Even if any terminology is a reflection of reality, by its very nature as a terminology it must be a selection of reality, and to this extent it must function also as a deflection of reality.
Burke (1966: 46)
Human beings are restless seekers of meaning—in life, in the universe, in their experiences. They seek it in the same way that they search instinctively for the necessities of physical survival and comfort. In their unquenchable search, they are guided by the remarkable ability to produce, understand, and use signs. Signs have helped human beings fill with meanings the immense void that would otherwise exist between their peculiar states of consciousness and the world. From time immemorial, these have given reassurance that there is continuity, purpose, meaning to life. The capacity for sign-making and sign-use has thus bestowed upon the human species the ability to cope effectively with the crucial aspects of human existence—knowing, behaving purposefully, planning, socializing, and communicating. Culture is a direct outcome of this capacity.
Culture is the system of shared meanings that is based on a signifying order, a complex system of different types of signs that cohere in predictable ways into patterns of representation which individuals and groups can utilize to make or exchange messages. The goal of this chapter is a technical one—to present and discuss the theoretical notions that permit a scientific analysis of this order. These allow the analyst to look systematically at a specific culture as a “container” of signs and their meanings. The goal of theoretical semiotics is to study the origins, nature, and properties of signs, that of cultural semiotics to examine their functions and uses within their “cultural containers.”
The primary objective of semiotics is to understand both the brain’s capacity to make and understand signs, and the knowledge-making activity this capacity allows all human beings to carry out. The capacity is known as semiosis, the activity as representation. Semiosis is the neurobiological capacity itself that underlies the production and comprehension of signs, from simple physiological signals to those that reveal a highly complex symbolism; representation is a deliberate use of signs to probe, classify, and hence know the world.
The difference, but intrinsic interconnection, between semiosis and representation can be seen in early childhood behaviors. When an infant comes into contact with an object, h/er first reaction is to explore it with the senses, i.e. to handle it, taste it, smell it, listen to any sounds it makes, and visually observe its features. This exploratory phase of knowing, or cognizing, an object can therefore be called sensory cognizing, because the child is using the sensory apparatus s/he was born with to cognize the object in terms of how it feels, tastes, smells, etc. The resulting sensory units of knowing apparently allow the child to recognize the same object subsequently without having to examine it over again with h/er sensory system. Now, as the infant grows, s/he starts to engage more and more in semiosic behavior that clearly transcends this sensory cognizing phase; i.e. s/he starts to imitate the sounds an object makes with the vocal cords and to indicate its presence with the index finger. At that point in the child’s development, the object starts to assume a new semiosic form of existence; it has, in effect, been transferred to the physical strategy used by the child to imitate its sound features or indicate its presence. This strategy produces the most basic type of sign, which, as Charles Morris (1938, 1946) suggested, allows the child from that point on to substitute the sign for the object. As mentioned in the previous chapter (§2.1), this is known as displacement. As the child grows, s/he becomes increasingly more adept at using signs to represent the world in a displaced manner. Incidentally, note that the word represent means, literally, “to present again,” i.e. to present some referent again in the sign.
The instant children start to represent the world with signs, they make a vital psychosocial connection between their sensory states to their conscious thoughts about the world. To put it figuratively, signs constitute the “representational glue” that interconnects their bodies, their minds, and the world around them in a holistic fashion. Moreover, once the child discovers that signs are effective tools for thinking, planning, and negotiating meaning with others in certain situations, s/he gains access to the knowledge domain of h/er culture. At first, the child will compare h/er own attempts at representation to the signs to which s/he is exposed in specific contexts. But through protracted exposure and usage, the signs acquired in such contexts will become cognitively dominant in the child, and eventually mediate and regulate h/er thoughts, actions, and behaviors.
The interconnection among the body, the mind, and culture can be shown graphically as follows:
The signifying order thus provides the means for the developing human being to organize the raw information that is processed by h/er senses into meaningful wholes. But as a consequence, the understanding of the world is not a direct one. It is mediated by signs and, thus, by the referential domains that they elicit within mind-space.
Carl Jung, the great Swiss psychoanalyst (chapter 1, §1.2), was fond of recounting how signifying orders had the power to affect even what one sees. During a visit to an island tribal culture that had never been exposed to illustrated magazines, he found that the people of that culture were unable to recognize the photographs in the magazines as visual representations of human beings. To his amazement, he discovered that they perceived them, rather, as smudges on a surface. Jung understood perfectly well, however, that their erroneous interpretation of the photographs was not due to defects of intelligence or eyesight; on the contrary, the tribal members were clear-sighted and highly intelligent. Jung perceptively understood that their primary assumptions were different from his own and from those of individuals living in Western culture, because they had acquired a different signifying order that blocked them from perceiving the pictures as visual signs.
The signifying order can be compared to the default mode of computer software. A computer is formatted in a way that is known as its default mode. This format can, of course, be changed intentionally by a human programmer. But if there are no changes made, the computer will automatically operate according to its original format. Analogously, the signifying order is the human being’s default mode for knowing the world. But in the same way that a human programmer can always choose to change a computer’s format, so too, the individual human being can always decide to alter h/er own “format” at any time. Indeed, therein lies the paradox of the human condition—throughout the life cycle, there is an unexplainable need within each person to transcend the categories of knowing provided by the signifying order. When the categories of the latter fail a human being in h/er search for new or more profound meanings, then s/he can resort to h/er innate capacity for semiosis to alter the default mode. Changes to the signifying order’s format, in fact, are what lead cumulatively to cultural change and evolution. Signifying orders are the products of human beings and, therefore, subject to being changed constantly by them to suit any new need or demand.
The interrelationship among semiosis, representation, and the signifying order can be summarized graphically as follows:
As mentioned, the human infant’s first attempts to know things constitute an instinctive cognizing strategy based on the sensible properties of things (i.e. properties that can be sensed). This clearly serves to enhance recognition of the same things without any further processing of sensory input. But the extraordinary feature of human development can be discerned in the child’s capacity to transcend sensory knowing and to engage in rudimentary representational activities (pointing and making imitative sounds). What feature of the brain endows human beings with this capacity? In our view, the most plausible answer to this question has been formulated by the American semiotician Thomas A. Sebeok, who argues that this capacity is the outcome of the presence of three types of semiosic systems in the human brain that allow for the modeling of experience:
- Primary Modeling System (PMS) = this is the neural system that predisposes the human infant to engage in simulative forms of semiosis.
- Secondary Modeling System (SMS) = this is the more complex neural system that predisposes the human infant to engage in verbal forms of semiosis as s/he develops.
- Tertiary Modeling System (TMS) = this is the highly complex neural system that allows the maturing child to engage in highly abstract forms of semiosis. As Sebeok (1994: 127) remarks, the TMS is “the most creative modeling that Nature has thus far evolved.”
Modeling systems theory is a powerful analytical framework for explaining the etiology of representational systems, behaviors, and activities. The PMS is an innate neural system that endows the human infant who has passed through the sensory cognizing phase with the capacity to represent objects through imitative semiosis. Now, as the child develops cognitively and socially, h/er representational activities become more and more abstract, i.e. progressively more reflective of displacement. This secondary mode of knowing and representing is a concomitant of the brain’s secondary modeling system (SMS). The SMS is a functionally more complex neural system that endows the child who has passed through the PMS phase with the capacity to represent objects primarily through verbal semiosis. Finally, at around 15-20 months the child starts manifesting increasingly the ability to engage in abstract symbolic representation. This is a concomitant of the maturing brain’s tertiary modeling system (TMS). The TMS is a highly complex neural system involving all areas of the cortex, endowing the verbal child with the capacity to know and represent the world in abstract ways.
The two crucial insights of modeling systems theory can be summarized as follows: (1) representation is tied to three semiosic phases; and (2) these phases are evolutionary—i.e. the development of complex symbolic activity (= a TMS endowment) is dependent upon a prior emergence of verbal representational activities (= a SMS endowment) which is itself in turn dependent upon the development of early imitative semiosis (= a PMS endowment).
As discussed in the previous chapter (§2.2), Ferdinand de Saussure and Charles S. Peirce were the two founders of contemporary semiotic science. Saussure’s definition of the sign, in particular, laid down the course that semiotic inquiry was to take in the first half of the twentieth century.
The Saussurean Perspective
In the Cours de linguistique, Saussure defined the sign as something perceivable (i.e. made up of sounds, letters, etc.), which he termed the signifier (= [A] part of the sign), that is used to encode a concept, which he called the signified (= [B] part of the sign). He named the relation that holds between the two signification (= [A ≡ B]).
Saussure considered the connection between the signifier and the signified an arbitrary one. To make his point, he argued that there was no evident reason for using, say, tree or arbre (French) to designate “an arboreal plant.” Indeed, any well-formed signifier could have been used in either language. A well-formed verbal signifier is one that is consistent with the phonological structure of the language in which it is coined (tree is well-formed in English; tbky is not). Saussure did admit, however, that there were some signifiers in a language that were obviously fashioned in imitation of signifieds. Onomatopoeic words (drip, plop, whack, etc.), he granted, did indeed attempt to reflect the sound properties that their signifieds are perceived to have. But Saussure maintained that onomatopoeia was a relatively isolated and infrequent phenomenon. Moreover, its highly variable nature across languages demonstrated to him that even this phenomenon was subject to arbitrary cultural perceptions. For instance, the word used to refer to the sounds made by a rooster is cock-a-doodle-do in English, but chicchirichi (pronounced “keekkeereekee”) in Italian; the word referring to the barking of a dog is bow-wow in English, but ouaoua (pronounced wawa) in French. Saussure suggested that such onomatopoeic creations were only approximate and more or less conventional imitations of perceived sounds.
Many semioticians have differed with this specific part of Saussurean theory. What Saussure seems to have ignored is that even those who do not speak English, Italian, or French will detect an attempt in all the above signifiers to imitate rooster or canine sounds—an attempt constrained by the respective phonological systems of the languages that are, in part, responsible for the different phonic outcomes. Such attempts, in fact, probably went into the making of most words in a language, even though people no longer consciously experience words as physical simulations of their referents. This is because time and constant usage have made people forget the simulative connection between signifier and signified. Consider the word duck. This signifier is indeed one of an infinite number of permissible phonic creations that can be envisioned to encode the signified in English. But it is implausible that duck was created arbitrarily. More than likely, whoever originated that signifier did so in an attempt to simulate the sound s/he perceived ducks to emit. Now, whether or not this is what actually happened is beside the point. The interesting thing to note here is that once people are told about this hypothetical scenario, they start typically to experience the signifier consciously as onomatopoeic, rejecting alternative candidates that could in theory have been chosen to refer to a duck (e.g. glop, jurp, flim, etc.) as somehow “unnatural.” Many semioticians argue that this kind of anecdotal evidence is rather extensive, and therefore that it strongly suggests that sign-creation is hardly an arbitrary, discretionary process, but rather one that is born of simulative primary modeling behavior.
The Peircean Perspective
Peirce called the perceivable part of the sign a representamen (literally “something that does the representing”) and the concept that it encodes the object (literally “something cast outside for observation”). He termed the meaning that someone gets from the sign the interpretant. This is itself a sign (or more accurately a signified in Saussurean terms) in that it entails knowing what a sign means (stands for) in personal, social, and context-specific ways.
Peirce then subdivided the representamina of human representational systems into qualisigns, sinsigns, and legisigns (Figure 3.1). A qualisign is a representamen that draws attention to, or singles out, some quality of its referent. In language, an adjective is a qualisign since it draws attention to the qualities (color, shape, size, etc.) of referents. In other codes, qualisigns include the colors used by painters, the harmonies and tones used by composers, etc. A sinsign is a representamen that draws attention to, or singles out, a particular object in time-space: e.g. a pointing finger, the words here and there, etc. A legisign is a representamen that designates something by convention: e.g. words referring to abstract concepts, symbols, etc.
Peirce then identified three kinds of ways in which objects can be represented (Figure 3.2): (1) through some form of replication, simulation, or resemblance, called iconic representation (e.g. a photo resembles its object visually, a word such as bang resembles its object phonically, and so on); (2) through some form of indication, termed indexical representation (e.g. a pointing index finger is an indication of where an object is in space); (3) by convention, called symbolic representation (e.g. a rose is a symbol of love in some cultures).
Peirce viewed icons as the most basic type of signs, because they are tied to sense-based representation. Hence, he called them firstness signs, being physical substitutes for the referents themselves. But since icons are fashioned in cultural contexts, their manifestations across cultures are not exactly alike. Peirce used the term hypoicon to acknowledge this fact. A hypoicon is an icon that is constrained by the signifying order of a culture. But it can nevertheless be figured out by those who are not a part of the culture, if they are told what the referent is. Peirce defined indexes as secondness signs, because unlike icons they are not substitutes for their referents. Finally, he defined symbols as thirdness forms of representation, because in this case the sign, the sign-user, and the referent are linked to each other by the forces of historical and social convention, not by any sensory, temporal, or spatial phenomenon, situation, or circumstance.
Peirce suggested, moreover, that there were three types of interpretant (what the sign-user or sign-interpreter understands through the sign): (1) a rheme is an interpretant of a qualisign; (2) a dicisign is an interpretant of a sinsign; and (3) an argument is an interpretant of a legisign:
Like St. Augustine before him (chapter 2, §2.1), Peirce did not see the sign as independent of human variability—i.e. he did not consider it to be a purely conceptual means of knowing. The sign, for Peirce, is only a means of ascertaining the meaning intended on the basis of personal and social considerations, relevant sources, and historical factors.
The basic Peircean typology of signs can be summarized in chart form as follows:
As mentioned in the previous chapter (§2.1), semioticians seek answers to the what, the how, and the why of meaning. But what is meaning? And indeed what happens when we define the meaning of a sign?
Take the dictionary definition of cat as “a small carnivorous mammal domesticated since early times as a catcher of rats and mice and as a pet and existing in several distinctive breeds and varieties.” The first problem that emerges with this definition is the use of mammal to define cat—i.e. it makes the assumption that one is familiar with this term. But, then, what is a mammal? Once again, the dictionary definition is of little use because it defines mammal as “any of various warm-blooded vertebrate animals of the class Mammalia.” And this leads to the question: What is an animal? The dictionary defines an animal as an organism, which it defines as an individual form of life, which it defines as the property that distinguishes living organisms. At this point the dictionary has gone into a loop—i.e. it has started to employ an already-used word, organism, to define life! This inbuilt circularity in dictionaries is even more apparent when the referent is abstract, as the following vignette devised by Hayakawa (1991) illustrates:
- What do you mean by democracy?
- Democracy means the preservation of human rights.
- What do you mean by rights?
- I mean those privileges God grants all of us—I mean man’s inherent privileges.
- Such as?
- Liberty, for example.
- What do you mean by liberty?
- Religious and political freedom.
- And what does that mean?
- Religious and political freedom is what we enjoy under a democracy.
Not only does it seem that pinning down the meaning of words is probably a futile enterprise but, as the psychologist C. K. Ogden and the philosopher and literary critic I. A. Richards showed in their classic 1923 work, titled appropriately The Meaning of Meaning, the word meaning itself has many meanings. Here are some of them:
|=||“has the importance of”|
So, like the axicms of arithmetic or geometry, the notion of meaning is best left undefined in semiotic theory. It is something of which everyone has an intuitive understanding, but which virtually no one can really explain. It is a given. On the other hand, the term signification, as used in semiotics, is much easier to define, even though meaning and signification are used interchangeably by many semioticians. Essentially, signification designates what is inferable from the relation [A ≡ B]. Signification is not an open-ended process; it is constrained by a series of factors, including conventional agreements as to what [B] entails in specific contexts, the nature of the code to which the sign belongs, and so on. Without such inbuilt constraints, determining what a sign means would be virtually impossible.
Signification is a relational process—i.e. signs acquire their meanings not in isolation but in relation to other signs. Some of the more common relations are as follows:
- Synonymy: The relation by which the meanings of different signs intersect (hide-conceal, big-large, etc.). The intersection is normally of the partial overlapping variety. The meanings are rarely completely coincidental.
- Homonymy: The relation by which two or more meanings are associated with the same sign (play as in Shakespeare’s play vs. play as in He likes to play).
- Antonymy: The relation by which different signs stand in a discernible “oppositeness” of meaning to each other (love-hate, hot-cold, etc.). But antonymy is a matter of degree, rather than of categorical difference.
- Hyponymy: The process by which the meaning of one sign is included in that of another: e.g. the meaning of scarlet is included in the meaning of red, tulip in that of flower.
- Proportionality: The process by which distinctions among certain subsets of signs are maintained by the components that make up their meanings. These components are isolatable through proportions that can be set up among signs that are similar to those used in logic and mathematics (mam:woman:child :: bull:cow:calf).
Whereas the first four relational processes need little comment here, the last one requires further elucidation. In the proportion man:woman:child :: bull:cow:calf six components of meaning can be factored out. These are: [male], [female], [adult], [nonadult], [human], and [bovine]. These now allow us to understand at what level the proportion holds:
From this graph we can see that the proportion holds at its lowest, or subordinate level, where the components (also known as features) manifest an isomorphic pattern: [adult male]:[adult female]:[nonadult]::[adult male]:[adult female]:[nonadult]. It is at the superordinate level that the main categorical distinction between the two sets of triplets is established by the features [human] vs. [bovine]. Like factors in arithmetical and algebraic expressions, these features allow the human mind to keep certain signs distinct by virtue of the fact that they enter with certain other signs into proportional relations. By virtue of these relations a manageable set of signs allows members of a society to represent economically an illimitable array of meanings, in the same way that an infinite set of numbers can be represented in normal decimal notation by different patterns among ten digits.
The foregoing discussion raises the question of what a concept is. In semiotics, philosophy, and psychology the term concept is limited to designating a general strategy for classifying things that are perceived to subsume some general pattern. Concept-formation can thus be characterized as a “pattern-extracting” or “pattern-making” process that appears to serve some survival function in the human species. Distinguishing, for instance, between living and nonliving things is a conceptual pattern that, obviously, serves a useful purpose for people everywhere.
The psychological work on the nature of concepts is laden with controversy. But when looked at cumulatively and impartially, the research seems to support the following general distinction between concrete and abstract concepts—a concrete concept is one that is empirically demonstrable and an abstract concept is one that is not. A concrete concept is, therefore, one that encodes a pattern that can be seen, heard, smelled, touched, tasted, or observed in some direct way, while an abstract concept is one that encodes a pattern that is not so readily demonstrable. So, for example, the word cat refers to a concrete concept because one can always demonstrate or observe the existence of a cat in the physical world. The word love, on the other hand, refers to an abstract concept because, although love exists as an emotional phenomenon, it cannot be demonstrated empirically (i.e. the emotion itself cannot be observed apart from the behaviors, states of mind, etc. it produces).
The relevant psychological research shows that all concepts are formed in one of three general ways. The first is by induction—i.e. by extracting the pattern from specific facts or instances. For example, if one were to measure the three angles of, say, 100 specific triangles (of varying shapes and sizes), one would get the same total (180°) each time. This would then lead one to induce that the sum of the three angles of any triangle is the same. Induction reveals a type of conceptualization process designed to tease out a general pattern from specific occurrences. The second way in which humans form concepts is through deduction—i.e. by reaching a conclusion, or deduction, on the basis of certain observable facts or premises. It is the opposite of induction in that it entails inference of a pattern by reasoning from the general to the specific. For instance, if one were told that A is greater than B, and that B is greater than C, then one would deduce that A is (much) greater than C. Finally, there is a third type of concept-formation process that has, until recently, been largely neglected by mainstream philosophy and psychology, but which is, in effect, the crucial one that is involved in a large portion of abstract concept-formation. This was called, appropriately, abduction by Peirce. It is a blend of analogical, associative, and iconic pattern-inferencing. It can be defined simply as the derivation of an abstract pattern on the model of an existing concrete, or already known, pattern. A classic example of abductive reasoning can be seen in the theory of atomic structure proposed by the English physicist Ernest Rutherford (1871-1937). Rutherford conceptualized the inside of an atom as having the structure of the solar system, with electrons behaving like little planets orbiting around an atomic nucleus. His planetary model of atomic structure was, in effect, an abduction of solar-system structure.
The distinction between concrete and abstract concept-formation is, needless to say, a general one. In actual fact, there are many degrees and layers of concreteness and abstraction in conceptualization that are influenced by social, affective, and other kinds of factors. Suffice it to say here that concept-formation serves the basic purpose of organizing most of the raw, unorganized sensory information that comes from seeing, hearing, and the other senses into meaningful forms. Moreover, the type of conceptualization process enlisted depends on the type of pattern that the human mind seeks from a specific situation. Often, all three processes—induction, deduction, abduction—are involved in a complementary fashion.
In the early 1970s, the psychologist Rosch (1973) came to the conclusion that concrete concepts such as the colors display a three-tiered hierarchical organization that varies from language to language, from person to person. At the highest level, which she called the superordinate level, concepts have a highly general classificatory function. So, in the domain of color, the concept encoded by the word color itself would be a superordinate concept, because it refers to the general phenomenon of chromatism itself. Then there is the basic level, which is where a word such as blue would fit in. This is a “typological” level—i.e. the level at which “types” of color exist. The third level, which Rosch called the subordinate level, is where more detailed ways of classifying something occur. There are, in fact, many shades of blue—dark blue, navy blue, sky blue, turquoise, etc.—which we might need for specialized purposes.
In semiotics, concepts are further classified according to the meaning patterns they exemplify. There are three general kinds of patterns: denotative, connotative, and annotative.
Denotation is the initial conceptual meaning that is established between a signifier and a signified. But the denoted signified or referent, [B], is not something specific in the world, but rather a prototypical category of something. For instance, the word cat does not refer to a specific cat, although it can in a specific usage of the term, but to the category of animals that we recognize as having the quality “catness.” The denotative meaning of cat is, therefore, really catness, a prototypical mental picture marked by specific distinctive features such as [mammal], [retractile claws], [long tail], etc. This composite mental image allows us to determine if a specific real or imaginary animal under consideration is an exemplar of [B]. Similarly, the word square does not denote a specific square, but rather a figure consisting of four equal straight lines that meet at right angles. It is irrelevant if the lines are thick, dotted, 2 meters long, 80 feet long, or whatever. So long as the figure can be seen to have the distinctive features [four equal straight lines] and [meeting at right angles], it is identifiable denotatively as a square.
A remarkable feature of human semiosis is that any sign can be extended freely to encompass other kinds of referents that appear, by association or analogy, to have something in common with [B]. This extensional process is known as connotation As an example of how connotation works, consider again the word square. Its extended uses can be seen in utterances such as:
- She’s so square (“old fashioned”)
- He has a square disposition (“forthright,” “honorable”)
- Put it squarely on the table (“evenly,” “precisely”)
Note, however, that the distinctive features of [B] are implicit in such extensional uses; i.e. an old-fashioned person, an honorable individual, and the action of laying something down evenly imply these features—a square is an ancient idea and known by everyone (hence “old-fashioned”); it is also a figure with every part equal (hence “forthright”); and it certainly is an even-sided figure (hence “evenly”). Any connotative extension of the word square is thus constrained by the original [B]. More formally, connotation can be defined as the mapping of [A ≡ B] onto a new referent, [C], (“old-fashioned,” “forthright,” “evenly,” etc.), if [C] can be seen to entail [B] by association or analogy. This can be shown formally as follows:
[A ≡ B]≡ [C] ⬄ [C⊇B]
This formula states that any sign [A ≡ B] can be applied to any other referent [C] by extension, if [C] entails the distinctive features of [B] ([C ⊇ B]). To use the above example of square connoting evenly as a concrete case-in-point, the above formula would be filled in as follows:
[square ≡ [four-sided plane figure of equal sides]] ≡ [evenly] ⬄ [evenly a foursided plane figure of equal sides]
Connotation is the operative signifying mode in the production and decipherment of creative texts such as poems, novels, musical compositions, art works—in effect, of most of the non-mathematical and non-scientific texts that a culture produces. But this does not mean that meaning in science is necessarily encoded denotatively. On the contrary, many of the theories and models of scientists, as the philosopher Max Black (1962) argued, are born of connotative and/or metaphorical thinking, even though they end up being interpreted denotatively over time. The theory of atomic structure, for instance, was fashioned through analogical extension. It was presented as a tiny solar system by physicist Ernest Rutherford (above, §3.4), with a sun (nucleus) and orbiting planets (electrons, protons, etc.). The end result was a theory that extended a model that at the time it was devised was already familiar to scientists.
There is another type of connotation that semioticians generally call emotive, but which we will call annotation instead. The word yes, for example, can have various emotive meanings, depending on the tone of voice with which it is uttered. If one says it with a normal tone of voice, it will be understood as a sign of affirmation. If, however, one says it with a raised tone, as in a question, Yes?, then it would imply doubt or incredulity. Such “added meanings” to the word yes are examples of annotation. This can be defined simply as the interpolation or assignment of subjective meanings to a sign or text.
In 1957, Osgood, Suci, and Tannenbaum invented an interesting technique for fleshing out the annotations that concepts entail, known as the semantic differential. This consisted in posing a series of questions to subjects about a specific concept—Is it good or bad? weak or strong? etc.—as seven-point scales, with the opposing adjectives at each end. The answers were then analyzed statistically in order to sift out any general pattern from them. Consider a hypothetical example. Suppose that various subjects are asked to evaluate the concept President in terms of seven-point scales such as the following:
An informant who feels that a President should be modern would place a mark towards the modern end of the modern-traditional scale. One who feels that a President should not be too young or old would place a mark near the middle of the young-old scale; an informant who feels that a President should be bland would place a mark towards the bland end of the attractive-bland scale; and so on. If a large number of informants were asked to rate the term President in this way, then it would be possible to draw an ideal profile of the presidency in terms of the statistically significant variations in annotation that the term evokes.
Interestingly, research utilizing the semantic differential has shown that, while the meanings of most concepts are subject to personal interpretation and subjective feelings, the range of variation in annotation is not simply a matter of randomness, but forms a socially based pattern. In other words, the experiments have shown that the annotations of signs are constrained by culture: e.g. the word noise turns out to be a highly emotional concept for the Japanese, who rate it consistently at the ends of the scales presented to them; whereas it is a fairly neutral concept for Americans, who place it in the mid-range of the scales.
In effect, the signs that make up signifying orders refer to those aspects of reality or experience that specific cultures deem important, relevant, or useful. Therefore, what signs detect or capture in the world is always but a portion of what is around. But, as we have seen in the foregoing discussion, through the phenomena of connotation and annotation even the limited set of signs that a culture makes available to its members can be used to cover a very large domain of meaning. Moreover, as we shall see in chapter 6, new meanings and new forms of reference can always be created through metaphor.
The semiotic investigation of how signs refer to the world (signification) has uncovered that signs have specific properties. It has shown, for instance, that signs can be classified as witting and unwitting (Sebeok 1994). Unwitting signs are those that are generated by bodily processes and are, ordinarily, beyond the conscious control of the individual. Most signals fall into this category. Witting signs are those that humans make and use intentionally to represent the world. The signifiers of unwitting signs are provided by biology; those of witting signs are created by individuals and cultures for specific purposes.
The bodies of all animals produce signals to convey certain needs, to respond to stimuli, etc., but what they mean will depend on the species. As the biologist Jakob von Uexküll (1909) argued, the signaling system is a derivative of anatomical structure. Animals with widely divergent anatomies will manifest virtually no signaling patterns in common.
All animals are endowed with the capacity to use and respond to species-specific signals for survival. Birds, for instance, are born prepared to produce a particular type of coo, and no amount of exposure to the songs of other species, or the absence of their own, has any effect on their cooing. A bird reared in isolation, in fact, will sing a very simple outline of the sort of song that would develop naturally in that bird born in the wild. This does not mean, however, that animal signaling is not subject to environmental or adaptational factors. Many bird species have also developed regional cooing “dialects,” apparently by imitating each other. Vervet monkeys, too, have the usual set of signals to express emotional states and social needs, but they also have developed a particular predator signaling system—a specific call alerting the group to eagles, one to four-legged predators such as leopards, another to snakes, and one to other primates. The calls seem innate, but in actual fact the young of the species learn them only by observing older monkeys and by trial and error. An infant vervet may at first deliver an aerial alarm to signal a vulture, a stork, or even a falling leaf, but eventually comes to ignore everything airborne except the eagle.
Most signals are emitted unwittingly in response to specific types of stimuli, urges, needs, and affective states. Because manifestations of animal signaling are truly remarkable, it is little wonder that they often trick people into seeing much more in them than is actually there. A well-known example of how easily people are duped by animal signaling is the case of Clever Hans. Clever Hans was heralded the world over as a German “talking horse” in 1904 who appeared to understand human language and communicate answers to questions by tapping the alphabet with his front hoof—one tap for A, two taps for B, three taps for C, and so on. A panel of scientists ruled out deception by the horse’s owner. The horse, it was claimed, could talk! Clever Hans was awarded honors and proclaimed an important scientific discovery. Eventually, however, an astute member of the scientific committee that had examined the horse, the Dutch psychologist Oskar Pfungst, came to suspect that Clever Hans would probably not tap his hoof without observing his questioner, since the horse had probably figured out—as most horses can—what the signals that his owner was unwittingly transmitting meant. The horse, Pfungst asserted, tapped his hoof only in response to inadvertent cues from his human handler, who would visibly relax when the horse had tapped the proper number of times. To show this, Pfungst simply blindfolded Clever Hans, who, as a consequence, ceased to be so clever. The “Clever Hans phenomenon,” as it has come to be known in the annals of psychology, has been demonstrated over and over with other animals as well (e.g. a dog will bark in lieu of the horse’s taps in response to certain signals unwittingly emitted by people).
A large portion of communication among humans also unfolds in the form of unwitting signals. It has been shown, for example, that men are sexually attracted to women with large pupils, which signal unconsciously a strong and sexually tinged interest, as well as making females look younger (Sebeok 1994). This would explain the fashion vogue in central Europe during the 1920s and 1930s of a crystalline alkaloid eye-drop liquid derived from belladonna (“beautiful woman” in Italian). The women of the day used this drug because they believed—and correctly so, it would appear—that it would enhance facial appearance and sexual attractiveness by dilating the pupils.
But humans are capable as well of deploying witting signals for some intentional purpose—e.g. nodding, winking, glancing, looking, nudging, kicking, head tilting. As the linguist Karl Bühler (1934: 28) aptly observed, such signals act like regulators, eliciting or inhibiting some action or reaction. Signaling systems can also be created for conventional social purposes. The list of such systems is extensive, and includes railway signals, smoke signals, semaphores, telegraph signals, Morse code signals, warning lights, flares, beacons, balefires, red flags, warning lights, traffic lights, alarms, distress signals, danger signals, whistles, sirens, bleepers, buzzers, knocking, gongs, bells, drums.
An icon is a sign made to reflect some perceivable property of a referent so that it can be figured out in the signifier. Photographs, drawings, Roman numerals such as I, II, and III are visual iconic signs because they are created to reflect their referents visually; onomatopoeic words are vocal iconic signs because they are created to reflect sound properties of their referents; perfumes are olfactory iconic signs because they are meant to be suggestive of certain natural scents; a block of wood with a letter of the alphabet carved into it is a tactile icon because the letter’s shape can be felt and figured out by touch.
Iconicity is seen by most semioticians as a primary strategy in representation—a view, incidentally, that has philosophical antecedents in John Locke (1632-1704), Giambattista Vico (1688-1744), Ernst Cassirer (1874-1945), and Suzanne Langer (1895-1985), among others. The English philosopher Locke argued, in fact, that words refer to sensible properties, with meaning being the internal operation of consciously recognizing these properties. The Italian philosopher Vico emphasized that the human mind “does not understand anything of which it has had no previous impression from the senses” (in Bergin and Fisch 1984: 123), because it is “naturally inclined by the senses to see itself externally in the body; and only with great difficulty does it come to understand itself by means of reflection” (Bergin and Fisch 1984: 95). The German philosopher Cassirer linked abstract forms of expression to an unconscious “grammar of experience” whose categories are not those of logical thought, but rather of an archaic mode of sensorial thinking that continues to gain expression through iconicity. The American philosopher Langer saw all efforts to know and understand through representation as essentially sensory-aesthetic reactions to the world.
The presence of iconicity in representational systems across cultures is strong evidence that human consciousness is attentive to the recurrent patterns of color, shape, dimension, movement, sound, taste, etc. detected by the human perceptual system. Archeological evidence attests to the ancientness of visual iconicity. The first inscriptions, cave drawings, small sculptures, and relief carvings of animals and female figures found in caves throughout Europe, such as those at Lascaux in France and Altamira in Spain, were created some 30,000 to 40,000 years ago. But even in the verbal domain iconicity was probably the primordial semiosic force in word creation (see chapter 5, §5.2). As Peirce so often remarked, the verbal symbols and abstractions that seem so remote from the sensorial realm were nonetheless born of iconic semiosis.
Indexes are signs created to identify something or someone in terms of its existence or location in time or space, or else in relation to something or someone else. Indexes do not resemble their referents, as icons do; they indicate or show where they are. The most typical manifestation of indexicality is the pointing index finger, which humans the world over use instinctively to point out and locate things, people, and events in the world. Many words, too, are indexes: e.g. here, there, up, down refer to the relative location of things when speaking about them.
Indexicality is known more technically as deixis. There are three types of deixis:
- Spatial Deixis is a form of reference by which the spatial locations of objects, beings, and events are either indicated or correlated by a manual sign like the pointing index finger, a demonstrative word like this or that, an adverb like here or there, etc.
- Temporal Deixis is a form of reference by which the temporal relations among things and events are either indicated or correlated by an adverb like before, after, now, or then, a timeline graph showing points in time as located to the left and right, or on top and below, of each other, etc.
- Personal Deixis is a form of reference by which the relations among participants taking part in a situation are either indicated or correlated by a personal pronoun like I, you, he, she, an indefinite pronoun like the one, the other, etc.
The presence of indexicality in representational systems across the world is evidence that human consciousness is attentive not only to patterns of color, shape, etc., resulting in iconic semiosis, but also to the recurrent cause and effect patterns that are contingent on time and space relations, resulting in indexical semiosis.
Symbols are signs created by conventional means. Most semioticians agree that the emergence of symbolicity in humankind is what has endowed it with the capacity to reflect upon the world in purely conceptual ways. Words in general are symbolic signs. But any object, sound, figure, etc. can be fashioned and/or employed symbolically. A cross figure can stand for the concept “Christianity”; a V-sign made with the index and middle fingers can stand for the concept “peace”; white can be symbolic of “cleanliness,” “purity,” “innocence.” These signifieds are established by convention.
The presence of symbolicity in representational systems across the world is evidence that human consciousness is not only attentive to physical and cause-and-effect patterns (resulting in icons and indexes respectively), but also to pattern in itself. The view of some semioticians—and it is ours as well—is that iconicity, indexicality, and symbolicity are interconnected in evolutionary terms. The anecdotal evidence to support this view is substantial: e.g. the child first learns to represent something by pointing to it (indexicality) and then naming it (symbolicity) later; people instinctively resort to iconicity (gesturing, making imitative sounds, etc.) and indexicality (pointing) when communicating with someone who does not speak the same language; iconic, indexical, and symbolic modes of representation often converge in the creation of a single sign; and so on. As an example of the latter, consider the common traffic sign for a crossroads:
The signifier of this sign consists of two straight lines, one with an arrowhead, intersecting at right angles. This cross figure is both an icon and a symbol—it is iconic because its shape visually resembles a crossroads, but since the figure could easily be used in our society to represent other signifieds in other situations, it is also symbolic insofar as we need to know that it has been chosen, by convention, to refer to a crossroads. Finally, the sign is also an index because when it is placed near an actual crossroads it indicates that one is about to reach it physically.
Symbolicity is the operative mode of representation in all convention-based systems. Consider, for instance, a typical high school problem in algebra:
Mary has a number of dimes and nickels worth $2.00. If she has twice as many nickels as dimes, how many of each does she have?
This is solved, of course, by setting up an equation. First, a letter from the alphabet is selected, say x, to stand for the number of dimes. This is, of course, an arbitrary choice. Any other letter, or symbol (a dot, a line, etc.) for that matter, could have been chosen to represent the number of dimes. Incidentally, a letter symbol in algebra can stand for any number of things: a number of dimes, of boxes, of shoes, etc. What it stands for is irrelevant. The connection between x and a quantitative referent is purely arbitrary. Remarkably, this is precisely what allows people to solve such problems effortlessly. The rest of the reasoning process involved in solving the problem is tangential to this main point. It goes like this. One dime is worth $0.10, so x dimes are worth $0.10x;there are twice as many nickels as dimes, or 2x nickels in total. One nickel is worth $0.05, so 2x nickels are worth 2($0.05x) = $0.10x. The two values add up to $2.00. So, the appropriate equation is $0.10x + $0.10x = $2.00. Solving for x, we get x = 10. Thus, Mary has 10 dimes and 20 nickels.
Nowhere has symbolicity borne more remarkable fruits, in fact, than in the development of mathematics and science. An early impressive example of what it has permitted humans to do is found in the annals of geometric history. Standing during the summer solstice at Alexandria, and knowing that it was due north of the city of Syene, with the distance between the two cities being 500 miles, the Greek geometer Eratosthenes (275-194 BC) used simple geometric reasoning to calculate the earth’s circumference—without having to do it physically. At the summer solstice, he reasoned, the noon sun is shining directly down into a well at Syene, since the sun is directly overhead at that time of day:
Eratosthenes, therefore, represented the direction of the sun with the straight line OBS’ (in the illustration). At the same instant in Alexandria, he reasoned further, the actual direction of the sun was representable with AS, and the overhead direction with OAD. AS stood, in effect, for the direction of a ray of sun, and since all rays are parallel, then Eratosthenes knew that ray AS was parallel to ray BS’ at Syene. From a theorem of Euclidean geometry, he was then able to conclude that the angles DAS and AOB are equal. On the basis of this knowledge, Eratosthenes proceeded to measure DAS, which he found to be 71/2°. This then was the size of angle AOB at Syene. But, Eratosthenes reasoned further, this angle is 71/2/360° (since the earth is virtually a sphere and therefore almost a 360° angle), or 1/48 of the entire angle at O. It followed from another fact of geometry that the arc AB, the distance between Alexandria and Syene, was 1/48 of the entire angle, which was, of course, the circumference of the earth. Therefore, Eratosthenes concluded, the circumference was 48 times the length of the arc: 48 x 500 = 24,000 miles. This is in close agreement with the actual known value today of 24,844 miles.
This story shows clearly how symbolic representation allows people to model the world in abstract ways, in accordance with established conventions (in this case of Euclidean geometry), and then discover facts about the real world through them. It has permitted people, in other words, to let go physically of their environment in order to grasp it in abstract ways. But symbolic methods of representation are not born symbolically. The early geometers of ancient Egypt, Sumer, and Babylon were concerned with such practical problems as measuring the size of fields and laying out accurate right angles for the corners of buildings. Their empirical discoveries, their meticulous diagrams, and their visual observations produced early iconic models that were refined and systematized later by the Greeks. By the sixth century BC the Greek mathematician Pythagoras (582?-500? BC) laid the cornerstone of symbolic geometry by showing that the various observations and iconic diagrams of the empirical geometers could be synthesized into a theorem. Other Greek geometers subsequently synthesized other observations into other theorems. The further synthesizing of theorems led to the establishment of geometry as a science. Only then could someone like Eratosthenes use its techniques to determine the earth’s circumference.
Recall from above (§3.3) that a legitimate verbal signifier in a language is one that shows consistency with the phonological structure of the language. The signifier duck, for instance, is an acceptable signifier to English-speaking ears because it conforms to English sound and word structure. The formation of any signifier, verbal or nonverbal, is constrained in fact by the structural requirements of the code (language, music, etc.) within which it is formed.
There are three structural relations that characterize all codes. One is called paradigmatic structure. Consider the following words:
- pin vs. bin
- fun vs. pun
- duck vs. luck
The different meanings of the words are detected, first, by virtue of the fact that they have different initial sounds that signal the differences. This differentiation feature of signs is known as paradigmatic structure. It is the relation whereby some minimal feature in a signifier is sufficient to keep it differentiated from all other signifiers of the same kind. Paradigmatic structure is found in all human systems. In music, for instance, a major and minor chord of the same key are perceivable as distinct on account of a half tone difference in the middle note of the chord; the left and right shoes of a pair of shoes are identifiable as different on account of their different orientations; and so on.
Paradigmatic relations do not operate in isolation; they interact with combinatory and organizational relations. These are known as syntagmatic. Paradigmatic structure involves distinctiveness and selectability; syntagmatic structure involves combination and organization. The words pin, bin, fun, run, duck, luck are legitimate signifiers because the combination of sounds with which they are made is consistent with English syllable structure. The latter is an example of syntagmatic structure. On the other hand, mpin, mbin, mfun, mrun, mduck, mluck would not qualify as legitimate verbal signifiers in English because they violate its syllable structure. Syntagmatic structure too is found in all human systems. In music, for instance, a melody is recognizable as such only if the notes follow each other in a certain way (e.g. according to the rules of classical harmony); two shoes (with different orientation) are considered to form a pair if they are of the same size, style, and color; and so on.
In essence, something is a sign if it has a discernible (repeatable and predictable) form and is constructed in a definable (patterned) way. Signs are like pieces of a jigsaw puzzle. These have visual features on their “faces” that keep them distinct from each other, as well as differently-shaped “edges” that make it possible to join them together in specific ways to complete the overall picture.
The third structural relation is called analogy. This is a replacement relation, by which one type of sign can replace another in a specific way. Thus, for example, European cards can replace American cards for playing solitaire if an analogy is made between European and American suits. The model of planets orbiting around the sun can be used by analogy to represent the structure of an atom, whereby the sun is replaced by the nucleus, the orbiting planets by electrons, and so on.
Analogy is a force of change in sign systems. Words are often reformed or created on the model of existing grammatical patterns in a language. For example, in Old English the plural of name was naman. This was changed over time to names on the model of nouns like stone—stones. Analogy is the operative force when children utter a form like goed, rather than went. This is created in analogy with forms like played, stayed, etc.
A code is a structural system, i.e. a system in which signs reveal a specific paradigmatic, syntagmatic, and analogical architecture. A simple example of a code is the game of solitaire. The cards in this code are legitimate signs if they have been constructed with distinctive features (in terms of suit and number value); if they can be organized in vertical columns to complete the game; and if they can be replaced by a set of cards with a comparable system of distinctive features. In other words, solitaire is a code in which the various cards are differentiable paradigmatically from each other by suit and number and placeable or organizable syntagmatically into columns in specific ways, in which they can be replaced by other kinds of cards with comparable features.
A code can be thought of as being like a formatted computer disk. The format of a phonological code, for instance, provides the differential phonic features (sounds) along with a finite set of combinatory syllable patterns for making words. Often, a code is made up of many constituent structural systems or codes (also called subsystems or subcodes). For instance, the language code consists of phonological, morphological, syntactic, and semantic subcodes, each with its own type of paradigmatic, syntagmatic, and analogical formats.
Knowledge of a code allows an individual to construct appropriate messages with the resources of the code. A face-to-face conversation, for instance, involves not only the simultaneous deployment of the subsystems of language (phonological, morphological, etc.), but also those that compose gestural, facial, and various other nonverbal codes. The verbal message is thus “woven together” with the resources of different codes. This “weaving together” is called a text. A text is a collation of signs taken from one or more codes in order to construct and communicate a message. When someone says something to someone else, writes a letter, or wears a certain kind of dress for an occasion, s/he is engaged in text-making. Routine conversations, musical compositions, stage plays, poems, dance styles, ceremonies are but a few examples of the many kinds of texts that people make on a regular basis, as individuals or as groups. These can only be understood fully if the codes used to make them are known. The term message refers to what one wishes to communicate with a text; the term text refers more specifically to how the message is constructed.
Texts bear meaning in specific contexts. The term context refers to the real-world conditions—physical, psychological, social, etc.—that influence, shape, and even determine how a text is made or what a text means. Consider a discarded and damaged beer can. If someone were to come across this item on a sidewalk on a city street, s/he would no doubt view it as a piece of garbage or rubbish. But if the person saw the very same object on a pedestal, displayed in an art gallery, “signed” by some artist, and given a title such as “Waste,” then s/he would be inclined to interpret its meaning in a vastly different way. S/he would, in fact, interpret it most likely as an artistic text, decrying a throw-away or materialistic society. Clearly, the can’s physical context of occurrence and social frame of reference—its location on a sidewalk vs. its display in an art gallery—will determine what it means.
It is now possible to define the signifying order more formally. The signifying order is the overall system, or macrocode, that supplies the signs, the specific codes in which they are organized structurally, and the texts they make possible to the members of a culture. This interrelationship can be illustrated graphically as follows:
Research in semiotics has shown that representation unfolds in terms of three dimensions which, as we discussed above (§3.3), Peirce called firstness, secondness, and thirdness. We will refer to this as the dimensionality principle throughout this book.
Consider, for instance, the temporal aspect of representation. In this case, a sign’s meaning in a specific instance is inferable relative to three temporal dimensions, which can be represented graphically as axes in three-dimensional space: (1) a synchronic (firstness) axis that provides its attendant meaning; (2) a diachronic (secondness) axis that provides its latent historical meanings; and (3) a dynamic (thirdness) axis that entails the potential for new meaning pursuant to its use in the specific instance:
Consider, as a second example, the type of dimensionality that can be called notational. This constitutes a representation space in which the various meanings of a sign are inferable relative to three notational axes: (1) a denotative (firstness) axis that provides its intended meaning; (2) a connotative (secondness) axis that allows for the sign’s extended uses according to context; and (3) an annotative (thirdness) axis that entails the individual’s own understanding of the sign’s meanings. The sign’s position relative to the axes determines its specific meaning: if it is closer to the firstness axis it is primarily denotative (intended meaning); if it is closer to the secondness axis it is primarily connotative (extended meaning); and if it is closer to the thirdness axis it is primarily annotative (personal meaning).
Consider, finally, a third type of dimensionality that can be called structural. This constitutes a representation space in which the various meanings of a sign are inferable relative to three structural axes: (1) a paradigmatic (firstness) axis that entails a selection operation; (2) a syntagmatic (secondness) axis that entails a combination operation; and (3) an analogical (thirdness) axis that entails a replacement operation:
These dimensionalities can be summarized as follows:
The dimensionality principle makes it explicit that there is an interconnectedness among the multifarious dimensions of representation and signification. It also allows us to establish a commonality among different representational systems. Because all such systems are composed of the same kinds of dimensionalities, the principle provides a basis for showing an interrelation among all areas of knowledge-making, from language to science and the arts. A digit in numerical representation, for instance, has the exact same structural features in dimensional terms that, say, a noun in language has—i.e. both are signs that exist in temporal, notational, and operational three-dimensional spaces, deriving their forms, functions, and meanings in terms of these dimensionalities. The difference between a digit and a noun is thus not to be located in structural patterns, but in the different cognitive functions of the representational systems to which they pertain. This is why, despite their different functions, both are understandable in exactly the same way. In essence, the dimensionality principle makes it obvious why such seemingly diverse forms of representation as poetry and mathematics are not mutually exclusive—with adequate exposure to both, people will be able to extract meaning from either one of them in remarkably similar ways.
The signifying order is both restrictive and liberating in human terms. It is restrictive in that it imposes upon individuals born into a specific culture an already-fixed system of signification. This system will largely determine how people come to understand the world around them—i.e. in terms of the language, music, myths, rituals, technological systems, and other codes that they learn in social context. But the signifying order is also liberating because paradoxically it provides the means by which individuals can seek new meanings on their own. The artistic, religious, scientific, and philosophical texts to which individuals are exposed in social contexts, moreover, open up the mind, stimulate creativity, and engender freedom of thought. As a result, human beings tend to become restless for new meanings, new messages. For this reason, codes are constantly being modified by new generations of artists, scientists, philosophers, and others to meet new demands, new ideas, new challenges.
Leaving aside this knack for creativity for the moment, the fact remains that culture structures beliefs, attitudes, worldview, and even sensory perception to varying degrees. As a concrete example, the reader should look at the following classic visual illusion. As s/he can probably confirm for h/erself, most people living in Western societies will see line AB as longer than line CD:
In actual fact the lines AB and CD are equal in length, but the orientation of the arrowheads fools the Western eye into seeing AB as longer than CD. In rural Uganda, on the other hand, psychologists have found that people see the lines as equal in length (Simon 1976: 19-20). The factor behind this illusion is that Western individuals are accustomed to seeing drawings in perspective. In painting perspective is the technique of creating an illusion of depth or length in two-dimensional surface drawings. As a historical footnote, it should be noted that the craft of perspective drawing dates back to the Renaissance, after the Italian artist Filippo Brunelleschi (1377-1446) discovered and then entrenched this technique in Western painting.
Visual illusions provide strong evidence in favor of the notion of structural effect, i.e. of the notion that signifying orders structure perception, beliefs, worldview. Consider, as another example, color perception. The light spectrum consists of a continuous gradation of hue from one end to the other. According to some physicists, there are potentially 8 million gradations that the human eye is capable of distinguishing. If one were to put a finger at any point on the spectrum, there would be only a negligible difference in gradation in the colors immediately adjacent to the finger at either side. Yet a speaker of English describing the spectrum will list the gradations as purple, blue, green, yellow, orange, and red. This is because the speaker has been conditioned by the English language to classify the content of the spectrum in specific ways. There is nothing inherently “natural” about the speaker’s organizational scheme; it is a reflex of English vocabulary, not of Nature.
By contrast, speakers of other languages are predisposed to see other color categories on the very same spectrum. Speakers of Shona, an indigenous African language, for instance, divide it up into cipswuka, citema, cicena, and cipswuka (again), and speakers of Bassa, a language of Liberia, segment it into just two categories, hui and ziza. The relative proportional widths of the gradations that these color categories represent vis-à-vis the English categories can be shown graphically as follows:
So, when an English speaker refers to, say, a ball as blue, a Shona speaker might refer to it as either cipswuka or citema, and a Bassa speaker as hui. But this does not stop an English speaker from relating h/er categories to those of the other two languages. The specific color categories one has acquired in a cultural context in no way preclude the ability to perceive the color categories of other cultures. This is, indeed, what a learner of another language ends up doing when s/he studies the new color system: i.e. s/he learns how to reclassify the content of the spectrum in terms of the new categories. Moreover, in all languages there exist signifying resources for referring to more specific gradations on the spectrum if the situation should require it. In English the words crimson, scarlet, vermilion, for instance, make it possible to refer to gradations of red. But these are still felt by speakers to be subcategories of red, not distinct color categories on their own. They are related hyponymically to each other (above, §3.4).
In 1969, the psycholinguists Berlin and Kay argued, moreover, that differences in color terms are only superficial matters that conceal general underlying principles of color perception. Using the judgments of the native speakers of twenty widely divergent languages, Berlin and Kay came to the conclusion that there were “focal points” in basic (single-term) color systems which clustered in certain predictable ways. They identified eleven universal colors, or focal points, which corresponded to the English words red, pink, orange, yellow, brown, green, blue, purple, black, white, and gray. Not all the languages they investigated had separate words for each of these colors, but there emerged a pattern that suggested to them the existence of a fixed way of perceiving color across cultures. If a language had two colors, then the focal points were equivalents of English black and white. If it had three color terms, then the third one corresponded to red. A four-term system had either yellow or green, while a five-term system had both of these. A six-term system included blue; a seven-term system had brown. Finally, purple, pink, orange, and gray were found to occur in any combination in languages which had the previous focal points. Berlin and Kay found that languages with, say, a four-term system consisting of black, white, red, and brown did not exist. Berlin and Kay’s universal color system can be represented as follows:
The intriguing implications of this research were pursued vigorously in the 1970s by many psychologists. Eleanor Rosch (1975), for instance, demonstrated that the Dani people of West Irian, who have a two-color system similar to the Bassa system described above, were able to easily discriminate eight focal points. Using a recognition-memory experiment, Rosch found that the Dani recognized focal colors better than non-focal ones. She also found that they learned new colors more easily when the color names were paired with focal colors. Such findings suggested to Rosch that languages provided a guide to the interpretation of color, but they did not affect its perception in any way.
But many problems remain to this day with the conclusions reached by color researchers. For one thing, the fact that the eleven focal points posited by Berlin and Kay corresponded to the color terms of their own language (English) is suspicious. Could the researchers have been predisposed by their own language to gloss all other terms according to the English categories? Many of the terms Berlin and Kay listed, some critics have pointed out, turn out to be borrowings (color terms taken and/or adapted from other languages), which would greatly undermine their theory.
Semiotically speaking, color terms are verbal signifiers, and the categories they encode are their referents. This means that people are predisposed to attend primarily to the gradations (referents) they have learned to discriminate through the color signifiers they know. This is a practical strategy; otherwise, millions of signifiers would need to be invented to classify the spectrum in terms of all the possible discriminations that can be made. But this simple semiotic explanation also makes it clear that to learn new ways of classifying the spectrum, all one has to do is learn new signifiers or invent new ones.
The underlying purpose of the foregoing discussion has been to show that the meanings captured by one signifying order in no way tell the complete story about the world. The signifying order always leaves gaps, offering up only a portion of what is potentially knowable in the world. Indeed, a little reflection will reveal that an infinite number of signifiers could be created without any signifieds. This is exactly what young children do when they make up “nonsense words,” creating them seemingly only for the pleasure of making imitative, pleasant, or humorous sound effects. The great British writer of children’s books Lewis Carroll (1832-1898) invented his own nonsense language, in his poem Jabberwocky, to show that the English language as constituted does not tell all there is to tell about reality. Using signifiers such as brillig, slithy, tove, wabe and others (from Through the Looking Glass, 1871: 126-129), Carroll showed that it is an easy thing to make up legitimate words that seem to beg for legitimate meanings.
In effect, Carroll had coined signifiers without signifieds; i.e. words that suggested ideas by virtue of the fact that they sounded like English words. Actually, Carroll provided his own signifieds for these words as follows to make his point even stronger:
Analogously, there are infinitely many signifieds that are not captured by a language. Indeed, there are still no words in English for “side of a hill,” “smooth and active,” and other Jabberwocky concepts. Here are a few other examples of potential signifieds not captured by existing English words:
However, even though gaps exist in a signifying order, humans have the ability to fill them any time they wish. They do this typically by inventing new signs, altering already-existing ones to meet new demands, borrowing signs from other signifying orders. One can always find ways to refer, for instance, to the above signifieds by paraphrase or some other verbal strategy—e.g. a pie without the top crust = a top-crustless pie. But the lack of signifiers to enshrine these concepts implies that they will not be anticipated by speakers of English within the scheme of things.
The notion of structural effects is not a theory of mind; it simply acknowledges what effects signs have on human thinking. In actual fact, there are creative forces constantly at work in individual human beings. The philosopher Giambattista Vico (chapter 2, §2.1) termed these the fantasia and the ingegno. The former is the capacity that allows human beings to imagine literally anything they desire freely and independently of biological or cultural processes; it is the creative force behind new thoughts, new ideas, art, science, and so on. The latter is the capacity that allows human being to convert their new thoughts and ideas into expressive representational structures—metaphors, stories, works of art, scientific theories, etc. So, although human beings are indeed shaped in large part by their particular biology and by the social system in which they are reared, they are also endowed with creative faculties that seem, in our opinion, to be well beyond the capacities of the current sciences of biology and psychology to explain. The human being is, indeed, a true enigma among living species.