![[syndicated profile]](https://www.dreamwidth.org/img/silk/identity/feed.png){kind=small}
languagehat_feed) wrote2025-09-14 09:09 pm
The Dream Songs as Epic.
As I said back in 2014, John Berryman is one of my favorite American poets, and I welcome the imminent appearance of Only Sing: 152 Uncollected Dream Songs; Shane McCrae, who edited it and wrote the introduction, has a Paris Review essay about it from which I offer a few excerpts:
It has taken me years to realize that The Dream Songs is an epic—and a successful, even great one. For years, I searched for the successful traditional epic I felt certain must have been written by an American, and although I more than once encountered poems that seemed to fit the bill formally, none of them seemed an artistic success to me. Most often, they were let down by their language, which was commonly pedestrian, almost as if it were a secondary or even tertiary concern of their authors. But, of course, the language of an epic poem must be, in its way, as compressed as the language of a lyric poem—and in those moments when it is not compressed, the language must strike the reader as relaxed from compression, and loaded with the certainty of future compression. The language of The Dream Songs is always either compressed or suggestive of compression. The poem has this, and little else, in common with traditional epic.
But The Dream Songs also, of course, features a hero, as epics traditionally do—Henry. […] Henry, of course, is no Odysseus, though he more closely resembles Odysseus than all other epic heroes, with the exception of the unnamed protagonist of Dante’s Commedia (indeed, Henry strikes me as a combination of both heroes, but sitting in an armchair, sometimes a desk chair, at the end of a long day, talking, sometimes singing, sometimes shouting, in an otherwise empty room). Henry is an unheroic hero—a heroic hero has in-narrative effects upon the physical world and the people in it; Henry, for the most part, does not. When he does, the reader must take his word for it that he does; he, rather than the narrative of the epic, describes the effects he has. He is, in other words, a twentieth-century white American male, not especially remarkable, the sort of person who doesn’t establish or recover a nation, or parley with angels, or explore hell, but the sort of common person of whom nations are constituted, to whom angels were once commonly believed to minister in small ways, of whom hell was once commonly believed to be full. Henry is a hero for a disenchanted nation, from which once-common beliefs have mostly fled. He does not mourn the disappearance of those beliefs; he has held on to the beliefs he could. […]
In a 1968 interview with Berryman, Catherine Watson wrote, “Not all the songs about Henry are in the books, Berryman said, but ‘if there is a third volume, it will not take him further. It will be up to the reader to fit those poems in among the published ones.’ ” Berryman understood his epic to be complete, but he did not believe that its completeness could have only one form—although his remark does suggest that it has an established beginning and end; note the phrase, “fit those poems in among.” Only Sing collects 152 possible additions to the epic, each of which is worth reading for its own merits. […]
In November of 2023—on the anniversary, although I didn’t know it at the time, of the date on which Berryman wrote Dream Song 29—I flew to Minneapolis for a daylong visit to the Andersen Library Reading Room at the University of Minnesota. There, Erin McBrien, then the interim curator, located the boxes of Berryman’s unpublished material and patiently answered all my questions, and I photographed each of the manuscripts of the unpublished Dream Songs. The next day, I flew home and began transcribing the Songs. Doing so, I made no effort to Americanize Berryman’s spelling—he studied for two years at Clare College, Cambridge, and often favored British spelling—and I left the entirely idiosyncratic spellings and words untouched (one example of the latter: the word sieteus in the poem beginning “Hearkened Henry,” which perhaps ought to be she tells, but is, in fact, sieteus in Berryman’s typescript). I corrected only obvious typos. Once the Songs were transcribed, I had to determine how to arrange them, and I settled upon ordering them alphabetically according to first line. I could not organize them chronologically, because most of them hadn’t been dated by the poet and I didn’t want to guess—my goal was to impose as little of my own will as possible upon the organization of the Songs. […] Although it was Berryman’s practice, when collecting the Dream Songs into books, to group the Songs in numbered sections, I haven’t done so, as to do so would be to impose the will I’m trying to minimize. These Songs are put together in the way that I hope best allows—or at least allows as well as any other way—readers to “fit [them] in among” the already existing Songs, so that each reader might expand the epic according to their own wishes, thereby laying claim to their particular sense of what The Dream Songs is.
I’m trying not to add to my mountain range of physical books, but I may have to get a copy of this one. (I linked to a clip of Berryman reading Dream Song 29 here.)
