Our concern in this chapter is with a set of issues central to the ontology of film. “Ontology” traditionally refers to the branch of philosophy that deals with questions about the being and nature of things. We can safely assume that films exist or even less controversially that there are some films. So, our first question in the ontology of film need not be “are there films?” but rather “what is film?” An even clearer statement of our primary question may be “what sort of thing is film?”
Armed with an answer to that question, we could wade more confidently into secondary issues regarding films’ constitutive parts or the conditions under which they persist through time. But alas, while we have no shortage of available theories about the nature of film, there is nothing like consensus around any one of them. This has not been an entirely unfavorable condition, however. A good number of critical insights and useful observations about film have emerged out of philosophers’ debates, arguments and disagreements over film ontology.
Another sort of disagreement, though, has been less beneficial. These are disagreements over the question itself, which largely emerge from two sources. First, we don’t enjoy antecedent agreement on the class of things that are films (the extension of “film”). And we can’t expect to agree on what makes things films if we don’t first agree on which things are films. Second, we don’t yet agree on what we’re asking. That is to say, we don’t agree on criteria for an adequate answer.
So, instead of offering or defending a particular theory of film ontology in this chapter, I’d like to discuss the proper extension of “film” and propose a set of criteria for selecting an ontological theory of it. I’ll then consider some current and likely candidate theories using those criteria. By way of conclusion, I’ll offer some suggestions for ways the discussion might move forward.
Determining the Extension of “Film”
We should notice right away that there is no natural or privileged use of “film” or set of films independent of some particular theoretical or conversational interest. A frustrated aficionado of celluloid and acetate film might be interested in the nature of the set of objects that ranges from Louis Le Prince’s experiments to those few hold-out 35-mm theaters scattered around the world. But most of us do not think that the history of film ended with the swift triumph of digital projection in this decade. The films that dominate the cultural landscape are still being made, distributed, seen, discussed and written about, even if there is no longer any film involved.
The “film” in this sense relies on Le Prince’s and Edison’s technologies, but doesn’t emerge until others use that technology to say and tell. And it is not entirely clear when that activity starts. Are the Lumiere brothers’ early projected films the start of this history or does it wait on the first narrative film (which may also have been theirs)? We can’t just be interested in narrative films per se, lest we exclude whole swaths of genuine films more likely to be seen at the gallery than the multiplex.
But then again, it is not merely the technology of moving images in which we’re interested. Whatever films are, they don’t include security camera footage or Skype conversations. We should think of ourselves as primarily investigating an art form. This is not because we should assume from the outset that there is an ontological difference between last night’s security footage and Rear Window , but because only the art film (understanding that term as widely as is reasonably possible) provides us with a special investigative interest.1 The cultural, aesthetic and historical interests we take in Hitchcock’s masterpiece2 motivate questions about its nature. And because we take similar interests in other films, we should not assume that Rear Window is ontologically sui generis.
Just as it is the art form status of films that arouses our interest in its ontology, the ontology of film has figured prominently in the establishment of film as an art. As moving pictures matured from curiosity to art form, the live-action narrative film naturally invited comparisons to two close cousins. Focusing on the technology and means of display, the film is kin to photography. It is in fact photography itself, multiplied, put in motion and coupled (eventually) with sound. Focusing instead on their content, films are in the family of theater and continue the ancient history of drama and comedy.
This situation presented something of a dilemma for the early advocate of films’ art status. On the one hand, the art status of photography was itself in question. On the other, if films are artworks in the theatrical tradition, then the artwork in question may just be the actions filmed rather than the film itself. Thus, the price for resting the possibility of films’ art status on similarity with another art form is that it makes room for a skeptic to claim that film is no distinct form, but rather a mode of presentation for the form to which it’s compared.
This central problem is still with us, even if the art status of (some) photography is more of a given today. If film can be an art form, it must be a kind of thing that is (a) capable of being art and (b) different in some significant and relevant way from existing art forms. Given their proximity, the pressing challenge for (b) is to distinguish film from photography and theater, though we want to be able to distinguish it from painting, poetry and music as well.3
We have good reason, then, to want an ontology of film as an art form. And, in fact, it is difficult to see what other compelling reason we could have for wanting an ontology of film. I don’t think I’m going out on a limb by thinking that the ontology of art is just more interesting than the ontology of security or communication technology.
Even so, there does not necessarily need to be an ontological difference between the film and the security footage in order for there to be a difference in their art status, unless there is an ontological difference between art and nonart (but that would be another matter). There is, after all, at least a prima facie distinction to be made between ontological and art-status questions. In the present context, the former ask “what makes this thing a film and not something else?” While the latter ask, “what makes this film a work of art and not something that is not a work of art?”
There are two reasons, though, to confine the extension of “film” in question to filmic works of art and exclude security camera footage, home movies, Skype conversations and the like. The first follows from the kind of interests we just discussed. It is captured in what David Davies calls the ‘pragmatic constraint’ on the ontology of art:
This “methodological principle guiding philosophical inquiry,” Davies notices, is at least implicitly at work in many and otherwise various ontologies of art.5 Sherri Irvin later makes explicit use of the same criterion, (rightfully) preferring to call it the “critical practice constraint.”6Artworks must be entities that can bear the sorts of properties rightly ascribed to what are termed “works” in our reflective critical and appreciative practice; that are individuated in the way such “works” are or would be individuated, and that have the modal properties that are reasonably ascribed to “works,” in that practice.4
Applying the critical practice constraint to film suggests an ontological difference between (for instance) security camera footage and live-action narrative films. It is inappropriate to criticize security footage for its aesthetic or otherwise formal faults. Regardless of how films are properly individuated, the question of individuation does not arise for security footage. Footage has duration, not individuality. And the same film could have different endings. If some duration of security footage has a different ending (and “ending” is not even the right word here), it is a different bit of footage.
The second reason becomes apparent when we consider the difference between actual security camera footage and security-camera-footage-like shots in a film. Certain shots in The Wire (2002–2008) or The Bourne Identity (2002), for instance, let you know that a character is being filmed by providing the perspective of the security camera filming. The entire movie Look (2007) is shot from such perspectives. But in these instances, the shots themselves establish the existence of a fictional and unseen security camera. Actual security cameras require the existence of an actual unseen security camera. These are not merely ways of seeing the film and the footage. Something is fundamentally different about the two. One camera is fictional and the other actual. Surely, the difference between the fictional and the actual is ontological.
This is not to say, however, that the ontological difference between security camera footage (and the like) and film as we mean to discover it must track the fiction/non-fiction divide. There are, after all, plenty of non-fiction films, both documentaries (in the traditional sense of “documentary,” we’ll consider another in a little bit) and nonnarrative (what are often and sometimes pejoratively called “experimental”) films. We could imagine, in fact, actual security footage being repurposed as such a film. But then it would have been transformed (or “transfigured” in Arthur Danto’s parlance).7 And that transformation is a transformation in the kind of thing that it is. It will go from a mere recording to something with a point, a meaning. It will go from having only duration to an individual identity. Its modal properties will change. It will become subject to aesthetic evaluation. It will, in short, become a new kind of thing.
So, only those moving images8 that might be reasonably thought of as artworks are films in the sense under consideration. But there is still some work to be done to say which artworks are members of that class. The first couple of questions involve the technology of displaying films—even (as we discussed above) when the technology that is standard is changing over time. First, is such technology necessary? For instance, if films involve moving images, then so do flip books, and we might be reluctant to admit flip books into the relevant class.9 But perhaps we shouldn’t be too hasty. Bill Brand’s “Masstransiscope” is a series of paintings that appear as an animation to riders of the Q train after they leave Dekalb Avenue on their way to Manhattan from Brooklyn. The moving-image effect is delivered by stationary pictures viewed from a moving train rather than moving pictures viewed from a stationary position. The same visual effect could have been carried out in the normal way, with those particular pictures in motion. In fact, it is theoretically possible to set up giant slides of each frame of The Godfather (1972) separated by black pillars at just the right distance and have a train pass them all at just the right speed while piping in synchronized audio across the loudspeakers. The overall effect wo...