Variations on the Musical Image:
An Interview with Philip Glass

James Tobias


Admittedly, the opportunity to interview composer Philip Glass gave me some apprehension. Through four decades of prolific output for concert stage, theater, opera, and film, Glass’s musical idiom, a disciplined calculus of textural forms mirrored in its own expressive flexibility, has only become more accessible, and influential, in its adaptations to film or theater in collaborations with equally prodigious figures in those fields—Robert Wilson, Godfrey Reggio, Martin Scorsese all came up in our talk. But his accomplishments pale before the central question—how do you ask questions about not simply the form of the music or its stature but what the music means? I needn’t have worried. Interviewing Philip Glass was something like throwing a log on a warming fire—with a bit of fuel added to the flames, Glass would simply blaze forth, illuminating through memorable anecdote or insightful observation the diverse aspects of the process of creating music for films, narrative and documentary alike. The interview took place November 1, 2001, during a visit by the Philip Glass Ensemble to Ann Arbor, Michigan. The Ensemble was performing three evenings of works scored for film projection—a practice revived from the silent era which has held increasing interest in recent years. A case in point, Glass’s re-scoring of the classic horror film Dracula (1931) gives pause to reconsider the use of music in film.

I’ve long considered the variants and derivations of what was once called “minimalist” music to be closely aligned with music for film. In dramatic films of the early years of sound, musical scores consisted of as little as a title theme and whatever diegetic music might have been necessary; and these were stripped down orchestrations suitable for tight budgets and low-fidelity equipment. But as full-length scores evolved for sound film, film music composers learned to emphasize flexibly repeatable motifs that might be extended, abbreviated, or otherwise remixed as final adjustments were made to both sound and image. The more modular motifs of filmic “background music” were generally less meaningful structurally than a leitmotif would have been in opera, but that didn’t prevent them from dramatic use: establishing location, mood, and so forth. These “underscored” motifs also provided pacing or cued audience orientation, yet only rise to a flourish at points of dramatic emphasis. Film music’s overriding aesthetic of dramatic ambience and modular structural development along loosely imagistic lines is shared by some strains of “minimal music.”

In this regard, Glass’s score for Dracula is a fascinating exercise in returning to early sound cinema the dramatic implications of musical developments that can only have arisen in its wake. Directed for Universal by horror specialist Tod Browning, the film made typically sparing use of music: only a title theme, and elsewhere, source music appropriate to scenes where it would be heard within the story. Glass’s Dracula, on the other hand, is scored through, and even without changes to image or dialog, the dark count appears in a new light: the film is at once less spooky, more expository, and more romantic. The imagery becomes more remote, its charms standing in relief. The score works like a score should: it sets scenes or carries us from one scene to the next; it suggests invisible events or character intent. But it is also more than a score. The musical pre-eminence of the soundtrack—its refusal to merely underscore actions onscreen—skews the image away from its moorings in the conventions of horror. Its generic identity stripped, the film accompanies the pulsating music as much as the reverse. With this musical appropriation of the image, we experience a historical dimension of cinematic narrative—much as when we see historical footage presented in slow motion to emphasize not so much a vision of the past but the past of our cinematic vision. Further, Glass’s score points to the work of the audience in interpreting sound and image.

Below, Glass makes the point that an active relationship with the work of art is the essence of a modern aesthetic. Hanns Eisler had the same point in mind with his soundtrack for Alain Resnais’s Night and Fog (1955), a documentary on Hitler’s death camps which deployed music against image to beg interpretation of the searing and still inconceivable events of the Holocaust. To sensitively provoke audiences to a deliberation of the impossible, Eisler scored, for example, tracking shots of the deserted camps from outside their wire fences with a sweet, almost summery setting of pizzicato strings suggesting a verdant country idyll—not industrial-scale murder. While Glass is rarely so univocal in his scores as to merely counterpoint the visual, his documentary work can be seen as a fully realized version of Eisler’s experiments in engagement through pointed combination of sound and image—whether depicting global contemporaneity in Reggio’s Qatsi Trilogy, with 1983’s Koyaanisqatsi being a late-twentieth-century exemplar of the modernist city symphony pioneered by Vertov or Ruttman; or, the serpentine self-questioning postmodernism of Errol Morris’s The Thin Blue Line, where Glass’s ambiguously reflective “Metamorphosis” theme quietly assists the film’s fraying of a narrative of crime and punishment.

Two overlapping dynamics can be pulled apart in regards to the open-ended relationship between audience and work Glass expects. In principle, this aesthetic of musical engagement suggests that music is as important as image in the meaning making process. The corollary in production terms is that the composer must share creative responsibility for the film to some appropriate degree. The way these matters are resolved in non-fiction film has tended to differ somewhat. First, in regard to the importance of music, traditional theories of documentary have maintained that to create an accurate historical record, the presentation should proceed without manipulation. Music in documentary can suffer from a bias in which the visual record is taken for historical record; this bias is compounded by the fact that if music is expected to contribute values considered emotional, its use may be considered an unnecessary, or worse, unethical, aspect of documenting events. In Glass’s (or Eisler’s) case, music is a powerful part of an account which may argue with presumptions of a secure documentary realism. Secondly, in regard to production, one materialist critique suggests that so as not to falsify its account of reality, the work itself must also issue from a non-authoritarian, or at least generally progressive, process of production (see, for example, the discussion of Barbara Hammer’s documentary on Ogawa Productions in Documentary Box #19 in this regard).

A Glass soundtrack which relies on the hearing audience to determine its meaning represents a re-working of these issues for film from the musical standpoint. Prioritizing the musical capacities of the filmic work has the merit of extending the engaged spectatorship of the more challenging documentary practices to the synthetic constructions of commercial film while simultaneously rescinding the compulsion in documentary or biographical film to impossibly fetishize the real after the moment of its occurrence has passed.

Still, music in Western thought or commercial filmmaking is most often considered incapable of depiction. So this auditory relief of the real in favor of audience interaction only begs the question of what, if anything, is to be interpreted in the work at all. If music helps constitute a non-hegemonic interpretive aesthetics, and at the same time, if musical meaning is non-representational or “merely” emotional —either outside history or not capable of expressing it—any work resting on musical interpretation, and any audience interaction with it, risks falling back into the category of the retrograde, the escapist, or the trivial.

Glass’s collaboration on Martin Scorcese’s biopic Kundun (1997), which dramatizes the Dalai Lama’s life and flight to India, is an instance of handling this problem. The film aims at perhaps the most difficult problem for realist biography: telling the story of a person who is revered as the embodiment of a divine being. Moreover, this sacred figure becomes, as a result of the crisis the film attempts to depict, the embodiment, for the Western audience likely to attend this film, of a national culture in the process of being extinguished from its geographical location. How does any actor act out the path to enlightened conscience? What does a cinematic spectacle of spiritual individuation mean for the mass audiences of cinema? How is Tibet to be located in contemporary world space? At the very least, given the West’s indifference toward the Tibet, the largely non-existent state of Tibetan filmmaking, and the very crisis of occupation and diaspora which the film depicts, Kundun might be seen as a response by Western artists to the neglect, exclusion, and violence which in fact make any filming of the Dalai Lama in Tibet unlikely. While the film’s aims are undercut by its own heady desire to resolve linear historical narrative with the realities of Tibetan culture in diaspora through the conventions of the biopic, the musical sequence depicting the climactic escape to India (to which Glass refers below) perhaps redeems the film in one important sense.

This sequence proceeds through a series of evolving and dissolving sand mandalas, along with fragments of dream or visionary sight. These formal and subjective elements all invoke repetition on several levels, interrupting the linear history that has structured the film until this point. The realist frame that had earlier worked to represent historical geography becomes a frame for representational and non-representational elements alike. Linear time is overcome by patterns of time emitted from essentially cultural identifications. That this overcoming occurs at the point of flight beyond national borders suggests that Tibet-in-diaspora calls on a heterogeneous historical subsistence enabling a transfer of its generative capacities beyond traditional geographical location. In a surprising way, then, the film equates a multiple logic of signification with the sacred in culture. We might say that, as Glass says of the Tibetan musicians who contributed to Kundun’s soundtrack, this sequence provides “a door into the work.” It asks us to consider what the combination of music and image can mean. One answer is given within the film, a biopic which can only become biographical on the condition that its own realist depiction of history is overturned: “non-violence.”


(James Tobias) JT: You compose highly textured, what’s been called architectural music that has many centers, many interlocking pieces, in which the audience has a very active role in terms of interpreting the structures, the forms, the emotional elements, the dramatics of the music. There are a million different ways I think of interpreting many of the pieces.

(Philip Glass) PG: That is very true. I think that’s a very accurate assessment of the dramatic position of the music. It’s very much the idea that the music is completed by the perception of the audience. I think that the essential feature of a modern aesthetic is the idea that works of art do not stand alone, but exist as an interaction between the perceiver or viewer and the work itself.

So when I go into a library and see books of music in the library, I don’t think of them as pieces of music. They become occasion for an event to take place. I don’t think of them as having an independent existence apart from the fact that they are being played and listened to. And I think that comes from Duchamp. And I think it was very, very expressed in a much more radical way by people like John Cage and Merce Cunningham when they worked together. But the idea is that there’s a distance, and I very often work with film music this way, that there is a metaphorical distance between the image and the spectator, and the way you cross that line, that distance, that becomes your interpretation of what you’re looking at. A lot can be indicated, but the final experience is personalized by the viewer.

JT: In your multimedia opera, The Photographer (1982), I hear the figure of a horse, Muybridge’s horse. I’m interested in this fact that it seems there is an image prepared there, that is very appropriate to the content. This piece happens to be about Muybridge, who is a photographer, who helps give birth to cinema. And I hear a horse galloping through the final dance in Act III of that piece.

PG: This is a very good example of what we’re talking about. There is no horse there—the horse is in your head! But then again, art is always in your head.

When Bob Wilson and I did Einstein on the Beach (1976), we’d have talks with groups afterwards. And very often the talks were people interpreting the piece for us. And Bob and I had no idea at all, we had no story in mind. In fact when we began working on it Bob said he wanted to pick a very well known person, and the reason is, he said if we do the story about Einstein we don’t have to tell the story because everyone knows it.

JT: The question of how music and images communicate is particularly important in film music, but it’s also important in music where viewers have an active role.

PG: The magic and art of what you do working in film, what I do, the art of it is implying these things. We don’t really tell the story. We point in a certain direction. You may go there and you may not. Composers want to know how to write music, how you get an idea, and I tell them—and they are a little surprised—don’t look at the movie too much. You start doing what Godfrey Reggio calls putting the bark on the dog. It’s not much fun for anybody anymore.

JT: A commonly used term in film music studies is “Mickey mousing.” Cutting to the beat. Making a character’s movements sound exactly as they look. And it takes some of the mystery out of the combination.

PG: Well, it takes us out of the combination. It leaves us as the spectator. You know the example of the opposite of that is anytime you see a commercial that is selling you something, basically there’s no room for any idea of your own. The strategy of the music and the image is to formulate one idea. And so you’re not supposed to have any other idea. It’s this kind of a dictatorial relationship between the image and music. And what we try to do in theater and music is to actually do the opposite, to leave a big space. The freedom of the imagination of the spectator is a precious part, an important part of the transaction. One of the reasons why we don’t like commercials is because it does that too us. It feels like you’re being manipulated. And you are! When you’re looking at works like this, I feel the space, and I encourage composers to look for that space. Of course, you may have a point of view of your own, and it’s important to have something—although John Cage indicated almost nothing.


JT: Can you tell us a bit about working on the Powaqqatsi (1987) and Naqoyqatsi (2002)?

PG: That probably became the model, one of the models of collaboration for me. I began in the late seventies with Godfrey Reggio on that piece. We had the great luxury of having a lot of time to do the film, partly because we didn’t have enough money to finish the film—there was no distributor, no one was waiting for the film. So we took three or four years.

JT: The film has since become a cinema classic. It has redefined documentary cinema in terms of the tradition of city symphonies, movies which explore geographies or urban landscapes.

PG: I see it in television commercials all the time—that fast moving stuff. That’s Godfrey’s stuff. With Godfrey I experimented to the point where in one case, I began writing the music before the film—people always assume that the image comes first, but it doesn’t at all. Powaqqatsi begins with a scene in a goldmine in northern Brazil, and I had images of that place that were filmed by Cocteau. And I wrote a ten-minute piece based on the images that I saw, and I recorded that piece and took it with us and the film crew down to Brazil, and when the cinematographer was making the movie, he had a headset and was listening to the music while he was filming.

JT: I’m interested in the score in the sense of the choral work. As frenetic as the pacing of the rest of the film as a whole, i.e. the images and the tempo of the music, yet there is something sublime and graceful about the [vocal] score, which in many ways accentuates the pace of the film, and also contradicts some of the frenzy in the visual. Do you write for the voice?

PG: Well, in this case, these are singers that are associated with the ensemble writing, and it’s different singing than you’ll find in the opera work that I’ve done. It’s much more instrumental, and I’m not setting words. So the voices are freed from the need to deliver meaningful sounds, in the sense that you know right away that you don’t have to understand what they are saying. It can be very annoying in opera if you don’t understand it. My view of it is that by being released from that particular obligation of making sounds of that kind, the voice is free to become, and they do have almost a kind of beatific sound to them for that reason.

JT: This movie is filled with contradicting messages in a way, and I think that part of the reason why it is interesting is because of that.

PG: I think you’re right, it is probably a classic movie in a way. It’s partly because every eight or ten years we reinterpret it. It looks different. Godfrey’s idea—and Godfrey’s ideas change also—was a film about the impact of technology on the ordinary life that we live. He thought that was what it was about, and I think to an extent that would still be true. But when people first saw it in 1980, most people thought it was a tripping movie, and they either got high or stoned and they didn’t pay much attention. And it wasn’t until seven or eight years later that people began to figure out the movie was actually about something. By the late eighties, people began to thinking about what that movie was actually really about, and then people began to think about it differently. We’re touring with this piece now—and of course, you have to remember that we did this in 1980—and there are a series of images of buildings collapsing, where people actually begin crying in the audience. Actually that scene of the fireman at the end was shot in 1978 when there was a brownout in Harlem, and the city was in a semi state of chaos, and the firemen were there trying to bring order to the streets. But when it looking at it now the film looks like it is about September 11th.

People say, aren’t you going to write any music about September 11th. And I said well actually, we did it twenty years ago. And the movie has that ability to be looked at. And I don’t know how Godfrey and I managed to do that. We weren’t that smart, we weren’t looking into crystal balls or anything. We were just dealing with material that was very potent, which was very visually strong. We were kind of lucky in a way, the way people can get lucky. Not that luck is the whole thing. But we hit our stride with that movie in a way. We didn’t know what we were doing, we didn’t understand it ourselves for years.

next >>

Philip Glass

Pioneer of minimalist theory and one of the twentieth century’s most significant artists. After attending the Julliard School he encountered the music of Ravi Shankar, which eventually led to the renunciation of his previous work and explorations in new musical directions. His prodigious body of work from the 1970s to the present includes works for orchestra, chamber orchestra, opera, theater and film. Also works in pop music (David Bowie, Aphex Twin), and his music continues to exert tremendous influence on succeeding generations of artists.


Selected scores for film and stage:

1982_ Koyaanisqatsi

1985_ Einstein on the Beach

1987_ Hamburger Hill

1988_ The Thin Blue Line

1991_ A Brief History of Time

1992_ Anima Mundi

1995_ Candyman: A Farewell to the Flesh

1997_ Kundun

1998_ The Truman Show

1999_ Dracula

2002_ The Hours