SOURCE: “Coleridge's “Kubla Khan” and the Fragment of Romanticism,” in Modern Language Notes, Vol. 96, No. 5, December, 1981, pp. 1035-50.

[In the following essay, Bahti examines the language and structure of “Kubla Khan” and notes that it is both a fragment and a whole.]

I wrote reflections that, in many ways, were even stronger than their origin.

—Derek Walcott

[Der] negative Sinn … entsteht, wenn einer bloß den Geist hat, ohne den Buchstaben; oder umgekehrt. …

—Friedrich Schlegel¹

When Coleridge's “Kubla Khan” appeared in 1816, the contemporary reviewers spoke of the poem's “nonsense.” This “nonsense” was immediately related to the ostensibly partial character of the poem: it was not wholly a meaningful poem, but only meaningless music; or else, Coleridge had dared too much, and therefore succeeded at only little, or even nothing at all, that was meaningful.² Even when the poem was soon judged very positively, the discussion remained within the confines of the question of partiality and meaning: “Kubla Khan” was so perfect because it was purely sensual music and imagery, and did not at all need to be more, or whole.³ In both cases the poem was considered as a fragment, while the possibility of one's understanding it laid claim to totalization. Either one could wholly understand it—but unfortunately there was no whole to understand—or one did wholly understand it, and that meant that one understood that it was not to be understood as a whole.

Nor do the later readings of “Kubla Khan” avoid this question of fragment and totality. One of the first great achievements of academic scholarship in romanticism (although widely surpassed today and condemned as misinterpretive) investigated the poem from the perspective of “source-study,” whereby J. L. Lowes valued it as a combination from parts of other texts, like a bricolage.⁴ One of the more recent, literary-historically more accurate studies understands the poem as a part of Coleridge's project for a new kind of epic (to be called The Fall of Jerusalem), but which, as a part, had already cancelled the whole of the projected epic: the “symbolic” history encompassing all ages is reduced to a visionary instant, and the two classical genres of the drama and the epic are reduced to the lyric—whereby E. S. Shaffer nonetheless still calls “Kubla Khan” an “epic fragment.”⁵ The more we know of this poem, of its sources and its author's intentions, the less we understand whether it is only a part or already a whole. This is particularly the case with the meaning of the poem: if we understand it ever better in part, then we still wonder whether there is a wholeness of meaning to it at all. The critic George Watson once said: “The fact is that almost everything is known about the poem except what it is about.”⁶

One could say the same of Coleridge. He stands as the fragmentary poet of English romanticism—perhaps, excepting Hölderlin, of European romanticism altogether—while a more precise overall interpretation of his oeuvre is still lacking. Rarely has one seen so many unaccomplished projects and unfinished texts: his writings lie there like a field of ruins and fragments. Yet within this “whole,” how does one characterize him? Is he mainly poet, or philosopher? Even if one does not deny the drive toward unity and totality in his poetic theories and speculative philosophy, one must concede that they remain fragments, and perhaps essentially fragmentary as well. But if according to the general English interpretation, Coleridge is essentially a poet—which means that he relates to particulars—and not a philosopher, this is often only the English prejudice against the “specious systems” of “empty” or “abstract” German idealism.

If in today's canon “Kubla Khan,” together with “The Rime of the Ancient Mariner,” is made to stand for Coleridge's poetry as a whole, so can every beginning student also say why the poem is representative of romantic poetry in general. Just looking at it superficially, one notices immediately the images, motifs, and ideas that today are held to be typically romantic. The Oriental setting on one hand, the emphasis on the sacred on another (“the sacred river” repeated three times, and also the ending)⁷ are commonplaces of romanticism, and here they are related to one another through the revised understanding of orthodox Christianity current around the turn of the century, when the “higher criticism” of Eichenhorn and Herder saw the Orient as an initial stage toward Western Christianity. The talk of “caverns measureless to man” points to romantic theories of the sublime as well as to the suprahuman as a familiar principle whereby romanticism distinguishes itself from a cliché for the renaissance (“man is the measure of all things”). The haunted, the Gothic, and the erotic (ll. 14ff.) often appear at this time, as does the animation of the earth (ll. 17ff.), which signifies not only pantheism, but also, much more far-reaching, a renewed kind of lyric: instead of the descriptive landscape poetry of the 18th century, there is now once again apostrophizing nature poetry in the sense of personification—one thinks of Wordsworth's Prelude, Shelley's odes, or many of Hölderlin's poems. The “romantic violence” of the second strophe—an uncontrollable outburst—may be easily related to the thematics of the French revolution, whereby Kubla Khan appears as the figure of a monarchical despot. At the end one notes the great estimation of the creative power of poetic “music” (ll. 45ff.), and also the adequation of speaking to seeing (l. 48), as typical of romantic and modern poetry: today one speaks of the English romantics as “the visionary company.”⁸ The figure of the poet as inspired visionary closes this highly romantic poem. If one adds to this the introductory note as well (we know now that its story of the creation of the poem is false, but what is more important is that the fiction presents itself for the reader as true), the representative character of “Kubla Khan” becomes even stronger: here one has the motifs of the illness (as later in the figure of the poète maudit) and solitude of the poet (as in the figures for the poet in Wordsworth, Hölderlin, and others); the “anodyne” as a narcotic (a type that persists from De Quincey and Baudelaire to today); the concept of “inner senses” (see Wordsworth on the imagination as “when the light of sense goes out,”⁹The Prelude 1805, VI, ll. 534-35); and lastly the image of composing poetry in the middle of a trance or sleep (which we recognize from Rimbaud and again from surrealism's écriture automatique).

I have bothered with this catalogue of commonplaces of European romanticism, not only to show how “Kubla Khan” can be taken as a part for the whole of romantic poetry, but also to be able to abandon such thematic remarks and analyses. For it is my opinion that working thematically with the question of fragment and totality in romanticism doesn't get one anywhere. This is above all the case when this question concerns the fragment or totality of meaning and understanding; it is then a hermeneutic and structural question, no longer a thematic one. To be sure, there is the word and image of “fragment” in the subtitle, the note, and the poem itself, but to understand this also means to interpret and to understand the whole poem and our own interpretation(s) as fragmentary or total. If one looks at the language and structure of the poem more closely, it may quickly be seen how many self-reflecting notions of part and whole, fragment and totality, come into play.

Later I will interpret the note more extensively, but first I would make just two introductory observations. Coleridge speaks of the composition of the poem as “the images [rising up] as things, with a parallel production of the correspondent expressions.” First one has images as things, which means that the images appeared and were taken as if they were things. To this are added the expressions corresponding to the images, where “parallel” is to mean as much as “simultaneous,” although the “expressions” only appear afterwards in the syntax of Coleridge's formulation. Thus one has expressions—the words of the poem—which correspond to images; and the latter appeared as things. The “correspondent” and “as” define a triadic relation of signification which is to be simultaneous, but in the “composition” (in the fiction of the composition) there were only two elements, and in the poem itself only one: in Coleridge's “sleep” there were only images and expressions, while the things were merely metaphorical (“as”); and now one has only the words of the poem “here preserved.” Although these words are to correspond to images, and the images appeared “as” things, there was and is no reference except for reference to the metaphoric in itself. The appearance of images is the semblance of things, and the “parallel” appearance of words is the correspondence to this “first” appearance.

This chain of metaphoric reference—or reference to metaphors—depends, as always with metaphors, upon apparently clearly distinguished categories of identity and of opposition (word and image, image and thing), which must at first appear single, separate, and distinct, in order then to become comparable and substitutable in metaphors. The poem itself begins as if various oppositions will be maintained and developed. Not only is the poem in its strophes clearly divided into three stages of setting, eventful narrative, and retrospect together with a wishful prospect, but the first strophe itself forms a pair of stable dichotomies. If on the one hand the setting of Xanadu is first described as the infinite, where there is neither spatial (“caverns measureless to man”) nor temporal (“a sunless sea”) measure, then on the other hand the “pleasure-dome” is the place of the finite, with spatial boundaries (“twice five miles … were girdled round”) and temporal categories (“blossomed” means seasons, and “forests ancient as the hills” introduces history). The clear opposition between “sunless” (l. 5) and “sunny” (l. 11) indicates the larger, categorical dichotomies which govern the opposition between Xanadu and the “pleasure-dome,” and the first strophe as a whole: the dichotomy between the infinite and the finite, and more precisely, that between the outside and the inside (“girdled round” and “enfolding”), and that between the hyperbolic and the defined.

Like the poem as a whole, the second strophe is itself divided in a threefold manner: it represents a sequence of fragmentation (ll. 12-24), repetition and attempted closing (ll. 25-30), and paradoxical reflection (ll. 31-36). But this division is not at all as sustaining and stable as that of the three separate strophes appeared to be. For what is fragmentation? In this strophe, as a scene of fragmentation, oppositions and dichotomies such as those of the first strophe are also split apart. The clear distinctions and oppositions introduced in the note and conditioning the whole poem are at stake here, and none more than the opposition between fragment and totality.

After the “deep romantic chasm” has run transversely and aslant through the pleasure-dome and has been personified (“As if this earth in fast thick pants were breathing”), this chasm becomes the locus of an outburst: from this chasm, within the whole of the chasm, “A mighty fountain momently was forced.” What now follows from this is one of our literature's most curious representations of an origin as the result of fragmentations. First, there is within this bursting-forth of the fountain (“Amid whose swift half-intermitted burst”) yet another outburst: “Huge fragments vaulted like rebounding hail.” Like the fountain appearing as a part of the whole of the chasm, this relation between whole and part repeats itself here through the word “Amid,” and indeed with the introduction of the word “fragment”: from the fountain—now construed as a whole—the “fragments” are thrown. We are at the fount of the fragments in the poem, or at their origin. One must now inquire how, in an ever-increasing self-reflexivity of the poem, this fount of the fragments is also the origin of the poem itself. For as this sequence of divisions develops further—whereby a part within a whole becomes a whole for yet another part¹⁰—these categories (of a part as something within a whole, a fragment as a part of a preexisting totality) invert themselves. With the repetition of the word “Amid” (“And 'mid,” l. 23), the third fragmentation is not one of a part within a whole, but rather a chiasmic inversion with the whole within the part: “And 'mid these dancing rocks at once and ever / It flung up momently the sacred river.” What until now has been a whole for the parts—the fountain for the fragments—is now, within these parts (“And 'mid these dancing rocks”), itself a part which at once produces and bursts a whole: the river; or, better, the river is produced in its bursting. And this happens precisely as the “fragments”—initially an undefined general concept—are identified as specific things (“rocks”): definitionally, as the whole of the word becomes a defined part. Thus, at the highest point of division and fragmentation, there is the production of that which the beginning of the first strophe already spoke of: the sacred river And thus one can say—finding once again a whole, or the beginning of a whole, within a fragment—that the poem and its first strophe begin “amid” the second strophe; they spring out of, or take their origin from, the fragmented fountain. Here there arises not just thematically the outer world (Xanadu) within the fragmentation of the inner world (the pleasure-dome), but in the perspective of structural self-reflexivity there also arises the possibility of the first strophe amid the second: only now, after its origin, is there the condition of possibility for a narration of the course of the river. This inverted temporality—the origin or the possibility of a beginning only after the beginning—is also noticeable in the temporal terms. The fragmentations are a series of instants (“Momently,” ll. 19 and 24), but amid these instants there appears one (“at once”) which suddenly becomes an infinity (“and ever,” l. 23). Once again, the possibility of a continuous temporality of the river springs forth from out of a fragmentation. Or does this not on the contrary mean that this origin-as-fragmentation perpetually remains just that, fragmentation, never achieving a fluid continuity?

What follows immediately thereafter is characterized by its slower rhythm, the calming repetitions of phrases from the first strophe and alliterations of m and r, as if the poem is now to “run onward” by way of a repetition of its beginning (see “ran,” l. 26). But is this the beginning of a narrative of continuity? On the contrary, the river runs toward “tumult”; and with the reappearance of an “And 'mid” (l. 29) this tumult is defined as a repeating present of the poem: the attempt at a whole comprehension or total representation of the narrative's time brings with it its own destruction, as one hears that the past (“Ancestral voices”) only leads to a prophesied future of returning tumult (war). The last third of the second strophe stays with this running river. At first there is yet another representation of a middle: “The shadow of the dome of pleasure / Floated midway on the waves.” This “midway” forms a centering between extremes—above (“dome”) and below (under the “waves”)—just as the following lines also represent a harmonizing (“the mingled measure”) of extremes (“From the fountain and the cave”: above and below, upwards and downwards). Even the “miracle” of the last couplet harmonizes extremes in the rhetorical form of a paradox (“sunny” and “ice”). Is this centering and harmonizing—a reestablished symmetry, even if a narrative standstill—the answer to the earlier splitting and fragmentation that were brought into the poem by the “slanting romantic chasm” and the various “Amids”? Is this fragmentation annulled and elevated (aufgehoben) precisely here into a new totality of symmetrical opposites?

The interesting thing in these lines is the “shadow.” For shadows are the inversions of reflections; instead of light thrown back, they are the interruptions of light thrown forward. But as such shapes they still refer back to their originals. Here the original is, thematically, the pleasure-dome itself, but as already mentioned, the centering and harmonizing of these lines is also the self-reflection of a figurative expression: the rhetorical figure of the paradox. This self-reflection occurs in the middle of the statement, “It was a miracle of rare device,” for “device” also means “devise” in the renaissance sense of a rhetorical figure. The paradox itself—“A sunny pleasure-dome with caves of ice”—is a familiar figure, one of Petrarch's favorite paradoxes, which often shows up in later European Petrarchism. But what does it mean that a possible narrative of the course of the river through the pleasure-dome should be brought to a standstill of rhetorical self-reflexivity, as if the poem finds itself in a restored totality and symmetry of rhetoric? To claim as much would be to succumb to the aesthetic seduction of these six beautiful, highly rhetorical lines. For in spite of all the symmetries, repetitions and apparently totalizing tendencies, the self-reflexive rhetoric is itself also a place of fragmentation.

In the subtitle the poem is called “a Vision,” and according to the note it is a part of a recollected and represented vision. One notices immediately that the last strophe mirrors this structure of the poem's genesis: as the “I” of the narrator appears for the first time, it is said he once “saw” something which he now wants to revive and represent. More striking is the parallel between the structures of the relations of signification in the note and in this last strophe. Like the poem that is to produce the vision of “images as things” through its “correspondent” words, here a poetic representation is intended to produce an image as a thing. Thematically, this refers to Xanadu—“That dome in air! those caves of ice!”—but in an entirely strict and literally self-reflexive sense it is also the poem “Kubla Khan”: the line just quoted repeats line 36 quite closely. Thus, to recollect, revive, and represent an image as a thing means to bring a previous image to consciousness—the paradox of the “miracle of rare device”—and to rewrite it again, or further. The figure of the narrator appears in this “I” as a self-reader; the “I” that has once seen a maid, and wants to revive and represent her song of the pleasure-dome, here names the poet who has “seen” (read) his own words in the first two strophes of his poem, and would now re-present them once again.

An interpretation of the structural self-reflexivity of the poem leads yet further into the thematics of this strophe. What the “I” saw and heard was “an Abyssinian maid, / And on her dulcimer she played, / Singing of Mount Abora.” Abora is the river of Eden, thus the fount of mankind's beginning, or in the strict sense: the origin (as if “ab-oraginal”). Similarly, Abyssinia—at this time being explored as the place of origin of the Nile river—was frequently used as an image of an origin.¹¹ Furthermore, one notes that each name begins with a and b, so that these names of origins include the beginning of the alphabet as well. Lastly, I would add that abba (a plus b) is the Hebrew word for “father.” Taken together, the two names and their significations refer to origins and founts, or more precisely, to a continuity between origins and beginnings, a successive, almost genealogical continuity, like that of the alphabet or a relationship between father and sons. Thus one can note here an attempted rewriting of the second strophe, where the origin arising from fragmentation only led to tumult and then to the standstill of the narrative. To remember, and to revive and repeat, the image of an Abyssinian maid singing of Abora, would mean to construct a double continuity between origin and image (what was there, and what is to be there again): like the maid's song of the origin, Coleridge's repetition of this poetic music is to represent the first, “original” images—those of the poem itself—as if they were real things (“I would build that dome in air / … And all who heard should see them there”).

The names' play of allusion to the alphabet refers back to the first strophe. For there, too, the first three lines of the poem played with alphabetical order: “Xanadu” is pronounced as z and there one finds the river Alph, which in Greek (alpha) as well as Hebrew (aleph) signifies the first letter. When one speaks of last things, then, one finds the first ones already included. Or put another way, to write and read words (those of the poem according to the note; those of the first two strophes from the perspective of the third; those of the “Abyssinian maid” from the standpoint of the “I”) means already to presuppose images (those of the vision and those of the “pleasure-dome” and the maid) and also images “as” things. Does this mean that the parts, like the letters, must already be understood within a delimited, finite whole? Or does the appearance of the words of the poem in the first strophe already presuppose their “earlier” appearance in the image of the singing maid as well as their later reappearance as actual, real images (“And all who heard should see them there”)? Ultimately, does this mean that fragments and fragmentation must again—or always already—be understood within a totality, so that “Kubla Khan” represents a dialectic of the apparent totality of the first strophe, the productive fragmentation of the second, and the restored, re-collecting totality of the third? And might this dialectic of self-reflexive writing and reading be the actual, proper meaning of the poem's self-reflexivity: one reads what one writes, and as a reading one produces positively what the writing itself had negated?

Here two doubting remarks may be added. In one of the poem's manuscripts (the so-called “Crewe MS.”) there is “Amora” instead of “Abora.” What is significant here is not the play on love (amor), but rather the allusion to Amara, which in Milton's Paradise Lost (IV, 281) names a false Abyssinian Eden. As also in the note's fiction, the “original” can be untrue, and its further repetitions and representations a mirroring of the false instead of a production of the true from out of its “negative” source or origin. The second point concerns the temporality of the repetition in this third strophe. It is unquestionable, not that something was there before—be it Coleridge's vision, or the pleasure-dome, or the singing maid—but rather that the temporality has been determinable and established up to this point: the whole poem up to line 41 is in the past tense, and indeed up to this point it is also self-reflexively past, that is, already written and read by Coleridge. The remainder of the third strophe, on the other hand, is in the conditional or the imperative; one could not rightly say that anything is present, for there is no present without the indicative. One can deduce the following from this distinction in temporalities. While the main part of the poem is in a narratable and readable temporal sequence—something was; now “I” am writing about it; “you” will then read it—the poem from line 42 onward is no longer narratable but only performable. The last thirteen lines are a matter of a possible writing and reading, wherein a first “reading” or recollecting—“Could I revive within me / Her symphony and song”—would be the condition of possibility for the further writing and reading, and not the inverse. Up to line 41, the whole poem was scarcely anything other than a perpetual self-reflexive self-reading and rewriting. Can the remainder now perform a further repetition? Only when a “reading” of Coleridge's revival of the “song” is performed—even if only implicitly—would the “images” and “things” be there about which he could write narratively. But such a “reading” would already be the production or the writing. One can always read and write backwards from line 41—as that which is past, the meaning to be achieved always remains “behind” one—but from line 42 on, one can only read and write forwards; which means that there is no meaning behind the lines which one could then represent or discover and interpret, but rather only a meaning that is still to be performed. Although the first five lines of this strophe refer thematically to the preceding part of the poem, thereafter Coleridge cannot read his own poem, but on the contrary, only the possibility of his writing. But this means not to read something, or to read a nothing. How could such a poem as “Kubla Khan” be at all possible? It is as if the “I” (Coleridge the poet or narrator) would only now write a poem about the pleasure-dome: “I would build that dome in air” presupposes that words could arise like images which are not yet there, and that is a question of the presuppositions of a reading which is also not yet there.

We stand before an ever-inverting mirror symmetry of writing and reading—to write further presupposes a reading, but in order to read something must first already be written—rather like the poem itself that first had to begin in order to arrive at the origin of its beginning. To play for a moment with a false etymology, is the “Abyssinian maid” the point in the poem where we came upon an abyss of sense? If we would understand this paradoxical symmetry, this standstill of self-reflection, we must avoid a premature, totalizing allegory of rhetoric. For if the narrative self-reflexively led to the unreadable and therefore unnarratable structure of the poem, so that in the self-reflection of rhetoric the possibility of narration was at once granted and taken back, the poem has still not arrived at a totality of its representation and meaning—even if as the representation of its non-sense.

If one now looks more closely at the note, a more precise image of the allegory of this poem and its mirror symmetry appears. On the one hand there is the story of Coleridge's failed attempt to write down the “distinct recollection of the whole,” on the other hand the counter-image of the self-restoration of the visions in the pool's mirror. After the interruption Coleridge retained only a “dim recollection of the general purport of the vision,” which means he fell from the “distinct whole” into an allegorical, more exactly, an aenigmatic part, for “dim” here indicates, almost like a technical term, the realm of the allegorical.¹² The “lines and images” left behind indeed point to the destroyed whole, but the movement from part to whole—from letters and words to purport or meaning—remains allegorically dimmed: Coleridge remains within a partial hermeneutic of the aenigma with written fragments which are supposed to signify a “whole meaning.” And when one stands (or has fallen) in an aenigma or allegory, the understanding of it remains similarly aenigmatic or allegorical.

Opposed to this, the counter-image in the other, adjoined bit of verse at first appears as the temptation of a restoration of a total understanding in the unifying of fragments into a renewed totality. But the allegorical fragments (again the word “dim”) would reunite not just into a whole reflection of the meaning of visions, but also—the allusion to Narcissus is unmistakable—into a mirroring of the observer as an image within the allegorical visions: and thus also, once again, only an image (a part) amongst other images. This idyllic counter-image is therefore just as reflective of partial character as Coleridge's own experience; the error of Narcissus was to take his image as an independent whole. When Coleridge writes in the note that some fragments were left behind for him, while the others passed away like disturbed mirror-images on water “without the after restoration of the latter,” he indicates that the parts—including those of this recollection or understanding—remain forever fragmentary, without so much as the temptation of Narcissus' error.

The reference to “images on the surface of a stream” reminds one of lines 31-32, where the shadow of the pleasure-dome floated midway on the waves of the river. We are now in a position to understand how this paradoxical centering remains in the middle of a fragmentation of attempted totalities. For as the shadow is an inversion of a reflection, its significance is here an inversion of its thematics: not the production of the pleasure-dome (as the meaning or referent of the shadow), but that of the rhetoric of the poem—an allegory of shadowy mirrorings. And when a fragment is precisely to mirror a missing totality, this means—in this logic of inversion—that the “totality” is inverted into a fragment. For in the “nature” (or rhetoric) of the case, mirrorings are always inversions. Ultimately it is the same with our fragmentary understanding of the poem, for to interpret the fragments—the various words and images which are to mirror a meaning—means to understand their inversions of meaning as totality into meaning as fragmentary, which then means forever to interpret their meaning in a fragmentary manner. As the adjoined bit of verse in the note says, the fragmented mirror-images “each mis-shape the other”: the conceptual pairs of fragment and totality, poem and meaning, text and interpretation perpetually mis-shape one another.

This is the case in this poem—but not only in this one—because it concerns rhetoric, by which I refer not merely to the specific rhetorical figures such as hyperbole, chiasmus or paradox, but rather to its misshaping or inverting organization of the fundamental structures of the true and the false, the original and the represented, the image and meaning. The poem is structured in a temporal sequence according to which there was first a vision and its “distinct recollection,” then the attempt at its representation; put another way, first came the images, then their verbal representation and narration. When this attempted representation and narration then fall into an allegory of rhetoric, one could perhaps still believe that the original would remain unrhetorical—or literal—in the conditions of its possibility. This would then in turn be the condition of possibility for a literal understanding of the allegory. But in the note, as the author falls ever further away from the original vision in the play of increasingly failing “recollection,” it is said that “the Author has frequently purposed to finish for himself what had been originally, as it were, given to him.” The important thing here is the “originally, as it were,” for the small proviso signifies that the allegory of rhetoric in the poem does not begin from a literal element, but rather takes its point of departure from an already metaphorical (“as it were” or “as if”) “origin”—or from a metaphor. As in the poem itself the origin of the river takes place only after the beginning, in a fragmentation and chiasmic inversion, the origin of the poem has “from the beginning” already rhetorically split or fragmented itself. Words are to be literal (“correspondent”) images of the original images, but since these “originally” original images are already figurative (“as it were”) expressions, the representational words of the poem are also always already unliteral, not wholly themselves, but rather rhetorically doubled and divided. If one begins from “metaphoricity,” one is already in a fragmented allegory of rhetoric.

This last inversion of the literal and the rhetorical, according to which the “original” that one was “wholly” to represent is already brought forth as rhetorically fragmented, also runs through the poem from the beginning onward. For when it is said, “In Xanadu did Kubla Khan / A stately pleasure-dome decree,” the word “decree” is already similarly divided. Let us recall the third strophe, where in the phrase “Could I revive within me” the performance of the narrative of the pleasure-dome and the narration of its performance came to a standstill together. On the one hand, “decree” here means to pronounce and to perform a decision: for example, a pleasure-dome is to be somewhere, and suddenly it is there, already performed and produced. On the other hand, “decree” means to separate, to distinguish and to decide: for example, the pleasure-dome should be itself and not something else. The two significations align themselves against one another as the possibility of a representation and the preceding judgment upon this possibility: if it were the case that a pleasure-dome is to be produced, then its representation and narration could also be performed; but this would also mean to decide what and whether a pleasure-dome would be at all. The crisis of the poem and its meaning is contained in this. To decide between “decree” as performance and “decree” as judgment is the same as to decide between the literal and the rhetorical, the original and the restored, thematics and self-reflexivity, or the already-narratable and the only-performable. It scarcely seems an accident that a “crisis” itself signifies such a separation and difference (krinein), or that the “device” of the poem—both the thematic as well as the rhetorical production of the pleasure-dome—itself already signifies “division.” For when one cannot distinguish and therefore cannot decide between two divided or fragmented significations, one then has a crisis: that of the separation or differentiation (the “original-partition” in Hölderlin's sense¹³) between judgment and meaning. And this crisis is as such—once again krinein—the critique in the middle of literary criticism itself.

Criticism encounters this crisis of self-inverting, but forever finds still divided parts of the text and understanding each time that it would speak of the fragment of romanticism. Romantic fragments—which, like Coleridge's “romantic chasm,” refer to previous totalities that are to be re-produced but at the same time invert the latter back into fragments through their mirrorings—never arrive at a critical understanding without mirroring criticism in its own crisis of fragmented meaning and divided judgment. When Friedrich Schlegel, in the famous Athenäum fragment 116, said of “romantic poetry” that it can “hover on the wings of poetic reflection in the middle [between the represented and the representing], forever re-empower this reflection and multiply it as in an endless series of mirrors,” this reflection—between totality and fragment, thematics and rhetoric—is exactly like Coleridge's paradoxical shadow, which floats mid-way on the waves, forever “empowers” itself in reflecting parts and images, but never arrives at the totality of a wholly self-mirroring meaning. This “perpetual alternation between self-creation and self-annihilation” (AF 51; see KF 28 and 37) of meaning is also, I suggest, Schlegel's concept of “negative meaning,” which is defined as “a presentiment or foretaste without a second proposition or postscript [Vorgefühl ohne Nachsatz]” (KF 69). This negative meaning of the romantic fragment and of romanticism in general does not belong to any understanding of the matter that could stand at an end-point, reach back and narrate; rather, it is always underway, pre-sensing or anticipating without being able to complete or fulfill—totalize—anything. Thus the title of the latest translation of Schlegel's fragments is perhaps a bit misleading: the romantic fragment points not so much to an absolu littéraire as to a lecture littéraire non-absolue.¹⁴ For this reason it hardly makes sense to speak of the “romantic fragment” or of “romanticism” as a historically-closed part or whole. Endlessly self-inverting and self-empowering readings are projected by romanticism's self-reflection and remain the fragments of its meaning. Romanticism itself thereby becomes the fragment of romanticism, and as readers who, in interpreting this meaning, themselves in turn become reflected fragments, critics and literary criticism are also the fragment of romanticism, parts left behind from a self-fragmenting allegory of meaning.

Notes

1. Kritische Fragmente, No. 69, in Friedrich Schlegel, Kritische Schriften, ed. W. Rasch (Munich: Hanser, 1970). Cited in the text below with the abbreviations KF (Kritische Fragmente) and AF (Athenäums-Fragmente).

2. Cf. William Hazlitt, The Examiner (2 June 1816) and Thomas Moore, The Edinburgh Review (September 1816), in Coleridge, The Ancient Mariner and Other Poems: A Casebook, ed. A. Jones and W. Tydeman (London: Macmillan, 1973), pp. 62, 65, and 74-76.

3. Cf. Leigh Hunt, The Examiner (21 October 1821), in Coleridge … A Casebook, p. 81; and John Bowring, The Westminster Review (January 1830), in Coleridge: The Critical Heritage, ed. J. Jackson (New York: Barnes and Noble, 1970), p. 550.

4. J. L. Lowes, The Road to Xanadu: A Study in the Ways of the Imagination (Boston: Houghton Mifflin, 1927).

5. E. S. Shaffer, ‘Kubla Khan’ and The Fall of Jerusalem: The Mythological School in Biblical Criticism and Secular Literature, 1770-1880 (Cambridge, England: Cambridge University Press, 1975), pp. 18 ff.

6. George Watson, Coleridge the Poet (London: Routledge and Kegan Paul, 1966), p. 119.

7. “Kubla Khan,” in The Complete Poetical Works of Samuel Taylor Coleridge, ed. E. H. Coleridge (London: Oxford University Press, 1912), I, 295-98. Subsequent citations of “Kubla Khan” refer to this edition.

8. Cf. Harold Bloom, The Visionary Company: A Reading of English Romantic Poetry (New York: Doubleday and Co., 1961).

9. The Prelude, ed. E. de Selincourt, 2nd ed. revised by H. Darbeshire (London: Oxford University Press, 1959), 1805 text, VI, ll. 534-35. Subsequent citations of The Prelude refer to this parallel text edition of the poem.

10. This is precisely the meaning of “half-intermitted burst”: the divisions further empower themselves through a splitting or halving of the middle, thus, “half-intermitted.” I thank Arden Reed for having called my attention to this.

11. Cf. Wordsworth's famous simile for imagination as a self-concealing origin, “like the mighty flood of Nile / Poured from his fount of Abyssinian clouds / To fertilise the whole Egyptian plain,” The Prelude 1850, VI, ll. 614-16.

12. I Cor. 13:12, for example, translates aenigmate as “darkly,” and an aenigma is called a “dark allegory” in the rhetorical handbooks.

13. I refer to Hölderlin's Homburger-fragment which today is called “Urteil und Sein”; he there understands judgment (Urteil) as the original partition (Ur-teilung) between being and consciousness.

14. L'Absolu littéraire—Théorie de la littérature du romantisme allemand, ed. P. Lacoue-Labarthe and J. L. Nancy (Paris: Seuil, 1978).