Volume 12 · Issue 2

THE ORDER OF NIGHTS VIIa AND VIIb IN BLAKE’S THE FOUR ZOAS

JOHN KILGORE

The presence of two seventh Nights in the perplexing manuscript of Blake’s The Four Zoas is a scholarly puzzle which has two theoretically distinguishable aspects. First there is the editorial problem of finding the least imperfect arrangement of the two Nights within the text of the poem; then there is the interpretive problem of deciding what imaginative place each Night has in the poem as a whole. The two problems really overlap, of course, for critical issues arise in editorial debate as “internal evidence,” and responsible interpretation must be consistent with a factual understanding of the manuscript. Even from a primarily interpretive standpoint, moreover, it is occasionally more illuminating to sift through the hectic evidence of Blake’s deletions, additions, and revisions in The Four Zoas than to treat the present text as if it were a fully intentional work. The study of the manuscript can itself become a kind of interpretation, for part of the significance of the poem resides in the story of artistic growth and struggle which the manuscript obliquely reveals.

Yet there has been a certain tendency to divorce textual scholarship from critical interpretation in the study of The Four Zoas, and in the debate over the VIIa-VIIb problem in particular.¹↤ ¹ An exception to this general tendency is Morton Paley’s Energy and the Imagination: the Development of Blake’s Thought (Oxford: Clarendon, 1970), chapter 6. See pp. 160-61 and 263 for vary succinct discussions of the ambiguity of the two Nights seven. A similar approach is taken by E. D. Hirsch, Jr., in Innocence and Experience: An Introduction to Blake (New Haven: Yale Univ. Press, 1964), chapter 7. In emphasizing the changes and inconsistencies involved in Blake’s “development,” however, both studies tend to exclude any sense of final coherence in the structure of the poem. All references to The Four Zoas are from David V. Erdman, ed., The Poetry and Prose of William Blake, 4th ed., rev. (Garden City: Doubleday, 1956). In its commitment to a synoptic overview of Blake’s entire canon, its preoccupation with the basic structural principles of his symbolic system, what one may call the main critical tradition in modern Blake studies has had very little to say about the manuscript of The Four Zoas. Prior to David Wagenknecht’s fairly recent discussion of the two Nights seven in Blake’s Night: William Blake and the Idea of Pastoral,²↤ ² Cambridge, Mass.: The Belknap Press of the Harvard Univ. Press, 1973. See pp. 225-26, 298-302, and for a critical reading which takes textual problems into account, pp. 210-50, passim. In much of what follows I am indebted to Wagenknecht. there had been to my knowledge no really detailed critical examination of internal evidence bearing on the problem. On the other side, Blake’s editors have been equally reluctant to become “readers” in the Damon-Frye tradition, preferring either to remain within the precincts of textual scholarship, or to employ such alternative critical methods as Erdman’s “historical approach.”³↤ ³ David V. Erdman, Blake: Prophet Against Empire, 2nd ed. (Princeton: Princeton Univ. Press, 1969). See pp. 293-96 for Erdman’s description of his method as applied to The Four Zoas. In some ways, no doubt, this division of perspectives is desirable; yet it has probably lent some obscurity to the problem of the Nights seven, and to the poem as a whole, which is not purely Blake’s own. In what follows I try to give a systematic account of the critical and editorial alternatives associated with the VIIa-VIIb problem, synthesizing previous arguments and adding a number of observations based on my own reading of the poem.

Both Nights are headed “Vala / Night the Seventh” and VIIb contains in addition a marginal note, “Beginning of Night VII,” and an interlinear note, “Beginning of the Seventh Night,” both inserted at line 153 after Blake decided to change the internal order of the Night. “End of the Seventh Night” is written at the end of VIIb, and the same phrase occurred after the last line of VIIa at two or possibly three different stages of revision, but Blake erased it each time to add new material, and after his latest additions he did not replace it.⁴↤ ⁴ See G. E. Bentley, Jr., William Blake: Vala or The Four Zoas: A Facsimile of the Manuscript, a Transcript of the Poem, and a Study of Its Growth and Significance (Oxford: Clarendon, 1963), p. 162 col. 2. I am, of course, indebted to Bentley throughout. All the leaves of VIIa except the last three bear stitch-marks which indicate that at one time the Night was bound up in a grouping which included most of the present Night III and all of IV, V, and VI; the unstitched leaves were presumably added after Blake had unbound the others. The fact that VIIb bears no stitch-marks indicates that it, like Nights VIII and IX and the added leaves of VIIa, was never bound.

The traditional explanation of the rival Nights is that Blake wrote VIIb first, later grew dissatisfied and wrote VIIa to replace it, and then simply failed to decide what should be done with VIIb. Thus in Erdman’s text VIIa is printed in sequence, on the supposition that it represents Blake’s later and more definitive thoughts on what should happen at this point in the poem, while begin page 108 | ↑ back to top VIIb is printed as an appendix after Night IX. The critical tradition which has grown up around the assumption that VIIa is a later “version” of VIIb interprets Blake’s need to rewrite the earlier Night as the symptom of an imaginative crisis, and thus looks to one or both Nights for some definitive turning point in the growth of the poem and the development of Blake’s vision. Frye, for instance, associates the textual problem with Blake’s invention of the Spectre of Urthona: “The conception of the Spectre of Urthona seems to have broken on Blake quite suddenly when he was proceeding to a simpler climax, and occasioned the rewriting of Night VII . . . . Eventually it burst the whole Zoa scheme altogether, and was one of the chief reasons for abandoning the poem.”⁵↤ ⁵ Northrop Frye, Fearful Symmetry (Princeton: Princeton Univ. Press, 1947), p. 298.

This account of the poem’s growth sounds persuasive, but more recent scholarship has severely questioned the assumption on which it is based, namely that VIIb preceded VIIa. Wagenknecht describes the emergence of the alternative view of the problem: ↤ ⁶ Wagenknecht, p. 226.

In 1956 . . . H. M. Margoliouth, in his edition of Vala . . . reminded scholars that, except for obviously late additions, the narrative of VIIa flows continuously into VIIb. This fact of course hinted that the long-held assumption that VIIb was composed prior to VIIa needed to be reexamined, as did the idea fostered by received opinion that the two versions were somehow thematically opposed to each other. . . . Margoliouth’s hint was developed into a death knell for received opinion by G. E. Bentley’s facsimile edition of the poem, which demonstrated conclusively on the basis of stitch marks on the manuscript that VIIb—as VIII and IX—had to be considered later than VIIa.⁶

The implication here is that the two Nights were never alternate “versions” at all, but simply two Nights written in sequence, with VIIa coming first; and that Blake gave both of them the number seven because, when he came to transcribe VIIb, he realized that he needed more than three Nights to finish his poem, and so must fit I-VIIa into six Nights.⁷↤ ⁷ See H. M. Margoliouth, ed., William Blake’s Vala: Blake’s Numbered Text (Oxford: Clarendon, 1956), p. xiii. Cf. Bentley, p. 163, cols. 1-2. In other words, VIIb was originally intended as part or all of a Night VIII, but had to be changed into a Night VII; the original Night VII, i.e. VIIa, had to be squeezed back into the preceding Nights.

It should be noted, however, that Erdman does not find Bentley’s dating of VIIb conclusive at all,⁸↤ ⁸ Nor, apparently, does W. H. Stevenson, editor of the Longman’s Annotated Edition of The Poems of William Blake (London, 1971). On p. 371 Stevenson asserts that VIIb “as a whole was undoubtedly written earlier . . . [and] rejected by Blake.” But I find Stevenson’s overall position rather confusing and equivocal. If he believes that VIIb was “undoubtedly” written earlier and rejected, why does he print the text VIIa-VIIb, and indicate in his notes (371) that “a continuity in the narrative” justifies this order? This treatment of the text corresponds to the view that VIIb is neither earlier nor rejected; and Stevenson hardly clarifies matters by remarking that “the first part of VIIa . . . appears to belong to the same date as VIIb” (371). My suspicion is that Stevenson really subscribes to essentially the view I give—that the Nights were originally written VIIa-VIIb, with the present ending of VIIa coming later—but has preferred to avoid controversy by not saying so in unmistakably clear terms. Alicia Ostriker sides with Erdman in William Blake: The Complete Poems (New York: Penguin, 1977), p. 945. but argues that Blake may have omitted it from the stitched grouping of Nights III-VIIa, not because it was later (i.e., as yet unwritten), but because Blake had already replaced it by the time he bound up the other Nights. This is not impossible, but the hypothesis by itself is not more convincing than Bentley’s simpler assumption that all of the unstitched Nights (VIIb, VIII, and IX) are later than VIIa, and it fails to account either for the narrative continuity of VIIa-VIIb, or for the use in VIIb of symbolic names which crept into the poem only at a late stage of revision (Bentley, 171 col. 2).

In short, though no really definitive evidence is available, it seems fairly clear that VIIa was composed before VIIb. Composed, that is, in some form: for the conclusion does not apply to the last three leaves of VIIa, which are unstitched and contain “obviously late additions” to the original Night (we will consider the significance of this fact in a moment). Margoliouth’s demonstration of the original narrative continuity of VIIa-VIIb remains the most important piece of evidence supporting the newer theory of the Nights seven, so it is worth glancing at this juncture in the text of his Vala:

. . . the immortal shadow shuddering
Brought forth this wonder horrible [Vala] a Cloud she grew & grew
Till many of the dead burst forth from the bottoms of their tombs
The Spectre smild & gave her Charge over the howling Orc

End of the Seventh Night
. . . . . . . . . . . . . . . . .

VALA

Night the Seventh

Now in the Caverns of the Grave and Places of human seed
The nameless shadowy Vortex stood before the face of Orc
The shadow reard her dismal head over the flaming youth
With sighs & howling & deep sobs that he might lose his rage
And with it lose himself in meekness she embracd his fire

(Margoliouth, 45-46)

In the original ending of VIIa, the Shadow of Enitharmon, having coupled with the Spectre of Urthona, gives birth to Vala, whom the Spectre then sets as a guard over Orc. At the original beginning of VIIb, Vala obediently stands on guard duty, and then embraces Orc, much as in the Preludium to America. In default of more conclusive evidence, the continuity of the narrative is a persuasive demonstration that at an early stage of the poem’s development VIIa came first and VIIb second. The argument can be reinforced by other, similar observations. In VIIa Urizen, arriving in the Caves of Orc after his journey through chaos, “took his seat upon a rock” (77:18); in VIIb we find him still seated there (95:16). In VIIa Urizen proceeds to create the precondition of war by enslaving Orc (for “War is Energy enslav’d”); in VIIb there is war. In VIIa Enitharmon tells the Spectre of her fear that Orc may rend her “to the winds of heaven” (82:38) and her hope that he will rend Vala instead (83:34); in VIIb Orc does rend Vala (93:21), and Enitharmon’s fear for herself reaches a climax in her “Watchman” speech (97:28ff.). Likewise, early in VIIa (78:4), Blake describes the Tree of Mystery in a way which strongly suggests that he is using the symbol for the first time; there have been no previous references in Nights I-VI, and the more cryptic allusions to begin page 109 | ↑ back to top the Tree in VIIb seem to presuppose the passage in VIIa.

This evidence of an original VIIa-VIIb order is further strengthened by the fact that Night VI is continuous with VIIa but not VIIb. Placed at the end of VI, either the original or the revised version of VIIb would produce a bewildering non sequitur. By contrast, VI-VIIa is a clear sequence. Night VI ends with a confrontation between Urizen, Tharmas, and the Spectre of Urthona; VIIa begins, “Then Urizen arose the Spectre fled & Tharmas fled.” It is rather hard to doubt that Blake conceived and wrote this transition all at once.

Again, the attempt to think of VIIa as a “replacement” for VIIb becomes very difficult when one notices how greatly the two differ in content. There is virtually nothing in either the original or the added passages of VIIb to indicate that the Night was an early model of VIIa. Given Blake’s usual habits of composition, the traditional account of the two Nights would lead us to expect an exceedingly complex entanglement of VIIa and VIIb, or perhaps a single palimpsest. Instead, we have two clearly separate Nights, involving largely different events and characters.

Clearly, then, we ought to abandon the idea that VIIa was originally written to replace an earlier VIIb, an idea which no substantive comparison of the two Nights will support. But this is far from being the end of the story, for Blake made major changes in both Nights as Vala turned into The Four Zoas. To VIIa he added a few small passages and then the long section of pp. 85-90, one of the crucial episodes of the present poem. He appears to have made few major additions to VIIb, but he left three notes directing that the first 153 lines be moved to the end, thus reversing the order of the two main episodes of the Night (Erdman, 762). These revisions so thoroughly undermine the original narrative continuity of VIIa-VIIb, and raise such doubts concerning Blake’s ultimate intentions, that the editor of The Four Zoas finds his basic problem still unsolved despite the discovery of the Vala sequence. If the order of the original Nights was clear enough, the order of the present Nights is not; it is, in effect, a separate question.

There is no real possibility of reading VIIb first and then VIIa; but once the added passages of VIIa are taken into account, the opposite sequence becomes equally unsatisfactory. It is clear enough why such critics as J. Middleton Murry, Northrop Frye, and Harold Bloom, once persuaded that it was necessary to choose between the two Nights, should be eager to believe that VIIa was the later and authoritative version. For in the present VIIa we have not only the great dramatic confrontation of Orc and Urizen, but the climactic rapprochement between Los, his Spectre, and Enitharmon. The latter event in particular, which takes place entirely in the added passages, provides a crucial link in Blake’s myth. Prior to VIIa, Los is a merely natural imagination, a formidable but basically inhuman demiurge. He is often called a “demon” and a “terror,” and he has helped Urizen to enslave Luvah and later Orc. Up to this point all his creations have been semi-instinctive, ad hoc responses to the fallen state, and the remainder of his energy has been spent in lovers’ quarrels with Enitharmon. But the sequence of events which begins with the Spectre of Urthona’s seduction of Enitharmon (84-86) works a profound change in Los. By embracing Enitharmon, the Spectre gains the power to “enter Los’s bosom,” and then is embraced in his turn by Los, “first as a brother / Then as another Self.” Under the force of the Spectre’s “inspired” persuasion, Los suddenly feels “A World within / Opening its gates” and vows to “teach / Peace to the Soul of dark revenge & repentance to Cruelty.” It is at this point that we hear, for the very first time in all Blake’s works, of the necessity of “Self-Annihilation”—henceforth one of Blake’s two or three most important themes.⁹↤ ⁹ Concordance, p. 1654. The new alliance and new vision lead directly to the discovery of Los’s true mission: he will progressively reconstruct the fallen Albion, first by weaving “bodies” for the “Spectres of the Dead,” then by creating “forms sublime” which the spectres can assimilate into (87:36-39 & 90:15-25). The Spectre of Urthona changes at this point from a “ravning” Selfhood to a helpful “medium,” and Enitharmon, still more dramatically, from an aberrant Female Will into a loving source of creative inspiration. Finally, in a fitting climax to these climaxes, the three-way alliance manages to regenerate and redeem the fallen Urizen.

Thus the extraordinary added passages on pp. 85-90 point the way, not only to the apocalypse of Night IX, but to the central myths and themes of Blake’s two subsequent epics. Given the importance of the Los-Spectre relationship in Jerusalem, and of the weaving of bodies in Milton, one gets the impression that in these pages of The Four Zoas Blake achieved an imaginative breakthrough which profoundly affected his conception of the later poems—unless, indeed, it was his work on the later poems which showed him how to solve his problems with The Four Zoas. It is understandable, then, that critics should find VIIa more faithful to the spirit of Blake’s mature vision than VIIb, and should therefore be reluctant to surrender the traditional view of the problem. The tendency can be observed in Harold Bloom’s discussion of the Nights seven in the Erdman edition of Blake:

The two sections of Night VII are alternate ways of preparing for the blackness of Night VIII and the liberation of Night IX. I find it difficult to believe that VIIa is not the later in composition, but in this belief I am guided only by what I take to be internal evidence. Whichever was later, our understanding of the poem’s total design can only benefit by a study of both versions. I do not think it accurate (or fair to the poem) to read Nights VIIa and VIIb as being an intentional sequence, in that order. VIIa in at least some respects is an imaginative advance on VIIb.

(Erdman, 876)

One cannot fault Bloom’s sense that VIIb makes an begin page 110 | ↑ back to top unsatisfactory sequel to the present VIIa; for in VIIb we confront the old Los again, a warrior Los with “his knees / Bathed in bloody clouds. his loins in fires of war where spears / And swords rage . . . ” (96:21).¹⁰↤ ¹⁰ Similarly, Stevenson notes that in VIIb, “The regeneration of Los and Enitharmon is ignored,” p. 389. Cf. Margoliouth, p. 144, and Paley, p. 161. Similarly, Tharmas, Orc, and Urizen, all of whom have been partially rehabilitated in the triumphant reconciliations of VIIa, reappear in VIIb, without explanation, in more primitive stages of fall. As a sequel to the present VIIa, then, VIIb would give us at best a very sour irony, sufficient to undercut the new ending of VIIa, and with it the justification for the poem’s apocalyptic conclusion.¹¹↤ ¹¹ Of course Night VIII is a war Night too; but here the battle scenes are matched in stirring counterpoint to scenes which depict Los’s mighty efforts to resist the forces of fall, so that the outlook is not nearly so bleak as in VIIb. On the whole, the tone of VIII seems to me one of breathtaking suspense tending strongly toward optimism.

Moreover, the present VIIa tallies fairly well with Night VIII, while VIIb does not. At the beginning of VIII we have a “Council of God” passage—one of Blake’s favorite setpieces—and then Los and Enitharmon in effect resume the conversation they have been having near the end of VIIa:

Then Los said I behold the Divine Vision thro the broken Gates
Of thy poor broken heart astonishd melted into Compassion & Love
And Enitharmon said I see the Lamb of God upon Mount Zion
Wondring with love & Awe they felt the divine hand upon them

(VIII, 99:15; cf. VIIa, 86:44-60)

References to the apocalyptic mission quickly follow, with Enitharmon once again “sighing forth” the spectres and Los “recieving them into his hands” (99:23-100:2; cf. VIIa, 90:35-67). Late additions throughout Night VIII continue to develop the same theme (cf. p. 113), supporting the conclusion that Blake was assuming the final version of VIIa as the immediately preceding Night. One could, conceivably, make a similar argument for the continuity of VIIb (in the revised sequence) and Night VIII, on the basis that the spectres of the dead descend from Beulah at both 95:11 and 99:19. But the spectres also descend late in VIIa (85:18), and on the whole a VIIb-VIII sequence would present an inexplicable shift from sardonic gloom to prophetic hope. The Los and Enitharmon of VIII are unmistakably the prophetic figures who emerge at the end of the revised VIIa; they have little in common with the savage Los and hysterical Enitharmon of VIIb.¹²↤ ¹² Erdman, p. 758, offers a rather curious argument to show that the two Nights were closely joined at a previous stage of composition. He theorizes that the marginal note Blake left on page 100, calling for the insertion of a passage beginning “Los stood” between 100:1 and 100:2, refers to most of page 90—i.e., basically to the present ending of VIIa—but that Blake later changed his mind and left the passage on 90 in its present position. His conclusion is that “this bit of masonry” (the provisional insertion of 90 into 100) “which cements VIIa closely to VIII, seems to indicate a time when Blake considered VIIb as abandoned (or moved).” Wagenknecht reviews the argument on pp. 298-300; I agree with his conclusion that Erdman’s theory about pages 90 and 100 is “extremely suppositious . . . on the basis of physical evidence alone,” and also improbable for stylistic and thematic reasons. But unless I misread, Erdman’s conclusion still stands, for the different and much more obvious reason that VIIa in its present state is still “cemented closely” to VIII. Cf. Paley, p. 161.

It seems, then, that at a late stage of revision Blake decided to route the poem around VIIb. This means that the traditional account of the two Nights may not be so far mistaken after all. Nothing says that Blake had to write VIIb first in order to be dissatisfied with it, or VIIa second in order to prefer it if a choice had to be made. It is quite possible that he first wrote both Nights (as consecutive rather than alternate episodes), next started feeling that his approach to apocalypse was unsatisfactory, and then saw that his best opening for a more suitable approach could be found in the earlier VIIa. Once he had made his changes in VIIa, he would have found that VIIb no longer fit as a sequel; since at this point he had ten Nights anyway, he might then have decided to keep VIIa as Night VII (reversing his earlier decision to squeeze it back into I-VI), revising Night VIII accordingly. Thus VIIa, though not originally a replacement for VIIb, would have become one by the time Blake finished his revisions.

This theory would seemingly allow us to go on reading VIIb in the way it has usually been read—as an alternate but rejected approach to Night VIII—if we stipulated that VIIb’s previous niche is presently filled by the added section of VIIa alone, rather than by the Night as a whole. Once again, however, matters cannot be so simple. Blake was not a poet to delete material he liked if he thought he could possibly find a place for it, and the very fact that he kept VIIb together with the rest of his manuscript testifies to his hope that it could somehow be reintegrated into the poem. Moreover, though VIIb seems to be excluded by the VI-VIIa-VIII sequence upon which Blake finally settled, its disentanglement from the other Nights is incomplete. For instance, the references to war in the added passages of VIIa come quite abruptly; when Los draws Urizen “from out the ranks of war,” it is very difficult to tell just what war Blake is talking about.¹³↤ ¹³ There are four previous references to war in VIIa, but these are brief and casual. Concordance, p. 2069. Then on reflection one has to conclude that it is the war described in VIIb. Of course the poem’s tendency to subvert linear chronology should never be underestimated, and it is just possible that we are dealing with a proleptic reference to the wars of Night VIII. Still, on the whole the climactic turn toward peace, mercy, and art at the end of VIIa seems more comprehensible if played against the backdrop of Orc’s violent rising in VIIb. It is as if Blake, once having written VIIb, could not wholly refrain from assuming it as a basis for his revisions to VIIa, or as if he worked with the intention of later transplanting parts of VIIb into VIIa.

Wagenknecht offers a substantial demonstration of VIIb’s importance to Blake’s overall argument (225ff.), and shows that Night VIII contains back-references to VIIb as well as VIIa; his conclusion is that Blake left “two integral, not alternative, Nights VII.” If Blake had suppressed VIIb, perhaps no one would ever have noticed that it was missing; with the Night before us, however, we see many things that the poem as a whole seems to need. The very dramatic meeting between Tharmas and Vala brings Tharmas up to date—he has been rather scarce since Night IV—and prepares for his next appearance, in VIII. The emergence of Vala herself as a fully distinct character obviously provides a crucial continuation to the story of her birth in VIIa, and a prelude to her appearance as Rahab in the apocalypse. In some ways her epiphany is redone in Night VIII, yet the later treatment is less full, and can be better understood by the light of the earlier one.¹⁴↤ ¹⁴ Cf. especially 101:20ff: . . . While the dark shadowy female brooding over
Measurd his [Orc’s] food morning & evening in cups & baskets of iron
With tears of sorrow incessant she labourd the food of Orc
Compelld by the iron hearted sisters Daughters of Urizen Here we have in little the Preludium of America—and the original beginning of VIIb—yet again. This could mean that Blake was thinking of VIIb as abandoned, and was trying to sketch in one portion of its contents which he considered vital. On the other hand, the passage is brief enough to have been intended as a refrain or echo, so it is equally possible that Balke was still committed to including VIIb. For instance, the descent of the spectres, “tempted by the Shadowy females sweet / Delusive cruelty” (VIII, 99:20), makes more sense if one has first read the exquisite description of Vala’s beauty which Blake provides in VIIb.

In sum, we have to regard VIIb as part of the poem, but there is no really satisfactory way to fit it in. Blake left the Night as a unit, but as a unit it resists placement anywhere in the present begin page 111 | ↑ back to top text. It seems that VIIb was originally designed to carry the poem all the way from the birth of Vala to the brink of apocalypse—albeit a simpler apocalypse than the one Blake finally imagined—and it has a cataclysmic finality which makes it seem to belong—somewhere—late in the present Night VIII.¹⁵↤ ¹⁵ In Margoliouth’s Vala, the ending of VIIb in its original order is joined directly to the original beginning of IX—so that VIIb is in effect the Night VIII of Vala. I do not find this arrangement of the text quite persuasive, even as representing an earlier state of the manuscript, but I do agree with Margoliouth’s general sense (and with Bentley’s) that the material of VIIb was originally conceived as part of the last Night before the apocalypse. It is mainly the emphasis on climactic war (a “signal of the dawn” for which Blake had the authority of Revelation) which persuades me of this. But by the time Blake made the present transcript of VIIb, with the heading “Night the Seventh,” another Night VIII must have existed. Yet its general pessimism and the unregenerate state of its protagonists make VIIb seem to fit best into the juncture between the old and new portions of VIIa, where it originally began. No doubt this is the least unsatisfactory niche; still, VIIb rises to a climax of its own which would throw the alternate climax of the revised VIIa off center. In VIIb the apocalypse seems imminent; in VIIa it is not to come for another six thousand years (87:26). In VIIb Tharmas and Vala seem destined to be key figures in bringing about the final consummation; in VIIa they are supernumeraries. Moreover, if restored to its original position, VIIb would interrupt the story of Los’s reconciliation with his Spectre and Emanation, spoiling the narrative rhythm of VIIa. It is certain that Blake himself, once he had conceived of the regeneration of Los, preferred to avoid such interruption and to pursue the narrative of VIIa directly to its present climax.

The puzzle grows still more perplexing when one tries to grasp Blake’s reasons for changing the order of VIIb. It seems logical to suppose that the change would have come after Blake had written the new ending of VIIa, since the original continuity of VIIa-VIIb would already have been lost. If at this point Blake was toying with the idea of a VIIa-VIIb-VIII sequence, the switch would have the advantage of moving the descent of the spectres to the end, creating a bridge between VIIb and VIII. Yet the corresponding transition from the new ending of VIIa to the new beginning of VIIb seems quite wayward. At the end of VIIa Urizen is an infant in Los’s arms; the reordered VIIb begins, “But in the deeps beneath the Roots of Mystery in darkest night / Where Urizen sat on his rock the Shadow brooded / Urizen saw & triumphd & he cried to his warriors.” Of course, this is just enough of a non sequitur to be possibly intentional—especially since Blake has noted at the end of VIIa that Los was able to redeem all of Urizen except “his Spectrous form which could not be drawn away” (90:60). The lines take us back to that troublesome crossroads in the narrative, Vala’s birth from the Shadow of Enitharmon, and it is this event rather than Los’s regeneration—or this event together with a misconception of what has happened to Los—which causes Urizen to triumph.¹⁶↤ ¹⁶ Vala was originally mentioned in the third as well as the first line of the passage, which read “Urizen cried to the Shadowy Female” rather than “Urizen cried to his warriors.” The change, however, does not affect the fact that Urizen is gloating over the birth of Vala: it merely means that his speech is now directed to his warriors rather than her. An additional, deleted line on the same page reads, “The shadowy voice answerd O Urizen Prince of Light,” showing that Blake at one time conceived of Urizen’s speech as part of a conversation between Urizen and Vala. Since no other trace of Vala’s half of the conversation remains here (it has conceivably been moved to VIII, 102:28 ff.), it seems probable that the change in the third line came quite early, before the reordering of the Night. It seems just conceivable that Blake was trying to rearrange VIIb in such a way that, having traced the consequences of the Enitharmon-Spectre union in the upper world of Los, the narrative now backtracks and begins to trace the consequences of the union in the lower world of the spectrous Urizen. Yet such a change, supposing that Blake once intended it, would do nothing to solve the weightier problems involved in a VIIa-VIIb sequence, leaving us with an inexplicable contrast between prophetic hope and ironic despair.

The other possible explanation, to which I tentatively incline, is that Blake changed the order of VIIb before writing the new ending to VIIa. On this hypothesis, the change would be the first expression of Blake’s dissatisfaction with the Vala sequence, and the first sign of his search for a more humanistic approach to apocalypse. The advantage of the new order is in moving the confrontation of Tharmas and Vala—which, though gloomy enough, is at least more hopeful than the frantic violence which dominates the other portion of the Night—to the end.¹⁷↤ ¹⁷ Stevenson, p. 390, gives the opposite opinion: “In the second half (as it now is) Orc appears. In transferring this passage to the end, Blake has delayed till last the revelation of the worst evil.” The idea is supported by Blake’s late addition, at 94:12, of a lament for Luvah which makes the portrait of Vala considerably more sympathetic. Moreover, the reversed VIIb would still be fairly continuous with the original VIIa (though less so than formerly), since the revised beginning returns to the moment of Vala’s birth. But if the reordering of VIIb did in fact occur at a fairly early stage, it probably corresponded to a plan of revision which Blake abandoned when he decided to merge Los and the Spectre. At whatever point it was executed, the flip-flopping of the Night was a mechanical tactic, incapable by itself of solving any of the serious problems confronting Blake. His failure to change the numbering of the Nights seven is, by itself, sufficient evidence that he was never finally satisfied to place VIIb between VIIa and VIII.

Still, it is not difficult to arrive at a general sense of Night VIIb’s place in Blake’s vision, even though it is impossible to find a physical place for it in the text of his poem. In effect, Blake’s narrative separates into parallel strands when Vala is born from the Shadow of Enitharmon, at the point of the original ending of VIIa. Frye’s account of this doubling refers only to the present VIIa, but seems valid even if applied to both Nights seven: “In Night VII [i.e., VIIa] . . . a double crisis takes place, one an imaginative advance, symbolized by the mingling of Los and the Spectre of Urthona, the other a consolidation of error symbolized by the birth of Rahab [i.e., Vala] from the Spectre of Urthona and . . . the Shadow of Enitharmon” (Frye, 278). Roughly speaking, the path of “consolidation of error” leads through VIIb, with its emphasis on climactic war, the catastrophic rising of Orc, the triumph of Urizen, the epiphany of Vala, and the inconsolable rage of Tharmas; the path of “imaginative advance” leads through the added passages of VIIa, where for the first time we find a fully prophetic Los, aided by his Spectre and Emanation, struggling to save mankind through creative action. The two strands come back together again in VIII (though with some fairly messy duplication of VIIb), where a sense of breakneck progress toward apocalypse is brilliantly captured in the counterpointing of Los’s furious acts of creation and Urizen-Satan-Rahab’s equally furious acts of subversion and destruction.

Likewise, though no exact chronology of the composition of these Nights can be given, it is possible to speak with reasonable assurance about the process of imaginative growth which must have been involved. At an early stage, it would seem, Blake conceived of an apocalypse motivated almost exclusively by the “consolidation of error,” a mechanical apocalypse entailing an almost instantaneous leap from nadir to zenith. What this vision lacked was begin page 112 | ↑ back to top the myth of the fully human and self-conscious Los, and thus any sense of the possibility for positive human action in the fallen world. Possibly Blake was astonished at the oversight once he recognized it; possibly he came to regard Milton and Jerusalem as categorically superior to The Four Zoas because in those poems the mature Los, the self-conscious visionary confronting human problems on the stage of human history, is dominant throughout, whereas the far-ranging fantasy of The Four Zoas deals largely with a sort of cosmic pre-history.

In any case, the new ending to VIIa marks a decisive change in Blake’s concepts of apocalypse and prophecy. By mingling Los and the Spectre Blake creates a human agency capable of turning fallen history in the direction of a Last Judgement, thus transforming the apocalypse into a quest story with the poet-prophet as its unequivocal hero. It was a change which strained the narrative structure of the epic to the breaking point, squeezing VIIb halfway out of the poem and causing major revisions in Night VIII. Yet the new vision does not really invalidate the old; rather, as a later stage in a dialectic, it necessarily implies and incorporates the earlier stages. The essential coherence of the structure which Blake finally left—VIIa, VIII, and IX, with VIIb lounging somewhere in the shadows—is not diminished by the fact that it presumably grew out of a poetic process whose beginning did not know its end.

Why, then, did Blake lack the stamina to make a final disposition of VIIb, if indeed that was all he had to do in order to finish his poem? The question is especially troubling in that Blake apparently had a logical moment at which to attempt the reintegration of the displaced Night when he revised Night VIII to bring it into line with the new VIIa. It would seem more natural to bring in passages from VIIb at this point than to plan on yet another revision of VIII later on. On the other hand, one can imagine that Blake would find the problem of VIIb an unwelcome distraction from the new vision which absorbed him as he reworked Night VIII. Moreover, he might have been rather slow to part with the idea that VIIb could be retained as a unit. The Four Zoas is, finally, a story, yet its narrative structure is formed by the spatial accretion of alternate myths and symbols around a central core of events (Fall, Incarnation, Redemption, Apocalypse) which they interpret and exemplify, as well as by the linear unfolding of the plot. It may have taken Blake some time to decide that the clash of the old VIIb with the new VIIa and VIII was a greater incongruity than even this structure could tolerate.

Nevertheless, the recognition of VIIb’s incompatibility eventually came, and one still must wonder why Blake never solved the problem. The speculation which suggests itself, supposing that we have guessed right up till now, is that the new inspiration proved all too potent. It drew Blake’s energies away from The Four Zoas toward Milton and Jerusalem, but persuaded him more and more that further revisions of the earlier poem were necessary, or at least desirable. Night VIII in particular makes extensive use of the symbolic “Machinery” of the later epics, in a way that sometimes justifies Margoliouth’s remark that “Blake abandoned the poem because the new wine was bursting the old bottles.”¹⁸↤ ¹⁸ William Blake (London: Oxford Univ. Press, 1951), p. 115. It is as if Blake could not content himself with completing The Four Zoas as such, but had to go on to attempt a wholesale demonstration of the poem’s consistency with its offspring; as if, after a certain point, everything had to be said over again from the stand-point of Jerusalem. Nights I and II contain certain late additions which suggest that Blake may have decided to work through his first six Nights yet again, installing passages which would anticipate the new vision, before tackling the problem of VIIb. Yet at the same time, judging by the virtually atemporal structures of Milton and Jerusalem, Blake was undergoing a crisis of disenchantment with narrative itself. In Jerusalem he seems completely unwilling to impose anything resembling a narrative chronology on the actions and events of his myth. As he grew more and more accustomed to letting spatial principles of symbolic association prevail over “clock-time,” Blake no doubt found himself less and less inspired to resolve the narrative tangles of the Nights seven.

How, then, should the two Nights be presented to the reader? My not very earthshaking conclusion is that K in this particular is superior to E. VIIb should come after VIIa, but is too important to the poem to be treated as an appendix. Both editions are undoubtedly correct in presenting each Night as a unit, rather than attempting to reintegrate VIIb into VIIa or VIII or both, for such an attempt would be highly presumptuous, and would obscure a problem it could not solve. Likewise, it seems best to print VIIb in the revised order, a step which is dictated by the general policy of following the latest intentions for which there is direct evidence in the manuscript.

Whether the critic should be bound by this arrangement of the Nights is another question, or perhaps no question at all, since the critic is always at liberty to choose the order of the passages which he wishes to discuss. But for the purpose of a linear commentary, or of having a theoretical model of the poem’s narrative sequence, I would recommend a critical order much like the one Wagenknecht uses in his reading of the poem.¹⁹↤ ¹⁹ See Blake’s Night, pp. 224-25. Having read VIIa up to the point of Vala’s birth from the Shadow of Enitharmon, I would then take up VIIb, considering it in its original position in order to get a coherent view of Blake’s first attempt to depict approaching apocalypse. I would choose the original rather than the revised version of VIIb (though with some hesitation), for the sake of preserving the original continuity of VIIa-VIIb, and because I suspect that the reordering of the Night corresponded to a plan of revision which Blake never carried out. Having read VIIb straight through in this order, I would return to the added passages of VIIa; then on to VIII and IX. I do not, of course, claim that this ordering of the text would have any authority; it is merely a sequence for which there is some scholarly justification, and one which would be useful for some critical purposes.