This book presents my current thinking about what is important in the psychology of thinking and decision making and how it relates to questions of public interest. I try to provide sufficient references so that an academic reader could track down the source of these ideas. The ideas here are a continuation of those presented in an article I wrote for Behavioral and Brain Sciences in 1994, titled ``Nonconsequentialist Decisions.''
I would like this to be read by everyone concerned with public affairs or the psychology of thinking and decision making. That is, of course, too much to expect.
In attempting to reach a somewhat wider audience than usual for me, I have tried to simplify the presentation by eliminating some of the usual academic qualifications, such as ``It could be argued that X'' when I really mean to say that I think X is true. I have also put references in endnotes so as not to clutter the text.
I am grateful for specific comments and general advice in the early stages of this project from Paul Rozin, Martin Seligman, and Karen Steinberg. Helpful comments on specific chapters came from Willett Kempton, Howard Kunreuther, Howard Margolis, Jay Schulkin, Karen Steinberg, and Peter Ubel. Judy Baron, David Baron, Deborah Frisch, Joshua Greene, Robert Jervis, and Joan Bossert and Nancy Hoagland (at Oxford University Press) provided helpful comments on the whole book. Mark Spranca convinced me of the importance of the intuition of naturalism, and Howard Margolis strengthened my belief that intuitions can affect public outcomes. Before and during the writing of this book, my research has been supported by the National Science Foundation. David Baron helped with typesetting, which was done with LATEX2e in Adobe Palatino font.
One way to make decisions is to weigh our options on the basis of their expected effects. We would favor options that we expected to have better outcomes. We do not always make decisions this way. Instead, we apply various intuitive rules to our decisions, rules that do not refer to outcomes alone. We also apply these rules when we evaluate the decisions of others, including government officials.
For example, we often consider the harm caused by our actions to be more serious and more to be avoided than harm caused by our omissions. We avoid positive options that have negative side effects, even if the positives outweigh the negatives. The resulting bias against helpful action is often reinforced by similar biases in favor of the status quo, of what is natural, or of what others have autonomously chosen. When we think about decisions affecting large groups of people, we tend to favor groups we belong to - such as nations or races - at the expense of outsiders. We judge fairness within these groups, attending less to the larger groups that contain them. Our judgments of fairness and justice are based on a kind of balancing - an eye for an eye - even when we could foresee that this attitude would make things worse on the whole.
The point of this book is that we should not be surprised when these intuitions - played out in the public sphere through the actions of individuals and government officials alike - lead to outcomes that are worse than the best we could have, often substantially worse. After all, these intuitions are not based on the principle of achieving the best. Sometimes they may lead to the best despite their apparent design, but this is not typical. If we want a better world, one relatively inexpensive way to get it is to improve the way we make decisions. We need to think more about their effects, and less about the rules that might guide them.
Consider again the intuitive bias against causing harm through action, as opposed to omission. As a result of this intuition, some people avoid taking protective measures that might cause harm, even though the same measures are more likely to prevent harm. When a vaccine - such as the DPT vaccine (diphtheria, pertussis, and tetanus) - causes rare but serious side effects or death, people resist using it because they want to avoid these effects, even though the vaccine can prevent a disease that is more likely and equally serious and deadly. Government officials resist requiring the vaccine. It is not that the government officials are wise but capitulate to public demands. They make the same intuitive judgment.
The intuition that distinguishes acts and omissions is a principle that people apply to their decisions. It has, through its effect on many people, brought about outcomes that nobody wanted, in particular, epidemics and deaths from preventable diseases. This is the pattern that I explore in this book. People follow intuitive principles of decision making that are not designed to produce the best consequences in all cases. Predictably, these principles sometimes lead to unhappy results that could have been avoided if people had focused more on how to produce the best results. So our intuitive principles have a cost. I focus on cases in which the cost is borne by many people - that is, public outcomes. Some of these people may not even agree with the principle that made things worse for them. The question I raise is why we should keep paying that cost.
People have an intuitive moral rule ``Do no harm'' or, more specifically, ``Do no harm through action.'' In some cases, this rule is sensible. If Tom pushes Dick into the lake and Harry fails to rescue him, we punish Tom more than we punish Harry. Harry, after all, might have thought that someone else would rescue Dick or that he might be sued if he tried and failed. In other cases, like vaccination, this rule is potentially harmful. It leads us to neglect things we could easily prevent, like disease and death. This has happened in whooping cough epidemics in England and Japan. If parents or pediatricians had questioned the do-no-harm intuition at the outset - asking whether they should just try to minimize children's risk - then many of these deaths might have been avoided.
Some parents resist the DPT vaccine for their children even when they know that the total risk is lower with the vaccine than without it. They do not want to see themselves as the cause of harm to their children; better that the harm should come from ``nature,'' even if they could have prevented it. Once people have made this judgment, however, they adjust their other beliefs to conform to it. They come to believe that vaccinating really has more risk than not vaccinating. Although the original intuition was not based on consequences, people convince themselves that following it will always lead to the best results anyway. My main point here is that this is not true. Intuitive principles that are not based on consequences do not always produce the best consequences, and we should not be surprised by this.
The do-no-harm intuition also affects the decisions of judges and juries when people sue the makers of the vaccines for brain damage and other long-term effects. These consequences are awful, but so are the consequences of not making any vaccine at all. Yet drug companies do not get sued for the injuries caused by their failure to make a product. So pharmaceutical companies take these lawsuits into account when deciding to invest research and development resources into more, possibly risky vaccines versus yet another drug to lower cholesterol. The same intuitive rule makes people resist government policies that help many people while hurting a few.
Notice that bad results come from well-intentioned intuitions about what is right. These intuitions play some role in a great variety of human tragedies. Wars - both military and trade - result because citizens support the belligerent stance of their government against the immoral behavior of another nation. People oppose regulations or agreements that could protect the environment because these regulations seem to violate some principle, such as autonomy or the right to self-determination. As a result, they get results that they do not want.
This is a kind of paradox because many of the opinions in question are moral beliefs and judgments. The capacity to form and espouse moral beliefs is one of the wonderful features of humanity. People do not feel they are doing wrong when they act on these beliefs, but these beliefs repeatedly cause trouble.
The basic problem is that many of our beliefs, like the distinction we make between acts and omissions, do not concern consequences or results. We could try to follow just those principles that bring about the best results, but our principles are not designed this way. So we are constantly facing conflicts between the intuitive principles that we all follow and the results we all want.
The problem is most serious when we can follow the principle without much risk of facing the consequences ourselves. In vaccination, for example, the risk of the disease and the risk of a serious reaction to the vaccine are both very low. If we ourselves were faced with a choice between certain death from a disease (which we would get if we failed to vaccinate) or a fifty-fifty chance of death from the side effects of a vaccine, we would probably think harder about the consequences, and we would not worry so much about the intuitive distinction between acting (vaccinating) and not acting.
One area where the risks of facing the consequences seem low is our political behavior. This includes voting, speaking to each other, making contributions and working for causes, and other things we do to try to influence public policy. It is difficult to think about the consequences of this behavior because we see it two different ways. In one view, because so many other people affect the outcome, the contribution of each person's action to the overall outcome is tiny. Thinking about public issues, and acting on these thoughts, is ``cheap.'' You don't have to pay for it by accepting the consequences of your mistakes. Even if your candidate for office turns out to be a disaster, you can console yourself by saying that your vote wouldn't have mattered. Even government officials and elected legislators may feel that their main task is to express their moral intuitions, for their vote is just one among many. Thus, the political sphere is one where intuitions tend to have free play.
In the other view, public action of many people - whether through voting, speaking, or doing a government job - affects so many people that the effect of everyone's behavior together is enormous. If a billion people together, through their political action, affect the outcome of a billion people (perhaps the same billion, perhaps not), then, on average, the effect of each person's action is just as noticeable as if that person were making a personal decision. Political action no longer seems so cheap when we take our effect on others into account. It may seem that voting is an exception here because each vote is rarely decisive; elections are hardly ever so close. Yet the margin of a vote is often important, aside from the outcome. The margin tells elected officials about the extent of their mandate and the actions that will make them popular, and it informs them and other candidates about prospects for the next election. Elected officials in modern democracies are, in general, highly sensitive to public opinion.
The same arguments apply to other expressions of political opinions and moral views, such as writing letters to representatives and newspapers, posting messages to news groups on the Internet, and just talking to people. These things ultimately have consequences. They are part of the total body of opinion that guides the behavior of nations and other institutions. In sum, we cannot ignore the potential consequences of our political action so long as we care about our effects on others. Yet the first view, that our voice has little effect, often encourages us to express our intuitions without even thinking about the consequences.
Some intuitive beliefs are held blindly. People do not know what gives them their authority, and often nothing does. This ignorance does nothing to weaken people's commitment. The abortion debate in the United States is a good example. One side thinks of the fetus as a human being with a right to life and protection of the law. The other side thinks that prohibition of abortion infringes on the rights of women to control their bodies. Extremists on both sides do not think that their views are amenable to argument or reason. Commitment flows from the strength of feeling, from a raw perception of rightness. People even pride themselves on the strength of their ability to resist reasoned arguments from the other side. Debates take the form of repeated assertions. Each side tries to wear down the other rather than to persuade it.
Part of the problem is that one particular intuition makes us believe more strongly in the others. This is the intuition, discussed later in more detail, that what is natural is good. We tend to see our intuitions as the product of some natural force that, in some sense, understands more than we do. It has a kind of authority, like the authority that religious leaders sometimes have, that allows it to make pronouncements to us, which we then accept without knowing fully the reasons for them, trusting that the reasons are there. I shall argue, however, that many of these intuitions arise in a much simpler way. They are the application of principles that are often consistent with bringing about good consequences but that are applied in cases where they do not do this. They are, in the language of psychology, overgeneralized.
My use of the term intuitive is meant to include both blind feelings and also more reflective beliefs. The term is meant only to capture the idea that the fundamental basis of these beliefs or principles is that they appeal to some judgment other than consequences.
Intuitive principles are surely not the only cause of human misfortune, even if we limit ourselves to human behavior as a cause. Bad events happen sometimes when individuals simply pursue their self-interest rationally. Financial markets crash when thousands of investors all try to get their money out of a falling market: the market crashes because everyone wants to sell and few want to buy. Other bad events result from the violation of moral standards that limit the pursuit of self-interest at the expense of others' interests. Some people seem to have very weak standards of morality to begin with, so they are easily swayed toward immoral behavior by the example of others or by a bit of benefit they might obtain. Some people knowingly violate their own standards of morality. Perhaps violent criminals do this, or soldiers who rape or torture their prisoners of war. Sometimes these violations result from social pressure, which itself results from other factors such as weak standards.
I certainly do not want to deny these other behavioral causes of harm. But they are the usual suspects when we talk about the human causes of human misfortune: self-interest, weakness of will, absence of self-restraint, lack of principles, and social pressure. We know about them already, and we have been trying to control them for centuries. Perhaps by working a little harder on a somewhat neglected cause of trouble - our intuitions - we can gain a kind of leverage over the human condition. Even if the effects of our intuitions are small in the grand scheme of things, we might get a handle on them more easily than we can on other causes of harm. And even a small benefit can help a lot when its effects are accumulated over great masses of people.
Moreover, our intuitions affect our ability to deal with the other causes of trouble. If, for example, we believe in the morality of retribution and in group responsibility for the acts of individuals, we may support excessive retaliation for clearly immoral acts against groups whose members were at fault, punishing the innocent along with the guilty, even though we know that such excesses will only lead to a cycle of escalating violence, as we have seen in the Middle East, the Balkans, Northern Ireland, Eastern and Central Africa, and India in recent years. Our beliefs in retribution and group responsibility are not the basic problem. But they exacerbate the original problem, making it worse than it would be if we took other goals into account, such as the goal of making a peace agreement.
In sum, intuitions may have only small effects on big outcomes, but they may also be more controllable than some of the other forces, especially because they have not been seen before as a source of trouble. Reducing the negative effects of intuitions might thus be a cost-effective way of improving the human condition. Such improvement gives us a kind of leverage. Although the effect is small, it is broad.
Harmful intuitions show up even in situations where one would think that self-interest was paramount. Consider the decline of fisheries in the Atlantic Ocean off the coasts of New England and Canada. Between 1963 and 1993, the number of flounder, haddock, and cod declined by more than 90%, mainly as a result of overfishing. It took almost 10 years for effective regulations to be imposed. Now the regulations must be so drastic that some fish cannot be caught at all until the stocks come back. Thousands of people are out of work.
Part of the problem was that each person pursued his or her economic self-interest. It was not in anyone's interest to cut back fishing, regardless of what others were doing. But democratic mechanisms were in place to impose limitations on fishing. Every time some regulation was proposed, many people thought that it was wrong, and they opposed it. The personal cost of supporting the regulation - or of assenting silently - would have been low, and almost all of the regulations would have been better for each person in the long run than no regulation at all. So we cannot explain this opposition in terms of the same simple self-interest that makes fishermen unwilling to cut back spontaneously. As I will argue in the next chapter, the fishermen opposed the regulations on the basis of their intuitions concerning personal autonomy and fairness, abetted by wishful thinking: that the decline in the fish population resulted from everything else aside from overfishing.
This pattern is repeated in a variety of social misfortunes examined in this book. People are gripped by some idea, a principle that has much to be said for it but that ignores some equally valid principle on the other side. On the basis of such principles, people commit themselves to one side of a debate. They want their side to be right, so they engage in wishful thinking to convince themselves that both the facts and the arguments support their view. They make up additional arguments on their side and fail to try to think of arguments on the other side. Some of the arguments they make are ones they would recognize as weak if they were not already committed to their positions.
Intuitions can be useful when we correctly perceive them as part of the story rather than as the whole story. They become dangerous when we think in a way that protects whichever idea grips us first. How can we keep these intuitions in check? For a start, it may help to be actively open-minded, to put our initial view to the test by seeking evidence against it as well as evidence in its favor. It may also help to ask whether there are possible answers other than our own, and whether we are ignoring certain goals or values - even the values of others - that would make some alternative answer seem more reasonable. When we find alternatives or counterevidence, we must weigh it fairly. Of course, there may sometimes be no ``other side,'' or it may be so evil or foolish that we can dismiss it quickly. But if we are not open to it, we will never know. When large groups of people fail to think in a way that is actively open-minded, social discourse breaks down.
Actively open-minded thinking must often be quantitative. When good arguments are found on both sides of an issue, we must often find a way to compare the arguments quantitatively. In the DPT vaccination example that began this chapter, an argument for the vaccine is that it prevents disease and resulting death. An argument against is that it causes side effects, which may also result in death. A simple quantitative comparison is to count the resulting deaths from vaccinating or from not vaccinating. A more complex quantitative comparison would take into account the severity and frequency of the symptoms and side effects other than death. This could be done informally, or formally by assigning numbers to everything. The important point is that we must be willing to think of decision making as a kind of balancing, with each argument put onto the scales and weighed.
I have suggested that intuitions can get us into trouble if we follow them blindly. So let us look at the most common ones that do this, intuitions that most of us apply frequently (and usually appropriately). The boundaries of each are fuzzy, some can be subdivided further, and some important ones are doubtless missing. But some list is probably better than no list. None of the intuitions in question is crazy or evil. That is the point. We all hold these, and most of the time they are reasonable or at least harmless compared to other ways of making decisions.
All of these intuitions are reasonable rules of thumb. Using them often leads to the best outcome, for good reason. For example, we should usually favor the status quo because we are not so good at anticipating the effects of changing it. Likewise, people usually know what's good for them better than other people do, so autonomy is a good idea, other things being equal. And evolution has created a kind of purposive design for living systems, one that is best left undisturbed - again, other things being equal.
The intuitions cause trouble because we conduct our thinking as if they were more than this, in several ways.
1. The intuitions become absolutes. Instead of thinking of these principles as rules of thumb, we elevate them to the level of absolute constraints on action. The do-no-harm principle, for example, becomes an absolute prohibition on hurting some people in order to help others, even when the help is great and the hurt is small. Thus, a trade agreement among nations, which will cause some workers to lose their jobs, may be rejected because of this, despite preventing many other people from losing jobs, as well as making more goods available at lower prices.
When absolutes conflict, compromise becomes more difficult. For example, a trade agreement involves economic benefits and increased autonomy on the one hand, but on the other, greater difficulty in enforcing environmental regulations (which are often challenged as restraints on trade). People who care about both the economy and the environment will be sensitive to the magnitude of each effect. Environmentalists who also care about economics might decide to accept the risks of a trade agreement because the environmental costs are small relative to the economic benefits. Other international environmental agreements might have great environmental benefits and small economic costs, and it would be better to work on getting these adopted rather than on opposing the trade agreement. When principles are held absolutely, compromise and logrolling are difficult. The end result is that, instead of either agreement, we end up with neither, and both economics and the environment may suffer.
The intuitive principles that people follow - autonomy, not going against nature, nationalism, preserving the status quo, etc. - are often good rules of thumb. In general, it is better to honor them than not to honor them. People usually know what is good for them, so autonomy leads to better decisions. Tampering with nature is risky. Citizens know more about what is needed in their own nation than in other nations. But these are rough guidelines - rules of thumb that are not always true. They become most problematic when people elevate these useful rules of thumb to inviolable principles, neglecting the big picture in favor of a small piece of it.
2. Intuitions define aspiration levels. Intuitive principles almost always define acceptable levels of some good. Once an acceptable level is defined, the principle obliges us not to fall below that level, but does not oblige us to rise above it. Thus, for example, we are obliged not to harm people through our actions, but we are not obliged to help people through our actions. The level of aspiration here is whatever results from doing nothing.
This sort of intuition is very strong. When I play tennis, I often open a can of tennis balls on the court. I feel a strong obligation to throw away the metal top to the can I just opened, rather than leaving it to litter the court. So I do this. But I often leave behind several tops left by others, which I could easily pick up and throw away. My intuitive sense says that I am obliged not to make the situation worse, but I am not obliged to improve it.
Such principles are convenient because they limit our obligations in our daily lives. If I felt just as obliged toward every lid, I would have no clear rule for stopping. I would have to judge, each time, whether the effort of picking up another lid was worthwhile. But the same intuition can be applied in matters of policy. If we are talking about governments regulating pollution instead of tennis players picking up their lids, we should ask how we can get the most pollution reduction for a given expenditure, and it may turn out that it is better to make companies clean up someone else's pollution rather than their own, regardless of our intuition to the contrary.
The status quo often defines an aspiration level. We are more upset about losing what we have than about failing to get what we do not have. Imagine the reaction if someone in the U.S. government proposed a new subsidy for tobacco growers. Yet the opposition to the current subsidies is so muted that nobody thinks they are threatened. Likewise, when a new law helps many people but hurts a few, relative to the status quo, we are reluctant to support the law because we take the harm more seriously than the gain. But if the law were already in effect, we would not want to repeal it because those who benefit from it would then be hurt.
In the vaccination case described earlier, the aspiration level is inaction, the result of doing nothing. (This is not the status quo because nobody is sick yet.) In other cases, the aspiration level is defined by some principle. One principle we shall see repeatedly is that of autonomy. Thus, interfering with autonomy is considered a great loss, although creating additional autonomy where it does not exist is seen as less important. Likewise, destruction of what is natural is particularly harmful, more so than failing to return something to its natural state. Finally, a distribution seen as fair can define an aspiration level. If it seems fair for two boys to get 10 peanuts each, and if one gets 12 and the other 8, then the 8 will be seen as a loss of 2, which will seem more serious than the gain of 2 for the other.
Once an aspiration level is defined, losses relative to it are taken more seriously than gains. This is called ``loss aversion.''
3. Wishful thinking. People tend to believe what they want to believe, which is often determined by their immediate self-interest. Credit-card interest rates are extremely high compared to other rates for borrowing, and banks try hard to sell credit cards because this interest is so profitable for the banks - several times the profit they make on other activities. According to economic theory, the interest rates and the profit ought to come down because of competition. It seems that this does not happen, in part because card users tend to think that they will not borrow on their cards, so the high rates are irrelevant to them. Card holders do not even admit to themselves the amount of borrowing they are doing: they drastically underestimate the amount that they already owe.
Likewise, most drivers believe that they are safer and more skillful than average, and most people believe that they are more likely than average to live past 80. There are limits on this. We can't convince ourselves of anything. But we can trick ourselves into believing what we want by paying attention to the evidence that supports a belief, ignoring the evidence against it and not worrying too much about what our beliefs might be if we were more objective.
Once we are gripped by an intuitive principle, wishful thinking sets in. As we shall see, when scientists told fishermen that they were overfishing and should cut back, the fishermen did not want to hear this particular news, so they looked for reasons why the scientists were wrong.
4. My-side bias and overconfident belief. Although people usually employ wishful thinking so that they believe good news and disbelieve bad news, the opposite can happen. Chicken Little misinterpreted the evidence to support his pessimistic view that the sky was falling. People suffering from clinical depression often try to convince themselves that they are to blame for some bad chance event, and sufferers from anxiety exaggerate their fears. It is thus sometimes more sensible to speak of ``my-side bias'' rather than wishful thinking. Even Chicken Little wanted his side to be correct, despite, we assume, wishing that it were incorrect. Once people have committed themselves to a belief, they tend to look for evidence in its favor and reject evidence against it, even if the belief makes them unhappy. This is called ``my-side bias'' because it favors the thinker's side of the argument.
My-side bias results in overconfidence, in unwillingness to listen and to compromise, in breakdowns in cooperation. The overconfidence in the rightness of the principles in question is, of course, one reason that people treat them as absolutes that cannot be compromised.
The mechanism of my-side bias is often subtle, involving several people. An interesting case concerns the question of whether global warming is resulting from human activity, particularly the burning of coal, oil, and gasoline, which produces carbon dioxide, which traps the sun's heat in the atmosphere. The mechanism of warming has been understood for some time, and computer models predict that considerable warming will occur over the next few decades. Although temperatures have increased slightly over the past century, most scientists who worry about global warming are more impressed by the computer models than by the warming so far, which may be the result of random fluctuation (although the most recent evidence says that even the observed trend is real). In 1990, an article in Science demonstrated that precise measurements of atmospheric temperature could be made from satellites by measuring microwaves. Such measurements had been made from 1979 through 1988, and the article remarked incidentally that during this 10-year period no overall warming was observed. The authors were concerned with the accuracy of their measurements. In press accounts, they acknowledged explicitly what they should have said in the article, that 10 years is simply too short a time to detect a trend, so their finding ``does not prove that there is not a global warming.'' Still, despite this clarification in the text, the headlines emphasized the lack of trend: the Washington Post account of March 30, 1990, was titled ``NASA satellites find no sign of `greenhouse' warming.'' An anti-environmentalist congressional staff member, however, sent out a letter to members of the U.S. House reprinting a key graph from the article and saying, ``The NASA study offers strong evidence that there is no `greenhouse effect.' '' It is easy to understand this event without assuming any deliberate attempt to deceive. The staff member involved saw the headline and the graph and, given that these supported his view, saw no need to look very hard for evidence on the other side, such as a disclaimer by one of the authors of the original study. Moreover, the staffer's confidence in his conclusion took no account of the fact that he had failed to look for such a disclaimer. He was as confident as if he had looked and failed to find it.
5. Belief overkill. Many controversial issues are controversial because there are good arguments on both sides. A rational decision would involve balancing the arguments in a quantitative way, a way that takes into account their strengths or the magnitudes and probabilities of the possible results. But people find ways to avoid this balancing. Through wishful thinking, they convince themselves that all the good arguments are on one side. Robert Jervis provides many examples of this kind of overkill in judgments about foreign policy. In discussions of a ban on testing nuclear weapons, ``People who favored a nuclear test-ban believed that testing created a serious medical danger, would not lead to major weapons improvements, and was a source of international tension. Those who opposed the treaty usually took the opposite position on all three issues. Yet neither logic nor experience indicates that there should be any such relationship. The health risks of testing are in no way connected with the military advantages, and a priori we should not expect any correlation between people's views on these questions.''
Attitudes about capital punishment provide a good example of overkill. It is possible in principle to believe that capital punishment is morally wrong yet effective as a deterrent against serious crimes, or morally acceptable yet ineffective. Yet almost nobody holds these combinations of belief. Those who find it morally wrong also think it is ineffective, and vice versa.
Here is an example of the process at work, from a subject in a
study of reasoning, asked his opinion about animal
experimentation: ``We actually have no right to do such things.
It's not even necessary. If it was necessary, maybe there would
be a reason for it, but, there's no need for it, I don't think.
We're sort of guardians here.'' The subject's intuition was that
animals have a right to our protection. He could believe
this
and still also think that we could gain some benefit from experimenting on them, a benefit that we ought to forgo. But, instead, he convinces himself that no benefit exists.
The idea that intuitions cause trouble is part of a relatively recent tradition in psychology, the study of heuristics and biases. In the past three decades, psychologists have discovered dozens of cases in which people use principles of thinking in ways that lead to errors. The principles themselves, heuristics, are typically good rules to follow in certain situations, but people use them where they are not so good. As a result, people display biases away from the best judgment. Biases are not just errors. Errors could go in either direction, but biases are in one direction.
The status-quo principle is an example. In this case, the heuristic principle is ``stick with what you have.'' The bias is that people tend to change less often than they should if they want to achieve their goals as much as possible. For example, in several experiments, half of the college students in a room were given a large candy bar and half were given a college mug. These objects were put on their desks, but the students could not touch them. Then the students were asked if they would trade the object they were given for the other one: a mug for a candy bar or vice versa. If the students had been asked in advance, before they were given anything, we would expect half of them to prefer the object they did not get. After all, the objects were assigned at random. Typically, in the experiments, only about a quarter are willing to trade. Here is a clear case where no arguments can be made for the status quo - the object each student has - yet people maintain it anyway.
One explanation of this effect is that the status quo defines a psychological aspiration level. If you have a mug - even if you would otherwise slightly prefer a candy bar - a trade will involve losing the mug in order to gain the candy bar. A basic principle, the principle of loss aversion , says that losses loom larger than gains. So the loss of a mug seems too great to justify the gain of the candy, even though the gain of the candy would be preferred to the gain of the mug.
This status-quo effect, and other effects, may also result from the misapplication of principles that are useful in many situations. In general, other things being equal, it is in fact better to keep the status quo, for many reasons: in organizations, constant changes make it difficult for people to learn how things work; it is good to be predictable; and the status quo was often chosen for good reasons that are now unknown. But people honor this principle even when all these arguments are carefully removed.
The reason for this kind of misapplication of principles is not fully understood, but it seems to be a general feature of learning. In arithmetic, for example, many children faced for the first time with ``17 - 9'' answer ``12.'' They apply a principle that was useful up to that point: when subtracting within a column, subtract the smaller number (7) from the larger (9). If they understood the purpose of this rule, however, they would realize that it does not achieve its purpose when the smaller number is ``on top.'' Perhaps the ultimate reason for this misapplication of principles is that we do not always fully understand the purposes of the principles we use.
The term heuristics is understood as equivalent to rule of thumb. I prefer the term intuitions here because I argue that the principles in question are not always used as rules of thumb. They are elevated to a higher status because they are perceived as being absolutely correct. When decisions affect many people, these intuitions take on moral force. We feel that they are more than just personal rules; rather, they should be accepted by everyone. When others go against them, that is a reason for righteous anger.
Effects such as the status-quo effect are often called biases because they move people away, in a systematic direction, from what is optimal. The interesting finding of the mug experiment was not that the students made random errors, sometimes one way and sometimes the other. (The experiment would not have revealed such errors in any case.) It was rather that responses were systematically biased toward the status quo more than they would be if the students chose what they would have preferred before being given anything. Thus, in general, biases often result from the use of heuristics.
I am going one step beyond most past psychology by extending this heuristics-and-biases approach to moral intuitions, as well as nonmoral principles of decision making. This extension has a problem. In the study of personal decision making, we can usually specify a ``right answer'' and then show that subjects aren't giving it. In the mug experiment, for example, we cannot give the right answer for each subject, but we can say that, if subjects had responded in terms of their real preference, half of them would have switched. In matters of morality, however, it is more difficult to specify a right answer.
I shall assume here that the right answer is the one that can be expected to yield the best consequences. More generally, the best principles to follow are those that lead to the best consequences. This claim is called consequentialism or (in one form) utilitarianism. So people are making errors, in a sense, if they do not follow these consequentialist principles. Although I cannot fully defend this assumption here, I adopt it because my main purpose is to show that intuitions often make consequences worse and to ask why we should make this sacrifice to the intuitions that do this.
The vaccination case provides an example. The lower the death rate, the better the result; this is the main consequence of the decision. So it is better to vaccinate as long as fewer deaths will result from it, even if the vaccine itself will directly cause some deaths. The do-no-harm intuition argues against vaccinating, so it leads to worse consequences. This does not imply that the people who follow it have bad intentions. Indeed, the interesting thing is that they make their choice exactly because they think it is morally right. But it turns out to be wrong according to the consequentialist standard. Thus, intuitions that are not justified in terms of consequences often result in consequences that we do not like. We pay a price for following our intuitions, and sometimes it is a heavy one. In view of this situation, we might want to reexamine the reasons we have for our intuitive judgments.
If we correctly follow the principles that lead to the best outcome, we will achieve the best outcome possible. If anyone argues for an alternative principle - such as ``respecting the right to autonomy even when it allows people to harm themselves'' - that person must explain how the harm is justified. This explanation must be addressed to those who do not accept the intuitive principle, so it cannot appeal to the principle itself. The whole issue is whether the principle can be justified. We do have reasons to oppose the principle, namely, that it produces results that we do not like. For example, if I put the long-run health of fisheries and those who work them ahead of the autonomy of fishing people, then, if you want to prove me wrong you must explain to me why we should let the fisheries die for the sake of autonomy, even though all the other costs of honoring this principle exceed the benefits, in terms of what matters to people.
One explanation is that, in the long run, it is better to respect autonomy because, that way, people learn to make decisions for themselves, and they make better decisions. Even if, along the way, some harm results, this harm is outweighed by better consequences. Notice, however, that this argument is made completely in terms of consequences. It is an argument within utilitarianism, not an argument against it, whether it is correct or not.
Utilitarianism is based on the idea that decisions in general should produce the best consequences. It extends this idea from ordinary decisions that affect mainly one person to decisions that affect many people. This principle, applied to ordinary decisions, explains why the status-quo effect is an error. If you judge that the candy bar is a better outcome for you than the mug, and you still keep the mug, then you are producing a worse result. You are adhering to a principle that, in this case, does more harm than good.
Similar things happen in the domain of moral judgment. One example is honoring autonomy even when people use it to hurt themselves. Another example concerns punishment. We have a moral principle that harm doing should be punished. This rule usually yields good consequences because punishment deters future harm doing. But when we punish - through lawsuits - the makers of beneficial and well-made vaccines because of harmful but rare side effects, we discourage companies from making such vaccines, so we do more harm than good. Many people do not think about justifying punishment in terms of its consequences, so they are willing to punish a company even when they believe that no good will result. This is even true of judges. The utilitarian principle here is that two wrongs do not make a right, unless the second wrong prevents another, greater, wrong.
The nonutilitarian moralist will argue that these bad outcomes are simply the price we pay for morality. If a regulation infringes upon the autonomy of fishermen, for example, then it should not be put into effect, even if its absence leads to a worse outcome for the fishermen themselves. If a vaccine harms some, we should not use it, even if many more will suffer from our omission. But why would we want moral principles that lead, over and over, to unwelcome outcomes? Where do these principles get their authority to override our interests? Of course, many have tried to answer these questions, and it would take at least another book to deal with all their arguments. I raise these questions here not to settle the issue but to hint at why I think the task for the other side is formidable.
To apply utilitarianism, we need a thorough accounting of consequences. We need to consider emotional effects, such as the bitterness that results from perceived unfairness or from the regret that comes from harming someone through an action. We need to consider the effects of our decisions on other decisions taken later: when people break a rule for the greater good, they make it more likely that they and others will break the same rule in the future, for what seems at the time to be the greater good but isn't. Sometimes it is best just to follow the rules, despite appearances to the contrary: don't lie under oath, don't commit adultery, and so on. But we can understand that we do this because following the rules does serve the greater good, not because morality is separate from consequences. When we make decisions quickly, when we have a strong self-interest, when our information is compelling but unreliable, and perhaps in other cases, we must take into account the possibility of error as part of our judgment. We must override the judgment we would make without taking error into account, exactly because that judgment is likely to lead us away from the best consequences.
Most moral systems make some room for consequences. The arguments here are not totally alien to anyone. Even those who admit some other principles aside from attention to consequences might come to feel, on the basis of looking at the effects of moral intuitions, that more attention to consequences is warranted.
Many of our problems and their solutions are rooted in our psychological makeup. Others have made this kind of argument before, but they have pointed to different aspects of human nature. The problem is not a matter of evil motivation, false consciousness, sexual repression, neurosis, or psychosis. Rather, our problems come - paradoxically - from some of the best aspects of our nature: our desire to be moral and virtuous. We are well intentioned, but our intentions go awry because we do not always think well. This is a hopeful view because the solution to our problems can be found, in part, in the standards of thinking that we adopt for public discourse.
The intuitive principles that cause trouble are not just the results of greed or impulsiveness. The trouble results from people acting on principles that they themselves believe to be right, if only because people have convinced themselves of the principles' rightness. The problem of concern here is not evil, but if anything, an overconfident morality that results from an excess of self-deception. Most people take stands on public issues on the basis of gut feelings or intuitions about what is right. These feelings often capture part of the issue - like the blind men and the elephant in the fable - but they miss the whole.
Of course, people sometimes do act out of bad intentions, knowing that what they do is morally wrong by their own standards. This is surely true of many criminals, especially those who attack and rob strangers. Even bad public outcomes, like disease epidemics and the decline of fisheries, are often blamed on some sort of malice on someone's part - greedy fishermen, corrupt officials or legislators, and so on - as if the intentions had to fit the outcomes. This is almost the standard story in news media and private discussions about such outcomes. And perhaps it is sometimes true. Yet many commentators neglect the possibility that these outcomes arise from actions fully congruent with the intuitive beliefs of the actors, and that the intuitions in question, though held through no malice, are misapplied.
The intuitions that drive us are often related to goals that we do not think about. For example, autonomy is a good principle because it usually helps to achieve better consequences. People usually know what is good for them better than other people do, and social support for autonomy helps people learn to make better decisions. So we can justify autonomy in terms of our goal that people should make good decisions. But when people think about autonomy as a value in its own right, rather than as a means to another end, they cannot easily make tradeoffs between autonomy and other values.
To see the whole, we must think more about consequences for everyone. The autonomy of fishing people is a good thing, other things being equal; it allows them to plan their work efficiently. But overfishing destroys the fish. Each side - fishermen and conservationists - transforms one of these consequences into a near absolute, and a compromise that respects both values becomes difficult.
Thus, our well-meant and deeply felt intuitions about what is right often prevent us from achieving the results we want. We should not banish these intuitions, for they are often important reminders of things we need to consider. But we should relegate them to a secondary role. Primarily, we should base decisions that affect the common good on our understanding of consequences, results, effects. This simplifies decisions, in a way. We usually ask about consequences and other principles. I am suggesting that we can omit the second step, although we might have to be more careful in how we think about the first.
The following chapters examine a number of issues in which intuitions lead to bad effects in the public domain. They consist mainly of a series of case studies, organized into chapters according to the type of intuitions involved in people's arguments. Each chapter has a substantive theme as well. For example, Chapter 2 is about social traps, such as the fisheries problem just discussed. That problem serves as the major case study, but other, related problems are discussed, too. A final chapter will include recommendations for action.
In most cases, the biases discussed are mainly on one side of a controversial issue. The main argument does not depend on which side is right. The examples I have chosen are just examples. They are not the entire set of possible examples. The reader should not assume that the presence of bias makes the side in question wrong. People can be right for the wrong reasons. But it is important to know what the right reasons are, for they affect future decisions. The point here is to get clear on the distinction between good and bad reasons for our attitudes. If we get rid of all the bad reasons, we will all be better off.
It may seem that a change in our way of thinking and talking will not have much effect. After all, the future of humanity depends mostly on economics, scientific progress, or the limits of the earth, not on the quality of our thinking. But small changes can help a lot. All of our major problems can be seen as races between constructive and destructive forces. Whether population growth outstrips food production, for example, is a matter of a fraction of a percent in the growth rate. A small change in an attitude toward birth control on the part of a mere billion people can make a big difference. Likewise, a small increase in world trade could allow some poor nations to change direction, if ever so slowly at first, from deepening poverty to economic growth. When we are dealing with long-term trends, small changes in rates are amplified after many years. When the rate of growth of world food production is about the same as population growth, a small change in one or the other can lead to famine or plenty. Our attitudes toward decision making act like a lever. With a place to put it, we can move the world.
Some warnings are worth heeding. In the 1980s, scientists such
asVaughn Anthony of the National Marine Fisheries Service in
Woods Hole, Massachusetts, began saying that the fish off the
coast of New England and Canada's maritime provinces were
disappearing because of overfishing.
Fishing has been a major industry in these areas for centuries,
and nothing like this had happened before. People did not act
fast enough. Now the numbers of cod, haddock, and flounder are
so low that it is not worth fishing for them.
Around the year 1500, the explorer John Cabot described theGrand Banks area as so ``swarming with fish that they could be taken not only with a net but in baskets let down with a stone.'' Fishing was a mainstay of the economy of the colonies, and then, after the Revolutionary War, of the United States and Canada. It stayed that way because the basic technology did not change that much.
The problems began in the 1960s, when giant ``factory trawlers'' from Japan, Poland, and other countries began fishing off the New England coast. Often these ships fished in teams, taking all the fish from one area and then moving elsewhere. In 1976, as part of a larger series of international changes, the U.S. government passed the Magnuson Fishery Conservation and Management Act, which extended coastal waters to 200 miles, evicting the foreigners. It also set up a system for regulating fisheries, which depended on eight regional management councils, consisting of representatives of fishers as well as others. The Commerce Department controlled the whole system. In 1977, to restore stocks of cod,
haddock, and yellowtail flounder, the New England Council adopted quotas. Five years later, it dropped the quotas because the fishermen thought that they were no longer needed. The U.S. government then made low-interest loans available to fishermen - in part to take up the slack left by the departure of the foreigners - and the number of boats increased. Between 1982 and 1986, the cod catch in Georges Bank dropped by 50%.
At this point, scientists such as Anthony urged the regional council to take drastic action, but fishermen thought that a series of milder steps would do the job. The Environmental Law Foundation sued the Commerce Department, and the New England Council began to develop a regulatory plan. Hearings were held. Agreement on the magnitude and methods of regulation was elusive. Meanwhile, the fishing kept on going.
Now the cod are nearly gone, and the halibut and haddock are ``commercially extinct.'' In 1993, the Canadian cod fishery was closed, causing the loss of more than 40,000 jobs. In 1994, U.S. quotas for haddock and cod were cut almost to zero off the New England coast, including, of course, Cape Cod, which was named for the once-plentiful fish. The fish are disappearing from overfishing.
This is a world phenomenon. In the past few decades, the world's people have acquired the technology to extract all the edible fish from the ocean in a few years, and population growth has increased the economic incentive to use the technology. More fishing is done for export rather than for feeding the fishing communities. Governments even try to subsidize fishing as part of a program of economic development. In Kerala, India, for example, the government tried to promote economic development by subsidizing fishing. First it subsidized large trawlers. Then, after protests from small fishermen, it began subsidizing their purchases of outboard motors and modern equipment. In both cases, overfishing resulted. The officials involved did not think of the fishery as limited, because it had seemed inexhaustible to the traditional fishing folk who had worked it for centuries.
In the Phillipe and Tongan Islands of the South Pacific, fishermen now pour bleach into the reefs - or even use dynamite - to make the fish come to the surface, where they are easy prey. This destroys the reefs, the habitat of future fish as well as places of great beauty. But once some fishermen do it with impunity, those who refrain will only lose out. The reefs, and ultimately the fish that live in them, will be destroyed by others.
This chapter will discuss these problems of fishing as examples of a general type of problem, typical of environmental problems of all sorts.
The main problem is that each person overfishing is best for him and worse for all the other fishermen. In other words, each would want all the others to stop overfishing. This kind of case is known as a commons dilemma. It goes by many other names and takes many forms: social dilemma, public goods problem, resource dilemma, free-rider problem, collective action problem, negative externality, n-person prisoner's dilemma, and so on. Scholars from psychology, economics, philosophy, and political science have recognized the importance of these dilemmas, and each discipline has taken a slightly different point of view on what the relevant situations are and what they should be called, hence all the different names.
In a commons dilemma, many people face the same pair of options: cooperation or defection. Defection (fishing as much as possible) is better for each individual, and cooperation (restraint) is better for everyone else. The term commons dilemma comes from Garrett Hardin's classic article, ``The Tragedy of the Commons,'' which used the problem of overgrazing cows on a common pasture as an example. Defection can involve taking from a common pool, as in the case of overfishing and overgrazing, or it can involve failing to contribute to the pool, as in shirking on the job. Such dilemmas exist when people cut down too many trees for firewood, have too many children, use too much water in a water shortage, cheat on their taxes, violate contractual agreements, or exaggerate their losses when they make insurance claims. Hardin argued that the natural course of a commons dilemma was to end in overuse of the commons: ``ruin is the destination toward which all men rush. . . . Freedom in a commons brings ruin to all.'' The only solution is ``mutual coercion, mutually agreed upon.'' Those involved must regulate themselves through collective institutions such as government.
Overfishing is a classic commons dilemma. For each person, catching fish helps him and hurts others. A single fisherman might think, ``I could fish only enough so that, if everyone else did the same, we could all continue to fish the same amount in the future, and I would benefit from this in the long run.'' But the future benefit from what a single fisherman does will spread around everyone else who fishes the same waters. The total effect of what a single person does will be so small that each individual would not notice. Unless the person has some sort of social bonds with the others, which make them all cooperate, the future benefit would be too little to make the person want to restrain his catch to the sustainable level.
An additional feature of some commons dilemmas is that the benefits of defecting are immediate and the costs to others are in the future. This makes the dilemmas more difficult to deal with. When this feature is present, the situation is called a ``social trap.'' Overfishing is a social trap because its effects are exerted in part by slowing the rate of reproduction of the fish, which affects the future. Fishermen themselves understand the problem quite well: ``Fishermen are hard pressed economically right now. They have to look at the short-term reality, that they have to feed their families like everyone else. It's tough to be an altruist and say, `Let's let the fish rebuild.' It is politically unpopular.''
The problem is more difficult when the short-term advantages of fishing are a matter of life and death. Shrimp fishermen on the West Coast of Mexico and their families were seriously malnourished for parts of the year because of regulations against fishing at those times. (In part, the problem came about because more shrimp were being exported for cash than were being eaten by the fishermen and their families.) Violation of the regulations was thus more tempting. This was so even with full and rational understanding that overfishing could deplete the shrimp. For children who might die from their weakness, the future was unimportant if they could not make it through the year. Under such conditions of high risk, neglect of the future is rational. The situation may sometimes be truly hopeless, but that is not the usual case.
Another source of difficulty is that it is sometimes in a person's rational self-interest to ignore the future. Older fishermen who plan to retire soon and who do not care about those who take their place have every selfish reason to take what they can while they can. If interest rates are sufficiently high, even younger fishermen may think that they could make enough money in a short time to retire early from fishing and live partly off the interest. But then the resource is lost to other fishermen, not to mention people who eat the fish.
We might think that all commons dilemmas would lead to ruin, but this is not so. Many have been solved because people have come to recognize them and come to some kind of collective agreement. In Alanya, Turkey, in the 1970s, the fishermen faced a situation of declining stocks and increased competition. Their local cooperative, after several trial-and-error efforts, adopted a set of rules that solved the problem. The rules, which are still in effect, involved identifying all the licensed fishermen in the area and assigning each one to a fishing location by lot. Because some locations are better than others, the fishermen rotate through all the locations, taking a new location each day throughout the fishing season. The technology ensures that plenty of fish will be left: it is impossible to take them all from any given site. The fishermen enforce the system themselves through formal and informal penalties.
Elinor Ostrom and her colleagues have described this and other examples of spontaneous solutions, not only in problems of fishing but also in use of scarce water resources (in southern California, for example) and agricultural land (in Switzerland, Japan, and elsewhere). They have contrasted these successes with other commons dilemmas that went unsolved, like that of the North Atlantic fisheries. The successful cases tended to be those with clearly defined boundaries (so that defectors could not enter after an agreement had been made), rules appropriate to local conditions (e.g., the assignment of locations would not work in an ocean fishery), democratic participation in modifying the rules, arrangements for monitoring and enforcement, gradual sanctions (not too severe for the first defection), mechanisms for conflict resolution (e.g., local courts), and noninterference from outside governments.
Ostrom's cases all involved small communities of up to a few hundred people, in which most people were known to other members of the community. This made enforcement easier, because defectors would lose their reputations. Other dilemmas, such as the North Atlantic fisheries, involve larger communities in which exclusion of outsiders is difficult without the cooperation of national governments. Yet many of these larger dilemmas have been solved too, even those that involve the whole world. A major international success story is the Montreal Protocol, in which the nations that produced chlorofluorocarbons (CFCs) such as Freon agreed to phase out their production. (CFCs destroy the protective layer of ozone in the upper atmosphere, leading to increased ultraviolet radiation, which has a variety of effects on plant and animal life, most of them harmful.) Other examples are international trade agreements and United Nations peacekeeping efforts.
The problem in the North Atlantic is thus not just that the fishermen overfished. It is unreasonable to expect them to do otherwise, in the absence of some agreement. If a single fisherman decided to stay home in order to help the fishery rebuild instead of going out and working as hard as all the other fishermen, his action would have very little effect on the fishery but a pronounced negative effect on himself. Only a saint without a family would do this.
But this was not the only choice. The fishermen are represented by various organizations. A second decision that each one faced is whether to try to influence these organizations one way or another, or not at all. Unlike cutting back on fishing in the absence of rules, such attempts at influence are not going to ruin anyone's livelihood. It does not take a saint to make the effort. Of course, if enough people make the effort and succeed, there will be cutbacks for all, but all fishermen will also benefit because of cuts made by others. Here is another example of how political action is effective, relative to its cost, because its benefits are spread over many people.
What happened was that many of these organizations opposed any regulations proposed by others and made little effort to design better regulations. Jake Dykstra, a fisherman involved in regulatory issues over a long period, said, ``The fishermen are indicating rather strongly that if they could get a management program that they believe in, they have a willingness as never before to self-enforce it.'' But, according to Paul Schneider, it ``became apparent that constructing an effective groundfish plan that the fishermen of New England `could believe in' was going to be extraordinarily difficult, and not only because of their understandable disinclination for a `stick in the eye.' For every suggested policy option, somebody - fisher, conservationist, or enforcer - had a reason why it just wouldn't work. For every measure amenable to the fishermen of one port, there were fishermen in another port who claimed they would be put out of business.''
An individual fisherman can ask those who represent him to fight regulation, or to lobby in favor of reasonable regulations proposed by others or constructively proposed changes that might be acceptable to all involved. It is perfectly reasonable for a fisherman to do this while, at the same time, fishing as hard as he can. Similarly, people concerned about their nation's budget deficit might write letters asking for a tax increase while, at the same time, filling out their own taxes carefully to pay the minimum legally required. Writing a letter to your representative surely has less cost to you than sending in an extra check for the taxes that should be collected, yet the benefits might be as great in increasing the chance of a change that will benefit everyone.
So why did the fishermen oppose regulations? I suggest that two intuitions were basic causes of their opposition: perceived unfairness and a belief in the rightness of autonomy. Once these intuitions had dictated a position, other reasons were recruited to justify it. These intuitions and others like them may characterize other commons dilemmas. Although these dilemmas arise because of the conflict between self-interest and the interest of others, the solution to them may sometimes have more to do with thinking differently, so that mutual regulation can be achieved, than with simply becoming less selfish.
When people must cut back, it is important that the pain be fairly distributed. Unfairness becomes an excuse for defection and for opposition to the proposed scheme. Each group evaluates fairness in its own favor.
A psychology experiment has made this point clearly. Subjects were instructed to fill out questionnaires until told to stop. They expected to be paid but did not know how much. Each subject was given either three or six questionnaires (depending on the experimental condition) and was told to stop after either 45 or 90 minutes. When a subject finished, she was told that there had been another subject who had had to leave before he could be told that he was supposed to be paid. The experimenter, who also said he had to leave, gave the original subject $7 (in dollar bills and coins) and asked her to send the other subject his money (in the stamped, addressed envelope provided). The subject was
told that the other subject had put in either more, the same, or less time and had completed more, the same, or fewer questionnaires.
At issue was how much money the original subject would send to the ``other'' subject (actually a confederate of the experimenter). Subjects who either worked longer or completed more questionnaires than the ``other'' gave the other less than $3.50. It just cannot be true that, if they had been asked before the experiment, the subjects who worked longer would have thought that time was more important and subjects who did more would have thought number of questionnaires was more important. Subjects apparently seized on any excuse to see themselves as deserving more. When the original subjects were equal to the other subject on both dimensions, they sent almost exactly $3.50, on the average. Only when subjects did worse on both dimensions (time and number of questionnaires) was there a slight tendency to send more than $3.50 to the other.
This experiment may tell us something very interesting about human nature. People are not simply selfish; they want to do what is right. Almost all subjects took the trouble to give the other subject half of the money when they had no reason to do otherwise. When people behave selfishly, then, it is likely because they have deceived themselves into the belief that they are behaving fairly. More relevant here, this experiment shows how people can convince themselves that, when different standards of fairness are available, the correct one is the one that favors them.
In 1981, the National Marine Fisheries Service proposed a regulation for striped bass. Fishermen in regions where the bass spawned would be allowed to catch relatively small fish, as short as 14 inches. The length limit for other fishermen was much larger. This was a compromise designed to protect the former group. Larry Simms, a Maryland fisherman who had led the successful opposition to previous attempts to raise the limit from 12 to 14 inches, said, ``You take [the 12-inch rule] away from us and you take away a million dollars. . . . We don't want to be the only ones sacrificing . . . . The fish swim away after they reach 14 inches.''
Undoubtedly, the Maryland fishermen thought it was fair to base limits on past catch levels, while others probably thought that the same length limit should apply to everyone. Other disputes involve groups who think that fairness involves some special provision for them. For example, many Japanese viewed proposed restrictions on the bluefin tuna as ``an outrage aimed at Japanese fish-eating habits, . . . a part of our national food culture.''
Unfairness is still the major argument against proposed regulations. Recent opposition to a quota system in the United States has come from small fishermen, who object to basing quotas on past catches, which they see as favoring large ships. They are supported by some environmental groups, particularly Greenpeace. Alaska Governor Tony Knowles has opposed the U.S. National Marine Fisheries Service efforts to save the chinook salmon by declaring it an endangered species and limiting Alaskan fishing severely. Knowles argued that the salmon declined because of hydroelectric dams on the Columbia River in Washington state, and that the restrictions would benefit Canadians, who would not be subject to them. Likewise, Connecticut's congressional delegation is working to change the quotas for fluke (flounder), which are now based on past catches. Connecticut had lower catches than other states and now wants quotas to be divided equally among the states.
The perception of regulation as unfair is particularly acute when those who bear the brunt of the regulation do not see themselves as the cause of the original problem. Often, in fact, their perception is correct - they are not the cause. But they may still be the cure. This is particularly common with small-scale fishermen, whose fisheries are invaded by large factory trawlers, often from other nations. Pollution also flows into the ocean and can ruin a fishery, just as it has almost destroyed the famous oysters in Chesapeake Bay on the East Coast of the United States. When fishermen or crabbers are told that they must cut back to preserve what remains, they feel justifiably angry. But those who are truly at fault cannot bring the fish or the crabs back. In the Chesapeake Bay, the problem occurred before most people were aware of the effects of pollution. Laws have now been passed to prevent more pollution, but the damage has been done. Likewise, in the 1970s, most nations prohibited foreign fishing within 200 miles of their coasts, but again the damage to coastal fisheries had already been done, and small-scale fishermen were asked to cut back. Those who took the fish cannot put them back, and most of the damage was fully within existing international law. Until the 1970s, nations accepted the idea that the seas were a common resource open to all.
In other cases, the unfairness is less blatant. Restrictions on seal hunting caused the seal herd to increase, and seals eat fish. In many cases, fish stocks declined for unknown but entirely natural reasons. Yet the same fishermen must bear the brunt here too, and it still seems somewhat unfair that they should suffer for what they did not cause. It seems even more unfair when another human being caused it and cannot be made to suffer. Yet from a utilitarian point of view, restraints on the fishermen are the best solution, if not the only possible solution. We cannot control natural fluctuations. In the case of the seals, it would be at best a short-term fix to kill the seals in order to help restore the fish.
A fundamental intuition about fairness is that those who cause harm should pay and those who do not cause harm should not pay. This intuition establishes an aspiration level for the payments made by each of the parties involved. When this intuition is necessarily violated, as it is here, anger results. People may then seek some other outlet for that anger in the form of some other way in which they are being treated unfairly, such as the imbalance between regions, between large and small fish, or between commercial and sport fishing. They will tend to insist that the imbalance against them be rectified before they accept regulation. Because most fishermen are able to find some way in which they are being treated unfairly, a majority will resist practically any regulation.
Perhaps the most important cause of the failure to demand regulation is that fishermen want autonomy. Small-scale fishermen may be particularly protective of their autonomy because of the nature of their work, and citizens of the United States are steeped in a political culture that emphasizes autonomy, yet this attitude is found in fishermen the world over, and, to some degree, it seems likely that nobody likes to be given commands, even in the most collective cultures. This desire for independence, which may be particularly strong in East Cost fishermen, is aggravated when the rules seem the slightest bit neglectful of the details of the local scene. ``The head of one East Coast fishermen's group called the proposed NMFS [National Marine Fisheries Service] attempt to curtail Georges Bank overfishing `moronic little bureaucratic games' and accused the agency of `a desire to take control of every aspect of the business.' ''
This intuitive attitude toward autonomy leads people to oppose coercive solutions, yet commons dilemmas such as overfishing require such solutions. We cannot depend on the goodwill of fishermen. The immediate benefits of excess fishing to the individual are too great, and the long-term benefits from a single fisherman's restraining himself are spread over so many people that no single person would even notice. Even if all current fishermen just happened to be saintly enough to cooperate on their own and reduce their catches voluntarily, we can be sure that others would move in to take advantage of the situation if there were no penalty to prevent them.
In a small fishing community without outside interference, the coercion may be subtle, taking the form of social disapproval or withdrawal of support (e.g., failing to help in an emergency), but it is coercion still. And the number of communities that can regulate themselves in this way is dwindling.
Some people have advocated ``market solutions'' to the problem of overfishing and other social dilemmas. For example, a market could be set up for the right to fish. This could take the form of a transferable license to fish some amount. The question of what ``amount'' to use is a technical one: it could be number of fish per year, pounds of fish, days fishing, number of boats, area of the sea, length of coastline, or some combination of these and other quantities. The number of licenses sold would be set so as to preserve the fish, to make fishing sustainable. Then the licenses could be bought and sold freely, just like any other productive property, such as land for agriculture. Although such schemes may work well, the idea that they are somehow based on freedom - because the market is ``free'' - is a play on words. Some authority must decide how many licenses to sell and, most important, must enforce the rules against taking more fish than the license allows and against fishing without a license. Such enforcement can be costly. And it is still inescapably coercive.
Yet many people resist the idea of coercive solutions in any form. Perhaps some people just have not heard of commons dilemmas (under any name). Their thinking about social issues is influenced by the examples they can think of, which happen to be cases in which a free market works best, and in which the support of property rights is so much in the background that it is not seen as coercive. Or people understand the concept of a commons dilemma but do not happen to think of it. More likely, they have heard of such situations but do not want to believe that overfishing, or whatever dilemma they are in, is one of them. They engage in wishful thinking.
Perceived unfairness and intuitive support for autonomy often lead us to wishful thinking. The fishermen were no exception. They convinced themselves that the proposed regulations were not needed, and they did this in several ways.
Fishermen wanted to believe the fisheries scientists were wrong when the scientists said that overfishing was causing a decline and that regulation was needed. Joseph Brancaleone, a former fisherman who chaired the New England Council, defended the council's moderate (and ultimately ineffective) controls in terms of scientific uncertainty: ``The data that we have are so slim that we can't put a number on [the effect of the controls]. By the third or fourth year, we'll have the data.'' Another fisherman put it more bluntly, speaking to government scientists: ``You fellows seem to have a lot, I don't know what you call it, somebody asks you a question and we ask you a question, `Well, I - I don't really know.' And you base everything you tell us on what you really don't know. It kind of bothers us fishermen. We're going to be told we've got to do all this stuff, but nobody really knows anything. Doesn't make much sense. You guys have got some jobs, but I don't think we're going to have any if you keep on. Save the fish, but what about the people?''
Uncertainty is used here as an argument against regulation. The default or status quo is assumed to be correct in the absence of clear evidence against it. (The status-quo bias is also at work here. If the regulations were already in effect, the arguments against them would probably be less vehement.) But uncertainty could go either way. The use of uncertainty to support the status quo is like not having bypass surgery because you might die from the operation. Decisions under uncertainty are all calculated risks. For both the fishermen and the surgical patient, both action and inaction are risky. Uncertainty leaves room for wishful thinking, but the world is deaf to our wishes.
Fishermen have an opportunity for wishful thinking in the fact that fish stocks fluctuate for reasons unrelated to overfishing, such as the weather or ocean currents. When stocks decline, fishermen can usually find some other explanation aside from overfishing, and they are often correct. But the stocks would still be higher without overfishing than with it, and they would recover more quickly.
Such wishful thinking is easier when people do not understand how expert knowledge comes about, through critical inquiry. Scientists and other experts frequently hedge their pronouncements with caution. They know they are talking about objective matters and could be shown to be wrong. Their opponents do not take this critical stance toward their own beliefs, so they express high confidence.
Many writers have pointed out that fishery regulation can work without changing human psychology if the proper administrative procedures are set up. Many fisheries throughout the world have been regulated on the basis of scientific advice, so this is humanly possible. Still, we are not doing nearly as well as we might, so some change in human thinking could help, too. All of the sources of resistance to particular regulations operate as well when changes in the law are proposed. Fishermen oppose these changes, too.
Fishermen also argued that proposed regulations were the result of political influence, not science. In one case, the political influence of sport fishermen was blamed. Of course, such influence is possible, but the fishermen's degree of confidence that the scientists were wrong was unwarranted because the scientists had been correct repeatedly.
Wishful thinking led to blaming others, with the implication that regulation of fishing was not needed: ``The finger is being pointed at commercial fishermen for overfishing, and it's not our fault. The real fault is pollution.'' Some fishermen and politicians blamed the decline on the environmentalists, who were successful in banning the annual Canadian seal hunt in 1983, leading to an increase in the seal population, which then competed with the fishermen for the fish. Others argue that violations of existing rules are the problem. Sometimes the blame was placed on nature: declines were the result of year-to-year variations or spontaneous increases in the number of predators.
In general, when a policy goes against some intuition, such as support for autonomy, blame and mistrust of others can justify advocacy of other solutions. Either the facts are wrong or someone is doing something even more insidious.
Another type of wishful thinking is what we might call the utopian fallacy: if a proposal is not ideal, then it should not be adopted. The wishful thinking is in believing that the ideal is possible. Of course, this is a good attitude if the ideal is possible, but typically it is not. In this way, the utopian fallacy resembles many intuitions I discuss. By adopting the ideal as an aspiration level, it views the proposed rule as a loss. But the real choice is between an imperfect rule, which will save the fish, and no rule at all, which will not. In the case of fisheries, the utopian fallacy is particularly clear because of the cost of delay: as regulators hold more hearings and strive for perfection, the fish continue to decline.
People support their intuitive desire for autonomy by convincing themselves that no coercion is required. They think that the goodness of people, suitably focused, will make coercion unnecessary. We see this in some followers of Ayn Rand, in radical free-market advocates, and in leftist communitarians. For example, an anonymous posting on several Internet news groups advocated the abolition of income taxes and of all money on the ground that
the restoration of economic freedom by no longer requiring people to use money . . . would establish the best possibilities for a civilized world . . . All right and sustainable societal change can be brought about by ``moralsuasion'' and a good character based economic life . . . If personal taxes are eliminated then people [can] set up a natural, more efficient and more truly profitable barter system which has at its center good character and personal trust . . . This type of civilization evolves locally self-sustaining inter-networked human eco-systems with naturally balanced resource distribution systems - i.e., the planet naturally works for 100% of humanity.
When people think about solutions to commons dilemmas, they frequently think of voluntary solutions; coercion is an option that rarely springs to mind. For example, in one study with Bill Hale (a student), I asked several college students about seven different examples of commons dilemmas: declining soil productivity, air and water pollution, depletion of fossil fuels, global warming, population growth, loss of animal and plant species, and overfishing. After we described each item, we asked subjects what actions might help the situation and who should take the actions, and we asked why the actions have not been taken yet. (We emphasized that we wanted all the actions that might be taken, not just one.) Several subjects failed, in one or more cases, to note the possibility of collective action as a solution. In answer to the question about what actions should be taken, they mentioned only individual restraint. For example, on the problem of soil productivity, ``The companies that own the land should use the land for grazing and producing for only a short time and then let the land regain its nutrients for a time.'' Or, on the global warming item, ``Logging companies can take steps to gradually reduce acreage of cut forests.''
I noticed a similar assumption in a video presentation on automobile pollution put on by the Franklin Institute Science Museum in Philadelphia. The presentation allowed an audience to respond to questions by pushing buttons on their seats, and their responses were immediately displayed. After warning of the dangers of pollution, the video asked, ``Would you use public transportation to commute to work?'' and other such questions. Of course, the audience answered yes because the wording of the question allowed them to fill in whatever ``if only'' conditions they chose (e.g., if only it weren't so expensive, if only it were more convenient). The presentation implicitly assumed that the solution was an increase in voluntary use of public transportation. Of course, for most commuters who drive, considerable personal sacrifice would be required for this solution to work. A significant change in transportation use requires significant incentives, in the form of making public transportation more attractive, making driving less attractive, or both.
Even Lester Brown, head of the Worldwatch Institute, does not seem immune to this kind of thinking. In an article on population, he says, ``If people know that maintaining current family size will reduce cropland area per person by a third or half during the next generation, they can see what this will mean for their children. If they know that large families almost certainly will bring more hunger and even mass starvation, they may well decide to shift to smaller families.'' Unfortunately, each family's food intake is determined mostly by other people's children. It would be nice if mere awareness of a social dilemma could solve it, but that is not always the case.
A special form of wishful thinking about the sufficiency of voluntary solutions is apparent when the benefits of cooperation are in the long run, a situation called a social trap. People may think that cooperation is in their long-run self-interest, even though they understand that it is not in their short-run interest. In a small group of people who interact regularly, this may of course be true for a variety of reasons: cooperators gain a reputation for cooperating, which makes them more desirable; and they also really do influence the tendency of others to cooperate, so they benefit somewhat from the later cooperation of others. But in large-scale social dilemmas involving thousands, millions, or billions of strangers, both of these effects are unlikely to pay off. People may still think cooperation pays off because they understand that collective cooperation pays off in the long run, and they apply this conclusion to the acts of individuals. In essence, they think about individual action without holding constant the actions of others. As a result, they believe that cooperation is in a person's long-run self-interest, so that no coercion is required. Instead of coercing people with rules, it is sufficient to explain to them where their real interest lies.
I looked for this belief in another questionnaire study, in which I told people about various social traps, including one involving a water shortage and another involving overfishing. I made it clear that people could overuse the water or overfish without anyone knowing about it, so that their reputations would not be affected and so that they could not set an example - good or bad - for anyone else. To check their understanding, I asked subjects whether it was in their immediate self-interest to restrain themselves, and most said that it was not. I then asked if it would be in their long-run self-interest to restrain themselves, and about half - more or less for various items - said that it was. I found this result even in a fishing scenario in which I asked whether cutting back would increase a fisherman's long-run income, so that it was clear what ``self-interest'' meant.
Subjects' explanations seem to confuse the effect of one person's actions with the effect of everyone's. They speak as if they were making the decision for everyone, not just one person: ``By not cutting back on my current level of fishing I would more than likely be hurting myself in the future, for sooner or later the fish population would dwindle down to zero, leaving me with absolutely no income. I am willing to give up short-term financial growth for long-term financial stability.'' ``For the short term I'd bring in more money [by fishing more], and be better off. But the more and more of this particular fish I caught, the less there would be breeding in the ocean. Eventually, they will die out and I will be left with none . . . .'' ``It would not increase directly, but in the long run you will end up with more money over the years.'' ``We shouldn't be selfish. We should look to the long term instead of the short term. . . . In the long run, it may [increase my income] because the fish would be less frenzied, and there would be more of them, and they would have time to reproduce more so there may be more fish to catch in the future.'' The last response illustrates the wishful thinking inherent in the idea that virtue is its own reward. If only this were true.
Many responses explicitly referred to the group even though the question was about the individual: ``In the long run, the water shortage would eventually affect me too. If everyone thinks the same way, . . . everyone will water their garden and the water shortage will continue and worsen, which will highly affect me.''
Thus, people confuse their own interest with that of the group, especially when the long-run is at issue. This confusion has advantages. It may make people more willing to cooperate spontaneously or even to take political actions such as voting, which are actually somewhat costly to them even when the selfish benefits are considered.
But it also has dangers because the confusion of self and group may work both ways. People may see their own interest as identified with a particular group that they are in, such as Cape Cod fishermen or Rhode Island fishermen (or Americans). They may thus think that their support of this group is justified in terms of their self-interest, even when the interest of this group goes against the interest of some larger group that also includes them, such as fishers as a whole (or all people). In such cases, it might be better for people to understand that actions on behalf of the smaller group are just as much against their narrow self-interest - if only because of the effort involved in taking action - as are actions on behalf of the larger group. If they understood that the good reasons for action were based on the benefits to others in both cases, they might be more willing to act on behalf of the larger group. Such action would smooth the path to agreement among groups, and it would do more good on the whole than would action for the smaller group only.
There is hope. Some fisheries have been saved by regulation - either by local indigenous groups of fishermen or by governments - and they are thriving. Even at the world level, agreements to protect dolphins and whales have had some success. Regulation of forests has at least reduced the rate of their destruction. Rich nations have taken other successful steps to protect their internal environments, such as laws that have reduced air and water pollution.
One of the great success stories of recent years is the Montreal Protocol of 1987, in which countries around the world agreed to phase out the production and use of chemicals that were harming the upper atmosphere. Ozone in the stratosphere reduces the ultraviolet light from the sun that reaches the earth. Increased ultraviolet light would cause higher incidence of skin cancer and many other effects on animals and plants around the world. In 1970, the chemist Paul Crutzen called attention to the deleterious effects of nitrogen oxides on ozone. This argument was one source of political opposition to the building of supersonic transport airplanes in the early 1970s. In 1974, F. Sherwood Rowland and Mario Molina found that chlorofluorocarbons (CFCs) were transported to the ozone layer, where they could also destroy the ozone. CFCs are inert chemicals that do not break down spontaneously, so their effects are long lasting. They were used in aerosol cans, in plastic manufacture, and in refrigerators and air conditioners. (Freon is a CFC.) Many countries soon banned their use in aerosol cans. In 1985, a pronounced reduction of the ozone layer over Antarctica was observed.
In 1987, representatives of 24 nations and the European Economic Community met in Montreal and agreed to reduce production and use of CFCs and other ozone-destroying chemicals. The agreement made provisions for developing countries, which were allowed more time and given some assistance in converting to CFC substitutes. It also called for frequent reviews based on scientific evidence. These have occurred, resulting in several more agreements and some speeding up of the process. The treaty is working. Declines have been found in ozone-destroying chemicals (those containing chlorine), and although others were still increasing (those containing bromine), the problem at least appeared to be getting no worse, and a reversal is within reach.
The agreement is interesting for several reasons. First, it is a solution to a commons dilemma that involves the whole world. Commons dilemmas are easiest to solve when they involve small communities; this was no small community. Second, the problem of fairness was solved by compromise. The poorer countries had some relaxation of the rules, but not as much as they would have liked. Third, the agreement was made on the basis of scientific evidence, before any epidemic of skin cancer or extinction of species from ultraviolet radiation actually occurred.
Some special factors contributed to this success: the problem was clearly delimited; the costs of solving the problem were manageable because there were substitutes available for most of the chemicals involved; and these chemicals were made by a relatively small number of companies, so that monitoring was relatively easy. The scientific evidence was also fairly clear about the harmful effects of CFCs. However, the U.S. government supported the agreement even before its cost-benefit analysis was completed (although it was clear what the results would be). Part of the support for the agreement among government officials came from the intuition I've called naturalism: ``Don't fool with Mother Nature.''
Of course, success is not perfect. Moreover, despite the fact that Crutzen, Rowland, and Molina won the 1995 Nobel Prize in chemistry for their discoveries, wishful-thinking opponents of government regulation continue to question the science. Tom DeLay, the Republican majority whip in the U.S. House of Representatives, called ozone depletion a ``media scare.'' And the state of Arizona in 1995 passed legislation declaring itself exempt from the federal law implementing the Montreal Protocol; Governor Fife Symington declared the ban on Freon production to be based on ``hokey science.'' Still, despite these problems, we must count the Montreal Protocol as a success.
Much less successful have been the world's efforts to prevent global warming from human activity. It seems likely that the earth's surface is going to get warmer over the next few decades. At least part of this warming will result from the production of carbon dioxide (CO2) and other gases such as methane and CFCs. These ``greenhouse'' gases reduce radiation from the earth into space without reducing the effectiveness of the sun at warming the earth, making the atmosphere more like a greenhouse.
People influence production of these gases in many ways, but mostly by burning fossil fuels such as oil, coal, and natural gas, which are oxidized to form water and CO2 when they burn. Burning wood does not have any effect unless we fail to let it regrow. We can think of trees and other plants as reservoirs of carbon, keeping it out of the atmosphere. So if the total mass of plants is reduced - including the residues of plants in the soil and the oceans - then atmospheric carbon must increase, and that happens through increases in CO2. Massive deforestation without replacement of the trees is thus part of the problem too.
The expected warming will have many consequences for us, some positive but more negative. Oceans will rise and coastal areas will become uninhabitable. Dikes will be needed to protect cities and some agricultural land. Some coastal countries, such as Bangladesh, will be particularly hard hit. Changes in weather patterns will make some land dryer and other land wetter. Current predictions suggest that the net change will be for the worse for food production. Even if it is not, time will be required to change the organization of agriculture. People will have to move away from areas that become too dry to provide sufficient drinking water.
All these effects are highly uncertain. The predicted effects could not occur at all, or the effects could be much worse. In the case of ozone, the chemistry was well understood, and action was taken only after reduction of ozone was observed. The predictions of global warming are based on computer models of global climate, models that are not fully based on theoretical understanding, although they do predict current seasonal weather patterns fairly well. Uncertainties have to do with things like how much carbon can be absorbed by the soil, how much of the sun's energy will be reflected away by increased air pollution from industry, and how arctic ice sheets will respond to whatever warming occurs. Specific predictions about particular regions, such as the North American grain belt, are even more uncertain.
The potential problems are exacerbated by other trends, particularly the increase in population. If population were stabilized, we would have a better chance of producing sufficient food. As it is, however, agricultural production is barely keeping pace with population growth, and global warming could upset the balance. Although much malnutrition results from inadequate distribution rather than inadequate supply, we cannot assume that the problems of distribution will be solved by any given time. To say that ``population growth is not the problem because the real problem is distribution'' is to engage in the utopian fallacy, if, as seems likely, controlling population is more feasible than redistributing wealth.
Another trend that creates problems is the economic growth of some countries out of poverty - for example, China, India, and the countries of Southeast Asia. This trend is to be welcomed, of course. But development brings increased use of fossil fuels, and more CO2. China, in particular, has vast deposits of coal, on which it depends for future development.
In sum, unlike the problem of the ozone layer, that of global warming is complex, uncertain, and linked with many other changes and policy questions.
Given these trends, we have several options, many of which may be combined with other options. First, we could do nothing special now. When the oceans start to rise, people will build dikes. When land becomes arid, people will move. When population outstrips food production, we could stave off mass starvation for a while by switching from meat to grains, and this might give us time to control population. Given the uncertainties, and given that the warming process is probably slow, this might be the best approach.
Second, we could prepare for the consequences. We could direct new development away from coastal areas and areas that are likely to become dry. We could try to reduce population growth. We could develop plans for gradual resettlement of those who would have to move. The trouble here is, of course, the very uncertainties that make the first approach attractive. We could take the wrong steps and waste resources doing so. Even planning requires resources.
Third, we could reduce the production of greenhouse gases. Some of the steps required to reduce greenhouse gases could be beneficial for other reasons. For example, increases in energy efficiency could be cost-effective in their own right. Such beneficial steps would - given the laws of economics - be taken anyway, eventually, but it might be worth putting some resources into speeding them up. Even here, the resources required, such as incentives and taxes, are not trivial.
Other steps are more costly still. Substantial reduction of CO2 production would weaken the economy of any country that tried it. Large amounts of resources would go into rebuilding and redesigning everything, from automobiles to the layout of residential patterns. Changes in social behavior would be required. For example, substantial energy savings could be realized by concentrating residential patterns, with more people living in highly concentrated developments (perhaps surrounded by nearby ``green belts''), linked by convenient systems of mass transportation. Such a step would require overcoming the reluctance of many people in developed countries to live in urban areas, a reluctance stemming from the current social decay of cities in many countries.
Fourth, we could try to remove CO2 from the atmosphere. We could do this by ``natural'' methods such as planting trees - thereby rebuilding the massive forests that people have destroyed in the past few hundred years - or by ``unnatural'' methods such as ``fertilizing'' the Antarctic Ocean with iron so that microscopic plants could grow there in abundance, sucking CO2 out of the atmosphere.
Fifth, we could try to change the course of development of those countries that would rely on increased use of fossil fuels, such as China. We - that is, the international community - could try to induce these countries to develop in other ways. Such induction could involve bargaining, in which we offer some benefit in return for cooperation.
Any of these actions to counter the problem require cooperation of the sort that occurred in the Montreal Protocol. Unlike what happened in the North Atlantic fisheries, people would have to be willing to accept the idea that some agreement is better than no agreement, even if it seems unfair and even if it threatens national and individual autonomy. I am not sure that any action is needed. The problem is that, if action were needed, and if that were to become clear to scientists, we would not be able to take it.
All the elements for an impasse are in place. Unlike the situation with stratospheric ozone, the magnitude and nature of global warming are highly uncertain. We know that increased CO2 must lead to warming eventually, that CO2 has been increasing, and that warming itself will have some of the effects listed earlier. We do not know how fast this will happen, or even whether the warming effect will be countered by opposite cooling effects such as those that caused past ice ages. Quantitative predictions are difficult because they require the use of computer models of the global climate, and the best models still do not work well. Most models require fudge factors just to explain the current temperature level in a way that is consistent with other data. They predict that greater warming should have occurred than what has been measured, so they seem to be taking insufficient account of some factor that is slowing things down. Or, alternatively, the data on past warming aren't really good enough to say that the models are wrong. The models also fail to incorporate various mechanisms that could lead to a sudden increase in warming.
Even as these uncertainties are resolved, the public and their representatives will be skeptical. Uncertainty provides room for wishful thinking, and it causes resistance to new evidence. Those opposed to international agreements or government intervention have found, and will probably continue to find, plenty of scientists who will tell them about the uncertainty, and that is all they need. Science is always uncertain. Rush Limbaugh, a popular and very conservative radio talk-show host, has argued that scientists disagree, so we need do nothing. Limbaugh's evidence comes from opinion polls of scientists, which show considerable disagreement on the question of whether human-caused warming has already occurred. This is, of course, a different question from the more important one, which is whether it will occur, or, even more to the point, whether it is likely enough so that we ought to worry about it. Thomas Schelling, in a very wise essay about global warming, argues that our real concern should be preparing for unexpected effects that are larger than what the models predict. ``There isn't any scientific principle according to which all alarming possibilities prove to be benign upon further investigation.''
Signs of problems to come are already present. Following the example of the Montreal Protocol, the United Nations Conference on Environment and Development took place in Rio de Janiero in 1992. Among other agreements made, the Rio Declaration on Sustainable Development, signed by 150 countries, committed the signers to reduce CO2 emissions to 1990 levels by 2000. As I write this in 1997, a conference just held in Kyoto, Japan, let to a tentative agreement to reduce CO2 and other greenhouse gases, but the agreement in its present form seems unlikely to be approved because the United States wants developing countries, such as China, to accept binding reductions and they do not think it is fair for them to deny themselves what the developed countries had in the past. Let us ask, though, what developed countries like the United States could do to reduce greenhouse gases.
One thing that might be effective is a tax on the carbon content of fossil fuels. This would discourage, in the most direct way, the emission of CO2. It would leave people and businesses free to decide, within this constraint, what is most efficient for them. Some might move from oil to natural gas. The latter has relatively more hydrogen (which produces water when burned) and relatively less carbon (which produces CO2). Some might replace fossil fuels with other sources of energy such as nuclear power, wind energy, or ``biomass'' (wood or other plants that are burned but then replaced). Even moving from coal to oil would help.
More generally, the idea of taxing whatever causes bad effects is economically efficient. It is more efficient than ``command and control'' regulation because it allows these bad effects to occur when there is enough benefit in return. (True, the test of ``benefit'' is what people can pay for. Although this test is crude, it is not so crude as to be useless.) If energy cost more because of the tax, people would still use it - they would just waste less of it. They would buy more efficient cars and refrigerators, for a start. Some economists have suggested that taxes of this sort could ultimately bring in enough revenue as to replace most other taxes.
In 1993, the newly elected Clinton administration proposed a complex tax on energy for the United States, a BTU tax with exemptions for solar, wind, and renewable energy sources. The tax was designed to reduce both the U.S. government's budget deficit and greenhouse emissions. The administration had already considered and given up the idea of a carbon tax and a gasoline tax. As Vice President Al Gore explained: ``A gas tax . . . is unfair to those in the rural areas and western states who drive long distances. A carbon tax . . . which might make sense for other reasons, hits especially hard the coal states. . . . A BTU tax . . . is fair in its impact to every region of this country.''
Although the BTU tax was designed to minimize perceived unfairness from the outset, the perception persisted, and the tax ultimately died. Energy producers claimed that some sources of energy would be hit harder than others, particularly coal, oil, and natural gas. The American Trucking Association complained that truckers would pay relatively more than other energy users because of the formula adopted. Senators from farm states - which rely heavily on diesel fuel for farming - and from oil-producing states ultimately formed a coalition to defeat the proposal in the Senate. Of course, any tax will hit some people harder than others and is thus subject to interpretation in terms of unfairness; indeed, a majority of Americans seem to agree that a ``fuel tax would be an unfair way to reduce fuel consumption because some people are forced to use more fuel than others by their business or personal needs.''
The farm-state senators were not, by themselves, enough to defeat the tax. The public was largely against the tax, perhaps because people are against taxes in general, but perhaps also because they they thought that it was unfair. Of course, those who were especially hurt by the tax screamed the loudest, but if fairness were not the issue, and if this was indeed the fairest possible tax increase that would reduce both greenhouse emissions and the deficit, then it should have won majority support. We do not know for sure, but the lingering perception of unfairness might have led to its demise. This was especially likely if each senator attended most to those constituents who were hurt relative to others. The situation is analogous to that of the fishermen, with each group applying a concept of fairness that favors itself, each group finding any agreement unfair, and no agreement coming about.
Also, like the case of the fisheries, the costs of the tax were immediate and the benefits - both economic and environmental - delayed. This made people more prone to wishful thinking, trying to find reasons for delay or inaction.
Some of the perceived unfairness was essential to the goal of reducing CO2 emissions. Any tax that does this will penalize those who emit more CO2 more than it will penalize other energy users. The idea is to shift the economy away from activities that cause emissions and toward other activities. This was perceived as unfair because it ``punished'' people for something they could not foresee. As Senator Max Baucus put it, ``Either we have an energy tax where everybody pays evenly, or we have no energy tax.'' Of course, such a tax would be impossible in principle: some people use more energy than others; and if the tax were equal for everyone, it would not provide any incentive to reduce energy use. Senator Baucus's intuition of unfairness worked against the greater good.
In the end, the administration settled for a 4.3 cent per gallon tax on transportation fuels and more budget cuts. (Senator Baucus, from the sparsely populated state of Montana, was able to kill a larger gasoline tax.) This left the United States with the cheapest energy in the developed world (except for Canada), despite importing half of its oil, and with little more incentive to reduce emissions. Worse still, in 1996, the price of gasoline began to rise, and Congress voted to repeal this small tax for one year, with the president's promise of support. Editorial opinion was largely against the repeal, but on grounds of budget balancing or fairness (noting that U.S. gasoline prices were low relative to other countries). I could find only one editorial that even mentioned the effect of gasoline use on the atmosphere, and that mention consisted of two words in a list of effects.
This, of course, is what happens within a single nation. Played out at the world level, distorted perceptions of fairness and wishful thinking will only be worse, if we do attempt to reduce global warming by reducing output of greenhouse gases. But we may have another way.
Marine phytoplankton, the little organisms that populate the oceans and provide food for other creatures, contain carbon. The limiting factor on the growth of these plankton is the iron content of the water.
If you put some ocean water in your bathtub and add iron, plankton will grow. In 1990, a panel of the U.S. National Research Council recommended that dumping iron in the ocean be studied as a possible solution to the problem of global warming. The plankton would suck CO2 out of the atmosphere. The scheme was called the ``Geritol solution'' after a patent medicine, an iron supplement for ``tired blood.'' As the late John Martin, one of the scientists who developed the theory, put it (half in jest, it seems), ``You give me half a tanker full of iron, I'll give you another ice age.'' As it turned out, the experiments were done. The iron did increase the growth of plankton, but it disappeared too quickly. Much more than half a tanker would be needed - in fact, so much as to make the solution impractical.
Many people, including scientists, were appalled at the idea of tinkering with nature: ``Environmental intervention on a global scale, no matter how well-intentioned, could prove worse than the global changes already in evidence. We simply do not understand the global ecosystem well enough to attempt large-scale quick fixes.'' Scientists' comments were echoed by others. A U.S. congressional staff member said, in an interview, ``Some scientists are suggesting that you could put 500 tons of iron into the ocean to make plankton bloom. The plankton would soak up carbon. It's obviously bullshit. On the face of it, it's ridiculous.''
But the idea won't die. In 1995, Kenneth Coale and his colleagues modified the procedures for adding iron, and the experiment yielded much more promising results. But Coale himself is worried: ``We are conducting research that may be used toward geoengineering and that does make me feel a bit uncomfortable. I don't feel we have the same dilemma as the scientists who worked on the Manhattan Project [building the first atomic bomb], but there are some similarities.'' He, and a few others, think that research should proceed.
How can the skeptics be so sure that the idea won't work? It seems that they started with their moral intuition against ``tinkering with nature'' and simply assumed that the scientific facts would bear them out by showing that the idea would fail. It may even be true that scientists gave up the idea too quickly because they did not try hard enough to make it work - to find ways of making the iron stay in place longer, for example.
What if the experiments succeed? Or what if some other idea, such as farming seaweed on a large scale, seems more promising? Of course, the uncertainties of tinkering are enormous, but so are those of doing nothing. It is true that the safest solution is to reverse course and decrease the production of CO2 and other gases, but suppose that turns out to be politically impossible or economically unjustifiable, as seems likely. The question is whether we will then be inhibited by a reluctance to risk harm through action, like those who will not vaccinate children even though the risk of vaccination is smaller than the risk of the disease. Such inhibition could be exacerbated by the belief that the solution in question is not ideal, even though the ideal is impossible. This was a problem for fishermen too.
Here the problem is particularly acute because the tinkering is with nature. Reluctance to tinker with nature is a strong intuition, little understood by those who are in its grip - or by anyone. It is widespread in modern society: many people oppose any attempt to modify the weather; many others strongly prefer to drink a glass of natural spring water rather than a glass of distilled water with chemicals added to make it chemically identical to the spring water. The same consequences are evaluated differently depending on whether people bring them about: ``I don't mind natural extinctions, but I'm not too enthusiastic about extinctions that are directly caused by man. I feel that a species has a right to survive and be able to survive on its own and be able to change and evolve without the influence of whatever man does. I don't want to see man kill [any species]. If it's going to happen, it should happen naturally, not through anything that man has an influence on.''
The Geritol solution to global warming is one of a host of proposals for solving environmental problems through active manipulation of nature. Alfred Wong has proposed that about 20 panels made of zinc, each the size of a football field, carried into the stratosphere by balloons, could inactivate the chlorine atoms from CFCs that are the root cause of ozone loss. Happily, for those not wanting to tamper with nature, this too was shown to be a bad idea. The idea that what is natural is good can influence the debate about global warming in more subtle ways, as well. For example, some of the debate about global warming is about the extent to which it is caused by human activity, as if it must be benign if it were naturally caused. In the long run, what matters are its effects, whatever its causes.
This bias toward the natural is puzzling because of its haphazardness and inconsistency: people have been tinkering with nature throughout history, through fishing, hun