To restate an important point, systems of belief such as cultural and ethical relativism are commonly seen as huge advances in moral thinking. To be sure, it is refreshing that we no longer stone adulterers to death. Gay and Lesbian activists no doubt welcome the slow move to tolerance for their orientation. And religious groups are well served by the increasing pluralism of our times: Nobody wants to be burned at the stake, or denied the rights to liberty and property because their beliefs are outside the norm.
However, in the sections to follow, I will argue that we should not confuse tolerance with relativism, that tolerance itself is a kind of “absolute” that seems to contradict the very foundations of relativism, and finally that judgments about human behavior—both epistemic and eventually normative—are inevitable and desirable if we must “act” in any way at all, and especially if we must act to make the world better.
Before I do this, however, I want to pre-empt the argument that advocating or enforcing particular moral norms is a remnant of totalitarianism, stone-age intolerance, or just plain “fussiness.” My argument is that the belief that certain things are morally right or wrong, desirable or undesirable need not cause us to behave in ways that are violent or unjust. True, some conflict is inevitable, and sometimes the majority can justifiably enforce its will on a minority, particularly when the minority is engaged in actions which cause direct harm to the majority (It works the other way as well: The majority cannot cause undue harm to the minority in a society containing a genuine rule of law.) But by and large, there is a demonstrable difference between advocating a moral view and violently enforcing that view.
For example: Many people in the elections of 2000 thought Ralph Nader a moralizing, absolutist stick-in-the-mud. Nader spoke in absolutes; he called corporate practices “evil” and clearly advanced an agenda that he thought to be both pragmatically, and morally, correct. In fact, accusations of his absolutism were the most effective forms of attack against Nader. “He is too absolutist,” people would say. “He could never be a good leader.”
However, the fact of Nader’s moral absolutism, even granting the possibility that such absolutism precluded him from being an effective leader in America’s political system, did not make him somehow “evil” or “unethical.” It is in such cases that the relativist conflates the discursive “violence” of moralizing with the physical or material violence, which comes when I strike you, shoot you, or slap you because I disagree with you. There is a clear difference between “rhetorical” or “discursive” violence (“You are a bad person. I hate you”) and material violence. Those who engage in absolutism in the public forum, but do not threaten people with force or the loss of freedom, are not violent totalitarians, and so needn’t scare those concerned with peaceful pluralism. In sum, it is possible, I think, to believe in moral absolutes and not be “intolerant” or totalitarian.
The inevitability of judgment
Having pre-empted the usual relativist appeal to fear, now I will explain why judgments are inevitable. By “inevitable,” I really mean two different things: First, judgments are inevitable in that our very linguistic, cognitive, and structural underpinnings require us to judge. The second sense of “inevitable” is more ethical in nature; I will argue that if humans are to act collectively to improve their world, then judging our actions, and the actions of other communities and individuals, is unavoidable, and—given that we can do so nonviolently—desirable.
Cognitively, I cannot help but make judgmental pronouncements. This is where the illogical nature of the relativist argument comes into play. Consider the following formula in favor of relativism:
All statements which judge the desirability or undesirability of particular actions are incoherent.
Premise (1) above is such a statement.
Thus, premise (1) is incoherent.
The obvious contradiction extends beyond the epistemic nature of the above formula. It also applies to the ethical assumptions of relativism. Consider:
All statements which judge the desirability or undesirability of particular actions are unethical statements.
Unethical statements ought to be rejected.
Premise (1) above is such a statement.
Thus, premise (1) ought to be rejected.
It is, therefore, cognitively impossible to sustain the notion that we can “never judge.” The relativist, however, will reply that such formulas do not really do justice to the relativist position. Instead, it will be argued that the relativist doesn’t even make either of the first premises above.
The simple and indisputable fact of the matter is that it is impossible for us not to “judge”—if by “judgment” we mean the cognitive act of assigning meaning to the things around us. If judgment (including judgment of the desirability or undesirability of a particular act) is inevitable, then the question is not whether or not we should judge. The question becomes: HOW should we judge? What criteria should we use? Should our judgments be collective or individual? Should they be charitable or tough-minded?
None of those questions preclude tolerance. But they do require us to admit that we will always, individually or collectively, see certain acts as desirable or undesirable, and whether we choose to call them “good,” “evil,” “smart,” “stupid,” or whatever, we ought to be fully aware of their inevitable role in our personal and political lives.
Relativists argue that “values cannot be universal,” since they vary according to history, culture, or individual preference. Relativists see universalization as the greatest conceivable evil, as an act of philosophical imperialism.
But just as it is impossible not to “judge” at some level, it is similarly impossible not to universalize. This is both cognitively and pragmatically true. It is cognitively true because humans share too many cognitive similarities not to universalize in their very linguistic and logical constructions of reality. It is pragmatically true because humans are not solitary actors: We form communities in order to solve problems or improve our lives, and in so doing, we “universalize” our goals and objectives, even our values. I will address both the cognitive and the practical inevitability of universalization presently.
Cognitively, our very linguistic and thought structures demand that we “universalize” our concepts. Linguistic philosophers such as Noam Chomsky have shown that we have faculties which are, for all practical purposes, “innate,” in that we all learn the same way. If this is true, then it seems that, insofar as we agree about more than we disagree about, or insofar as we see the world in much the same way (for all practical purposes), the very act of sharing knowledge is an act of “universalizing” (http://www.britannica.com/seo/n/noam-chomsky/).
But it is in the pragmatic sphere that we see the act of universalizing values. In order to overcome world hunger, people must first agree that hunger and malnutrition are bad things. Although there may be some disagreement, say, from people who believe in the Malthusian thesis that we ought to let people die now to prevent greater death in the future, even these objections assume the same value judgments and vocabulary as those who reject starvation at the outset.
Similarly, anthropologists agree that every culture has prohibitions against various transgressions. As philosopher Joseph Grcic writes, these “core beliefs” might vary in some local manifestations, but it contains more similarities than differences. He writes: “This core consists of: 1) prohibition of murder or the killing of in-group members except within parameters specified in the group (e.g. as punishment, self-defense, or other socially accepted rituals); 2) prohibition of random bodily violence, harm or insults (harm to prestige or self-esteem); 3) rules requiring some degree of work from the able bodied to meet survival needs; 4) a prohibition of theft and establishment of some level of private property; 5) rules requiring some level of care for others, especially infants, the old and infirm; 6) knowledge is valued at least insofar as assisting in the provision of food, shelter and healing illness; 7) truth telling and promise keeping are generally valued except in specific cases; 8) the encouragement of some form of marriage and mating where sexual needs are met, reproduction and nurture of children take place; 9) some restrictions on sexual intercourse with the rule against incest most universal” (Joseph Grcic, DIALOGOS 74, 1999, pp. 125-6).
There are, as I said, pragmatic reasons for these universals: Societies that reject prohibitions against murder will eventually no longer be societies. Societies that contain no provisions for helping their weakest members will not, in some Nietzschian fashion, “become stronger,” but will in fact die out as more and more members fall through the cracks, and as the diversity and strength necessary to face unexpected social challenges dissolves.
I hope that this section has especially shown that “moral universals” do not require us to believe in those very different and absolutist moral codes which usually form the foundational justification for moral doctrine. Instead, whatever their various metaphysical underpinnings, societies universalize their values because they need to in order to survive.
Finally, none of this suggests that differences do not exist, or should be swept under the rug of public discourse. People agree, disagree, and agree to disagree, but in those very acts of public discourse which define and refine social values, the inevitable fact of universalization remains.
Share with your friends: |