The Enigma of Reason: A New Theory of Human Understanding
Page 32
Argumentation in East Asia
On March 11, 2011, one of the strongest recorded earthquakes wreaked havoc in Japan’s Tōhoku region. The Fukushima Daiichi power plant was badly damaged, leading to the worst nuclear accident in the world since the 1986 Chernobyl accident in Ukraine. The Japanese political world was badly shaken, as people discovered glaring gaps in the regulatory process supposed to keep nuclear power plants safe.
The aptly nicknamed “nuclear power village” was one of the culprits.15 In this gathering of private companies and regulatory agencies, peace reigned. And that was the problem. For the Japanese scholar Takeshi Suzuki, the village was governed by kotodama—a belief in the mystical power of words.16 By making taboo the words that can lead to disputes, kotodama “serve[s] a social role to discourage or hinder argumentation.” More generally, because of the intense pressure to maintain social harmony, “the Japanese are not trained to argue and reason.” Japanese culture would have precluded the members of the nuclear power village to discuss the dangers of nuclear energy, leading them instead to construct a “safety myth of nuclear power plants.”17
Suzuki is not the first to deplore the “lack of argumentation and debate in the Far-East,” to quote the title of an article by Carl Becker.18 Becker lays the blame for this lacuna on age-old cultural precepts prevalent in Eastern cultures, encapsulated in such dictums as “Eloquence … is nothing but hell-producing karma” (from the Rinzai school of Zen). Other authors suggest that East Asian languages are poorly equipped to deal with logic and argument. In his wide-ranging comparison of Eastern cultures, Hajime Nakamura noted the vagueness inherent in the Chinese and the Japanese languages and the obstacle this vagueness sets to “expressing logical conceptions.”19
Because they are often singled out for the opprobrium they cast on argumentation, East Asian cultures offer a good case study for our claim that some traits of argumentation (and to begin with, obviously, its very presence) are universal. Do people in East Asian cultures argue and benefit from doing so? Do they even have the option of properly arguing, given the alleged vagueness of their languages? The answer is a resounding yes. East Asian languages did not stop the development of logic and rhetoric. In the fifth century BCE, the Chinese Mohists created logical systems as complex and abstruse as those of Western scholastics. Two hundred years later, the rhetoric of the great Han Feizi was so rich that no Roman rhetor could have matched the “subtlety of [its] psychological analysis.”20
What about the alleged cultural taboos against argumentation, castigated as an egotistical breach of social harmony? Ironically, for all their proclamations against eloquence and argumentation, East Asian intellectuals never stopped confronting one another. Confucius’s Analects state that “the superior man is slow to speak but quick to act.” This did not stop Confucianists from writing treatise after treatise “reacting to criticisms of opponents” and “engaging in philosophical debate with rival doctrines.”21 The history of Daoism is equally full of arguments, in spite of the Tao Te Ching’s assertion that “a good man does not argue; he who argues is not a good man.”22
Berating argumentation didn’t stop East Asian intellectuals from arguing. No big surprise here. Arguing is what intellectuals do for a living. Are other people more inclined to respect the precepts erected—and flouted—by the wise men? Hardly. Recent historical work has revealed that even a country with such tight social control as Japan “is not the ‘relatively peaceful’ arhetorical society described by Becker and others, but a country whose past three hundred years have been marked by great ideological and often physical conflict, and whose disputes have often been conducted and recorded in the form of debates.”23
Even early Japanese tradition reflected an understanding of the benefits of group discussion. At the dawn of the seventh century, Prince Umayado set out a new constitution. Its last article reads:
Decisions on important matters should not be made by one person alone. They should be discussed with many. But small matters are of less consequence. It is unnecessary to consult a number of people. It is only in the case of the discussion of weighty affairs, when there is a suspicion that they may miscarry, that one should arrange matters in concert with others, so as to arrive at the right conclusion.24
What little experimental evidence there is regarding the benefits of argumentation in East Asian cultures converges with the results obtained in the West. When Japanese students are given the Wason four-card selection task to solve on their own, they perform as badly as their American or European counterparts. But when the same students must discuss the answer in small groups, they enjoy the benefits of argumentation, and most groups converge on the right answer.25
The scorn that Japanese culture supposedly displays toward argumentation is not to be blamed for what happened in Fukushima. It was a run-of-the-mill case of regulatory capture. Through more or less direct forms of bribery, regulatory agencies can come to serve the private interests they are supposed to regulate. In Japan, civil servants who play along with companies can expect to be offered cushy jobs upon retirement—a practice called amakudari. Other countries can do it differently, but they also suffer from regulatory capture. Let’s not blame Rinzai Zen for a sadly widespread political practice.
Argumentation in Small-Scale Societies
For millennia, Eastern and Western cultures have relied on writing to develop complex rhetorical traditions and argumentation-centered institutions. The human species did not evolve in such a culturally sophisticated context. The environment of our ancestors was closer to that of Luria’s unschooled peasants, and even closer to the conditions in which modern-day hunter-gatherers live. Even if we accept that everyone, schooled or unschooled, can reason, that doesn’t mean that everyone argues.
Of the two classic depictions of the original state of humanity—the “noble savage” and Thomas Hobbes’s “war of all against all”—neither offers a good context for argumentation. The noble savage would not have had the motivation to engage in argumentation. The war of all against all offers even fewer opportunities for debates to flourish.
These views are nowhere near the truth. Our ancestors were neither living in harmony with one another nor waging constant war against one another. And argumentation may have played at least as important a role in their social lives as in in ours. When a collective decision has to be made in a modern democracy, people go to the voting booth. Our ancestors sat down and argued—at least if present-day small-scale societies are any guide to the past. In most such societies across the globe, when a grave problem threatens the group—ecological crisis, war, protection of common resources—people gather, debate, and work out a solution that most find satisfying.26
Even the most egalitarian of our modern societies looks quite hierarchical compared with a typical small-scale society. Hunter-gatherer societies have no king, no general, no manager to boss people around. If some members have more influence, it is not because they are invested with some supernatural or birthright authority but because of the services they render to the community, for instance, in hunting, in war, and in coming through discussion to good collective decisions. This generalization holds even when a rigorous chain of command might be expected—as in societies plagued with constant warfare.
In the many years anthropologist Napoleon Chagnon spent among the Amazonian tribe of the Yanomamö, he witnessed so much conflict that he attributed about 50 percent of adult male deaths to violence.27 Yet this permanent state of conflict did not give rise to a strong hierarchical structure. When K%obawä, a village headman, wanted to be heard, he couldn’t simply raise his voice or threaten with his club. He had to rely on argumentation. As Chagnon reports, “should someone be planning to do something potentially dangerous, [K%obawä] simply points out the danger.” Despite his position, “he so diplomatically exerts his influence that the others are not offended.”28
We could pile up examples of sophisticated argumentative practices in small-scale societies. In
the Trobriand Islands, Edwin Hutchins reported convoluted legal argumentation and complex chains of reasoning.29 Among the Lozi of Zambia, Max Gluckman discovered a culture focused on debates with a rich vocabulary to describe the quality (or lack thereof) of someone’s argumentation:
kuyungula—to speak on matters without coming to the point
kunjongoloka—to wander away from the subject when speaking
kubulela siweko—to talk without understanding
muyauluki—a judge who speaks without touching on the important points at issue
siswasiwa—a person who gets entangled in words
siyambutuki—a talker at random30
To show that argumentation can be as effective in traditional, small-scale societies as it is in ours, Thomas Castelain visited remote groups of K’iché Maya in rural Guatemala, people who practice subsistence farming and, in most cases, can neither read nor write, and only speak their native language.31
With local help, he asked K’iché participants to solve so-called conservation tasks on their own. These tasks require understanding that a given property of an object is conserved across some changes. In the present case, participants were presented with two glasses containing the same amount of water and asked in which glass the water would rise more: in the first glass, in which a ball of Play-Doh was about to be plunged, or in the second glass, in which the two halves of a ball of Play-Doh of identical size would be plunged.
Only a third of the participants answered that the water would rise as much in both glasses. But when they had to discuss the task in small groups, over 70 percent got the right answer. In fact, as soon as a participant had the correct answer, she was nearly always able to convince other group members to change their mind—exactly what had been repeatedly observed in WEIRD cultures.
How to Reconcile Evolutionary, Cognitive, and Anthropological Perspectives on Reasoning
We have focused on the universal traits of reasoning and argumentation, debunking the idea that schooling is necessary to understand simple reasons or that, in some cultures, argumentation might be suppressed or even never develop. Reasoning and argumentation are found everywhere, as we should expect if reason is an evolved module and if the production and evaluation of argument is one of its two main functions.
To say that reason is a universal mechanism does not imply that it works in exactly the same manner in all places or that specific societies can affect reason only in superficial, merely cosmetic ways. Reason is deployed in a variety of cultural practices and institutions that may inhibit some of its uses, enhance others, and provide socially developed cognitive tools that complement and extend naturally evolved abilities.
In all human societies, for instance, there are rules and institutions for resolving issues of rights, going from private disputes to criminal cases. The way these issues are being argued varies greatly across cultures. In many small-scale societies, there is no court system; the parties make their case without the help of lawyers. Elders, local assemblies, or political leaders play the role of arbiter or judge. Still, institutional forms of epistemic vigilance may play a role in these proceedings. There may be, for instance, culturally developed forms of vigilance toward the source such as oaths or even ordeals, which are believed to deter lying. Vigilance toward the content, on the other hand, typically relies just on commonsense production and evaluation of arguments. There are typically no rules regarding admissible evidence and no standard of proof. In larger societies with a state organization, by contrast, arbitration, litigation, and criminal justice are in the hands of complex institutions and obey a whole range of precise rules (with much cultural variation).
One of the most famous criminal trials in recent history, that of the American football star O. J. Simpson, provides a striking illustration of the degree to which legally regimented argumentation may depart from commonsense reasoning. Simpson was accused of having murdered his ex-wife and a friend of hers. In October 1995, at the end of an eleven-month trial where his lawyers argued that he was not involved in any way in the murders, he was acquitted. The families of the two victims, however, filed a civil suit against him for “wrongful death.” Simpson lost this civil trial, and in February 1997, he was ordered to pay $33.5 million in damages for the two deaths the jury concluded he had caused.
To people unfamiliar with the American legal system, this may look like blatant incoherence: how can the same man be acquitted of a double murder and yet made to pay compensation for it? The key consideration here is the difference in the standards of proof that apply in a criminal and in a civil trial. In the criminal trial, the standard being “proof beyond reasonable doubt,” Simpson’s lawyers had argued that there was indeed reasonable doubt and won. In the civil trial, the standard being “clear and convincing evidence,” the civil parties pleaded that there was such evidence and won against Simpson. Even so, most people, even in the United States, found it hard to make any intuitive sense of this double verdict. Legal scholars, on the other hand, could easily find reasons for these two divergent decisions by reasoning at a more abstract level: there are good reasons for different standards of proofs in a criminal trial (where you might sentence someone to prison or even death) and a civil trial (where what is at stake is merely money). Legal arguments in O. J. Simpson’s case may well have been at odds with commonsense reasoning. Still, like all arguments, these arguments were ultimately rooted in intuitions about higher-order reasons.
The use of argumentation for resolving conflicting interpretations of events and rights fits quite well the interactionist approach to reason and the argumentative theory of reasoning. Not all culturally developed forms of reasoning provide such obvious illustration of the approach. We mentioned in Chapter 9 the puzzles, riddles, paradoxes, and other brain twisters that in some cultures are produced as stimuli for the pleasure of solitary reasoners. How is this supposed to fit with the interactionist approach? Some cases of argumentative interaction do not fit the approach in any obvious way, either—consider the debating societies such as the Oxford Union, founded in 1823 and currently the world champion of competitive debating. In competitive debates, two or more teams try to best each other through arguments, arguing for a point of view that they have been arbitrarily assigned to defend, and on a topic the audience might not care much about, such as “Should states construct false historical narratives that promote social cohesion?” and “There is a potion which can stop you falling in love. As an eighteen-year-old, should you take the potion?”32 Why should people bother to argue for opinions they do not hold on issues indifferent to them? Why should people pay any attention to such exchanges of arguments?
We argued in Chapter 10 that reasoning has a double argumentative function: for a communicator, reasoning is a means to produce arguments in order to convince a vigilant audience; for the audience, reasoning is a means to evaluate these arguments and accept them when good, or reject them when bad. In the case of solitary reasoning on puzzles and enigmas pursued as a leisure activity, producing arguments to convince others or evaluating others’ arguments needn’t play any role. In the case of competitive debating, arguments are produced not in order to convince an audience of the truth of their conclusion but in order to convince a jury of one’s argumentative skills.
These interesting cases, however, do not present to the argumentative theory of reasoning a greater challenge than do recreational sex, masturbation, and pornography to the claim that the main function of sex is reproduction. Some evolutionary considerations should help make the point.
Any evolved mechanism is adapted to the environmental conditions in which it has evolved and may malfunction or produce nonfunctional effects in different conditions. Breathing more carbon dioxide than is found in normal air may lead to suffocation. Breathing more oxygen may cause euphoria. These abnormal effects do not challenge the standard view that the breathing mechanisms of mammals are well adapted to their function of delivering oxygen to the body and removing carbon dioxide
as needed.
Evolved cognitive modules are typically adapted to processing information belonging to a given domain and to drawing specific inferences from it. We called such a domain the “proper domain” of a cognitive module.33 In the proper domain of a snake avoidance module, for instance, are snakes present in the environment. In the proper domain of a mindreading module are the mental states of people with whom the individual is or might be interacting. The operations of such cognitive modules are triggered by input information provided by other modules, lower-level perception modules in the case of the snake module, modules processing information about, for instance, the behavior, speech, or visual expression of others for the mindreading module.
Inputs that trigger the operation of a module are imperfect. They do not pick out all and only cases at hand that fall within the proper domain of the module. There is a cost in time and energy to detecting what belongs to the proper domain of a module. This cost makes it more efficient to accept a certain rate of detection errors. The operations of most modules are, in fact, triggered by simple diagnostic criteria rather than by a complex pondering of a variety of factors (a pondering that might need a cognitive mechanism of its own and might not be that efficient anyhow).34 Often, the trigger is oversensitive—think of the trigger of jealousy or of danger detection—but this oversensitivity may well be adaptive.
The range of inputs that actually activates a module is its actual domain. There is no way the actual domain would exactly correspond to the proper domain. Mistakes in detection are unavoidable and result in either false positives (false alarms) or false negatives. As we noted in Chapter 11, depending on the relative cost of false negatives (such as mistaking a snake for a piece of wood) and false positives (mistaking a piece of wood for a snake), an efficient module may well be biased toward making the less costly type of error so as to avoid as much as possible ever making the more costly one. In such cases, the mismatch between the proper and the actual domain of a module is in fact advantageous.