Documentaries, Docudramas, and Perceptual Beliefs: Terrone Documentaries, Docudramas, and Perceptual Beliefs

ABSTRACT

The main accounts of the documentary in contemporary analytic aesthetics have difficulties in dealing with the distinction between documentaries and docudramas. On the one hand, the assertion‐based accounts proposed by Noël Carroll, Trevor Ponech, and Carl Plantinga cannot properly differentiate documentaries from docudramas. On the other hand, Gregory Currie's account can do so by relying on the notion of trace, but this involves an undesirable side effect, namely, the exclusion, from the documentary, of those documentaries that do not include traces of their subjects, as, for instance, documentaries that resort to reenactment. Is there a way to exclude docudramas from the documentary without also excluding other films that we usually treat as documentaries? This article affirmatively answers by relying on the notion of perceptual belief, that is, a belief that one can form by endorsing the demonstrative content of one's perception. While documentaries primarily involve the formation of perceptual beliefs, docudramas invite us to form nonperceptual beliefs by primarily involving the formation of perceptual imaginings. That is why docudramas are not documentaries. By contrast, a documentary that resorts to reenactment primarily involves the formation of perceptual beliefs, namely, perceptual beliefs about the reenactment itself.

Some twenty years ago, Gregory Currie and Noël Carroll engaged in an insightful discussion on the nature of documentaries. Currie (1999, 2000, 2001) proposed an account based on the notion of trace while Carroll (1997, 2000) centered his account on the notion of assertion. As pointed out by Plantinga, both these accounts help us to better understand what a documentary is but “fail as traditional definitions of the documentary” (2005, 105). Plantinga thus proposes an account that aims to combine the explanatory power of Currie and Carroll, thereby solving their respective problems. Still, I contend, there remains a crucial issue to be addressed, namely, differentiating the documentary from the docudrama, which Currie characterizes as “the re‐creation, by dramatic means, of certain actually occurring events” (1999, 295). In this article, I propose a new account of the documentary that is based on perceptual beliefs instead of on assertions or traces. I argue that this account can distinguish the documentary from the docudrama more effectively than its predecessors.

In order to properly appreciate a film about certain actually occurring events, one should know whether this film is a documentary or a docudrama. One should be aware, for instance, that films such as All the President's Men (1976), The Doors (1991), and No (2012) are docudramas, whereas films such as High School (1968), In the Land of the Deaf (1992), and Capturing the Friedmans (2003) are documentaries.¹ That is because in the relevant appreciative practice documentaries and docudramas are governed by different norms of appreciation that are inherent in these categories.² Thus, an account of the documentary that aims to comply with the relevant appreciative practice should take the distinction between documentaries and docudramas into account. This is what I do in this article.

After presenting the analytic debate on documentaries in Section I, and after introducing the notion of relevant appreciative practice in Section II, I argue, in Section III, that both Carroll and Plantinga's accounts fail in complying with the relevant appreciative practice as regards the distinction between the documentary and the docudrama. Currie, instead, succeeds in differentiating the documentary from the docudrama by means of the notion of trace. Yet, this notion leads him to deny the status of documentary not only to docudramas but also to certain films that are usually appreciated as documentaries. This makes his account also at odds with the relevant appreciative practice. Is there a way to differentiate documentaries from docudramas without resorting to the notion of trace? I propose to address this issue by relying on the notion of perceptual belief, which I introduce in Section IV and exploit in Sections V–VI in order to properly distinguish the documentary from the docudrama. In Sections VII–X, I develop this account and defend it from possible objections. In Section XI, I draw my conclusions.

I. TRACES AND ASSERTIONS

Currie characterizes a documentary as a film about a subject that predominantly exploits photographic traces of that subject. “About,” here, designates an intentional relationship: the filmmaker intended to foreground something, namely the subject (often indicated by the title), about which he or she communicates information and possibly opinions. “Of,” instead, designates a causal relationship: the photographic apparatus recorded things on whose appearance photographic traces counterfactually depend; if those things had not been so, these traces would not have been so. For instance, The Armstrong Lie (2013) is a documentary because it is about cyclist Lance Armstrong and exploits photographic traces of Armstrong, whereas The Program (2015) is not a documentary but rather a docudrama because it is about Armstrong but exploits photographic traces of Ben Foster, the actor who plays Armstrong in this film.³

Still, some films that we usually treat as documentaries are about their subjects and yet do not exploit photographic traces of those subjects. This is what usually happens when a documentary is about an “out‐of‐reach” subject, that is, a subject that cannot be filmed, as, for instance, the life of Napoleon, the extinction of dinosaurs, or the origin of the universe. Furthermore, Currie's account can hardly deal with those documentaries—like the classic ones by Robert Flaherty, John Grierson, and Humphrey Jennings—that heavily exploit reenactment, that is, the practice of resorting to filmed reconstructions of past events. In Choi's terms, “Currie rejects documentaries with performances or reenactments as proper documentaries, since they are not literal traces of people and events that a film is concerned with” (2001, 318). In fact, a reenacted scene is about a certain event but is not a trace of that event; rather, it is a trace of the reenactment itself.⁴

On the other hand, Carroll can accommodate out‐of‐reach subjects and reenactment by conceiving of a documentary as a film that makes assertions, that is, uses images and sounds to articulate a propositional content to whom the audience is meant to respond by forming beliefs having that content.⁵ Thus, an alleged documentary about an out‐of‐reach subject actually is a documentary because it makes assertions about that subject even though it does not exhibit traces of that subject. For instance, a documentary can make assertions about dinosaurs even though it does not exhibit traces of dinosaurs. Likewise, Carroll's account can accommodate reenactment because a reenacted scene makes assertions about a certain event even though it does not exhibit traces of that event, but only of its reenactment.

Still, according to Plantinga, Carroll's account finds it hard to accommodate what Nichols (2001) calls the “observational documentary mode.” While documentaries in the “expository mode” explicitly make assertions, often by means of a voice‐over narrator, documentaries in the “observational mode” limit themselves to showing us events. For instance, Pennebaker's Dont Look Back (1967) is a documentary about Bob Dylan that shows his 1965 concert tour in England rather than making assertions about him.⁶ A metaphor that is often used to characterize the “observational mode” is the fly on the wall, which suggests that the filmmaker limits himself or herself to observing the events filmed as if he or she was a fly on a wall in the place where these events occur. In fact, flies do not make assertions and so filmmakers who behave like flies do not make assertions either.

In order to solve this problem, Plantinga proposes to conceive of documentaries as films that can make not only canonical assertions about their subjects but also peculiar metarepresentational assertions about themselves. Although a documentary in the observational mode does not explicitly make assertions about its subject, it implicitly asserts something about itself, namely, “that the use of motion pictures and recorded sounds offer an audiovisual array that communicates some phenomenological aspect of the subject, from which the spectator might reasonably be expected to form a sense of that phenomenological aspect and/or form true beliefs about that subject” (Plantinga 2005, 111). Going back to our metaphor, the fly on the wall does not make assertions and yet the filmmaker can assert that he or she is behaving like a fly on the wall.

II. THE RELEVANT APPRECIATIVE PRACTICE

Since the beginning of the article, I have claimed that the relevant appreciative practice draws a distinction between documentaries and docudrama, and I am going to rely on this distinction in order to criticize Carroll and Plantinga's accounts of the documentary. Still, before going into this, the notion of relevant appreciative practice requires some clarification.

According to a view that is gaining more and more consensus in contemporary aesthetics, art appreciation is a sort of social game governed by norms that usually remain implicit in a network of shared attitudes, such as beliefs, intentions, and expectations.⁷ An important task of the philosophy of art consists of trying to make such norms as explicit as possible. In this sense, the philosophy of art contributes to an investigation on culture (the realm of norms) that has interesting analogies with the investigation on nature (the realm of facts) carried out by science and scientifically minded metaphysics. Just as the latter disciplines aim to “carve nature at the joints” (Sider 2011, 3), the philosophy of art aims to carve culture at the joints.

Specifically, I conceive of works of art as essentially social and normative entities that can be traced back to the category of “public artifacts,” which Thomasson characterizes in these terms: “While all artifacts are indeed mind dependent, public artifacts do not depend merely on the individual intentions of their makers; they also depend on public norms” (2014, 47). Following Houkes and Vermaas, I conceive of the public norm on which a public artifact depends as the “use plan” that prescribes how to use it (2010, 7–8). While use plans of technical artifacts are quite often made explicit by user manuals, those of works of art tend to remain implicit in networks of shared attitudes. Assuming that the primary use of works of art consists in their appreciation, I call “appreciative practices” the network of attitudes that establish their use plans. In particular, I call “relevant appreciative practice” the network of attitudes that establishes the use plans that are relevant to a specific form of art.⁸

Since documentaries and docudramas basically are films, the appreciative practice that is relevant to them is the network of attitudes that establishes the use plan of films. Outstanding works in the philosophy of film, such as Currie's (1995), Carroll's (2008), and Gaut's (2010), can be seen as attempts to make that use plan explicit. Still, in addition to the “global” norms that govern the use of a work as a film, there are “local” norms that govern its use as a specific kind of film. My working hypothesis is that the documentary and the docudrama involve different local norms. In the next section, I rely on this hypothesis in order to criticize the main philosophical accounts of the documentary for failing to comply with this normative difference. Then, I propose a new account that can not only accommodate the difference between the local norms of the documentary and those of the docudrama but also figure out such norms, thereby turning the working hypothesis into a full‐fledged thesis.

III. THE DIVIDE BETWEEN THE DOCUMENTARY AND THE DOCUDRAMA

Currie's account sharply differentiates the documentary from the docudrama because docudramas do not predominantly exploit traces of their subjects. On the other hand, Carroll's assertion‐based account tends to treat docudramas as documentaries since the former films also invite the audience to take propositions about their subjects as asserted. Given that complying with the relevant appreciative practice is among the desiderata of an account of the documentary, and assuming that this practice differentiates documentaries from docudramas, Currie's account should be preferred to Carroll's in this respect.

For the same reason, Currie's account should be preferred to Plantinga's account, which succeeds in reconciling Carroll's assertion‐based account with the “observational mode” but cannot properly differentiate the documentary from the docudrama. Plantinga himself acknowledges this when he states that, in wondering whether a certain film is a documentary, “we might accept actors playing historical figures if we were convinced that quality research had figured into the historical accuracy of what the actors wore, said, and did” (2005, 114).⁹

In principle, Plantinga has a way to differentiate the docudrama from the documentary. He should state that documentaries must make both (1) canonical assertions about their subjects and (2) metarepresentational assertions about their communicating some phenomenological aspect of their subject by means of “phenomenological approximations of the look, sound, and some other sense or feel of the pro‐filmic event” (2005, 115). Yet, Plantinga refuses to do so since such an account would have the same problems as Currie's with respect to documentaries about out‐of‐reach subjects. The latter films, in fact, do not satisfy (2) because the subject cannot be the pro‐filmic event, which is out of reach. Thus, Plantinga prefers to formulate his account as a (nonexclusive) disjunction between (1) and (2). Yet, in this way, docudramas can count as documentaries in virtue of satisfying (1), just as documentaries about out‐of‐reach subjects do.

Docudramas are problematic also for the account proposed by Ponech who characterizes documentaries as constituted by “cinematic assertions,” which have “the principal goal of letting spectators know that it is appropriate to take the attitude of belief toward the greater part of that which is either explicitly shown or implied by the depiction” (1997, 216, my emphasis). In fact, docudramas also may share this “principal goal,” unless one unpacks the expression “the greater part” in a way that allows one to differentiate docudramas from documentaries, but Ponech does not do that.

At the end of the day, Currie's trace‐based account seems to be the one that can effectively distinguish between docudramas and documentaries. Yet, this explanatory benefit involves an unsustainable cost, namely, the impossibility of accommodating documentaries involving reenactments or out‐of‐reach subjects. One might wonder whether there is a way to exclude docudramas from the domain of the documentary without excluding also some genuine documentaries. In what follows, I affirmatively answer to this question by relying on the notion of perceptual belief.

IV. PERCEPTUAL BELIEFS

One can form a perceptual belief by endorsing the demonstrative content of one's perception. If, for instance, one perceives a dog jumping, one can form the perceptual belief that this dog (having the sensory features supplied by perception) is jumping in this way (having the sensory features supplied by perception). In Brewer's terms, when a person forms a perceptual belief:

His experiences contribute essentially to his grasp of certain perceptual demonstrative contents. These contents refer to particular mind‐independent things in the world around him, of which they predicate determinate mind‐independent properties. In doing so, they give him a reason to endorse those very contents in belief. Simply in virtue of grasping the content that that thing is thus, he has a reason to believe that that thing is indeed thus. (1999, 204–205)

One can conceive of a perceptual belief as essentially involving predicates such as “like this” or “in this way” that Heal (1997) calls “indexical predicates.” While nonperceptual beliefs are just caused by perception (I form the belief that a dog is jumping because I perceived a dog), perceptual beliefs are constituted by perception (I form the perceptual beliefs that this dog is jumping in this way by deploying a demonstrative content provided by perception itself).

From this perspective, a documentary can be characterized as a film that enables its audience to form perceptual beliefs concerning the events depicted. The spectator of a documentary that depicts a dog jumping can form not only the perceptual belief that this film is screened in this way, but also the perceptual belief that this dog is jumping in this way. I call the latter a pictorial belief, that is, a perceptual belief that one can form by endorsing what one perceives in a picture. We usually form pictorial beliefs by relying on photographic traces but the two notions are distinct; the former is a cognitive notion concerning the endorsement of picture perception, whereas the latter is an ontological notion concerning the causal relation between pictures and what they depict.¹⁰

Although pictorial beliefs are a kind of perceptual beliefs, there is a crucial difference between ordinary perceptual beliefs (that is, those derived from face‐to‐face perception) and pictorial beliefs. If one perceives a dog jumping in front of him or her, one can form the perceptual belief that this dog is jumping here and now. Yet, one cannot form this sort of belief by endorsing a pictorial perception. The spatial location “here” and the temporal location “now” cannot be included in the content of a pictorial belief. One can only believe that this dog is jumping somewhere and sometime in the actual world. At most, that belief can be enriched by extraperceptual pieces of information about space and time (for instance, information that this dog jumped in this way on Christmas Day 2017 at noon in Berlin). Ordinary perceptual beliefs, instead, directly provide spatial and temporal information.

A documentary is a film whose use plan primarily involves the formation of pictorial beliefs and their possible enrichment with extraperceptual information. By “primarily,” I mean that, in the use plan of the documentary, the formation of pictorial beliefs has priority over that of pictorial imaginings, whereas in fiction films it is the other way around. At this point, the notion of pictorial imagining requires clarification.¹¹

Pictorial imaginings are similar to pictorial beliefs since they also represent things as having their place in a spatiotemporal framework and as possessing the features revealed by the picture. However, in the case of pictorial imaginings, the spatiotemporal framework is not that in which the picture, its maker, and its viewers have their place. The key difference thus lies in the nature of the spatiotemporal framework, which for pictorial beliefs is the actual one, whereas for pictorial imaginings is an alleged one, which one might call a fictional world.¹²

Suppose that you have a time machine at your disposal. A pictorial belief warrants that you can travel thereby perceiving with your own eyes the events depicted from the same standpoint from which you were perceiving them on the screen. By contrast, a pictorial imagining cannot warrant this because the standpoint from which you perceive the events depicted is not in the spatiotemporal framework where you have a place but rather in an alleged spatiotemporal framework severed from yours.¹³

Pictorial beliefs, as well as pictorial imaginings, may be indeterminate not only with respect to space and time but also with respect to other features. For instance, the pictorial beliefs elicited by a black and white documentary such as Dont Look Back are indeterminate with respect to color. One can properly form pictorial beliefs only if one has a way of filtering out the picture's features that are not to be ascribed to the scene depicted. In the case of a black‐and‐white film, one can easily do so, thereby forming perceptual beliefs that concern shape and light intensity but not color. In fact, something similar may occur also in the case of ordinary perception, for instance, when one wears sunglasses of a certain kind.

This notion of filtering out allows us to address an objection that one might derive from Plantinga's criticism of Currie's account. Plantinga argues that documentaries involve a “mediation” that makes them essentially different from traces: “Documentary films are also edited, and editing almost invariably further interprets the event and involves intentionality in a way that indexical signs such as traces do not. When one adds music or titles or voice‐over narration, additional mediation between documentary and subject is added” (2005, 107). Does such “mediation” also prevent the spectator of a documentary from forming pictorial beliefs? No, it does not, because the spectator can easily filter out features such as editing or music or titles or voice‐over narration, thereby endorsing only the relevant elements of his or her pictorial perception. By contrast, the contribution of actors to the depiction of real individuals in docudramas cannot be filtered out in the same way.¹⁴ That is why docudramas do not primarily involve the formation of pictorial beliefs and therefore are not documentaries.

V. DOCUMENTARIES AS GENERATORS OF PICTORIAL BELIEFS

Currie can effectively differentiate documentaries from docudramas since the latter do not predominantly exploit traces of their subjects. Yet, in order to accommodate out‐of‐reach subjects and reenactments, we should acknowledge that traces, in a documentary, are a means to the end of forming pictorial beliefs, which are not forced to concern the subject of the documentary. A film that primarily involves the formation of pictorial beliefs is a documentary even though the subject of the documentary is not included in the content of those beliefs.

Too Much, Too Young: Children of the Middle Ages (2011), for example, is a documentary that includes traces of the speeches on an expert of the Middle Ages but surely is about childhood in the Middle Ages, not about that expert or his speeches. From Currie's perspective, this is not a genuine documentary since its traces are not of its out‐of‐reach subject. By contrast, in my account, this remains a documentary since it still primarily involves the formation of pictorial beliefs, even though these beliefs are about the expert instead of about childhood in the Middle Ages.¹⁵

The case of reenactment is more articulated. A documentarist who reenacts an event instead of recording it should choose whether to (1) present the reenactment as such or (2) try to deceive the audience by presenting the reenactment as if it was the original event. However, the documentary would primarily involve the formation of pictorial beliefs in both cases, in spite of the fact that it is not using traces of the event it is about. Specifically, option (1) would involve the formation of veridical pictorial beliefs concerning the reenactment itself, whereas option (2) would involve the formation of deceptive pictorial beliefs about the original event.

Suppose that you are watching a documentary in which the sinking of the Bismarck battleship is represented by film of sailor's cap being dropped into a water tank. There are two basic ways of regarding the cap in the tank.

If we comply with option (1), we recognize the tank for what it actually is, that is, an object having its place in our spatiotemporal framework. We thus acknowledge that the filmmaker exploits the tank as a reenactment aimed to represent a historical event, which in turn has its place in our spatiotemporal framework. This leads us to form nonperceptual beliefs about the original event by forming veridical pictorial beliefs concerning the reenactment.¹⁶

If, instead, we comply with option (2), we are somehow deceived by the tricky reenactment devised by the filmmaker, thereby treating the picture of the tank as if it was an actual recording of something happening during the sinking of the Bismarck; we thus form deceptive pictorial beliefs concerning the original event.¹⁷

Ultimately, both option (1) and option (2) primarily involve the formation of pictorial beliefs. The difference is just that in (1) we form veridical pictorial beliefs about the reenactment, whereas in (2) we form deceptive pictorial beliefs about the original event.

VI. DOCUDRAMAS AS GENERATORS OF NONPERCEPTUAL BELIEFS THROUGH PICTORIAL IMAGININGS

As a limit case of reenactment, one might consider an ideal docudrama in which the actors are perfect look‐alikes of the real individuals portrayed and the events are perfectly reconstructed in all their details, and the audience is aware of this. This ideal docudrama would primarily involve the formation of pictorial beliefs not only about the actors but also about the reenacted events. Thus, according to the account of the documentary that I am proposing, it would count as a documentary, and not as a docudrama.

However, this is not an undesirable consequence. First of all, one might wonder whether the ideal docudrama represents a plausible possibility. There seem to be good intuitive reasons for a negative answer. But suppose, for the sake of the argument, that ideal docudramas can be made, or at least that a film can be presented as an ideal docudrama to the audience. Just as an observational documentary, an ideal docudrama provides us with allegedly reliable sensory information about the real events constituting its subject in virtue of a causal chain to them. The difference only lies in the fact that, in the case of the observational documentary, the causal chain is warranted by the film itself as photographic trace, whereas in the case of the ideal docudrama the chain involves a stock of accurate historical knowledge (arguably based on photographic traces) that allows the filmmaker to perfectly reconstruct real events.

Since the ideal docudrama shares the use plan of the documentary, we can treat the former as a special case of the latter. By contrast, the use plan of the real docudrama primarily involves the formation of pictorial imaginings whereby we can form nonperceptual beliefs about its subject. A docudrama thus differs from documentaries since it does not involve the formation of pictorial beliefs. However, a docudrama also differs from other fiction films since it invites us to form beliefs about its subject. In particular, this differentiates docudramas from historical fiction films such as The Leopard (1963) or Barry Lyndon (1975). The latter just use counterparts of real characters and events as a background for fictional characters and events that have no real counterpart, whereas a docudrama brings counterparts of real characters and events to the foreground even though it does not enable us to form pictorial beliefs about them.¹⁸ Relying on the notion of subject of a film introduced in Section I, one might say that docudramas, just as documentaries, are about real individuals and events, whereas historical fictions are about fictional characters and events that significantly interact with real ones. It is worth noting that the distinction I am proposing involves only the film's subject, not its accuracy. In this sense, Marie Antoinette (2006) is a docudrama despite its deliberate inaccuracy since its subject is the real queen, whereas Barry Lyndon is a historical fiction despite its accurate historical reconstruction since its subject is a fictional social climber.

Although docudramas are different from historical fiction films, they remain instances of fiction—even when their reconstructions are very accurate—in virtue of the priority of pictorial imaginings in their use plan. This is emphasized by those docudramas that, during the end credits, show us photographic images of the individuals who have been so far portrayed by actors, thereby signaling the fictional nature of the docudrama. One might mention, for instance, the end credits of Milk (2008), American Sniper (2014), Snowden (2016), or Bohemian Rhapsody (2018).¹⁹ It is worth noting that docudramas like these invite spectators to compare the content of their pictorial imaginings with the actual features of the subject only at the end of the screening. During the screening, instead, spectators are invited to focus on the content of their pictorial imaginings, that is, on what is going on in a spatiotemporal framework that has relevant analogies with theirs but is not theirs.

On the other hand, documentaries that exploit reenactment explicitly present the reenacted events as the outcome of a reconstruction carried out in the actual world thereby inviting the spectator to form pictorial beliefs about the reenactment itself.²⁰ That is to say that the pictorial imaginings elicited by a reenactment are not “constitutive imaginings” that constitute a fictional world in which we are invited to locate the events the film is about; rather, those are “ancillary imaginings” that just enrich our experience of events located in the real world.²¹ Thus, documentaries‐cum‐reenactment do not invite us to focus on a fictional world and to locate events in that framework in the way the docudrama does. Their main focus of attention is the actual world, not the fictional one. They normally signal this by grafting reconstructed scenes onto a patently documentary backbone. The docudrama, instead, is reconstruction all the way through.

In sum, both docudrama and documentary‐cum‐reenactment lead the spectator to form nonperceptual beliefs about some real events but the former does so via pictorial imaginings about events in a fictional world, whereas the latter via pictorial beliefs about reconstructions in the actual world. A docudrama presents reconstructions as if they were freestanding events, whereas a documentary‐cum‐reenactment presents reconstructions for what they really are. That is why a documentary‐cum‐reenactment is not fiction in spite of resorting to reconstructions whereas a docudrama is fiction in spite of concerning real events.

VII. WHY FICTION FILMS ARE NOT DOCUMENTARIES ABOUT ACTORS

According to a quip attributed to director Jean‐Luc Godard, fiction films are nothing but documentaries about their staging. Yet, this is just a quip that provocatively contradicts the relevant appreciative practice in which fiction films surely are not appreciated in this way. Thus, an account of the documentary should avoid treating fiction films as documentaries about their staging even though they are traces of their staging. For this purpose, Currie requires that a documentary exploits photographic traces of its subject. Although most fiction films also exploit photographic traces, these are of the staging, not of the subject.

However, if we characterize the documentary in terms of pictorial beliefs, we do not need this requirement anymore. Although in watching a fiction film one can form pictorial beliefs about its staging, this film does not primarily involve the formation of those beliefs but rather that of pictorial imaginings. Spectators can also form pictorial beliefs, for instance, if they are interested in actors or locations, but they are not forced to do so in order to enjoy a fiction film.

Here is the priority of pictorial imaginings over pictorial beliefs. A spectator of a fiction film who only forms pictorial imaginings, not pictorial beliefs, is still correctly appreciating the film in spite of overlooking some relevant aesthetic aspects (for instance, performances of actors). We might call “immersion” this special way of enjoying fiction films by only forming pictorial imaginings.²² Immersion, so understood, is an attitude that the relevant appreciative practice usually accepts and often encourages. On the other hand, a spectator of a fiction film who only forms pictorial beliefs about actors, not pictorial imaginings about characters, surely is getting things wrong. According to the relevant appreciative practice, this is not a special way of enjoying fiction films but rather a way of misusing them.

The priority of pictorial imaginings over pictorial beliefs in the fiction film can be stressed by considering the phenomenon that Robert Hopkins (2008) calls “collapsed seeing‐in.” The idea is that in engaging with a fiction film we usually experience a pictorial representation of fictional events, even though we know that this actually is a pictorial representation of a theatrical representation of fictional events. At the phenomenological level, the theatrical tier disappears—or, in Hopkins's terms, collapses. We are thus discouraged from forming pictorial beliefs about the staging in spite of the fact that we are actually facing photographic traces of it. This shows that a fiction film primarily involves the formation of pictorial imaginings in spite of being made of photographic traces. Docudramas are like all other fiction films in this respect, whereas documentaries, even when they use reenactments, do not allow pictorial imaginings to overtake pictorial beliefs.

At this point, one might call attention to those fiction films as, for instance, O Lucky Man (1973) or Dogville (2003), that encourage the spectator to form not only pictorial imaginings but also pictorial beliefs thereby making him or her fully aware of the fictional nature of what is depicted. Hopkins dubs such films “Brechtian” and he acknowledges that they constitute an exception to the dominance of collapsed seeing‐in as regards fiction cinema (2008, 155).

Does my account oddly cast Brechtian films as documentaries instead of as works of fiction? No, it does not, because Brechtian films do not go so far as to give up the priority of pictorial imaginings over pictorial beliefs. They limit themselves to putting pressure on such priority in a way in which a paradigmatic fiction film does not. In fact, they put pressure on such priority by endorsing it, not by dismissing it. In this sense, we might say that Brechtian films play, in the fiction domain, a role similar to that played, in the documentary domain, by documentaries‐cum‐reenactment. Just as the latter supplement the formation of pictorial beliefs (which is the backbone of the documentary as such) with some ancillary pictorial imaginings (which is the specific contribution of reenactments), the former supplement the formation of pictorial imaginings (which is the backbone of the fiction film as such) with some ancillary pictorial beliefs (which is the specific contribution of Brechtian elements).²³

Brechtian fiction films can also be docudramas. The Big Short (2015) or Vice (2018), for instance, are Brechtian docudramas in which the spectator is invited to form pictorial beliefs in some special moments during the screening (instead of just at its end, as in the abovementioned docudramas that show photographs of their subjects during the end credits). In spite of inviting us to form not only pictorial imaginings but also pictorial beliefs, Brechtian docudramas remain sharply distinct from documentaries‐cum‐reenactment inasmuch as their main focus of attention remains the fictional world.

Still, one might consider a borderline Brechtian film in which the parts that invite us to form pictorial beliefs are as relevant as those that invite us to form pictorial imaginings. Is this a docudrama or rather a documentary‐cum‐reenactment? An interesting example of this kind is (unsurprisingly) the film Brecht (2019), which tells the biography of Bertolt Brecht by combining staged reconstructions and recordings of real events in a way that prevents us from establishing which predominate. A film like this allows the spectator to choose whether to appreciate it as a docudrama or as documentary‐cum‐reenactment, just as the Jastrow picture allows the viewer to choose whether to see it as a duck or as a rabbit. That is to say that one is entitled to see Brecht either as a docudrama with Brechtian (in Hopkins's sense) inserts or as an expository documentary that massively exploits reenactment. Ultimately, one is entitled to see this film either as a docudrama or as a documentary because, as such, this is neither a docudrama nor a documentary.

Here is what films like Brecht teach us. Although the documentary and the docudrama are distinct categories, there can be films such that it is indeterminate whether they belong to the former or to the latter. This is a consequence of the fact that, as argued earlier, the distinction between the documentary and the docudrama rests upon cultural norms implicit in practice. Films like Brecht show that such norms, which in the vast majority of cases determine whether a film is a documentary or a docudrama, are not fine‐grained enough to do so in any case. However, this does not lead us to the conclusion that there is no relevant boundary between the domain of the documentary and that of the docudrama. The right conclusion is, rather, that such a boundary involves a thin border area that belongs to neither.

VIII. ANIMATED DOCUMENTARIES

The distinction between pictorial beliefs and pictorial imaginings has allowed us to distinguish two categories of films made of photographic traces, namely, documentaries and fiction films. But what about animation films? These, in fact, are not made of photographic traces (or are just photographic traces of pictures, not of real events). However, the distinction between documentaries and fiction films should also apply to animation, so as to draw a boundary between animated fiction films like Bambi (1942) and documentaries like Chicago 10 (2007) that exploit animation instead of photographic traces. One might thus object that animation does not involve the formation of pictorial beliefs, and therefore documentaries that exploit animation cannot primarily involve the formation of pictorial beliefs. This would lead my account to the unwelcome conclusion that animated documentaries are not genuine documentaries.

My reply is along the lines of the strategy whereby I have argued that documentaries‐cum‐reenactment are genuine documentaries. Just as documentaries‐cum‐reenactment, animated documentaries resort to reconstructions without primarily involving pictorial imaginings. Here is the difference between an animated documentary and an animated fiction.²⁴ The latter primarily involves the formation of pictorial imaginings concerning events in a fictional world, whereas the former primarily involves the formation of perceptual beliefs concerning an animated reconstruction in the actual world that provides us with information about the subject of the documentary. That is to say that the pictorial imaginings elicited by animated documentaries are not constitutive of a fictional world but rather ancillary to our experience of the actual world. Thus, animated documentaries must be underlain by perceptual beliefs, whereas animated fictions do not need them, as shown by the fact that the latter films, unlike the former, enable immersion.²⁵

The primacy of perceptual beliefs in animated documentaries can be emphasized by considering that these films often graft animations onto a nonanimated documentary backbone, or at least resort to a voice‐over narrator whom we are invited to treat as a real person, thereby leading us to also form perceptual (auditory) beliefs about him or her. For instance, the animated documentary Sunrise over Tiananmen Square (1998) invites us to form auditory beliefs about the filmmaker Shui‐Bo Wang who is telling his autobiography.

Still, most perceptual beliefs that underlie animated documentaries are different from the pictorial beliefs that underlie paradigmatic documentaries. The latter beliefs concern what is depicted, whereas the former concern the animated pictures themselves. Thus, animated documentaries primarily involve pictorial beliefs in a peculiar sense; perceptual beliefs about pictures, not about what is depicted. An animated documentary differs from an animated fiction by inviting spectators to treat pictures primarily as objects that have their place in their actual world and allow them to form nonperceptual beliefs about individuals and events that have their place in their actual world as well.

Borrowing an example from Currie, let us consider a “documentary about Disneyland in which Mickey Mouse acts as the narrator” and “we see Mickey on screen, as he takes us through the location” (1999, 292), and let us compare this with a fiction film like Who Framed Roger Rabbit.²⁶ The latter invites us to focus on a fictional world inhabited by both flesh‐and‐blood individuals and animated creatures, whereas the former invites us to focus on the actual world thereby forming perceptual beliefs about a Mickey‐image that is helping us to form pictorial beliefs about Disneyland.²⁷

Although animated documentaries are genuine documentaries in virtue of a primacy of pictorial beliefs over pictorial imaginings, we must acknowledge that animated documentaries are peculiar nonparadigmatic documentaries since they primarily involve the formation of a peculiar sort of pictorial beliefs. However, this ultimately is a welcome conclusion since it contributes to explain why animated documentaries are usually considered peculiar nonparadigmatic documentaries in the relevant appreciative practice.

IX. LIVE TELEVISION AND CCTV

So far, I have considered objections according to which my account of the documentary is too narrow, and I have addressed them by showing that it can include documentaries about out‐of‐reach subjects, as well as those that exploit reenactment or even animation. However, one might still object that my account is too broad in the sense that it would be forced to include “informational pictures” such as those of live television or CCTV, which primarily involve the formation of pictorial beliefs and yet are not normally treated as documentaries in the relevant appreciative practice.

In order to reply to this objection, I analyze the term “film” in my characterization of a documentary as a film that primarily involves the formation of pictorial beliefs. This term can have two distinct meanings. As Meskin and Robson point out, it can be used “to pick out both a medium (by which we mean not the particular materials from which they are made but a practice, or set of practices, for using some materials to make artifacts which may or may not be art) and an art form (that is, a particular sort of appreciative art kind)” (2010, 562, my emphasis).

If one conceives of the film as a medium, then also informational pictures satisfy the definition of documentaries as films that primarily involve the formation of pictorial beliefs. Yet, this is no longer the case if one conceives of the film as an art form, that is, a kind that groups works in such a way that “we normally appreciate a work in the kind by comparison with arbitrarily any other works in that kind” (Lopes 2010, 17). Since informational pictures such as those of live television or CCTV do not belong to the film as an art form, they do not belong to the documentary either, provided that we define documentaries as films (that is, members of the corresponding art form) that primarily involve the formation of pictorial beliefs.

In this way, we can highlight both similarity and difference between documentaries and informational pictures. They both primarily involve the formation of pictorial beliefs, but documentaries do so in the framework of the appreciative practice in which fiction films also are evaluated, whereas informational pictures are mainly meant to supply information. For instance, both the documentary Zidane: A 21st Century Portrait (2006) and an ordinary live broadcast of a soccer match primarily involve the formation of pictorial beliefs about what is happening in the soccer field, and yet the former, unlike the latter, is a documentary.

On the one hand, one might characterize documentaries as “upgraded” informational pictures that are candidates for appreciation in the practice in which fiction films also are evaluated. On the other hand, one might say that informational pictures are “utilitarian” documentaries that do not play any role in that practice. This allows us to capture at the same time what documentaries and informational pictures have in common (that is, primarily involving the formation of pictorial beliefs) and what differentiates the former from the latter (that is, the relevance of appreciation).

X. MOCKUMENTARIES

Another problematic case—one might finally object—are fiction films such as Forgotten Silver (1995) or Borat (2006), which are usually called “mockumentaries.” These disguise themselves as documentaries, thereby primarily involving the formation of pictorial beliefs. Therefore—so the objection runs—my account would wrongly cast them as documentaries instead of as fiction.

My reply is that the use plan of mockumentaries should distinguish them from deceiving documentaries, otherwise we would wrongly appreciate them as documentaries instead of as mockumentaries. On closer inspection, the use plan of the mockumentary involves the recognition of an attempt to deceive the spectator. In principle, the spectator can carry out such recognition in two different ways, that is, either by being deceived by the mockumentary and then retrospectively unmasking it, or by enjoying the mockumentary as if he or she is being deceived by it though he or she actually is not. The latter case is much more common in the relevant appreciative practice, in which mockumentaries are usually presented as such to the audience.²⁸

Anyway, a mockumentary involves primacy of pictorial imaginings over pictorial beliefs in both cases. If the spectator is actually deceived by the mockumentary and then unmasks it, the use plan involves forming pictorial beliefs and afterward turning them into pictorial imaginings. If, instead, the spectator enjoys the mockumentary as if he or she was being deceived by it, the use plan involves forming pictorial imaginings that he or she treats as if they were pictorial beliefs. Ultimately, in both cases the spectator pretends to form pictorial beliefs but at the end of the day actually forms pictorial imaginings.²⁹

Conceiving of a documentary as a film that primarily involves the formation of pictorial beliefs allows us to show in which way mockumentaries mock documentaries thereby differentiating themselves from other fiction films. While an ordinary fiction film invites us to locate the depicted events in a spatiotemporal framework different from ours, a mockumentary invites us to locate the depicted events in our own spatiotemporal framework, just as documentaries do. Yet, in the case of the mockumentary, such cognitive activity does not end up in the formation of pictorial beliefs but only in that of pretended pictorial beliefs that ultimately reveal themselves to be pictorial imaginings.

XI. CONCLUSION

In this article I have argued that a documentary, as a work of nonfiction, primarily involves the formation of pictorial beliefs, whereas a fiction film primarily involves the formation of pictorial imaginings. To put it more vividly, a documentary provides one with viewpoints warranting that one might reach them if one possessed a time machine, whereas a fiction film does not.

This distinction holds even if one considers films such as Life, and Nothing More … (1992), Boyhood (2014), or Entre Dos Aguas (2018), which play with the divide between fiction and nonfiction (see Di Summa‐Knoop 2014). Although the latter films are so accurate in depicting their stories that you might be tempted to treat them as documentaries, you cannot do so because even a time machine would not allow you to reach a viewpoint from which you could perceive the events depicted by these films. A time machine would allow you to go see only the actors, not the characters; the staging, not the events that constitute the story.

At most, if you are watching a docudrama, you can use your perceptual experience of the events depicted to form beliefs about certain events in your world, and yet there is no warrant that you could travel with a time machine thereby directly perceiving these very events from the same viewpoint from which you were perceiving them on the screen. This means that the beliefs you are forming while watching a docudrama are not pictorial beliefs but rather the nonperceptual outcome of pictorial imaginings. Therefore, docudramas remain on the fiction side of the border between fiction and nonfiction in cinema. One might say that docudramas are closer to the border than paradigmatic fiction films, just as, on the nonfiction side, documentaries that exploit reenactment or animation are closer to the border than paradigmatic documentaries. There can even be films that have their place exactly on the border thereby preventing us to determine whether they are documentaries or docudramas. Nevertheless, I have argued, the divide between fiction and nonfiction remains crucial to the appreciation of cinema.³⁰

REFERENCES