On the Sonnet

by David Solway (August 2021)

A Dead Poet being Carried by a Centaur, Gustave Moreau, 1890


I have always loved the sonnet—the sonetto, or “little song”—with its traditional (though no longer canonical) fourteen lines, or quatorzain, as the apotheosis of poetic carriage, not only for its prosodic requirements but for its logical structure. Its history and practice show it to be one of the most exacting and yet adaptable of English poetic forms.

        The form originated in 13th Century Sicily in the innovative work of Giacomo da Lentino, a notary in the court of Frederick II, who added a rhyming sestet to the eight-line verse found in the common peasant songs, or strambotto, of the time. He is mentioned in Canto XXIV, line 56 of Dante’s Purgatorio as “the Judge,” an allusion to his court functions. Certainly, his judgment regarding the evolution of the musical strambotto was impeccable and, as it turned out, poetically momentous. The fourteener reigned for centuries as the principal orientation of the craft.

        The form was cultivated and slightly modified by Petrarch in the 14th Century and passed into the English tradition via the creative interventions of Thomas Wyatt and the Earl of Surrey in the 16th. As is well known, the two major paradigms—the Italian (or Petrarchan) octave-and-sestet and the English (or Shakespearian) three quatrains bound by a red ribbon couplet—each in its own way develops an argument according to a series of rules. In the Italian mode, the subject is elaborated in the first eight lines and qualified after a “turn” or volta in the latter six. The English sonnet ideally develops its theme through the first three units (a slight “turn” is sometimes detectable after the eighth line) and clinches its burden with a short rhetorical indentation. A third minor mode can be found in the intersecting, Venn-type rhyme scheme of the Spenserian sonnet, as in the Amoretti where the argument advances seamlessly toward the concluding couplet, or as in the most literate of English monarchs James I’s borrowing of the model.

        Such verse schematisms are ways of thinking. As Don Paterson writes in the Introduction to 101 Sonnets, the sonnet “represents one of the most characteristic shapes human thought can take.” Paul Oppenheimer in his magisterial The Birth of the Modern Mind: Self, Consciousness, and the Invention of the Sonnet succinctly defines it as “an instrument of self-reflection,” a vehicle for silent reading and quiet contemplation. In Oppenheimer’s view, the sonnet is not simply a respected verse contrivance but integral to the development of the Western meditative tradition.[1]

        Of course, not everyone—and certainly not every literatus—is on board. In a Dictionary of the English Language, Samuel Johnson described the sonnet as “not very suitable to the English language.” And in his Introduction to Collected Poems, William Carlos Williams was of the similar mind: “To me all sonnets say the same thing of no importance.” But then, Dr. Johnson was not a poet and Dr. Williams was a bad one.

        Readers of poetry will have their favorite sonnets and sonnet writers, of which there are a rich abundance in the Western canon, from the Elizabethan and Jacobean era to the present moment. Shakespeare’s Sonnets are clearly the gold standard. John Donne’s Holy Sonnets, though gnarled and difficult, are a milestone in religious sensibility, comparable in reflective power to his more famous 17th Devotion. The sonnets of Elizabeth Barrett Browning and John Keats are among the most affecting in the corpus. More recently, who could not admire poets like Edna St. Vincent Millay, Wendy Cope, Tony Harrison and Joseph Heithaus, whose wit, skill and love of the form are palpable?

        Plainly, to avoid redundancy and broach new possibilities, supple variations can be introduced into the logico-thematic apparatus (nod to Wittgenstein), as in Hopkins’ “curtal” sonnet, a mathematical reduction of the Petrarchan design (itself a close relative of the Golden Section), or Milton’s “caudate” sonnet exhibiting a set of extra rhymed lines,[2] or George Meredith’s envelope-rhymed sixteen-liners from Modern Love, considered quasi-sonnets for their thematic progression. (Harrison is notable for his “Meredithian” sonnets from The School of Eloquence.) John Berryman’s gasp-and-choke Berryman’s Sonnets tend to follow the basic Italian model and its multiple provisions—the syntactic clotting that violates the metrical norm is a purposeful expression of passionate emotion, “Crumpling a syntax at a sudden need,” as he confides in #47. His 18-line verses in The Dream Songs, for all their obscurity, break new ground, bearing a distant relation to the sonnet mode, “a stripped-down and rebuilt sonnet,” as April Barnard writing in Poetry Foundation described them. His hauntingly lovely envoi #171 seems almost Spenserian in its nimble, interweaving rhyme scheme.

        It would be remiss not to mention Wilfred Owen’s “Dulce et Decorum Est,” one of the great poems of the language. A double (28 line) sonnet, the unrhymed final couplet of the first rhyming with the initial two lines of the second, it is a masterpiece of both technique and feeling, sinuousness of form and beauty of diction accentuating the horror it portrays, pitting the ideal against the real.

        More recently, Yannis Livadis’ “fusion sonnet” with its plinth of seven extra lines, a half sonnet beginning with the first and ending with the fifth line of the original, has added to the sonnet thesaurus, as in the work of Bengalese poet Sonnet Mondal. (“Should I not learn to create being formally strong?” Mondal asks, rhetorically, in “Centuries of Creation.”) Other sorts of internal mutations, including breaking with classical prosody in conformity with English patterns of stress and accentuation, are possible as well as desirable so long as they comport with the alleles of the sonnet’s genetic code.

        It should be clear, then, that taken by themselves fourteen lines do not a sonnet make, since quantity can be accidental. More is required. Here I believe that Paterson erred in including such instances in his anthology, explaining that “they serve to show how fuzzy the definition is.” Ted Berrigan’s “justly lauded”[3] The Sonnets is a classic example of fourteen-line (sometimes fifteen-or sixteen-line) fractured ramblings that are not sonnets. (His repeated phrase “the sonnet is not dead” possesses a certain antiphrastic value, at least with respect to so deceptive a production.)

        Similarly, Katie Ford’s much celebrated but profoundly self-indulgent garland of 39 sonnets in If You Have to Go, detailing a failed marriage, have been called heartbreaking, but they shouldn’t be called sonnets, even if they are Meredith-themed. Fourteen lines minus evident rhyme, metre or cadence, syntactic order and recognizable “turn” may be attractive to some readers but they do not marry well to the chosen form. The same is true, mutandis mutatis, of Terrance Hayes’ acclaimed American Sonnets for My Past and Future Assassin, verbal receptacles stuffed with political hyperbole and fashionable racial complaint exhibiting, be it said, a certain raunchy exuberance—but they are not sonnets, not by any stretch of the literary imagination. Are Paul Muldoon’s 14-liners in Quoof sonnets? Hard to tell, despite the line count and the canny half-rhymes. Borderline maybe, but interesting, with a Seamus Heaney-like sound.

        Nevertheless, fourteen lines, or some mathematical or symmetrical variant thereof, is a basic or default property of the sonnet, which leads us to conclude that Shakespeare’s Sonnet 126 is not a sonnet but an odd little poem consisting of twelve rhyming couplets. Well, why not? Give the Bard a break.

        The sonnet production of two poets in particular has fascinated me for many years. Are Robert Lowell’s blank verse sonnets sonnets? The honest answer would have to be: No. They are 14-line hodgepodges of scrambled logic and grammar, featuring almost none of the sonnet’s constitutive attributes. Lowell worries in Notebook 1967-68 that he has failed to avoid the sonnet’s “gigantism,” but the gigantism is all in his own, often indecipherable, confessional self-regard. Similarly, are Canadian icon Milton Acorn’s Jackpine Sonnets sonnets? Again, the honest answer would have to be: No. They are, rather, a miscellaneous collection of ramblings of variable line lengths and numbers of lines, some collated prose paragraphs, others mere political manifestoes. Acorn’s riffing on prime numbers—13, 17, 21 and 25—as somehow representing “my free form sonnet” is amusingly flamboyant nonsense. Jackpines grow every which way with wayward abandon. Sonnets don’t.

        The next step would be to dribble fourteen words or letters down the page and call it a sonnet, if the definition is really that fuzzy. AdzTeal’s tacky Down the Dark Memory Lane…, self-described as “poems written by me when utterly bored. Don’t expect them to make much sense, especially in the rhyming part,” would then presumably answer as sonnets, given a penchant for elasticity. Muldoon tried one in Quoof, but it should be judged a mere trinket. I regard such trickle-downs as counterfeits, emanations or knockoffs.

        When all is said and done, one appreciates the poet who sets himself the task of adhering to the sonnet’s staple book of rules, hewing to the decasyllabic line (Shakespeare’s Sonnet 109 doesn’t, but we’ll let that go), the standard line count, a strict rhyme scheme (including full rhymes), and internal verse-and-discursive patterning, while at the same time striving to maintain a colloquial inflection and, obviously, avoiding the archaisms of poetic diction—the obsolete subject-predicate inversion (unless there is an ironic, recursive or aphoristic purpose, as in Frost’s “Something there is that doesn’t love a wall”), upper-case abstractions (except as common acceptations), and the démodé if once-handy contractions (o’er, ne’er, e’en, etc.) and preterite accenting (turnèd) which can no longer be used except satirically. (David Novak works the device in his 2000 Sonnets, a quaintness which detracts from their virtù.) Such exclusions make the poet’s job even harder in the effort to retain the metrical beat and syllable count which the classical sonnet favours and e’en demands.

        Of course, as we’ve seen, the modern sonnet can go in different directions, but the poet must be careful to guard against the risk of producing a generic freak. A good example of a freak sonnet is Simon Armitage’s take off on Shakespeare’s Sonnet 20, which runs di-di-di-dit dit di-dah-di-dit dah-dah-dah all the way down to the concluding couplet’s resonating di-di-di-dit di-dit dah-dah/dah-dit dah-dah-dah dah. Armitage is playing with beats and accents, not altogether accurately, in a cutesy attempt at fey avant-gardisme that leads nowhere. Just as an important aspect of invention lies not in playing indiscriminately with the external or sense-making forms of the sonnet but in working dexterously with and inside the framework itself, so, regrettably, one can no longer avail oneself of the toolkit of cheap tricks and artificial constructs in order to conform to the template, which plainly makes the sonnet-work more strenuous and challenging.

        This, I believe, is a pivotal reason that the sonnet is among the most difficult of poetic forms to master and domesticate. The compass is small, compression is necessary, tactical contractions and syllabic toothing-stones are taboo, syntactical liberties are sensitive, the rules governing structure and argument are intricate and severe. High modulus is the rule. Still, the whole must seem effortless, conversational, “natural,” reproducing the pulse of living speech yet preserving its own lexical and phonetic valence. “Invention,” said Sir Philip Sidney in Sonnet 1 of Astrophil and Stella, flees “step-dame Study’s blows”; rather, “look in thy heart, and write.” And yet the 108 sonnets of the compilation are held in place by a classical scaffolding—Italian and English—bringing in the “dictionary’s method” and “rattling rows” he affects to disdain in Sonnet 15. The two, Nature and Art, are compatibly wedded in the best sonnets.

        In some ways the sonnet reminds me of a Rolex watch; as horologist William May instructs the buyer seeking to determine authenticity, look at the writing: the engraving is convex and there is no bubbling. The same is metaphorically true of the genuine sonnet in our time: a precision-geared object that eschews a stiffly formulaic rendering yet remains technically unblemished. There is no bubbling.

        In short, the contemporary sonnet that honours its ancestry is a paradoxical enterprise, a rigorously codified ritual performance conducted, for the most part, in ordinary language and common syntax. It is enormously versatile even within the rigid parameters of its historical mintage, both austere and companionable, and capable of treating practically any subject under the sun. It is a little song with a big voice.

[1] Oppenheimer is an excellent sonneteer as well. He writes in his 2010 volume In Times of Danger:

          I need a form that I did not invent—

this shrewd eight-hundred-year-old slippery one,

its loose firm music that seems half-heaven-sent…

…like some good argument,

     this form is worn with rubbing bright as steel…


[2] Hopkins was also partial to the caudate sonnet. See his “That Nature Is a Heraclitean Fire and of the Comfort of the Resurrection.” “Caudate” means “with a tail,” “a coda”; “curtal,” of course, signifies “curtailed,” “without a tail.” The Golden Section is a proportional ratio of approximately 8:5, inherent in natural objects as well as architecture, painting and music.


Admittedly, “Heraclitean Fire” is a rushing, headlong, difficult poem to parse, an excessive sonnet par excellence, but its sentiment is profound. It’s interesting to compare Kim Addonizio’s (non-sonnet) “Heraclitean,” a pedestrian effort to deal with change and evanescence. Her concluding line Fish pulse slowly under river ice matches rather poorly with Hopkins’ masterful finale:


                 This Jack, joke, poor potsherd, patch, matchwood, immortal diamond    

                                           Is immortal diamond.


It is sobering to reflect how poorly the contemporary, with few exceptions, tends to stack up against the classic, ice against diamond.


[3] The epithet is from Andy Frazee writing in The Kenyon Review. A sign of the sonnet’s cachet is how the word itself is often conscripted to cover improvisations that bear no likeness to the sonnet’s anatomy and articulation. (To give Berrigan his due, there is one “sonnet,” LXX, that can stand on its own as a decent poem.) The Poundian urge to “Make It New” will often result in work that resembles Tracey Emin’s unmade bed—surely the kind of bed no sensible person would want to sleep in or make love in, or even contemplate, let alone buy. I would prefer Make It True. Making it new does not necessarily make it good. Paul Oppenheimer is on point when he writes in a sonnet titled “Making It Newer than New”: “what nonsense…to argue that the singer give up singing.”

Table of Contents




David Solway’s latest book is Notes from a Derelict Culture, Black House Publishing, 2019, London. A CD of his original songs, Partial to Cain, appeared in 2019.

Follow NER on Twitter @NER


Leave a Reply

Your email address will not be published. Required fields are marked *

New English Review Press is a priceless cultural institution.
                              — Bruce Bawer

Order here or wherever books are sold.

The perfect gift for the history lover in your life. Order on Amazon US, Amazon UK or wherever books are sold.

Order on Amazon, Amazon UK, or wherever books are sold.

Order on Amazon, Amazon UK or wherever books are sold.

Order on Amazon or Amazon UK or wherever books are sold

Order at Amazon, Amazon UK, or wherever books are sold. 

Order at Amazon US, Amazon UK or wherever books are sold.

Available at Amazon US, Amazon UK or wherever books are sold.

Send this to a friend