University of Virginia Library

The Composition and Revision of Coleridge's Essay on Aeschylus'Prometheus
by
S. W. Reid

Some thirty years ago Sir Edmund Chambers conjectured that Samuel Taylor Coleridge's 1825 lecture on Aeschylus' Prometheus Bound included material prepared in 1821 for his son Hartley.[1] To support his conjecture, Chambers referred to the substantial, but inconclusive, external evidence of the letters of 16 May 1821, 26 April 1825, and 19 May 1825.[2] Yet, documents recently brought to light make Chambers' reasonable guess convincing indeed, and it is the purpose of this article to show that the 1821 transcription of the essay originally written for Hartley was revised by Coleridge in 1825, and is the chief ancestor of the text of the 1825 lecture published in the Royal Society of Literature's Transactions and later in Coleridge's Literary Remains.

The evidence consists of two manuscript fragments, one of thirteen leaves and the other of two. The first of these, now in the Duke University Library, has hitherto been catalogued as "a collection of MS. notes on philosophical subjects by S. T. Coleridge"; the second fragment is two leaves of Notebook 29, otherwise known as the "Clasped Vellum Notebook," now in the Berg Collection of The New York Public Library. The thirteen leaves at Duke and the two leaves in New York apparently were once together in Notebook 29, and reunited they contain the complete text of the essay as it was prepared for Hartley in 1821.

The Duke manuscript contains not only the major part of the essay in its early form but also the revisions made by Coleridge for his lecture in 1825. Its thirteen leaves are gathered in quires of two (conjugate) leaves each; thus the manuscript is constituted of six gatherings plus a disjunct leaf, folio 1. The watermark reads eads & pine 1802. The leaves, unnumbered, are roughly bound in a coarse brown paper. According to the records of the library's manuscript division, they arrived at Duke in this wrapper when the manuscript was purchased; this information is verified by the fact that mathematic equations which appear on the wrapper in Coleridge's hand were included in the prefatory remarks which he wrote for the 1825 lecture.[3] Folio 1 recto contains the words 'Mr John Anster | 30 Thornhaugh

Street | Bedford Square' in Coleridge's hand in the upper right corner.[4] Across the page are the words 'by the Author, S. T. Coleridge' apparently in the hand of Joseph Henry Green, Coleridge's amanuensis.[5] The main text, in Green's hand, begins at the top of folio 1 verso and continues to the bottom of folio 13 verso. The last sentence is incomplete. Revisions in ink and in pencil in Coleridge's hand occur throughout the manuscript up to the last third of folio 11 recto, where a new paragraph in the main text begins. Two other scraps of evidence, both uninformative, deserve mention: one, a sentence about the document on the wrapper in an unidentified hand, and the other, a brief description of the document in the hand of J. Dykes Campbell on a separate sheet that has been kept with the manuscript.

Inscribed in Green's hand upon two leaves of Notebook 29 in the Berg Collection of The New York Public Library is the remainder of the essay. This notebook was kept by Coleridge from 1814 to 1825, according to the Berg Collection's records. As customarily, Coleridge turned the book around when he had filled about half its pages, and consequently the text in question begins at the top of folio 118 verso with a fragment of a sentence, continues on 118 recto, and ends at the bottom of 117 verso. There is no inscription on these pages in any hand but Green's.

II

The evidence is unequivocal for concluding that these two leaves of Notebook 29 provide the continuation and completion of the incomplete text of the unrevised essay contained in the Duke manuscript. For example, the crucial sentence fragments (the one at the end of the Duke manuscript and the other at the beginning of the Berg) constitute a coherent sentence when joined together. Beyond this quite substantial evidence, there are the facts that Nature, the subject of the Berg pages, is clearly the subject of the final Duke pages and that the general theme of the essay is carried to its conclusion in the Berg pages. And conclusive evidence is found in the physical properties of the documents. Besides what one would normally expect — the facts that the documents match in their papers, in their formats, and in the general spacing of the inscriptions made by Green — one piece of physical evidence clearly indicates that the Duke pages formerly were bound in Notebook 29 just before folio 118. For between folios 119 and 118 a conspicuous separation in the binding of the book

occurs. From the fragments of paper clinging to the binding thread at this point, one plainly must conclude that leaves have been torn out of the book. These fragments belong to one and the same gathering of two leaves, and they adhere to the 119 side of the break. Naturally, one would look for the mates of these fragments in the initial leaves of the Duke manuscript. As previously noted, the initial Duke leaf is disjunct. Furthermore, it has been trimmed along its gutter edge in a straight line, and, as a result, it measures 7¾ inches from this inner edge to the outer edge. But all other leaves of both the Duke manuscript and the notebook measure 8 inches. Thus, the initial Duke leaf wants ¼ inch, and it is no accident, certainly, that ¼ inch is the exact width of the largest fragments adhering to the binding of Notebook 29.

It is safe, therefore, to assume that the Duke leaves were once bound in Notebook 29 between the folios now numbered 119 and 118, and that they were torn out and rebound after the entire essay had been copied into the notebook. For the word 'by' on the first page was covered when the leaves were rebound, and it is not at all likely that a certain number of leaves were first torn out, the essay subsequently copied onto them, and the copying continued on the succeeding leaves yet part of the book. We may conclude, then, that Green transcribed Coleridge's essay into Notebook 29 and that either he or Coleridge failed to extract folios 118 and 117 when tearing out the thirteen leaves. These leaves were thereupon bound as a unit. The continuation and completion of the text of these leaves is provided by the pages remaining in Notebook 29; the entire extant document of the essay in its early form — as conceived by Coleridge and transcribed by Green in 1821 — has now been recovered.

Unfortunately, it is not quite as apparent that this early form of the essay was composed in 1821 to aid Hartley in writing his poetic adaptation of Prometheus Bound, for we have no manifest proof that it was. Although second-person pronouns appear, Hartley is not addressed by name in the manuscript, and neither he, his father, nor anyone else whose mere word might have sufficed has identified this particular manuscript with the essay reportedly written for Hartley in or around May 1821. However, the evidence which does exist suggests strongly that this essay and our manuscript are identical.

We know, for instance, that the transcription of the essay must have been made between November 1817, when Coleridge met Green, and early 1825, when Green ceased acting as amanuensis for Coleridge.[6] It is not likely, therefore, that Green made the transcription for the 1825 lecture. Nor, to take another tack, is it likely that Coleridge composed the essay prior to 1819. Before this year there is no mention of a composition on Aeschylus' Prometheus, whereas Coleridge's 1818 letters abound with references

to the work on the two courses of lectures of 1818 and the revision of The Friend. This evidence, then, indicates that the essay in its earlier form was composed and transcribed sometime after 1818 but before early 1825.

Other evidence points more precisely to a probable date of composition and transcription. The letters of Hartley and of his father reveal that in early 1820 Hartley was contemplating a translation of Aeschylus in which the Prometheus, he wrote, would "serve as a sort of text, for some observations on the sacerdotal religion of Greece, and on the sources and spirit of mythology."[7] Hartley's project was to be entered in competition for the prize essay at Oriel. Greek drama was on Coleridge's mind also, and he mentioned specifically Hartley's and his own work on an essay dealing with the prosody of the Greek dramatists, though he seems to have supported the Prometheus project as well.[8]

These plans were upset, of course, by the dismissal of Hartley from Oriel in the spring of 1820, ostensibly on account of his unscholarly ways and his weakness for drink. By August 1820, his father had settled him with Basil Montagu in Bedford Square. There he was to continue his work on Greek drama. Though Coleridge continued for a while to concentrate on the essay on prosody,[9] Hartley's interests turned him in the direction of a poem based on Aeschylus' Prometheus, and on 19 February 1821 he wrote to his brother Derwent, "Prometheus prospers" (Letters of Hartley Coleridge, no. 18). Apparently it did indeed prosper, for according to Derwent the poetic fragment was a principal occupation during Hartley's stay in Bedford Square and had been completed virtually, though perhaps not in every detail, by May 1821, before Coleridge wrote his essay on the metaphysics of Aeschylus' tragedy.[10]

Derwent's testimony is confirmed by Coleridge's own remarks on the subject. He wrote to Derwent in a letter postmarked 16 May 1821:

Set beside this statement, the words of the manuscript immediately following the point where the revisions cease and where the lecture ends take on substantial significance. Clearly the essay in its early form was composed for someone close to Coleridge who was writing a poem the "subject" of which was Aeschylus' Prometheus;[12] clearly his father knew that Hartley was writing a poem on this subject at this time. Moreover, it is evident that the manuscript additionally fits the description that Coleridge gave of his essay: the accumulated materials amount to "a small volume almost" on the subject,

As a matter of fact, these accumulated materials seem to have crippled the wings of Hartley's Pegasus, as Coleridge himself intimates,[13] and indeed Derwent, who was kept informed throughout of the project and was at Highgate with Coleridge and Hartley in the summer of 1821, affirms that the views of the Prometheus expressed in Coleridge's lecture are identical with those which in May 1821 discouraged Hartley's poetic efforts.[14] Since, as will be indicated shortly, the manuscript contains the germ and much of the substance of the lecture, Derwent's affirmation is plainly tantamount to an identification of the essay of 1821 with that of the manuscript at Duke.

Whether or not the manuscript at Duke was the very one given to Hartley is another matter. The evidence is not conclusive but indicates that it was. Coleridge's principal aim was to present Hartley with the "small volume," and the manuscript appears to be just that. The first page, containing the words 'by the Author, S. T. Coleridge' but no text, gives the effect of a title page, and had the purpose of Green's transcription been to preserve a copy of the essay in the notebook for Coleridge's use, it is unlikely that the text would have been begun on the second page and that Green would have identified the author. The wrapper, which contains Coleridge's inscription of equations that later became part of the lecture

(Works, p. 347), had certainly been added to the manuscript by the time Coleridge revised it. Presumably Coleridge, accidentally overlooking the final two leaves of the essay, tore out the Duke leaves, had them bound, handed the manuscript to Hartley in 1821 during one of his visits to Highgate, and in late 1822 repossessed the manuscript before Hartley left London for a teaching position in the Lake Country. All that we know, then, points to the conclusion that the leaves at Duke and those at the Berg Collection constitute the "small volume" written for Hartley in 1821.

III

This early form of the essay was revised by Coleridge for his lecture before the Royal Society of Literature in 1825. A comparison of the manuscript with the printed lecture reveals that the text of the personal essay formed the basis of the public lecture and that this second and final form of the essay incorporated most of the holograph revisions in the Duke manuscript. Not only did almost all of these alterations turn up in the text of the lecture, but the undeleted text of the essay also became part of the text of 1825. One would not wish to have to argue in each of these cases of agreement between the substantive readings of the manuscript and of the lecture that the lecture's reading originated independently of the very same form in the manuscript. When Coleridge stated at the end of the lecture that he could "trace the persecutions, wanderings, and migrations of the Io, the mundane religion" in a "future communication," he was no doubt thinking of the passage on this very subject in the unpublished part of his earlier text.[15] Both the lecture and the holograph revisions in the manuscript terminate at the end of the tenth section. It may even have been the relatively extensive personal reference to Hartley in the first sentences of the following section which made Coleridge decide to put a stop to his discourse at this point. In short, there is every reason to believe that Coleridge revised the initial text of the manuscript and then made it the main body of his lecture for the Royal Society of Literature, and the utter lack of evidence of Coleridge's attention to the essay at any time other than during the composition of it and the use of it for the lecture obliges us to conclude that the revision was undertaken specifically for the purpose of the lecture.

Two distinct stages of revision in the manuscript can be identified. Most of the alterations are in a dark ink, and these occur on folios 1 verso through 11 recto. Some alterations in pencil followed these on a few pages (folios 2 verso through 4 recto, 9 recto, and 11 recto). Of this order there is one piece of evidence that alone practically constitutes demonstration. It involves the only instance of revision in which the very same change has been made in both ink and pencil. On folio 2 verso, Coleridge in ink has

mended a 't' to a 'T', but in doing so he has not made the new inscription very legible; at the same place he has satisfactorily traced another 'T' in pencil. Evidently Coleridge thought the inscription in ink not sufficiently clear and therefore sought to improve it by retracing it. Fortunately, the inference that the alterations in ink were made before those in pencil is supported by other evidence in the manuscript. For example, in one case considerable alteration in pencil is interrupted at a passage which has been deleted during revision in ink, and in another certain elements of a passage have been altered in ink, and the whole passage, specifically including some of these alterations, has later been marked in pencil for revision.[16]

In addition to the revisions contained in the manuscript itself, more extensive alterations and new passages must have been written out on other sheets and combined with the revised manuscript text in a fair copy which was turned over to the Royal Society of Literature. Coleridge had accepted appointment to the Society in March 1824, but he apparently did most of the work on the lecture in early 1825. On 8 April he was busy with the text but envisioning completion of his task, and two and one half weeks later he was finished (Unpublished Letters, 333, 345, 346; Letters, II, 737-738). Delightedly aware of its difficulties, he delivered the lecture on 18 May 1825 (see Unpublished Letters, 351; Letters, II, 739-740).

It has long been thought that the lecture was printed at this time and that at least one copy of this earliest edition survived. These suppositions are based on T. J. Wise's statement in A Bibliography of the Writings in Prose and Verse of Samuel Taylor Coleridge that a "Private Edition was printed and circulated in the customary manner, in advance of the reading" and upon his description of a copy of that "Edition" once in his possession and now in the British Museum. According to Wise, only twenty-five copies were printed.[17] No external evidence has ever been advanced to support Wise's assertion.[18] But internal evidence gained by collating the issue of the lecture described by Wise as a "Private Edition" and the alternate issue of the lecture in the Society's Transactions (1834) proves that Wise's statement is false.[19]

Collation of the two issues of the lecture on the Hinman Collating Machine revealed, first of all, the fact that the pages throughout the lecture

were printed from the same setting of type. In addition, I noted a number of aberrations such as would only arise from broken or worn type, and I found no textual variants. These findings clearly can lead to only one conclusion: the two issues of the lecture, printed from the same type setting, are distinct and distinguishable impressions and issues of the same edition printed contemporaneously. The Yale copy of the lecture (as well as Wise's description of this issue) reveals that the essay was at one point printed as a discrete unit, beginning with B1 and continuing through D3 (quarto); on the other hand, when issued in the Society's Transactions, the essay began on 3C2 verso and ended on 3E4 verso. Since the lecture was printed from variant impositions, standing type — and not stereotype plates — must have been used in printing the second issue. It is, of course, preposterous to assume that the type of this lecture was kept standing over a period of eight years. Hence we must conclude that both issues of the lecture were set up and printed in early 1834.

There is really no evidence for any further conclusions, such as the order of these issues through the press. Aberrations of the kind referred to above are almost all identical, and those few that vary in one edition or the other constitute conflicting evidence supporting one order of printing as much as the other. Of course, it might be argued that the lack of evidence of deterioration itself tends to point to the "Private Edition" of twenty copies as the one with priority, but such negative evidence is of a low order of probability indeed and fortunately does not have to bear any textual weight, since there are no textual variants at stake. All we can say is that one issue, possibly the private one, was set up and printed not earlier than the first months of 1834 and that thereupon the type pages were re-imposed to print the lecture for alternate issue.

The details of the composing, transcribing, revising, and printing of Coleridge's lecture are now before us, and the complete essay in its early form has now been recovered. We have traced the essay from Coleridge's composition in May 1821 for Hartley, through J. H. Green's transcription in Notebook 29 and his or Coleridge's oversight of folios 118 and 117 when extracting the Duke leaves, to Coleridge's revision of this bound "small volume" in 1825 for his Royal Society of Literature lecture. The printing of the lecture in 1834 closes the history of the producing of the essay in its authoritative forms, and there now remains the task of editing the authoritative documents for publication. When publication of both versions appears, it should allow us to view one of our great nineteenth-century writers developing an idea over a span of years, explicating and revising his explication of one of his favorite topics, the Prometheus myth.

Notes