Symbolic Modeling and Aesthetic Creation: With a Critique of “Global Semiotics”

I. Sebeok and Deely: “Global Semiotics”

At the turn of the twentieth and twenty-first centuries, semiotics exerted an ever broader and deeper influence and gradually won a scholarly status not to be overlooked. John Deely holds: “The idea of the universality of the sign is today taken for granted as the badge of postmodernity; this is the innate right of semiotics.” Susan Petrilli, taking up Deely’s view, further points out: “The postmodern stage is called ‘the way of signs’ … As an interpreter of sign-action living in the twenty-first century, ‘now’ means the line dividing the modern from the postmodern, means the beginning of a new, truly global epoch of intellectual culture. ‘If there is a single core idea for the emergence of postmodern consciousness’ … ‘then it is the idea of the sign.’ The idea of the sign, and the foundation of this idea in sign-theory, together constitute the key context of our present, particular standpoint, in order to make a new understanding of the history of understanding.” To replace the view of language with a broader view of the sign, and to replace the single linguistic medium with complex, plural signs as the medium between human and world, and thereby to reinterpret the history of human understanding—this will bring us into a new era of understanding. As semiotic research continually deepened, the domain of semiotics also continually expanded, and semioticians went on to propose the conception of a “global semiotics.” Sign-activity is thereby no longer related to human beings alone, but is regarded as a phenomenon ubiquitous in the biological world and indeed the whole of nature. In this, Thomas Albert Sebeok played a decisive role.

Sebeok’s theoretical horizon was extremely broad; his scholarship touched on linguistics, semiotics, anthropology, biology, and cultural studies—and it was precisely this rich accumulation that enabled him to cross the boundaries of the humanities, the social sciences, and the natural sciences, to rethink semiotics with a macroscopic vision, and creatively to propose a methodological framework, “Modeling Systems Theory,” which profoundly revealed the biological basis of sign-activity. If Saussure, laying weight on the analysis of linguistic structure, thought about the sign and its properties statically, with language as the paradigm, and laid the foundation for structuralism, then Peirce dynamically described sign-activity from a phenomenological angle, proposing the triadic relation of sign, object, and interpretant, and opening another line of semiotic thought. Sebeok inherited Peirce’s semiotics and, taking cognitive science as his basis, turned his research vision to the genetic process of the sign, not only deepening semiotic theory but effectively expanding its domain.

Sebeok thought about the genesis of the sign within the relation of interaction between organism and environment. He held that sign-activity is a capacity possessed by all organisms, closely linked and interacting with vital activity; apart from sign-activity, life too ceases. The sign-activity of the organism is at the same time the process of modeling. “The ability to make models is in fact a derivative of semiosis, and can be defined as a species’ ability to produce and comprehend, in its own distinctive way, the specific models it needs to process and integrate perceptual input.” That is, the organism processes and encodes perceptual input through the modeling mode peculiar to its species, and thereby experiences, understands, and grasps the world, and regulates and guides the organism to make a fitting response. The cognitive form peculiar to a species is therefore embodied in that species’ modeling behavior. The process of vital activity in a particular situation is at once the process of modeling, the process of dealing with the world, and the process by which sign-activity produces and experiences meaning. “Semiosis is the neurobiological capacity to produce forms (signs, texts, and so on); modeling is the channel that connects the capacity for semiosis with the representation of a referent (that is, the actual act of creating form).” Sebeok divides modeling systems into three levels: primary, secondary, and tertiary. The primary modeling system constructs forms through “permeation” (spontaneous simulation) and “imitation” (intentional simulation); it is the most basic mode of sign-activity for the human being and for organisms, and all other levels of modeling are founded on it. The secondary modeling system is that which produces indexical (or indicational) and extensional modeling processes. The tertiary modeling system refers chiefly to the highly abstract, symbol-based modeling system. These modeling behaviors of different levels do not exclude or replace one another, but are interdependent, mutually influencing, and mutually cooperating, constituting an interactive system. Except for language, meta-form, and meta-symbol, which are modes peculiar to human sign-activity, the other organisms too enjoy the three levels of modeling system.

In Sebeok’s modeling-systems theory we see several features. First, the understanding of sign-activity is no longer confined to communicative activity, no longer laying weight on sign-representation, but is regarded first of all as the cognitive mode of the organism—the distinctive modeling behavior and mode of life by which the organism deals with the world. This fundamentally changes people’s conception of the sign and expands the domain of semiotics. Second, symbolic modeling is regarded as the medium of the organism’s cognitive activity, as the premise and basis of communicative activity, with an analysis of the inner logical relations of the various types of modeling behavior, so as to construct a complete theory of symbolic-modeling systems. Third, Sebeok both inherits Peirce’s semiotic tradition and does not cling to certain of its concrete practices: unlike Peirce, who divides sign-activity, at the logical level, into the “triadic relation” of sign, sign-object, and interpretant, Sebeok explores the sign-process of how the organism dynamically grasps the world through modeling. More importantly, Peirce’s semiotics is founded on phenomenology, studying the “formal element of phenomena”; Sebeok adopts the research vantage of cognitive science, combining it with neurobiology—that is, he no longer confines sign-activity to the domain of “phenomena” closely linked to “consciousness.” It is precisely these fundamental changes that enabled Sebeok effectively to open up the semiotic horizon, to bring the semiosis of animals into it, and thereby to construct a “global semiotics.”

In the historical context of critiquing linguistic centrism, rational centrism, and anthropocentrism, Sebeok’s “global semiotics” had enormous appeal, enticing some scholars, after “zoosemiosis,” successively to propose “phytosemiosis” and “physiosemiosis,” extending the sign-activity Sebeok had originally confined to the biological world to the whole of nature, including the physical world. But the problem is this: these expansions often graft “global semiotics” onto the root of Peirce’s triadic relation of the sign, which leads to internal contradiction and error in the theory. In his influential Basics of Semiotics, Deely devoted a chapter to expounding “physiosemiosis.” Unlike Sebeok, who remade Peirce’s semiotics with modeling theory, Deely still follows Peirce’s practice, interpreting sign-activity on the basis of the triadic relation. And Peirce’s triadic relation of the sign is in turn founded on phenomenology—that is, it thinks sign-activity within the framework of the relation between consciousness and the object of consciousness. Whether organisms other than the human have consciousness is an unsettled question; therefore the interpretant in the triadic relation seems necessarily to have to be borne by a conscious human being. In a letter to Lady Welby, Peirce says: “I define a sign as anything which is so determined by something else, called its object, and which so determines an effect upon a person—an effect I call its interpretant—the latter being thereby mediately determined by the former. I added the words ‘upon a person’ as a sop to Cerberus, because I despair of making my own broader conception understood.” On one hand Peirce wished to propose a “broader conception” of semiotics; on the other, he held fast to the phenomenological basis and could not break out of the confines of human conscious activity. Once he was compelled to call in a conscious “person” in the sign-process, he in fact confined the sign-process within the circle of human activity. To break this limitation and construct the grand plan of a “global semiotics,” Deely’s practice was to find a way to remove the “person.”

Deely remade the triadic relation of the sign: he replaced the “person” in the interpretant position with a “thing,” so as to remove the “person” Peirce had been compelled to add. For example, in the sign-relation concerning dinosaur fossils, sign (bones), object (dinosaur), and interpretant (suitably activated paleontological training) constitute a semiotic triangle; a person with paleontological knowledge interprets the object dinosaur through the sign bones. This is an instance of human experience. Deely “cleverly” replaces “the paleontologically trained person” with “the geological-fossil structure of the bones,” so that the interpretant is converted into a physical fact rather than a human psychic structure; thus the human is removed from the triadic relation and the physical world’s own triadic relation is re-established—this is what Deely calls “physiosemiosis” (Fig. 1, Fig. 2). On the surface, Deely seems reasonably to expand semiotics to the whole of nature, establishing a “global semiotics” that includes the physical world; yet it contains a huge logical hole: the relation between “the geological-fossil structure of the bones” and “the bones” still needs a “trained person” to discover and interpret—the human remains an irremovable factor in the sign-relation. Only the semiotic triangle has changed: from one triangle it has increased to two interrelated triangles. First, “the geological-fossil structure of the bones” becomes a new sign, and the original bones become the object; their relation must be interpreted by a “geologically trained person” as interpretant, who understands the geological features of the bones (object) through interpreting the information contained in the “geological-fossil structure” (sign)—this constitutes the first semiotic triangle (I). Then a “paleontologically trained person” (who may be the same person before and after, or different cooperating persons) further interprets, on the basis of the result of the first triangle (I), the relation between the bones (sign) and the dinosaur (object), constituting the second triangle (II). In triangle (II), the bones are converted from “object” into a “sign” used to interpret the dinosaur (object). The two interpretations integrate, yielding a fuller interpretation and deeper understanding of the dinosaur (object) (Fig. 3). Although these relations are physical relations that already existed, the discovery and establishment of all of them require human participation, and are not merely a physical process. Not only can the human not be removed, but this human must also have received the necessary professional training.

To evade this difficulty, Deely holds that the triadic relation of the sign is not the theoretical point; the theoretical pillar is the interpretant, and it is the interpretant that determines the sign-process. But if the so-called “sign” (representamen) cannot be distinguished from the object, then the sign is the object itself, and the interpretant is immediately cancelled, for there is no room or need for its existence. The object may be real, imagined, illusory, vaguely chaotic, metaphysical, or even absent—but in every case there must be posited an object distinct from the sign; this is the very condition of legitimacy for the interpretant’s existence. Peirce’s proposal of the triadic relation is precisely the necessary premise for sign-activity to be generated and to exist; none of the three may be lacking, and this triadic relation can apply only to human sign-activity and can hardly be extended to the biological world.

But for an animal, even one lacking the capacity for consciousness and unable to establish a relation between consciousness and the object of consciousness, it can after all make a fitting response to particular stimuli; in Morris’s words, the behavioral response is itself already “interpretation.” (We regard “interpretant” and “interpretation” as two different concepts: “interpretant” refers to a specific position in the triadic relation of the sign, while “interpretation” may refer to the organism’s instinctive response to a stimulus, with no independent position.) Such a situation, though different in nature from human sign-activity and not constituting the triadic relation, is yet similar to sign-activity, and in human eyes may seem interpretable as a sign-process. Morris inherited Peirce’s definition of the sign; to clarify these two different types of semiosis, he distinguished “animal sign-process” from “human sign-process.” He held: “The continuity and discontinuity between human sign-behavior and animal sign-behavior are equally real; the similarity and difference between human sign-behavior and animal sign-behavior are equally real.” “In animal behavior what predominates is the signal, not the sign, while language-signs (and post-linguistic symbols) are mainly, perhaps uniquely, a human achievement.” Morris correctly distinguished “signal” from “sign” and regarded the sign as a “post-linguistic” product. Regrettably, he did not further clarify the difference in nature between the two types of sign-behavior, their mutual relation, and the root of the difference.

As neuroscience, brain science, and cognitive science developed, Sebeok no longer stayed at the phenomenological angle to think about sign-activity, but set out directly from cognitive science, absorbed the views of neurobiology, and entered semiotic analysis by examining modeling behavior. Once he abandoned the phenomenological basis of sign-analysis and took the standpoint of cognitive science, Sebeok transcended the fence of “consciousness,” linked the sign-behavior of animals with that of humans, and successfully constructed his “global semiotics.” Only, from the neurobiological angle one cannot explain the problem of consciousness, nor the root of the distinctiveness of human language. Yet Deely and other scholars, on one hand, inherit Peirce’s view of the sign and follow the phenomenological framework of the relation between consciousness and its object; on the other hand, they take no account of Sebeok’s cognitive-science evasion of the problem of consciousness, while on a phenomenological basis insisting on a “global semiotics” encompassing all organisms and indeed the whole of nature—and so cannot but fall into self-contradiction. Not only does “physiosemiosis” become an erroneous view, but it also effaces the difference in nature between animal and human semiosis. This view, to be sure, caters to the intellectual current of critiquing anthropocentrism, but at the cost of misinterpretation. Deely and others entirely ignore the different natures of the information-process, the signal-process, and the sign-process.

This misinterpretation has its cause, within which lies a confusion of two different research vantages: first, objectively studying the sign-behavior of the organism setting out from the organism itself; second, the organism’s sign-behavior as seen in human eyes. The human being has a special capacity for “symbolization”: he can discover or set up all manner of relations, thereby endow objects with sign-features, and interpret the human world by constructing triadic sign-relations. In the human field of vision, any object, including a physical object, has already been symbolized or re-symbolized. It is precisely through all manner of triadic sign-activities that the human grasps and interprets the world. The human world is the symbolized world. But for the physical world itself, or for the animal itself, this is not so. In the physical world, the relations among things, or the exchanges of energy and information, are by no means sign-behavior; and the animal’s own sign-behavior does not enjoy Peirce’s triadic relation. It is precisely the confusion of these two different vantages—directly equating the phenomena the human observes and understands with the substance of the organism’s sign-behavior, or with the interrelations of the physical world itself—that inevitably leads to crudely granting sign-behavior to all organisms and even to the physical world, and so falls into error.

At the same time, there exist two different definitions and standards of semiotics. One adopts the definition of Peirce’s triadic relation, in which case animal sign-behavior is not sign-activity in the true sense, or, as Zhao Yiheng says, can with some strain be classed under “broad semiotics”; the other adopts Sebeok’s modeling-systems theory, removing the boundaries between consciousness and the unconscious, between intentional behavior and instinctive behavior, between objective and non-objective relations, in which case one may legitimately use “global semiotics” to cover all organisms. Our approach is to synthesize and coordinate the semiotic theories of Peirce, Morris, and Sebeok. In sum: to define sign-activity with Peirce’s triadic relation of the sign; to distinguish, in Morris’s way, human sign-behavior from animal sign-behavior—affirming their continuity and similarity while stressing their discontinuity and difference, and clarifying their qualitative distinction; and to use Sebeok’s notion of “modeling” to regard all the different types of sign-behavior as the organism’s continual process of modeling.

II. Man Is a Language Animal

To clarify the relevant problems concretely, we must begin the discussion from the genetic process of sign-behavior—that is, the process of modeling. Sebeok set great store by the role of “simulation” (imitation) in modeling; here we must make a certain correction. Simulation must be the simulation of an “object,” which already presupposes that all organisms enjoy the capacity for consciousness; otherwise there would be no object of consciousness and no simulation of an object. We hold that modeling occurs in the process of the organism’s unfolding behavior and dealing with the world; it is precisely the relational mode between behavior and world that lets modeling form. A particular behavioral pattern is correlated, coordinated, and matched with a particular external stimulus, establishing a relation of identity—this is modeling. Modeling is the organism’s establishing, by behavior, a mutually adjusting, mutually matching relational model with external stimuli. The appropriate behavioral patterns construct the model, sediment into the organism’s survival experience, and become the organism’s instinct. For the organism, especially the single-celled organism, it is neither necessary nor possible first to simulate the object and then make a behavioral response; rather, on receiving a useful stimulus it directly makes an instinctive response, and the whole process needs no participation of conscious activity—it is itself an unconscious, instinctive process. “An organism does not perceive the object itself, but perceives according to its own peculiar modeling system, which enables it to understand and model the world in its own distinctively biological way.” Other animals are likewise. A cricket has no auditory consciousness of its “song”; through touch it directly relates to its own action. A frog can keenly sense very small, fast-moving objects, such as edible insects, but ignores large, slow-moving ones; the frog’s world is constructed through the correlation of its own behavior with useful stimuli, not through simulation. A herring-gull chick recognizes its parents on the basis of the vertical line of the beak and a horizontal red dot; when it senses a parent alighting beside it, it gives a food-begging cry and pecks. A wooden beak achieves the same effect. That is, the chick recognizes its parents not through simulation, but through a simple useful stimulus that arouses a behavioral response. The organism establishes a particular correlation between its behavior and external stimuli, joining itself with the world into one, mutually adapting, completing the modeling process—and this process passes through no conscious cognition or simulation; it is merely an unconscious instinct, the whole process still below the threshold of consciousness.

Let us further probe the root of the nature of behavior modeling. Behavior is bodily behavior; it cannot exist independently of the body. This nature of behavior also determines the nature of the relational model (modeling) the organism establishes with the world. The organism deals with the world through the medium of behavior and constructs a relational model of identity. Because behavior can only be bodily behavior and lacks independence, the world this relational model constructs likewise cannot be separated from the body, cannot be external to the organism; rather it structures the world and takes it into the body, connecting them into one. That is, the world becomes the natural extension of the organism’s body (Marx). “To perceive the world immediately, at the level where I and the world are not yet distinct and things have not yet been lived in their individuality, means not to stand before something but to be ‘in it, inside it.’” The process by which the organism establishes a particular relation with its world is the modeling process, and this process is itself an instinctive process—the process by which the organism takes the world into itself and constructs unconscious experience. The organism, by its own behavior, connects the inner and outer worlds into one and models them, thereby shaping both, constructing a corresponding, overlapping, fused inner world and outer world. These relational models are transmitted across generations through the genetic mechanism, activated in youth by means of “play,” and continually revised and reconstructed in particular situations so as to adapt to a changing environment, undergoing hierarchical evolution over a long historical process. In this process, not only is the organism’s own experience shaped by behavior, but external stimuli too are screened and reshaped, thereby establishing a relation of mutual adaptation. The reason we stress that the modeling of all other organisms is an unconscious process, constituting instinct rather than a conscious process, is that once an organism enjoyed consciousness, it could equally take its own behavior as an “object of consciousness” and cognize its own behavior—and the cognition of one’s own behavior is reflection, which could generate reason—and this is precisely contrary to the facts.

The human being, as a distinctive organism—with group life, upright walking, tool use, and increased brain capacity—underwent a qualitative leap in modeling. Group life made communicative activity ever more important, and vocal behavior, by its convenience, began to stand out from the other behaviors and gradually became the chief medium of communication. Originally, vocal behavior was only a secondary “accompaniment” appearing alongside bodily behavior, and so it refracted bodily behavior and reflected inner bodily feeling. But vocal behavior is not merely bodily behavior; it is also a behavior that can leave the body and propagate, possessing a dual character—which provided the condition for its gradually gaining relative independence. On one hand, as the initial “language,” vocal behavior is always affective, a sign conveying the actor’s own feeling; on the other, vocal behavior is not subordinate only to body and action—it always concerns the object and situation to which behavior is directed, is subordinate to the relation generated by behavior modeling, and is united with object and situation. For human existential activity, the utility of vocal behavior in expressing subjective state is after all secondary, while how to represent the object is primary. Upright walking and tool use freed the human hands, making behavior ever more refined, diverse, and complex, and the human accumulation of experience about the world ever richer. This experience is also sedimented within vocal behavior. Thus the meaning vocal behavior expresses gradually shifts from bodily feeling toward experience about the world, and with the condensation and abstraction of experience, finally forms the concept. The use of tools especially is like driving a wedge between human and world, both relating and splitting the relation. The disembodiment of vocal behavior, the formation of the concept, and the dissolution of the unified human-world relation by tools—all of this prompts vocal behavior to bud off from the integral bodily behavior, becoming a behavior with relative independence, a symbolizing behavior carrying concepts laden with rich world-experience, a speech-act establishing a brand-new relation between human and world. This is Saussure’s “concept / sound-image.” Between language and the world of meaning there exists an interpretant, which in turn constitutes Peirce’s triadic relation of the sign—that is, language modeling, a distinctive mode of modeling.

Saussure says: “The dualism of linguistic phenomena resides in this duality: sound as sound and sound as sign—the former a physical phenomenon (objective), the latter a physical-psychic phenomenon (subjective), but never the opposition of the ‘physical’ of sound and the ‘spiritual’ of meaning. There is one domain, the inner spiritual domain, where sign and meaning lodge, interdependent and inseparable; and a second domain, the external domain, where there are only ‘signs,’ yet signs reduced to a string of sound waves, which we can only name the sound-image (figure vocale).” This “duality” of language—at once an “inner spiritual domain” and an “external domain”—wins it relative independence and makes it the effective medium for constructing consciousness and the object of consciousness, thereby establishing the objective relation. In fact, in the development from the modeling shared by organisms to the language peculiar to humans, communication plays an extremely important role. It is precisely the need for communication, and the advantage of the dual character peculiar to vocal behavior (at once embodied and disembodied), that let it stand out from the various coordinated bodily behaviors and shift from an “accompanying” behavior to a “dominant” one. When language, by means of sound or other images, gains a materialized body of relative independence, it becomes a relatively stable carrier continually condensing human experience; and only when interpersonal communication continually demands determinacy of meaning does the concept finally take shape. Thus language at last separates from the human body and becomes the primary tool of interpersonal communication. Without the need for communication, without the impetus continually to improve and perfect the communicative tool, there would be no generation and development of language. Conversely, the generation of language as communication reshapes the mode of modeling between human and world, making the world present itself in human consciousness in a brand-new way. From this angle, it is precisely the need for interpersonal communication that prompts the transformation of primary modeling into language modeling in the true sense.

Language is a mode of modeling enjoying relative independence; it differs fundamentally from the original modeling, being a transcendence of the organism’s original modeling mode and the ground of the human being’s becoming human. As a relatively independent mode of modeling, human language no longer takes the world into the self through modeling, constructing it into a unity, but sets the world up as the human being’s object-world—a modeled object-world—so as to interpret and grasp the world. When the language in which sound-image and concept combine enjoys relative independence and externality, and when it projects particular perceptual experience onto this materialized body, it already appears in the manner of the human’s object; and so to name the world is to set the world up as object. The relative independence of language endows language itself with the capacity for “objectification,” and the world thereby becomes the human being’s object-world; from then on an objective relation is established between human and world. At the same time as setting up the object-world, human consciousness too is born. Language, consciousness, and the object of consciousness are synchronously constructed; they are all fruits of the transcendent evolution of modeling activity. Peirce says: “There is no element whatever of man’s consciousness which has not something corresponding to it in the word … the word or sign which man uses is the man himself.” For the first time the world presents itself clearly in human consciousness in the manner of the human’s “object-world.” The relative independence of language also wins language itself a metalinguistic capacity, and in language’s self-consciousness and self-reflection its conventionality and logicality are steadily strengthened. (Some scholars oppose Saussure’s binary structure of language—“concept / sound-image” or “signified / signifier”—to Peirce’s triadic relation of sign, object, and interpretant, holding Peirce’s definition superior. This is in fact a great misreading of Saussure: the two do not constitute an opposition, and there is no question of which is superior—rather they are precisely complementary, the two theories standing at different levels. What Peirce speaks of is the triadic relation of sign-activity constituted by sign, object, and interpretant; what Saussure calls “language” (concept / sound-image) is the “sign” (representamen) element within Peirce’s triad, with object and interpretation bracketed, concentrating on the structure of the sign (representamen) itself. It is precisely the distinctive structure of the linguistic sign (representamen) that endows language with its distinctive nature and function setting it apart from other signs. Only by combining the two theories can we examine more deeply the characteristics of the various sign-activities.)

The generation of human language modeling did not abolish or replace the original modeling modes of the organism; only, in the process of interpreting the world, language modeling occupies the more prominent position, while the various types of modeling still coexist and cooperate, so that the human being enjoys the most diversified modes of modeling. That is, while language modeling proceeds, behavior likewise participates in modeling activity. The human interpretation of the world thus contains intricate and diverse modeling behaviors. Our proposal that “man is a language animal” is no relapse into linguistic centrism or anthropocentrism; rather we hold that the human being has at once multiple modes of modeling: language modeling endows the human with distinctiveness and transcendence, while behavior modeling lets the human keep the features shared with other organisms and preserve sympathy for all organisms and nature. Thus the reflexivity of language prompts the human continually to adjust his own behavior, so as to overcome the maladies of the various centrisms and relieve the predicaments and disasters they cause. This change is substantial and decisive. As a relatively independent mode of modeling, language lets the human transcend present, actual existence, transcend the real world, and construct all manner of possible worlds—an illusory world, and even a purely metaphysical world. The generation of language-modeling activity lifts the human above the whole biological world. Only the human can transcend the self and the real world to create his own worlds in myriad forms and endow the human world with cultural character. And it is precisely the conscious activity born together with language that lets the human, while interpreting the world as an object of consciousness, also interpret and reflect on his own behavior as an object of consciousness, set purposes for behavior so as to adjust and master it, and place behavior between purpose and reason, and within interpersonal interaction, to interrogate its rationality. Human reason is established in just this way. Language-modeling activity changed human behavior itself, so that behavior is no longer merely an instinct but is at the same time conditioned by human consciousness and reason.

III. Behavior, Language, and the Sign

After the generation of language modeling, human behavior not only changed but can be regarded as sign-activity, regarded as “behavior-language”—a mode of sign-representation similar to language. The relative independence of language wins it the capacity for objectification, and so human behavior can be peeled and abstracted from the body to become an object of consciousness. Because human unconscious experience is constructed through behavior, and the modeling process is itself the process of dealing with the world through behavior and of gaining unconscious experience, once behavior is separated from the body and becomes an object of consciousness, it necessarily embodies and represents unconscious experience, and is therefore regarded as a sign-activity, regarded as “behavior-language.” Through behavior-language—including posture, action, expression, and so on—we can discern another’s inner state. Only, behavior-language is still behavior, with the mode of operation and the features of the original behavior modeling. Thus, from the angle of genesis, speech-act buds off from bodily behavior, the fruit of the transcendent development of behavior modeling; while from the angle of the sign, it is language that symbolizes behavior, prompting behavior to be transformed into behavior-language. Not only is human behavior so; the behavior of other organisms too can be viewed in this way. Under the illumination of human consciousness, within the relation-system the human establishes, all organisms can be brought into the human sign-field and given a semiotic interpretation. Behavior originally unconscious and instinctive can, in the human field of vision, be analyzed by Peirce’s triadic sign-structure. Take Pavlov’s experiment with the dog: a bell sounds while food is given; after repeated training, whenever the bell sounds the dog will go to seek food. That is, the dog has reconstructed a relational model. For the dog itself, the bell and the food are one; the bell is part of the food; the dog’s response to the bell is a response to the food. This is a “stimulus–response” instinctive process, not a “sign–interpretant–object” sign-process. But in the human field of vision, bell and food are distinguished as two interrelated, wholly different objects of consciousness; the bell is the representamen of the food, a sign; the dog responds to the bell as sign (representamen) and seeks food, and this behavioral response is the interpretation of the bell-sign. To borrow Morris’s terms, the organism’s modeling behavior originally belongs to the signal-process, but in human eyes is interpreted as a sign-process; the human grants it sign-character, symbolizes it. As the standpoint shifts, the same modeling behavior is given a different interpretation.

The objectifying capacity of language can set up almost all things as the human being’s objects of consciousness, whereby the human can go on to observe, analyze, and cognize the object. This lets the human not only discover the various objective relations existing in nature but also subjectively set up all manner of relations. Once a relation is set up and one attempts to use it to interpret one object by means of another, the triadic relation of the sign is established: one element is regarded as sign (representamen), the interpreted other element becomes the sign-object, language itself bears the most effective interpretant, and the human enters the interpretant position. For instance, between “smoke” and “fire” there is an original natural correlation; when we discover this relation, on seeing the sign “smoke” rising from afar we may infer that “fire” is burning. The relation between “Chanjuan” (the fair lady of the moon) and “the moon,” however, is wholly virtual and non-natural: it is the human who forcibly establishes a correlation between the two through a fictional story, interpreting “the moon” through the story of “Chanjuan,” so that “Chanjuan” becomes a metaphor of “the moon” and a triadic relation is constructed. We may further take “Chanjuan” as “object” to make a further interpretation, so that sign-activity is continually extended in meaning. Signs of the first kind—where the human discovers an objective relation in nature and builds a triadic relation to unfold interpretation—are called “natural signs”; signs of the second kind—where the human subjectively sets up a relation, interprets another object by means of the sign within it, and gains collective recognition—are called “conventional signs.” Moreover, it is precisely the birth of language, the splitting of consciousness from the object of consciousness, that makes it possible for modeling behavior by way of “simulation” to appear in quantity. To set up a relation by constructing a model through “simulation” is itself already a sign-process. In this sense, language is the mother of signs.

“The fact of human evolution not only reaffirms the sign-evolution that had always existed before, but continues a coexisting dual-track sign-evolution: one track needs no language (animal semiosis), the other relates to language (human semiosis). Semiosis should be regarded as a universal phenomenon belonging at once to nature and to culture.” The generation of language does not replace the original modeling modes but works together with them. Before language presents the world as an object-world, behavior has already related to the world, connecting world and human together, fusing them into one, modeling the relation of world and human, “anthropomorphizing” and “vivifying” the world—this is the “humanization of nature” and the “naturalization of the human.” The construction of the various signs is therefore the process in which language (speech-act) and behavior (behavior-language) jointly operate. “Neuroscientists have now found that we possess a tissue called ‘mirror neurons,’ whose function is to respond to the behavior of others … ‘because they blur the boundary between self and other … because these neurons cannot recognize the difference between self and other.’ The nature of empathy is inherent.” One may say that mirror neurons are the fruit of the organism’s behavior modeling in the course of activity. Between human and world two modeling activities of different tendencies are interwoven and overlapping at once: behavior modeling fuses world and human into one, constructing the non-objective relation; language modeling splits the two, constructing an objective relation. A great many symbols are generated through this two-way construction. Facing lofty peaks, what we feel is not closeness but a sense of the sublime; the character of this emotional experience originates in the particular relational model constituted by human behavior modeling, embodied as the astonishment, awe, and consequent passion when behavior modeling is blocked. When language modeling presents the peaks as object, the peaks are already permeated with the emotional color the human has experienced and thereby become a symbol of sublime feeling. The relation between peaks and human emotion is both a natural relation—unconsciously constructed by the behavior modeling of the organism as a part of nature—and a humanly made cultural relation—consciously reconstructed and reinterpreted by the human through language modeling.

“Synesthesia” too originates in this double modeling and is rooted in the homology and similarity of behavior modeling. “In everyday experience, sight, hearing, touch, smell, and taste can often interpenetrate or communicate; the domains of eye, ear, tongue, nose, and body may be without boundary. Colors seem to have temperature, sounds seem to have shape, cold and warm seem to have weight, smells seem to have substance. Such things constantly appear in ordinary language.” Precisely because behavior modeling is homologous and similar, the different senses can transpose and convert into one another. This is also what Rudolf Arnheim calls “isomorphism,” and the “iso-” depends precisely on the homology of behavior modeling. One may say that all signs are rooted at once in the double process of behavior modeling and language modeling, the joint fruit of double modeling. Whether a note, a shape, or a stroke of color, each has already passed through the double modeling of behavior and language to be closely related to the human, and so possesses the sign-function of emotional representation. Lipps’s “empathy” and Groos’s “inner imitation” originate in the double modeling between the human and the object-world. A metaphor that truly moves the heart is not merely constructed by language’s use of the similarity, contiguity, and relevance of objects; behavior modeling necessarily intervenes—it is behavior that first establishes relational models between tenor and vehicle, and because of the commonality of the constructed models, the two then constitute a deep-level metaphorical relation. The metaphor thus constructed not only enlightens the intellect but can shake the human body and reach the depths of the soul; the emotion it gives is hard for words to exhaust. Although language modeling is founded on behavior modeling, because it is the condensation and abstraction of experience and has already been conceptualized, language’s interpretation of behavior modeling is bound to be inadequate, unclear, and plural—always with vagueness, dimness, and meaning-beyond-words, often admitting of multiple interpretations. For instance, the hibiscus blooms in the morning and falls in the evening; this natural phenomenon establishes a double-modeling relation with the human and becomes a representation of his inner emotion. Yet this representation is ambiguous, polysemous, and indeterminate. Qian Zhongshu says: “Just as the hibiscus blooms in the morning and falls in the evening—hence its name ‘lasting-but-a-day’—the Yiwen Leiju, scroll 89, records Su Yan’s poem-preface: ‘I delight in its blossom, yet sigh that its glory does not last the day,’ loving the morning bloom yet, in the end, resenting the evening fall; it also records Dongfang Shuo’s saying: ‘The hibiscus dies at evening and blooms at morn; the scholar too is not forever poor,’ like Bai Juyi’s fifth ‘Fang Yan’: ‘The pine of a thousand years rots in the end; the hibiscus of a single day is yet a glory’—knowing it falls at evening, one still envies the morning bloom.” Qian Zhongshu uses this to illustrate the “two handles and many sides” of metaphorical meaning.

Speaking of bi, xing, and fu, Qian Zhongshu notes that Hu Yin’s Feiran ji, scroll 18, “Letter to Li Shuyi,” records the words of Li Zhongmeng: “Seeking the object to lodge feeling is called bi; touching the object to arouse feeling is called xing; narrating the object to speak feeling is called fu”—words of considerable merit. “‘Touching the object’ seems an unintentional coincidence, picked up at random and put down again, attached to the later text yet not seamlessly joined to it, unlike ‘seeking the object,’ which is deliberately managed, its reasoning smooth and its diction continuous.” On the surface, xing’s “touching the object” seems “an unintentional coincidence, picked up at random”; but in fact there exists between the two objects an isomorphism of behavior modeling. This isomorphism lets two objects, seemingly unconnected and unrelated, “coincide,” even fuse into one, producing emotional resonance—and so it constitutes poetry’s arousal of feeling by touching the object (gan-wu qi-xing). The human grasps and interprets the world through modeling; when he produces all manner of modeling types—behavior, language, and the various other modeling modes—all of them join in the human grasp and interpretation of the world, so that the human world is jointly constructed by all manner of modeling. “For the ordinary person, the great majority of the versions from science, art, and perception differ in some way from that familiar everyday world, which he builds, cutting corners, from fragments of scientific and artistic tradition and from his own struggle for survival. In fact this world is most often taken to be the real one; for reality in a world, like realism in a picture, is largely a matter of habit.” Because signs too are jointly generated by the double modeling of behavior and language, they can in the end all be interpreted from the interaction of behavior-language and speech-act. If we adopt the definition of Peirce’s triadic relation, it is not hard to see that, in the generative process of human modeling, a sequence is formed: behavior modeling, language modeling, other symbolic modeling. Among these, the birth of language causes behavior modeling to undergo a qualitative transformation, so that behavior can at last be regarded as a sign-activity, as behavior-language; and the two, hand in hand, jointly generate all other types of sign-activity. The relation between human and world is woven by all the modeling modes—behavior modeling, language modeling, other symbolic modeling—a net of relations by which the human captures and constructs the human world. Popper’s cultural theory distinguishes “three worlds”: “In Popper’s scheme, World 1 corresponds to the preconscious world of instinct and sense-impression, belonging in essence to the so-called ‘immune’ or ‘biochemical’ self with a sign-like significance; World 2 corresponds to the ‘sign’ self; and World 3 corresponds to the representational end-products of human sign-activity—symbols, institutions, cultural behaviors, and so on. The genetic code of World 1 and the sign-activity codes that govern the conscious states of World 2 and World 3 are qualitatively distinguishable, but, as we have seen, among the three worlds there exists an interconnection formed through extended modeling. In fact the biological dimension, the sign-activity dimension, and the representational dimension of the self are interwoven in a ‘modeling symbiosis,’ in which no dimension (such as the cultural) can replace another (such as the genetic).” Clearly, the animal can enjoy only “World 1,” while the human enjoys three mutually overlapping, mutually influencing worlds.

IV. The Construction of the World and Aesthetic Creation

The human being’s modeling modes are most fully developed, and among them the birth of language is decisive; thereby behavior begins to be regarded as behavior-language and, in coordination with speech-act, jointly creates all manner of sign-activities. It is through all types of sign-activity that the human grasps and understands the human world. The world is jointly constructed by all manner of signs—behavior-language, speech-act, and the other sign-activities. Likewise, the human inner world is jointly constructed through symbolic modeling, that is, through behavior-language, speech-act, and the various sign-activities. Behavior-language constructs a non-object world and the human unconscious world; speech-act chiefly constructs the object-world and the human conscious world; the various sign-activities, in different ways, interweave and overlap the object-world and non-object-world, the conscious world and unconscious world. In this sense, the human world is always already constructed: whatever your eye reaches, your ear hears, your body touches, your nose smells, your tongue tastes, has already been constructed by all manner of signs—and you cannot at all feel the existence of any medium, having long grown accustomed to this constructed world. A distinctive contribution of Sebeok’s modeling-systems theory lies in its changing the theoretical basis of semiotics, regarding the sign as the medium by which the organism deals with the world—a medium that, through modeling, sets up the relational model between organism and world, so as to grasp and understand the world. What people call the “a priori” is in fact the various symbolic modelings, which preset our modes of sensation. “Before we are able to regard the something that lets us make signs as a dynamic object, a noumenon, a brute matter, the pending something has already drawn our attention; it is prior to perception, already sign-like.” Behavior, language, and the other symbolic modelings have long determined the range, form, and order of the senses’ perception of the world. Time-space relations are generated in the process of the double modeling of behavior and language, modeled by modeling activity—for the modeling process is the process by which human and world establish a relation, and so the human is set within time-space relations. The modeling process itself already contains logical rules and presupposes the a priori. And because of the plurality of symbolic modeling and its interaction, it inevitably drives the change and differentiation of that unified, determinate mode of perception, and even finally deconstructs its stability—which precisely provides the occasion for literature and art to innovate using symbolic modeling.

The distinctiveness of the human being lies further in this: he can continually produce all manner of signs and, in all manner of ways, reconstruct the world—reshaping the human world with scientific signs, religious signs, literary-artistic signs, purely abstract signs, and various practical or non-practical signs, even creating all manner of non-existent, infinite possible worlds, and making these created possible worlds stand out before us in an unfamiliar way, compelling us to re-examine, rethink, and reinterpret them. The freedom of sign-creation comes precisely from the relative independence the sign possesses. Were this relative independence lost, the human would have to crawl back onto the ground of reality, utterly unable to realize even the slightest transcendence. Literature and art are the defamiliarized possible worlds the human creates using signs; they arouse the desire to re-contemplate and reinterpret this world, thereby changing the human being’s habituation and unconsciousness, and letting him view the real world in a brand-new way. “A work is used to change the way those who see it view the world. To hold that the world is an unjust place is, in fact, to move people and then to change it.” Although literature and art are a transcendent possible world, even one that can run parallel to the real world without intersection, both worlds are constructed by human sign-activity; symbolic modeling and its normativity determine both the real world and all possible worlds, and the literary-artistic world. Our discussion of the various worlds must therefore first discuss human modeling and its normativity—must explore the norms of behavior-language, speech-act, and the various other sign-activities. Common modeling norms establish an inner connection between the literary-artistic world and the real world, so that the two can mutually map and mutually interpret. And it is precisely the normativity of behavior that connects the relation between a work’s characters and real persons: our interpretation of the behavioral norms of a work’s characters at the same time reveals the behavioral modes of particular groups in real life and their implicit ideological meaning. Two originally separate lives communicate and understand each other by way of behavior and its normativity. As Merleau-Ponty says: “What he can see is the factor of the inner becoming outer, by which we enter into others and into the world as world, and others enter into our transformation or reorientation—that is, behavior.” The reason literature and art can react upon real society lies in this: the work, by means of detailed description, profoundly displays the behavioral norms of its characters, foregrounding those behavioral norms that have already been habituated and made unconscious in everyday life, and interpreting and critiquing them—in fact thereby already interpreting and critiquing the behavior of particular groups of people, thus raising human behavioral self-consciousness, critiquing the ideological content implicit therein, and correcting the various alienated behaviors. Here we discover the relation between the autonomy and heteronomy of literature and art: from the angle of literature and art’s constructing a possible world parallel to the real world, literary-artistic activity is an autonomous activity within this possible world, undisturbed by the real world, enjoying the features of freedom and autonomy; but from the angle of the normativity of the modeling activity in the two worlds, in both worlds the norms are shared, all deriving from real modeling practice, possessing contextuality and heteronomy, and being only the norms of particular social groups in particular contexts—hence “institutional,” refracting an ineradicable ideological character, and thereby producing the connection and mapping between the literary-artistic world and the real world. Literature and art are therefore at once autonomous and heteronomous.

Fundamentally, the aesthetic world of literature and art is jointly constructed by behavior-language and speech-act, because all artistic signs—literary, musical, painterly, sculptural, dance signs—are founded on behavior-language and speech-act, the fruit of the double modeling of behavior and language. Literary-artistic activity therefore first of all embodies the features of these two utterly different modeling modes: speech-act lets the appreciator place the work in the position of object to contemplate and appraise, keeping a rational distance from the object, savoring every word and phrase, every chapter, every note, every melody, every brushstroke, every color, every form, every dance-posture, pondering the mystery contained within and seeking intellectual enlightenment; behavior-language lets the appreciator throw himself in, immerse himself, forget himself, and receive the tremor of life and the cultivation of feeling. Human behavior is bodily behavior, and the form of behavior-language is the form of human life; the literary-artistic world therefore enjoys what Susanne Langer calls the “form of life.” In the two-way construction of aesthetic activity, the boundary between consciousness and the unconscious is dissolved, the boundary between self and non-self is effaced, the cleavage between body and spirit is healed, the object-world and non-object-world overlap; literary-artistic activity involves all the human being’s modeling modes, so that all the human being’s creativity is mobilized and all his vitality is kindled, thereby unfolding one boundless, vitality-filled possible world after another. Literary-artistic activity dwells in these possible worlds, in this both-this-and-that transitional state and marginal space; it paradoxically coordinates and fuses the seemingly conflicting modeling modes, endowing the human with the most richly diverse modes of existence and the broadest, most boundless space of existence. This is the aesthetic state and the aesthetic space, the freest and most splendid blossoming of human life. Aesthetic activity therefore has at once the contemplation, appraisal, and practice colored by epistemology and praxis, and the immersion, lived experience, and apprehension colored by ontology; it is the shifting and fusion between the two modes and the two relations, and to neglect either side is already to deviate from the essence of aesthetic activity.

Literature, as an art of language, must have language modeling as its dominant. Yet the construction of the literary world is by no means the single behavior of language modeling; it likewise contains complex and plural sign-activity. What we call “poetic language” is a language that fully uses all manner of means to summon the other modeling modes—especially behavior-language—to participate in creation; it is a language in which speech-act invites behavior-language to cooperate. Literariness and poeticness are born precisely in this process of cooperation, deeply generated by the cooperation of two modeling activities. These means refer not only to the rhetoric of language itself but also to context, mode of reception, and many other factors. When speech-act summons behavior-language to construct a literary possible world together, it also sets the necessary condition for the human to experience this world deeply. It is precisely the participation of behavior-language that fuses human and world into one, lets the human immerse and be intoxicated within it, the soul enlightened, the body shaken—and literariness and poeticness thereby arise of themselves. Scholars have always focused only on language itself to seek literariness and poeticness, which is bound to be futile; for when you examine the text and language as “object,” stripping text and language away from context and receiver, they instantly wither and shrivel, behavior modeling having ceased to exist in this state of cleavage, and literariness and poeticness thereby vanish. The literary world jointly constructed with the participation of behavior-language is a world that language alone cannot interpret, brimming with a misty atmosphere and an inexhaustible flavor—an atmosphere and flavor that can only be experienced in person and are hard to put into words.

Hegel made this demand of the artist’s creation: “The only requirement is that the content of art should be, for the artist, the substantial, the inmost truth of his consciousness, and that he should recognize the necessity of a certain mode of expression. For the artist in his creation is also a natural thing; his artistic skill is also a natural (innate) talent; his activity is not a conceptual activity wholly opposed to its sensuous material, joined to the material only in pure thought, but is one that has not yet lost the face of nature, closely combined with the object, believing in the object, identifying his inmost self with the object. At this point subjectivity wholly permeates the object, and the artwork too springs wholly from the genius’s complete inner state and force; such a product is firm, unwavering, concentrating its full robust spirit into itself. This is the basic condition of complete art.” Hegel vividly describes the artist’s creative activity, pointing out that although artistic creation is closely related to the object of consciousness, it is yet not a conceptual activity—the artist has become a “natural thing,” identified with the object, “wholly permeating the object.” This clearly expounds that artistic activity needs not only observation and cognition of the object, not only speech-act, but is likewise inseparable from behavior-language, needing, through behavior modeling, to combine and identify the creative subject with the object, fusing into one. Only thus is a complete, vitality-suffused artwork created. In aesthetic activity, the contest between the two different modeling modes (speech-act and behavior-language) and the different relational modes (objective relation and non-objective relation) becomes the irrepressible main impetus of the genesis and evolution of literature and art. Hegel divided art into three developmental stages—symbolic, classical, and romantic—holding that they embody three relations of idea and form. “Symbolic art gropes for the perfect unity of inner meaning and outer form; classical art finds this unity in expressing the individuality of substantial content as an object of sensuous contemplation; and romantic art, in foregrounding spirituality, again exceeds this unity.” Because spirit finally transcends sensuous form and forms art’s self-consciousness, “art ends with the appearance of its own philosophy.” Although Hegel’s deductive, subjective summary of the three stages of artistic development has been criticized by later generations, it reasonably reveals the overall trend of art history. In fact, behind the development and evolution of art lies precisely the change of the modes of symbolic modeling and their normativity, especially the change in the contest between behavior-language and speech-act. In the course of artistic evolution, speech-act occupies an ever more important position while behavior-language progressively declines, thereby producing the staged phenomenon of artistic development and transforming art from appreciation and lived experience into pure meaning-interpretation, and even into self-interpretation of the artistic process, producing the “end of art” and the “end of aesthetics.” Alongside this overall evolution, it also incurs the resistance of behavior-language; we are not hard put to see the rise of body aesthetics, the expansion of aesthetics, and the aestheticization of everyday life. As for the change in the normativity that symbolic modeling itself enjoys, it is in a relatively secondary position, displayed mainly as the evolution of artistic style. In the internet age, signs are over-produced, over-copied, and over-transmitted, and the mode of the body’s presence, the relational mode, and the mode of modeling also undergo great change—all of which makes our world increasingly an aestheticized, illusory world.