(ISSN 1466-4615)

Vol. 4 No. 19, August 2000



C. Paul Sellors

The Nature of Film Spectators




Francesco Casetti

_Inside the Gaze: The Fiction Film and its Spectator_

Translated by Nell Andrew with Charles O'Brien

Bloomington and Indianapolis: Indiana University Press, 1998

ISBN: 0-253-21232-4 (pb); 0-253-33443-8 (hb)

xviii + 174 pp.


With _Inside the Gaze_ [1] Francesco Casetti develops enunciation theory of

earlier film semiotics in an effort to explain satisfactorily the

complicated web of cinematic spectation. Rather than possessing a textual

code for understand films, he argues, spectators must bring a *competence*

to their engagements with these texts. [2] In principle, this is certainly

a reasonable modification to enunciation theory. It first identifies the

act of spectation as a process, rather than a product, and second, locates

an intersection between the filmic text and an empirical, unique, flesh and

blood spectator. In its broadest understanding, it is therefore amenable to

both film theory and history. Although he locates a position for actual

spectators in his model, he brackets them from his present argument,

focusing instead on understanding the real spectator's textually

constructed avatar, the being that appears during spectation, 'by the way

it [the film text] says *you*' (15).


Casetti constructs his challenging and complex argument with detail and

skill, weaving together numerous intricate narratological components of

filmic texts in an effort to explain how they designate and place

spectators within the narration and then set them out on their journey

through the narrative (14). Unfortunately, Casetti's remarkable ability to

animate the formidable body of these narratological concepts results more

in a display of rhetorical agility than analytical rigour. His laudable

dissection of the intersection of gazes and narrational entities in film

leaves many problems faced by enunciation theory in general untouched.


Casetti begins with three important, and I believe correct,

presuppositions. First, textual studies of films and reception studies of

audiences are complimentary endeavours and must be compatible with each

other. Second, the interaction between a text and spectator implies that

cinema is in some way a communicative medium. Third, since films are

communicative, then the content communicated must originate from an

intending entity of some sort. 'In so many reception studies', he laments,

'the text simply disappears or becomes a completely indifferent object,

whereas I insist that the text is crucial, if only because its presence is

what turns a *social* situation into a *communicative* situation' (xvii).

What he does not make clear is why enunciation theory is the correct tool

to defend and develop these presuppositions. Indeed, it is precisely the

nature of the enunciator and enunciatee, especially when saddled with the

notion of *the gaze*, that causes a number of inconsistencies in Casetti's



A year before Casetti published the original version of _Inside the Gaze_

in Italian in 1986, David Bordwell provided some concise arguments against

enunciation theory in his _Narration and the Fiction Film_. [3] Bordwell

contended that even if we can accept enunciation theory for literary

fiction, it does not necessary follow that it can be adapted to explain

narration in filmic fiction. One of the fundamental problems, he insisted,

is that the theory's linguistic concepts do not easily map onto a

moving-image medium. Most notably, cinematic enunciation theory has not

been able to coherently translate from literature to film one of literary

enunciation theory's most basic components: the *enonce*, the utterance

itself. In literature, the enonce is a basic analytical unit, and can be

constituted by a word or many sentences. However, cinematic enunciation

theorists have not made clear what a filmic utterance *is*. Indeed, given

the complexity possible of a shot, for instance, it is difficult to see

what this minimum, basic unit might be for film. Bordwell concluded that

'because a film lacks equivalents for the most basic aspects of verbal

activity, I suggest that we abandon the enunciation account' (26). Despite

this suggestion, Bordwell has not given sufficient reason to abandon

enunciation theory. Instead, he has identified a substantial difficulty

that must be resolved if cinematic enunciation theory is to proceed.

Casetti, however, does not offer any insight to how such problems can be

solved. [4]


There are other difficulties. Equally troublesome is the notion of

spectator positioning that Casetti develops. He founds his analysis of the

*spectator in the text* on the principle that films generate a textual

*you* as the logical consequence of the enunciating *I*. Considered most

broadly, the basic assumption here is reasonable enough. If the act of

communicating does not assume an addressee, then the act is trivial. What

is not so straightforward about this assumption is that this addressee

should be conceived of as an abstract, singular, *ideal* individual: the

enunciatee. Within his theory, Casetti portrays this entity as the textual

instance of a spectator that real flesh and blood spectators are supposed

to develop some sort of relationship with. There are some quite substantive

difficulties with this structure. Most notably, it seems at odds with the

common experience of watching films. The attention of normal spectators, I

would think, is directed to the fictional contents of the film directly,

not to a process of negotiating some sort of correspondence with a textual

doppleganger. Casetti could be suggesting, and at times he appears to, that

when we see a film, we see it through the enunciatee. 'At this point, the

problem of the spectator becomes more particular, defined now as a problem

of a mark, within the film's interior, that indicates an invitation to see

and hear' (30). But in order to facilitate direct seeing through this

invitation, the spectator would have to entertain some notion of being

identical with the enunciatee. Casetti does not explain how such an

identity can be achieved. He may be addressing this problems when he

mentions *the game*, of which spectating seems to be a part. Unfortunately,

he neither clearly explains what the game is precisely, nor justifies why

fiction film should be cast as a game in the first place. Casetti needs to

state a great deal more about the relationship between the enunciatee in

the text and the real spectator in the theatre to diffuse such

difficulties. All he suggests is that the real spectator 'recognizes'



'by successfully placing himself within the scope of a reception

continually subject to images and sounds. The spectator, in sum, becomes

engaged in the act of gazing, responding to the availability of the

screen's world by assuming certain responsibilities according to the

demands of a true vocation' (9).


Rather than offering a clear analysis of spectator positioning in fiction

film, Casetti's explanation raises a rather troubling ontological problem:

spectators must somehow recognise themselves within a textual, non-existent

and abstract entity.


Equally suspect issues appear in the structure of his enunciative model as

a whole, especially once *the gaze* is superimposed onto it. The

enunciator, according to Casetti, possessing a unique visual perspective on

the fictional world of a film, enunciates an enonce to the enunciatee, with

whom we, as spectators, develop some kind of relationship with. But because

Casetti insists that actual spectators *see* fictional contents through a

network of gazes, his enunciator and enunciatee risk collapsing onto one

another. It is the enunciator who is responsible for directing the complex

structure of gazes through which the enunciatee and subsequently real

spectators have visual access. Because the enunciatee is the logical

implication of enunciation, then all the gazes that the enunciator offers

in the process of enunciating will be equally the visual content of the

enunciatee. In short, the ideal receiver of the enunciator's discourse is

identical with the discourse itself, since this receiver is wholly

determined by it. The enunciatee, being abstract and not sentient, brings

nothing to the enonce, and nothing, therefore, to a theory that seeks to

explain how spectators engage and understand fiction films. The enunciatee

is a trivial component of the model. As a result of this structure, it

appears that the content of the enunciatee's gaze is directly correspondent

with, indeed the same as, that of the enunciator's. Functionally,

therefore, it seems that the enunciator and enunciatee are identical, since

no two entities, abstract or not, can share the same temporal and spatial

co-ordinates (and therefore the same content of a gaze) and not be

identical. If this is the case, then, from the perspective of visual access

to fiction, the spectator also risks disappearing due to the enunciator's

gravity. In short, the theory appears to implode.


Casetti has not explained why a theory of cinematic spectation needs *an

ideal* receiver in the first place. His objective is to outline a model

that enables reception theory without losing sight of the text. In other

words, he seeks to understand how filmic texts impact real spectators.

Immediately, however, he substitutes this notion of impact or influence

with the metaphor 'position'. He then begins to treat this metaphor

literally. It is possible, however, to retain a notion of influence and the

structure of the communicative model on which enunciation theory is based

without constructing a virtual spectator.


Imagine, for the moment, that I am presenting this review at a conference.

It would certainly be odd to think that an ideal conference participant was

part of the communicative chain, especially for the receivers of the paper.

Competent conference participants can attend to the contents of the

presentation directly, without having to negotiate some form of

correspondence with an ideal entity. The situation would not change if I

could not attend the conference and instead sent a videotape of my

presentation. Casetti would certainly be ready for these counter-examples,

and point out that in both cases I am discussing communication, not

enunciation. Although these models are related, they differ in that the

former has real participants, the latter abstract entities. Simply shifting

from a real presenter to an abstract enunciator, however, does not explain

why we need to sacrifice general spectatorship for an ideal spectator as

the direct recipient of enunciation. The problem with enunciation theory,

at least as Casetti presents it, is it does not respect its kinship with

communication theory.


Now imagine a fiction film where this review is presented in a single long

take from the visual perspective of lecture hall seating. The real

spectators of this likely dull fiction can attend to the content of the

presentation, contemplate its visual point of view, and question the

reasons for its inclusion in the fictional film *directly*, without having

to understand how the filmic structure creates an ideal, non-identical

surrogate. It is certainly reasonable to assume that a communicative or

enunciative *I* implies a *you*, a receiver, as Casetti suggests. It may

just turn out, though, that this *you*, in both cases, is one and the same,

the real flesh and blood humans who receive the communication. Locating

real rather than ideal spectators need not risk losing the film text in the

process of spectation, as Casetti fears, just as the arguments of this

review need not succumb entirely to the agendas of its readers.


There is one further difficulty with Casetti's notions of seeing and the

gaze that I would like to address. One of the core problems with his

argument, and indeed enunciation theory in general, is it proceeds without

first understanding the nature of the content of an enunciation in fiction

film. This lack of attention to the nature of fiction generates severe

shortcomings. Casetti is clear that the enunciator rests outside of the

fictional world. This seems a necessary move on his part, because it allows

him to explain how the gaze can be directed freely, independent of

fictional events. It also permits him to draw a correspondence between

perception and belief, at least in certain circumstances like the

'impossible objective view'. This phrase indicates the type of difficulties

that Casetti's analysis manifests. Such a view, he argues, can only belong

to the enunciator, and presents an ideal view of characters resident in the

fictional world from outside of that world. By being outside, however, the

enunciator surrenders the possibility of any physical, including,

therefore, visual interaction between worlds. This gaze is impossible if it

belongs to the enunciator, not because it cannot belong to any participant

in the fiction, but because it is impossible that it is a view at all, at

least in a strictly visual sense. Once this is recognised, the notion of

the gaze, and seeing fictional contents in general, becomes suspect.

Although such a suggestion may seem blasphemous to some film theoreticians,

it cannot be denied that when we see a fiction, say _Citizen Kane_ for

instance, we only ever actually see Welles, his cast, and wonderfully

elaborate sets and props. We can no sooner *see* Xanadu than we can visit

it. [5] The problem is that there exists no Kane for either us or the

enunciator to hold in our gaze.


There are ways out of this difficulty, such as Kendall Walton's analysis of

make-believe, Gregory Currie's interest in the imagination, and, more

generally, cognitivists' focus on the mental processes of recognition,

among others. It is certainly not the current task to critique these

approaches here. Rather, I wish to indicate that they address spectatorial

access to films by taking into account how real spectators approach

engaging fictional objects and events represented by actual world

components. Undertaking such a methodology would permit Casetti an

analytical, rather than rhetorical description of the relationships between

spectators and narratological entities. [6] As it stands, however, he

promotes a thesis arguing, ostensibly, that spectatorial engagement is

positioned by the gaze enunciated to the enunciatee by the enunciator,

without any explanation of how we can have gazes of things that do not



Finally, I would like to add a note of complaint and one of praise about

the book's layout. I found the omission of a bibliography very frustrating.

Granted, detailed endnotes are included, but complete references to

citations are often difficult to locate because repeated references are

abridged. Such an omission is not only inconvenient for the book's reader,

these days it is lazy and inexcusable. In contrast, however, the glossary

near the end of the book is most helpful and clear.


Considered within the historical period of its initial publication in the

mid-1980s, _Inside the Gaze_ offers some valuable insights into and

modifications of cinematic enunciation theory. However, this theory as a

whole has not aged well over the last fifteen years. The problems it faces

cannot be accounted for simply under shifting trends in film scholarship.

Rather, the basic concepts on which the theory is based have proved

remarkably difficult to develop and defend. Bordwell was certainly well

justified to dismiss advancing enunciation theory further, even if he did

not offer decisive arguments against it. To be fair to Casetti, the

presuppositions that he launches his text from are as valid now as when he

first asserted them. The problem is that cinematic enunciation theory no

longer seems sufficient to undertake the tasks that he is asking of it. It

is a pity that this book took so long to appear in English after its

Italian publication. This was a timely book whose time, unfortunately, has

now past.


New York University, USA





1. Originally published as _Dentro lo squardo: il film e il suo spettatore_

(Milan: Bompiani, 1986); English edition prepared from the French

translation by Jean Chateauvert and Martine Joly: _D'un regard l'altre: le

film et son spectateur_ (Lyon: Presses Universitaires de Lyon, 1990). This

English edition also has an Introduction by Christian Metz, and a Preface

written by Dudley Andrew.


2. He defines 'competence' as: 'The knowledge required to relate the story;

a modality of narrational authority. Those responsible for telling or

hearing a story are endowed with the duty, desire, ability, and know-how to

carry out their mandate' (136).


3. See especially pp. 21-26.


4. Casetti defines the enonce rather loosely, describing it as: 'The

enunciated; any result of the enunciation; the utterance itself, which may

refer to the film, sequence, shot, etc.' (136)


5. In a footnote on _Citizen Kane_ Casetti makes one of the most obscure,

tangential suggestions I have ever read in a film theory text. Discussing

the relationship between the internal diegesis and external conditions of

production in _Citizen Kane_, he argues that 'for us, the letter 'K',

beyond being the hero's initial, can be understood as a coded allusion to

Welles. Both the letter 'K' and 'W' are constituted by means of a

conjunction of two acute angles; a simple rotation would turn one into the

other' (152, n. 48). I have attempted this geometrical surgery

unsuccessfully, only to discover that I was hampered by one letter being

constituted by three lines, the other by four.


6. It also would allow Casetti to avoid strange conceptions such as 'seeing

oneself see' (14).





Bordwell, David, _Narration in the Fiction Film_ (Madison, Wisconsin: The

University of Wisconsin Press, 1985).


Currie, Gregory, 'Visual Fictions', _The Philosophical Quarterly_, vol. 41

no. 163, April 1991, pp. 129-143.


Walton, Kendall L., 'Fearing Fictions', _The Journal of Philosophy_, vol.

75 no. 1, January 1987, pp. 5-27.




Copyright © Film-Philosophy 2000


C. Paul Sellors, 'The Nature of Film Spectators',

_Film-Philosophy_, vol. 4 no. 19, August 2000





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