When I decided to put Mary Elizabeth Frye's famous poem Do Not Stand at my Grave and Weep to music, I had no idea what challenges I would face in turning it into a lyric. A casual reading of the poem cannot but move even the most hardened of hearts and the thought of turning it into a song got me excited. But poems are usually very different from lyrics. You could say they range in closeness from loving brothers and sisters to distant cousins that hate each other. In rare cases, a poem can double as a lyric, but rarely.
Lyrics have many more constraints than poems for they are meant to be sung and therefore have a whole new set of rules that govern their creation in addition to the ones for good poetry: songs have melodies and rhythms that are often repeated which require sections with matching prosody, musical accents should be in sync with verbal accents, consonant clusters are often hard to sing, certain vowels are awkward on certain notes, to name just a few.
Here is the original poem I decided to musicalize:
Do not stand at my grave and weep;
I am not there. I do not sleep.
I am a thousand winds that blow.
I am the diamond glints on snow.
I am the sunlight on ripened grain.
I am the gentle autumn rain.
When you awaken in the morning's hush
I am the swift uplifting rush
Of quiet birds in circled flight.
I am the soft star that shine at night.
Do not stand at my grave and cry;
I am not there. I did not die.
Before I set words to music, I first look at the rhythms and the overall structure. Rhythm is especially important for it is the one commonality that music and lyrics share. They are like two wings of a bird that when synchronized rhythmically enable the bird to soar.
Do Not Stand at my Grave and Weep is considered Frye's first poem. She wrote it to console a guest whose mother had died. She wrote from her heart succinctly and beautifully and its timelessness has been evinced by its propagation over seven plus decades the world over. But Frye's lack of knowledge of poetic or lyrical conventions shows up in the details where, as we all know, the “devil” resides.
In looking at Frye's poem, the first thing I noticed is that it consisted of twelve lines composed almost entirely of iambs; a metrical foot consisting of an unaccented syllable followed by an accented syllable. Nine out of the twelve lines employ this pattern: u / u / u / u /. Rhythmic repetition is often useful in setting words to music but three of Frye's lines have one or two extra syllables. Rhythmic variety is generally a plus as contrast in all art engages the eye, ear, heart and mind, but when the rhythm of words is haphazard with no differentiated sections and they go against the rules that were originally set in the first few lines, they become decidedly unlyrical. Adding one or two syllables to an established rhythmic pattern is not contrasting enough and ruins the flow and, hence, the potential emotional impact.
Fortunately, composers have poetic license when it comes to setting public domain works. The obvious solution was to eliminate words or syllables that weren't that important. The fifth line “I am the sunlight on ripened grain,” which had one too many syllables, was truncated by eliminating the syllable “sun” in sunlight. Iambs preserved. Problem solved. The seventh line, “When you awaken in the morning's hush” was shortened to “When you awake in morning's hush.” Same meaning, fewer syllables. And the tenth line, “I am the soft star that shine at night” was fixed by eliminating the unessential word “soft.” Now all the lines have a consistent flow.
At the same time I was fixing the prosody, I felt that the overall structure screaming to be implemented was: A B A—twelve lines divided into three sections of four lines each. The A sections would establish and revisit the key center while the B section would modulate to other key centers. Without this departure, the repetitive rhythmic structure would cause the song to stagnate and would get boring very quickly.
I also employed a musical conceit that I've used in a number of my songs: establish the root of the key in the first line or two and then don't come back to it until the end of the song. This creates buoyancy and a sense of journey so, when the song returns to the root, there's a feeling of arrival and resolution. In “Do Not Stand...” the root is used in the first line—2 measures. It doesn't occur again, except in first or second inversions, until the last word of the lyric, “die” after which there is a coda with the last few lines repeated where it returns on the very last chord of the song. It's like going hungry for a little longer than usual before you eat. Your meal is then that much more satisfying.
So now I had a general plan. However, there was an inconvenient anomaly that threw a monkey wrench into the works: all individual lines were complete sentences except the seventh, eighth, and ninth lines which, together, composed one sentence. Additionally, they straddled what I planned to be two musical sections which made it doubly awkward. Rarely (if ever) does a chorus or a verse end in the middle of a sentence. In this case, it did. This was my biggest challenge.
My options were either to nix my musical structure and write a through-composed piece, rewrite chunks of the poem to fit my musical structure, or find a compromise between the poem as it was and my musical structure. These are challenges one deals with when trying to turn a poem into a lyric.
The most important thing to me was for the piece to be memorable, so option one was out; through-composed pieces are rarely memorable. And, I preferred to not rewrite anything (other than the syllable omissions) for the piece had been recited and sung for 87 years in its original form and messing with it too much could bring haters out of the woodwork, so option two was out. So, I decided to compromise. The question was twofold: how do I end the musical B section in the middle of a sentence and where in the middle of that long sentence should I let the singer breathe and still keep the continuity?
The last line of the B section goes, “I am the swift uplifting rush.” It might be intuitive to decide to have the singer take a breath after the musical phrase where there is room to breathe but the downside is that it would imply a period to the listener who is not reading the lyric which would make the last phrase of the sentence, “of quiet birds in circled flight” feel like it starts a new sentence making it sound like a non sequitor. The next line, “I am the star that shine at night” already sounds like a non-sequitor no matter what I might do with the previous lines but at least it was a complete sentence without any additional problems.
So after experimenting musically, I found an unintuitive yet somewhat effective solution to my dilemma: breathe before the word “rush” which would be delayed a beat to make room for a rest for the singer to breathe and then add an additional measure to give the relentlessly repetitive rhythmic scheme a slight but needed variation and extra time to ramp up to the last A section. In this way, the word “rush”—despite being held out—would connect to “of quiet birds in circled flight” without a breath interrupting the thought therefore preserving continuity of meaning.
These solutions, born of compromise, helped the various song elements coalesce but I still needed an ending. I decided to repeat the last A section with the first two lines serving as an instrumental and then end with the most potent lines of the song, “Do not stand at my grave and cry; I am not there. I did not die.” But rather than tack on these four lines, I created an elision to keep it flowing. An elision in music is where the last note or measure of a phrase overlaps with the first note or measure of the next phrase. In this case, the word “die” (the end of the last line) occurs simultaneously with the word “quiet” (the downbeat of the first phrase in the last A section) although the word “quiet” and the following two lines are tacit. This elision not only keeps things flowing but it breaks up the relentless four bar phrases throughout the song in the opposite way the eighth line does; the line ending with “rush” adds a bar whereas the twelfth line leading into the instrumental takes away a bar.
There were other compositional/lyrical challenges that I dealt with in ways that made sense to me, but it's important to note that there are many ways to skin a cat. Other composers might find other elegant solutions to the structural eccentricities of Frye's poem. Like not buying the cat in the first place (I couldn't resist. Sue me.). In any case, they will all most likely require massaging the words and/or the music in order to find compromises that work. But songs are not the sum of their parts. They comprise hundreds of creative decisions that collectively take on a life of their own. Sometime flaws become so memorable or endearing that you wouldn't want it any other way. But it usually works better when “flaws” are written on purpose rather than unconsciously.
Musicalizing poems offer formidable challenges. Many composers fail. This is why most prefer to write or work with lyrics from the start. But when any worthy content is musicalized thoughtfully, the whole world can enjoy a “measure” of success.