How Poems Work: Excerpt from "On ending a poem"

by Barbara Myers

See how many ends this stick has


Essay in excerpt

To come to the end. To stop. Not necessarily the same thing, as far as poems are concerned. In fact, a frequent criticism of a poem is that its stopping place creates a “weak”ending or one that “doesn’t work.” I stumbled into this muddy field recently in asking poet-editors to read and comment on a book I was working on. Critiques of endings dotted the pages, rarely the same view, occasionally even contradicting each other.

How to go about “fixing” a poem when the ending doesn’t “work?” Whose advice to follow, and how does that advice sit with the impulse that wants to express something true and real?

The reader-critic seems to say, I want the satisfaction of knowing the dimensions of this piece, that it isn’t partial, not a draft, and that the writer hasn’t walked away before sufficiently attending to the birth. Conversely, the reader may find that the poem ends too abruptly, or is too neatly tied up, a package with no openings to enter into.

It was Barbara Herrnstein Smith who made the observation that stopping and ending are two different things. Her book Poetic Closure was published in 1968 and is still the only full-length work on the subject. (It’s the sole referent on the entry in the Princeton Encyclopedia of Poetry and Poetics.)

Of a poem, she asks, “What keeps it going?” Then: “What stops it from going?” These questions, she says, suggest the close relationship between poetic structure and closure, the degree that a poem appears “closed” when it is experienced as an integrated piece, coherent, complete, and stable. The way a poem ends is part and parcel of its overall structure. And the degree of closure achieved at the end of a poem, she says, depends upon how the piece is structured throughout, both in its form and in its content.

Consider these last five lines from Keats’ “Ode on a Grecian Urn:”

When old age shall this generation waste
Thou shalt remain, in midst of other woe
Than ours, a friend to man, to whom thou say’st,
‘Beauty is truth, truth beauty,’ that is all
Ye know on earth, and all ye need to know.

Look at the structure of this poem, which Smith declares “the most securely closed poem ever written.” The “terminal features,” she suggests, include the use of a regular metre, as well as antithesis and allusions to death. Other signals of the end could include alliteration and repetition. On the question of content, the end of the poem is also marked by its epigrammatic pronouncement: “… all ye need to know.” Epigrams are definitely a sure-fire way to close off a poem (though not usually recommended).

At the opposite end of the spectrum is anti-closure, where poems are open-ended, lacking terminal features in structure, especially in content or theme. One of the bestknown proponents of this approach is American poet Lyn Hejinian. In her 1998 essay, “The Rejection of Closure,” she argues for dispelling any complacency that takes language for granted. In an “open text,” she says, “the writer relinquishes total control and challenges authority as a principle, and control as a motive.” Hejinian addresses herself to closure in general, not specifically to poem endings, but they can’t help but be implicated. Her definition (and criticism) of a “closed text” is one in which all the elements of the work are directed toward a single reading of it, without any “lurking ambiguity,” while in an “open text” ambiguity is desired. Repetition as a device, for example, “postpones completion of the thought indefinitely” and, as Keats said, the mind should be “a thoroughfare for all thoughts.” Keats, again. Repetition, again. Illustrating diametrically opposed ways of ending a poem. Smith agrees, to a point, that relatively weak closure can be the more successful kind, if the reader senses it is so because of the poem’s overall design. Otherwise it can be a failure resulting from a “deficient, defective, or spurious relation between the structure of the poem and its conclusion.” Open endings, as opposed to closed endings, “convey doubt, tentativeness, inability or refusal to make absolute or unqualified assertions.” They imply questions such as, “What do we know? How can we be sure we know it? They challenge the validity and even the possibility of unassailable verities, the moral or intellectual legitimacy of final words.” The degree to which a poem is closed off to the reader, how finished the work is, how much or how little room there is for anyone except the poet to participate, covers far more than how a work ends–but how a work ends reflects and clinches just how open or closed a work is. And like most of us writing and rewriting poems, I was thinking less of closure when it came to endings, and more in terms of successful versus unsuccessful. I was interested in having each piece cohere, in feeling that the poem satisfied itself. The terms “weak” and “strong” seemed too pejorative and vague.



an Arc Essay [read more essays]
Published in Arc 63: Winter 2010.
Skip to content