You can view a
version of this
article here.
The Why and the How of Movie Trailers

14 June 2004 | 4697 words

An essay on trailers, new technologies that effect trailer viewing, and signs of emerging factors involved in our contemporary viewing of film narrative.

Throughout their exhibition history, trailers have been seen as advertisements selling films to the public. My goal is to show that this is not the only thing trailers do. By considering movie trailers in the light of recent developments in trailer viewing, specifically, the downloading of trailers from the internet, I hope to show that trailers also always provide the narrative information and genre contexts that attentive spectators can—and do—use to structure their viewings of films.

This may at first seem contrary to the lived experience of millions of trailer viewers: trailers, after all, seem to do nothing but sell their film. My goal is not to debunk this very real experience. Instead, by showing how this experience is linked to a particular site of exhibition and how, in turn, new experiences take its place when trailers are viewed in new places with new technologies, I hope to point out an emerging factor in our contemporary experience of film narrative.

Cinema v. Internet Trailers

Cinema trailers have typically been viewed in theatres before the screening of the main attraction, and it is to this context that their appeal is pitched. When making a trailer, producers attempt to “‘[fill] the quadrant:’ that is, [reach] out, in 150 seconds, to younger men, older men, younger women and older women [1]”. But making a trailer is just the beginning. Once the trailer is finished, it must be placed in a theatre and be seen. Although an object of intense negotiations between distributors, theatre owners, and even sometimes filmmakers, the placement of trailers is probably at best haphazard; in a best case scenario, the trailer will be correctly placed in front of its target audience and they will watch it. But even this audience includes only potential viewers for the film it advertises; of those who watch, only a few will be interested in what they see, and there is no guarantee that even these few potentially interested viewers will pay attention while the trailer plays. Trailers must compete after all with soda commercials, animated theatre logos, other trailers, and last minute runs to the concession stand. Within this context, it seems natural that trailer producers would devote most of their energy to capturing the attention of theatre-goers and that their product would be seen universally as an advertisement.

Potentially useful strategies—suggested by Gerard Genette’s discussion of sales pitches in book prefaces—include selling the value of the subject matter, the reputation of the author, and the novelty of the treatment [2]. Trailers clearly can and do employ all of these. They present their films, for example, as important, truthful treatments of lightning rod issues (dead man walking, 1995). Or, they can present them as the product of established, popular, or important directors or stars (eyes wide shut, 1999). Writers occasionally receive similar treatment, and producers too, though these last are usually billed as “the makers of” a previous film. Selling the novelty of the film’s treatment of its material is especially wide-spread. One writer has noted for example, that “as years passed and studios keyed in on the importance of advertising, a distinct, hard-sell style of trailers emerged. Every movie coming down the pike was touted to be more ‘spectacular’ or ‘hilarious’ than anything seen before [3]”. This strategy has led to a general tendency in trailers to celebrate spectacle and excess, but novelty manifests in other guises as well: most often as notices that a film is the newest in a series (star wars: episode 1— the phantom menace, 1999) or the product of new technology (final fantasy: the spirits within, 2001).

In addition, trailers often sell films in terms of their likeness to other films or film forms. Thus comedy trailers will display the presence of gags and jokes; romantic comedies will establish love and impediments to it. In the same vein, successful predecessors can be identified. Thus the trailer to maid in manhattan (2002) consciously evokes the story of pretty woman (1990) while being shown before screenings of sweet home alabama (2002) [4]. All of these together, comprise a repertory of strategies trailers use to tell viewers why they should watch their film. Given the difficulties inherent in the theatrical presentation of trailers and the styles they encourage, they can also reasonably be assumed to mark the limits of what viewers have typically expected of trailers. But the internet, with the new viewing contexts and behaviours it supports, seems to be changing that.

A recent study reported in The New York Times claims that “movie trailers are the most-watched video material on the Web [5]”. Likewise, a major ad campaign for Bell Canada presented in movie theatres and in mass-mailings in winter 2003 marketed high-speed internet service almost exclusively as a means of more easily downloading trailers. If the shift in trailer viewing indicated by this study and Bell Canada’s ad campaign is to be believed, several changes can be expected in the ways viewers regard movie trailers.

At its most basic level, the downloading of trailers transforms the trailer audience from a potentially interested target group into already interested, attentive viewers. Unlike cinema trailers, internet trailers are not encountered by chance [6]. Downloads must be sought out and take time to view even over high-speed connections. As a result, internet downloaders are more likely to view trailers for films that have already been pitched to them and that already interest them. Thus they likely view internet trailers seeking something other than a sales pitch. Given this new situation, I would suggest that internet trailers, no longer burdened with selling a film, instead provide guidance for how best to watch the film, a supposition supported by the numerous web sites where fans, anticipating upcoming releases, pour over downloaded trailers to figure out what to expect from the upcoming release [7].

Trailers as Viewing Guides

In general, downloaders seem to turn to internet trailers to learn basic story information: an introduction to principle characters, the sense of the general setting including time period or region, and a basic story abstract. Characters and setting can be presented in a variety of ways: most simply a montage of images can show principle characters, iconic landscapes, or period clothing; but more typically, characters and setting are identified by name or described by a voice over or segments of film dialogue. In the same way, story abstracts vary greatly in detail, though in general they at least identify a narrative arc: characters will race to a goal (brewster’s millions, 1985), a little guy will overcoming great odds (rocky, 1976), a crime will be solved (seven, 1995), two men will fight for the love of a women (the talented mr. ripley, 1999) or two lovers will find each other despite the obstacles in their path (my big fat greek wedding, 2002). These story abstracts naturally situate the film within appropriate genre contexts, and these contexts, like the basic story information that suggest them, are an important part of the viewer’s preparation to view the film. Once evoked genre expectations elicit the complex, variable viewing strategies that viewers will use to make sense of the film’s narrative and style.

This information of course will also be provided by the film itself, and so in a sense, the preparation provided by the trailer is unnecessary. However, when the trailer does prepare viewers, its work is not without influence, especially since basic information and story abstracts are usually supplied by third person omniscient voice-overs that tend to become “the voice of the image maker [8]”. This natural authority of the voice-over is emphasized by its place at the center of the audience’s attention. As Jonathan Glenn, a trailer producer, has pointed out:

The big secret of movie trailers is that they are essentially long radio commercials with pictures… Try closing your eyes during a trailer—you still know what’s going on with the story just by virtue of what you’re hearing [9].

Given the attention given to spoken elements in trailer production and the ease with which voice-over conveys basic story information, it is not surprising that they “tend to voice the ideological and/or moral agenda behind the film” in the same way that similar voice-overs do in narrative film [10]. These statements, which seem to come from the image-maker and thus echo the function of prefaces’ statements of authorial intent, unavoidably frame viewer responses to the film text by forcing viewers “to take a position, positive or negative, in relation to [them]” [11].

Viewers of internet trailers pay similar attention to the image track, as the image stills on fan sites attest. Even imponderable aspects of the film are the object of speculation: one analysis of lord of the rings: the two towers (2002) trailer asks whether a figure barely perceptible through smoke in one still might be a villain, while another reads several segments of the trailer as suggesting that Liv Tyler’s character might give Viggo Mortenson’s his sword. Questions such as these are obscure, but also, oddly, at the heart of narrative experience. Rightly or wrongly, they shape the way the people asking them will view the film. There is nothing new in this recognition that expectations shape narrative. What is new is that in this case, the questions are not provoked by the narrative. They are provoked by a particular method of trailer viewing that treats trailers as viewing guides rather than sales pitches. It is exactly this approach to trailers that is encouraged by internet downloading.

Cinema to Internet Trailer

Of course, claiming that cinema trailers sell films and internet trailers prepare viewers to watch a film (even though they are the same text) is one thing; identifying what this implies about film viewership is another. To consider this question, I would like to take up the trailers for two films: natural born killers (1994) and velvet goldmine (1998). Both of these films focus on controversial subject matter but were distributed nationally by major film companies. As a result, it is reasonable to suggest that special care was given to the sales pitch presented in the cinema trailer. By the same token, both of the films were produced by writer-directors who can lay legitimate claim to being film authors. Therefore, it seems reasonable to assume that their treatment of the controversial material in their films is carefully constructed toward a particular end. By considering the trailers of these two films, it may be possible to ascertain to what extent cinematic trailers can effectively prepare viewers to produce a preferred reading of an authored text when viewed through the emphases encouraged by internet downloading.

Because of its ambivalent treatment of excessive violence, the trailer for natural born killers was farmed out by Warner Brothers to Global Doghouse, a trailer production company generally known for its successful treatment of “difficult movies [12]”. The entire first half of the resultant trailer is comprised of shots of the American flag and news clips of events discussed in earlier Stone films. When images from the film finally appear, they unroll at a breakneck pace and, in accordance with Jonathon Glenn’s equation of radio ads and trailers, provide primarily tonal emphasis to the voice-over. This voice-over, spoken by one of the 10 to 15 paternal male voices that narrate all American trailers [13], reads as follows (ellipses mark points where pauses allow the noise or dialogue of the film images to be heard):

platoon. wall street. born on the fourth of july. jfk. Oliver Stone’s vision has changed the way we saw our past; now he takes a look at where we are and where we’re going. And you’ll be shocked at what he sees. Micky and Mallory. Feared by thousands…watched by millions…Woody Harrelson. Juliette Lewis. Robert Downey Jr. Tommy Lee Jones….natural born killers. In the media circus of life, they were the main attraction.

The sales pitch of this voice-over is clear. People should watch this film because it has been made by a respected director and is part of an important series of films. Furthermore, it is deals with an important issue; it is timely; and it is relevant to how we live our lives. This claim to importance is supported by the flag imagery, the integration of news footage, the list of celebrated actors, and the title card of simple white text on a black background that closes the trailer and reads “an Oliver Stone film.” Even the expressionist aesthetic of the selected film images and editing pattern of the closing montage support this claim, suggesting the film provides an artistic and intellectual (rather than popular) treatment of violence in the media. This trailer was considered a success and “helped cement [Global] Doghouse’s reputation [14]”. What’s more important for our concern is that it seems to hold up when subjected to the new expectations of the internet viewer.

When the question asked of the trailer is shifted from “Why should I watch this film?” to “How should I watch this film?”it quickly becomes clear that this trailer’s sales pitch is also a near perfect presentation of a preferred reading; this trailer sells Oliver Stone’s vision. All basic story information is clearly and directly provided by the trailer. Images of the two principle characters are shown repeatedly, and they are introduced by name in both the voice over and in a brief moment of dialogue. Furthermore, all of the secondary characters are presented without names under the list of actors in the voice-over. Settings common to road movies and crime films are suggested through images of jail cells, car interiors, and the iconic desert Southwest. The story abstract is very weak, but the relatively extended view of Mickey and Mallory in the press mob that closes the trailer’s violent imagery could perhaps be read as suggesting a serial killer chase film à la bonnie and clyde (1967) or badlands (1973). Of course, the absence of a clear story abstract, though atypical, here serves to reinforce even further the “artistic” implications of the stylized imagery and the auteur focus of the sales pitch.

The violence of this artistic imagery likewise announces the key ambivalence of the film to viewers. The film and the trailer clearly elicit and maintain interest through their representations of extreme violence, a classic example of the spectacle encouraged by the pressure for novelty typical of film trailers. Yet, this use of violence is the object of the film’s criticism. This paradoxical use of violent spectacle and simultaneous critique is likewise present in the voice-over. If the narrator suggests Stone’s film will change the way we look at our world, he also pauses to let Mallory absurdly confess her love to Mickey in the midst of carnage and for Mickey to turn to the camera and say “You ain’t seen nothing yet.” Thus the trailer not only clearly states the overt moral agenda of the film (a critique of media’s exploitation of violence), it also directly presents the moral failure of the film as it exploits violence for its own purposes. Thus, this cinema trailer prepares film viewers not only for the film’s content and Stone’s message, it prepares them for the ambivalence of the image-maker’s paradoxical celebration of this content. And all of this while selling tickets to the film.

With velvet goldmine the situation is very different. In fact, the trailer for velvet goldmine is a perfect example of the relationship most people accuse trailers of having with their film: i.e. the trailer misrepresents the film in order to sell it. Todd Haynes’s film is a complicated treatment of queer precedents for and responses to the sexual trangression of 70s glam rock, and at its release, posed a series of marketing problems for Miramax Films. The most obvious, of course, is the gay content in the narrative, which, unlike violent content, tends to limit rather than elicit viewers. Speaking in 1998, the director Bill Condon claims that gay films in national release generally make a maximum of three million dollars at the box office, an amount that, at the time, apparently marked the size of the gay market [15]. Thus, any attempt to sell a preferred reading of the content of velvet goldmine posed the risk of limiting the potential profits of the release. But perhaps more importantly, any attempt to market gay material by Miramax in 1998 posed the risk of antagonizing its parent company, Disney.

In the months prior to the release of Haynes’s film, Disney, because of its perceived support of homosexuality, became the target of a widely publicized boycott by the Southern Baptist Convention, the largest protestant denomination in the United States, a boycott which was subsequently endorsed by a number of high-profile, nationally active religious and political organizations. Miramax’s films were considered particularly offensive in this regard and thus seemed ill-placed to market Haynes’s film in terms of its content. (Miramax was one of only two subsidiaries mentioned by name in the resolution authorizing the boycott) [16]. Warner Brothers had marketed Stone’s offensively violent imagery under the cover of its artistic merits; but Haynes lacked such celebrity. Worse, what celebrity he did have created more problems as it arose from the accusations of obscenity directed at his first feature, poison (1991), accusations made by the same people now supporting the Disney boycott. Thus Miramax faced the awkward prospect of marketing an overtly queer film by an unknown or, worse, infamous director at exactly the moment when its films were being singled out and faulted for tarnishing an American institution (Disney). Apparently, the decision was made to skirt this issue entirely [17].

The trailer eventually released obscured the gay content at the film’s center and sold the film as pure spectacle and nostalgia. The trailer’s sales pitch was conveyed through a voice-over and was supported by images and dialogue fragments arranged to produce a “call and response” format. With these moments of interposed dialogue indicated by italics, the voice-over reads as follows:

It doesn’t really matter much what a man does with his life. What matters is the legend that grows up around him. Brian Slade was the wildest rock star to come out of London. The biggest thing since slice Beatles. But that wasn’t enough. We set out to change the world. What happened? Who did it? And why? Next week is the 10 year anniversary of the whole shooting incident. One journalist is about to look into the mystery. I was trying to contact you about a story. From the moment Brian Slade stepped into our lives, nothing would ever be the same. He was, in the end, like nothing he appeared. Right after everything crashed, Brian seemed to get lost in the lie; he became something else. Miramax films invites you to throw away your expectations and take a magical trip back to the 70s, when the glam scene rocked London and the outrageous music, fashion, and behaviour shocked the world. I knew I should create a sensation.

Unlike the one presented in the trailer to natural born killers, this trailer’s sales pitch makes no claim that the material is important, new, or timely. Haynes’s name never appears, and there is no reference to earlier works by Haynes that might suggest a frame of reference for this film. Instead, the film is pitched (and here I may be exaggerating, but not by much) as Disney with an edge: this will be an “outrageous” “magical trip” to a world of rock music, sequins and misbehaving celebrities offered up by the makers of the transgressive but fun pulp fiction (1994) and clerks (1994). This final invitation to view the film clearly dominates even the elaborate story abstract that defines the film’s narrative as a murder mystery. It goes without saying that it skirts the gay material that pervades the film. It is obvious therefore that the trailer for this film misrepresents the film in order to sell it to potential viewers.

But misrepresentation is not inaccuracy. A murder (hoax) does offer the occasion for the narrative and spectacle is a central pleasure of the film’s visual style. The costumes are outrageous—and won an Oscar for it—and the gay content that pervades the film is not its exclusive subject matter. Thus an evaluation of the trailer’s faithfulness to the film is beside the point, and worse than this, avoids a much more important issue.

As a sales pitch cinema trailers are successful and useful to the extent that they make members of the public into members of a film audience. Accuracy and inaccuracy (and this must be a key point in any discussion of trailers) are therefore irrelevant because they become an issue only after a ticket is bought. The cinema trailer’s function is to put bodies in a theatre, not be fair to the film. New viewing contexts, of course, change all that. Viewed over the internet, cinema trailers no longer simply put people in a cinema; they become a guide for watching the film. Considered as explanations of how to watch the film rather than as reasons to watch the film, the misrepresentations of the trailer for velvet goldmine become important because they badly prepare viewers for what they will see.

So what film does velvet goldmine’s trailer prepare the viewer to watch? It seems clear that, on a basic level, viewers are told to expect a film that will be fun, celebratory, and maintain a certain lightness of treatment. Furthermore, the fact that potential viewers are invited to “relive” the glam rock scene suggests that the film is a nostalgia piece that will present something familiar, something remembered, even if only vaguely. In terms of genre, the detailed story arc very clearly elicits the murder mystery. This familiar context conjures specific viewer expectations (e.g. heavily plotted narrative) and behaviours (e.g. a tendency to search for objective facts or clues and to engage in puzzle solving) that will set the stage for viewing of the film as surely as do the questions posed by fans awaiting the two towers. Perhaps most importantly, in a contemporary cinema culture where “lesbian, gay, bisexual, and queer understandings of popular culture…exist as appropriative of and subsidiary to taking things straight,” the complete absence of gay material in the trailer necessarily suggests that there will be none in the film [18].

It seems fair to say, that anyone prepared to watch a film in this way, will be bewildered by what confronts them when Haynes’s film is projected. The film is a spectacle and, I would argue, fun, but it is not light and certainly not nostalgic in the way that this term is generally used in discussions of contemporary movies (think forrest gump, 1994, or almost famous, 2001). Viewing from these contexts is sure to be frustrating. There are too many layers beneath the surface spectacle, the narrative is too convoluted, and the film too long to allow simple attention to the costume and music. Worse for the nostalgia viewer is the fact that Haynes seems bent on revising rather than remembering glam rock, which reveals the expectations raised by the trailers story abstract as especially misconceived.

Patterned after citizen kane (1941), the narrative works more as a group memoir than a murder mystery (not least because there is no murder). As such its construction is sprawling, at times redundant, at other times obscure, and in general lacks the forward drive that is emblematic of the mystery genre. It is also likely that viewers drawn to this film to relive a memory are probably the viewers who will be the least open to the film’s revision of this memory, a revision that begins with the opening sequence and continues unabated to the end of the film: the film opens with emblematic queers Oscar Wilde and Jack Farie and ends with Christian Bale’s character (a figure in the film of the glam rock fan and the film’s audience) reclaiming his own memories of glam rock as the sign of an emerging gay identity.

This mismatch between the sales pitch of the trailer and the way in which it prepares viewers to watch the film was most likely obscured when velvet goldmine was released; and without an examination of the film’s reception upon its release, it would be impossible to determine what influence it may have had on its success at the box-office. Yet, examining it in the light of internet downloading still helps to clarify the distinction between two functions and to suggest what importance they each have on the viewing of narrative film.


Recent developments in trailer viewing, specifically, the downloading of trailers from the internet, are revealing the notion of trailers as mere advertisement to be only a partial understanding of the work of these devices. Far from being a simple sales pitch that accurately or not vends its wares, the trailer provides the narrative information and genre contexts that attentive spectators can—and do—use to structure their viewings of the film. But what are we to make of all of this since, after all, the study of trailers can be—and should be—only a small component of film study?

To answer I would suggest that as trailers migrate into new sites of exhibition through new technologies, the changes they undergo may serve as key indicators of emerging film viewing practices. If this is true, and I hope that my analysis at least suggests that it is, then trailers’ functions and uses should not go unexamined. Distinguishing between a sales pitch and a viewing guide and identifying the contexts that make trailers first one and then the other, seems to be a fundamental first step in this direction. But this is only a beginning. Further study would hopefully identify how particularly important preparatory functions such as genre identification, story descriptions, and ideological declarations do their work. In addition, it would be important to examine how other presentations of trailers alter their function. But other potentially rich areas of analysis abound. To cite only two examples, I would call attention first to their universal incorporation into DVD releases, where they are an integral part of an extended paratextual system of interpretation; and secondly, to the unexamined but important role they play in the creation of commercial auteurs, and therefore, our contemporary cinema culture. It is through consideration of these and related questions about the work trailers are doing that these mysteriously captivating but frustratingly banal pieces of cinema will begin to be useful for film studies.

1 Marshall Sella, “The 150-Second Sell, Take 34” The New York Times July 28, 2002.

2 Gerard Genette, Paratexts, Trans. Jane E. Lewin (Cambridge: Cambridge University Press, 1997), pp. 198-208.

3 Sella, “150-Second Sell.”

4 Rick Lyman and Laura M. Holson, “Holidays Turn into Hollywood’s Hot Season” The New York Times November 24, 2002.

5 Amy Harmon, “Movie Studios Provide Link for Internet Downloading” The New York Times November 11, 2001.

6 The trailer text in both cases is the same. “Cinema trailer” and “internet trailer” will be used to distinguish between different methods of viewing the same text.

7 In the Spring of 2003 a series of sites were dissecting the trailer for the two towers: one of the most extensive frame by frame analyses of all of the available trailers. This particular site also directed interested downloaders to movies where each of the available trailers was being screened, an important service for those willing to buy a ticket to a movie they wouldn’t otherwise care about in order to watch a particular trailer on the big screen (i.e. the way it was meant to be seen.). The trailer breakdowns can be viewed at:

8 Sarah Kozloff, Invisible Storytellers: Voice-Over Narration in American Fiction Film (Berkeley, Los Angeles, and London: University of California Press, 1988), pp. 74, 75.

9 Jonathan Glenn, interview by David Geffner in “Anatomy of a Trailer,” RES: Resolution Independent (July/August 2002), p. 43.

10 Kozloff, Invisible Storytellers, p. 80

11 Genette, Paratexts p. 224.

12 David Geffner “Anatomy of a Trailer,” p. 42.

13 ibid., p. 44.

14 ibid., p. 42.

15 “Director’s Commentary,” Gods and Monsters, dir. Bill Condon, 1998. DVD.

16 The censure of Disney can be read at: The resolution that started the boycott can be read at:

17 Any links between the boycott and marketing decisions made by Miramax or Disney are obviously conjectural. Still, one week after the commencement of the boycott, Disney recalled a hip-hop album already in distribution, claiming it was inappropriate to release obscene material through Disney subsidiaries. Despite denials, this move was viewed as a response to the boycott. The story as reported by ABC news can be viewed at:

Circumstances surrounding Miramax’s film marketing strategies are also suggestive. In the three years prior to the 1997 boycott, gay films such as priest (1994) and lie down with dogs (1995) were released by Miramax and marketed as such. In the three years following the boycott, its explicitly gay releases, velvet goldmine and the talented mr. ripley were not.

18 Alexander Doty, Flaming Classics: Queering the Film Canon (London: Routledge, 2000), p. 52.

Brian Crane is currently a PhD student at Université de Montréal.

<<< :: Previous entry
Les DVD qu’on peut ne pas acheter

Next entry :: >>>
The Provocateur Auteur

[ Back to Top ]

ISSN 1715-7641
Copyright © 2004-2005 Synoptique and its respective authors. All rights reserved.
Reproductions of any portion of this website only with the
expressed permission of Synoptique and its respective authors.

 * * *