Music Theory in Film Music

I've been trepidatiously getting my feet wet with Neo-Riemannian music theory since, for better or worse, it seems to have become a primary method for analysing film music. Unfortunately a straightforward, concise introduction to this subfield of theory doesn't seem to exist, so I've had to piece my understanding of it together from a few different sources. Neo-Riemannian theory, the dominant branch of so-called transformational theory, is named for Hugo Riemann, a late 19th-century German music theorist who came up with a system in which triads can be transformed into other triads by various standardised operations. It is most particularly applicable to late 19th-century music that stretches the boundaries of common-practice harmony (especially Brahms and Wagner, who stretched it in very different ways). Film music of the Korngold-Steiner-Williams-Goldsmith-Horner-Shore lineage has been seen to fit these transformational paradigms very well, which makes sense because these composers were influenced by the very type of music for which Riemann's theory was built.

I got started in this exploration by the use of this method in an article I stumbled upon in the most recent issue of Music Theory Online, Frank Lehman's 'Hollywood Cadences: Music and the Structure of Cinematic Expectation'. Lehman doesn't go too deeply down the transformational rabbit hole in this excellent article, as he focusses mainly on just a few typical cadences that show up frequently in Hollywood scores. He labels these by their Neo-Riemannian names, but a deep understanding of the method isn't required to understand the article. Another article by Lehman, 'Transformational Analysis and the Representation of Genius in Film Music' in Music Theory Spectrum, is another story entirely. This is a very dense analysis of James Horner's score for A Beautiful Mind, and to follow it you have to have substantial grounding in the theory.

This is where the problems start. How does someone, perhaps a Monteverdian who has recently developed an interest in film music, make use of such an analysis? And before one can make use of it, how does one make sense of it? Is it even worth the trouble? The answer to the last question is 'yes': it is indeed worth it. Too much film musicology never actually gets to the music, so any corrective to this lack is worthwhile. The same is true about studies in musical theatre, an issue to which Stephen Banfield devotes much space in a recent mostly damning review of the Oxford Handbook of the American Musical in JRMA. The lack of music in musicology is something that has been bothering me lately, especially as I've been preparing my teaching about musical theatre and film music. The focus is mostly on the words and/or narrative, where we really need to be doing what Banfield calls melopoetic analysis, looking at both words and music at the same time, or Michel Chion's audio-vision, exploring the sound- and image-tracks simultaneously. This also comes out of my music theory teaching, something I at first thought I wouldn't like but which I've actually been rather enjoying (other than dealing with voice leading, which is no fun for anyone).

So, musical analysis matters, but what about the Neo-Riemannian stuff? Lehman's footnotes pointed me towards an article by Richard Cohn in the Journal of Music Theory with the promising title 'Introduction to Neo-Rimannian Theory: A Survey and a Historical Perspective', but it is actually no such thing. It is only an introduction in that it introduces a series of articles in a special issue of the journal, and while it does give a history of how the theory developed it doesn't actually explain what it is. No use at all to the beginner, then. My next step was the Oxford Handbook to Neo-Riemannian Theories, which sounds good but is mostly preaching to the converted. Again, this large edited collection gives historical perspective and offers a few analyses without actually explaining what the theory consists of. You have to know the ins and outs of the theory for the book to be of any use; otherwise it's like reading a book in a foreign language knowing only the present tense.

I then finally found a chapter that might have been just what I was looking for: a chapter in the Oxford Handbook of Film Music Studies on 'Transformational Theory and the Analysis of Film Music', by Scott Murphy (author of an interesting study of James Newton Howard's music for Treasure Planet: The Major Tritone Progression in Recent Hollywood Science Fiction Films, which like Lehman's article on cadences uses Neo-Riemannian labels but is limited enough in scope that it doesn't get overwhelming if you're not familiar with the theory). The chapter is a bit more useful because it assumes little prior knowledge of transformational theory, but it stays too much on the surface and is really just a literature review, rather than explaining the theory and showing how it can actually be used. I need something with musical examples: here is what a Leittonwechsel actually looks like in a musical context.

The next step was to look backwards, to what is always cited as the article that introduced Neo-Riemannian theory, David Lewin's 'A Formal Theory of Generalized Tonal Functions.' Like many foundational texts, it doesn't have all that much to do with the subfield it helped create. The Neo-Riemannian side is only one of many possible applications of Lewin's system, which can also be applied to post-tonal pitch-class set analysis. The article is as mathematical as it is musical and is therefore very tough going. Sample sentence: 'To see the salient mathematical aspect of the conjugate relationship, notice that (C,1,4) and (C,1,9) are paired, and that 4+9=1.' (That last bit isn't quite as weird as it sounds, because pitch-class notation only goes up to 11.) So it was an interesting thing to read, but didn't help much with trying to grasp Neo-Riemannian theory.

Oddly enough, though, after having read all of this material that doesn't make much sense on its own it's all starting to come together in my mind. I do have some idea as to what this stuff is about, but I would still find it very useful to have a concise introduction, or at least a book that doesn't already assume familiarity with the system. The lack of such a thing is a general problem in music theory, and perhaps even musicology more generally (though I'm so deep into that by now that it's harder to notice the lack). When applying a theoretical (or historical, political, social) paradigm to music, we really ought to be careful not to alienate too many readers. This is not at all akin to 'dumbing down': theoretical rigour is still necessary and there is no reason not to assume that readers are generally well-educated and capable of understanding complicated things, but some good, clear introductory work is needed. I am constantly reminded of this when teaching; if I want to assign a reading on post colonialism in music I'd better make sure I've told my students what post colonialism is, or I need to be sure that I've assigned something that introduces the field. Susan McClary remains a model that we should all follow in this regard: whether she is writing on gender theory or modal theory she always gives a clear introduction before she undertakes deeper analysis. Likewise Eric Clarke with ecological perception theory in his Ways of Listening, and Peter Franklin with philosophies of Modernism in Seeing Through Music. These introductions are useful refreshers for those of us who already have the theoretical background, and they don't leave newcomers high and dry. With such broad concepts as gender, Modernism, and transformational or Schenkerian theory, each scholar will have his/her own version of what it means, so an introduction is doubly important.

I'm now going to have a look at David Kopp's Chromatic Transformations in Nineteenth-Century Music to see if he sheds any more light on this group of theories. Whether or not I actually use any of this in my own studies of film music (and I suspect I probably won't), I think this exploration into theory-land has been worth the trouble, in spite of the lack of a good map.

Some Clever People (Part One: Germans)

A few scholars and theorists have been coming up, or probably will come up, frequently in these pages, so it is probably worthwhile to take a moment and briefly explain why I find their work so compelling. In some cases I like their work a great deal and in other cases I take issue with it, and I am more familiar with some than with others, but I find all of the following very stimulating to read.

Theodor Wiesengrund Adorno (1903-1969)

German musicologist/sociologist/philosopher. Since I read his essay 'Types of Musical Conduct' (part of Sociology of Music) in my Masters year I've had an intense love-hate relationship with Adorno. He is known for different things depending on whom you ask: a musicologist will say that he is one of the most important theorisers of Modernism in music, while a sociologist will focus on his work in founding what is called the Frankfurt School of criticism. A philosopher will probably take note of his aesthetic theory, and a political scientist will see him as a major voice against fascism. For me, it's the way his Frankfurt School work illuminates his musical work that is most interesting. The Frankfurt School philosophers (who also include Max Horkheimer, Herbert Marcuse, and tangentially Walter Benjamin) were pessimistic about what they called the Culture Industry: they thought that art had been commodified and transformed into an industrial product to be bought and sold, and which was then used unintelligently by its bourgeois purchasers. Adorno labeled jazz as the worst offender, but his ramblings against jazz are the weakest aspect of his thought: he simply didn't understand it and he made no effort to do so. He really just used 'jazz' as a label for most things that he didn't like, which was an awful lot of things. The constant pessimism is very grating, but Adorno writes in such a way that I find myself agreeing with him even though I don't want to. His parenthetical writing style is both frustrating and engaging, and I'm sure it has rubbed off on me to some extent. In addition to 'Types of Musical Conduct', where he creates an elaborate typology of people by the way they listen to music, I love his Wagner book, which must be the most brilliant hatchet job ever written. Adorno hated Wagner even more than he hated jazz, honing in on the principle of the Leitmotiv, which according to Adorno is music commodified, turned into nothing more than a jingle. Adorno's favourites were Beethoven (of course), Berg, and early Schoenberg (before he adopted the 12-tone method, which for Adorno was an aesthetic betrayal because it was too strict, an attitude captured by Thomas Mann in his novel Doktor Faustus). Pretty much any other composer was weak, at least most of the time. Eventually I plan on tackling his Aesthetic Theory and I still need to read Dialectic of Enlightenment rather than just reading about it. For the musically-inclined I would suggest Sociology of Music, even though it is out of print (libraries should have a copy). This is a series of posthumously-published lectures, and since they were originally delivered orally they are a bit easier to understand than his written works. Adorno is unavoidable in musicology, and even Richard Taruskin, who claims to want to leave him behind, comes to many of the same conclusions and mentions him frequently (they are very much on the same page when it comes to to the early music movement, which was just beginning when Adorno was working).

Bertolt Brecht (1898-1956)

Like Adorno, another difficult writer with whom I often agree against my will. He was of course a playwright, probably the most influential of the twentieth century, but his more abstract theorising has been just as influential as the plays. Ever since his death theatre and opera production in Germany has been dominated by Brecht's principles. The much-hated (in the English-speaking world, at least) Regietheater stems directly from Brecht, though it is just as frequently a betrayal of his ideas as it is an exemplar. What Brecht hated most of all is what he called 'culinary' theatre, theatre that an audience consumes as they would consume a meal: without really thinking about what they were seeing. By forcing alienation (Verfremdung) via various devices such as projected titles, presentational acting styles, masks, and aggressively non-melodic music, Brecht gets his audience to think critically about what they are seeing. If you see The Threepenny Opera or Mother Courage and are moved by the characters' plights, you've seen a bad production, at least if the goal was to be Brechtian about it. Brecht's plays are extremely difficult to perform well, so strong is our will to identify with characters in the theatre or the cinema. When Brechtian techniques are applied to a play or opera by someone other than Brecht, it almost never comes off and, ironically, becomes culinary itself as the director ticks the Brecht-effects off his list. This is the problem with so much Regietheater: it becomes about how the director can force the text into the Brechtian orthodoxy rather than about a deep exploration of the text itself. Brecht isn't to blame for this; he himself wrote that Brechtian opera was probably impossible, and considered his own forays into the genre (Threepenny Opera, Mahagonny) to have missed the mark, though he blamed Kurt Weill for that rather than himself. He wasn't a very nice man, and he rarely gave credit where it was due: look carefully at the published scripts and you can often find 'Collaborator: Elisabeth Hauptmann' in very small type at the bottom of an introductory page. In the case of Threepenny Opera she was actually the translator and adaptor of the play, Brecht having only written the song lyrics, but he was happy to take credit for the whole, especially when it became a hit. His personality was so strong and compelling, and his self-mythologising so effective, that he was able to get away with it. They are fascinating plays no matter whether it was the empirical Bertolt Brecht who wrote them or 'Brecht' the author-function, but I much prefer the real 'Brecht' to his many latter-day imitators.

Next time I'll write about some Mediterraneans.

When is a musical not a musical?

In studies of film genre, the musical is nearly always presented as a case study for analysis. Scholars and critics have spent a lot of time seeking to define what constitutes a film musical, but the more I think about it the more I feel they are asking the wrong question. Rick Altman's The American Film Musical (1987) is a seminal text not only in studies of the musical but in genre studies per se, and his structuralist narrative categorisation is still used. In its most basic form, the musical according to Altman stages the formation of the heterosexual couple through music. This often takes the form of a plot that starts with a man and a woman who dislike each other, but as soon as they sing or dance together they fall in love. Every Fred Astaire-Ginger Rogers movie shares this narrative. In Shall We Dance (1937), for example, Astaire is a ballet dancer and Rogers is a tap dancer; their dance styles, as well as their personalities, seem incompatible but they discover in dancing together that they are in love. This plot was reversed in The Band Wagon (1953), with Astaire as the tap dancer and Cyd Charisse as the ballerina. Again, they don't get along until they dance together. There are plenty of musicals, though, that don't involve this narrative or even where it breaks down, but which everyone would still call a musical, so there is clearly something amiss with this categorisation (to be fair to Altman, he does recognise this). In Meet Me in St. Louis (1944) the heroine and hero never sing together in the final film (their only duet was cut). One could argue that this leaves a hole in the film, and the boy-next-door John Truett does always seem a bit of drip when compared to Esther. John never sings at all, and non-singing characters in musicals never seem fully-realised. In The King and I (1956) and My Fair Lady (1964) everyone sings but the leading man and lady never have a duet. In both of these cases, though, something stands in the way that would not allow a marriage to take place within the world of the film and, significantly, the world of the original audience (race difference, class difference, age difference). The lack of duets in these cases is actually using the convention against itself, the exception that proves the rule.

If we decide to define the musical based on this narrative structure (boy meets girl, boy and girl fight, boy and girl sing/dance together, boy and girl get together) we run into some troublesome cases. This is exactly the structure of David O. Russell's Silver Linings Playbook (2012), but not many people would call it a musical. (That said, I'm sure some producer somewhere is thinking at this very moment of adapting it for Broadway.) Why not? Though there is a lot of music in the film, it is all diegetic, 'real' within the world of the film. The characters never break into song or dance spontaneously, without being able to see/hear a band or recording accompanying them. Raymond Knapp calls this MERM, musically-enhanced reality mode. So let's re-define a musical as being a film in which MERM occurs. That way we include all of the couple-narrative musicals as well as the ones without that plot (Annie [1982] comes to mind). Yet there are plenty of film 'musicals' that do not evoke MERM. One such is Cabaret (1972); nearly all of the music (in the film, though not in the stage musical on which it is based) takes place within the cabaret itself, where it is entirely diegetic. Even though director Bob Fosse intercuts cabaret scenes with outside scenes, the cabaret can always be conceived of as going on in the background. Cabaret does not fit the narrative pattern either; Sally (Liza Minnelli) and Brian (Michael York) never make music together, nor do they get together at the end. So we need a different definition that will fit something like Cabaret. Perhaps a musical is best defined as a film that features a lot of musical performance, which can be based either in MERM or in reality. That sounds good at first, but the complications continue. Take the recent example of Inside Llewyn Davis or its stylistic prequel O Brother Where Art Thou, which do indeed feature many musical performances but which most viewers probably wouldn't consider to be musicals. A definition needs to be found that includes Cabaret but excludes Inside Llewyn Davis, so we need to think about the difference between the two films and their use of music. In Cabaret the music, even though it is diegetic, comments directly on the story (cabaret numbers as symbols of Weimer-era decadence), while in Llewyn Davis the music itself serves a plot function in that the characters are musicians but its subject matter does not do much plot-bearing in its lyrics or notes. Of course there is some connection; a song in one part of the film wouldn't work as well in another part, but without the songs Cabaret would have a very different plot (it would be like John van Druten's toothless play I Am a Camera, on which the musical is based) while Llewyn Davis would have the same plot (though it would be a pretty dull film). So can we define a musical as a film where music has an important plot function? That fits everything we've seen so far, but exceptions can be found. In some of the early Busby Berkeley musicals like Footlight Parade (1933) the music has no plot function at all, and  musical scenes are usually just tacked on at the end when we see the big show the characters have been working on throughout the film. Footlight Parade is about putting on cinema prologues, big musical stage shows that were performed in large cinemas before the movie in the early days of sound film. The musical numbers have no bearing on the movie's plot: the film is about the process of staging the numbers, not the numbers themselves, and the only music is whatever the characters happen to be working on at the time. Footlight Parade is more like Llewyn Davis than like Cabaret, but most would still consider Footlight Parade to be a musical.

I could go on searching for a definition of a 'musical film' but I think I've made my point: when dealing with generic categories there usually isn't a single blanket definition that can fit everything that it needs to. Genre is best seen as a constellation of definitions that overlap depending on the audience. And yet 'musical' seems a relatively straightforward category! In the film world 'film noir' is much more contentious, and the music world is chock-a-block with problematic genres. There's a very good article by Joti Rockwell that deals with some of these questions in relation to bluegrass:http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8711239. But if it's so problematic, what's the use of the concept of genre anyway? It is of course of great use: without categorisations like genre we couldn't talk to each other about much of anything. Think of the word 'tree', which is really a generic category. There are some things that we can all agree on as being trees, and there are other things we can all agree on as not being trees. The California redwoods are definitely trees just as Shall We Dance is definitely a musical, and a water lily is definitely not a tree just as Taxi Driver is definitely not a musical. Genre allows us to say things like 'New Zealand has a lot of trees' without having to spend hours saying exactly which trees are in New Zealand. If every different tree had a different name, if every pencil had a different name, if every piece of paper had a different name, we could never get anywhere because we would always be listing things. But even these seemingly foolproof categories like 'tree' can break down. Outside my front door is a plant that I assumed was a bush when I moved in, but in December it bloomed with bright red bristly flowers: my bush was actually a pohutukawa tree disguising itself as a bush. I still resist calling it a tree because it fit so well my previously-established category of 'bush', being a short dense plant with leaves close to the ground. It's the Silver Linings Playbook of the plant world. To sum up, I think that we should stop thinking of genres (or any other categorisation) as autonomous platonic concepts and start seeing them as socially-constructed concepts. Much more interesting than trying to decide once and for all what a musical really is is to ask why it is perceived as such. What is the history of the category? Where did it come from? How has it changed over time? I can't come up with a good concluding sentence, so here's what is probably the best of all the couple-forming scenes, Fred Astaire and Cyd Charisse in 'Dancing in the Dark' from The Band Wagon.