To illustrate social universals, all societies are structured by statuses and roles and possess a division of labor. These phenomena lie near the core of the social realm, which consists in essence of social statuses and their relationships but also of the interrelationships between the individuals (who “inhabit” statuses). Although individuals are normally affected by them, anthropologists probably think of social universals most frequently as traits or complexes attached to and defining collectivities rather than individuals. But certain mental and behavioral mechanisms, present in all normal individuals, are undoubtedly also involved in human sociality. Since statuses are also cultural, and generally possess linguistic labels, social phenomena touch upon the individual at one end and culture and language at the other.
Although most anthropologists are comfortable talking about societies in the plural, as though they were discrete and countable entities, attempts to specify the boundaries of societies pose many difficulties. There are, for example, societies within societies, and there are individuals who belong to more than one society at a time. In important senses, societies transmute themselves from one into another as time passes. (I will continue to write about “peoples” and “societies” in spite of these difficulties.)
Culture consists of the conventional patterns of thought, activity, and artifact that are passed on from generation to generation in a manner that is generally assumed to involve learning rather than specific genetic programming. Besides being transmitted “vertically” from generation to generation, culture may also be transmitted “horizontally” between individuals and collectivities. Examples of culture are tools, kinship terminologies, and worldviews—which in each case may take distinct forms among peoples who are genetically indistinguishable. Culture is divisible into “traits” (single items) and “complexes” (more or less integrated collections of traits) and typically is thought of as though it were attached to collectivities rather than isolated individuals. This deemphasis of the individual stems not from an anthropological belief that individuals do not create culture but from the observation that any given individual receives much more culture than he or she creates. Because so much culture is imposed upon rather than created by any particular individual, anthropologists (and others) often think of culture as a sort of supraindividual entity in itself, or as something dictated by that supraindividual entity called “society.”
Since conventional social arrangements are by definition a part of culture, this further confounds the social and the cultural. There are, thus, many contexts in which “social” and “cultural” are used interchangeably (particularly when stressing a contrast with things “biological”). Furthermore, anthropologists often use “a culture” and “a society” as synonyms, stressing by the former the entire collection of cultural traits and complexes associated with a particular society or people. Some parts of culture are not very profitably understood in social terms—some items of material culture, for example—and it is these nonsocial elements that are understood when one contrasts culture and society.
Because any language possesses many conventional traits that are transmitted within populations much as culture is transmitted, and because the lexicon of any language has a close relationship to the culture of its speakers, language has often been thought of as closely related to or even a very important part of culture. At any rate, all peoples use one or more particular languages; all languages have phonemes, morphemes, and syntax1; and each of these aspects of the structure of language contains further universal elements. In terms of its structure, language is normally understood as a more or less closed system whose parts are defined or understood in relation to one another; few parts of any particular language are normally understood as phenomena connected to the individual. But social factors are often important in understanding the syntax and semantics of particular languages, and some features of language (such as the “marking” that will be discussed later) can only be understood when it is borne in mind that individual organisms employ and shape it. Thus sociolinguistics and psycholinguistics have emerged as subdisciplines within, or adjuncts of, linguistics.
Although anthropology literally means the “study of humanity” in the widest sense, twentieth-century anthropology—at least in the English-speaking world—has been primarily concerned with the social and cultural aspects of human affairs. As I noted earlier, human anatomy and physiology, identified as “biological” in contradistinction to the “social” and “cultural” realms, have been left to other disciplines or given a somewhat marginal existence within anthropology. Moreover, as a matter of practice in anthropology—at least until the 1960s—if a universal was to be of interest it usually could not be too obviously or completely determined by human biology, nor could it be a “lower” mental function; “higher” mental functions have been a bit more acceptable as topics of anthropological concern. What is probably the most famous list of universals in anthropology—George P. Murdock’s list published in 1945 (quoted and discussed in the next chapter)—is specifically a list of cultural universals.
All humans breathe, and yet no list of universals mentions the fact: the linkage to biology is too close, the sociocultural influence too negligible, the distance from “higher” mental functions too great. This is not to say that anthropologists are never professionally interested in breathing: the effects on the human body of life at high altitude—generally resulting in increased lung capacity—is, for example, an area of research for physical anthropologists. But most of the remaining anthropologists would find breathing of interest only to the extent that it is modified by social or cultural conditions (as in yogic practices).
To explore a more complicated example, sexual activity is, in some senses, as physical as breathing. Sexual activity occurs in all societies, but it too does not appear on most lists of universals. Unlike breathing, however, sexual activities show many obvious social and cultural modifications, and the anthropological literature on human sexuality is considerable. From the viewpoint of anthropological interest in universals, however, it has not been sex itself that has been of most interest (actual descriptions of coitus are quite rare), but certain phenomena associated with it, such as incest regulations and male-female differences in temperament and behavior.
Universals may in fact be linked to human biology—and this is sometimes stated as an anthropological assumption (e.g., Sahlins 1976; Shepher 1983; and discussions in subsequent chapters). But to judge by much of the existing practice, universals must not be so closely linked to biology that there is nothing left to say about them from the perspective of social and cultural anthropology.
For a considerable period the term “universal” was used without anyone thinking it needed to be defined. During that period the implicit definition was approximately as follows: a trait or complex present in all individuals (or all individuals of a particular sex and age range), all societies, all cultures, or all languages—provided that the trait or complex is not too obviously anatomical or physiological or too remote from the higher mental functions.2
I repeat that this is a definition that fits much existing practice; it may not prove to be useful in the future. One of the reasons why it will probably require revision is the difficulty of making useful distinctions between biology and culture. I write at, and am a product of, a time when the distinction remains fundamental to most anthropologists—even though it is vaguely and falsely conceived. Nothing in human culture comes into being or gets transmitted without consideration of the specifically human genetic makeup. Yet significant aspects of human anatomy and physiology can only be fully understood with some consideration of human culture, which always and everywhere is a crucial part of the environment that interacts with human genes to produce human organisms. Any hypothetically conceived boundary between the “thoroughly genetically determined” and the “not too obviously biological” is more likely to be a boundary between what has and what has not been interesting to anthropologists. An a
nthropology less concerned with the opposition of culture to biology, and more concerned with their interaction, may well arrive at many new conceptions, including new conceptions of universals.
For example, consider the views of two of the most important present-day contributors to understanding universals, Noam Chomsky in linguistics and Robin Fox in anthropology. Both distinguish “substantive” universals (Fox 1989:113; Chomsky 1965:27–30; Chomsky and Halle 1968:4), which are what anthropologists usually mean by universals, from universals at a deeper level. For Chomsky these are “formal”; for Fox they are universals at the level of “process” (examples will be given below). Both Chomsky and Fox assume or find that these deeper and more significant universals, which on the “surface” do not necessarily or even typically manifest themselves in substantive universals, are rooted in human neurobiology. Fox (1989) goes on to argue that the distinctions between the individual, society, and culture are entirely artificial and a barrier to the development of social science.
But, as I said, I write at a time when these distinctions are still fundamental to the way social scientists think. Furthermore, my discussion follows anthropological practice in giving serious weight to substantive universals—even though it does not exclude formal or processual universals that may lie at deeper levels.
In addition to the distinctions Chomsky and Fox make between substantive and deeper universals, there are quite a number of other kinds of universals that must be examined. To begin with, some anthropologists draw attention to the distinction between universals and “near universals,” generally to argue that the distinction is not important.3 Various lines of reasoning support this position. One is that, given the quality of ethnographic reporting, the distinction could reflect error of reporting. We know for certain that many of what we expect to be universals could not be shown to be present in all societies from the ethnographic record as it presently stands. But the reason the trait or complex appears to be absent in some society or societies is either that the record is silent on the matter or that the record is wrong. If, for example, the rate of murder were a constant in all societies, those of very small scale might not have a murder in several generations. Under these circumstances the members of the society might very well say that they never murder, and the anthropologist might find no contrary evidence. But both anthropologist and native could be wrong in saying murders don’t occur in the society in question. As a concrete example, Roberts and Sutton-Smith (1962:167) originally concluded from a broad-ranging study of games that they were not universal. But in a later publication (Sutton-Smith and Roberts 1981:437), they decide that the adequacy of the reports of societies without competitive games should be “questioned seriously.”
Another line of reasoning is that human behavior is so complex and malleable that any trait or complex can only approach universality. Thus men nearly universally find lighter skin pigmentation attractive in women (van den Berghe 1986). Van den Berghe provides an evolutionary explanation for this preference, and yet Western societies—where tanned skin has had an appeal—indicate that it is not universal. Presumably, a countervailing tendency—to admire signs of high status, such as tanning indicated—can override a tendency for men to prefer lighter-skinned women. This line of reason presumes that there are so many human tendencies (each of which, by the way, may be a universal at some deep level) that few if any can override all the others to manifest itself as a (substantive) and absolute universal. Even if a trait or complex had a universal distribution, the argument goes, we can imagine the conditions (unusual though they might be) that would eliminate it in some particular case.
Closely related is the argument that universals lie at the end of a continuum: traits and complexes can be scaled all the way from those that are unique to a particular individual, society, culture, or language up to those found everywhere. The distinction between a near-universal and a universal (or absolute universal), then, is not significant, or imposes an artificial break in an unbroken natural continuum. A near-universal is universal enough.
The dog, for example, was absent from some cultures, probably less than 5 percent of those known to ethnography. If, however, it had spread everywhere, what difference would it make in our understanding the dog-human relationship? If some people really did get along without fire—and it is possible that some branches of early Homo sapiens did not have it—we wouldn’t understand the uses of fire or its apparent universality any the less. It is certainly fair to say that many universals and near-universals are likely to have very similar explanations.
No one seems yet to have thought about where to place the cutoff that distinguishes near-universals from merely widespread traits and complexes. Perhaps a cutoff at a 95 percent distribution, by analogy with the 5 percent rule for statistical significance, might make sense in some cases. In the case of the distribution of cultural traits or complexes, such as the domestic dog, it would.
A trait or complex more widespread than chance alone can account for is called a “statistical universal” (Greenberg 1975:78). The near-universal is, in a sense, an extreme form of the statistical universal. A remarkable example of a statistical universal is the use of words with meanings closely related to “little person” to label the pupil of the eye—as indeed is the case in English. This occurs in approximately one-third of all world languages (Brown and Witkowski 1981)—far beyond what is expected, given that the alternatives are limitless. Words for small animals are disproportionately represented as the sources of words for muscles—again English is an example since “muscle” is from the Latin for mouse, and of course “calf” is a further example. Some such label for muscles occurs in almost 20 percent of a sample of world languages (Brown and Witkowski 1981). The obvious reason for “pupil” and its semantic analogues is that close scrutiny of the pupil reveals a “little person” looking out at you: your own reflection. The apparent reason for “muscle” and its analogues lies in the similarity of motions: small animals dart about in a manner analogous to the motion of muscles under the skin.
Another conception of universals distinguishes “implicational” or “conditional” universals from “unrestricted” or “non-conditional” universals (Greenberg 1966; 1975:77–78). An implicational universal is a trait or complex that always appears when certain conditions obtain. It takes the form “if A then B,” in which A is not an individual, society, culture, or language and is not itself a universal. It is a rule that is universally applicable. An example is that “all societies possessing paved highways possess centralized government.” Rules of this sort are common in anthropology, and are often convincingly demonstrated. Implicitly they are even more widely used, as Hempel (1942) has shown for historical explanations in general.
But this conception of universals presents a sort of optical illusion in that it illustrates relativity when looked at in one way and universality when viewed from another angle: some peoples have paved highways, but some don’t; some have centralized government, but some don’t (relative statements); yet all that have one have the other (universalistic). When one examines the causes of universals, it is apparent that the relativistic image that results from implicational universals is a surface appearance; fundamentally there is little difference between unrestricted and implicational universals.
Implicational and statistical universals in combination are particularly common. That is, statements of the sort “if A then a tendency to B,” in which A is not an individual, society, culture, or language, probably constitute the single most common form of cross-cultural generalization. That particular kinds of kinship terminologies tend to be found in matrilineal societies while alternative kinship terminologies tend to be found in patrilineal societies are but two of the no doubt hundreds of generalizations of the sort that are familiar in anthropology. If anthropologists typically thought of these as kinds of universals, universalistic thinking would be far more prevalent than it is. But because of the optical illusion mentioned above, and because of
the statistical rather than absolute form of these statements, their kinship with universals is rarely noted.
Generally, I provide little discussion of implicational or statistical universals that are little more than conventional cross-cultural generalizations. But it is important to be aware of the extent to which these kinds of generalizations imply a connection with universals or universalistic perspectives.
An important variant of the implicational universal consists of universal evolutionary sequences. These take various related forms. They may be “if A then B” statements in which it is asserted that A can only emerge after B, as when it is asserted that the locomotive can only emerge after the wheel; or they may be the same form of statement in which it is asserted that both A and B emerge in tandem as the result of some prior factor C. Alternatively, they may take the form of “if at developmental stage A, then trait or complex B will be found.” The various “unilineal” schemes of sociocultural development proposed by nineteenth- and twentieth-century anthropologists are cases in point (a lengthy quotation of A. V. Kidder in the next chapter is a good illustration). The evolutionary stages in which color terminology develops, as noted earlier, are further examples. Closely related sequences have been posited for the development of botanical and zoological life forms (Brown 1977b, 1979).
What is perhaps another variant of the implicational universal posits what might be called a universally fixed “pool” of sociocultural elements, from which all traits or complexes of a given type are formed. Linguists, for example, find that all phonemic systems are based on a finite list of possible speech sounds or contrasts of speech sounds (Jakobson, Fant, and Halle 1967). Any particular language employs a selection from the universally given possibilities. The International Phonetic Alphabet, thus, is meant to have universal applicability.
Human Universals Page 7
