Film Worlds: A Philosophical Aesthetics of Cinema

IN THE FIRST CHAPTER I ACCEPTED WITH CERTAIN QUALIFICATIONS the view widespread among theorists and philosophers that most films classified as narrative ones project fictional realities that serve to ground the existence and actions of characters in a relatively comprehensible time and space—and that in this special sense films (in common with literary narratives and perhaps certain other representational art forms), construct and possess “worlds.” But, taking into account not only their narrative structure and fictional contents but other aspects, not least of which are those elements that are, broadly speaking, perceptual, formal, affective, and otherwise extranarrative, I have also offered the suggestion that, to an unprecedented and perhaps unique degree among works of fiction, films not only contain but are worlds. To be sure, these entities are more in the nature of virtual than material or physical realities, existing (as we say) on a plane of shared human intentions, invoking the imaginative responses of viewers.

Yet even granting this much, what justifies the stronger claim that in speaking of the worlds of films, we are speaking, in some sense, literally, as distinct from entertaining no more than a suggestive metaphor? How does a two-dimensional play of light and shadow translate into such a “global,” enclosed, and singular reality? After all, as even the sympathetic reader may be inclined to point out, although some of us may seem to live for films, no one is quite able to literally live within them in anything like the intriguing ways in which a number of films have imagined.1

In this chapter I will attempt to further articulate and defend the concept of film worlds by introducing a general perspective in the philosophy of culture and art that affirms the perceptual and symbolic construction of not only many kinds of artworks but other sorts of social and cultural realities (and, in some senses, human “reality” itself). Like the heterocosmic model but with different emphases, this view suggests that, as known and experienced, these can and should be conceived—and experienced—as singular worlds.

SOME GENERAL FEATURES OF WORLDS

In reflecting on what the world of ancient Rome, the fashion world, the world of cyberspace, the fictional worlds discussed in the previous chapter, and a great many other worlds all have in common, it becomes apparent that the English word world is an unusual and intriguing one. The Oxford English Dictionary lists more than twenty-five primary definitions. The broadest and most all-encompassing refers to the earth (“or other heavenly body”) as the physical totality first charted in its full extent by the great explorers and cartographers of the sixteenth century.² According to a number of its relatively more modern definitions, however, as in the worlds mentioned above, the term refers not just to the globe that we all inhabit, or to any other potentially habitable planet, but to specific dimensions of the life of human beings who possess a common culture—or, hypothetically, the alien cultures of alien beings, as in the “strange new worlds” of Star Trek and other science fictions.

Along these lines, and reflecting a similar distinction made by the ancient Greek philosophers, both Heidegger and Gadamer (whose writings on art we will have occasion to return to) distinguish between a cultural “World,” as the realm of human intentions and actions, and the naturally given “Earth” (Heidegger) or “Environment” (Gadamer), as the material substructure that supports and sustains (but also limits and constrains) human actions and possibilities.³ As this broad distinction reflects, in speaking of such worlds as those of art, literature, sports, politics, business, and various academic and scientific disciplines, we are not referring to any preexistent material or biological realities but rather to the shared works and products of human minds in the exercise of their cognitive, affective, and creative potentialities. In this primarily cultural sense or senses of world there is always an implied multiplicity: the world becomes many worlds, denoting (again from the OED) a “particular division, section, or generation of the earth’s inhabitants or human society; with reference to their interests or pursuits” (or, alternatively, “a group or system of things or beings associated by common characteristics” denoted by a qualifying word or phrase).4

The existence of such common frameworks of meaning and action implies that the particularly human version of Reality-at-large may be carved up into different sections or areas, or partitioned into compartments, separating sets of events or experiences from one another. Such divisions occur on the basis of ideas, plans, purposes, and values as much as on any physical or geographical facts. In many contemporary academic disciplines and intellectual discourses the pluralism of worlds in this sense is frequently acknowledged, as it perhaps cannot fail to be, under modern and postmodern conditions of inquiry. But this occurs most often only implicitly, as a kind of background assumption of all specialized or technical ways of thinking, speaking, and writing. Broadly speaking, one of the most compelling (if also challenging) aspects of the experience of modern social and cultural life is that of a vast multiplicity of worlds, beyond the capability of the individual even to list, let alone to participate in. These worlds organize, divide, and frame the phenomenal appearances and meanings of things in many different ways. While some worlds are the products of collective cultural activity and intentions spanning generations, others, like the worlds of films, are the deliberate products of a small number of individuals or the creative offspring of a single, fertile imagination.

Worlds, in this cultural sense, then, including those created by filmmakers, have an external, extrasubjective, and social existence. As such, they are distinguished from the internal, private mental realm(s) of the self, as well as what phenomenological philosophers refer to as the Lebenswelt (lifeworld), not least because any number of distinct worlds of experience may either be a part of, or totally absent from, a single individual’s consciousness and putative lifeworld.⁵ Yet while always intersubjective and possessing some external, physical basis or grounding, the worlds under discussion are never material or physical realities (alone) in the way that natural or physical spaces and places are. Indeed, almost all worlds that may be pointed to as existing within present or past cultures have, instead, a highly composite character. They reflect a meeting of material and “spiritual” forces of the very sort that human beings—and works of art—may also be said to possess.

Films certainly have such a necessarily dual, ontological status, as not only intention-bearing but “embodied” or “incarnate,” in the ways argued by Margolis, for instance, with reference to all artworks.6 That is, they comprise both material and directly sensible phenomena, as well as a range of psychological and ideational (that is, perceptual and conceptual) aspects—significations, ideas, beliefs, histories, traditions, and so on—that bring their emergent nature to the plane of our common mental and emotional life. The relations and interactions among these fundamental dimensions or components of worlds may be quite complex, and they establish their own unique patterns, given basic objective parameters of space, time, and causation, on one side, and our cognitive constitution and subjective conditions, on the other. The purely physical support structure of worlds may be quite minimal: the words of a book that, when read, manage to bring an entire fictional world into existence or a few props and background settings for a dramatic performance, sufficient to evoke our often latent but powerful “imaginary forces” (as famously called forth in the prologue for Shakespeare’s Henry V).

Worlds are discrete and singular, yet they also clearly overlap with one another, more or less extensively, as if to form extended families. And certain numbers of them share the same contents (people, places, objects, representations), which may be integral parts of more than one world. The existence of multiple worlds (actual, and not merely logically “possible”) entails that they are bounded or framed, that is, that there are physical or psychical borders that divide one from others. The worlds of films are bounded by the physical border of the screen, the filmstrip (in celluloid filmmaking), and the camera’s viewfinder; the experiential boundaries of cinemas, or movie theaters (in traditional film viewing); and also the psychological boundaries of relevant, common beliefs and attitudes of viewers, such as, for example, that the two-dimensional action of the film is not physically contiguous with the three-dimensional space occupied by the viewer’s body.

Not all cultural artifacts, perceptual objects, and purposeful activities amount to worlds, since most, in fact, are constitutive parts (more or less central) of a world, or of many. Pragmatically speaking, and as Goodman—the leading philosophical advocate of “world-theory” in the arts and sciences, alike—maintains, whether or not a given symbolically constituted reality qualifies as a world of its own ultimately appears to come down to whether the experiences associated with it are illuminated by being conceived in such a way (together with precedents and expectations of historically given cultural and linguistic practices).7

In addition to having borders and an internal coherence and unity, a world in this most general sense must, it would seem, be (or appear to be) not adequately reducible to, or fully explicable in terms of, any set of intentional phenomena external to itself. (For if it could be so reduced or explained, it would, instead, be regarded as forming only part of some larger world-like totality). However, while worlds are marked by a pronounced structural and experiential alterity, in being other than or different from one another, the perceptual, cognitive, linguistic, or other boundaries between and among worlds are always only partial, relative, and shifting (and there may be uncertainty concerning exactly where a world begins and ends).

Continuing with this partial list of attributes considered in abstract terms, a world would appear to be a unified whole with distinct parts or aspects. Its unity is profoundly holistic in that it is more than one of the sum of the parts considered as parts. This is the case because each world sustains a unique network of internal relations, including such basic relations as meanings, or the symbolic significance that attaches to objects, events, and behaviors (and that may be found in that world and perhaps no other). Worlds also possess, or create, a location or setting where something, most typically concerted actions and interactions, happens (either in perceptual or bodily reality or, as we say, in imagination). This consists of some demarcated spatial or quasi-spatial area or dimension that helps to define the objects and activities present. Of particular importance in the cinematic context is that such spatiality may be virtual and non-three-dimensional in physical fact, as in the case of other imaginative, perceptual, or representational spaces. Finally, worlds are structured or ordered in many ways: they harbor cognitive categories of their own, often but not necessarily shared among other worlds, and recurrent patterns of action governed by accepted beliefs, norms, values, and so forth.

These features of cultural worlds in general—consisting of various admixtures of mind and matter, the sharing of contents (or “materials”) among them, the qualities of singularity, boundedness, irreducibility, a whole with distinct yet integrated and functional parts, alterity, actual or virtual spatiality, forms of order and rule-governed activity—are found, on reflection, to have a prominent place in the specially created and experienced worlds of films on multiple levels. In fact, a number of these features have already been broached in the first chapter, in relation to the differences between the worlds-in and the worlds-of narrative fiction films. Their more specific, cinematic manifestations and consequences will continue to have an important part to play in the discussions of the artistic structures, features, and elements of films that follow.

Yet it is clear that conceived only in terms of such abstract properties, worlds are still essentially empty, lacking any concrete “content” by which alone they can have an actual existence as experiential realities for inhabitants, members, players, viewers, and others who in some way share them. At a fundamental level, whether relatively more virtual or actual (e.g., in accommodating the presence of our bodies), all worlds in our present conceptions are composed of signs and representations (whatever else they contain). They always involve and require the symbolization of aspects of concrete, immediate sensory experience. In fact, as both Goodman and the neo-Kantian philosopher of science, history, and culture Ernst Cassirer have shown to great effect, all worlds in the sense that I am attempting to describe and define may be properly qualified and spoken of as symbolic.

SIGNS, SYMBOLS, AND ARTWORKS: THREE TRADITIONS

Compared with other animals man lives not merely in a broader reality; he lives, so to speak, in a new dimension of reality. . . . No longer in a merely physical universe, man lives in a symbolic universe. Language, myth, art and religion are parts of this universe. They are the various threads which weave the symbolic net, the tangled web of human experience.

—Ernst Cassirer

To simplify greatly, but not inaccurately: throughout the latter half of the twentieth century, and into the twenty-first, two approaches in general semiotics have dominated scholarship in the humanities and social sciences. What they share is a concern generally, and paradigmatically, with all phenomena of intersubjective messaging, anchored in the fundamental capability of an individual to make mental associations, which are raised to the level of intersubjective communications and the development of systems (or codes) for sharing intentions. This much applies not only to so-called natural signs and our varied responses to them but conventional relations among conventionally established signifiers and their purported objects. Both traditions hold to the insight that the study of such pointers and surrogates for aspects of direct experience is fundamental, a major key to our understanding culture and our claims to knowledge. From this common and now familiar ground, however, the two modern traditions in semiotics (or semiology) and several distinct intellectual paths within them, diverge.

What may be termed the “pragmatic-instrumentalist” tradition and paradigm has its root source in the seminal writings of the American scientist, pragmatist philosopher, and polymath C. S. Peirce. At the heart of much recent linguistic semantics and philosophy of language, the pragmatic-instrumentalist view favors so-called communication models of symbolic reference and tends to be empirically based and naturalistic (or “precultural”) in orientation. Ultimately, in this view, signs of all kinds owe their existence to the survival needs for cooperation among the individual members of our biological species. They are conceived as cleverly invented tools that facilitate the sharing of subjective intentions, thus allowing for common understandings and concerted actions in a physical, Darwinian environment, translating into multiform social behaviors that are distinctly human.8 Peirce’s theory of signs, and its general orientation, may be considered at least one progenitor of some contemporary, naturalistic, and empirically based approaches in some disciplines within cognitive science. Through this route and others, this first tradition also exercises present-day influences in cognitivist or perceptual approaches to a range of issues and problems in the philosophy of art and film and in some film theory.

With reference to the latter it should be noted, however, that there is also much wider general agreement among film theorists of various stripes that film images in their communicative dimension frequently possess characteristics of all three of Peirce’s canonical, major sign types: icons, indices, and symbols.⁹ They may, in other words, stand in the iconic relation of perceptual resemblance insofar as they “merely” denote the objects they “contain”; an indexical relation insofar as many of their properties are determined by photographic and related causal processes or conditions, forging an “ontological link” between object and referent; and (generally closer to our main concern in relation to cinema as art) a symbolic relation insofar as they evince culturally and cognitively mediated connections to what is potentially the entire realm of human-recognized, intentional (mental) objects. Thus, no matter how remote it may also sometimes seem from specifically artistic realities and issues, this tradition (like its main rival) has valuable insights to contribute to a theory of the worlds in and of films and to cinematic art.

The other dominant and to some degree philosophically opposed tradition, as it were, traces its origins to Saussure’s linguistics and largely (although not entirely) French structuralist and poststructuralist thought. It has exerted an enormous influence on cultural anthropology, sociology, psychology, and cultural and literary theory. As is well known, this large branch of semiotic theory centers on relatively autonomous systems of communication (symbolic structures) that are culturally and ideologically coded in various ways, above or below their immediate pragmatic and social-communicative functions. Such structures and codes, including languages and bodies of myth (and interpretive or ideological formations), as well as texts and other cultural productions are seen as rooted in the use of what Saussure has famously defined as the arbitrary sign. This is a sign that derives its meaning via its relation to other signs in a closed system rather than through any “natural” perceptual correspondence (e.g., verisimilitude) to the empirical reality to which it refers (or, more accurately, to which actual speakers choose to apply it). For many theorists within this second broad and historically ramified tradition, the collective signs and codes existing within systems of communication are cultural and ideologically freighted; they shape, and often constrain, human thought through the propagation of certain binary distinctions and dichotomies, for instance. Such processes, operating in one way or another in all cultural media and artistic production, including cinema, are often seen to reinforce particular worldviews, ideologies, and sociocultural narratives; one major task of the theorist is to identify and lay bare these structures for critical scrutiny.

Although Metz’s Saussurean (and later Lacanian) semiotics is less concerned with the specific represented content of films vis-à-vis their cultural messages for audiences (the focus of most film semiotics), and more with the formal, narrative-cinematic and quasi-linguistic structures governing their presentation (in the form of a diegetic reality), it has exerted a major influence in this theoretical vein. Metz’s complex and, as some might argue, at times convoluted negotiations of the extent to which cinema might be considered and studied as “language” highlights that whereas the pragmatic-instrumentalist tradition was, from its very nineteenth-century origins, concerned with all nonlinguistic forms of communication (e.g., gestures, visual depictions) as much as discourse proper, the linguistic-structuralist tradition attempts to apply insights originally formulated in relation to linguistic semantics to other forms of cultural communication and representation, including the nonverbal arts.

Perhaps inevitably, insofar as narrative filmic representation is perceptual, or iconic, but also highly reliant on cultural conventions, within film theory there is also substantial overlap and intellectual hybridity between these two preeminently “naturalist” and “nonnaturalist” semiotic traditions. Peter Wollen, in his influential Signs and Meaning in the Cinema, for example, turns to Peirce and his famous classifications of symbols, icons, and indices to modify certain aspects of a still generally (and self-identified) structuralist account of cinematic communication and expression, thereby differentiating it from Metz’s semiotics.10 In a very different way, Deleuze’s philosophy of film adopts Peircean concepts and vocabulary, which are innovatively combined with Bergson’s theories of time, memory, and imagination—all in a way explicitly distinct from, and critical of, structuralist and semiotic approaches to cinema in the Saussurean tradition.

However, often overlooked today, in film theory, as elsewhere, there is yet a third general tradition of thought on the symbolic construction of reality, bound to what might be called the symbolic-expressive—or, for short, “expressivist”—paradigm. German in origin, this approach descends from the philosophy of Kant and his immediate contemporaries and successors (e.g., Herder, Goethe) and runs through to the thought of Hegel and Alexander von Humboldt. In the twentieth century it is reflected in Cassirer’s philosophical anthropology (articulated in his seminal, multivolume Philosophy of Symbolic Forms), Langer’s aesthetic theory, and, in some important respects, Goodman’s philosophy of symbolic reference, art, and “worldmaking.” George Lakoff’s and Mark Johnson’s influential writings (jointly and individually) on the power of metaphor to shape our conceptual categories and, hence, to “make worldviews,”¹¹ has also been justly seen as a legacy of this Kantian and post-Kantian tradition (which, within the fields of social theory and history, respectively, also counts Norbert Elias and Lewis Mumford, among its notable members).

As I have mentioned, the pragmatic-instrumentalist paradigm is anchored in empirical and naturalistic thought (in some respects behaviorist or neobehaviorist), stressing what are deemed to be species-wide, pan-human communications needs. Structuralism and its many successor positions are often associated with radical forms of historical, cultural, and epistemic relativism. The third tradition may be seen to occupy an intermediate position between the natural and culturally specific, or “normative.” It has been concerned, since its inception, with the basis and rational justification of the Geisteswissenschaften (the humanities) and, as part of this post-Enlightenment interest, to identify the correct placement of art amid the major, symbolically conditioned avenues for the representation of human experiential reality.

This standpoint locates the origins of human signs of all types in certain processes of abstraction, classification, or categorization, taken by the mind with respect to the presentations of sense experience. As ubiquitous as pan-human symbol making and use is, it is regarded less as an aid and accompaniment to higher level perception and more reflective thought, as existing in some way apart from it, than one of its foundation stones or enabling conditions. Semiotic devices and constructions, in other words, are more than a set of exceptionally helpful and versatile cognitive tools, serving to make the natural environment more navigable; nor are they conceived as a highly consequential adjunct or addition to already fully formed minds and societies.12 Symbols, in the broad and inclusive sense favored by this tradition, are, by contrast, the initial means and method by which an otherwise disarrayed, ephemeral, and generally chaotic sensory manifold comes to some semblance of organization and, therefore, to intelligibility. Rather than preexisting, or somehow preformed, that which is symbolized in one or another media of representation is only capable of formation and articulation on the basis of prior possession by the cognitive subject of schemes of categorization, that is, some so-called grid-work of interpretation placed over the deliverances of immediate experience.¹³ From this originally Kantian perspective, then, the generic concept of the symbol, as already incorporating other sign types and functions, is not a reflection on pragmatic communication and its contexts, the ways and means of transferring and exchanging information among subjects per se. Rather, it is the (arguably) logically prior representation (or categorization) of the immediately given material of sensible experience. Also distinguishing it from the structuralist-semantic model, the “symbolic” in this sense is not one form of code but the cognitive basis for the use of any code.14

It is this last-introduced general framework of ideas—as taken up in aesthetics and the philosophy of art, together with film theory—that offers the most promising support for an improved articulation of the full symbolic dimension of cinematic works and their worlds (or world-like nature). This perspective also converges with a prominent tradition of thought within film theory itself, focused on the symbolic basis of both basic filmic communication and cinematic art (as involving artistic “expression” in a number of senses). Here, as we will discuss shortly, we may find thematic unity and common purpose in the writings of such figures as Mitry, Pasolini, and Deleuze, who reject orthodox film semiotics, whether recognizably Saussurean or Peircean, as it is transmitted through relevant writings of Metz, (early) Barthes, (early) Wollen, Eco, and several other theorists. Mitry, Pasolini, and Deleuze each see Saussurean semiotics as failing to sufficiently appreciate, or engage with, the “expressive-constitutive” and pre- or protolinguistic grounds of symbolization in cinema. Prior to pursuing this subject in detail, however, and as one final preliminary to it, it is well worth devoting the remainder of this chapter to some of the specific, cinematically relevant insights concerning symbolization and art that the tradition in question, as it developed through the twentieth century, has offered.

SYMBOLIC FORMS AND FEELING

Exerting an important intellectual influence on figures as diverse as Albert Einstein and art historian Erwin Panofsky, Ernst Cassirer has recently been called a “singularly important and underappreciated thinker” both as a philosopher and as one of the greatest twentieth-century intellectual historians.¹⁵ Attempting a grand synthesis of Kant’s critical Idealism (positing innate mental categories of understanding) and Hegel’s conception of the historically evolving human mind or spirit (Geist), his extension of the aforementioned expressivist paradigm with respect to conceiving the nature and function of symbolic thought eventuates in a position today often referred to as “cognitive pluralism.”¹⁶

Cassirer maintains that our symbolic categories, which have their original root in “expressive, nonverbal symbolism,”17 free the human subject (or potential subject) from the immediate conditions of moment-to-moment, here-and-now experience characterizing a precultural, or natural, presumably animal mode of existence. Myth, language, art, religion, and science as the symbolic forms, or basic cognitive orientations that Cassirer identifies, are seen to emerge progressively and to supplement one another in collective mental development. Given general spatial, temporal, and other structuring principles of the mind, each symbolic form is predicated on a distinctive set of relations among the human subject, the intentional symbol, and the symbolized object or phenomenon. In truly dialectical fashion, these symbolic channels represent the ways in which we come not only to know the world but also to know ourselves, through a pronounced objectification of direct perceptual and bodily experience. To create and use symbols, which Cassirer memorably describes as the “organs of reality” is to take up a position in relation to sensible reality that defines both reality and the self.¹⁸ The focus in his version of Lebensphilosophie (life-philosophy) is not confined to just the thinking self, as a disembodied, so-called epistemological or Cartesian subject, but the integral feeling, desiring, and, above all, expressing self.

For Cassirer a symbol, no matter how abstract and conceptual, is in its actual use a concrete and stable externalization of individual expression, upon which all dynamic aspects of cultural life depend. Each of the major symbolic forms provides a kind of truth that the others (by virtue of their different cognitive structures) simply cannot reach. In this sense each has its own epistemic charter and value, as an interpretative mode or relation to immediately given sensation. In the case of the concrete products of the different major forms of symbolic thought and expression identified—for example, each individual work of art, scientific model, and mythic tale—these different, basic modes of human symbolic relation to experience are combined with (a) a particular content and subject, and (b) a unique mode of presentation and structure; the latter is readily accessible to our faculties of perception (as it must be) but also partly dependent on the nature and properties of (c) the physical medium of symbolization in question.

Art, for Cassirer, is historically rooted, like myth, in “expression” proper (Ausdruck). At first closely combined with but then gradually separating off from language (as “representation” or Darstellung) and myth (as Lévi-Strauss also maintains), art comes to be fully recognized as an autonomous and coequal form of expression and cognitive representation.19 In his view the symbolic representations (and interpretations) that we (properly) recognize and appreciate as works of art are among the most exemplary of all symbols, with regard to the accomplishment of the primary symbolic function of mediation between our direct sensations and our self-conscious reflections. That is, they are found to be situated midway between the Kantian poles of the human mind of “sensibility” and “understanding.” Indeed, although there is some precedent in Hegel’s aesthetics, Cassirer is seemingly original in theorizing the work of art as a kind of intermediate, Janus-faced symbol, pointing to both itself and its subject of representation simultaneously. It rests approximately halfway between the concrete and qualitative, but only vaguely self-referential and self-centered, protosymbols of myth, ritual, and religion, on the one hand, and the fully “self-conscious” but also arbitrary (or “free”) and self-effacing (“transparent”) symbols of rational discourse, wherein individual, signifying elements are entirely system-dependent or bound in both their syntactic and semantic relations (e.g., the fully developed linguistic sign), on the other.

The most relevant difference between myth (and to some degree religious thought and feeling) and art, as symbolic forms per se, lies in the fact that the expressive images of the latter always involve (some degree of) awareness of the constructed and mediated nature of their represented realities, on the part of both artwork creator and perceiving subject. Such a self-consciously mediated presence of the object-image in the work of art, and that which it represents, is thus, for Cassirer, of a different order than in mythic expression.²⁰ The symbolic form and, by extension, the created worlds of art claim for their audiences a truth, to be sure. But this is much less a direct and literal truth than that which the products of the vast domain of what Cassirer terms the “mythic consciousness,” past and present, claim (and also notably different from the sort of objective truth that scientific representation allows for). In this account, if aesthetic symbolism instead seeks truth only “within appearances,” it is also very typically, if not invariably, suffused with a pronounced affective content. For art has not wholly relinquished its aboriginal ties to a mythical consciousness that is a preconceptual and prerational one, entirely lacking any general separation of fact from value, or concepts or propositions from subjective feelings, attitudes, desires, and so forth. It should be noted that despite its stress on feeling in art, this view contrasts with the so-called Canonical Expression Theory of art, associated with the aesthetic idealism of Benedetto Croce and his followers, wherein the genuine artwork is seen to exist not in the physical world but in the mind of the artist whose feelings it expresses through the work as but a conduit for them. As Cassirer argues, however, this is to neglect the full input of the medium and of formal structure in art.21

Coming closer to our primary concerns, it was left to Langer and Goodman, two North Americans in the analytic philosophical tradition, to substantially build on Cassirer’s positioning of art as among the major symbolic forms of representation, in ways that are only touched on by the Weimar philosopher. While both later thinkers incorporate these major insights into much more detailed and developed theories of the subject, they draw different lessons and proceed in very different (if at points also intersecting) directions. Whereas Goodman pursues the primarily perceptual, formal, and rational orders and structures of artworks in their reality-shaping power and in the form of singular worlds constituted by processes of symbolic reference relations, Langer pursues the realm of subjective feeling and its objective expression in the essentially “presentational” form of art (which even extends, in a particular way, to literature).

A student of Cassirer, as well as a translator and interpreter of his work, Langer sees Cassirer’s understanding of symbolization as “hewing the keystone” for a new aesthetics.²² In her Philosophy in a New Key and Feeling and Form, two canonical, if today far less frequently cited and discussed, works of twentieth-century aesthetics, she argues that it affords the basis for a more comprehensive and persuasive expression-rooted theory of art than had previously been forwarded. The realities that works of art symbolize and present are, for Langer, aspects of the life of subjective “feeling,” defined as nothing less than all human experience, which defies satisfactory articulation in language as the cognitively dominant, discursive form of representation.²³ In general accord with Baumgarten’s central “heterocosmic” distinctions, yet still stressing the “rationality” of art, she holds that while also selectively abstracting from, objectifying, and thus transforming experience, the “presentational” (analog) symbolic mode of art does so in a fundamentally different way than do language and logic as “discursive” (digital) symbolism (or, indeed, than do nonartistic visual and environmental signs and signals).²⁴ In marked contrast to Goodman, whose cognitivist aesthetics focuses on the artwork as a complex network of symbols (in various kinds of more or less direct reference relations), a work, in Langer’s view, is a single indivisible “symbol,” or objectification, of subjective feeling given fixed and external form. As existing in a concretely realized object or performance with a special aesthetic sort of existence, such feeling is thus amenable to intersubjective access and reference.25

The brief appendix of Feeling and Form is entitled “Note on the Film.” It represents an ancillary effort on Langer’s part to make a place for cinema within her general philosophy of art. Despite a film’s photographic and indexical ties to the physical reality before the camera, and its narrative and dramatic aspects, she sees cinema as, like all art (representational and abstract), a matter of the objective conveyance of subjective (felt) reality. Looked at critically and with the benefit of hindsight, Langer’s account of cinema is suggestive and flawed in equal measure.²⁶ One of its main merits is a stress on the fact that a film work as a whole is a form of “presentational symbolism” that transforms its represented contents (taken from the world of actuality). Beyond the verisimilitude of cinematographic representation (in and of itself), it creates and sustains an artistic “illusion”—in the very specific sense of a virtual, highly affective, and immersive aesthetic reality, or “appearance.”

In keeping with Cassirer’s philosophy of symbolic thought, Langer also rightly (if somewhat implicitly) upholds the crucial distinction between the medium of cinema and its “formal” (i.e., artistic) uses, with the admission, however, of certain necessary constraints and advantages the latter places on the former in the interests of artistic creation. As Carroll has discussed, the crucial distinction in question has periodically been overlooked or marginalized in film theory. This has frequently come at the price of avoidable confusions and misplaced reductions of properties of cinema’s creative use (formal, artistic, narrative) to medium: prompting, for instance, Carroll’s proffered slogans for film theory of “forgetting the medium” and of attending to “use rather than medium.”²⁷

Finally, while the symbolic dimension of film worlds on which I will elaborate is closer to Goodman’s aforementioned understanding of a work of art as a complex network of interconnected referential functions (symbols), it equally supports (and is supported by) Langer’s contention that an artwork, as a whole and as experienced, conveys, through a presentational mode of abstracting and objectifying symbolization, a unique, embracing, irreducible, and work-generated “feeling.” This view anticipates Dufrenne’s more or less simultaneous (but independently arrived at) concept of total aesthetic expression and the world atmosphere of an artwork, as well as what I will later describe as the “world-feeling” of a cinematic work.

ART-MAKING, FILMMAKING, AND WORLD-MAKING

Nelson Goodman begins his study Ways of Worldmaking with an acknowledgment of his intellectual debt to Cassirer’s philosophy of symbolic forms (via Langer’s translations and advocacy).28 From one vantage point Goodman’s scheme of major symbolic relations, or functions, as first advanced in his highly influential Languages of Art, is a kind of abstract formalization of much of what Cassirer earlier argues concerning differences and similarities between art and language. This includes the ways in which at certain times, and in certain contexts, symbols may refer to their own properties (what Goodman refers to and theorizes as exemplification). With good reason Paul Ricœur has suggested that in some respects Goodman both extends and radicalizes Cassirer’s account, in arguing that language, science, and art are constituted by different conceptual and symbolic systems that construct worlds (or specific “versions” of the world) according to the functioning of the cognitive and referential categories and frames of reference particular to them.²⁹ The underlying unity of these most fundamental forms of representation is preserved in Goodman’s view, however, by the common presence of the same basic types of symbolic reference relations that he identifies (viz., denotation, exemplification, expression, and allusion).

Yet, and as he stresses, Goodman approaches the problems of art and art-making with a very different philosophical orientation, seemingly far removed from the critical idealism of Cassirer and the naturalistic “life-philosophy” of feeling and art espoused by Langer. As advised in the explanatory preface to Ways of Worldmaking, whereas Cassirer’s explorations of symbolic forms are deeply historical and anthropological, his own approach to the concept of a vast plurality of symbolically created worlds is formal, analytic, and synchronic. It is, in other words, almost completely ahistorical, as well as intentionally “nonintentional” (in philosophical parlance)—that is, nonpsychological—in orientation. As a logically minded Humean empiricist, Goodman addresses the interactions between “nonverbal symbol systems” from the vantage point of his particular brand of philosophical analysis, as rooted in a pronounced nominalism in logic and ontology.30

Providing sometimes radically different, even seemingly opposed, ways of ordering and rendering experiential or empirical reality more comprehensible, worlds, for Goodman, are made entirely of symbols that function together within larger systems. Each such system, or scheme, presents a distinct but potentially “right”—that is, explanatorily sufficient, coherent, and illuminating—version of reality, or “the way things are.”³¹ There is no means of reducing such different world-versions to a single set of master symbols or concepts, a set that, in any case, would have no more direct relation to a precategorized reality than the versions in question.³² The focus on the making, perceiving, and understanding of the symbolically constructed worlds to be found in the arts and sciences alike both supports and is a consequence of the major theme of Goodman’s aesthetics, that “the arts must be taken no less seriously than the sciences as modes of discovery, creation and enlargement of knowledge in the broad sense of advancement of the understanding” (102).

With its romantic-sounding ring and suggestion of godlike or Faustian powers of creation out of thin air (or “nothing but symbols”), the title of Goodman’s cited study can be somewhat misleading. “Worldmaking,” as a cognitive activity, as well as sometimes also a practical or technical one (involving, as in the case of art, physical and material realities) is primarily a matter of creatively transforming chosen features of cultural realities that already exist. Goodman stresses that just as scientific worlds, and our “everyday, practical” worlds are built on or over their historical predecessors, so, too, are specially, self-consciously made artistic worlds created from parts of other, older ones, as well as the established, external realities to which they refer (17). Thus, from a comparative perspective, at least, world-making in art is always fundamentally a “remaking,” whereby the worlds of existing artworks, and other relevant symbolic worlds or world systems (and the patterns and conventions that they have established) are transformed in accordance with the artist’s intentions, skills, and individual style (6). Through such transformations and innovations (often beginning with the commonplace), new artistic worlds are founded, bringing with them new interpretations of that which is most real in experience.

Peter Greenaway, however, provides perhaps the clearest example of the translation of these general, symbol-centered themes and ideas into filmic practice. Greenaway’s approach to cinema, rooted in an evident preoccupation with symbolic representation, is deeply “antirealist” in the general senses of both the core arguments and the assumptions of realist film theory, as well as any philosophical position tantamount to so-called naive (or commonsense) realism. While he is on record as being inspired by French structuralist and poststructuralist thought in this respect, his films also (and without contradiction) provide a particularly clear and highly self-reflective example of filmmaking as symbolic “world-making” in Cassirer’s and Goodman’s senses. From early, neo-avant-garde works such as A Walk Through H and The Falls to later, more conventionally narrative ones, Greenaway’s films, in both form and content, call special attention to artistic filmmaking as a higher-order creation of new forms and meanings, derived from the more extensively shared, suprasubjective, and fully formed materials of entire world systems made up of the individual “signs” of cultural life.

As numerous critics and theorists have noted, and as the director himself has suggested in interviews and DVD commentaries, at the center of Greenaway’s cinematic corpus are the myriad ways in which symbol systems impose order on what is “naturally” disordered and unknowable, through objectification and repetition, among other means.33 Their specific forms and subjects, drawn from the major symbolic forms of art, myth, language, religion, and science that Cassirer identifies, have ranged from the theory of evolution (a paradigmatic taxonomical enterprise) in A Zed and Two Noughts to architecture in The Belly of an Architect, mathematics and the iconography of sex and death in Drowning by Numbers to the seventeenth-century artistic, literary, and mythological archetypes of the Draughtsman’s Contract (figs. 2.1 and 2.2). For film scholar David Pascoe, “above all, Greenaway’s films offer an inventory of the tools for representation” and both present and reflect on numerous other “artificial orders and structures” through the lens of those cinematic ones that he, and other filmmakers, consciously create.³⁴ Yet in the context of fictions more or less fully realized, they also consciously foreground the relation between and among existing classificatory or representational systems and the ways in which such world-making schemas may be creatively combined in the medium of cinema, as its own hybrid vehicle of simultaneous cultural and individual construction, reflection, and expression. All of this is part and parcel of what has been variously referred to as Greenaway’s “taxonomical” (Pascoe 21), “encyclopedic,” “cartographic,” and “museum film” aesthetic, one that has also been compared with that of Joyce, Borges, and British painter R. B. Kitaj in these respects (Pascoe 42, 52). Such an approach to film and art-making is also highlighted in the forms and contents of Greenaway’s interactive artistic installations and exhibitions—as well as clearly signaled in some of their (Goodman-esque) titles such as “Some Organizing Principles” and “100 Objects to Represent the World” (Pascoe 204–6).

FIGURES 2.1 AND 2.2 The world as representation in Greenaway’s The Draughtsman’s Contract.

As his critics and supporters alike often claim, Greenaway’s films are clearly far removed in form and content from a great deal of relatively more conventional narrative cinema, as well as many prominent “art cinema” styles. In ways to be discussed, however, and with reference (in chapters 4 and 5) to Goodman’s epistemologically rooted theorization of art-making as “worldmaking,” and the particular processes and major types of symbolic functions it is argued to involve, what Greenaway’s filmmaking exemplifies about art and cinema as symbolic communication and expression amounting to creative transformation (as the foundation for the meaning and experience of his own films, and others) may still be justly considered paradigmatic of all cinematic world-making, from the most story- and character-driven to the most reflexive, conceptual, and abstract.

To pause now and take stock: I have offered in this chapter what may be considered a preliminary or background analysis of worlds as cultural and symbolic constructions. The position I have adopted is that works of representational art, including films to a preeminent degree, are in important respects continuous with such culture-forming and -sustaining domains of human meanings, interests, and actions: notwithstanding their primarily virtual as distinct from physical mode of existence. Cinematic works contain signs (capable of being conceived and analyzed in accordance with various traditional semiotic approaches) but also make use of more interpretation requiring symbols. Moreover, cinematic works are themselves symbolic representations and presentations of experience taking a specifically artistic and aesthetic form. As the “third tradition” of reflection in general symbol theory that we have identified maintains, this involves self-reference on the part of films to their own created forms and experiences as objects of attention, as much as to their represented contents. While inescapably “cognitive,” in the sense of involving, and greatly contributing to, reflective knowledge, neither in theory nor practice must the “symbolic form” of art, in cinema or elsewhere, entail any diminution of an artwork’s feeling dimension. In contrast, in singling out the unique and characteristic properties of artistic symbolization (as distinct from other forms of representation), the expressivist tradition in question helps provide a way of seeing the intellectual and affective poles as not only entirely compatible and copresent but necessarily conjoined. Finally—and further supporting what I suggested in my overview of existing conceptions and models of films and (or as) worlds in the first chapter—the views here discussed also point to various processes of creative transformation and audience immersion as central to the dynamics of artistic world-creation and experience as forms of symbolic understanding and experience. Our next task is to examine the specific cinematic manifestations of these transformative and immersive processes and their observable results.