by
This manuscript has been read and accepted for the Graduate Faculty in English in satisfaction of the dissertation requirement for the degree of Doctor of Philosophy.
Supervisory Committee: Professor James P. Crutchfield, Professor Angus Fletcher, Professor Joan Richardson, Professor Eve Kosofsky Sedgwick
Although telos has been variously interpreted throughout history, I argue that it has consistently involved the concept of chance. To link telos with chance goes directly against the grain of contemporary thought that associates teleology with rigidity. Thus, I must ask the reader to be prepared to meet with what will appear at first to be conceptual noise, defying and contradicting what everyone believes to be true. In time, the ideas should become clear. It should also become clear that no one definition of telos that has been made so far corresponds with the definition that I present here. I analyze a number of different teleologists in light of recent advances in the sciences of complexity and evolutionary theory. The reader should not expect to find the traditional Aristotle, Kant, or Emerson here. What the reader will find, however, is that there are new ways to consider the moments of genius and of folly in traditional teleological arguments.
The first chapter functions as an overture to the entire work. The reader should plow through it without worrying too much about comprehension at first. The same ideas will be repeated in various forms and examples again and again throughout the later chapters. Since my thesis explores unfamiliar territory for most literature scholars, my strategy has been simply to repeat myself enough times and in a variety of ways that, in the end, the argument will, I hope, seem like an old one. I have chosen to use the term telos, even though I am redefining it, in order to link my subject with the long history of teleology.
Part One consists of three chapters that define telos and summarize its history. The next four chapters, which make up Part Two, each further define and reexamine a different kind of teleology. In each of these chapters, a narrative is analyzed in terms of the teleology it supposes.
All seven chapters begin with an abstract, and the reader may find it helpful to look at each of these before beginning the book. I have not provided a concluding chapter summarizing the work. For that purpose, I suggest the reader reread to the introduction.
I would like to thank the Jewish Foundation for the Education of Women, the Dactyl Foundation for the Arts and Humanities, the Santa Fe Institute, and the Art and Science Laboratory for supporting this research. I would also like to acknowledge Jim Crutchfield for his patience and willingness to work with me on a daily basis for more than a year, helping me understand the strange new world of deterministic chaos, computational mechanics, and complexity. I also thank Neil Grayson and Michelle and Angus Fletcher, who spent many long hours fleshing out these ideas with me, and especially Joan Richardson, who gave me my first real introduction to science and whose own work confirms my belief that science is indispensable to the study of literature.
Abstract. One of the prime concerns of narrative theory is the issue of intention. A crisis situation has existed in the arts since the time Roland Barthes in "The Death of the Author" (1968) argued that the notion of authorial intention in narrative is as untenable as teleology in the natural world. In response, contemporary movements have attempted to promote "non-teleological" art. The resultant work may appear less predictable but does not necessarily appear less willed. "Chance operational" art can seem teleological if coincidental patterns happen to seem meaningful. Such patterns seem caused only by the (reader's) purpose they come to serve in the end. Conversely, predictable works of art may not appear intentional, in the sense of willed by the artist, but determined merely by known laws of genre, grammar, convention, and so forth. The following study attempts to identify the characteristics of art that make it seem teleological. I argue that telos involves two distinct mechanisms--which are seldom both noted in traditional arguments--one for the maintenance of order and one for the discovery of new order, which I refer to as directionality and originality, respectively. It is here assumed that telic systems are formed according to mechanistic laws that arise spontaneously from disorder. In turn, law-abiding systems come to function in advantageous ways not predicted by those laws. These two aspects, emergent lawfulness and adaptability, make natural systems telic, that is, progressive or creatively organized toward goals. In this view, only when activity involves both directionality and originality can it be called intentional or, in my view, artistic. With these considerations in mind, I reinvestigate the history of teleology and its influence on narrative aesthetics.
Telos is a concept with a complicated history. A number of distinct species of the term have evolved since Aristotle attempted to describe final causality, the purposes of nature.[1] Under Christianity, telos was associated with the mysterious ways of Providence. Under classical determinism, it referred to sequential, mechanistically predetermined design. Under postmodernism, the teleological view of things seemed a mere culturally-induced hallucination. More recently, a new use of telos, referring to the phenomenon of spontaneous self-organization, has emerged in the science of nonlinear dynamics. The fact that this single word can refer to several apparently contradictory ideas signals a hidden richness. Dismissed by Charles Darwin, and then by Jacques Derrida, telos keeps returning like a misunderstood ghost. It demands a clearer conceptualization, especially considering the fact that contemporary proscriptions against so-called teleological art are based on a limited understanding.
Generally speaking, critics claim teleology assumes some original external cause, some Designer that sets the whole machinery of the universe in motion. Misunderstood this way, final causality is said to be "linear," consisting in direct cause and effect relationships, predictable, proportional, and reductive. As I hope to demonstrate, teleology and intention are not bound up with the classical notion of causality and "linear" determinism. Teleologists throughout history have typically described telos as some sort of internal constraint or guiding principle, not as the direct effect of the hand of an external agent.
Postmodern theories assert that the conception of intention assumes a rational static Humanistic "self," a conception that is no longer tenable (under the early 20th century understanding of causality). Subsequently, the long-running efforts to remove teleological language from biological and historical discourses,[2] were extended into the aesthetic discourses. To put it another way, by the same logic that argues against the existence of an Author of the universe, it is supposed that there is no author of a work of fiction. Nowadays, cultural theorists try to avoid discussing the function, utility, end, intention, or logical reasons for various aspects of visual, narrative, or poetic language. The most enlightened director, painter, or writer, according to many institutions that now support the arts, does not pursue a "teleological" final product, but instead attempts to represent "life as process" of ever changing interpretations.[3] Teleological explanations are said to be "linear" and, as such, artistically inferior. It is assumed that telos should be expunged from a true work of art, since it would represent an attempt to impose an external, artificial, and subjective principle upon reality.
Such a rejection of teleology shows that these critics are unfamiliar with the practices of teleology. Historically, teleologists have represented their explanations with a reflexive series, parts < -- > whole < -- > parts, or a cyclical series, A - > B- > C-> A, not by a linear causal chain, A-> B-> C-> D. Or, to be more precise, in teleology, A is thought to be one of a number of indirect causes of C, but A, as such, cannot exist independently of C because C determines the function of A. Because A is a part of C, it owes its existence to C as the whole. Organisms, for example, are telic in the sense that their individual cells and organs owe their existences to the animal as a functional whole. Many teleologists have thought of telos as a product of feedback, not as a cause separate from or in the future of the process it guides. The most remarkable and most overlooked fact is that telos is always described as the result of the random interactions of the parts. The importance of chance in teleological explanations is something that has been generally ignored, if not denied. I seek to change this situation.
Postmodern literary and cultural critics are not the only ones who have given teleology an undeserved bad name. Generally speaking, 20th century philosophers have tended to consider intention in terms of "goal direction." This metaphor conjures up an actually existing material object that is spatially and temporally removed from the agent. In their discussion of basic conceptual metaphors, George Lakoff and Mark Turner write, "Take for example, PURPOSES ARE DESTINATIONS… It is virtually unthinkable for any speaker of English (as well as many other languages) to dispense with [this metaphor] for conceptualizing purpose…. To do so would be to change utterly the way we think about goals…" [1] Changing the way we think about intentional behavior is precisely what this work seeks to do. Those that try to defend teleology along these lines claim an agent does A in order to do B in order to accomplish the final goal of C, assuming that A and B would not have occurred without the intention toward C.[4] The goal metaphor makes it appear as if C exists in the future and attracts the actor toward it. But the "goal" may be understood as just another motivating factor, an idea in the agent's mind, existing in the present, which propels the agent along. Upon reviewing this literature, I find none makes a valid argument for the necessity of nonreductive teleological explanations.[5] C, as described in these analyses, is just the last point in a linear sequence, not something radically different from what came before. This would have ended the discussion entirely were it not for changes in science, which have led to the reassessment of reductionism.
It is pointless to try to "naturalize" teleology, as so many 20th century philosophers have attempted to do, by making it conform to classical determinism. All true teleologies must be nonreductive: teleologists claim the whole is greater than the sum of the parts because telos adds "something more." My belief that teleology needs to be reconsidered and redefined is based on recent findings in nonlinear dynamics research which indicate that complex systems are indeed "something more" than reductive analyses of their individual parts would predict. I reformulate the question of intention this way: Can an end state C emerge that cannot be reduced to the sum of the states that precede it? Does this emergent property, C, therefore, require that some internal constraints be posited to guide the irreducible interactions of all that contribute to C? Posed this way, I avoid psychologizing the problem: I do not use the misleading "goal" metaphor; I do not ask who does what for the sake of what. When I ask the question, Are humans intentional beings? I am asking whether or not humans are capable of emergent (and thus free) organized and directed behavior that cannot be described reductively. My answer is yes.
In regard to literature, to an extent, I can agree with Barthes that at times an author can be an automaton, and the forces of language, culture, and literary convention may merely interact in him or her producing various textual effects.[6] A Barthesian text is "a multi-dimensional space in which a variety of writings, none of them original, blend and clash. The text is a tissue of quotations drawn from the innumerable centres of culture" (146). I depart from Barthes in my argument that multi-dimensional spaces can be constrained in such a way that they tend to arrive at relatively few end states (I refer to this as "directionality"). I also depart from Barthes in my belief that original writings are possible. Sometimes the normal forces that compel language use can interact in an unexpected way and this creates a new linguistic effect (I refer to this as "originality"), not unlike a Darwinian adaptation.
The misinformed prejudice against "teleological art" is unfortunate because teleology is actually a useful and extremely subtle tool for exploring what it means to create something truly new and yet meaningful. In particular, teleology has shown that the concept of intention is inhabited by two distinct kinds of behavior. The first is associated with directionality, or the mechanisms for maintaining order. The second is associated with originality, or the discovery of wholly new functions. As I argue in Chapter Two, teleologists, from Aristotle, to Immanuel Kant, to Henri Bergson, have noted that the mechanistic laws of nature seem to develop in a more limited direction toward forms with increased functionality than the laws themselves imply. There seems to be an internal force or mechanism that guides development, in biology for example, both in the individual and in species adaptation. This directional-internalist theory of change suggests that intrinsic factors drive evolution in predetermined directions.[7] Teleologists have also noted that nature seems intentional because it often creates original systems that seem to anticipate unpredictable future needs by reinterpreting old forms for new uses. That is, species seem to be able to plan for the future; they adapt to their environments, often becoming more sophisticated.
My twofold view of teleology has been visited before, but studies have tended to see these two aspects as contradictory rather than working together. As just one example, in the early 19th century, teleologist Richard Owen conceived of two principles: one, which he considered directional -- but not teleological -- brings about stability and similarity or "a vegetative repetition of structure"; the other, which he called simply teleological, brings about change and diversity by shaping a system according to its function. He claimed the former is illustrated in the conception of a groundplan or archetype and in the mathematical symmetry of some organisms and crystals. The latter is illustrated in the conceptions of adaptation and progress. Owen argued that directionality results from a "polarizing force," which produces similarity of forms across species, while originality results from an adaptive "special generalizing force," which produces the diversity of organic forms. Owen did not view these forces as sides of the same coin as I do.
The distinction between directionality and originality is perhaps partially derived from Kant's conceptions of the aesthetic and the teleological judgments, which informed the arguments between transcendental morphologists (for form) and Kantian teleomechanists (for function).[8] We will return to this dialogue between formal causality (involving a governing eidos as rational principle) and final causality (involving a governing eidos as purpose) throughout. It is also noted that Aristotle's notion of final cause cannot be clearly separated from his notion of formal cause. As Aristotle explains,
If then it is both by nature and for an end that ... plants grow leaves for the sake of the fruit ..., it is plain that this kind of cause is operative in things which come to be and are by nature. And since "nature" means two things, the matter and the form, of which the latter is the end, and since all the rest is for the sake of the end, the form must be the cause in the sense of "that for the sake of which."[9]
In this work, formal cause is somewhat conflated with final cause, insofar as the "end state" ("perfection" to Aristotle) may be considered the laws of structure (form) as a whole that, through feedback, determine the parts. This describes the aspect of telic directionality. The ultimate utility or purpose that the end state or whole happens to serve (a purpose other than that of merely maintaining itself as a whole) further determines the parts. This describes the aspect of telic originality. Although some teleologists, like Owen, have considered only original changes telic, originality can only occur against a background of directionality.
19th century teleology was mostly concerned to account for the differences and relationships between species. The transcendental morphologists were concerned with structures as such, not with the meaning or purpose of structure. They were interested in directionality, in the neutral interrelations between parts and wholes, in how, for example, the increase in size in one organ might be correlated with a decrease in size in another organ. They believed that the direction of evolution was predetermined by intrinsic factors. They argued that there exist archetypes from which the forms of animals are derived. They thought that original (Platonic) forms were varied according to mathematical laws that govern species differentiation. (They did not consider why change might occur in response to the way a particular structure functions in an environment. Thus, they were not, like other teleologists, interested in what I call originality.)
Although the transcendental morphologists' claims were rejected after Darwin, research in phenomena involving directionality has made a recent comeback. Nonlinear dynamics theorists now argue that there are relatively few archetypical kinds of behaviors and patterns that result from self-organizing feedback processes. This leads to the appearance in nature of what transcendental morphologist-teleologists have called "variations on a theme." For example, similar classes of chemical reaction-diffusion processes are responsible for stripes on a tiger and stripes on a tiger swallowtail butterfly.[10] Similar laws guiding biological development produce "gills" on mammal embryos as well as gills on fish. These similarities do not necessarily imply a common ancestor, but a common law. The idea of variations on a theme is further explored in terms of its use in literature in Chapter Five.
While the transcendental morphologists were primarily interested in similarities between species (how individuals conform to an ideal), other teleologists were interested in the differences between species (why individuals depart from an ideal). These teleologists also considered the role of function in determining the morphological changes in animals. They were interested in the original aspect of telos, which consists in how an organized whole or neutral pattern (produced by directionality) of one system is used by another system to its advantage. Work in this field was, of course, taken over by Darwinian evolutionary theorists.
The original aspect is apparent when extrinsic factors (e.g., environmental pressures) drive evolution in new directions, which are not predetermined but are constrained by available patterns. Spontaneously formed "eye-spots" on a butterfly wing scare off a predator that mistakes them for owl's eyes. This predator's use may be considered a form of interpretation because it makes the systems being used (i.e., the butterfly's "eye-spots") function in a way that that they could never have functioned on their own. In both cases, involving directionality or originality, the resulting extremely fortunate function or behavior may seem, in retrospect, predetermined or intentionally designed for an ultimate purpose. In this way, products of nature seem like works of art.
If a work were merely directional, like genre fiction,[11] it would be too predictable and conventional to be considered art. If, however, a work were completely original, like a dream, eschewing the use of any (or any known) pre-existing structures, it would be unintelligible. Therefore, art must be both directional and original if it is to be perceived as communicating a new message, having an intention. This may be an obvious statement within the discourse of aesthetic theory, and it should be as obvious within the context of the freewill debate, but it is not. Though the notion of human intention (of behaving in a directed but free manner) should include descriptions of both directional and original behavior, most theories of intention tend to argue for one type of behavior at the expense of the other. This insensitivity to the different aspects of intention is pervasive throughout discussions of teleology. The two distinct aspects of telic behavior that I have identified are frequently collapsed into the single term telos and used in a confused manner. It is time, then, to look again at the history of teleological arguments with these two aspects in mind. If we do this, we will find that the field is actually much richer than we ever suspected.
Although Aristotle's telos was long ago rejected as a scientific principle, it cannot be denied that there has emerged the narrative of humankind, which, as physicist Paul Davies has gravely remarked, "seems almost contrived."[12] Some believe that story-like organization cannot arise spontaneously in a world that is governed by chance and necessity alone. Thus, when lucky accidents occur, they suppose that some other force is at work, limiting or guiding chance and putting the fix in on physical laws to suit purposes. According to Jacques Monod, there is no explanation for the fact that the universe is more organized than seems probable.[13] He insists that our existence is simply as lucky as winning a lottery. In Laws of the Game: How the Principles of Nature Govern Chance (1975), Manfred Eigen and Ruthild Winkler object to Monod along structural evolutionary theory lines, arguing,
it is not a matter of hitting the jackpot right off.... The competitive behavior inherent in the laws of selection and evolution limited the number of blanks [nonwinning numbers] and prevented the majority of them from even getting into the game in the first place. [14]
The fact that organization and complexity arise through selection is not as surprising as Monod claims.
Today the goal of structural evolutionary theorists, like that of some 19th century secular morphologist-teleologists, is to elucidate the "principles of organization" that result in the repeated appearance of similar biological structures. They study the energetic, mechanical, morphogenetic constraints that limit the kinds of biological structures that nature can produce. As noted above, like teleologists, the structuralists contend that these constraints result in a relatively small number of structural archetypes considering the multi-dimensional space in which they evolve. Thus, if there were a film version of Earth's evolution that could be rewound and run again, many of the forms we know today would reappear.[15] Structural archetypes occur throughout nature, animate and inanimate, at the microscopic as well as macroscopic level. Also called structural attractors, they are sometimes compared to Platonic forms because they exist, as concepts, prior to the process of natural selection.[16] It turns out that evolutionary forms are not as deeply contingent upon external agents and environmental pressures as Darwinists have argued. The task of the biologist today, then, is to discover which forms are likely to appear. Only then is it worth asking which of them will be selected by environmental conditions.
Monod was wrong about our luck. The odds are stacked in our favor. No teleologist was ever able to discover the mechanisms that result in this telic directionality. But today we know that the nonlinear relationships of elements in dynamical stochastic systems can produce order spontaneously. That is, more order arises than a simple probabilistic calculation would have predicted from the initial and boundary conditions.
Thus to conclude this section, the same kind of organization can arise from a number of initial conditions. Thus, one can also expect that various readers with various interests and experiences can arrive at a similar meaning of a text. Meaning, in turn, can be original insofar as it is emergent, not reducible to the sum of its parts. What others have identified as telic order is seen here as a relatively objective phenomenon determined by the global effect of local stochastic interactions over time within a single system (directionality) or by the way a singular coincidence between two separately evolved systems is interpreted or used by one or the other (originality). The former focuses on the phenomena of spontaneous self-organization, the latter on adaptive evolution.[17] This twofold view of telos does not assume emergent order is inevitable or necessarily progressive, but the fact that it appears so retrospectively is what defines the telic aspect as such.
Teleology's twin aspects of directionality and originality involve a complex contest between structure and randomness, causality and chance, mechanism and intention. It is not surprising then, that while some argue intentional behavior is compatible only with determinism,[18] there are equal numbers who argue that it is only possible if there is some degree of indeterminism.[19] These contradictory views have long since been part of the paradox surrounding the notion of telos. It is my feeling that these contradictions were unknowingly planted by Aristotle in Physics with his argument that telos guides events that are probable, which for him seems to have meant both morally sensible[20] and aesthetically pleasing and, what we would now call, statistically likely. Aristotle's twofold sense of probability contained an unarticulated connection between utility, determinism, and chance that has variously expressed itself over the years. The "sensible" side of telos is here associated with originality, willfulness, and the reasons why things are the way they are. The "statistically likely" side of telos is here associated with directionality, physical necessity, and descriptions of how things become the way they are.
There are a number of distinct versions of teleology--each corresponds roughly to a different theory of causality--which I group into five broad categories: Aristotelian, analogical determinism, deterministic fortuity, pragmatism, and self-organization. (There are also the theories of causality associated with mechanistic determinism and radical indeterminacy, which do not have a teleology attached.) I explain these distinctions in detail below. But first, I would like to note that teleologists can be parsed into two main schools. Some, which I call the mentalists, were concerned with originality; and others, which I call the nonmentalists, were concerned with directionality.[21] We may say that the first speaks more to the interests of art, the second to science. We may further say that the mentalists tend to posit an actual extrinsic agent as efficient cause of telic phenomena. The nonmentalists tend to argue telos is an intrinsic rational principle, not due to an external efficient cause or a physical agent.
A simple analogy will help to illustrate the distinction between nonmentalism and mentalism. The sequence 2, 3, 5, 7, 11, 13, 17, 19, 23 is governed by a rule. Each number is a prime number. This sequence may be compared to the laws that govern systems that spontaneously self-organize; you can discover the laws by examining the process of pattern formation itself. (This is just an analogy; nonlinear dynamics is more complicated.) I also note that the rule is rational and therefore seems like something produced by intelligence. If we examine the sequence 14, 23, 28, 33, 42, 51, 59, 68, 77, 86 we find a different case, which may be associated with mentalism. It is also governed by a rule, but its rule cannot be discovered by examining the sequence itself: the numbers represent stops on the Lexington Avenue subway line in New York City.[22] The rule for this sequence is imposed on the sequence from without by an intelligent agent.
In the nonmental view, teleological phenomena are analogous to, though categorically different from, products made intentionally by human artisans. Thinking of telic phenomena in the natural world as if they were created by an intentional being or organizing intelligence gives one to understand them as designed objects that have rational functions. Kant's teleology, for example, distinguished between works of art and organisms. Art implies an extrinsic cause "distinct from the matter, or parts." Organisms "combine of themselves into the unity of a whole." Art is designed. Animals are not. According to Kant, "We only read this conception into the facts as a guide to judgement."[23]
In the mentalist view, as developed largely by Christian theologians, such as St. Augustine and St. Thomas Aquinas, everything is caused by the (sometimes strange) logic of divine will. God exists outside of human time. He experiences eternity in an instant, while human beings experience sequential time. Because God is not constrained by time or space, He can make causal connections that do not conform to the physical laws (cause precedes effect) in human existence. This kind of teleology is referred to as mentalism, because telic phenomena are controlled by the thought of an external supernatural agent, an efficient cause. A nonmental version of Christian teleology is also possible if God is considered to be an immanent nonphysical intrinsic force rather than an external agent.
Another kind of externalized telic cause needs also be mentioned. Although Kant argued that telic forces were given in the interactions of the individual parts of a system, a few of his followers in biology began to look for an actual physical source of the "principle of organization" in a germ cell or a seed. In this view, teleological phenomena are set in motion by a physically well-defined initial condition. A seed unfolds its already determined future. However, this just begs the question of what caused the seed. Again, it seems one would have to posit some external intelligence.
It may be true that there were a few so-called teleologists who sought a fixed, stable source for the organizing principle, but generally, only those making arguments against teleology tend to characterize it this way. Genuine teleology seeks to escape reductive analyses. It seeks, in additional to material causes, evidence for an emergent vital force immanent in the process itself. It seeks internal final causes not external efficient causes. As vitalist Bergson writes in 1907, predetermined teleology
implies that things and beings merely realize a programme previously arranged Ö As in the mechanistic hypothesis, here again it is supposed that all is given. Finalism thus understood is only inverted mechanism. [24]
Another example of a predetermined mechanistic form of teleology would be Monod's notion of telemony. Monod attributes the telemonic character of all living things to the fact that the structures of molecules allow only one kind of behavior in a particular situation. These powers of "discrimination" result in limiting factors and an orientation toward invariance of molecular behavior and, subsequently, in an organism's ability to maintain homeostasis and produce offspring like itself. Furthermore, in Monod's reductive view of self-determination, a so-called "agent" is compelled to act in a specific way by virtue of its physical constitution. Thus, he sees a type of telic "predetermination" in biological organisms. [25] However, the power of "choice" that Monod attributes to molecular activities is really no choice at all for the laws of physics, chemistry, and so forth compel the behavior. Monod's agents are essentially automata. Although Monod has argued that "chance alone is at the source of every innovation, of all creation in the biosphere" and that predeterminism is transcended when a new mutation, resulting ultimately from "quantum perturbations," allows for adaptability and the apparently progressive nature of evolution,[26] Monod claims the concept of chance is opposed to the notion of intention. Thus, we may say that his view of intention, which is mainly directional, excludes the possibility of innovation, creativity, and originality.
To reiterate then, the weakest kinds of teleologies are those that posit either a metaphysical presence, or an actual physical agent or a physical "seed" or blueprint that determines telic behavior. This begs the question of what is the cause of this originary cause. To explain such things one may feel compelled to posit an Author or Designer. Strong teleologies require no further explanation of final cause; it is an emergent phenomenon, given in the interactions of the individual parts of a system. In my opinion, weak teleologies are not true teleologies at all.
Furthermore, it is argued here that the weakest forms of teleology preclude a notion of human intention. intention cannot exist in a timeless Christian universe where everything has always already occurred. This is the point of difficulty with all religious systems that espouse an omniscient and beneficent god who determines all things from the outside.[27] Likewise, intentional selfhood cannot exist in a classical mechanistic universe where every event is essentially the direct result of the sum of what went before. There is no room for a theory of human intention in Monod's model (in which there is a direct relationship between a gene, its product, and its function) any more than there would be in the Christian model (in which all is directly determined by God). In a classical determinist universe, presumably, one could predict how an agent would react to new situations, as long as one has sufficient information. Insofar as this strong deterministic view holds, actions as well as natural events, then, would be automatic, neither creative nor free.
Opposed to the deterministic model of intention is the equally dubious model of intention based on indeterminacy. The "Action Painters" of the 20th century, for example, believed that an act of will (which constitutes an artistic act) must be freely executed, not inevitably determined by convention, past experiences, or habit, or otherwise compelled by any external cause. The "artistic" works produced in this way appear random, arbitrary, and undetermined and do not seem any more telic than completely predictable works do. Moreover, unless an inherent self as sole cause of impulsive or "intuitive" actions can be proved to exist, this notion of intentional behavior is groundless. The lesson to be learned here is that neither completely determined nor completely undetermined actions constitute intentional behavior.
Radical indeterminism and radical freedom of action seem as opposed to intention as physical predeterminism and predetermined action are. What both views of intention lack is a conception of how probabilistic necessity and deterministic chaos work together, providing direction and originality.
How is the emergence of telos from chaos to be explained? To answer this, I turn to American pragmatist C. S. Peirce, as I will do quite often in this work. Peirce seems to me to be an important precursor to those studying nonlinear dynamics today. He provided many concrete illustrations to explain his theories, which I use throughout. I refer to his position as indeterminism, which I contrast to the view of French deconstruction and poststructuralism or radical indeterminism.[28]
First, let me define a concept important to indeterminism: objective chance. Objective chance is not based on the ignorance of the observer; it is inherent in the physical condition. This is opposed to the classical conception of chance, held by determinists, which arises from incomplete knowledge of the initial state of the system. In "A Guess at the Riddle" (1887-88), Peirce describes objective chance, which he calls "absolute chance," as chaos that, while "real," cannot be experienced. His description bears resemblance to the then-yet-to-be-discovered quantum mechanical world. It is pure potential that has not yet had an effect.
The existence of things exists in their regular behavior. If an atom had no regular attractions and repulsions, if its mass was at one instant nothing, at another a ton, at another a negative quantity, if its motion instead of being continuous, consisted in a series of leaps from one place to another without passing through any intervening places, and if there were no definite relations between its different positions, velocities and directions of displacement, if it were at one time in one place and at another time in a dozen, such a disjointed plurality of phenomena would not make up any existing thing. Not only substances, but events, too, are constituted by regularities. The flow of time, for example, is itself a regularity. The original chaos, therefore, where there was no regularity, was in effect a state of mere indeterminacy, in which nothing existed or really happened. (278)[29]
However, within this multi-dimensional and discontinuous reality, continues Peirce's argument, some coincidentally regular patterns may occur, resulting in the "existence of things." Although chaos, by Peirce's definition, is that which is utterly homogeneous in its lack of structure, chaos does not have any rules to govern its behavior or to make it continue to behave in an ideally random (non-repeating) way. Thus, even chaos can produce coincidental regularity, a pattern, or a piece of "primal matter." As soon as primal matter (also called the arche) is put into relationship to chaos or a different kind of primal matter, there is a sense of change, or "polarity," or differentiation. If this polarity is self-reinforcing or persists (through a kind of selection process), it has begun to "take habit." The original chaos has become predictable. This is the effect of telos, for the cosmos has begun to spontaneously organize itself. It moves from a state of high entropy to a state in which some structure, differentiation, and heterogeneity exists.[30]
Derrida assumes a classical reductionist view when he supposes telos is paradoxically the "very thing within a structure which governs the structure, while escaping structurality" (960).[31] Derrida refers to the concept of a "center," which, he claims, may be described "as readily as archÈ as telos." He fails to distinguish between an originary cause and final cause (960). They are not equivalent. If telos is understood as the emergent universal laws governing complex nonlinear processes, the paradox of telos, as that which governs the structure without having a preexisting structure itself, disappears. The concept of telic structure does not depend, as Derrida argues, on the existence of "a linked chain of determinations of the center" (960). Quite the opposite. Telic structures arise from nonlinear processes.
Peirce used an analogy with gambling to illustrate how differentiation, structure, and habits could emerge through chance. If we think of indeterminate states as the outcomes of throwing of a single die, a sequence of two throws would more often produce a sum of seven than two. Seven is more common because it can result from a variety of combinations: one and six, six and one, two and five, five and two, three and four, and four and three. Peirce also imagines that through feedback processes nature builds upon the past; thus, his die, like natural process, are not ideal, that is, they would have memory and would tend to repeat what has gone before, further increasing the chance occurrence of sequences totaling seven. In this scenario, over time deterministic probability (habit-taking) results, and this, unlike objective chance, would be part of human experience because it would result in regular predictable events, the structure of which we would be able to grasp.
Peirce's nesting of probabilistic determinism within indeterminism is to assert that order can spontaneously arise out of disorder. Unlike the radical indeterminists who focused on the idea that objective chance means that individual events are unpredictable, Peirce recognized that the very unpredictability of objective or "absolute" chance leads to the most predictable kinds of statistical regularities in time.[32] Although this claim may at first seem counter-intuitive, the logic becomes clear if one considers the kind of systems that exhibit the greatest degree of statistical regularity. For example, although no one can predict exactly when a person will meet his or her end, insurance companies can predict the average age of death with remarkable accuracy. Life expectancy statistics are predictable because the individual events that contribute to the average are uncorrelated.[33] We can arrive at a useful macroscopic description only if there is microscopic uncorrelatedness that can be averaged out. If deaths were causally connected, that is, if the result of the first death had an effect on the occurrence of the second, and the second on the third and so on, it would be much more difficult to predict the outcomes (one would have to rely on nonlinear analyses). It would be a little more like trying to predict the stock market where earlier trades have an effect on later trades. Similarly, the various states of an atom that has "no regular attractions and repulsions," whose mass may be "at one instant nothing, at another a ton, at another a negative quantity," and whose "motion instead of being continuous, consisted in a series of leaps from one place to another without passing through any intervening places" would be radically uncorrelated. The average of a large number of atomic states would consequently exhibit strict statistical regularities, as they indeed do.
Beyond the province of absolute chance, events begin to be correlated: the past affects the future. According to Peirce, then, causality at our level of experience is probabilistic. Out of the original chaos emerges probabilistic necessity, our physical laws. And as Edward Lorenz demonstrated in 1963, in addition to these physical laws, the way nature processes information also allows for the emergence of deterministic chaos.[34]
Peirce's pragmatic view of causality allows for plasticity and stability, original behavior and directional behavior, and it is these combined effects that, I argue, define intentional behavior. We will not be resurrecting the Humanistic self pronounced dead by Barthes. Instead, we, as literary theorists, have available a new meaning of selfhood and, consequently authorial intention. This new author and the author's meaning are dynamically stable, not static; and deterministic, but not predictable.
Because the meaning of intention, on the human or cosmic scale, seems (to most) to require the transcendence of reductive determinism, in the absence of a theory of emergence and spontaneous organization, the task of explaining telic phenomena has been left in the past primarily to artists, spiritualists, and madmen who have credited special genius, gods, or conspiracy. As I show in Chapter Two, Aristotle argued vehemently against this move. It is ironic then that Aristotle's nonmentalistic teleology was colonized by a thoroughly mentalistic one.
In the 17th century, around the time of Francis Bacon, father of empiricism, science and teleology were separated into different discourses. Investigations of intention and originality were left to teleology, but investigations of directionality were reassigned to science. Henceforth, science would describe how events occur, physically speaking, while teleology would seek to explain why they occur, irrespective of physical conditions. Empirical science would describe, for example, the rules of morphology that give rise to insects that happen to look like leaves or sticks. Teleology would attempt to explain why some butterflies look like dead leaves or why praying mantises look like twigs: in order to help them hide from predators, as if the utility were determined by a creator before selection had the opportunity to act on it.
A distinction similar to that which exists between science and mentalistic teleology exists in literary studies. A scientific brand of literary criticism would describe the historical precedents of the Polonius character in Shakespeare's version of Hamlet; whereas teleological criticism would explain why Shakespeare chose a name for the king's counselor (whom Hamlet kills) that sounds like Poland (which Fortinbras conquers), e.g., in order to link Hamlet to Fortinbras. The verbal echo seems to mean something. It seems intended to inform us about the way we are supposed to view the hero's actions. Historical criticism has not turned up a "cause" for the pattern, that is, the country of Poland does not figure into any of the sources for the play, and there is no record of the name "Polonius" before Shakespeare used it. One can suppose that either he contrived this pattern for some sensible reason or it is an unintended coincidence.[35] One can never be certain since patterns formed strictly by chance, divine intervention, and artistry are empirically unpredictable and, as such, appear equally whimsical, enigmatic, or poetic.
As I show in Chapter Three, literary critics have been trained to discover and to interpret pseudo-logical patterns of this kind. When they do so, they might as well be following the old tradition of hermeneutics conventionalized by the early church fathers in their readings of scripture. In the fifth century, St. Augustine saw foreshadowings of Christ all throughout the Old Testament. When Abraham extended his arms in a gesture that happened to resemble a cross, the act was meant, argues St. Augustine in City of God, to predict Christ's crucifixion. St. Augustineís God had strong symbolic and poetic leanings.
Christian teleology typically makes God the author of nature's book, whose intention manifests itself acausally through coincidences. Providence can behave predictably, in the sense that everything will work out for the best, but how this is supposed to be an inevitable process does not lend itself to empirical description. Retrospectively, we may patch together a sequence of events, but accidental functions can only be described after an effect is produced; they cannot be predicted. Furthermore, this concept of telos as an interceding mechanism of Providence is not sufficiently complex since it equates telos exclusively with originality and disassociates it from directionality. We may recall that Aristotle had argued that only regular predictable events that occur "always or usually" have purpose;[36] this excludes accidents and coincidences. Christian teleology, in stark contrast, often argues that especially fortunate or unfortunate accidents are caused by (or at least justified by) telos alone. This distortion is also reflected in other contemporary manifestations of telos, for example in novelist Thomas Pynchon's understanding of a telic event as "a Hollywood distortion in probability," which might implausibly delay the death of a character so that he can make a farewell speech.[37]
In such a view, an agent (a writer or God) external to the sphere of action arbitrarily determines events. It would be a mistake to associate this kind of telos with linearity, or predictable, proportional cause and effect. Nevertheless, the association is common in contemporary literary theory. In Ariadne's Thread: Story Lines, for example, J. Hillis Miller claims a linear narrative "tends to organize itself or to be organized in a causal chain" and follows "inevitable sequence," according to a "telos, arche, or ground."[38] At first blush, Miller's description would seem to be more appropriate for a reductive or mechanistic view of teleology, such as found, for example, in Monod. In addition, he, like Derrida, does not distinguish between an originary cause, the arche, and an emergent cause, telos.
Angus Fletcher explains what is meant (or should be meant) by "linear" by critics such as Miller. He notes that history in terms of a story with a well-defined plot involving creation, fall, redemption, and judgment is "somewhat misleadingly called 'linear.'" In this type of teleological narrative,
[s]eemingly chaotic and unrelated events are shown to have a progressive character; history appears to move in a certain direction. Because wandering bulks large in this story, the form of history in this tradition should be called "linear" only with the express understanding that with it the line is not a very straight line. ...
... By showing that the wanderings of the chosen ones are momentously linked to the all-known but veiled design, the prophet "straightens" the twisting, labyrinthine shapes of profane time. When the children are lost, he unveils his prophetic gift, an inspired sense of direction.[39]
Fletcher shows that there are two different ways of thinking of "linearity" that must be kept distinct. As I further argue in Chapter Three, one involves a deterministic chain of events such as is found in a world governed by physical laws. (Recall the sequence of prime numbers.) The other involves a linking together of unrelated events and associations in the mind by someone outside the system who is not constrained by physical laws, space or time. (Recall the sequence of numbers that corresponded to the stops on the subway line.)
In addition to the categorization I make between mental and nonmental teleologies, a number of subtler distinctions can also be made. Each distinct type of teleology is a function of a different theory of causality. As mentioned above, I group teleologies into five broad categories: Aristotelian, analogical determinism, deterministic fortuity, pragmatism, and self-organization. (I also briefly mention mechanistic determinism and radical indeterminacy, to provide contrasts to teleology; therefore, seven different theories of causality are briefly reviewed in this section.) It will be instructive to revisit these teleological arguments with the distinctions between directionality and originality in mind. Though my portraits of these teleologies may be painted with rather rough strokes, my intention is to bring out their differences with respect to their stances on directionality and originality, nonmentalism and mentalism. Though I give them below in an apparently chronological order, some may exist, in one form or another, throughout history. Sometimes the worldview describes the common one, sometimes the one held by an intellectual elite.
First, the doctrine of creation ex nihilo is, like telos, also associated with Aristotle, and he developed this concept as a way of avoiding the problem of having to provide causal explanations that infinitely regress. In Metaphysics, he argues the universe began with a First Cause, which, as Waterfield has noted, may be considered a pure form of telic originality or motivation and does not require a further explanation in the way that material and efficient causes do.[40] With the exception of this metaphysical digression, Aristotle confines himself to a more or less organic-mechanistic description of final cause that I have described above as nonmentalism. Aristotle's nonmentalism is further explored in Chapter Three.
Second, according to the teleology of what I call in Chapter Four analogical determinism, the universe also has a supernatural beginning in time. Analogical determinism is associated with theology and contrasts sharply with Aristotelian teleology in that all matter and all events can be related in a fashion that transcends space and time. Events are not connected by physical causal chains so much as through resemblances.
In this view, things are the way they are because of their ultimate ends; therefore, in a sense, the future determines the past. This makes perfect sense if there is a supernatural being controlling events who exists beyond time. Analogical determinism is distinguished from predeterminism, since time is irrelevant to a creator who experiences the past, present, and future all at once. (It is also important to remember this distinction with regard to telic order in a fictional world, which is also created by an author outside of the narrative time.) In this supernatural view of telos, the reasons for events are the primary consideration. Only as a secondary consequence of final cause are events related materially or mechanically in a continuous fashion in space and time.
Although there is supposed to be no such thing as chance in this view, human beings, existing within time, have little power to make predictions[41] because utility-determined events do not necessarily follow from physical laws. They are determined by divine mind. One can only recognize the reasons for events after the fact. An empiricist who has complete knowledge of the example of Abraham's attempted sacrifice of Isaac is powerless to predict that God would sacrifice his own son. However, this is exactly the kind of causality that exists in a universe determined by analogies. Only prophets and literary theorists are in the position to read events this way. Ordinary folk generally recognize the prophetic purposes of events after the fact. Interpreting "chance" patterns, then, becomes a way of discovering divine intention.
Third, if mechanical determinism includes a notion of divinity it is associated with orderliness and is equated with physical laws, which simply are out of necessity, and are not accounted for by utility. Utility is more or less a convenient (if not divine) side effect. There is, arguably, no place for telic originality in this view. In a mechanically determined universe, all matter acts in a physically continuous series, and each event is entirely predictable. As in an analogically determined universe, time is here again irrelevant since the future is already given in the present, and the appearance of chance events is due to lack of knowledge of causal factors. The defining moment for this philosophy came when Pierre-Simon Laplace asserted that any one who had knowledge of the forces in nature and position of every thing in the universe could predict all future behavior.[42]
However, if everything is determined, sequential, continuous, and predictable, one is faced with the impossibility of explaining, scientifically, how everything began in the first place. Such a unique event that, unlike all that follow, has no "before" and is inexplicable in terms of classical determinism, which says that every effect has a cause. The origin of the universe would have to be explained by a singularity or a miracle (which both defy the laws of classical physics). Or it might be supposed that the universe has always existed and always will. None of these explanations is particularly satisfying. Therefore, telos has been brought into the mechanical deterministic view to explain why the universe began (as Aristotle had found it necessary to do), if nothing else. Proponents of material determinism would include Isaac Newton, many of the 18th century philosophes, and even Albert Einstein, whose theory of relatively leads up to the edge of classical determinism.
Analogical determinism and mechanical determinism are two extreme theories of causality, and, in the literary domain, they can be associated with poetic and realistic fiction, respectively. If a narrator represents an analogical deterministic view, verbal coincidences and other poetic resemblances may provide the key to understanding the meaning of events. In a narrative written within a material deterministic view, in contrast, the plot will tend to be motivated by physical causality and will be perfectly linear, in the sense that each event will follow inevitably from the last.
Forth, the various philosophies that I group in Chapter Five under the term deterministic fortuity accept the tenants of material determinism, but have used the concept of fortuity to argue for human, or even cosmic, intention. Fortuity involves the coming together of unrelated factors that together result in a particular state. For instance, various genetic and environmental factors determine one's height. Telos is thought to be what actively and continuously causes the various factors to be present in the right proportions so that the correct state, say five foot two inches, will result.
According to a postmodern representation, the telos of deterministic fortuity may be compared to the running of a computer program. In this extreme version of deterministic fortuity, the common cause of two coincidental events, the genetic sequence for femur length and a diet containing a certain amount of calcium, would be physically inscribed in the initial conditions of the universe, but the separate causal chains (ending in a particular gene and a particular diet) that ultimately converge to produce a specific height would have unfolded according to a finely-tuned plan. Though the "decision" at each fork along the way (1 or 0) might involve some degree of chance, the odds are actually biased in favor of the design of the original program, which has the inevitable ending encoded in the beginning. Narratives written under this paradigm would not make use of magical coincidences, as in analogical deterministic narratives. However, predetermined coincidences, synchronized according to a prespecified design, would be used to further the action. (Here we see most clearly how deterministic fortuity differs from mechanical determinism: chance plays a role in determining context; the context, or whole, in turn determines the parts.) In retrospect, once all the necessary information had been revealed, one would be able to decipher the end in the beginning. Unlike in the analogical deterministic narrative however, the causal chains here would be caused by describable physical constraints. Many Victorian novels follow this scheme, in so much as they are more empirical than analogical.
The ideas behind deterministic fortuity may be traced back to Kant, although the computer program metaphor falls short of his conception of teleology, for he would not have accepted the idea of an actual physically inscribed program as initial cause. This is where postmodernism has incorrectly interpreted Kant's teleology. Instead, he imagined that limiting principles, inherent in ongoing natural processes themselves, guided events. According to Alicia Juarrero RoquÈ, "Kant's emphasis on recursive causality, wherein the parts are both cause and effect, precludes the existence of a preexisting whole" (113).[43] And as Ernst Cassirer explains, the Kantian whole is "contained in them [the parts] as a guiding principle."[44] In Kant telos is emergent, given in the interactions between parts and the whole.
It was the fact that fortuitous events can collectively result in the existence of sentient beings, the artistry of nature, and the apparently progressive character of history that seemed to Kant to require a guiding force. As he writes,
...it may be held that from an Epicurean concourse of causes in action it is to be expected that the States, like little particles of matter, will try by their fortuitous conjunctions all sorts of formations, which will be again destroyed by new collisions, till at last some one constitution will by chance succeed in preserving itself in its proper form,--a lucky accident which will hardly ever come about!"[45]
Teleologists, like Kant, who believe in deterministic fortuity, do not usually believe that God is supposed to exist beyond time, external to the human world. Instead, a nonphysical cosmic intention is believed to be immanent in physical events themselves. The role of telos in this case would be to balance the odds; its medium is statistics. Proponents of variations of deterministic fortuity would include, in addition to Kant,[46] Charles Bell,[47] William Paley,[48] Ralph Waldo Emerson,[49] and, with certain qualifications, Karl Marx.[50]
Fifth, as I show in Chapter Six, a teleology based on pragmatism characterizes the period of modernism in literature. Teleology in this view inherits most aspects of the 19th century deterministic fortuity version, sketched above; however, it is changed somewhat by the fact that positivism had started to give way to probabilism.[51] The new century began with the introduction of quantum mechanics. The notion of telos as consisting in the order and arrangement of the original configuration of well-defined particulate matter was no longer tenable. Thus, telos and agential selfhood could not be exactly prespecified and had to be reconceived as the emergent products of contextualized interactions.[52] Early pragmatism offered a way of making sense of the way matter now appeared to be information defined by use or by the effect produced from interactions with other matter. In doing so, pragmatism modified Aristotle's original teleological argument that what a thing is and its function are one.[53] But the help this philosophy offered was misunderstood. The interpretation of information does not imply a conscious interpreter, any more than natural selection implies a conscious selector; however, pragmatism was subsumed by the postmodern notion of radical subjectivity. The idea of the effective "observer" began to haunt science.
Although one should resist the impulse to personify nature, it is true that, in some sense, nature does perform simple kinds of interpretations. For example, an inanimate system can react to another system following the rule, if A then B; if water temperature goes below 32ÄF, then water freezes. Such outcomes are usually well defined and predictable. However, another kind of "if A then B" situation may arise in which one system sees an "A" in another system that is not an "A" in the usual sense but only coincidentally functions as an "A." For example, a few mutations can result in a butterfly that looks almost exactly like a dead leaf. (It is apparently not a gradual adaptation shaped by selection, but simply the sudden result of a very lucky mutation.)[54] Would-be predators pass the dead-leaf butterfly by because they interpret it according to the rule: if it looks like a leaf, then do not try to eat it. Such outcomes are not well defined and predictable. Nevertheless, these kinds of misinterpretations can have a significant effect on the direction of evolution by causing a new useful mutation to proliferate.
A coincidental resemblance between a butterfly and a dead leaf might be called a stochastic resonance, an important concept in literary modernism[55] as well as in numerous areas of contemporary science. An illustration will help clarify the concept of stochastic resonance. Someone several feet away from you on a crowded street is saying, "Hey, you in the blue shirt. Stop." However, you cannot quite hear the entire sentence, so it just sounds like noise to you. Someone nearer to you coincidentally says (to someone else) the words "blue shirt" at exactly the right moment, completing the other, partially inaudible sentence, thereby making it intelligible. You stop and avoid stepping in front of a speeding car.
Sometimes a coincidental pattern or a side effect is found to be useful in a given context. Many elements in these late 20th Century ideas can be traced through C. S. Peirce,[56] William James,[57] and Bergson,[58] whose philosophies derive from evolutionary theory, generalizing it and making it applicable to all domains, not just the biological domain. Narratives written under this paradigm would include the notion that the use (or misinterpretation) of noise or error can lead to the creation of what had not existed before and, as I hope to show in Chapter Six, even what could not have existed before.
intention, according to a teleology that incorporates pragmatism, is not innate in the thinker, as Kant would have it, but, rather, the ability to think freely is an evolved "gift," as it were, compliments of the thinker's dynamic situation. In this view of intention, making an interpretation may not be the result of conscious choice. The opportunity to (mis)perceive a pattern according to a particular frame is merely due to the fact that, as neural biologist Jean-Pierre Changeux notes, the brain is constantly and randomly switching and evolving frames. The instance of the brain resting in one particular frame and therein-determining action does not require any special cause or explanation.[59] Such actions are not realizations of predetermined goals. Nevertheless, over time there may emerge an identifiable tendency to choose that gives meaning to the notion of a teleological self. Because mind and intention are emergent phenomena here, there is no need to search for an underlying efficient or material cause of a final cause.
Not all modernist writers fit this description. For instance, I would be