With Pleasure: Gertrude Stein and the Sentence Diagram

After expatriating herself, Gertrude Stein stayed away for over thirty years. When she did return, it was to deliver several lectures, one of which, “Poetry and Grammar,” fondly recalls a favorite childhood activity, sentence diagramming.
xxxxxxxxxI really do not know that anything has ever been more exciting than diagramming
xxxxxxxxxsentences. I suppose other things may be more exciting to others when they are at
xxxxxxxxxschool but to me undoubtedly when I was at school the really completely exciting
xxxxxxxxxthing was diagramming sentences and that has been to me ever since the one thing
xxxxxxxxxthat has been completely exciting and completely completing. I like the feeling
xxxxxxxxxthe everlasting feeling of sentences as they diagram themselves. In that way one is
xxxxxxxxxcompletely possessing something and incidentally one’s self.
Diagramming sentences, for Stein, was not a matter of playing with words. It wasn’t even a matter of transforming the horizontal, linear sentence into a tree with multidirectional branches, though I am sure that pleased her a good deal. For Stein, “a sentence is carefully made” but a “sentence comes to be for use.” As such, the sentence, both “made” and “used,” reflects and subsumes its maker and its user. The sentence diagrams itself, asserting its will, leaving the parser with an everlasting feeling of complete excitement and complete possession. But if the parser sets out to diagram a sentence and the sentence ends up diagramming itself, what is the parser in possession of? Stein’s answer is “something and incidentally one’s self.”
How curious the activity of diagramming a sentence is. We endeavor to figure out how to use the thing we’ve made and we find ourselves both used and made by that very thing. Stein says this is the point—“a sentence should force itself upon you, make you know yourself knowing it.” And so, she implores us again and again to “think of a sentence,” and to think of it completely. But a sentence, despite what the grammarian might say, is not a complete thought, and the sentence diagram knows as much. Watch:

The history of the American parse tree, or colloquially the sentence diagram, begins not with the familiar Reed-Kellogg diagrams, but with S.W. Clark’s earlier balloons. Clark lifted this auxiliary sentence from Scottish poet Thomas Campbell’s “To the Rainbow,” a poem included in Emerson’s Parnassus anthology. “And when its yellow luster smiled o’er mountains yet untrod, each mother held aloft her child to bless the bow of God.”
The classical definition of the sentence, established by the Hellenistic Grammarian, Dionysus Thrax in 100 bc, is that it is a complete thought. Clark’s treatment of the sentence debunks the classical definition. First and foremost, a sentence is not a thought; it is grammatical construct that thinks itself into existence. “A sentence thinks loudly,” Stein insists, and we must be willing to think with it as it exists. “What is a sentence,” Stein asks, “With them a sentence is with us about us all about us we will be willing with what a sentence is.” A sentence concerns us just as it surrounds us, and if we are willing, thinking a sentence, for Stein, is thinking a self.
But perhaps more importantly, a sentence can never be complete. It is more and less than complete, or as Stein says, “A sentence must hold another. This is a sentence but not restful.” A sentence is restless and multiple. Clark’s diagram acknowledges that this sentence, as is the case with all sentences, exceeds and falls short of whatever it is that the poet might have hoped to say, and thus whatever it expresses is neither complete nor singular. As Clark writes, “Diagrams are figures so arranged that the Elements of Sentences may be placed apart from one another—and in such a position that their offices and relations may be indicated.” Or, to use Stein’s words, a sentence diagram provides “more grammar for a sentence,” more offices, more relations.
In this diagram, words, dangling as they do from the center of the sentence—held—are stored in two misshapen sacs. Outgrowths give this particular sentence balloon its unique shape. Relationships are formed that might otherwise be ignored—who is doing what to whom and when and where and why now have more complex answers, and words and phrases are shown to both be in possession of themselves and possessed by another.
But, you might protest, this sentence is clearly complex. Perhaps a simpler sentence might be a complete thought. Let’s see, shall we?

This example, from the more familiar Reed-Kellogg system, is straight-forward enough. “Very quietly” modifies the predicate; “very” modifies quietly; “the leaves” is the modified subject; “fall very quietly” is the modified predicate. “Quietly” is an adverb that modifies “fall,” telling the manner; “very” is an adverb that modifys quietly, telling the degree. This web of relations demonstrates that completion and containment, linearity and clarity, are contradicted by the very combinatory act of the sentence. To read any sentence with more than one word, we must move back and forth, leaping over words to bring others together. It is not as if these relations aren’t communicated in the sentence itself, but the diagram makes them visible in a way that they might not otherwise be. By isolating and then reconfiguring words, sentence diagrams make the often-invisible force of syntax visible.
Like the sentence diagrams Stein adored, her prose poems resist an easy integration whereby words and phrases are understood only in terms of their contribution to the “complete thought.” As such, they stand in stark contrast to the more conventional prose poem, wherein, to use Ron Silliman’s words from The New Sentence, “There is no attempt whatsoever to prevent the integration of linguistic units into higher levels. The sentences in conventional prose poems take us not toward the recognition of language, but away from it.”
Our Reed-Kellogg example, because of its simplicity, does not give us a full picture of what these grammarians offer. And while I will not go into great detail here, I will simply show you what I mean:

With their broken lines for coordinating conjunctions, their small x for what is not said but implied, their angled lines for participles, their hovering lines for direct address or interjections, and their tails for articles, it isn’t hard to see why Kellogg and Reid’s sentence diagrams take hold while Clark’s balloons float away in the distance. Beyond their reassuring symmetry, they offer a hierarchy that attempts to make even the most unwieldy sentence manageable.
Reed-Kellogg’s hierarchical system would have been the one with which Stein was most familiar. How could she have fallen in love with a system that seems to reaffirm the patriarchal nature of English grammar? Well, as Stein says several times in “Sentences,” “a sentence is very manly,” and, on some level, Reed-Kellogg’s system merely reflects this fact. After all, the very word sentence has a dubious heritage. It is the child of the Latin word for “maxim” and the twin of the other sentence, the one that denotes judgment passed on to a transgressor of the law. As much as Stein loved diagramming sentences, she left off diagramming them and started writing them.
But her earlier experiences with the sentence diagram offered her something she wasn’t seeing in the conventional writing of her time. When asked what she meant by her famous line, “a rose is a rose is a rose,” she replied, “now listen! Can’t you see that when the language was new, as it was with Chaucer and Homer the poet could use the name of the thing and the thing was really there…We all know that its hard to write poetry in a late age; and we know that you have to put some strangeness, something unexpected, into the sentence in order to bring vitality back to the noun. Now its not enough to be bizarre, the strangeness in the sentence structure must come from a poetic gift.”
No doubt, the sentence diagram returned a sense of strangeness to the sentence, but ultimately, it was too systematic, too rigid and hierarchical. In short, it did not come from a poetic gift. While the sentence diagram may have demonstrated that a sentence restlessly holds another, it did little to bring back vitality to written language. That was up to the poet. Stein occupied the sentence as a site of resistance, one that allowed her to challenge both normative sentence structures and patriarchal imperatives. She took up the sentence as a poetic unit, but she did so more often in prose than in lineated poems. As Silliman writes, “This continual torquing of sentences is a traditional quality of poetry, but in poetry it is most often accomplished by linebreaks.” In Stein’s prose, sentences are torqued, not by lineation, but by other means, some of them poetic, like repetition and rhyme, and some of them, like syntax and punctuation, belonging equally, if not predominately, to prose. Perhaps it was her early experiences with the sentence diagram that first suggested the pleasures of such work. Rather than a complete thought, the torqued sentence becomes a song that allows us to think with it, as it exists. “A sentence.” Stein insists, “With pleasure and with pleasure with pleasure.”

Clark, S.W., A Practical Grammar, (New York: A.S. Barnes & Co, 1866.), p. 62.
Reed, Alonzo and Kellogg, Brainerd, Higher Lessons in English, (New York: J.J. Little & Co., 1885), pp. 39, 48.
Silliman, Ron, The New Sentence, (New York:Roof Books, 1995), pp. 82, 90.
Stein, Gertrude, “Poetry and Grammar,” in Towards the Open Field: Poets on the Art of Poetry, 1800-1950, ed. Melissa Kwansy (Middletown: Wesleyan University Press, 2004), pp.290, 295.
Stein, Gertrude, “More Grammar for a Sentence,” in A Stein Reader, ed. Ulla E. Dydo, (Evanston: Northwestern University Press, 1993), p. 561.
Stein, Gertrude, How to Write, (New York: Dover, 1975), pp. 35, 125, 136, 140,162, 185, 211.
Wilder, Thorton, in Introduction to The Geographical History of America by Gertude Stein, (New York: Random House, 1936).