If a writer of prose knows enough about what he is writing about, he may omit things that he knows and the reader, if the writer is writing truly enough, will have a feeling of those things as strongly as though the writer had stated them. The dignity of movement of an iceberg is due to only one-eighth of it being above water. (Ernest Hemingway)

A stubborn after-image, which comes from all the previous modes of writing and even from the past of my own, drowns the sound of my present words. Any written trace precipitates, as inside a chemical at first transparent, innocent and neutral, mere duration gradually reveals in suspension a whole past of increasing density, like a cryptogram. (Roland Barthes)

Hemingway and Barthes have radically different styles. But do they stem from radically different ideas about the nature of writing, from opposing views about a writer’s problem? That’s what has been interesting me these days. While I often find myself citing Hemingway when giving advice to students, I am hesitant to invoke Barthes because his philosophical and literary sophistication can be daunting. Even in their use of analogy, as in the examples quoted above, Hemingway’s point is much more straightforward and easier to put into practice. Barthes is more likely to impress our students with what Robert Graves called “the huge impossibility of language”. Hemingway simply tells them to put their asses in their chairs and write what they know.

In my last post, I tried to show that the difficulty of writing prose is always relative to your ambition. If you want to write a paragraph that says that Hamlet loved his mother, you can, in principle, simply say it six times:

Hamlet loved his mother. Hamlet loved his mother. Hamlet loved his mother. Hamlet loved his mother. Hamlet loved his mother. Hamlet loved his mother.

This, I would say, is scholarly prose at “degree zero”. It is the minimum effort we can make as writers to say something we know. I also suggested what we might do to take it up a level by addressing the reader’s possible objections to our thesis about Hamlet’s feelings:

Hamlet resented his mother’s marriage. She had married his uncle. His uncle had taken his crown. But Hamlet loved his mother. Her betrayal hurt him. It broke his heart.

But this leaves a great deal under the surface, as it were. It is merely the tip of Hemingway’s iceberg, and scholars do like to be a bit more explicit, a bit more elaborate. They belabor the point, if you will, as a matter of professional pride. They don’t just tell you what they know (and presume that you know enough to make sense of it); they tell you also how they know (and therefore give you an opening to tell them what they’ve gotten wrong). They are also more upfront about what they think you think about their views, whether they are offering support, elaboration or a defense of their central claim. Consider:

It is sometimes argued that Hamlet’s pain stems from an ambivalence about his mother. To be sure, Hamlet resented his mother’s marriage. She had married his uncle and his uncle had taken his crown. But Hamlet loved his mother to the end. Indeed, his resentment does not belie his love but confirms it, just as jealousy is often evidence of a profound attachment. Her betrayal hurt him, but only someone we love can hurt us like this; only love can break our hearts. We must conclude that Hamlet was not ambivalent about his mother but suffered, precisely, because of his constancy. He was repulsed by someone he could not help but love.

Whatever you may think about its content, I hope you will agree that this paragraph is now more recognizably “scholarly” or “academic”. (Adding a few sources would cinch it, I think.) It makes explicit what the previous effort leaves implicit, namely, the reasoning by which we turn Hamlet’s apparent ambivalence into an argument for the constancy of his love for Gertrude.

I think Barthes would say that these reasons are already implicit in the simpler statement. If the reader is sufficiently steeped in the tradition of Elizabethan drama, then, given time, the “innocent, transparent” mind would soon dissolve the kernels of truth, rendering it cloudy, which is merely to say more “learned”. But scholars protect their statements from being dissolved any which way, which is to say, they don’t want to see them diluted, so they tie them to a particular past, they inscribe them in a particular code. As I noted in my earlier posts, they “situate the nature of their language in a social area”. If you want to take the argument apart, you now have to do it in a particular way. You have to respect the terms of the discourse.

Notice that Hemingway is not interested in the fact that the iceberg melts, but in the style of its motion. His scale is macroscopic, almost geological. Barthes would have considered how things dissolve, the chemistry of writing. He would put it under a microscope, approach it at the molecular level.

This is also why Barthes has so much more pull in the academy than Hemingway. “Papa’s” advice seems banal and always-already understood–in a word, unscientific. We want to say that we don’t need to be reminded about what writing is in that sense. We don’t want to “reduce” the problem of writing to merely being, say, knowledgeable and honest — which is one way to interpret the injunction to “write truly”. We want writing to be something much more subtle, much more sophisticated. We don’t want to practice; we want a theory. We want, I suspect, our failings to be understandable and even our failures to be commendable. It is not that we don’t know enough, we insist, or that we don’t mean what we say. It is that writing is difficult. But Hemingway is ready to grant us this too. It’s hard to tell the truth. But it is not impossible and we must never resent the difficulty. We must learn — and teach our students — to face it as authors. We must not let it be the death of us.