The Lost Art of Songwriting

Reading Evert Cilliers’ eccentric take on the history and condition of pop songwriting is the most taxonomical fun I can remember having since Joe Carducci’s Rock and the Pop Narcotic. Outrageous omissions bump up against blatant contradictions busily competing with absolute howlers of bad judgement. But the passion and sheer audacity is winning. 

One principle implicit in Cilliers’ argument strikes me as true. The principle is that volume wins. The more songs one writes, the more likely one is to write a classic. This makes sense, both in terms of probability, and of perfecting one’s craft over time. Irving Berlin, for example, wrote 1,500 songs. Is it surprising that a few of those are still with us? This is something I’ve thought a lot about, especially regarding the relative paucity of my own output.

I also happen to think that Cilliers’ loose and simple and imperfect definition of a great song is probably the best one we have. He defines a great song as a tune that a whole lot of people, over a long period of time, can’t seem to stop humming. He doesn’t think that lyrical, harmonic or rhythmic complexity matters as much as a sort of melodic “stickiness”. That explains a lot of things, including why dumb lyrics aren’t necessarily a barrier to a song becoming a classic.  “Wimoweh” anyone? (And let’s not talk about “Smooth Operator”…)

At the end of the essay Cilliers asks the question “Why aren’t there any great songwriters today?”, and attempts a few answers. One or two are plausible. But he misses what I believe is the biggest reason. I believe the decline of songwriting as Cilliers defines it is almost certainly related to the rise of the writer-performer, the self-contained rock band and singer-songwriter, and the abandonment of the division of labor between composer, lyricist, and performer.

Once upon a time, a composer teamed up with a lyricist. They wrote lots of songs, either on spec., or  on commission for a specific purpose: a Broadway show, say. But the job didn’t end there. They had to then find the right person to perform a given song. Not to mention other variables: the right producer,  arranger, band, studio, etc. A good tune performed well not only sold, it turned on other performers to the song. The more singers covered a tune, the more that tune became a rite of passage, a common “standard” by which to judge a performer. It’s not hard to see why this system resulted in what we now call the American Songbook.

Why did they do it this way? Why didn’t Leonard Bernstein just write his own lyrics to West Side Story, especially if lyrics aren’t the end all be all? Because Sondheim was so much better at it, and he was young and hungry and available, and  given the ruthless competition for the best tunes, Lenny couldn’t afford to chance it on his second rate lyrics. And of course the thought of either or both of them performing this material themselves in anything approaching a professional venue was inconceivable. After all, they would say, that’s what singers are for.

Pre-rock songsmiths had to write tunes which would appeal not just to many listeners, but to many potential performers. Diction had to be simple. Lyrical themes had to be broad enough for a range of personalities to put across convincingly. Why would a singer want to work with a tune they couldn’t add something to? The music had to be catchy and basic enough to attract a wide public, but sophisticated enough to engage or challenge a performer. This kind of tension resulted in some amazing sleight-of-hand. The next time you hear Dionne Warwick doing a Bacharach-David tune, for example, see if you can identify the time signature. Odds are it’s wacky, but you’ve spent your entire life hearing it as straight 4/4.

The writer-performer, on the other hand, is harshly constrained. Their musical options are limited to their voices’ range, the chords they know, and the accompaniment they can manage. Lyrically, they are thwarted by the irresistible temptation to write the most personal, particular, and poetic words they can get away with. It’s difficult to bring any sort of intriguing distance to one’s performance when singing one’s own words. Your Mom might be a brain surgeon, but you don’t want her operating on you. Your lyrics would likewise prefer to be treated by a disinterested professional.

Songs written for oneself, then, are almost by definition less flexible than songs written with no particular performer in mind. Songs written for anyone to sing must be more musically and lyrically universal than songs by writer-performers tend to be.

The reality is this: the odds that a single person or even a small group of people will simultaneously be first class composers, lyricists, and performers, are vanishingly slim. It was true pre-1962, and remains so, although it might take a few more decades for us to realize that the Lennon and McCartney were the apotheosis, not the advent, of a songwriting golden age. They are exceptions that prove the rule: it’s been mostly downhill for traditional songwriting since the early 1960s.

Then again, all of this might be poppycock. Maybe all we should be “lamenting” is the natural passing of a particular way of making and consuming and thinking about music. Maybe all we’ve done is turned a corner. Culture evolves. Maybe it’s as absurd to compare rap and metal and techno and reggae to Tin Pan Alley as it would be to compare Homer, Ziryab, Provencal trouveres, John Dowland, and African griots to Tin Pan Alley. Apples and oranges and bananas and pears.

Science and history show that human beings have to make music and dance. The kids will rock out to the most compelling sounds available. The sheer instrumental necessity of this trumps whatever qualitative judgements or historical-critical significance we would like to assign to a given era or movement. The fact that I might sympathize with Cilliers’ wistful glance backwards tells me more about myself as a singer and a listener than it does about whatever “truths” he might expect his argument to yield.