begin page 74 | back to top


Must a Poem be a Perfect Unity?

This question, of interest once again because of the project of deconstruction, is the apparent subject of a curious little engraved prose work by William Blake called “On Homers Poetry,” which I quote in its entirety:

  1. Every poem must necessarily be a perfect Unity, but why Homers is peculiarly so, I cannot tell: he has told the story of Bellerophon & omitted the Judgment of Paris which is not only a part, but a principal part of Homers subject

  2. But when a Work has Unity it is as much in a Part as in the Whole. the Torso is as much a Unity as the Laocoon

  3. As Unity is the cloke of folly so Goodness is the cloke of knavery Those who will have Unity exclusively in Homer come out with a Moral like a sting in the tail: Aristotle says Characters are either Good or Bad: now Goodness or Badness has nothing to do with Character. an Apple tree a Pear tree a Horse a Lion, are Characters but a Good Apple tree or a Bad, is an Apple tree still: a Horse is not more a Lion for being a Bad Horse. that is its Character; its Goodness or Badness is another consideration.

  4. It is the same with the Moral of a whole Poem as with the Moral Goodness of its parts Unity & Morality, are secondary considerations & belong to Philosophy & not to Poetry, to Exception & not to Rule, to Accident & not to Substance. the Ancients call it eating of the tree of good & evil.

  5. The Classics, it is the Classics! & not Goths nor Monks, that Desolate Europe with Wars (E269-70).

One is confronted here at first with what appears to be a critical essay presenting an argument. In the usual argument, one begins knowing where one will end, that is, the arguer knows this and the movement and shape of the argument implies it. The form of the argument is enthymemic, as is characteristic of Blake in his treatises; and this supplies a certain amount of difficulty. Yet here something else seems to be going on simultaneously with argument, forming two lines of development. This I shall call dramatic search. It is connected here with the apparent instability of the word “unity.” The conclusion we reach may be foregone, as in argument, or come-upon, as in search. It may or may not be self-contradictory, as in argument. It may or may not be something satisfying to discover, as in search. In the exclamation of the last paragraph, one has the sense of a search pleasurably completed as well as an argument concluded. Very rarely, it seems to me, do Blake’s apparent arguments not have this parallel dramatic development. The two lines I mention suggest that Blake’s text is ordered by a sort of (dis)unity.

The search is one to constitute the meaning of “unity,” which keeps threatening dispersal of the argument, and to discover the consequences of notions of unity in human actions. We have the impression that in the course of dramatic search Blake discovers something for the first time, while if we are correct about the argument, we recognize also that we know its point from having read other works of Blake. It might be said, then, that the eureka is one of the revival of recognition rather than new discovery. For us, the conclusion is a surprise, in that to find that particular conclusion as a result of the problems initially having been posed is a surprise. In the usual argument, as in a debate, the reader begins knowing where he expects to end. In the meandering sort of argument infused with dramatic search, such as we seem to have here, the ending seems uncertain after all. Yet it is also not uncertain. As a cunning piece of persuasion it is certain; as drama and search it is not, except that there is something inevitable in the way any Blakean seed sprouts the same tree. This process is synecdochic, by which I mean here that the small issue with which we begin seems to have become the larger with which we end. In the case of Blake, we have the impression that no matter what the issue with which we begin the process will take us to this conclusion.

All the way along there are questions. The first of these is one unresolved in critical theory: is unity to be located in the work or only in the critical constitution of it, or somehow in both? The second is: what is this unity being mentioned? The situation is confounded by what appears (if we view the text as argument) to be an equivocation on “unity.” It appears that the first statement says that in order to qualify as a poem a work must have unity. But perhaps it says that the idea of unity is a hypothesis with which anyone reading a text begins (which is the way Northrop Frye has read it). This is to say that one searches the text for some principle of unity that will be appropriate to it.

It is as if Blake has been trying out ideas of unity with the Iliad as we enter this scene of search, continues in the first paragraph, and declares perplexity. The first unity he seems to propose is unity of plot. On the basis of this idea, at least as it is initially conceived, he concludes that the Iliad does not seem to have unity. Not only is part of Homer’s subject missing; it is a principal part. Why? Because the judgment of Paris ought to be, Blake thinks, part of the plot of the Iliad, for it is implicit in the unraveling, which involves the enmity of Athena against Paris and therefore Troy. Now the story of Bellerophon is not claimed by Blake to be inappropriate to the Iliad. Blake claims only that if one includes it one is hard pressed to explain why the judgment of Paris is not included. The Bellerophon story is told at modest length in Book 6 of the Iliad by Diomedes to his enemy in the field, Tydeus, and causes Tydeus to recognize a link of kinship. They agree to avoid fighting. The story explains why they do not fight, and it is also an example begin page 75 | back to top of Homer’s including something that fills out the complex web of relations that characterizes and guides behavior in the Hellenic world.

Blake mentions none of this, which could be used to explain its presence; but none of it would be relevant to the question of unity of plot. The episode seems to Blake less necessary to unity than the judgment of Paris, which Blake regards as an element of plot, presumably since it includes the motivation of a goddess to affect events that are told. It is therefore a “principal part of Homers subject,” that is, his story. Athena’s motivation stretches through the later events of the text and in that sense includes those events as the seed includes the oak. This appears to be Blake’s notion of unity at this stage of his essay. It is a causal situation that can be treated as a synecdochic unity, where the fragment implies the whole while still being a part. Blake’s own great poems are based on a synecdochic notion of unity.

Blake maintains this notion in his second paragraph, implying with “but” that what he next has to say is not always clearly recognized: the Belvedere Torso, a fragment, is as much a unity as the complete Laocoon. A fragment can implicitly project and thus include its absent whole. One thinks of the charming story told by Castelvetro of Michelangelo’s restoration of the lost part of a river god’s statue by quick study of the surviving fragment.11 Lodovico Castelvetro, “The Poetics of Aristotle Translated and Explained,” in Critical Theory Since Plata (ed. Hazard Adams), New York: Harcourt Brace Jovanovich, 1971, p. 151. Blake seems to think that the judgment of Paris is implied in the Iliad and ought to be there. In this sense, the Iliad is a fragment, like the statue Michelangelo restored, that Blake constitutes as complete. This completing act presents an interesting—and I think characteristically romantic—extension of the synecdochic notion that a part (fragment) projects the whole by bringing into presence by implication the absent part. The notion accounts in one way for the interest in fragments and tolerance of unfinished works in the romantic age.

But if this is the case, on Blake’s own principle (enunciated in the second paragraph) the Iliad may be unified after all. Blake seems to have restored to the text a unity he seems also to have denied to it. But is this unity the same unity to which Blake’s third paragraph refers? It does not seem so. The “unity” of paragraph three—unity as the cloak of folly—Blake definitely attributes to impositions on the text by certain readers of Homer. It is not clear whether this form of unity is really in Homer or not. That is, are readers who find this different sort of unity in Homer imposing it ruthlessly, or is it there to be found? From this point on, there are two meanings of “unity” at odds in the text. We have trouble deciding which one to apply: (1) a unity of synecdoche, (2) a unity which Blake seems to identify with a consistent moral allegory either imposed on or found in the text. Perhaps unity2 lurked all along in Blake’s first paragraph and we didn’t see it. Indeed, the first use of “unity” seems problematic when we reread the whole first paragraph, with its own “but.” In the second paragraph unity1 gains control, and it only appears to give way in paragraph three, where unity2 takes over and is apparently the imposition of moral allegory on the text. At this point, Blake’s text is unsettled and the uncertainty of the first use of the term reinforced.

The reason is that Blake’s introduction (with a certain violence) of unity2 forces us in rereading to impose the notion in the first paragraph, even as we had been invited to settle on unity1 as the only possible way finally to make sense of the passage. But this unity2 Blake abhors. It is for him imposition of moral allegory on materials that embody a different logic. The logic, Blake seems to believe, is misapplied in such situations and is viciously reductive. It converts interpretation into a witch hunt which allows the knave’s morality to prevail. This is why such notions of unity are the “cloke of folly.” The folly of misreading is cloaked by the appearance of moral rectitude. Imposition of unity2 turns texts into the moral precepts extracted from them, privileging precept over the minute particular.

Unity2 is associated with philosophy, unity1 with poetry; and, as Plato remarked, there has always been a struggle between them. In the fourth paragraph the text is further complicated by Blake’s introducing rule and substance on the side of poetry, exception and accident on the side of philosophy, reversing the classical locations and once again quarreling with Aristotle. Rule, now identified with poetry, must be the rule already applied in the text when unity was an acceptable term under its first meaning. This is the rule of synecdoche; substance becomes the unity of part and whole, or identity. This notion of identity includes both individuality and sameness; to traditional philosophy is relegated the either / or of difference and indifference and the necessity of a negating choice. In a companion piece called “On Virgil” (Virgil seems to have irritated Blake more than Homer, representing for him a more decadent form of the classic), Blake identifies unity2 with Grecian “mathematic form” and unity1 with Gothic “living form.” From that he proceeds to identify the classics with “war and dominion”—on the ground that “mathematic form,” being abstract, is like moral allegory, which leads to the attempt to bury everything under one law or negation: “Virgil in the Eneid Book VI. line 848 says Let others study Art: Rome has somewhat better to do, namely War & Dominion” (E270).

In observing the struggle that goes on in Blake’s search between the two meanings of unity we come to see that from the point of view of unity2, the sort of literary work of which Blake approved is not unified: meaning appears dispersed, variable, unstable, undecidable, and resistant to allegorical reduction. We could read the first paragraph from this point of view and be begin page 76 | back to top puzzled as to why the exclusion or admission of the stories Blake mentions are offered as evidence of the failure of the Iliad to measure up to unity2, unless we were to decide that Blake thinks a unity predicated on reason requires the presence of the judgment of Paris actually to demonstrate causal relations. But then Homer would come out all right in Blake’s view, and surely on the ground Blake offers in the second paragraph—the ground of synecdoche—for that concept of unity, added in, restores the judgment of Paris to the Iliad in its demonstrable absence. Why otherwise would Blake have noted its absence? The presence of Bellerophon is not a flaw. It is only that if it were missing we could not imagine its absence. Up to the third paragraph, Blake’s apparent complaint about Homer seems to turn upon itself to reveal the two faces of unity. At this point unity2 takes the stage, and we learn that it is the classical imposition of unity2 on poetic texts that is wrong. It is wrong because it is criticism imposing itself as reason on a work whose logic is synecdochic.

The romantics’ introduction of a new interest in the fragment as a work of art was an expression of unity1 against unity2. But, poor Homer! He is the victim of classical allegorizing toward unity2. Is he to be responsible for his interpreters? Blake apparently thinks so, and in “On Virgil” he simply condemns the whole classical tradition, which he identifies with sources in Babylon and Egypt, as allegorical: “Rome & Greece swept Art into their maw & destroyed it.” This domineering tradition, which imposes meaning “like a sting in the tail,” created the sort of tyranny and negation that has led and always leads, according to Blake, to war and repression.

If readers could read, that is, constitute a text purely as a (dis)unity1 then everything would presumably be all right. This is never entirely possible. If we are to talk about texts (i.e., converse over them, as surely Blake would want us to do) we must constitute them as well as recognize them and thus raise all of the problems that Blake dramatizes and argues over here. We must speak to some extent in a logic of unity2 (or, logic2), while yet respecting the text’s logic1 where we recognize it. One must say, therefore, that it is not merely a text that is deconstructed but our own constitution of a text as purely a piece of logic2. If this were all that deconstruction is then William Blake himself might well be placed among deconstructors. But deconstruction, with its radical rejection of the written as in any way connected with speech and its adoption of an infinite regress, at one time phrased as differance, detaches the text from any capacity to project human action. Theoretically, according to this view, there is no way for us to infer that, say, Browning’s “My Last Duchess” has within it a speaker and an auditor. Without this inference we are left only with the text, and there quickly rises to the surface a pure tropological substratum now unaffected by questions of who “speaks,” who listens, where all this occurs, etc. In such a text the tropological is certainly there, in the way that Blake seems to privilege the rule of synecdoche in works of art, but there is also the dramatic inference. From the burning fountain of pure trope to the icy regions of the extremity of logic2 there is a continuum, and the reading of a text is initially a placing of that text on this continuum, during which sensible decisions about internal intentions and so forth must be made. Some things in this process can be inferred, some things are uncertain, and some things are undecidable.

My sense of deconstruction is that in deconstructing logic2 it depends on that logic, continues to play logic2’s game, because in spite of its careful attention to tropes it can never posit, that is, establish as positive a logic1. It seems to me that Blake took tropes seriously as capable of constituting experience in a certain way, but not just tropes—also drama. Dramatic and synecdochic literality seems to me what holds Blake’s Jerusalem together (in its terms, not the terms of logic2) rather than diffusing the text endlessly. I acknowledge gratefully that deconstruction confirms me in my view that this is at the very least a very tricky matter. If deconstruction were to attempt to take seriously logic1, as a logic, or as I have called it elsewhere a mythic antithetical (to use Yeats’s term) logic, rather than drowning it in infinite dissemination, its aim, albeit ironic with respect to its own curious assumption that logic2 is really the only positive logic after all, would become an effort to engage in a positive conversation. This conversation would be about how logic1 uneasily contains logic2 in texts we have traditionally called literary and how logic2 even more uneasily (and this is what we have really learned from deconstruction) contains logic1 in texts we have thought of as philosophical or scientific. That there are different forms of narrative, dramatic, and argumentative progress, including intriguing mixtures like Blake’s, and that we can infer their natures with enough confidence to have sensible conversations about them and that these inferences enable us to arrest the flow of dissemination of meaning seem to me certain enough that we should go on making such inferences. This critical constitution of the text is more fundamental, even as it is more tentative, than the establishment of a determinate meaning, which threatens always to be an imposition of logic2 upon the text like “a sting in the tail,” the old romantic “allegory.” W.J. T. Mitchell notes the current anxiety about not being able to fix finally the meanings of texts “as if there were a time when we could.”22 W. J. T. Mitchell, “Visible Language,” in Romanticism and Contemporary Criticism (ed. Morris Eaves and Michael Fischer), Ithaca: Cornell University Press, 1986, p. 87. But this does not mean that the inferences of action and internal intention I have mentioned cannot be made or that the text shakes its finger silencing our conversation about it. The text is a potentiality for conversation. A text so constituted—and such a constitution is always temporary—comes into contrariety with other cultural begin page 77 | back to top objects, sometimes as a restraint, sometimes as liberation, always as an antitheticality. Antitheticality or contrariety resists romantic allegory and abstract law based on reason. It insists on the particular and exercises its ability to provide the other (but an involved other) in any cultural situation, any cultural moment always threatening the establishment of an external authority and the negation of freedom. But it is more than this resistance. It is also the ground of creation. Because it does not fix meaning according to logic2, it allows always for possibility, though its use will be likely eventually to die into a tyranny and require a repetition of the antithetical gesture, which is the gesture Blake makes when he dramatizes his argument.

Print Edition

  • Publisher
  • Department of English, University of Rochester
  • Rochester, NY, USA
    • Editors
    • Morris Eaves
    • Morton D. Paley
    • Managing Editor
    • Patricia Neill
    • Bibliographer
    • D.W. Dörrbecker
    • Review Editor
    • Nelson Hilton
    • Associate Editor for Great Britain
    • David Worrall
    • Contributors
    • Hazard Adams
    • D.W. Dörrbecker
    • Catherine Haigney
    • Andrew Lincoln

    Digital Edition

    • Editors:
    • Morris Eaves, University of Rochester
    • Robert Essick, University of California, Riverside
    • Joseph Viscomi, University of North Carolina at Chapel Hill
    • Project Manager
    • Joe Fletcher
    • Technical Editor
    • Michael Fox
    • Previous Project Manager and Technical Editor
    • William Shaw
    • Project Director
    • Adam McCune
    • Project Coordinator, UNC:
    • Natasha Smith, Carolina Digital Library and Archives
    • Project Coordinator, University of Rochester:
    • Sarah Jones
    • Scanning:
    • UNC Digital Production Center
    • XML Encoding:
    • Apex CoVantage
    • Additional Transcription:
    • Adam McCune
    • Jennifer Park
    • Emendations:
    • Rachael Isom
    • Mary Learner
    • Adam McCune
    • Ashley Reed
    • Jennifer Park
    • Scott Robinson
    • XSLT Development:
    • Adam McCune
    • Joseph Ryan
    • William Shaw
    • PHP and Solr Development:
    • Michael Fox
    • Adam McCune
    • Project Assistants:
    • Lauren Cameron,
    • Rachael Isom,
    • Mary Learner,
    • Jennifer Park,
    • Ashley Reed,
    • Adair Rispoli,
    • Scott Robinson
    • Sponsors
    • Funders
    • Blake/An Illustrated Quarterly
    • William Blake Archive
    • Carolina Digital Library and Archives
    • Use Restrictions
    • Copyright © 2015 Blake/An Illustrated Quarterly, all rights reserved. Items in this digital edition may be shared in accordance with the Fair Use provisions of U.S. copyright law. Redistribution or republication on other terms, in any medium, requires express written consent from the editors and advance notification of the publisher. Permission to reproduce the graphic images in this digital edition rests with the owning institutions.