Journal | Salon | Portal (ISSN 1466-4615)

Vol. 5 No. 26, August 2001



Warren Buckland

Problem Formation in the Analytic Philosophy of Film




_Film Theory and Philosophy_

Edited by Richard Allen and Murray Smith

Oxford: Clarendon Press, 1997; pbk 1999

ISBN 0-19-815921-8 (hbk); 0-19-815988-9 (pbk)

474 pp.


What are the issues that drive a philosophical analysis of film? In this review I have set myself the task of trying to identify those issues as formulated by the contributors to Richard Allen and Murray Smith's edited volume _Film Theory and Philosophy_. But first, I shall say a few words about the particular approach this book takes, outline my particular approach in reviewing this book, and then examine the issues discussed in each chapter.


In the jointly written Introduction, Allen and Smith offer a brief outline of their philosophical framework -- analytic philosophy, present a detailed critique of film theory influenced by Continental philosophy (particularly Althusser and Lacan), and end by discussing the problematic relation between science and analytic philosophy. Allen and Smith present analytic philosophy as a non-doctrine based discipline, defined by its method of argumentation, or reasoning strategies. In addition to disambiguation of concepts, 'typical strategies for authors in the analytic tradition in pursuit of logical precision would be the discrimination of strict, logical implication from implicature (the implied or indirect assertion of a proposition), and of deductive from inductive and abductive reasoning; the unearthing of hidden premises . . .; the scrutinizing of an argument for the presence of circularity, or question-begging; the attention to paradox, but also to its dissolution . . .; and the discrimination of logical contradictions from logical contraries' (5). The reader encounters many of these reasoning strategies throughout the entire volume.


In terms of my reasoning strategies, I shall use Rudolf Botha's philosophical study into the conduct of inquiry to analyse the way the various authors in _Film Theory and Philosophy_ formulate conceptual and empirical problems. [1] I focus on problem formation because I am convinced by Botha's (and Larry Laudan's [2]) argument that the rationality of a theory is based on its problem-solving effectiveness. Theories are important, therefore, to the extent that they provide solutions to conceptual and empirical problems. Botha focuses on problem formation in linguistic inquiry, although his analysis is of course applicable to other fields of research. I have already used Botha's work to analyse the formation of theoretical problems in film theory -- most notably in my chapter 'Film Semiotics' in Toby Miller and Robert Stam's edited volume _A Companion to Film Theory_, [3] as well as in my forthcoming review of Francesco Casetti's book _Inside the Gaze_. [4] For the sake of the coherence of this review, I shall reproduce my short summary of Botha's work as found in my review of Casetti's book before reviewing _Film Theory and Philosophy_.


Botha lists four activities involved in formulating theoretical problems:

(a) analysing the problematic state of affairs;

(b) describing the problematic state of affairs;

(c) constructing problems; and,

(d) evaluating problems with regard to well-formedness and significance. [5]


This list is based on the distinction between a 'problematic state of affairs' and 'problems'. Whereas the former refers to an aspect of reality a theorist does not understand, a problem formulates what a theorist needs to look for in order to resolve the problematic state of affairs.


In carrying out (a), analysis, the theorist must know exactly what is problematic, isolate each component of the problematic state of affairs, determine how they are interrelated, and identify the background assumptions informing his or her inquiry, such as the nature conferred upon the object of analysis.


In carrying out (b), description, the problematic state of affairs must be accurately recorded and formally described. For Botha, this involves three processes: (i) collecting data; (ii) systematising data; and (iii) symbolising the results. [6] In collecting data, the theorist must determine whether the data or the theory generates the problematic state of affairs. Systematising data involves the activities of classifying, correlating, and ordering. These activities enable the theorist to identify common properties among data, put similar data into classes, and determine the relations between the classes. Finally, symbolising involves representing data in a concise and accurate manner.


In carrying out (c), constructing problems, the theorist employs several different concepts (since a problem is made up of concepts). Botha identifies four types of concept involved in constructing problems (here I have modified his list to fit film theory): phenomenological concepts, which concern factual data and are intuitively known; filmic concepts (what Botha calls grammatical concepts), general background assumptions concerning the nature of individual films; cinematic concepts (what Botha calls general linguistic concepts), which concern background assumptions about the nature of film; and metatheoretical concepts (what Botha calls metascientific concepts), which concern the aims and nature of theoretical inquiry. [7]


In carrying out (d), evaluating problems, Botha recognizes that only problems satisfying the criteria of well-formedness and significance are relevant problems worth pursuing. A well-formed problem is solvable -- that is, is based on correct assumptions, and is clearly formulated. A significant problem is one that expands our existing knowledge of film. A problem may, therefore, be well-formed, but may not be significant.


Authors do not formulate and write out their theory in the manner made explicit by Botha's systematic and logical steps; such steps are the privilege of the philosopher. This adds indeterminacy when analysing any text in terms of Botha's categories. Furthermore, the stage of theorising I concentrate on here, the formation of problems, is only one stage -- albeit one of the most important -- in the development of a theory. Other stages include: giving descriptions of the object of study, giving explanations, making projections, justifying hypotheses, and so on. While analysing the formation of problems in a number of chapters in _Film Theory and Philosophy_, the reader should bear in mind that each chapter can be analysed again in terms of other stages in the development of a theory. To make this review manageable, I have primarily limited myself to the way authors formulate problems and, furthermore, I have limited myself to half the chapters, making do with a paragraph summary of the other half. This decision to analyse some chapters closely and summarise the others is not necessarily a judgment of quality -- since the editors have done a remarkable job in commissioning a consistently high standard of chapters. Instead, some chapters are discussed briefly for reasons of space and time.


Finally, I should add that my aim in reviewing this book is not, in fact, to pass judgment on the essays. (I shall only make direct evaluative judgments in terms of Botha's criteria for well-formedness and significance.) Instead, my aim is more 'object-focused', in that I am more interested in identifying the various issues or problems that emerge from the intersection of analytic philosophy and film theory. My aim in this review is therefore to offer a comprehensive survey of a long (474 pages), complex, and dense book. I have attempted to read the essays on their own terms in an attempt to convey to the reader the directions film theory is currently taking according to the contributors to this book.



Part I: What is Cinematic Representation?


_Film Theory and Philosophy_ contains 19 chapters divided among 5 parts. Part I: What is Cinematic Representation? contains chapters by Gregory Currie, Kendall Walton, Richard Allen, and Edward Branigan.


In 'The Film Theory that Never Was: A Nervous Manifesto', Gregory Currie outlines the philosophical starting point for film theory by listing a series of questions, and then summarises the results of his work to date that answers some of those questions.


(a) Analysing the problematic state of affairs. Currie outlines several types of interrelated questions that identify problematic states of affairs for film theory to tackle, organised from the general to the specific -- questions about: 1. film's nature; 2. modes of filmic representation; 3. standard types of engagement with film; 4. the individuation of filmic elements and their connections; 5. film production; 6. film kinds (grouping by genre, author, etc.); 7. film style; 8. individual films. Answers to these questions will constitute a philosophy of film, which Currie conceives as a grand theory (in opposition to David Bordwell and Noel Carroll's call for piecemeal theorising).


(b) Describing the problematic state of affairs. Currie presents in summary form his answers to questions 1, 3, and 6. And because he focuses on the results of his reasoning, more than the way he formulated problems, then much of his essay falls into a later realm of theorising, namely giving descriptions and explanations.


In relation to question 1, Currie argues that film is a realistic moving picture, that we literally perceive movement on screen, rather than an illusion of movement. He then qualifies this proposition to argue that we perceive apparent (non-illusory) motion. In terms of realism, Currie argues that film images present a likeness of the objects represented, and that spectators comprehend images in the same way they comprehend objects, although of course he acknowledges the many differences between images and objects. His theory of film realism simply posits a common property between an object and an image of that object. He also argues that individual shots depict space and time realistically.


Further, Currie argues that few conventions, understood as arbitrary rules, are involved in the comprehension of films (51), leading him to argue that genres are not governed by arbitrary conventions, a partial answer to question 6. In another response to question 6, Currie defends the concept of author intentionality for it is useful in understanding what stories are told in films (enabling the theorist to classify them) and how spectators comprehend films (54).


The issue of how spectators react to films (question 3) is the final part of Currie's theory outlined in this chapter. He summarises his theory of impersonal imagining in film viewing, in which the film spectator 'imagines the events of the fiction taking place, but does not imagine being in specific spatio-temporal relations to those events . . . What I deny is that the standard mode of audience role-play in film watching consists in playing the role of someone who is seeing those events as they happen' (55). Currie gives an example: 'the film viewer imagines Marion being attacked in the shower, but does not imagine being there in the shower to share the experience' (55). He compares and contrasts his theory of impersonal imagining with Kendall Walton's theory of fiction as make-believe, in which fiction is compared to games of pretence that children play. Currie agrees that fiction can be understood as make-believe, but argues that Walton's theory involves personal imagining, of the individual being involved with the make-believe. (The distinction between personal and impersonal imagining is taken up again by Murray Smith in his contribution to Part V.)


Currie ends by arguing that it is empirically decidable whether film spectators engage in personal or impersonal imagining, and suggests that research into autism can aid research into imaginative engagement with fiction films.


(c) Constructing problems. Question 8, about individual films, and question 3, about standard types of engagement with film, rely on phenomenological concepts, whereas the other questions rely on filmic and cinematic concepts. Furthermore, Currie favors theories that stick close to everyday, intuitive, common sense understanding (and he approvingly cites the philosophy of G. E. Moore (45)). The more abstract a theoretical concept, the more unreliable it is. In terms of metatheoretical concepts, Currie defends a grand theory approach to theorising, because 'a theory that is strongly integrated across the domain of film and strongly linked to successful work in other areas will be better -- more simple, coherent, and therefore more credible -- than a bunch of disparate theories isolated from other branches of knowledge' (43). Currie also justifies the philosophical approach to these questions because many of them (study of essence, representation, kinds of film, etc.) are philosophical questions from the outset, and successful answers will depend on sound philosophical reasoning.


(d) Evaluating problems. At the beginning of his essay Currie states: 'What I aim to do here is to offer an absurdly ambitious conception of what an analytical philosophy of film would look like if it took seriously the aim of constructing a systematic and globally connected theory' (43). Because of the scope of his manifesto, Currie's work has a potentially huge significance for film theory, for it aims to reformulate old questions and provide new answers. The individual theoretical questions he asks are also well-formed, although the ambition to interrelate them into a new grand theory means that, taken together, they will be difficult to solve.


In 'On Pictures and Photographs: Objections Answered', Kendall L. Walton summarizes his theory of depiction in images, and then addresses Gregory Currie's and Noel Carroll's responses to his theory. For Walton, images are props in visual games of make-believe: 'By this I mean, in part, that in looking at a picture the spectator imagines seeing what it portrays' (60). He then gives an example: 'seeing a picture of an ox involves thinking of oneself as looking at an ox. We can put this by saying that one *imagines* seeing an ox, as one looks at the picture' (61). Walton then challenges Currie's theory of impersonal imagining (61-65) and Carroll's objections to the concepts of make-believe and imagining seeing to explain pictorial representation (65-66). Walton ends by defending his theory of the transparency of photographic pictures (67-72).


In 'Looking at Motion Pictures' Richard Allen addresses the paradox common to the four types of causal theory of perception and then argues that, by abandoning these causal theories the paradox they create simply dissolves. Allen is therefore using the reasoning strategies of analytic philosophy (especially the later Wittgenstein) to show that the problematic state of affairs is created by faulty reasoning, not by the object of study.


(a) Analysing the problematic state of affairs. Allen's essay is distinctive in that it is explicitly based on identifying what is problematic in traditional causal theories of perception -- identifying the background assumptions behind these theories, and determining how these background assumptions generate problematic states of affairs. Allen begins by asking: 'What is it we see when we look at a motion picture?' (76), and then argues that such a question depends on a more fundamental question, of what we understand by the activity of seeing. Philosophical inquiry into this question has traditionally been dominated by causal theories of perception, which define seeing as 'a form of experience, a perceptual one, that is caused by the presence of the object in front of one's eyes' (76). Causal theories, Allen argues, lead to a paradox when describing how one perceives pictures, because the perceiver commonly reports on seeing, not the picture, but what the picture depicts.


(b) Describing the problematic state of affairs. In terms of systematising data, Allen has systematically identified all the main causal theories of perception, identified their background assumptions and diagnosed their flaws. Here I shall systematise his work further.


Allen argues that visual theorists have developed four causal theories to overcome the paradox of what we perceive when looking at a picture:


1. Illusion theory (dominant in Continental film theory), in which a picture causes the perceiver to have the same visual experience to the experience generated by the absent objects depicted. It attempts to overcome the paradox by arguing that we see the object a picture depicts because we see an illusion of that object.


2. Transparency theory (common in the classical film theory of Bazin, Kracauer, and Cavell, as well as in Kendal Walton's work), in which the photographic image makes objects present via photographic reproduction. It attempts to overcome the paradox by arguing that we see the object a picture depicts because what we see is the object itself.


3. Imagination theory (Walton) argues that we do not see what a picture depicts, although a picture enables us to *imagine* seeing what it depicts. It attempts to overcome the paradox by arguing that we do not see the depicted object, but only imagine that we see what the picture is of.


4. Recognition theory (Noel Carroll) also argues that we do not see the depicted object. However, it attempts to overcome the paradox by arguing that we only see a disposition of shapes and colors on a flat surface, which nonetheless afford us a recognition of what it depicts.


For Allen, all these theories are mistaken. Their causal background assumptions lead them to posit a causal link between the perceiver and a physical object. (Allen nonetheless finds value in Walton's transparency thesis, after it is shorn of its causal background assumptions (92).) Allen's solution is to argue that: 'we require an understanding of seeing pictures that, contrary to imagination and recognition theorists, respects the fact that seeing what a picture is of is a genuine case of seeing, without commitment to the idea that what we see is the object itself or an illusion of it' (77). The result is that we no longer need to think of the object of sight as a physical object, which then dissolves the paradox, for the perception of pictures is no longer understood to be a problem that needs to be solved.


(c) Constructing problems. Allen does not so much construct problems, as dissolve them. This leads him to propose a radical metatheoretical solution to the analysis of what we perceive when looking at motion pictures: 'abandoning the causal theory of perception does not simply involve relinquishing *a* theory of motion picture perception, it involves the abandonment of film theory itself as the route to understanding what it is we see when we look at motion pictures' (78).


(d) Evaluating problems. Allen uses the reasoning strategies of analytic philosophy to argue that the problem or paradox that has beset causal theories of perception is not a genuine philosophical problem. Abandoning causal theories of perception leads to the dissolution of the problem. In Botha's terms, Allen would say that the problem is not well formed or significant, and Allen's purpose in his essay is to demonstrate this point through reasoned argumentation.


In 'Sound, Epistemology, Film', Edward Branigan investigates the physics and phenomenology of sound and light in order to determine, in an exact and exhausting manner, the perceptual and material similarities and differences between sound and image in the cinema. Branigan begins by asking a number of epistemological questions that address problems about the relation between light and sound:


'Is sound less closely tied to the Kantian category of substance than vision? If so, what presuppositions about sound direct our search for knowledge from the visual features of film? May these presuppositions be altered to change our perception of the relationship between sound and light? More generally, how do we expect sound to be of use to us in describing the world and in imagining a real world through the fictional depictions of a film? How does sound relate to the structures of language?' (96).


Although both sound and image have the same physical basis in wave motion, sound appears to be transitory and contingent, while light appears to be more permanent and bound to material objects, which accounts for the traditional privileging of image over sound. One of Branigan's main arguments is to overcome this traditional way of thinking about sound, by rethinking the relation among sound, motion, space, and time.



Part II: Meaning, Authorship, and Intentions contains chapters by Paisley Livingston, Berys Gaut, Noel Carroll, Trevor Ponech, and George Wilson.


In 'Cinematic Authorship' Paisley Livingston addresses the problem of whether the concept of authorship can be applied to the -- especially commercial -- cinema. The first problem is one of an agreed definition, and Livingston offers a broad definition relying on the understanding of an author as a rational agent who expresses or communicates an intended meaning (134). The next problem is to apply this intentionalist definition of author to mass-produced commercial films, which are -- to a greater or lesser extent -- still unique films. The issue is to identify an author from the numerous makers of or contributors to a commercial film. Livingston presents four thought experiments, or hypothetical cases studies, of the different power relations that can operate between different agents in film making as a way of distinguishing makers from authors.


In 'Film Authorship and Collaboration' Berys Gaut also takes up the issue of authorship in the cinema, and proposes that it can only be understood in terms of multiple authorship.


(a) Analysing the problematic state of affairs. Like Livingston, Gaut argues that the everyday, intuitive understanding of 'film authorship' is too vague. He then diagnoses its problematic state of affairs, and formulates a number of problems. In terms of problematic state of affairs, Gaut writes: 'It has been held that the film author is the director, the screenwriter, the star, or the studio; that the film author is an actual individual, or a critical construct; that there is not one film author, but several; the claim of film authorship has been held primarily as an evaluative one, or an interpretative one, or simply as the view that there are authors of film as there are authors of literary works' (149). There are two problems to address: firstly, to determine whether these various theories are drawing on some core truths, or whether there is no truth in the concept of film authorship, and each successive theory is simply based on a foundation of (hot) air and rhetoric. Secondly, if there is some truth to the concept of film authorship, how can we identify singular authorship in a film? Gaut rejects the concept of single film authorship and argues for multiple authorship in mainstream cinema.


(b) Describing the problematic state of affairs. In terms of systematising data, Gaut identifies five basic 'ingredients' or concepts of film authorship, with sub-categories: 1. The kind of claim auteur criticism makes, divided into three sub-categories: a. existential claim (that film authors exist), b. hermeneutic claim (films can be interpreted by relating them to their makers), and c. evaluative claim (auteur criticism aims to evaluate films). 2. The ontology of the author, divided into: a. actual persons, or b. critical constructs. 3. Authors and artists, divided into: a. the author is an artist, or b. a literal author. 4. Occupiers of the authorial role: a. directors, b. screenwriters, c. stars, and d. producers. 5. Number of authors: a. single authorship, or b. multiple authorship. Any particular auteur theory can be characterised by specific combinations of these basic concepts -- and Gaut argues that there are in total 180 possible combinations (169 n. 13).


(c) Constructing problems. Gaut attempts to dispel the intuitive, phenomenological understanding of film authorship. In terms of filmic concepts, he argues that some films are works of art (and are therefore made by artists). In terms of more general cinematic concepts, he argues that mainstream cinema is inherently a collaborative medium.


(d) Evaluating problems. Gaut's chapter is extremely well-formed, for it clearly identifies the problematic state of affairs surrounding the concept of film authorship, identifies two main problems to solve, systematises the concepts surrounding film authorship by breaking it down into its basic components, and then offers a solution to the problems (multiple authorship). Furthermore, because the idea of the film author is a long-standing issue in film studies, Gaut's chapter is significant for it clarifies this idea and furthers our understanding of it.


In 'Fiction, Non-fiction, and the Film of Presumptive Assertion: A Conceptual Analysis', Noel Carroll argues that the term 'documentary' is too narrow to define the current practices of films we attribute the label 'documentary', and that the term 'non-fiction film' is too broad. He proposes the concept of the 'film of presumptive assertion' as an alternative.


(a) Analysing the problematic state of affairs. For Carroll, the commonly accepted definition of documentary (by Grierson) is no longer applicable to current 'documentary' practices. Rather than stretch the meaning of the word 'documentary', Carroll thinks we need a new term: 'we find ourselves in a situation where we have, on the one hand, the relatively precise notion of the documentary that Grierson has bequeathed us, and, on the other hand, another more ambiguous idea [about current 'documentary' practices]. This at the very least courts confusion. I propose to relieve that confusion by granting Grierson his definition for what he was talking about and by introducing a new concept for what we wish to speak about' (174). Carroll then rejects the label 'non-fiction film' because it is too broad. His label, the 'film of presumptive assertion', is a sub-category of the non-fiction film. This in turn requires that the distinction between fiction and non-fiction be clearly defined and upheld.


(b) Describing the problematic state of affairs. Carroll's essay consists of a conceptual analysis of the fiction/non-fiction distinction (which he defends against the Continental film theorists' assertion that this distinction is no longer valid), and a definition of 'the film of presumptive assertion'.


After describing the problematic state of affairs, Carroll begins to solve them, by adopting Paul Grice's pragmatic theory of meaning, which is based on authorial intentions (which, in film studies, translates into the film-maker indicating to spectators how they are intended to respond to his or her film, and spectators recognising that intention). With a fiction film, spectators are intended to respond to the film in such as way as to understand the content to be unasserted (that is, to comprehend them as suppositions). For non-fiction films, Carroll simply reverses the proposition: spectators should not understand the film's content to be unasserted; that is, spectators should not comprehend non-fiction films suppositionally.


Carroll then defines films of presumptive assertion. Simply put, spectators are intended to comprehend the content of these films as asserted. Moreover, 'I call them films of *presumptive* assertion not only because the audience presumes that it is to entertain the propositional content of such a film as asserted, but also because such films may lie. That is, they are presumed to involve assertion even in cases where the film-maker is intentionally dissimulating at the same time that he is signalling an assertoric intention' (187). Carroll then briefly contrasts films of presumptive assertion with what he calls films of presumptive trace (188-91), a subset of films of presumptive assertion in which every image is an authentic document or trace of its content. This term is therefore close to Grierson's meaning of documentary, and is too narrow to cover current 'documentary' practices, which may involve staging, re-enactments, animation, and so on. Hence, Carroll prefers the term 'films of presumptive assertion' because such films are not always based on authentic documents or historical traces, but are defined by the film-maker's assertoric intention.


Finally, Carroll practices an additional stage of theorising -- the justification of his ideas. In particular, he justifies his reliance on the concept of intentionality, of maintaining the fiction/non-fiction distinction, and for asserting that films of presumptive assertion are necessarily objective.


(c) Constructing problems. Carroll argues that the everyday, intuitive (phenomenological) understanding of documentary and non-fiction film-making is too vague. His chapter operates on the level of cinematic concepts to offer a rigorous definition of a major film-making practice. In terms of metatheoretical concepts, Carroll implicitly generates his argument and identifies problems by means of 'The Topics' (as do a number of authors in _Film Theory and Philosophy_). The topics name a grid for generating arguments, a system that classical rhetoricians devised to help find something to say. Topics enable the writer or speaker to conduct a conceptual analysis of the subject under discussion. Definition is one of the common topics for conducting a conceptual analysis of any subject matter.


(d) Evaluating problems. As one would expect from Carroll, he clearly diagnoses the problematic state of affairs in current thinking about film, and formulates well-formed problems. Moreover, because he is analysing and revising current thinking on a major category of films, his essay also has significance, although it is difficult to assess at this stage whether his ideas will actually change current academic thinking on documentary film, and whether his commitment to objectivity (and intentionality) will be accepted -- or, indeed, whether the unwieldy term 'films of presumptive assertion' will catch on.


In 'What is Non-fiction Cinema?' Trevor Ponech aims to specify 'what it is that causes a movie to be non-fiction' (203). His approach is very similar to Carroll's, to the extent that he makes almost identical points and develops the same arguments concerning non-fiction films. For example:


'A cinematic work is non-fiction if and only if its maker so intends it' (204). 'Documentaries . . . are cinematic assertions, naturally meaningful images being among the elements employed by the communicator toward assertive ends' (205). 'In asserting that something or other is the case, cinematic agents typically expect audiences -- employing a combination of perceptually derived beliefs about the depiction, non-perceptual beliefs and background knowledge, and inferences -- to arrive at particular cognitions regarding not only what is shown on the screen but also how things stand in the world' (207). '[T]he difference between fiction and non-fiction is one of force, not of style, form, or content' (216).


Ponech's chapter begins to differ from Carroll's when it introduces two main strategies of non-fiction film-making, which are distinguished on the basis of the degree of control the film-maker exercises over their subject matter: 'distinguishing between Type I and Type II non-fictions reflects the intuition that there are genuine if not radical differences between, for instance, works of observational cinema and more theatrical documentaries' (217). Ponech ends by justifying his adherence to author intentionality (although he is in good company in _Film Theory and Philosophy_). The main problem with this chapter is that, although Ponech developed it independently of Carroll's chapter, it says much the same things, and the discussion of different types of film is underdeveloped, and certainly does not challenge the far more sophisticated taxonomy of types of documentary developed by others, such as Bill Nichols.


In 'On Film Narrative and Narrative Meaning' George Wilson focuses on the distinction David Bordwell draws between comprehending film and the interpretive activity of explicative criticism, which analyses a film's implicit meanings. Through the analysis of a scene from Nicholas Ray's _Bigger than Life_ (1956), Wilson challenges the viability of Bordwell's distinction. One of the major problems with Bordwell's proposal, for Wilson, is that implicit meanings are not limited to the act of interpretation, but are part of the activity of comprehending a film as well: 'explicative interpretation is a natural, although sometimes more sophisticated, extension of familiar and absolutely basic strategies by means of which typical audiences comprehend a movie' (234). This is because the spectator's main activity in watching a movie, according to Wilson, is to identify and construct a pattern of explanatory, causal connections to assess characters' actions.



Part III: Ideology and Ethics contains chapters by Jennifer Hammett, Hector Rodriguez, and Tommy Lott.


In 'The Ideological Impediment: Epistemology, Feminism, and Film Theory', Jennifer Hammett analyses the consequences of the feminist film theorists' problematising of representation as an alienating illusion. She then proposes this to be a false problem.


(a) Analysing the problematic state of affairs. The problematic state of affairs Hammett identifies lies at the heart of feminist film theory. The premise of feminist film theory is that, like language and other forms of discourse, cinematic representations are ideological constructions that naturalise patriarchal ideology. One dominant issue to emerge from this theory is, therefore, how to evade patriarchal ideology. The two dominant responses -- an appeal to the radical potential of developing critical distance (the idealist route), and an appeal to authentic, everyday reality (the realist or essentialist route) -- both 'assume that our embeddedness in language and representation matters' (245). Hammett finds this assumption to be in error and, in accordance with the agenda of analytic philosophy, she argues that '[o]nce we abandon the premise that our embeddedness in language matters, then the need to escape that embeddedness -- to achieve a neutral encounter with phenomena -- simply disappears' (246). She then adds: 'If I am right, feminists can give up the epistemological ambition of feminist film theory, and pursue feminist critical and political goals unencumbered by fears of the alienating effects of representation' (246).


(b) Describing the problematic state of affairs. Hammett collects her 'evidence' by outlining the position of two key feminist film theorists, Mary Ann Doane and Christine Gledhill. She then orders their work under the headings 'idealism' and 'realism'/'essentialism' respectively.


(c) Constructing problems. As with Richard Allen's contribution, Hammett does not so much construct problems, as use the methods of analytic philosophy to dissolve theory and the problems it raises. Hammett's contribution to feminist analysis is to recommend a shift in focus from cinematic concepts to filmic concepts (from a feminist theory of 'the cinema' to a feminist critique of individual films, for the general issue of representation is not a problem per se).


(d) Evaluating problems. Hammett dissolves problems, and so we need to evaluate the problems she dissolves. She argues that the problem dominating feminist film theory -- how to escape patriarchal ideology (plus the two solutions: an idealist and a realist one) -- is not well formed or significant: 'In searching for a feminist response to the reign of representation, feminist film theory has been pursuing a futile and pointless goal' (257).


In 'Ideology and Film Culture' Hector Rodriguez challenges the standard definition of ideology as false consciousness (a view Noel Carroll upholds), in which ideology consists of erroneous propositions. The problem with this standard definition for Rodriguez is that it assumes ideology simply consists of propositions that can be empirically tested, and that challenging someone's ideological beliefs is simply a matter of presenting to them factual information that falsifies those ideological propositions. This empirical view of ideology does not sufficiently encompass what it means to hold ideological beliefs: 'Evidence will often fail to convince because factual information is not a necessary ground of ideological belief . . . The underlying assumption here is that a belief can be rationally evaluated without taking into account its place within a field of human practices and concerns' (262-63). To challenge someone's ideological beliefs requires a more fundamental transformation -- of how they perceive, feel, make choices, act, and live. In other words, ideology is a moral issue, rather than an issue in inductive reasoning, and Rodriguez uses the term 'moral picture' to capture his sense of ideology as morally unjustifiable (268). He then conducts a short analysis of _The Wind and the Lion_ (John Millius, 1975), a film that 'invites us to see colonial history in terms of a moral picture that endeavours to legitimize [Millius's] country's foreign policy by idealizing military conquest as the expression of an existential confrontation with risk and death [and which] systematically overlooks the patterns of systematic exploitation and abuse imposed on colonized peoples, as well as the economic institutions and interests that encourage and subsidize military expansion' (277). For Rodriguez, therefore, ideological film analysis 'brings out the [moral] pictures that undergird a certain pattern of social and political commitment, so as to reveal something morally undesirable about those pictures and that commitment' (277).


In 'Aesthetics and Politics in Contemporary Black Film Theory' Tommy Lott diagnoses a number of problems in black film theory, problems caused by its reliance on the premises of Continental film theory, namely: black film theory's conflation of film aesthetics and ideology, and its automatic critique of commercially produced black Hollywood films and automatic valorisation of black independent films. Lott then relates these two problems: 'A too rigid distinction between studio-produced and independent black films has often been a source of confusion regarding the aesthetic value of films that have been produced by both groups of film-makers' (283). Lott argues that '[t]he growing number of Hollywood movies by independent black film-makers seems to demand a more nuanced black film commentary than the standard critique of Hollywood that was fostered by earlier blaxploitation era films' (283). Lott reviews the paradigms upheld to define black cinema (Third cinema, the films of Melvin Van Peebles), plus the critical methods of analysis, particularly reception studies. Lott argues that the reception studies approach to black spectators is insufficient in itself to determine the political and aesthetic value of (particularly studio made) black films. Like may contributors to _Film Theory and Philosophy_, he advocates that the intentions of the film-maker override audience response.



Part IV: Aesthetics contains essays by Peter Kivy, Flo Leibowitz, and Deborah Knight.


In 'Music in the Movies: A Philosophical Enquiry', Peter Kivy begins by arguing that film music has its roots in eighteenth century melodrama (spoken drama with musical accompaniment in the background). From this historical beginning, he identifies a problem: in silent film, music served a melodramatic function by filling the vacuum of silence. But with the advent of talking pictures, musical accompaniment continued beyond its obvious function of filling the silence: 'My question -- the question of this chapter -- is *why* the music plays on when the sound comes in? If the filmic function of music, in the silent era, is to fill the vacuum left by the total absence of expressive sound, why does it outlast its function when the full resources of expressive sound fill that vacuum in the era of the talking picture?' (314). Kivy answers that a gap still exists in the film after the introduction of speech, a gap that music attempts to fill. He identifies these gaps as subtle cues of human emotive expressions that film cannot capture, but which music offers a substitute (322).


In 'Personal Agency Theories of Expressiveness and the Movies' Flo Leibowitz examines three theories of expressiveness -- by Richard Wollheim, Bruce Vermazen, and Stephen Davies -- and then considers their relevance for developing a theory of film expressiveness. She spends a lot of time on Wollheim and Vermazen before rejecting them, and then turns to Davies's theory of expressiveness in music as the most relevant for film. For Davies, it is not the composer who is expressive, but the sound of the music itself. Intentionality is therefore attributed to the music, not to a personal agent. Leibowitz find this to be a suitable theory of expressiveness in film.


In 'Aristotelians on _Speed_: Paradoxes of Genre in the Context of Cinema', Deborah Knight analyses the consumption of popular genres. In particular, she diagnoses a paradox that Carroll has formulated in relation to the consumption of popular genres as an ill-formed paradox. She then addresses the 'paradoxical' state of affairs that Carroll analysed and describes it differently.


(a) Analysing the problematic state of affairs. The problematic state of affairs Knight analyses has already been formulated into a problem by Noel Carroll: Is the behaviour of spectators watching genre movies rational? The premise of this problem or paradox, for Carroll, is that, because spectators know the genre formulas, they will already know the formulaic stories in genre films (and in pulp fiction). So why watch them?


(b) Describing the problematic state of affairs. As well as collecting data from Carroll's essay, Knight analyses several popular genres and films, and focuses on the action film _Speed_ (Jan de Bont, 1994). Carroll solves his own paradox by arguing that, although genre texts are completely predictable, spectators instead consume them primarily on the level of plot, by anticipating what will happen next. Knight argues that Carroll's paradox is ill-formed because no genre film is completely predictable and cannot, therefore, be reduced to its generic formula. An adequate theory of consumption of genre films should combine the spectator's expectations set up by genre formulas as well as the way the spectator anticipates future actions on the basis of the canonical story format. Neither are completely predictable, and so both actively contribute to the consumption of genre films.


(c) Constructing problems. Both Knight and Carroll are centrally concerned with the relation between filmic concepts (the individual genre film) and cinematic concepts (or at least the cinematic as divided into several genres), although each relate the filmic to the cinematic differently. Furthermore, the 'paradox' they focus on is phenomenological to the extent that their main object of study is the consumer's everyday experience of genre fiction.


(d) Evaluating problems. Knight has already judged Carroll's paradox to be ill-formed, although she thinks it is a significant problem because she reformulates it and addresses it herself.



Part V: Emotional Response, the final part of the book, contains papers by Carl Plantinga, Dirk Eitzen, Murray Smith, and Malcolm Turvey.


In 'Notes on Spectator Emotion and Ideological Film Criticism' Carl Plantinga addresses the issue of emotion in ideological film criticism, and considers two examples: sentimentality, and emotions that accompany screen violence. He first clears the conceptual groundwork by critiquing ideological film criticism as developed within Continental film theory, particularly its suspicion of all emotions elicited by mainstream cinema (a suspicion fostered through Bertolt Brecht), and its ideological formalism, in which particular filmic techniques are conceived to inherently embody ideological or progressive political effects. The main problem here is not only determinism, but also that film content is not taken into account. Plantinga takes a more piecemeal approach to film emotions, by arguing that only some emotions elicited by mainstream films are ideologically suspicious. He therefore challenges the determinism of Continental film theory, and defends a cognitive theory of emotions. He formulates these issues into a problem in the following sentence: 'It is curious, then, that film theory has tended to neglect 'content' in its descriptions of spectator response, and has failed to theorize the place of cognition in spectator emotion' (379). Later he brings into focus the issue of the cognitive theory of emotion: 'The relevant issues then become not how films determine spectator response but what spectators bring to a film that influences their response, and how contextual factors delimit spectator responses and interpretations' (382).


In examining sentimentality and violence, Plantinga attempts to work out the relations among cognition, emotion, and ideology. Although inconclusive, his discussion attempts to avoid determinism by arguing that emotions do not in themselves necessarily embody any particular ideological position (384), and therefore that emotions such as sentiment can be used for beneficial or questionable ideological ends. In terms of violence on screen, the issue is not so much the violence itself, but the emotional involvement with that violence that spectators are encouraged to experience.


In 'Comedy and Classicism' Dirk Eitzen reassesses Bordwell, Staiger and Thompson's model of classical Hollywood cinema [8] and seeks to refine it, for it does not take into consideration the role and popularity of comedy in classical Hollywood. In particular, Eitzen questions the centrality of a psychologically-motivated cause-effect narrative logic in Bordwell, Staiger and Thompson's model, and argues that this model needs to be revised to take into account the counter-current of comedic elements such as gags, exaggerated behaviour, and parody. In an argument similar to the one Deborah Knight formulated against Carroll, Eitzen argues that spectators do not go to the cinema simply to experience a narrative and to solve its problems. Narrative causality is merely a means to an end -- to generate emotions in spectators. To ignore this end is a fatal flaw in (particularly) Bordwell's cognitive theory of film narrative. Referring to a study by the neurophysiologist Antonio Damasio, [9] Eitzen proposes that Bordwell implicitly posits film spectators as psychologically damaged, as spectators who are unable to comprehend emotions (406-7). In Eitzen's account, it seems that Bordwell has constructed a hypothetical spectator analogous to the one posited by Continental film theorists, for whom the film spectator is inflicted with 'deviant' sexual behaviour such as voyeurism and fetishism. Of course, it is also possible to argue that Bordwell recognised the importance of emotions in film viewing at the outset, but delimited the scope of his research in order to make it manageable. Eitzen is questioning whether Bordwell's artificial separation of cognition from emotion fatally distorts the study of spectatorship in the cinema.


In 'Imagining from the Inside' Murray Smith develops and refines his idea of 'central imagining' (a form of imagining from the inside of a fiction) as presented in his book _Engaging Characters_, [10] in light of the development and critique of this or similar concepts by other film scholars (most notably Gregory Currie).


(a) Analysing the problematic state of affairs. Smith begins by analysing a scene from _Dead Calm_ (Phillip Noyce, 1989), noting that the scene had a physical effect on him, a 'visceral flinching' (412). The problematic state of affairs Smith identifies from this experience have to do with the power of cinema to create such an experience. He then formulates two questions that address the problems he pursues in this chapter: 'what is the place of imagining a character 'from the inside' in engaging with a fiction? And, what is the function of POV [point-of-view], and other striking devices like sudden movements and loud noises, with respect to imagining a character's experience 'from the inside' in our engagement with cinematic fictions?' (412-13). In effect, the problem Smith addresses is the relation between textual structures (such as POV sequences) and psychological processes such as imagining.


(b) Describing the problematic state of affairs. Smith collects data from several film examples (especially POV sequences) and from his own intuitive reaction to these examples. He then systematises his data by introducing a number of conceptual distinctions, which serve to describe the data in a more adequate manner than has previously been described using other concepts (most notably, the concept of 'identification' with characters). Smith defines imagination as the film spectator's inferential activity, in which we simulate either the beliefs and emotions of characters, or beliefs about the fictional characters and events. The first activity defines central imagining (in which we imagine experiencing the fictional events from within a character's perspective) and the second acentral imagining (in which we simply imagine that something occurs in the fiction, outside a character's perspective). In terms of emotion, central imagining offers an imagined, self-directed emotion, and acentral imagining an imagined, other-directed emotion (426). Smith argues that both types of imagining/emotion are important in engaging with film characters, in which central imagining is framed by, or assimilated into, acentral imagining.


The above will be familiar to readers of _Engaging Characters_. One major problem in the book that Smith now addresses is that he treated textual structures and psychological processes as identical. In this chapter he concedes that 'certain kinds of textual structure may foster or predispose us to imagine in one way rather than another, as distinct from determining the nature of our imaginative response' (416). In other words, he now posits a contingent relation between textual structures and psychological processes. He also clarifies the relation between central imagining and the concepts of simulation, mimicry, and autonomic responses (416-17).


In terms of POV shots, Smith privileges them because they 'promote central imagining as a part of a larger structure of multifaceted alignment' (417) (where 'multifaceted alignment' refers to a spectator's multiple access to a character -- not only by seeing what they see, but by seeing how they react). Gregory Currie develops a similar distinction between central and acentral imagining -- namely, the distinction between personal and impersonal imagining respectively, [11] except that he downplays the significance of personal imagining and textual devices such as POV shots (as we saw in Currie's paper at the beginning of _Film Theory and Philosophy_). Smith contests Currie's privileging of impersonal over personal imagining.


(c) Constructing problems. Firstly, Smith relies on phenomenological concepts, to the extent that his data partly consists of the experience of watching films. He also relies on cinematic concepts to the extent that he focuses on spectators' experiences of general cinematic structures such as POV shots, and how those general structures give spectators access to the experience and emotions of fictional characters.


(d) Evaluating problems. Character identification is a major issue in the study of narrative film, making Smith's critical analysis and replacement of the concept of 'identification' a significant problem. Smith's distinctions in _Engaging Characters_ are logical and internally consistent, organised around clear distinctions such as 'empathy/sympathy', central/acentral imagining', and 'recognition/alignment/allegiance', which means that his theory is well-formed. In this chapter he clarifies his concepts and defends them in light of Currie's work.


In 'Seeing Theory: On Perception and Emotional Response in Current Film Theory', Malcolm Turvey presents a 'fundamentalist' Wittgensteinian critique of Carroll's and Smith's theories of visual perception. The chapter is in many ways an extension of Richard Allen's chapter on looking at motion pictures. The main problem with Carroll's and Smith's theories, at least according to Turvey, is that they postulate an unnecessary entity -- thought or imagination -- between the spectator and the film, because of the theory of perception they adopt. The result is that the spectator's physical perception of and emotional reaction to film images is downplayed, and instead this mediating entity takes precedence. In other words, for Carroll and Smith spectators respond to an abstract mental entity, not to the film itself. Following Wittgenstein, Turvey critiques their psychological/mentalist theories of seeing. According to Wittgenstein, no theory of the imagination or mental interpretation is required to explain standard visual experience, because we directly see the objects. Similarly, Turvey argues that film theorists do not need to posit the existence of the imagination or mental interpretation to explain standard visual experience of images, because we directly see the objects in the images:


'When confronted by an unambiguous image of a lion or any other object, the beholder does not behave as if he is subjectively interpreting the material properties of the image that he objectively perceives . . . Consequently, the aspect which we see in an image cannot be a mental entity, an indirect, subjective interpretation supplied by the mind of the beholder following the direct perception of the material properties of the image. It is not something we *think*. Rather, it is something that beholders *see* directly and instantaneously *in* images' (447-8).


Turvey concludes that his critique does not entail an outright rejection of Carroll's and Smith's theories, merely a modification of them:


'Rather than 'entertaining' the 'imagination' or 'thought' produced in the spectator's mind by the concrete cinematic representation and then responding emotionally to it, the spectator can directly 'entertain' and respond emotionally to the concrete cinematic representation of the fictional referent. He has the capacity to do so because he *regards* the concrete cinematic representation *as* the fictional referent it represents' (456).



In Conclusion


Film theory is no longer under the sway of Continental theory, for it unapologetically adheres to concepts such as likeness, intentionality agency, expression, cognitivism, the fiction/non-fiction distinction, film content, and emotional response, while debunking concepts such as illusion, deception, the 'death of the author', Brechtian alienation effect, subject positioning, identification, and the unconscious. It also asks more fundamental questions such as: What in itself is the act of seeing? (before considering the ideological or patriarchal meanings of looking). How do light and sound differ on the physical and phenomenological levels? Why do talking films require musical accompaniment? On a number of occasions the contributors to _Film Theory and Philosophy_ also fundamentally challenge the very act of theorising 'objects' such as 'looking' and 'representation', and they are not afraid to spell out the limitations of their research or to revise, reformulate, or reject their previous positions. In terms of constructing problems, a number of chapters focus on metatheoretical concepts, for the book is laying out the issues and problems for an analytic philosophy of film, which raises the activity of theorising to a level of explicit reflection.


Finally, as with this review, I feel that a number of chapters are too long (for example, Carroll's section entitled 'some objections', where he responds to several possible objections to his chapter, is excessive at seven pages; and Turvey's critique of Carroll and Smith, plus his outline of Wittgenstein's account of visual experience, is at times repetitive, making the chapter's logical progression very slow). And I get the impression that the editors encouraged a number of contributors to expand their chapters, which has made some of them unwieldy (Peter Kivy's chapter comes immediately to mind). Despite these minor quibbles, readers can nonetheless be confident that, by reading this book, they will encounter a coherent and internally consistent agenda, with well-formed and significant issues and problems to address, plus logical reasoning strategies, of a new phase of film theory.


Liverpool John Moores University, England





1. Rudolf Botha, _The Conduct of Linguistic Inquiry: A Systematic Introduction to the Methodology of Generative Grammar_ (The Hague: Mouton Publishers, 1981).


2. Larry Laudan, _Progress and its Problems: Towards a Theory of Scientific Growth_ (Berkeley: University of California Press, 1977).


3. Warren Buckland, 'Film Semiotics', in Toby Miller and Robert Stam, eds, _A Companion to Film Theory_ (Oxford: Blackwell, 1999), pp. 84-104.


4. Francesco Casetti, _Inside the Gaze: The Fiction Film and its Spectator_ (Bloomington: Indiana University Press, 1998). My review will appear in a forthcoming issue of _Semiotica_.


5. Botha, _The Conduct of Linguistic Inquiry_, p. 54.


6. Ibid., p. 66.


7. Ibid., p. 85.


8. David Bordwell, Janet Staiger, and Kristin Thompson, _The Classical Hollywood Cinema: Film Style and Mode of Production to 1960_ (London: Routledge, 1985).


9. Antonio R. Damasio, _Descartes' Error: Emotion, Reason, and the Human Brain_ (New York: G. P. Putnam's Sons, 1994).


10. Murray Smith, _Engaging Characters: Fiction, Emotion, and the Cinema_ (Oxford: Clarendon Press, 1995).


11. Gregory Currie, _Image and Mind: Film, Philosophy, and Cognitive Science_ (New York: Cambridge University Press, 1995).



Copyright © _Film-Philosophy_ 2001


Warren Buckland, 'Problem Formation in the Analytic Philosophy of Film', _Film-Philosophy_, vol. 5 no. 26, August 2001 <>.



Save as Plain Text Document...Print...Read...Recycle


Join the Film-Philosophy salon,

and receive the journal articles via email as they are published. here


Film-Philosophy (ISSN 1466-4615)

PO Box 26161, London SW8 4WD, England



Back to the Film-Philosophy homepage