About CA
Contact CA

Search Journal

Editorial Board

Permission to Reprint


Site Map


Ariadne at the Movies

  John Dilworth

Films are usually assumed to be types, with their templates or performances being tokens of those types. However, I give a counter-example in which two different films are simultaneously made by different directors, with the outcome of this process being a single template length of film which, I claim, embodies both of those films. But no two types could thus have a token in common, and hence type views of films must be incorrect. I further explain and defend the legitimacy of the example, and conclude by offering an alternative, representational view of the nature of films that can also more adequately explain the interpretive nature of experience of a film.

Key Words
Film, ontology of film, types and tokens, Noël Carroll, Gregory Currie, representation, interpretation

Several categories of artworks, such as films, novels, plays or musical works, are such that works in those categories seem not to be particulars, but instead to have multiple copies, instances or performances, so that ontologically speaking such artworks are naturally regarded as being types, which have such copies, instances or performances as their tokens.[1] In addition, there are those, such as Gregory Currie, who argue that even apparently particular artworks such as individual paintings are really tokens of some relevant painting type.[2]

However, in some recent papers I have argued that such type-theoretic approaches to the arts are seriously flawed in various ways.[3] In this article I shall extend those criticisms to apply more directly to films[4] by presenting a counter-example to views such as that of Noël Carroll, who has proposed a 'type' account of the ontology of film according to which films are types whose tokens include both 'templates' such as a film print or videotape, and individual film performances or showings generated from such token templates.[5]

A type view such as that of Carroll might seem to be a natural, or even unavoidable, basic position on the ontology of film. However, I shall attempt to show that that it is vulnerable to the following kind of counter-example. This vulnerability arises because of a logical feature of types, namely that there could not be a single token that was simultaneously a token of two distinct types of the same general kind. For example, if dogs and cows are two distinct types of animal, then there could not be a single animal that was both a dog and a cow. And similarly, if films are types, then there could not be a single film token, such as an individual template film print, that was simultaneously a token of two distinct films.

But my counter-example will involve just such a case, in which two distinct films are associated with a single physical template, which therefore could not be a token of both distinct films considered as types. Here is the counter-example, which as far as I know has not previously been envisaged or investigated by others.[6]

1. Example

The example concerns the making of two films. Two directors Leslie and Steve are independently approached by a scriptwriter Carla, who pitches a different script idea to each. Leslie's film, to be called Greed, would be about a cold-hearted financier whose only interest is in money, who ruins the lives of all around her in her efforts to accumulate more wealth; on the other hand, Steve's film, to be called Sacrifice, would involve a story about a woman driven to great financial success by a desperate need to love and help those about her.

Both Leslie and Steve are sold on the different script-ideas presented to them by Carla, but there is a potentially crippling practical difficulty: neither can afford to make his or her film, since the cost of hiring Carla to write it, plus of hiring actors, studio technicians, etc. is for both Steve and Leslie almost twice the amount that each could raise from backers or other sources to pay for their respective films.

Well, desperate people will resort to desperate remedies. One of the three of them, it doesn't matter who, comes up with the following unconventional solution to their dilemma. To start with, Leslie and Steve will each hire Carla to write their respective films for them, but to save costs Carla will write the two scripts with a large amount of overlap between them. Specifically, all of the characters, situations and dialog will be identical in each - they will differ only in stage directions and other materials explaining to the actors each director's different point of view on the characters etc. Thus Carla has to very carefully write the two scripts in such a way that each of the characters, incidents and dialog in them is susceptable to both of the different interpretations she originally pitched to Leslie and Steve.

Is this possible? Yes it is, because human actions are subject to different interpretations. What Leslie sees as selfish, acquisitive behavior in the leading character (who of course could only be named 'Jana', for two-faced reasons) could be viewed by Steve as selfless behavior done with its only object being to benefit others. Of course, not anything that Carla writes is susceptable to both interpretations, and as a result of this Steve and Leslie are often at odds with each other. Hence they suffer through many awkward and even acrimonious sessions with Carla in an attempt to arrive at a single line of dialog in the two scripts which each can accept as embodying their differing overall conceptions of their respective films.

Once the scripts are finished, the next stage of their unconventional agreement must be embarked on. To save costs Steve and Leslie have also agreed that they will use the same actors, studio technical facilities and shooting schedules for each of their films, with Carla serving as casting director and liason with the studio for each of the films.

By now Leslie and Steve are hardly talking to each other, given their divergent artistic conceptions and the great difficulties under which they are working, so the burden falls on Carla to see that everything is brought about in a manner which is satisfactory to each of them. As casting director for both films she has to audition actors and explain to them the uniquely confining working conditions, under which (we recall) each has to satisfy not just one but two separate film directors, each of whom has divergent conceptions of what kind of human character each line of dialog in the two scripts is to reveal and develop.

After screening out those actors who will not agree to work under these conditions, or who fail the initial auditions or screentests, Carla then presents each surviving candidate actor separately to Steve and Leslie for their respective approval. Naturally, each will approve an actor only if convinced he or she gives an effective portrayal of the character in each of their respective films. Hence many candidates are turned down by one or both of Leslie and Steve, with the result that they are rejected altogether. Only when both Steve and Leslie are independently satisfied with an actor's portrayal of a role in each's respective film is he or she hired by each director to play each of their respective roles.

Once casting for each film is completed, the (if possible) even more awkward combined shooting sessions are embarked upon. By this stage Carla has to really crack the whip over Leslie and Steve, to force them to come to agreement on the specifics of the shooting schedule for each film, the order of the scenes, technical and interpretive matters of lighting arrangements and camera angles, and so on.

But even with these prior arrangements finally ironed out, the actual shooting sessions often turn out to be abortive, since of course each director has veto power over any session which does not individually satisfy each of them as properly carrying out their individual artistic intentions for that shooting session as part of each's respective film. Hence many retakes of scenes are often necessary before both directors are satisfied (both on the film set itself and after viewing the filmed outtakes from these sessions) with their adequacy for her or his own artistic purposes.

The final artistic battles between Steve and Leslie take place in the cutting room. Originally each director had intended to edit his or her own film from the surviving film stock produced from those shooting sessions which were jointly acceptable. However, after initial trials both Leslie and Steve are quickly convinced that they lack many of the necessary but rare skills required to convincingly edit each of their films. They realize that each needs to hire a professional film editor with a proven artistic track record to complete each of the two films.

However, here as before, financial constraints force them to economize and hire a single film editor, Jack, to complete both films. But what is worse, because of his great reputation, Jack's time is extremely expensive and, so as before, both directors are forced to use the very same extremely confining joint-approval techniques in making use of Jack's services. They are forced to individually work with Jack in a way similar to that in which each originally worked with Carla in producing the script: each individual segment or scene edited by Jack for each of them must be approved by each of Steve and Leslie for their own separate artistic purposes before it is allowed to remain in either's final edited film as spliced together for each by Jack.

A final point: since neither director wants his or her film to be associated with the name of the other director (since, above all, they wish to avoid the accusation that they have merely succeeded in producing a jointly directed single film), they agree that neither of their names will appear on the screen in the film credits. And for the same reason they agree that the titles for their respective films will also be omitted from the on-screen credits. For naturally Leslie does not want the audience to be informed onscreen that they are watching Steve's film Sacrifice, for she intends that they should instead be watching her film Greed; while, as one might expect, Steve has similarly strong views about potential onscreen confusion for the audience with respect to his own film.[7]

The outcome of these processes is a single spliced length of film which embodies[8] two distinct, separate films (Greed and Sacrifice). Thus ends the example.

To conclude this section, I claim that if this example really does show what it seems to, then we already have a conclusive refutation of the view that films are types. Because if Jack's final, single spliced length of film is indeed a single putative token both of Steve's film Greed and of Leslie's distinct film Sacrifice, then those distinct films cannot be types, for the purely logical reasons already given.

But this conclusion might be resisted in two ways. First, the genuineness of the example might be questioned, and second, even if the legitimacy of the example is conceded, it might be argued that it is a mere logical oddity that has no implications as to the status of films in general.

A necessary ingredient in an adequate refutation of any such potential objections is a reasonably clear basic account of the identity conditions for films, which I shall now provide.

2. The identity of a film

The identity criteria for a film, such as Leslie's film Greed, involve two main kinds of factors, namely factors relevant to the identification of the particular finished film template L itself, and to the identification of the relevant actions and successfully realized intentions of the director[9] (Leslie) insofar as they help to explain how or why that finished template has the features that it actually has.

To see that both kinds of factors are relevant and in what ways, note first that it is conceivable that there might be a numerically distinct but qualitatively identical film template, L'', that had been produced by some completely different causal process having nothing to do with Leslie's intentions and actions in directing her film. In such a case, that template L'' would not count as embodying Leslie's film (and nor Steve's either), because it would not satisfy a 'causal particularity' criterion that is satisfied by Leslie's template L, namely one that causally links the length of film in question to Leslie's actual film-making activities, which thereby caused that particular length of film L to come into existence.[10] Thus, being the causal outcome of Leslie's directorial activities is a necessary condition for the template L to embody Leslie's film Greed.

At the same time, if we give due consideration to some relevant possibilities or contingencies associated with film-making, it would be too constraining to require that only template L could have served as the template for Leslie's film. For example, since L consists of various spliced lengths of film derived from various shooting sessions, it seems clear enough that, if in fact the whole film had been shot from a single batch of film B from a given manufacturer M, then if the studio had instead received a qualitatively identical but numerically distinct batch B' of film from M, then Leslie's film template would not have been L but instead a numerically distinct (though presumably qualitatively identical) template L', whose spliced sections came from batch B' rather than batch B of film.

Thus Leslie's film-making activities could have produced a numerically distinct template L', but since L' would have still been the causal outcome of Leslie's activities, it would have still counted as embodying Leslie's film Greed, unlike the first-mentioned template L'', which had no causal connection whatsoever with Leslie's film-making.

In addition, it would be metaphysically too demanding to require qualitative identity for any template that could have been the causal outcome of Leslie's activities. For just as the particular batch B of film stock might have been different, so also various other conditions or causal factors in the film production might have been different. For example, in one scene a slightly larger sofa might have been used for characters to sit on, so that the relevant spliced segment of film from that shooting session would have had minor qualitative differences from the corresponding actual segment. Or a scene might have been shot at a different time of day from the actual time, so that minor qualitative differences in the lighting of the scene could have occurred.

Nevertheless, there are limits to the amount of possible variation in the resulting template or templates. Only those possible templates which would both be the direct causal outcome of Leslie's film-making actions and intentions, and which would embody her relevant intentions, would count as legitimate templates of the film Greed. And of course, exactly similar points could be made for Steve's film Sacrifice. This concludes my basic account of the identity of a film.

3. The counter-example is genuine, and general

In brief, I claim that the counter-example is genuine because the single final film template produced does, according to the identity criteria for films just given, embody both Leslie's film Greed and Steve's film Sacrifice. The example carefully described the (often difficult and acrimonious) process through which both Leslie and Steve did manage to each carry out his or her distinct film-making intentions, and how each did so by producing a single final film template that was the direct causal outcome of the distinct film-making intentions of each of them.

I shall consider a more specific objection to the genuineness of the counter-example later, but for the present I shall use the discussion of identity criteria for films both to further demonstrate the genuineness of the example and also to show its generality.

The 'generality' objection to be considered is that even if the legitimacy of the example is conceded, it might be argued that it is a mere logical oddity, that has no implications as to the status of films in general. Presumably the objector's point is that there is something logically unique or deviant about either or both of Steve and Leslie's films, so that their non-typehood, even if it is conceded, shows nothing about ordinary, non-simultaneously produced films.

In reply, the logical flexibility built into the identity criteria for films as discussed above can be used to show the normality of each film, as follows: It was pointed out that there might have been some differences in the causal chain that produced each film, without this thereby destroying the identity of either of them. But most importantly for our purposes, the causal chain which produced Leslie's film might not have involved Steve's film-making at all, nor any of the (admittedly somewhat mind-boggling) complications resulting from his concurrent attempts to make his film while Leslie was making hers. It seems undeniable that Leslie could have made substantially the same script, casting, scene-shooting and final cutting decisions that she did, which contributed to her successful realization of her film-making intentions, even if Steve and any activities of his had played no causal part whatever in her film-making.

Indeed, not only could Leslie have done this (in some abstract empirical sense, such as that her doing so would not have been incompatible with any known physical laws), but we have good evidence that it is a realistic artistic possibility that she would have done this, assuming that her general artistic intentions were the same in each case. For at each and every stage in the film-making process Leslie freely made the artistic decisions that she did, even though she could have made different ones if she had wanted to. As a competent, serious artist, Leslie would not have allowed her overall artistic purposes to be deflected or compromised by any of the difficult practical hurdles or stumbling-blocks which are part of the environment in which every artist must work--including those unusual difficulties which arose because of Steve's film-making activities. Hence there is no reason to think she would have proceeded diffferently, even if Steve's activities had been causally absent from her film-making.

To be sure, in the absence of Steve's activities it seems likely that parts of the causal history of Leslie's film-making would have been different. For example, presumably the film would have been made more quickly, in the absence of any acrimonious debates with Steve, and so various specific differences both in the numerical and qualitative identity of spliced segments of the final film template could be expected to result from the more condensed time-frame. But as I have argued in my account of the identity criteria for films, those criteria are flexible enough to allow for such counterfactual numerical and qualitative variations.

The upshot of this discussion is that Leslie's film is indeed both a normal film and a film independent from that of Steve because it is a realistic possibility that Leslie could have, and would have, made the same film, Greed, with the same overall artistic intentions, even if Steve and his activities had played no causal part in her film-making.

And of course a similar line of argument to the above, as applied to Steve's film Sacrifice, would also establish its normality, and artistic independence or distinctness, from Leslie's film Greed. Hence I conclude that the counter-example is both genuine, and generalizable to apply to any films.

4. Qualitative differences

However, so far no mention has been made of what, intuitively speaking, is the strongest argument for the distinctness of the two films, Greed and Sacrifice, namely that the intentions of the two directors, as realized in their films, are qualitatively different in significant respects.[11] For recall that Leslie's film, Greed, is about a cold-hearted financier whose only interest is in money, while on the other hand, Steve's film, Sacrifice, involves a story about a woman driven to great financial success by a desperate need to love and help those about her. Furthermore, one of these films could be much better or more artistically successful than the other, which would serve to emphasize their artistic distinctness. Indeed, it would be surprising if some divergence in success did not occur, given the unique artistic limitations under which each director was working.

Also, it is not inconceivable that the lead actor in each film might, in the case of one of the films, win an Oscar for her brilliant, convincing portrayal of the main character's single-minded egotism, whereas in the other film she might generally be condemmed for her shallow, unconvincing portrayal of a woman desperately trying to help those around her. Indeed, arguably any greater success in her portrayal in one film is bound to correlate with some lesser success in her different portrayal in the other film, given the divergent interpretations given to the two characters by each director.

Thus, the distinctness of the two films is a simple matter of logic: the lead actor's portrayals could not both be convincing and unconvincing unless they were portrayals of distinct characters in films that are themselves distinct because of (among other things) the differing underlying intentions of the two directors with respect to those two characters.[12]

In addition, there is another significant, though perhaps more technical, argument for why the two films must be qualitatively distinct. It is generally agreed that the title of an artwork can play an integral part in characterizing the aesthetic qualities of any artwork--and so much so, that even two otherwise identical artworks must be distinct if they have been assigned different titles by different artists having distinct artistic intentions, as Jerrold Levinson has convincingly argued in the case of musical works.[13] But in the present case, the two films in question--Leslie's film Greed and Steve's film Sacrifice--do indeed have distinct titles assigned by different artists with distinct intentions,[14] and hence must be aesthetically distinct works.

Indeed, arguably this point alone is sufficient to refute any attempts to explain away the current two-film case as really a case of a single, jointly directed film. Thus my further argument below could be regarded as being not strictly necessary to further secure their distinctness, but instead merely as showing how it is possible for the directors to individually realize their own directorial intentions in thus producing their own distinct films.

5. The 'frustration of intentions' argument

Here is a specific genuineness objection, which could be called the 'frustration of intentions' argument, which proceeds as follows. At every stage in the production of the two films, the artistic intentions of each film director are being continually frustrated by the activities of the other. Since each has veto power over the other's artistic decisions, neither can be said to be acting in a genuinely artistically independent manner.

This is so (the argument proceeds) because in the given situation neither can unconditionally realize his or her intentions, but instead can only realize them conditional upon the approval or acceptance of the other. Hence their supposedly independent artistic intentions are a sham, because in practice their intentions are not independent at all. The two directors can (according to this line of argument) succeed only in jointly producing a single film which is the outcome of the lowest common denominator of the intentions of each, namely that set of intentions which each can minimally agree to, but which set contains no distinctive intentions from either (upon which the directors would disagree) of a kind such as would be needed to establish the directorial independence of either of them for the supposed films in question.

In reply to this skeptical argument, note first that a main part of the force it may seem to have comes from assuming the very point at issue concerning the example. This point is that it is possible to have a final single length of film which in some way embodies or results from two distinct sets of directorial intentions which serve to distinguish two separate films. However, the sceptical argument seems to assume that any evidence that the directors have agreed upon the inclusion of a given length of film in their joint projects (whether it is the final print, or the result of some intermediate film-making stage) is ipso facto evidence that they have must have agreed on a common set of directorial intentions with respect to that length of film. But the whole point of the example is to demonstrate that this need not be so.

As a specific aspect of why this need not be so, note that we must distinguish between a specific intention, and the choice of a specific cinematic way in which to express that intention (which could be called the 'mode of expression' of the intention.) Though it is true that each director has the power to veto the other's choice of a mode of expression for an intention, it does not follow that thereby he or she has vetoed the intention itself, for there may be at least one other mode of expression of that intention which is not in fact vetoed by the other director during the making of the films (who, recall, is concurrently pursuing his/her own modes of expression of her intentions.)

Second, note that even if a given intention of a director is completely frustrated (in that no amount of reshootings etc. result in the intention being expressed in any possible mode of its expression,) it does not follow that all of the the intentions of a given director will be thus frustrated. It is true that in practice each director may be able (purposely or inadvertently) to completely shut out any expression of a given intention by the other, but there may still be a coherent set of film-making intentions for each director, each of which sets is distinctively different in various ways, yet the members of which are not in fact shut out by the activities of the other director. But this situation is all that is needed to underwrite the directorial independence of the film-makers and the legitimacy and genuineness of the example itself.

Another way to express this present point is as follows. If a given initial set of directorial intentions by director A is taken as providing necessary conditions for the identity of director A's film X, then it is true that director B might prevent A from making film X by her own activities in making her film M. But A might still succeed in making a film Y during the same process whose intention-set resembles that for film X in enough significant ways for us to judge A's film-making efforts to been artistically successful on substantially the same film project during the joint film-making projects of A and B. And a similar point of course applies to B's film-making efforts on her own film during the same period. Thus I conclude that the current objection has been disarmed.

6. A non-type, representational explanation of films

In this section I shall briefly describe an alternative account of the nature of films, since without a plausible alternative a type-theoretic view of films is likely to persist, no matter how convincing any attempted refutation, such as the present one, may seem to be.

To begin, if a single length of film can embody or be the vehicle for two or more distinct films, then clearly it cannot be identical with either of them, since the two films are not identical with each other, and there could be no good reason to identify the strip of film with one of them in preference to the other. Hence films cannot be concrete particulars such as strips of film, or other physical entities or processes.

But what then is a film, if it is not a type? In several recent papers[15] I have argued for an approach that provides a broadly logical rather than an ontological answer to such questions about the nature of artworks.[16] According to this approach, the things that are usually considered to be artworks (or to be tokens or instances thereof, as in type-theoretic approaches), such as paintings, individual performances of plays or music, copies of novels etc., are not themselves artworks, but instead they are representations of artworks. Thus on this view the film template that 'embodies' both Leslie's and Steve's films in fact represents each of those films.

As to a film itself, on this view a film is whatever is represented by representations of the film in question. (Such representations would include both templates and individual film performances or showings). This view is deliberately and appropriately uninformative about the substantive nature and ontological status of a film, since in general the ontological status of various representations of any thing X tells us nothing whatever about the ontological status of X itself. Nevertheless, once that is understood, it is perhaps intuitively appropriate, and harmless, to say that on my view a film is the representational content shared or represented by all representations of the film in question.

Moreover, for present purposes this representational approach does have one all-important virtue (in addition to solving or avoiding problems that undermine a type theory), namely that it can easily explain how a single length of film (the single physical outcome of Leslie's and Steve's film-making) can both 'embody' or represent Leslie's film Greed and Steve's film Sacrifice. For in general it is unproblematic that a single representational object may represent two quite different things, such as in the case of a 'duck-rabbit' picture, which represents, or can be seen as, either a duck or a rabbit.[17]

And what is more, the availability and general feasibility of this representational view of artworks provides further evidence that a case such as the current 'double film' case is indeed legitimate (rather than logically deviant), and in general that such cases are to be expected in various art forms if the representational approach to the nature of art is correct.[18]

7. Interpretations of films

So far this article has been narrowly focused on logical and ontological issues about film. But it would not be complete without some brief discussion of broader issues concerning our actual experience and appreciation of films. In particular, what aesthetic difference should it make to the viewing of films in general--which are not ambiguous in the way discussed here--if the current representational view, rather than the type view, is correct?

This is a large topic, so one central issue only will be discussed here, concerning various kinds of interpretation of films. My argument will be that a type-based view of artworks would make genuine interpretation of films impossible, but that there are no corresponding problems for a representational view.

As a preliminary, it will be helpful to distill the conflict between the type and representation views down to its simplest elements. Consider again an ambiguous drawing, such as a duck-rabbit drawing that can be seen either as a duck, or as a rabbit.[19]

Now no such drawing could actually be an instance or token both of a duck and a rabbit, because ducks and rabbits are different types of animal, and hence can't share a common instance or token. But of course a single drawing can represent both a duck and a rabbit, even though it cannot be both a duck and a rabbit. My more extended example of two films being associated with a single length of film is an elaboration of this basic difference between the type and representation views: the length of film cannot be an instance of two different films, but it can represent two different films.

To continue, the issue of interpretation arises as follows. It is generally agreed that the concept of interpretation of something as an X can only be applied if the object is not actually or literally an X; for example, one cannot interpret or see a duck as a duck, simply because it literally is a duck. Thus in the case of a duck-rabbit drawing, it is only possible to interpret the drawing as a duck (or as a rabbit) because the drawing is not itself an instance or token of that animal.[20]

But similarly, I would argue, it is only possible to interpret what one views on a screen as a particular film if the viewed object or state is not itself an instance or token of that film. The problem is that the type-token or instantiation relation for a type X results in a physical object or state that concretely is an X, and hence which cannot be interpreted as an X. Thus the type view leaves no room for interpretation--whether aesthetic, or of other kinds--in that it reduces the aesthetic enjoyment of a unique filmic experience to the prosaic facticity of what kind of object or state it is that one is viewing on the screen. Or, otherwise put, it commits a Whiteheadean fallacy of misplaced concreteness[21] in that it confuses what are in fact affective, perceptual, conceptual and representational matters--the core of interpretive issues--with concrete physical states that are too determinate to be interpreted.

This interpretive criticism of the type view can be sharpened as follows. There are at least two distinctive kinds of interpretation involved in the viewing of a film--identifying versus critical interpretation--neither of which can be explained by a type view.

The first kind--identificatory or identifying interpretation--is logically prior to ordinary critical interpretation. For example, if one identifies a duck-rabbit drawing as being a drawing of a duck, one has identifyingly interpreted the drawing, subsequently to which one can proceed to critically interpret the aesthetic qualities of the thus-identified duck. Or in the main case of the ambiguous length of film, one could identifyingly interpret what one sees on the screen either as Leslie's film Greed or Steve's film Sacrifice--and having done so, one could then go on to critically or aesthetically interpret the relevant film as thus identified.

As for more usual films or pictures with which only one artwork is normally associated, these cases arguably require a prior identifying interpretation as well, in that there are various ways in which a picture or film can be viewed, such as merely still or moving shapes on a surface, or as a deadpan documentary, that would not involve an identifying interpretation of what is seen as specifically being a given visual artwork having aesthetic qualities. Thus an identifying interpretation is a necessary preliminary to any critical interpretation of a film as such.[22]

However, on a type view, no such identifying interpretation is possible, in that on its view, what one is looking at simply is an instance of the film, leaving no room for any such further interpretation of what is seen. While at the same time, no critical interpretation is possible either, because on the type view the relevant aesthetic properties of the film are concretely instantiated on the viewed screen for all to see, so that the type view has no theoretically principled way in which to provide room for legitimate differing individual critical interpretations of a film that go beyond mere correct versus incorrect perceptions of it.

Thus, in sum, given the centrality of various concepts of interpretation in discussions of artworks, the inability of a type theory of film to accommodate such concepts is a serious theoretical failing, quite apart from its logical flaws as discussed earlier. It is therefore fortunate that an alternative representational theory is available which is immune to these particular problems, in that representational aspects of objects or events are always open both to the possibility of diversity of items represented, and to identifying interpretations of them that pick out such items--along with subsequent critical interpretations that assess their aesthetic worth.[23]


[1] Support for a 'type' view as applied to the performing arts is provided by (among others) Noel Carroll, A Philosophy of Mass Art (Oxford: Clarendon Press; Oxford University Press, 1998), Gregory Currie, An Ontology of Art (New York: St. Martin's Press, 1989), Joseph Margolis, Art and Philosophy (Brighton, Sussex: Harvester, 1980), and Richard Wollheim, Art and Its Objects : With Six Supplementary Essays 2d ed. (Cambridge; New York: Cambridge University Press, 1980).

[2] Currie, ibid.

[3] Such as in my article "Artworks Versus Designs," The British Journal of Aesthetics, 41, 2 (April 2001), pp. 162-177, where I show (among other things) that two distinct sculptures could be associated with the same artifact, so that on a type view of sculptures (per impossibile) a single artifact would have to be a token of two distinct types of the same general kind. In the present paper I use a similar strategy as applied to films.

Some further criticisms of 'type' views in various categories of art are given in my article, "A Representational Theory of Artefacts and Artworks," The British Journal of Aesthetics, 41, 4 (October 2001), pp. 353-370. And in my "Theater, Representation, Types and Interpretation," American Philosophical Quarterly, 39, 2 (April 2002), pp. 197-209, some other problems for a type view as applied to a performing art such as theater are discussed.

[4] I offer some brief criticisms of type views as applied to films in "A Representational Theory of Artefacts and Artworks," but the present argument will be more specific, and arguably more conclusive as well.

[5] See his article "The Ontology of Mass Art," The Journal of Aesthetics and Art Criticism, 55 (1997), pp.187-199, the core of which is also reprinted in his book, A Philosophy of Mass Art, ibid.

[6] That is, beyond the loosely related scenario of Richard Strauss's opera 'Ariadne Auf Naxos', as invoked in the title of this paper, which involves the simultaneous presentation of two incongruous operatic works.

[7] The distinctness of the film titles is the basis of an important argument for the non-identity of the corresponding films: see Section 4.

[8] I use the term 'embodies' here merely as a neutral stand-in for whatever relation actually holds between the length of film, and each of the films in question. (In Section 6 I claim that the relation is one of representation).

[9] Of course, there are deeper philosophical issues concerning exactly how a director's intentional actions relate to, or are involved in, the causal chain of events in question, but for present purposes perhaps the current account is precise enough.

[10] To be sure, I would argue that Steve's film-making activities related to his own film equally caused the length of film to come into existence; but this is unproblematic, since such cases of 'causal over-determination' are common, and explainable in ways that make no appeal to the special logical and aesthetic features of the current 'double film' example.

[11] For the preceding arguments for the distinctness of the films (other than the 'different titles' argument) would apply even if the directors had had substantially the same film intentions, since arguably the two films would be aesthetically different simply because of each director's own identity, distinctive background and style of film-making, even if their overt intentions were the same in a given case. Jerrold Levinson argues for a similar view of musical works in J. Levinson, Music, Art, and Metaphysics: Essays in Philosophical Aesthetics (Ithaca, N.Y.: Cornell University Press, 1990), Ch. 4.

[12] I assume here that at least some such critical judgments are capable of being true or false, and of being inconsistent with other critical judgments, so that inconsistent pairs cannot be true together of the same film. See R. A. Stecker, "Incompatible Interpretations," Journal of Aesthetics and Art Criticism, 50, 4 (1992), pp. 291-298, for arguments in support of this view.

[13] Levinson, ibid., Ch. 8, 'Titles'.

[14] Note that it would not be convincing to argue that really both films are untitled, merely because their titles do not appear onscreen (as noted in the description of the case). For it is artistic intentions that determine whether, or what, title an artwork has, rather than any further contingent issues about how the artist decides to communicate the relevant title to her audience.

[15] Including "A Representational Theory of Artefacts and Artworks," "Theater, Representation, Types and Interpretation," and "The Fictionality of Plays," The Journal of Aesthetics and Art Criticism, 60, 4 (October 2002), pp. 263-273.

[16] A type theory of films cannot avoid being an ontological theory, in that the ontological status of the tokens of a given film presumably should settle any questions as to the ontological status of the film itself, when it is considered as a type.

[17] On which see E. H. Gombrich, Art and Illusion; a Study in the Psychology of Pictorial Representation, 2d ed. (London: Phaidon Press, 1962), pp. 4-5.

[18] See my "Artworks Versus Designs," as cited in fn. 3, for a sculptural example. For a comprehensive discussion of the representational approach to the arts in general see my forthcoming book,The Double Content of Art.

[19] Gombrich, ibid. Wittgenstein also discusses such cases, e.g. in his Philosophical Investigations, 3rd ed., trans. G.E.M. Anscombe (London: Basil Blackwell, 1958).

[20] E.g. see Wittgenstein, ibid., on how one cannot see a fork as a fork.

[21] A.N. Whitehead, Process and Reality (New York: Harper, 1929), p. 11.

[22] Compare Danto on the differences between a 'mere real thing' and an artwork, e.g. in his book The Transfiguration of the Commonplace (Cambridge, Mass.: Harvard, 1981).

[23] My thanks to the Editor and two anonymous referees for this journal for very helpful comments, as well as to my commentator Melissa Zinkin and the very responsive audience in an ASA Eastern Division session in Philadelphia, April 2003 on a shorter version of this article.

John Dilworth

Western Michigan University