Dark Matter of the Mind

by Daniel L. Everett

  Moving on to the next bowl of porridge, Clifford Geertz (1973, 5) offers a definition of culture that surpasses the previous ones, in my opinion, because it focuses on meaning: “Man is an animal suspended in webs of significance he himself has spun . . . I take culture to be those webs, and the analysis of it to be therefore not an experimental science in search of law but an interpretative one in search of meaning. . . . Once human behavior is seen as . . . symbolic action—the question . . . to ask [of actions] is what their import is.” Geertz’s definition of culture as “webs of meaning” is at least partly correct. Surely, as I argue below, meaning is a vital component of human life that culture furnishes. But meaning is not all there is to culture. Culture also establishes constraints, a variety of socially vital roles, and values, as well as conventions, symbols, and many if not all of the forms—grammatical, lexical, societal—used to encode and decode meaning.

  In spite of its shortcomings, however, Geertz’s definition is particularly interesting to me because it is the first propose that culture is a way of seeing meaning in the world. Yet it places too much emphasis on the symbolic. It ignores, for example, the “grammar of culture” that Pike proposed and that in some form must be part of our understanding of culture. Even grammar itself, although it manipulates symbols, is not itself purely symbolic. Yet without a grammar, we cannot share the meanings of the social and physical world around us that constitute Geertz’s notion of culture.

  Finally, let’s examine Marvin Harris’s (2001, 47ff) definition of culture: “Culture . . . refers to the learned repertory of thoughts and actions exhibited by the members of social groups—repertories transmissible independently of genetic heredity from one generation to the next.” This is another great start. But like the others, Harris’s definition falls short. It seems to make culture little more than our overt, or “exhibited,” behavior. But it is much, much more than this. In fact, taken literally, Harris’s concept of culture lacks an adequate role for dark matter (not unintentionally). Yet in my view, culture is defined primarily through dark matter.

  Another feature absent from previous definitions of culture is culture’s dynamicity, its refusal to remain static and fixed, its variability even within individual members of the culture, where members’ values, roles, and symbols fluctuate frequently.

  In light of the lack of fit of previous definitions of culture to our program here, let’s revisit the definition of culture that was proposed above:

  Culture is an abstract network shaping and connecting social roles, hierarchically structured knowledge domains, and ranked values. Culture is dynamic, shifting, reinterpreted moment by moment. Culture is found only in the bodies (the brain is part of the body) and behaviors of its members. Culture permeates the individual, the community, behaviors, and thinking.

  From this definition, it follows that culture is a relative concept and that people can share culture to a greater or lesser degree, even if they are not members of a well-defined social group or community or geographical area. There are various components to this definition, each subject to a range of (mis)interpretations, so let me unpack it before trying to apply it.

  By “abstract network” I mean that culture is a postulate to help explain what it is that people share when we say that they are members of a particular society. Culture resides exclusively in individuals, though its artifacts, rituals, tools, and so on, provide concrete evidence for it. And no two individuals are exactly alike culturally. There are only idiosyncracies. My next-door neighbor and I may share many similarities, but we are never identical. Yet we are part of a network—people who have some overlapping values or social roles or knowledge structure or all three.

  The more our values, roles, and knowledge structures overlap, the more connections we share and, therefore, the stronger our connection in a cultural network. Thus we can form a generational network, a CEO network, a rap-lovers network, a “Western culture” network, and industrialized society network, and even a Homo sapiens network, so long as we share values, knowledge structures (not merely knowledge), or roles, however slight the overlap.

  This is recognized by many laypeople when they claim that “people are all alike.” We do all share some values. Likewise, the other extreme—represented by cultural relativists—is also right when it claims that no two cultures are alike; no two cultures (or even individuals) share all the same values, all the same social roles, or all the same knowledge structures.

  Values are the assignment of adjectives of morality for the most part (more clarification of values will come directly) to specific actions, entities, thoughts, tools, people, and so on. They are also statements about how things should or should not be. To say, “He is a good man” expresses a value. This can be broken down into finer-grained values such as “He treats his children well” or “He likes stray animals” or “He gave me a ride home” or “He is polite.” Values are also seen in the tools we choose—a bat instead of a gun for home defense, or a machete instead of a hoe for digging vegetables in the garden. They are seen in the use of our time. Value sets are vast and varied. (We will return to them below.)

  “Hierarchical knowledge structures” refers to the idea that human knowledge, at least—perhaps this also applies to other animals—is not an unordered set of ideas or skills. What we know is broken down in various ways according to context. All is structured in relation to all. And this hierarchy inescapably produces a gestalt output.

  Consider playing the guitar. I say “I know how to play the guitar.” I do not describe what I know as “I know where to find the frets as I need them.” Take my knowledge of a specific song, such as the ability to play the peerless classic “Louie Louie.” If you ask me if I know it, I will say yes. If you ask me to perform it, I will. And in this exchange we treat it as a unitary item. But if I were to teach you how to play it, it would be of little use to say, “Here is ‘Louie Louie’” and just run through the entire song. I first need to explain its component parts: “Start with a G chord. Hit it three times. Then play a C chord, likewise striking it three times. Finally, form a D-minor chord, also hitting it three times. Then repeat as many times as you like, in 4/4 time, singing a number of lyrics (which don’t matter because no one knows them).” Then I might show you the melody, the lead guitar part, the drum part, and so on. And each of these subparts is also broken down into component parts. All arranged hierarchically, just like the stories and sentences we utter. The parts as we first hear them are etic experiences. As we internalize them and learn their role in the whole, the parts and the whole together fall within our emic understanding. And only then can we understand that to go beyond the basic performance of a song to “playing it with soul,” as we internalize the song and arrange it among our values, knowledge, roles, and so on.

  Moreover, the song is explicable ultimately by appeal to dark matter. We first learn the chords, guitar lead, and words in small bits. Then we learn to play them together, very gradually, without thinking of them consciously. These small parts—as Gestalt psychologists would have recognized via their ideas developed more than one hundred years ago—are not the song. The song as a whole is not dark matter knowledge. But its component parts become dark matter after they are learned. They were not originally tacit knowledge. But then they undergo the transmogrification from overt to covert knowledge, illustrating the effects of emicization (taking on the insider’s perspective of guitar playing) on access to knowledge.

  “Social roles” describes actions as conforming to a particular node one occupies in an abstract network. Any grouping of people will be defined by its values, the knowledge structures it devolves from and develops, and the expected duties of each of its members by virtue of their membership classification. Whether the group is a college department faculty (each is expected to specialize, and their specialization is one of their roles—e.g., syntactician, Amazonianist, chairperson, administrator, and so on).

  Department chairs in the North American academic
system, China, or the United Kingdom will differ in many of their values, administrative knowledge, and more, but in their roles (independent of what they are called), they share some aspects of administrative knowledge and values. Moreover, when there exist larger homogenizing forces across communities, roles and knowledge will grow even more alike. An example of such forces in modern academics is international accrediting bodies, such as the AACSB (Association to Advance Collegiate Schools of Business) or EQUIS (European Quality Improvement System) for business schools. They grant accreditation to schools for sharing and implementing the agencies’ values.

  No two individuals will share all values, identical value rankings, the same set of acknowledged social roles, and so on. And each individual will regularly update or modify their knowledge, the concepts of social roles, their values, and their rankings. This dynamicity is another reason we say that culture is an abstraction—it is a generalization with considerable smoothing.

  In order to better understand the model of culture I am suggesting, let’s consider how it might solve a problem proposed by Marvin Harris (2006, 23ff):

  Much evidence exists that the cultural information stored in the brain contains contradictory information. For example, in a study of how Americans conceptualize the family, Janet Keller (1992:61–2) recorded these competing “schema”:

  Family members should strive for the good of the whole group

  but

  The good of the individual takes precedence over the good of the whole group.

  Family is permanent

  but

  Family is always in transition.

  . . .

  Family is nurturant

  but

  Family is smothering.

  From such “contradictions,” Harris concludes: “Indeed, from my cultural-materialist perspective, the emphasis on the proposition that ideas guide behavior, but not the reverse, is the mother-error of contemporary anthropological theories” (22).

  Although my remarks here are not intended as a criticism of Harris’s overall theory, they do intend to show that his arguments here have no force. Everyone appears to hold mutually inconsistent values. But the inconsistency is not always real. For example, consider the last contradiction above, “Family is nurturant/family is smothering.”

  Most people would talk about their family in a similar way but not as a contradiction, saying instead something like, “Families are often nurturing/should be nurturing, but unfortunately they are often smothering.” “Nurturing” and “smothering” express value judgments. But because they are values, in my model we need to know how they are ranked. And because all ranked values are violable, such that lower values can be violated by higher-ranked values, it is possible for higher-ranked values to mask the effects of lower-ranked values, as we see in the rankings following this paragraph. Getting back to the terms at hand, the idea of nurturing includes widespread positive judgments in society as a label for a set of more finely grained values, such as {love, assistance, sharing of wisdom, independence, imposition of limits, trust, financial support, judgment, etc.}. Ironically, smothering can follow from the same values, differently ranked (and these are just two of many possible rankings that could derive these two concepts, meaning that we expect similar concepts to be different at the level of dark matter even when producing superficially similar results):

  NURTURING (rough pass): TRUST >> FINANCIAL SUPPORT >> LOVE >> ASSISTANCE >> SHARING OF WISDOM >> IMPOSITION OF LIMITS >> JUDGMENT, ETC.

  SMOTHERING: IMPOSITION OF LIMITS >> LOVE >> JUDGMENT >> SHARING OF WISDOM >> FINANCIAL SUPPORT >> TRUST, ETC.

  The >> follows from the use of this symbol in optimality theory (Prince and Smolensky [1993] 2004) and means that any value to the right of the double arrow can be violated in order to obey a value to the left. Thus a smothering parent may love and trust their offspring, as a nurturing parent does. But love and trust may be ranked differently in each parental behavioral set, leading to very different types of families or parenting. Moreover, as our definition of culture makes clear, different times or circumstances can have their own rankings, such that the same parent can be smothering for one request and nurturing for another from the same child.

  There are many different rankings possible for any set of values. And rankings can vary from individual to individual and even from situation to situation for the same individual or group. A family will contain nurturing and smothering values simultaneously, with the difference being in how the values are ranked for a given family member relative to a particular act (work or play) or person (son or daughter, for example) involved—your mother may be smothering regarding your love life and nurturing regarding your choice of profession, for example. Rankings and roles lead to seeming contradictions but, contra Harris, many ideas, once broken down into their component, violable value hierarchies are not contradictory at all. (Of course, some ideas may turn out to be mutually contradictory; e.g., “The earth is flat, but I think I can sail around it,” but we cannot even conclude this without careful analysis.) It turns out that people are rational and that the notion that ideas control much of behavior makes a good deal of sense. Assume rationality and work backward—from behavior and a theory of culture and the individual—to understand others.

  The same goes for all the other examples Harris uses. For example, he lists considerations in selection of a location for defecation for Hindu farmers in India:

  A spot must be found not too far from the house.

  The spot must provide protection against being seen.

  It must offer an opportunity to see anyone approaching.

  It should be near a source of water for washing.

  It should be upwind of unpleasant odors.

  It must not be in a field with growing crops. (Harris 2006, 24)

  He then observes, “Fulfilling all of these rules on a small farm leads to behavior that violates the rule of fecal avoidance, as evidence by the elevated incidence of hookworm.”

  Again, Harris cites this seeming set of contradictions as evidence against an idea-over-behavior understanding of culture. His own view of culture is also given as: “a culture is the socially learned ways of living found in human societies and that it embraces all aspects of social life, including both thought and behavior” (ibid., 19).

  But neither his criticism nor his definition of culture get us to the conceptual place I think we need to arrive at. First, the criticism doesn’t follow. Assume that the Hindu farmer’s desiderata for dump-taking are in fact ranked, violable values. Assume that “avoid fecal matter” is not the highest ranked of the values (outranked perhaps by “stay close to home”). Then the ranked-values approach accounts easily for the behavior, sans contradiction, in spite of what Harris claims. It also simultaneously shows his definition of culture to be underarticulated and thus inadequate. This is not to claim that all behavior is value- or idea-driven; for example, a baby’s grasping may be gene-driven. But most behaviors probably are value-driven when we take on the concept of violable constraint ranking. The same transition from inviolable, contradictory rules and constraints to violable, ranked constraints (“linguistic values,” from one perspective) is what has made optimality theory (OT) so influential among linguists—it solves the apparently intractable. Neither linguistic ranking nor the violable value ranking I am proposing here are original to the humanities. Ultimately they derive from “Hopfield Nets/Networks” (Hopfield 1982), which in turn emerged from the desire to understand why some physical states are achieved rather than others (why does molten glass dry “smooth as glass,” for example).

  The theory of culture defended here, however, is not simply an application of OT tout court. For example, OT assumes that linguistics constraints are universal—that every person is born with the same set of constraints and they simply have to learn how their particular language ranks the universal set. To capture this, the theory contains a GEN(erate) function to ensure that all languages have the same constraints. The
theory developed here, on the other hand, has no place for universal values, and therefore it has no need of this function. The anthropologist must discover the values and their rankings of each individual in each society.

  VALUES EXPLAINED

  With this brief introduction to my violable value-ranking model of culture, we can consider in more detail just what values are. Values, as I have stated, are fundamental to culture in various ways. They shape cultural forms, group intentions, meanings, aspirations, conventions, and so on. Before discussing their role in more depth, however, it is important to understand what a value is.

  There are several kinds of values, and each has an important role in society and culture, though there are a couple of broad sources. The first kind of value to mention are terminal values. Terminal values include “leading a comfortable life,” “having a sense of accomplishment” (see Rokeach 1973, 160ff), “freedom,” “security,” and so on—they are what society and the individual sanction as laudable goals. Instrumental values, on the other hand, are how we think it is best to achieve our goals; for example, “ambition,” “cleanliness,” “honesty,” “politeness,” “self-control.”

  There are also what I refer to as biological values. These are likely shared in one form or another by all humans, though that is an empirical hypothesis. These include things like self-preservation, not going hungry, being warm, and staying healthy. To the degree that these are universal, they are less interesting for the present discussion. But they are definitely part of our dark matter—more connected with emotions and bodily functions than higher cognition, perhaps. Their satisfaction is essential to our well-being, however, and they can override all other values in some circumstances. So it is well to remember them. At the same time, cultures rank and interpret these differently, so comparison is not trivial. What one considers healthy, another may consider disgusting. What one considers pain to be avoided, another may understand as pain to be sought.