Writing the American Ghazal

Since the publication of the late Agha Shahid Ali’s anthology Ravishing DisUnities: Real Ghazals in English (Wesleyan, 2000) and his own collection of ghazals, Call Me Ishmael Tonight (Norton, 2003), the traditional formal ghazal as helpfully defined for us by Ali has become a popular form, widely taught and practiced in poetry classes. Briefly, the formal characteristics are these: the poem consists of an unspecified number of couplets, each a separate “poem” with no obvious connection to the others (one metaphor is that each couplet is a separate stone in a necklace, beautiful in its own right, regardless of its placement); therefore, one couplet does not run on into the next; the lines are metrical and of equal length; both lines of the initial couplet, and the second line of each subsequent couplet, end with an unchanging refrain word or words, preceded by a changing rhyme (that is, the word or words in this position all rhyme with one another); frequently, but optionally, the poet “signs” the poem in the final couplet with his or her name or some reference to it. There are more subtleties than this, and anyone interested should read Ali’s anthology with its introduction and back matter. In his introduction, it should be noted, Ali appears to expect strict adherence to the formal rules, but in the anthology he has assembled he is far more forgiving of ghazals in English that follow only some of them.
     As for its content, Ali tells us, the ghazal is emotional, romantic, melancholy, often a poem of longing; a traditional metaphor is that it is the cry of the hunted gazelle when it knows it will die.
     Ali seems to regard the lack of connection between couplets (the “DisUnity” in his title) as the ghazal’s central characteristic. Before Ali informed us of the other formal dimensions, Adrienne Rich and Jim Harrison, among others, wrote fine poems they called ghazals that keep only this feature and the (unrhymed) couplet as stanza unit. Robert Bly has done likewise, adding the signature but, characteristically enough, omitting rhyme and refrain. It is these practices, as well as our typical mispronunciation (which he transcribes as guh-zaal or g’zaal) of the name that Ali sets out to chastise and correct. Here is his correction of our pronunciation: “ghuzzle, the gh sounding like a cousin of the French r, the sound excavated near unnoticeably from deep in the throat.” He must have known it was futile to hope that we Americans would get that initial consonant right, and those who have heard the news and abandoned guh-zaal have by and large settled for guzzle, as in what we do with beer. It’s a pity that the nearest thing to an English homophone is so inelegant, but there it is.
     For American poets, one path forward from Ali’s work is beautifully exemplified by the ghazals that appear throughout Marilyn Hacker’s Names (Norton, 2010). In addition to Hacker’s great formal skills, which allow her to adhere strictly to the formal rules and conventions without apparent effort, the book shows a high degree of engagement with the peoples, cultures and politics of the regions in which ghazals have flourished for centuries. There is little doubt that Ali would consider these to be “real ghazals in English.” Hacker signs the first and last ghazals in the book, the signature of the last addressing Ali himself: “Sháhid, if my name were Witness I would sign it./ I leave when the tired jokes about ‘Marilyn’ begin.”
     In Today (Sheep Meadow, 2008), her collection of 101 ghazals, Suzanne Gardinier is equally engaged with the cultural and political contexts of the ghazal, but observes only some of the formal specifications. She writes in discrete couplets; she ends each with a repeated refrain. She omits the rhyme that precedes this refrain, and does not use the refrain twice in the opening couplet. These decisions obviously give the poet far more freedom, but for me the plain arational charm of the form collapses. Some of Gardinier’s ghazals are quite powerful, but the effect of the rhymeless refrain is far less distinctive, much like that of our old friend anaphora, sent to the back of the line from its traditional place at the front. Another result of her formal choices is a degree of sameness from poem to poem that a different rhyme in each might mitigate. Gardinier never signs her poems directly, but has developed an interesting signature that might well identify her to those familiar with her work. Two poems end with a couplet containing the phrase “your gardener,” an obvious pun; many others use a parallel phrase: your shepherd, your stumbler, your listener, your collaborator.
     The differences in the formal choices of Hacker and Gardinier only hint at the range of possibilities that can be expected now that the formal ghazal is being practiced so widely. Few American poets bring to the form the degree of engagement that these poets bring. We’ll be seeing not only what Ali would accept as “real ghazals in English” but American ghazals that have been transformed as radically from the classic models as the American-born children of immigrant parents are transformed—to the consternation of their parents—by attending an American high school. I see no point in either condemning or applauding this development. It’s inevitable; it’s what happens here. For me, in fact, this is the most realistic and most interesting prospect for Americans: borrowing only the formal characteristics of the ghazal and filling it with American matter. I like the formal characteristics; I don’t know the tradition. To set out to learn it well would be a virtuous undertaking, but for me, late in life, it’s not going to happen.
     What I like most, I think—the source of the ghazal’s charm—is the way the addition of the refrain moves the rhyme away from the line-end while keeping its regularity. This is my understanding of why the ghazal penetrates, effortlessly, the resistance of so many students to ordinary end-rhyme. Whenever I’ve asked students to write ghazals on the spot, in class, one or two at least have come up with something delightful.
     It is time to disclose that my own interest in the subject comes from having written, in the last couple of years, many ghazals indeed. I had no idea, writing the first, the second, and so on, that I might go on for so long, but the poems kept coming. One decision I made, unconsciously at first, was to avoid, as best I could, sounding anything like Agha Shahid Ali, or like any of the classic masters of the ghazal I’d read in translation. It never occurred to me to go for any “Eastern”—coming from me, it could only be faux-Eastern—matter or manner. But I set out observing the formal rules as strictly as I knew how. I wrote discrete couplets with both rhyme and refrain, in metered lines of equal length, and I signed almost every poem with a form of my name or some oblique reference to it (I’m blessed with having my first name embedded in the name of my country). In what remains of this essay I want to describe some things I learned in the process.
     As the number of ghazals mounted and I began to consider assembling a whole book of them, my strict adherence began to waver as what my instincts led me to want to do for the poem-as-poem clashed with what I was supposed to do for the poem-as-strict- ghazal. The first time I noticed that I’d lapsed was when I fell into the Western habit—as Ali would have it—of writing poems as tightly organized and thematically unified as, say, the traditional sonnet. They observed the other formal strictures, but not this essential one. The best of the results, I decided, I liked very much. What would you do? I accepted them for what they were, and kept writing, sometimes observing various injunctions, other times not, but never without rhyme and refrain, and rarely without signature. This is my first observation: we are going to see many American ghazals in (metered or unmetered) couplets, with rhyme-and-refrain in their proper places and often the signature at the end, that observe the letter of Ali’s formal description—the outer shell of the ghazal, we could say—without that essential spirit: the distinct identity of each couplet. Some will deplore this development, I expect, and others accept it. But the ghazal is perfectly capable of telling a story, elaborating an argument, or embodying a dramatic monologue; the couplets even lend themselves to the ghazal-as-dialogue, with two speakers alternating stanzas.
     Working to establish connections, I toyed with the order of stanzas and began to violate Ali’s injunction against run-ons from couplet to couplet (which Ali’s introduction licenses, though only as the rarest exception).
     On the subject of rhyme, it is worth remembering that monorhymes—a ghazal is a monorhyme-plus-refrain—are uncommon for good reason. The refrain softens the effect, but a ghazal of eight couplets must rhyme nine times on the same sound without, if humor is not intended, any groaners or forced and unconvincing rhymes. Here, the rhyme is supported by the injunction that each couplet be a separate poem: if you can change the subject constantly, it’s much easier to come up with nine rhymes than if you have to stay on one. The handed-down practices of English verse, like widening the definition of rhyme to include various approximations, will be as useful here as in writing other Western forms. Another approach, however, is to embrace the predicament and allow the rhymes one is tempted to reject to bring in matter which one ordinarily wouldn’t touch, perhaps because it is gauche, or transgressive, or simply unacceptable to one’s ordinary poet-self, with whom it’s a good idea to be a tough negotiator.
     Fairly early in the sequence I ended up, without planning to, writing three or four light-verse ghazals, and found the external form, the shell, well suited to humor. The rhyme-plus-refrain, repeated often enough, has Ogden Nash-like potential for the intentionally ridiculous rhyme. Note that the most famous quote from Byron’s Don Juan works perfectly as the opening couplet of a ghazal: “But—Oh! ye lords of ladies intellectual,/ inform us truly, have they not hen-pecked you all?” (I admit to having completed a ghazal beginning this way, but my ordinary poet-self vetoed the result, and he was right this time.) Ali seems to have been sufficiently aware of this light verse possibility to include in his introduction an injunction against it—an injunction that, like most of his others, has little hope of being universally honored in America.
     One comic effect works with the reader’s own mental rhyming: when the rhyme proceeds from luck to truck to duck, is it going where I think it is? One effect arises when it does go there, and another, perhaps more interesting, when it doesn’t. Or if one takes on the dreaded poet rhyme, will we or won’t we get the standard know it and show it?
     Early in the process, the ghazals I wrote germinated in largely the same way that other poems did: for the first, I had a dream with a murderer in it, and started from the line, “Tell me this morning, how did that murderer find my dream?” But after a while, once I came to understand that I’d stumbled into a long-term project, poems more often began with just scraps of language, mostly a refrain plus a rhyme to which a few other rhymes cohered. I have many journal pages, Post-it notes and scraps of paper on which I’ve written “cell of ghazals smell of ghazals hell of ghazals well of ghazals” or “believe in leave in grieve in.” The arbitrary nature of such beginnings can have a positive value if it leads one away from one’s own conscious preoccupations and standard subjects, even from oneself, into territory one might not reach by more purposive approaches.
     The unrhymed ghazals of Rich, Harrison and Bly are exercises in association, or perhaps what Bly has called leaping poetry. Each couplet has a new subject, but a buried connection is assumed—there is some unstated, probably unconscious reason why couplet A gives rise to couplet B. Typically, then, the order of the associations would be assumed to have significance, however buried, resulting in a reluctance to change this order in revision. Ginsberg: “First thought best thought.” Once one lets go of this axiom, though, and acknowledges the role of rhyme, a given of the language, in initiating content, the order of stanzas becomes negotiable, and at times the source of much laborious revision.
     The ghazals in Ali’s anthology tend to have a rather long line, as do Ali’s own; one must, after all, try to get something like a whole poem into two lines. They are also, typically, not insistently metrical. These choices have the effect of lightening the impact of the monorhyme by placing the rhymes farther apart and outside of any conspicuously repeated metrical pattern. My own habits, for whatever reason, led me to shorter and more prominently metered lines, mostly pentameter and even tetrameter. The form is already a highly wrought one, especially with a metered line; the shorter the line, the more insistently one feels the rhyme-and-refrain, the rhythm and the artifice. Short lines give less room for narrative or image; the poems tend to become more rhetorical. For some, this damns them by definition, a conclusion that requires no thought; it is not inconceivable, though, that some poets might recognize and accept this tendency, then persist in order to see what they can do with it.
     Working with rhyme-and-refrain, one becomes aware of the way they form repeated units of rhythm at the end of each couplet and how these can dominate, for better or worse, the poem’s rhythmic presence. The refrain may be as little as a single unstressed syllable:

     He never meant to trot but she was hot to,
     he said. Her parents’ bed, the perfect spot to.

     He drank too much and lost his head (again),
     and as for the precautions, he forgot to.

Rhyme and refrain together might be as much as three stressed syllables in a row:

     Hang him from a tree he hasn’t hung from yet.
     Fling him off a bridge no one’s been flung from yet.

     Send succor, in whatever dark disguise:
     a hornet’s nest he’s not gone running, stung, from yet.

I find that these insistently repeated units, rather than the overall meter of the full line, tend to become the dominant rhythmic presence, and a source of variety when the same basic meter is there in every poem.
     In my own experience, one strategy for working with these effects is Frost’s basic trick of working American speech rhythms against the meter. Here (in a poem that clearly has a linear organization and couplets that are nothing like separate poems) I tried very plain (even clichéd and dated) speech in tetrameter lines like these:

     I was just an average Joe back then.
     I had no plans or dough back then.

     Family gone, no friends to speak of—
     I was feeling pretty low back then.

     I started hanging out too much;
     I had no place to go back then.

This ghazal does have a character, a persona; the ghazal lends itself as readily as any other strict form to the persona poem.
     It is perfectly possible, I’ve found, to use rhyme-and-refrain in a poem with a prose line, with the sentence or short paragraph replacing the couplet. The repeated final rhythm-units are one reason why the result still feels like a ghazal, only a bit more unbuttoned or stretched out.
     I adhered almost always to the convention of the signature, because I found that it tended to produce an interesting turn at the end. There are good reasons for signing few ghazals, as Hacker does; anyone who signs a large number will find it difficult to avoid making the same set of moves over and over. One may use one’s first name in various forms, one’s last name, a pet name or nickname, perhaps a pun or allusion. Another possible variation might be to sign the poem elsewhere than in the final couplet. (As the habit of signing grew on me, I found myself signing a poem that was otherwise nothing like a ghazal.)
     Sometimes the signed final couplet is just one more, in the spirit of the others, perhaps saving the best for last; sometimes it makes some final or summary statement to which the poet seems committed; sometimes it turns on the poet to admonish or chastise. It may present the signature in first, second, or third person, with significantly different effects: a statement by the poet, a statement to the poet, a statement about the poet. The couplet with a second or third person signature, while assuming the voice of some Other, remains the poet’s creation, and is thus ultimately a self-confrontation; writing many ghazals, one may be led to invent or discover many selves. The constant presence of assertions of the self may be rejected as terribly Western, or else embraced as a process of self-invention or -discovery. The poet who signs most of the time also creates the possibility of occasionally playing with the reader’s expectation of a signature by declining to meet it.
     There are some sounds for which the rhyming dictionary—writing many ghazals, one turns to it sooner or later—serves up just a few useable rhymes, and others for which it serves up very many indeed. The latter tempt the writer to go on and on, with one ingenious rhyme after another. Ali tells us that the effect of the ghazal as oral performance includes the audience’s waiting to see what the poet can possibly come up with next. This is both an opportunity to continue and a temptation to go on too long; for me it has generated some of the trickiest editing dilemmas.
     The writer of many strict ghazals intended to be published together must finally hope that readers will grow increasingly accepting of the need for this intricate dance, this insistent repetition—an acceptance that has for centuries been given without thought to the demands of the sonnet—as well as the occasional need to deviate, improvise or transgress. Their resistance to meeting so much insistent rhyme may soften with practice. To put it another way, the more we write, perhaps, the more we can get away with.
     Now that the form has become so widespread, and so many American poets have had time to get the hang of it, I suspect that there are a lot of interesting things out there, with all degrees of adherence or non-adherence to the rules. I see no point in insisting on either the strictest adherence or the varying degrees of its absence. I value those like Marilyn Hacker who are most able, as I am not, to write formal ghazals that carry on the traditions of the form as they emerged in its various homelands, but I do not think such capabilities are a requirement. Whatever one may think of the approach or approaches I’ve described here, it seems clear that the evolution of the ghazal in America will be interesting and instructive. Perhaps it is already time for a new anthology of (for better or worse) American ghazals.