Building Jerusalem: Composition and Chronology
Of all the puzzles surrounding the dates of Blake’s poems, that of Jerusalem seems the most intractable. In the absence of a draft, and with only a few contemporary references whose interpretation itself is problematic, critics and editors have settled for the stretch of time from 1804, the date on the title page, to 1820, the latest watermark in the first four complete copies printed, as the probable limits between which Blake composed, etched, revised and printed the hundred plates of the volume.1↤ 1. See the table in G. E. Bentley, Jr., Blake Books (Oxford: Clarendon Press, 1977; henceforth BB) 226. But within these limits there is great disagreement as to the specific years in which Blake actually composed Jerusalem2↤ 2. Both Christine Gallant (Blake and the Assimilation of Chaos [Princeton: Princeton University Press, 1978] 49) and James King (William Blake: His Life [New York: St. Martin’s Press, 1991] 172-74) argue that it was begun even earlier, at Felpham, 1800-03, as Peter Ackroyd also implies (Blake [London: Sinclair-Stevenson, 1995] 174). Ackroyd thinks it was started “at the same time” as Richard Brothers published A Description of Jerusalem, i.e., 1801 (cf. Morton D. Paley, “William Blake, the Prince of the Hebrews, and the Woman Clothed with the Sun,” William Blake: Essays in Honour of Sir Geoffrey Keynes, ed. Morton D. Paley and Michael Phillips [Oxford: Clarendon Press, 1973] 273). —and indeed, exactly what is meant by the composition of any of Blake’s illuminated works. The process includes a number of stages separable from each other by varying lengths of time: initial inspiration, whether verbal or visual; drafts of text, sketches of designs; inscribing more or less finished portions of text and accompanying designs on copper; possible revision; etching the plates; printing, coloring, then arranging and binding the printed pages. Indeed, one of Blake’s descriptions of the process raises doubts about the reliability of any of his reports on the composition of a specific work.3↤ 3. “When I am commanded by the Spirits then I write, And the moment I have written, I see the Words fly about the room in all directions[.] It is then published[.]—The Spirits can read and my MS: is of no further use[.]” Henry Crabb Robinson, Reminiscences (1852), in G. E. Bentley, Jr., Blake Records (Oxford: Clarendon Press, 1969; henceforth BR) 547. His apparent references to Jerusalem are too vague to be useful—whether in his 1803 letters to Butts, once alluding to a “long Poem” descriptive of his “Spiritual Acts” during his three years at Felpham, and two months later to “a Sublime Allegory which is now perfectly completed into a Grand Poem” speaking of his “three years trouble,” or in his Descriptive Catalogue of 1809, where he refers to a “voluminous” work which he “will, if God please, publish,” or in his Public Address of 1810 mentioning a “Poem concerning my Three years <Herculean> Labours at Felpham,” soon to be published: all these allusions point more clearly to either The Four Zoas or Milton.4↤ 4. The Complete Poetry and Prose of William Blake, ed. David V. Erdman, newly rev. ed. (New York: Doubleday Anchor, 1988; henceforth E) 728, 730, 543 (cf. BB 457), 572. In fact, his only direct reference to Jerusalem occurs in a letter to George Cumberland of 12 April 1827 mentioning it as “the Last Work” he produced, one copy of which he has “Finishd,” i.e., colored (E 784). Two possible allusions also occur in A Vision of the Last Judgment (1810), where his statement “Imagination is Surrounded by the daughters of Inspiration who in the aggregate are calld Jerusalem” (E 554) sounds remarkably like a description of the title page to Jerusalem, while his later reference to Jerusalem as the Emanation of Albion, which he immediately cancels to identify her as the daughter of Albion and Brittannia (E 558), suggests a stage in the composition of the poem before her role was fixed.
However, a more precise dating of Jerusalem was offered by G. E. Bentley, Jr., in his Blake Records of 1969 which has since been widely adopted. In 1964 Bentley discovered in an 1807 notebook of George Cumberland’s an isolated entry that “Blake has engd. 60 Plates of a new Prophecy!” This, he concluded, must refer to Jerusalem, since it is the only poem of Blake’s containing sixty plates or more (BR 187 and n4). He communicated his discovery to David Erdman, who noted it in his 1964 article on Jerusalem but did not adopt Bentley’s conclusion as to the date in his edition of Blake’s works in 1965 or thereafter—nor did Geoffrey Keynes in his 1966 edition.5↤ 5. David V. Erdman, “The Suppressed and Altered Passages in Blake’s Jerusalem,” Studies in Bibliography 17 (1964): 1-54 and E 730; Geoffrey Keynes, ed., The Complete Writings of William Blake, 2nd ed. (London: Oxford University Press, 1966) 918. However, in 1977 Bentley extended his argument by specifying thirty-seven “late” plates which for various reasons of internal evidence he believes were added to Jerusalem after 1807,6↤ 6. BB 225-28: pls. 8, 10, 11, 16, 19, 33, 34, 35, 42, 43, 46, 47, 53, 55, 56, 59, 60, 61, 63, 64, 65, 66, 67, 70, 71, 72, 77, 78, 82, 89, 92, 93, 94, 95, 96, 99, 100. leaving some sixty-three as the “early” plates supposedly shown to Cumberland in 1807. Since then, most scholars and editors have followed Bentley’s hypothesis that more than half of the work was completed and etched by the summer of 1807.7↤ 7. Most importantly, Morton D. Paley, ed., Jerusalem, Blake’s Illuminated Books, vol. 1 (Princeton: Princeton University Press/Blake Trust; London: Tate Gallery/Blake Trust, 1991) 12, and in The Continuing City: William Blake’s Jerusalem (Oxford: Clarendon Press, 1983) 2.
Yet there are strong reasons for disagreeing with Bentley’s theory. First, there are twenty-six plates containing references to Hand scattered through the poem, which must have been written after Robert Hunt’s devastating review of The Grave in 1808 (followed by his still crueler notice of Blake’s 1809 Exhibition),8↤ 8. The Blake concordance (ed. Erdman, 2 vols. [Ithaca: Cornell University Press, 1967] 1: 860) lists twenty-five plates (to which 43:50 should be added) referring to “Hand (Symbolic Name),” in thirty-seven separate lines: 5, 7, 8*, 9, 15, 17, 18, 19*, 21, 26, 32, 34*, 36, 42*, 43*, 58, 60*, 67*, 70*, 71*, 74, 80, 82*, 83, 84, 90. The ten plates marked with asterisks are from Bentley’s “later” group. but of these twenty-six plates only ten are begin page 184 | ↑ back to top in the group of thirty-seven that Bentley identifies as “later,” i.e., post-1807.9↤ 9. The ten “later” plates referring to Hand are 8:41, 43, 44; 19:18; 34:22, 36; 42:47; 43:50; 60:14, 43; 67:62; 70:1; 71:11, 22; 82:63. The other sixteen plates mentioning Hand (in twenty-two separate occurrences) are in the remaining group of sixty-three plates which Bentley would presumably designate as “early” (i.e., 1804-07).10↤ 10. 5:25; 7:18, 71; 9:21; 15:1; 17:62; 18:36, 39; 21:28; 26; 32:10; 36:15, 23; 58:29; 74:49; 80:57; 83:30, 86; 84:20; 90:23, 40, 49. Bentley notes that “The references throughout to Hand, who evidently represents the three Hunt brothers, must date from 1808 or later” (BB 228-29), but does not correlate these references with his division of early-late plates. In addition to these sixteen problematic “early” plates should be added a group of six other “early” plates alluding to events occurring after 1807. Plate 27:25-26 refers to excavations in Paddington in 1811;11↤ 11. David V. Erdman, Blake: Prophet Against Empire, 3rd ed. (Princeton: Princeton University Press, 1977; hereafter BPAE) 474. plate 38:6-7 refers to the War of 1812 and the executions of Mexican insurgents in 1811 and 1813 (BPAE 482); plate 40:19-24 describes the abolition of the slave trade in 1807 (BPAE 429); plate 45:25 alludes to the new (1815) Bethlehem Hospital as well as to the Battle of Waterloo (lines 55-56), and includes one of many warnings against taking vengeance on the defeated French (lines 36ff.) (BPAE 469, 466, 470). Plate 68:10ff. contains a song of victory for 1814 or 1815 as well as an allusion to the Russian campaign of 1812 (line 52) like that of plate 86:46.12↤ 12. BPAE 466, 482. Note also pl. 63:5-6, alluding to the Peace of Paris (BPAE 466), previously counted (note 6). Beside these post-1807 allusions, the character of Erin (mentioned seventeen times in twelve different plates)13↤ 13. Pls. 9*, 11*, 12, 44, 48, 50, 72*, 74*, 78*, 86, 88, 94*. Asterisks indicate plates counted previously, leaving eight new post-1807 allusions. symbolizes the new hope for freedom springing up in England with the revival of the cause of Irish independence in the decade of 1810, as Erdman suggests (BPAE 481-84), while the repeated warnings to Albion not to take vengeance on his defeated enemy (BPAE 462, 470) must date from 1815 onward.14↤ 14. Pls. 25:3-6, 36:41-42, 45:29-38, 47:12-16*, 63:8, for a total of four additional post-1807 allusions, not counting pl. 47 (cf. note 6). Thus a total of thirty-four or more plates which apparently were written in 1808 and after should be added to Bentley’s group of thirty-seven “later” plates. If approximately seventy-one out of the hundred plates of Jerusalem are then to be dated after 1807, Cumberland’s 1807 memorandum that “Blake has eng.d 60 plates of a new Prophecy!” cannot refer to Jerusalem.
It has been suggested that Blake simply revised many of these sixty etched plates in the years following 1807 to incorporate the thirty-odd references to Hand and post-1807 events.15↤ 15. E.g., W. H. Stevenson, ed., The Poems of William Blake (London: Longman, 1971) 622: “To include Hand as an important figure B, must have revised the poem after 1808.” But both Robert Essick and Joseph Viscomi, with their expert knowledge of Blake’s printing methods, have described the “extreme difficulty” of making revisions on already etched plates,16↤ 16. Robert N. Essick, William Blake Printmaker (Princeton: Princeton University Press, 1980) 109, demonstrating the difficulty of adding new relief lines to the already etched text of pl. 25 of Jerusalem; Joseph Viscomi, Blake and the Idea of the Book (Princeton: Princeton University Press, 1993) 278, 342. and none of the small number of such revisions which they detect involves the allusions cited above. If it is supposed that Blake revised his text after 1807 by simply discarding already etched plates for new ones, it must be countered that no evidence of such discarded plates survives, and that he was of necessity extremely parsimonious with copper. Almost all of the plates of Jerusalem (like those of most of the other illuminated works) were etched back-to-back in order to save expense, and it appears he was running out of copper as he completed Chapter IV.17↤ 17. Cf. BB 113, 145, 166, 208, 235-36, 308-09, 335, 381-82; also 225-28 on Blake’s shortage of “other, more perfect” plates in etching pls. 33, 56, 63, 71-72, 92-93, and 100. As for revising previously etched or “early” plates to incorporate the twenty-two lines referring to Hand (to say nothing of all the other post-1807 references), it should be noted that only seven “etched variants” are listed by Bentley, including none of these allusions.18↤ 18. Bentley, BB 237, adds seven other plates containing pen-and-ink variants and mentions about thirty more minor variations in the designs in his catalogue of the separate plates (238-58); none of these makes significant changes in the text except for the deleted mottos on pl. 1. Viscomi (338-39) claims forty-six plates bear deletions but mentions specifically only a few of them, such as the notable gougings-out on pl. 3; the few additions to etched plates consist of page numbers, corrected spellings, and two lines (37/33:1, last four words, and 45:40) etched in white line on a raised area (Viscomi 254, 256, 340-41, 354; cf. BB 246, 248; Essick 110-11, 156; Paley, ed., Jerusalem 188, 203). Moreover, Hand is an integral part of the poem from the beginning: he first appears on the sixtieth line of the text (5:25), and the thirty-seven separate allusions to him are scattered fairly evenly throughout the poem as a whole. Revising the etched plates of Jerusalem to incorporate such a number of textual additions after 1807 is thus a virtual impossibility: the “new Prophecy” cannot have been Jerusalem, and can therefore only have been Milton.
How then are we to interpret Cumberland’s statement? First the possibility of human error or misunderstanding in one form or another should be admitted. Cumberland may have misremembered sixty for fifty, in which case his statement readily applies to Milton, which may well have contained nearly fifty plates in the summer of 1807. Or, since it does not appear from Cumberland’s statement that Blake actually showed him sixty plates of an unnamed “new prophecy” but merely mentioned the figure, he may have simply been referring to intermediate stages of its design. Or the “new prophecy” may have denoted an amorphous state in the evolution of both Milton and Jerusalem, in which portions of both were mingled.19↤ 19. Bentley, BB 307, suggests the existence in draft of “a poem which comprehended the actions that were later divided between Milton and Jerusalem”; Essick and Viscomi (eds., Milton a Poem, Blake’s Illuminated Books, vol. 5 [Princeton: Princeton University Press/Blake Trust; London: Tate Gallery/Blake Trust, 1993] 36) describe “both Milton and Jerusalem as a single project” at one point.begin page 185 | ↑ back to top
But it is also possible that Milton at one time may have contained something like sixty plates, as Bentley himself has thought possible (BR 187n4). Milton clearly went through several stages of expansion and contraction during the course of its composition, as reflected in the varying number of plates in the different copies, and one of the early stages apparently included material about the English Civil War (BPAE 423-24), subsequently deleted, which linked it with the Lambeth Prophecies as one more chapter in the history of revolution. Erdman’s speculations on “the missing ten books of Milton” (BPAE 423-24) which linger in the background of “the early and late versions,” i.e., the “revolutionary version” and the pacifistic version (BPAE 425-27), suggest that at least a few pages of the early twelve-book form were committed to copper before being erased. In 1845 J. T. Smith recalled that Blake often “rubbed out” already etched plates “to enable him to use them for other subjects.”20↤ 20. BR 26n†, and cf. Viscomi 389-90n4 and BB 225-28, 335. Though he did not describe how these etched plates were erased, presumably an already etched plate could be polished to a clean surface like a new one more readily than it could be revised for corrections or additions.21↤ 21. See Viscomi 48-49 on the preparation of the plate for etching or engraving. Viscomi nowhere refers to Smith’s statement quoted above, from his 1845 Book for a Rainy Day. If so, this would supply the ten or more pages added to the existing forty-five to fifty pages of Milton (copies A-C) to make up the mysterious “60.”
One final consideration in the case against the half-finished Jerusalem of 1807 is the date on the title page. How is the date of 1804, inscribed at the bottom of the page, to be understood if the poem was not actually begun until at least 1808? Butlin has commented on Blake’s “habit of pre-dating works unparalleled until its adoption by some of the pioneers of the modern movement,”22↤ 22. Martin Butlin, William Blake (London: Tate Gallery, 1978) 27. which may well be the case here. In fact, the only reason for taking 1804 as the date at which Blake started work on Jerusalem is simply the date “1804” of the title page, squeezed in above the signature in a cramped space at the bottom of the page, looking almost like an afterthought. The date is not an integral part of the design of the page, as it is so strikingly in the title page of Milton, or even of the signature itself, as in Visions of the Daughters of Albion (“Printed by Will: m Blake : 1793.”), The Song of Los, The First Book of Urizen, The Book of Ahania, and The Book of Los. Instead, the date “1804” is etched in white line, i.e., incised on a large dark area, which could readily be done any time after 1804, if indeed the page as a whole was designed and etched that year rather than in 1808 or 1809 or whenever he did in fact begin Jerusalem. The addition of the 1804 date presumably was caused by his desire to link the two poems, as they seem to have been linked in their gestation, or perhaps to tie them both to a significant date in Blake’s life, the rebirth of inspiration following his return from Felpham and his acquittal at his January 1804 trial, or still more the mysterious experience of renewal at the Truchsessian Gallery in October 1804.
To place Jerusalem firmly in the decade of the 1810s sets it in a far more meaningful context of Blake’s life than that of the mid-1800s. It links the poem to the mood of contrition and forgiveness of his last work, as contrasted with the spirit of condemnation and self-justification of the ending of Milton. It also answers the question of what Blake was doing in the years from 1810 to 1818, when he virtually disappeared from the London art world and apparently produced little work of significance compared with that of the decades immediately preceding and following. Only the hundred plates of Jerusalem, in both its verbal and visual dimensions, fill this puzzling gap in the output of a major artist at the height of his powers. Its record of profound spiritual struggle and triumph in the face of worldly defeat was a labor calling forth all of Blake’s imaginative strength. Seen from this perspective, the composition of Jerusalem appears a striking example of what Charles Lamb was the first to describe as the sanity of true genius: as the creative means by which Blake regained his creative wholeness after the crisis of the years preceding 1810.