My review essay about Greil Marcus' The History of Rock 'n' Roll in Ten Songs is up at Public Books. Go there, read around; it's a great site with excellent contributors, and I'm pleased to have my work published there.
In the course of editing the original essay, which was 3600 words (ie. too long) when I first submitted it back in August, Stephen Twilley suggested the deletion of a major argument, one that drifted too far from the needs of the review. And he was right; the resulting review is stronger for the cut. I'm grateful to Stephen and also to Ed Winstead for their wise input.
But what good is having a blog if you can't wax poetic about outtakes?
The deleted argument centers on a question that's still important to me: Is music best understood as an event or a construction? Of course music is both a made thing, a constructed song of chords and words, and a performance that constitutes an event (including a boring event, like the time my friend Eric Nassau and I excitedly went to see Prisonshake, took heart that they seemed to be getting drunk before their set, and were shocked when they played too quietly…the only time I've ever experienced that: a rock band playing too quietly because they were thoroughly soused, but anyway….) So music is both, but do we get more out of music criticism that approaches music one way or another? One of the arguments I'm making in "Nothing Has Been Done Before" is that Marcus' book, and his entire viewpoint as a critic and cultural historian, emphasizes music-as-event while the majority viewpoint sees music as primarily a construction, which serves a consumerist culture incredibly well and strips away the humanity, the surprise, and the threat of music.
I wouldn't have thought to frame the question this way if not for a review by the music critic Simon Reynolds of Marcus' previous book, The Doors: A Lifetime of Listening to Five Mean Years, and while I disagree vehemently with Reynolds, I give credit where it's due. In his review, which appeared first in Bookforum in 2011 and then in expanded form on his own website, Reynolds argues that Marcus too often mythologizes songs by dramatizing them into discoveries if not revelations. This is the key passage:
A long-established Marcus technique is his knack of writing about a song as if it were a drama unfolding in real time, as though the band were discovering what the tune is about during its recording. This is fiction, of course: in most cases, songs are written and honed further through live performance long before being taken into the studio, where the band runs through multiple takes and builds up the sound through overdubs until the recording achieves the definitive and polished form that the world hears. Marcus’s odd insistence on treating song as events rather than constructions can get wearing.
I alluded to this in my essay at Public Books, in the following passage and a footnote at the end of it:
The dissent of The History of Rock ’n’ Roll in Ten Songs lies in its argument against this growing perception of music as primarily a construction and a product, something that is new but never very different. For Marcus, the event of performance is necessary to give the constructed song a life, but, more importantly, every performance contains the potential for a radically new effect.
My original thinking was that Marcus might have been responding to Reynolds' criticism in The History of Rock 'n' Roll in Ten Songs, at least to some degree. Maybe Reynolds framed the issue in a way that Marcus found helpful as he put together his book. But I really have no idea. When Marcus visited CCAD in spring 2013, he told me about The History of Rock 'n' Roll in Ten Songs as I tried to drive us out of a weirdly dense Columbus afternoon rush hour, but mentioned nothing about event v. construction or Reynolds. But then, as Reynolds suggests, this has always been a key issue in Marcus' writing, and it's a helpful organizing principle with which to think about Marcus' latest book—though you can see why, after this lengthy explanation, the argument was cut from the published essay.
In writing that essay, I've realized how important this question of event v. construction has always been to me. It's clarified what has always been the way I think about music, whether I'm making it, listening to it, or writing about it. My experiences as a musician have showed me that there's always a push and pull between the constructed song and the event of its performance. No single performance of the song is the same. I don't just mean that in terms of live performances, but as a way of thinking about recorded performances, too. The key is to consider not just the multiple versions of a song, but the many ways a song is listened to. How the song is heard is part of its event. And we forget that artists hear songs, their own, other musicians', and that this hearing informs their own songs, both in the writing and in the performances.
When I initially read the passage from Reynolds's review, I was bewildered and dismayed by his concept of how music is made. He makes it sound like the workflow of an industrial design firm: write the song, beta-test it, adjust, record, repeat. There's no small amount of condescension, intended or not, in the word "odd." Everybody knows, Reynolds seems to say, that songs are really constructions, and treating them as events is a dalliance that rational materialists will tolerate only so much.
There's a simple out for Reynolds here: "in most cases." It's a common technique in criticism; you allude to the corners but write the blanket statement anyway, to such an extent that the corners are ignored. It's not an attempt at a holistic understanding of music, one that really acknowledges the exceptions to the rule. And I think there are more exceptions than we might think. Yes, the process and qualities Reynolds describes are common, but even reading only Marcus' various books provides plenty of examples to the contrary. How often does Bob Dylan road-test songs before recording them anymore? In what way is Heavens to Betsy's Calculated "polished"? Where are the overdubs in the songs included in Harry Smith's Anthology of American Folk Music? There are multiple ways music is recorded, even today. Multiple goals, multiple angles on what a song should do, and multiple strategies to achieve those desired effects.
But my argument, one that I'll keep developing, is that music is fundamentally an event. It is the performance of a song that matters most. So even if the song is crafted the way Reynolds describes, it's still performed, and in the best, most thrilling cases, its performance creates the illusion, the suspension of disbelief, that the song just materialized, that it's a raw assertion from the singer, an act of will and imagination. This is how it can seem that nothing has been done before.
And if "event" sounds too heavy, too laden with importance, then call it an action.
The debate has its own merits, but it becomes compelling when applied to the question of why a work of art does or does not seem new. That's how I applied the event v. construction question to Marcus' book, which is very much about how rock 'n' roll can continually invent itself, and it's what I find most interesting as I do my own work. It seems to me that when we perceive music primarily as a construction, a product, it loses its ability to seem new. But that's on us. That's a result of our culture, our unlimited and all-encompassing consumerism. In my essay, I wrote that, "As the new becomes all, history becomes irrelevant—and history is crucial to understanding why, in fact, nothing has been done before." We are losing our sense of history. I'm convinced of this. The irony is that with history's dissolve, we're losing a sense of what it means to be new. Instead we only know what's next.
There are musicians out there, lots of them, who are still trying to make something new. This is the story I'm going to be chasing down over the next year or so.