quite common. The most direct evidence for the influence of this mechanism on the false consensus effect is the finding that people’s estimates of the prevalence of smoking are positively correlated with the number of people they know who smoke.
¹⁰

The false consensus effect is also partly a product of the type of causes people believe to be responsible for why they believe or act the way they do. When we think our beliefs or actions are the result of external elements like the situation or issues involved, we assume that those elements would have a similar influence on others and so we infer that other people would tend to think or act likewise. We believe, in other words, that what are powerful situational influences on our own behavior should govern the behavior of others as well. Alternatively, when our beliefs or actions seem to us to be more the product of personal dispositions or idiosyncratic past experiences that do not pertain to others, we have less reason to believe that others would think or act similarly. Overall, the false consensus effect should be quite pervasive because there is a well-documented tendency for people to be generally more inclined to explain their own behavior in terms of external, situational causes than in terms of internal, personal dispositions.
¹¹
*
Nevertheless, people do not
always
cite situational elements as the cause of their actions, and so the false consensus effect should vary in strength with the extent to which people do in fact identify such external factors as the cause of their beliefs, attitudes, or behavior.

Research that my colleagues Dennis and Susan Jennings and I conducted a number of years ago supports this analysis.
*
In one experiment, individuals who were induced to explain their preferences in terms of personal causes exhibited less of a false consensus effect than those who were led to explain them in terms of external factors. In another study, the size of the false consensus effect for a variety of different issues was related to the extent to which those issues typically prompt situational explanations for a person’s responses. Items that tend to elicit external reasons for one’s choice (“Would you rather buy stock in Exxon or General Motors?”) yielded larger false consensus effects than items that elicit more personal reasons (“Would you rather name your son Jacob or Ian?”).

Finally, there is yet another determinant of the false consensus effect, one that may be the most interesting and may have the most far-reaching consequences. This mechanism involves the resolution of ambiguities inherent in most issues, choices, or situations. Before we can decide what we think about some issue, we must first arrive at an exact definition or specification of its meaning. When deciding whether we prefer French or Italian films, for example, we must first determine exactly what the terms French and Italian films mean. The precise way that we interpret these two categories will not only decide our own preference, but will exert a parallel influence on our estimates of the preferences of others. If we think of
The Bicycle Thief
and
La Strada
when we think of Italian films, for instance, we may be more likely to prefer Italian films ourselves and to estimate that a larger percentage of the general population would have the same preference than if we construe Italian films to mean spaghetti Westerns.

Note that this interpretation of the false consensus effect rests on two assumptions: a) different people construe the same choices quite differently, and b) people generally fail to recognize this fact and thus fail to make adequate allowance for it when making consensus estimates. It seems that the process of interpretation is so reflexive and immediate that we often overlook it. This, combined with the widespread assumption that there is but one objective reality, is what may lead people to overlook the possibility that others may be responding to a very different situation.