So, Robert Pinsky wrote a cool little piece in Slate the other day titled “In Praise of Memorizing Poetry – Badly.” In it he argues for a particular benefit to be gotten from misremembering a poem: that it brings into focus the choices that were made in the poem, the the consequences of using one word rather than another. He illustrates his argument with Yeats’s “On Being Asked for a War Poem,” which he presents like this:

“On Being Asked for a War Poem” 

I think it better that in times like these
A poet’s mouth be silent, for in truth
We have no gift to set a statesman right;
He has had enough of [something] who can please
A young girl in the indolence of her youth,
Or an old man upon a winter’s night.

He talks about misremembering the [something] as “glory” or “indolence” or “striving” before rediscovering Yeats’s original “meddling.”

In the case of “meddling,” the result of the exercise is to highlight the historical context in which Yeats was writing. Yeats was an Irish poet writing about World War I in 1915. At the time, Ireland was still part of the United Kingdom, and was actively involved in the war. However, some Irish nationalists used the war as an opportunity initiate a rebellion against English rule. And, in fact, the Irish War for Independence began pretty much as soon as World War I ended.

During Easter week of 1916, Irish rebels seized control of several key buildings in Dublin and declared independence from England. Yeats wrote a poem about it.

Yeats’s poem was written in response to a request by Henry James, and was originally titled “To a friend who has asked me to sign his manifesto to the neutral nations.” In all of this context, the choice of “meddling” seems to point to a degree of ambivalence towards the war, even presaging Ireland’s own neutrality in World War II.

Now, of course, all of this information is, in principle, available to anyone who has both the original text and access Wikipedia. However, for Pinsky, it is this forgetting, the substitution of “meddling” with “glory,” that serves as the catalyst for this particular close reading. And I doubt that, in the absence of some similar impetus, very many people would have focused on this particular aspect of the poem.

In biology, similar mistakes, in the form of mutations, provide one of our most important windows into the structure and function of biological systems. These mutations are sometimes the product of targeted mutagenesis, but can also result from naturally occurring mutations.

A lot of our coarse-grained knowledge of many systems comes from loss-of-function, or knockout mutations, where a mutation removes a particular gene, or renders it nonfunctional. For example, in 1976, Sharma and Chopra first described a recessive mutation in the fruitfly Drosophila melanogaster. Flies inheriting two copies of the mutation exhibited various developmental defects, the most obvious of which involved wing formation. So, the mutation, and later the gene, became known as “wingless.”

This is typical in genetics, where a gene will be given a name based on the phenotypic consequences of losing that gene. So, a gene required for wings becomes “wingless,” a gene required for heart formation might be called “heartless,” and so on.

Kim Jong Il relaxes with some brews. Due to the nature of the discovery process in biology, many genes wind up with names that are more like the opposite of what the gene actually does. This is sort of like how the least democratic countries always wind up with the word “Democratic” in their names, or how Citizens United succeeded in dramatically curtailing most citizens’ abilities to control their own government.

More subtle mutations, which alter the behavior of a gene or its gene product without completely eliminating it function, are more closely analogous to the misremembering that Pinsky is talking about, however. In a way, a knockout mutation of an important gene is more like just removing one whole line from Yeats’s poem, without regard for grammar, rhyme scheme, coherence, etc. What you would wind up with is a mess that fails in many ways, and is probably not terribly instructive – just like in biology.

Point mutations, which might alter a single amino acid in a protein, provide a more targeted and interpretable set of changes. Such a mutation might cause a small shift in the binding behavior of the protein, or might cause a slight change in the timing of the gene’s expression.

Like in the poetry case, these mutations are more likely to be revealing of the fine tuning part of the creative process, where mutations of small effect arise and are subjected to natural selection. In some populations – things like certain viruses, which have a very large population size and strong selective constraints – it might even be reasonable to think that these alternate, mutant forms have been explored and rejected by past natural selection. In other cases (e.g., large mammals, with relatively small effective population sizes), the most common form we find in nature might not represent some finely tuned optimum, but may simply be a form that works well enough.

Similarly, when we read a Yeats poem, we are inclined to assume that every single word has been chosen with extreme care, that a host of plausible alternatives were considered and rejected by the poet before he settled on just exactly the right word, in this case, “meddled.” I think we are inclined to agree with Pinsky’s final assessment, that “by memorizing his poem imperfectly, I had received a creative writing lesson from a great poet.”

However, a lot of poems in the world, even very good ones, are probably more like large mammals, with many of the word choices working well enough, but not necessarily representing some optimum, even a local one. (There is of course, the question, in biology and in poetry, of to what extent one can talk coherently about optima, but that’s a post for another day.) But this process, deliberate or accidental tinkering, is critical both to the creation of great things, and to understanding how greatness is created.

Sharma RP, & Chopra VL (1976). Effect of the Wingless (wg1) mutation on wing and haltere development in Drosophila melanogaster. Developmental biology, 48 (2), 461-5 PMID: 815114