Charles E. Robinson: Introduction to The Frankenstein Notebooks (Garland, 1996)

Mary Wollstonecraft Shelley (1797–1851) was born Mary Wollstonecraft Godwin on 30 August 1797 in London; she was the daughter of two famous writers, William Godwin (1756–1836) and Mary Wollstonecraft (1759–1797), who died eleven days after her daughter Mary was born; she eloped with the married twenty-one-year-old poet Percy Bysshe Shelley (1792–1822) in July 1814, when she was sixteen (they were accompanied by Mary Shelley's slightly younger step-sister Claire Clairmont [1798–1879]¹ on a six-weeks' tour of France, Switzerland, Germany, and Holland); she gave birth to a premature daughter in February 1815 (who died twelve days later) and a son in January 1816. Less than four months later, the unmarried Mary Godwin together with Percy Bysshe Shelley, their three-months-old son William, and Claire Clairmont (unmarried but pregnant with the child of Lord Byron [1788–1824], who had recently separated from his wife) journeyed to Geneva, Switzerland, where they joined Byron and his doctor John William Polidori (1795–1821) for that famous summer of 1816, a summer that has since been celebrated by these writers' own recollections, by biographies, by fiction, and by film.

Among the many works conceived and/or written during that wet and stormy summer² were Byron's Childe Harold's Pilgrimage: Canto the Third, his drama Manfred, and his Vampire “Fragment”; Percy Bysshe Shelley's poems “Hymn to Intellectual Beauty” and “Mont Blanc”; Polidori's tale The Vampyre and his novel Ernestus Berchtold; or, The Modern Oedipus (both published in 1819); and Mary Shelley's Frankenstein; or, The Modern Prometheus. She began that novel as Mary Godwin in June 1816 when she was eighteen years old, she finished it as Mary Shelley in April/May 1817 when she was nineteen (she had married Percy Bysshe Shelley on 30 December 1816 after his first wife committed suicide), and she published it anonymously on 1 January 1818 when she was twenty. Now, exactly 180 years after she began writing Frankenstein, the publication of The Frankenstein Notebooks will provide a unique opportunity to study the creative process that went into the making of Mary Shelley's monster.

Many of those who have previously studied Frankenstein will know that there are a number of “texts” of the novel, each with significant textual flaws: the three-volume 1818 first edition,³ with its printing errors; the “Thomas” copy of 1818, with Mary Shelley's own holograph corrections and additions; the two-volume 1823 second edition (which used 1818 as setting copy), with at least 120 variant words occasioned by William Godwin, who had arranged publication of this edition while Mary Shelley (widowed since Percy's death on 8 July 1822) was still in Italy; and the one-volume 1831 revised edition (which used 1823 as setting copy), with its many changes that Mary Shelley herself introduced into the text. I say Mary Shelley herself and will from this point denominate her as MWS (as she herself did) thereby providing a shorthand to distinguish MWS from her husband PBS, who also had a considerable “hand” in the earlier 1818 text.

This edition of The Frankenstein Notebooks will introduce yet more “texts” of the novel, not all of which are extant: the now lost original 1816 ur-text or “story” that MWS wrote in the summer of 1816; the two-volume 1816–17 “book” or manuscript Draft that MWS wrote for approximately nine months from c. August 1816 through 17 April 1817, during which time PBS made and suggested alterations to the manuscript; and the three-volume 1817 manuscript Fair Copy (transcribed directly from the Draft) that MWS (with some assistance from PBS) prepared between 18 April and 13 May 1817 for the printers; the now lost proofs for 1818 that were set from the Fair Copy; and the now lost revises (that is, corrected proofs) for1818. The extant texts reproduced here in The Frankenstein Notebooks are the surviving manuscripts of the 1816–17 Draft and the 1817 Fair Copy, manuscripts owned by Lord Abinger and now on deposit at the Bodleian Library, Oxford University.

The Draft was originally written in two large (hard-cover) notebooks, here denominated Notebooks A and B, now disbound and consisting of a total of 152 leaves (with text on 301 pages) that, together with three insert leaves and two insert slips (with text on a total of 8 pages), account for approximately 87% of the 1818 text (missing are the leaves for Walton's four introductory Letters; for the beginning of Draft: Vol. I, Ch. [1]; for the end of Draft: Vol. II, Ch. 3; and for all of Draft: Vol. II, Ch. [4]). The Fair Copy was originally written in [?eleven] smaller (soft-cover) notebooks, the only surviving ones here denominated Notebooks C1 and C2. These two notebooks are now disbound: 19 leaves are missing; the 29 surviving leaves (with text on 58 pages), together with 1 insert leaf (with text on 1 page) as well as 1 substitute leaf (with repeated text on 2 pages), contain approximately 12% of the received 1818 text.

In an attempt to simplify the complicated sequence of these texts, I offer the following outline.

(For more information on each of these texts [including the dating thereof], consult the "Frankenstein Chronology." See also “Hypothetically Reconstructing an Ur-Text” and “Hypothetically Reconstructing the Fair Copy” below.

Ur-text: Draft(s) of a “story” ([?17] June–[August or September] 1816)—not extant⁴
↓
1816–17 Draft: 2-volume novel in two hard-cover notebooks ([August or September] 1816–17 April 1817)—most of Notebooks A and B survive
↓
1817 Fair Copy: 3-volume novel in [?eleven] soft-cover notebooks (18 April–13 May 1817)—parts of Notebooks C1 and C2 survive
↓
Proofs ([?23] September–[?3] November 1817)—not extant
↓
Revises ([?23] September–[?20] November 1817)—not extant
↓
1818: 1st Edition in 3 vols. (1 January 1818)—published by Lackington et al. in 500 copies
↓
↓ 1818 Thomas: “Thomas” copy of 1st Edition (corrected by MWS
↓ before July 1823)—survives at Pierpont Morgan Library
↓
1823: 2d Edition in 2 vols. (11 August 1823)—published by G. and W. B. Whittaker (but not set from 1818 Thomas)
↓
↓ [1826]: [?Re-issued 2d Edition] (4 April 1826)—apparently issued by Henry Colburn
↓
1831: Revised Edition in 1 vol. (31 October 1831)—published by Henry Colburn and Richard Bentley in 4020 copies

A listing of all the variants (both accidental and substantive) that exist among these extant texts is beyond the scope of this (and, possibly, any) codex edition, and no editor has ever collated all the accidental variants of punctuation, capitalization, and spelling between any two editions. However, nearly complete collations of the substantive or word variants between1818 and 1831 have been published: see the collations in 1818 Rieger (pages 230–259 [and page 288]) and in 1818 Macdonald (pages 317–359)—Rieger but not Macdonald incorporated the readings from 1818 Thomas in his text; for a shorter collation that is prefaced by a useful summary of the type of changes involved, see 1818 Butler (pages 199–228); and, for the most comprehensive collation, one that incorporates the autograph corrections in 1818 Thomas and that also includes substantive variants in 1823 as well as 1831, see 1818 Crook (pages 182–227).⁵ In this edition, I offer for the first time the parallel texts by which the extant parts of both the 1816–17 Draft and the 1817 Fair Copy can be collated with the 1818 text, and Appendices A and B provide a means for other collations (Draft with extant Fair Copy; and PBS Fair Copy with MWS re-transcription of the Fair Copy).

It should be stressed that these collations and parallel texts are valuable tools by which a reader can and should learn that a text or literary “product” involves a literary “process” that must be understood before any attempt is made to “fix” a text, whether “fix” is understood to mean “establish” or “make firm and unchanging” or “emend and therefore change.” In the case of Frankenstein, the substantive changes that MWS made in her revised edition are so extensive that many teachers and students of Frankenstein consider 1818 and 1831 as two different novels. Among the significant “alterations” made in 1831 are the following: a transformation of Elizabeth from a cousin to a foundling having no blood relation to Victor; a new chapter to accommodate a much longer exposition about Victor's childhood; and a more explicit and earlier introduction of the doppelgänger theme when Victor alludes to Aristophanes' myth of the circular and then divided primal human beings in Plato's Symposium.⁶ The already published collations between 1818 and 1831 provide an opportunity for any reader of Frankenstein to experience the creative process by which MWS transformed an already published edition into another text, and this edition of The Frankenstein Notebooks extends that opportunity backwards: it publishes for the first time the extant portions of the 1816–17 Draft and the 1817 Fair Copy (both printed parallel to 1818) so that the reader can “see” the transformations of one text into another and, at the same time, “see” the degree of PBS's involvement in these transformations.

In addition to collations and parallel texts, there have been other studies that have addressed the texts of Frankenstein by way of the manuscripts, but each of these has been flawed, in one way or another, because the full texts of The Frankenstein Notebooks, here published for the first time, were not available to benefit the earlier studies. As this edition will make evident, much of the criticism to date on Frankenstein will have to be re-evaluated in light of the collaboration between MWS and PBS, a collaboration that can now be “seen” in this edition by way of the different fonts used to distinguish the Shelleys' often similar manuscript hands. Those who have previously attempted to make those distinctions include James Rieger, E. B. Murray, Ann Mellor, Marie Hélène Huet, and David Ketterer, each of whose contributions will be briefly addressed below in “MWS and PBS's Collaboration in The Frankenstein Notebooks.” Because so many critics of the novel have made factual errors (repeating, e.g., that Frankenstein was published in March 1818 rather than its actual publication date of 1 January 1818), I also refer the reader to the "Frankenstein Chronology;" for a record of the facts about the novel as they can best be determined. Some readers may prefer to go immediately to that Chronology in preparation for reading The Frankenstein Notebooks.

Manuscript Photofacsimile Pages (versos)

These manuscript pages originated in the following notebooks: the 1816–17 Draft in hard-cover Notebooks A (in Dep. c. 477/1 and Dep. c. 534/1) and B (in Dep. c. 534/1); the 1817 Fair Copy in soft-cover Notebooks C1 and C2 (both in Dep. c. 534/2). All of the covers have long been discarded: those on the Draft Notebooks possibly in April/May 1817 when the Draft was transcribed into the Fair Copy; and those on the Fair-Copy Notebooks most likely in September–November 1817 when the Fair Copy was used as printer's copy.

Headers

The diagram below may simplify what is a set of very complicated relationships, but I ask the reader's patience to consider first that the two physical Notebooks A and B are paginated not as notebooks but as a two-volume novel. Consider, for example, that the draft of Volume I of the novel concludes not at the end but rather in the midst of Notebook A, at a point (MWS's page 160, a verso) where MWS would soon turn another page: she left blank her next page 161, a recto; then she immediately began the draft of Volume II on the next page (a verso), which she paginated with the number “1” although the preceding page in the same notebook was “161.” (A verso is an unusual place both for a page 1 and for the beginning of a new Volume, either of which would ordinarily be graced by a recto or a right-hand page.) Complicating this matter further is that the division between the two Bodleian deposits of The Frankenstein Notebooks in 1974 and 1976 comes close to but does not quite equal the division between Volumes I and II of the Draft novel. * * * * * Put simply, the Abinger and Bodleian division of Dep. c. 477/1 and Dep. c. 534/1 does not equal the physical division of Notebooks A and B, nor does either of these divisions quite equal MWS's conceptual division of Draft Volumes I and II. These divisions, yet again complicated by the three-volume division of the 1818 first edition (Volume and Chapter numbers may be found at the top of each transcription page) as well as by the two-part division of this edition of The Frankenstein Notebooks (revealed in part by the italicized pages 266 and 268 in the above illustration), demand the following diagram, arranged in chronological order:

Bodleian Foliation

Although all leaves of The Frankenstein Notebooks are now foliated in pencil at the top right of each recto (and are cited as reference points in the headers, textual commentary, and Quiring Charts), the photofacsimiles of the manuscript leaves (photographed in 1993) reproduce the Bodleian folio numbers only in Dep. c. 477/1 (deposited in 1974 and then foliated by Margaret Crum); Dep. c. 534/1–2, although deposited in 1976, was not foliated until 31 January 1995, when, after a discussion about a suitable ordering of the leaves, Bruce Barker-Benfield penciled the folio numbers onto each recto. The Library's purpose in adding such foliation was not to dictate an order of leaves (the presence of insertions allows for several alternatives) but to enable a neutral mode of citation for each surviving piece of paper.

Notes

Notes that address the physical conditions of a manuscript page are positioned with all other textual notes at the bottom of each transcription page.

The remnants of the original Frankenstein Notebooks tell a fascinating tale about the way that MWS created her novel. Part of this tale is evident in the preceding advice on how to use this edition, and the fullest expression of the narrative about these notebooks is to be found in the Frankenstein Chronology; that concludes this Introduction. Here, however, the technical details about the number and makeup of these notebooks at least provide an outline to the creative process involved in the making of Frankenstein. The technical details concerning each of the four extant notebooks are presented separately below, but it is worth mentioning here that all four of these notebooks in which MWS drafted and then fair copied her novel in 1816–17 were originally bound; that the bindings and the covers of all four notebooks have long since disappeared; that the surviving manuscripts in the Abinger papers are all loose leaves, all of which are “singletons” (i.e., single leaves), with the exception of two bifolia that are still joined; and that watermarks, torn edges, glue residue along some edges, sewing holes, and wet offset ink blots provide sufficient evidence by which to reconstruct the original makeup of the quires in each of these notebooks. Moreover, the very nature of the writing in these notebooks proves that MWS drafted and fair copied her novel (and that PBS made his changes and suggested his alterations) while the notebooks were still intact: each main group of now detached leaves includes at least one reconstructed bifolium where the writing overlaps across the two halves; the writing at what would have been the inside edges of some versos displays a wavering and a trailing off, indicating that the writer's hand was impeded by the gutter of the still bound notebook; the writing at the outside edges of the rectos frequently carries over to the outside edge(s) of what would have been the notebook's next leaf (or leaves), a very natural effect when a large volume is splayed open during the process of writing; now missing leaves that were apparently discarded during the composition process have sometimes left wet offset ink blots, suggesting that the leaves were torn from a bound notebook; and different leaves from later quires or from other sources were sometimes inserted into the main sequence of pages in such a way (leaving the evidence of wet offset ink blots or pin holes) to indicate that the main sequence of pages was bound in a notebook. All of this evidence leads to the inescapable conclusion that Frankenstein was drafted and fair copied in notebooks that were purchased as ready-made, bound blankbooks from commercial stationers in England or abroad. Further details about the size and source of the notebooks, the type of covers, the number of pages in each quire, the watermarks, and other particulars are provided separately for each notebook below.

§ Naming in The Frankenstein Notebooks

It is dangerous to speculate about the ur-text of Frankenstein, for it could have been very short and inconsequential; and it is possible, therefore, that everything mediating between such an ur-text and the Draft was nothing more than now discarded rough drafts of pages and sections that never had any integrity or identity as a separate text. However, there are clues in the Draft that MWS may have been copying from and expanding upon what she called her “story”—an ur-text of some length that she had written in the summer of 1816. The remarks on the names above suggest that Frankenstein spent his childhood with Myrtella and Carignan in that ur-text, and that maybe his mother's maiden name was Beaumont. However, it should be stressed that every changed name does not an ur-text character make. Other evidence suggests, e.g., that the very long chapter about “Maimouna” (who then became “Amina” and then “Safie”) was not drafted or even conceived until late November or early December 1816.

It is easier to speculate about what was not in the ur-text than to prove what was there. One way for MWS to enlarge her “story” into a “book” or novel was to add new characters and incidents to the outside and to the inside of her frame tale. That is, it is likely that neither the Captain (Walton from the outside narrative frame) nor the Arabian (Safie from the innermost tale) was in the original ur-text—and there is no persuasive manuscript evidence to suggest that they were.³⁰ Whether the other parts of the De Lacey episode were contained in the ur-text, with a possibly different mechanism for the monster learning to speak and read, I cannot determine. Everything else, however, may have been sketched out in one form or another in the summer of 1816. MWS tells us that the tale began with the line “It was on a dreary night of November.” Reasoning from that fact, we can speculate that the ur-text had three distinct parts (although it is impossible to determine just how long each one was): the creation scene; the scenes before the creation; the scenes after the creation. Evidence for each of these parts in the ur-text comes from the following sources: (1) MWS's statements in her Journals, her letters, and her 1831 Introduction; (2) mistakes in transcription that are best explained as copying rather than drafting errors; (3) a Draft page or chapter that underwent only minimal cancelation and revision when it was initially drafted into the Notebook (suggesting that it was an intermediate Draft that had been copied from an earlier version); (4) a Draft page or chapter with relatively full and/or precise pointing, suggesting that MWS and/or PBS had worked through the punctuation in an earlier complete draft; and, finally, (5) references in a part of the ur-text (hypothetically reconstructed by means of the evidence listed above) suggesting that other characters and incidents were elsewhere in the ur-text. With these evidences in mind, I offer the following observations about the ur-text, again reminding the reader that some of the evidence of an earlier state may merely mean there were intervening rough drafts rather than an ur-text.

The Creation Chapter

The creation chapter (Draft: Vol. I, Ch. 7a) is the one for which there is the strongest likelihood of its having existed in the ur-text. MWS indicated in her 1831 Introduction that she began her original tale on [?17] June (the morning after her dream) with the words “It was on a dreary night of November” (see transcription page 97 in this edition). These words may have begun the ur-text itself or, over the summer, they may have been preceded by introductory sections or chapters leading up to the creation scene. That the ur-text contained a description of that “dreary night” and its immediate aftermath is suggested by the relatively “clean” Draft: Ch. 7a, which has few MWS interlineations and cancelations made in the process of drafting (the additional interlinear and marginal alterations were a consequence of PBS's hand in the manuscript at a later time).

Chapters (or Characters and Incidents) before the Creation

If “It was on a dreary night of November” were preceded in the ur-text by other sections and if Ch. 7a was mainly a transcription of the creation scene and its immediate aftermath, then the other references in Ch. 7a to Ingolstadt, to the Swiss diligence, to Clerval, to Elizabeth, and to Victor's father, brother, and dead mother suggest that the ur-text would have also depicted Victor's adolescence in Geneva and his later experiences with Professors “K” and “W” in Ingolstadt. The relatively “clean” and accurately pointed pre-creation chapters (Draft: Vol. I, Chs. [l]–6) also argue for this ur-text about the young Victor and Myrtella and Carignan in Geneva. Two not-so-clean aspects of these early chapters are worthy of note here: the rough draft (folios 1–3) that MWS added to Notebook A in late October/early November 1816 offers further evidence that the originally drafted and “cleaner” Chapters were based on an ur-text; and any reconstruction of the drafting process must mentally eliminate the frequent and later interventions by PBS in the Draft. The existence of the relatively “clean” Chs. [l]–6, however, does not necessarily mean that they were mere copies of fully articulated chapters in the ur-text. That is, it is unlikely that an ur-text behind all of Draft Chs. [1]–7a (what originally occupied MWS pages [?34]–84) consisted of 50 pages; more likely it was no more than half that length, which would still have made it possible for MWS to expand her ur-text into her Draft without much cancelation and revision as she went along. Determining which of the pages or incidents or characters were expanded may prove impossible, but future speculation should be based on such concrete evidence as name changes and transcription mistakes.³¹

Chapters (or Characters and Incidents) after the Creation

If the ur-text contained most of the creation chapter (Ch. 7a) and some of the preceding chapters (Chs. [l]–6), it is likely that the ur-text also dealt with the consequences of the creation. If we use the same standard of a “clean” text for Draft: Vol. I, Chs. 7b–12, then the incidents of William's murder and Justine's imprisonment and death were part of the ur-text, a possibility supported by apparent copying errors in the Draft (see especially the mistranscription of the name “Myrtella” in the Justine section on pages 180–181 in this edition; see also pages 156–157, lines 31 and 36). The next Chs. 13–14 have relatively more copying errors³² and suggest that MWS was drafting new material at the same time that she was copying some things from earlier incarnations, possibly from the ur-text. An index of the kind of copying from an earlier text is to be found at the end of Vol. I in Notebook A (see note on transcription page 253 in this edition). MWS had finished drafting Ch. 14 (the last chapter of Vol. I) on three leaves that she paginated 153–157; for some reason she tore pages 153–156 (2 leaves) from the notebook, canceled the concluding paragraph she had just written on page 157 (which she then renumbered “153”), and then drafted a second and longer conclusion to this last chapter of Vol. I on pages 153–160. Although the shorter version originally written on pages 153–156 is missing, the extant pages 153–160 reveal what rewritten and expanded pages would look like when they are based on what was apparently a fairly good draft.

If, as is likely, parts of Chs. 13 and 14 (which record Victor's excursion to Chamounix) were in the ur-text, then MWS probably wrote it while she was still in Switzerland, possibly when she was at Chamounix (see "Frankenstein Chronology;" for 21–27 July and for 24 July 1816). If that were the case, then in late July and early August she may have written the first portions of the monster's narrative, the Draft of which dramatically opens Vol. II, where Chs. 1–3 have the same “clean” and well-pointed text by which an ur-text could be inferred. If that were the case, then the ur-text had at least introduced the De Laceys into the monster's life. After that, inferences get much murkier, because what appear to have been original pages 22–61 of Vol. II (which linked Notebooks A and B) are not extant. That there was no ur-text for much of these now missing pages is suggested by three circumstances: the missing pages suggest a trauma in the newly conceived text that required considerable drafting and redrafting; the missing pages were the first to introduce the story of Safie (initially named “Maimouna”), the innermost character whom MWS apparently created to expand her “story” into a “book”; and it appears that MWS was excited enough by writing this new section about the Arabian to announce to PBS on 5 December 1816 that she had “finished the 4 Chap. of Frankenstein which is a very long one & I think you would like it” (see "Frankenstein Chronology;" for that date). By hoping that PBS “would like it,” MWS is telling us that the Safie episode (and therefore the chapters in which the monster learned to read) were not in the ur-text. That all of this text was relatively new is also suggested by the rewrite Inserts “X” and “Y” (see pages 314–341 in this edition). In short, the missing forty pages and the rewrite sections are probably typical antecedents to the kind of “clean” Draft that makes up most of Notebook A.

Notebook B is a very different matter, and the very fact that it does not have a vertical pencil rule creating a left margin may tell us something about how MWS saw the remainder of her task: that she was not working from an ur-text in the same way she had been in most of Notebook A. The evidence for such a distinction is circumstantial, but there is a clear difference between the two manuscript versions of MWS's 1819 novella “Mathilda”: her revised and later draft of “Mathilda,” which resembles something between a final Draft and a Fair Copy, has a distinct pencil ruled margin similar to Notebook A of Frankenstein, whereas the earlier and rougher Draft that MWS entitled “Fields of Fancy” was drafted on pages that much more resemble Notebook B of Frankenstein, having a margin with no rule and a text in a greater state of revision and flux.³³ It is possible that this distinction between an unruled and a ruled set of pages might have enabled MWS to keep her drafts in order as she transcribed one from another.

How much of Notebook B was in the ur-text is difficult to conjecture: the sometimes “clean” text might mean that the monster's confrontation with the De Laceys (minus Safie) might have been sketched out in the summer of 1816—as most likely was the monster's murder of William (if it is inferred that Justine's trial was part of the ur-text). Proceeding further down this road would transform the ur-text into a mini-novel of some length, with only the Captain on the outside and the Arabian on the inside missing from what MWS had drafted in Geneva—that is to say, the ur-text may also have contained a skeletal form of the chapters in England and Scotland where Victor began making a mate for the monster (Draft: Vol. II, Chs. 11–15) as well as of the chapters back in Geneva and Evian where the monster murdered Elizabeth on her wedding night (Chs. 16–17). If these chapters were in the ur-text, the “busier” look to Notebook B (and, in many places, its more inexact pointing) suggests that MWS was doing much more than copying from or embellishing a well-developed ur-text at this point. Finally, arguing from the likelihood that the beginning of the Walton frame was never in the ur-text, we can conclude that most or all of the last scenes in the Arctic were not there as well. So after all this conjecturing, we end where we began, with the suspicion that neither Walton nor Safie were part of the ur-text, a text that may have contained in a shorter form much of the novel as it is represented by Notebooks A and B.

It is possible to deduce that MWS used eleven Fair-Copy Notebooks when she (and PBS) transcribed the Draft into a Fair Copy that would become printer's copy. The easiest way to argue that point is to note how close the pagination of the extant Fair-Copy Notebooks C1 and C2 was to the pagination in the 1818 edition: Fair Copy: Vol. III, Ch. [V], page 99 = 1818: Vol. III, Ch. V, pages 100–101 Fair Copy: Vol. III, Ch. VII, page 175 = 1818: Vol. III, Ch. VII, pages 176–178. Because MWS took painstaking care in transcribing her text into Notebooks C1 and C2, we may surmise that the same care and precision were employed in earlier notebooks and that she fair copied her Draft in the same way, with a new pagination sequence for each volume and with a relatively rigid number of words and lines per page, per leaf, and per notebook. If, then, the page numbers of the Fair-Copy Notebooks (even allowing for the blank spaces at the ends and beginnings of chapters, a characteristic of the notebooks as well as of 1818) were always close to the page numbers in 1818 (which had 181 pages in Vol. I, 156 pages in Vol. II, and 192 pages in Vol. III), then there would have been approximately 529 pages (181 + 156 + 192) in the Fair Copy. If the other notebooks were identical or similar to C1 and C2 (each of which had 48 pages), then the total of eleven notebooks would have proved just right: 11 x 48 = 528. It almost seems too perfect that the one extra page in 1818 meant that PBS had to find one extra page (folio 30 in Dep. c. 534/2) in order to write the last page of the Fair Copy. Such a calculation is, of course, too perfect, because we know that MWS and/or PBS made additions to the proofs (that is, there was text added in 1818 that was not in the Fair Copy). And, of course, MWS and/or PBS may have totally eliminated (rather than replaced) text in the proofs that had been in the Fair Copy. But neither of these actions cast much doubt on the approximation here, that eleven Fair-Copy Notebooks would have been sufficient to contain the text that produced 1818. In fact, there is evidence that some of these notebooks had 52 rather than 48 pages—in each of those cases, a notebook consisted of thirteen folded sheets that made up 26 leaves or 52 pages.

Further proof that the pagination in both Fair Copy and 1818 were nearly identical is revealed by the marginal numbers that MWS and PBS entered on the Draft pages in April/May 1817 when they were recording and estimating the proportions of the three-volume Fair Copy that they were creating out of a two-volume Draft. The meaning of all these numbers eluded me until I recognized that the Fair-Copy pagination approximated the 1818 pagination, enabling me to offer the following gloss for these otherwise inexplicable numbers.

“52”

This PBS number on Draft: Vol. I, Ch. 7a, page 80, apparently records the “52” leaves or “104” pages that had been used to fair copy the Draft up to this point— the 1818 equivalent is page “105.” It is possible that the “52” meant that the Fair Copy up to this point had used two complete notebooks of 24 leaves each, plus an additional four leaves in a third notebook, but there is no end of chapter or like stopping point in the Draft to warrant the notation. It is more likely that “52” designated two complete notebooks of 26 leaves each. As will be the case for the numbers discussed below, a deduction based on number of words rather than pages would support the same conclusion: the number of 1818 words up to this point and the estimated number of Draft words (estimated because half of the pages up to this point are missing) would occupy approximately two 24- or 26-page notebooks (or ²/₁₁ of the novel) at MWS's average of 285 words per leaf evidenced by extant Notebooks C1 and C2 (the estimate has to be for a leaf because MWS wrote more lines on one side than the other—see “The Makeup of The Frankenstein Notebooks&rdquo).

“52” together with calculation up to a canceled “156” and an emphasized “180”

These PBS numbers on Draft: Vol. I, Ch. 7a, page 81, appear to calculate the number of pages that were used to transcribe all of Fair Copy: Vol. I, which amounted to 181 pages in 1818 (see transcription page 221 in this edition), a number suggesting that Fair Copy: Vol. I also had approximately 181 pages). If Fair Copy: Vol. I occupied 180 pages, then it would have filled approximately three and one-half Fair-Copy Notebooks: in PBS's calculations, one number “52” denoted 52 leaves that generated “104” pages (that is, two 26-leaf notebooks); another “52” appears to denote “52” pages (one 26-leaf notebook); the number “24” appears to denote 24 pages (that is, half of a 24-leaf notebook or almost half of a 26-leaf notebook); and the sum “180” denoted the estimated or actual total of 180 Fair-Copy pages used for Vol. I.³⁴ These calculations were very important to the Shelleys, who were attempting (in April/May 1817) to divide the two-volume Draft into a three-volume Fair Copy, a process that demanded accurate estimates of the number of pages per volume. Moreover, the evidence of Notebooks C1 and C2 suggest that MWS was preparing very fine fair copies to impress a publisher—and that (as a daughter of a publisher) she might have been writing to rule, transcribing in such a way (with faintly ruled lines or with a template) that she knew approximately how many words she was getting to each page, to each notebook, and to each volume (Godwin may have been equally aware of such matters when he entered page numbers of his drafts into his Diary).

“Ch V- 113”

This MWS number on Draft: Vol. I., Ch. 7b, page 87, was not written in the Autumn of 1817 and did not refer to the fact that 1818: Vol. I, Ch. V, began on page 114 (with the implication that there was a change from proof to revise, or that MWS got the number wrong by one); rather, because the Fair-Copy pages approximated the 1818 pages, this number reveals MWS recording in April/May 1817 that she was beginning this Fair Copy: Vol. I, Ch. V on page 113—that is, 113/180ths through Vol. I. (With a little more maths, it would be possible to estimate just what day between 18 April and 13 May that MWS made this entry.)

“20” and “15”

These [?MWS] numbers on Draft: Vol. I, Ch. 8, page 102, seem to be echoed by the PBS numbers of “20” + “37” = “57” on page 137 (for which, see below), but I am not certain what they mean. Although they may refer to the number of pages MWS had just transcribed, they more likely suggest that it took approximately twenty pages of Fair Copy to transcribe fifteen pages of Draft. That 75% relationship may have been become clear to the Shelleys as they continued to transcribe. Although the Fair-Copy pages are not extant at this point to offer exact percentages, there is a 76% relationship between the estimated 180 Fair-Copy pages for Vol. I and the 143 pages of Draft up to that point (if we, as the Shelleys might have done, forget about the 6 insert pages at the beginning of the extant Notebook A).

“20” + “37” = “57”

These PBS numbers on Draft: Vol. I, Ch. 12, page 137, seem to echo the “20” + “15” on page 102 (for which, see above). It is possible that “57” may have something to do with the 57 pages between page 80 (where PBS first entered a number) and this page 137—or with the 58 pages between page 102 (where MWS last entered numbers) and page 160 (where Vol. I. ends).

“100” and “130” with calculations to a sum of “205”

These PBS numbers on Draft: Vol. I, Ch. 14, page 152, seem to indicate that MWS had used “205” Fair-Copy pages, an amount that approximates the 1818 pages used to this point: 181 pages (in Vol. I) + 21 pages (in Vol. II) = 202 pages. This number also suggests that MWS had just finished the 4th Fair-Copy Notebook (three notebooks of 26 leaves each [=156 pages] plus one notebook of 26 or 24 leaves [=52 or 48 pages] would equal 208 or 204 pages).

“169” and “191” and the sums leading up to these numbers

These two sums on Draft: Vol. II, Ch. [6], page 62 (the first extant page of Notebook B), seem to be estimates of the number of pages that will be occupied by Fair Copy: Vol. II and Fair Copy: Vol. III. MWS started her sums leading up to “169” with the number “98” and ended it with “97”—one of these numbers suggests the now missing Fair Copy: Vol. II, Ch. [VI], page 97 or page 98 (a number in tandem with 1818: Vol. II, Ch. VII, page 98, where Ch. VII begins (see transcription page 343, which is the first extant page of Notebook B). The sum of “169” either records or predicts the total number of pages in Draft: Vol. II (in that 1818: Vol. II had a total of 156 pages).

PBS seems to have used a similar form of calculation to reach his sum of “191,” a number that most likely records or estimates the 191 pages in Fair Copy: Vol. III (in that 1818: Vol. III had 192 pages). The fact that these important numbers are on the first extant page of Notebook B suggests that this page 62 was in fact the first page of Notebook B, at least in April/May when the Shelleys transcribed the Draft into a Fair Copy (and when they were doing these calculations). This conclusion supports my hypothesis elsewhere that the missing pages that at one time joined Notebooks A and B were probably from Notebook A.

“310”

This PBS number (which could possibly be “320”) on Draft: Vol. II, Ch. 8, page 87, is most likely a record of just how many pages had been used in the Fair Copy up to this point, a number confirmed by the number of equivalent 1818 pages: 181 (in Vol. I) + 136 (up to this point in Vol. II) = 317. This number may also tell us that the Shelleys had just finished their 6th Fair-Copy Notebook: we already suspect the first two notebooks used 52 pages each, and the third notebook may have had 52 pages, yielding 156 pages in all—if the next three Notebooks had the same number of pages, the six notebooks would have had a usable total of 312 pages, 310 of which were used to transcribe the Draft up to this point. Even if the number were “320” rather than “310” and the transcription had continued into the next notebook for a few pages, the estimated 529 pages of Fair Copy (based on the total number of 1818 pages) minus 310 pages (or 320) already used, meant that there would be 219 (or 209) pages yet to transcribe, an amount that could have been accommodated in the five remaining notebooks (#'s 7–11), even if all of them were 48-page notebooks: 5 notebooks x 48 pages = 240 pages.

A final word on PBS's involvement in the Fair Copy: because he recorded most of the numbers discussed above and made most of the calculations, it may look like he was involved in transcribing the Draft into the Fair Copy at these points. However, all other evidence suggests the contrary: PBS would not have wanted his hand to be in a Fair Copy that was to be kept anonymous; and the kinds of embellishments that PBS introduced into MWS's text when he did fair copy the Draft at the end of Notebook C2 are not present in the 1818 text around the places where PBS made the computations. Consequently, it is more likely that the PBS numbers are there because MWS asked him to make the computations for her as she fair copied her Draft in April/May 1817.

The computations that MWS and PBS performed in the margins of the Notebooks A and B indicate what numbers can do (and reveal). Other numbers (e.g., the very exact transcribing pattern of 19 lines on one side of a Fair-Copy leaf and 14 lines on the other, yielding an average of 285 words per leaf—see transcription page 649), together with other evidence, lead to the conclusion that there were eleven Fair-Copy Notebooks. And the average of 285 words on each Fair-Copy leaf permitted accurate estimates of what was written on now missing pages from Fair-Copy Notebooks C1 and C2. In short, much of what is missing from The Frankenstein Notebooks can be reconstructed from available evidence.

In other places, MWS used a different form of pagination (citing earlier pagination) to leave herself a trail through the convoluted Insert sections “X” and “Y” (see pages 314–341). She also left a trail of numbers “1” through “5” in Draft: Vol. II, Ch. 10, pages 416–441, revealing the sequence of revisions that she made to her Draft between 10 and 17 April 1817. Finally, the accident of MWS writing on a slip of paper on the verso of which was a reconstructible postmark of 9 April 1817 sheds a great deal of light on the makeup of The Frankenstein Notebooks.³⁵

The specific and important date of 9 April 1817 calls to mind an important moment in the text of Frankenstein, when Victor related his story to the magistrate “briefly, but with firmness and precision, marking the dates with accuracy” (see transcription page 577 in this edition). As is so often the case, what happens inside the text reflects what happens outside or with the text; that is, MWS herself in The Frankenstein Notebooks marked the dates with precision, and the alterations she made in the dates (together with other clues) reveal that she was very concerned with the chronology in her novel. Even though Walton's letters are dated with the decade of the eighteenth century missing, MWS has given enough information for us to determine three important points about the chronology: that the novel begins with Walton's letter dated 11 December 17[96] (which is, or is very close to, the day on which MWS imagined she was conceived);³⁶ that Walton's tale takes the gestation period of 276 days in the telling; and that the novel ends with Walton's last letter of 12 September 17[97], thirteen days after MWS was born and two days after her mother died. The clues to this reading of the chronology have long been available in the printed texts of 1818 and 1831—what the Notebooks supply is graphic evidence that MWS, like Victor, was carefully “marking the dates with accuracy.”

A perpetual calendar gives more than enough evidence to establish a chronology for all the plots in Frankenstein (although there is one major inconsistency). Walton leaves a clue in the outside frame when he remarks that he first saw the monster on “Monday (July 31st)” in the year “17—” (see Walton's Letter IV on transcription page 5); among the years in the eighteenth century when 31 July fell on a Monday is 1797, a year that is confirmed by a number of facts, one of which (as Mellor has remarked) is that the monster read a copy of Volney's Ruins of Empire, which was not published until 1791, thereby eliminating all other possible years in the 1700s when 31 July fell on a Monday. MWS left a second clue in Alphonse's letters to Victor in which it was announced that Victor's brother William had been murdered by the monster on “Thursday (May 7th),” the reading in 1818. A glance at pages 154–155 in this edition will show that MWS approached this date with considerable care: she first wrote “May 26” (which would have been the year 1796) and then immediately canceled “26” and replaced it with “28^th”, thereby indicating that she wished to get the year right (there would be little reason else for the two-day change). Eventually, in the missing Fair Copy or in proofs, the date was changed to “May 7th”, a date still consistent (exactly three weeks earlier) with the year that MWS had in mind, apparently the year 1795 (when the 28th and the 7th fell on Thursdays). I could invoke Volney again to establish the decade as the 1790s, but I am reluctant to use that means because some critics have raised questions about MWS quoting from not only Coleridge's Rime of the Ancient Mariner (1798) but also such poems as Wordsworth's “Tintern Abbey” (1798) and PBS's Alastor (1816). That kind of anachronism, however, is not at all unusual in fiction, when a narrative that is set in the past might contain a quotation from or allusion to a later poem. The monster reading Volney is quite a different thing indeed, and an anachronism there would have been a serious fault. But MWS was too careful for that fault, and the proof that the setting of Frankenstein is in the last decade of the eighteenth century is revealed in a manuscript change that shows again the care with which both MWS and PBS attempted to address this issue. See transcription page 459, lines 4–5, in this edition for the 1818 text where Victor remarks that he was in Oxford “more than a century and a half” after Charles I collected his forces there—since that year was 1642, then Victor must have been in Oxford between 17[93] and 17[99], and other evidence in the internal chronology would establish that Victor and Clerval visited Oxford in 17[96]. That both MWS and PBS had the 1790s in mind is further revealed by their alterations to Draft: Vol. II, Ch. 11, page 111: MWS first wrote “above a century before” and then revised it to “two centuries before”; then PBS more precisely altered the phrase to “century & a half before,” thereby designating the last decade of the eighteenth century for the temporal setting of Frankenstein, a decade that had personal and political events that MWS embedded in her narrative.³⁷

That MWS may have created her own private chronology of events in Frankenstein is further evidenced by these Draft and Fair-Copy Notebooks. See, e.g., transcription pages 15 (lines 31–32), 23 (line 13), 127 (line 25), and 537 (lines 30–31), where crucial ages and dates are altered or omitted in order to accommodate some sequence of events that had already been or was being established. In other places, changes were made not in Draft or Fair Copy but between the earlier state(s) and the first edition, as a glance at the parallel text of 1818 will reveal: see, e.g., transcription pages 153 (line 6) and 159 (line 24) for someone altering the date of Alphonse Frankenstein's letter from 2 June in Draft to 12 May in 1818.³⁸ Sometimes the Fair Copy is involved in this—e.g., MWS dated one of Elizabeth's letters 18 February in Draft, altered it there to 18 May, introduced the date of April in the Fair Copy only to cancel it and settle on 18 May, which became the date in 1818 (see transcription pages 537 and 647). It is clear something was going on with these changes: there may have been not only a chronology but also, and simultaneously, a personal numerology dictating some of the choices. Most of the evidence suggests that it was MWS rather than PBS who was making the choices, but there are times when PBS, as collaborator, was respecting what appears to have been MWS's plan.

A central issue in The Frankenstein Notebooks is the evidence they provide for the degree of collaboration between MWS and PBS in the making of the novel. The nature of PBS's involvement has already been suggested in various remarks above, the evidence of that involvement awaits the reader in the texts of this edition, and the specifics of his involvement are offered in the "Frankenstein Chronology;" that follows this Introduction. But a few general remarks are appropriate at this point in order to give the reader an overview of what might be called the conception, gestation, quickening, lightening, and birth of MWS's monster. MWS herself encouraged the use of such a metaphor by which to understand her Frankenstein: in her 1831 Introduction (pages v, xii) she described how she did “dilate upon, so very hideous an idea,” and she called her novel not only “the offspring of happy days” but also a “hideous progeny” that she bade to “go forth and prosper” in the literary world. If, as this edition will make evident, MWS is the creative genius by which this novel was conceived and developed, we can call PBS an able midwife who helped his wife bring her monster to life. His “hand” is in evidence in each of the extant Frankenstein Notebooks, and the "Frankenstein Chronology; below demonstrates his involvement in the printing, publishing, and reviewing of the novel. That PBS collaborated on this novel should come as no surprise to anyone, because the Shelleys left a long history of their shared activities as creative artists. They transcribed and they edited each other's works; they encouraged each other to undertake or to modify major works; and they even collaborated in the publication of History of a Six Weeks' Tour at a time when Frankenstein was being readied for the press. It is hoped that this edition will encourage someone to undertake a major study of this collaboration, which extended from the sublime (MWS's Frankenstein; or PBS's later response to this “Modern Prometheus” in his Prometheus Unbound) to the ridiculous (the two Shelleys engaged in a rhyming game in one of their pocket books, MWS providing the rhyming words and PBS filling out each line).³⁹

A reading of the evidence in these Frankenstein Notebooks should make clear that PBS's contributions to Frankenstein were no more than what most publishers' editors have provided new (or old) authors or, in fact, what colleagues have provided to each other after reading each other's works in progress.⁴⁰ What MWS actually thought of PBS's involvement in her text we will never know, no matter what methodology or facts we bring to this issue. I personally prefer to give both of them the benefit of the doubt and conjecture that (1) PBS suggested and made alterations to the text of Frankenstein for the purpose of improving an already excellent narrative (in [?February 1818] he wrote a review that judged the published novel “one of the most original and complete productions of the day”) and that (2) MWS accepted the suggestions and alterations that she agreed with. What this edition of The Frankenstein Notebooks provides for the first time is the available evidence by which to trace MWS's developing artistry as she drafted and then fair copied her novel. Even though she grew up in a family of writers and publishers (her father and step-mother ran the Juvenile Library and published books for children), this was her first experience writing and preparing a major work for publication. The fact that PBS had greater experience, having seen two of his own novels through the press as well as a number of volumes of poetry, might have given him a professional edge in their relationship, but that experience did not make PBS into a better novelist (his own novels, written before he was 20 years old, have little merit).

One of the purposes of this edition is to provide a solid and accurate foundation for future scholarship on MWS and on Frankenstein, in effect to counteract the increasing number of misinformed and misleading judgments about the novel, about the manuscripts, and about the Shelleys' collaboration. James Rieger (1974) was the first to mislead when, in his otherwise excellent edition, he provocatively phrased and overstated PBS's involvement with the text and subordinated MWS to her husband. According to Rieger, PBS “oversaw his wife's manuscript at every stage,” corrected “her frequent grammatical solecisms, her spelling, and her awkward phrasing,” and assisted to such a degree “at every point in the book's manufacture … that one hardly knows whether to regard him as editor or minor collaborator” (page xviii). Although these judgments reflect a great deal of truth (and collaboration merely means to “work with” not to “equally divide the labor”), their repetition in his “Note on the Text” that then challengingly asked “Do we or do we not owe [PBS] a measure of ‘final authority'?” (page xliv) seemed to devalue MWS's accomplishments.

More damaging to future scholarship was Rieger's misleading statement that “it was Shelley's idea that Frankenstein journey to England for the purpose of creating a female Monster” (page xviii). This judgment has misled many who without benefit of the manuscript have attempted to understand the contribution of PBS, who did in fact write in the margin that “I think the journey to England ought to be Victor's proposal” (see pages 424–425 in this edition). But PBS wrote this only after MWS had plotted the journey for the creating of the female and only because MWS originally had the idea of the trip drop casually from Victor's father, who suggested that his son take a two-year holiday while accompanying Clerval to England. All PBS did here was to ask MWS to change the source of the original idea for the trip, to have Victor “lead his father to this [idea of a journey] in the conversations” (PBS's marginal note to MWS) so that Victor would appear to be more in control of his destiny. When PBS wrote this note, the monster had already asked for a mate, Victor had already promised to create one, and Victor had “resolved to fulfil [his] promise while abroard [sic]” (see pages 398–399, 410–411, and 440–441 in this edition). In fact, PBS probably did not make his marginal suggestion until after MWS had written the later chapters in which Victor left Clerval behind in Edinburgh and journeyed on alone in order to create the mate; in any event (as most likely would be evidenced by the missing ur-text), PBS was not responsible for Victor's idea of creating a mate, and Rieger's unfortunate phrasing has misled a number of critics (see especially “Huet” below).

After Rieger's edition came E. B. Murray's “Shelley's Contribution to Mary's Frankenstein” (1978), which was based on some but not all of the Abinger Frankenstein manuscripts (Dep. c. 534/1–2 only), leading Murray to conclude misleadingly that PBS was responsible for a “thousand or so” words in Frankenstein—whereas he actually wrote and was responsible for more than 4000 words. Like Rieger, Murray judged many of these alterations to be minor: they “merely rewrite what Mary had written in order to provide a context for a given correction or addition,” or they offered “dictional, grammatical, or syntactical variations.” Murray added, however, that many changes were “creative additions which … help to shape atmosphere, incident, character, reader-response, and, consequently, aid in establishing the moral and aesthetic tone of the novel” (page 51). Then, to illustrate the way that PBS fair copied the Draft, Murray provided representative passages in parallel texts, keying the manuscripts not to 1818 but confusingly to 1831 Joseph.

Comparing Murray's transcriptions to the evidence in this edition will show that Murray, like all others who have ventured into The Frankenstein Notebooks, failed at times to distinguish PBS's hand from that of MWS. In those cases where I myself am unsure, I have employed footnotes to query the attribution that the fonts seem to prescribe, thereby alerting the reader to the problem of attribution. Any count of PBS words in the Notebooks has to be factored through these queries: that is, should only half of all the words queried as either “?PBS” or “?MWS” be counted as PBS's? or should 75% of those queried “?PBS” be counted and only 25% of those queried “?MWS”? There are other circumstances that make any attempt at an exact count foolhardy. Easiest to count, of course, are the words that PBS added to the Draft, which were then retained in the Fair Copy and printed in 1818. Also easy to count are those words that PBS was responsible for when he transcribed the final section of the Draft into the Fair Copy. Still, each of these cases involves two different kinds of words: those that are totally new, having been added to the text; and those that are merely modifications of MWS's words, some of which are merely relative pronouns: see, e.g., transcription page 455 where PBS altered MWS's “source of amusement” to “source of observation & amusement” and where he altered MWS's “that” to “which” (an alteration that frequently punctuates these Notebooks). Did PBS add two or three words to this page (that is, should we count “which”)? Whose word is it if PBS transformed MWS's “men” into “fishermen” (see pages 478–479 in this edition)? Even more difficult in the counting are those places where PBS transformed one MWS word into two PBS words, changed an MWS tense, corrected an MWS spelling, canceled an MWS word without making any substitution, or revised an MWS passage only to have MWS learn from it and rephrase not her old text but the new PBS text. Establishing a final and exact count for the PBS words in 1818 is impossible, although above I have hazarded a number in excess of 4000.

After Rieger and Murray came Anne K. Mellor's Mary Shelley: Her Life, Her Fiction, Her Monsters (1988), which offered the most extensive analysis of the Shelleys' collaboration to date. Mellor's study had the benefit of being a balanced one, for it stressed that PBS's alterations to MWS's novel “both improved and damaged the text and … must be analyzed with care.”⁴¹ On balance, however, although Mellor correctly judged that PBS was “responsible for much of the most inflated rhetoric in the text,” her Appendix on “Percy Shelley's Revisions of the Manuscript of Frankenstein” (pages 219–224) seems to credit PBS for many improvements to the text. Consider the topic sentences that Mellor used to introduce a series of paragraphs: “On several occasions, Percy clarified the text” by eliminating such things as a “vague comment” here or a “sloppy” phrase there; “Percy frequently substituted more precise technical terms for Mary's cruder ones”; “Some of Percy's revisions markedly improved the coherence and narrative continuity of the text”; “Percy occasionally improved Mary's paragraph transitions”; “Percy several times enriched the thematic resonance of the text”; “Percy further developed the complexity of the monster's character in several revisions”; and “Percy stressed Victor Frankenstein's unique responsibility for his creature.” Anyone objecting that Mellor gave too much credit to PBS in these pages is reminded to check all of her readings of PBS's hand against the evidence in this edition and also to consider that MWS most likely would have made some or many of the PBS improvements if she alone had been responsible for correcting the Draft and transcribing it in the Fair Copy.

If Mellor inadvertently gave more credit to PBS than either Rieger or Murray, then Marie-Helene Huet has outdone them all in her Monstrous Imagination (1993), which misread the evidence in Rieger and in Murray and possibly in Mellor and which changed PBS from collaborator to co-author. Had Huet consulted the manuscripts before publishing her book, she would not have rashly claimed that PBS was responsible for the “contrast between Victor and Elizabeth and the suggestion that Victor should create a female” (page 154). Both judgments are false, for MWS had established that contrast (what she and not PBS had called “a great dissimilitude”—see pages 10–11 in this edition) before PBS made any alterations, and it was MWS not PBS who introduced Victor's decision to create a female (see above on “Rieger,” who apparently misled Huet). Huet compounded a misreading of evidence when she claimed that PBS wrote the monster's “final words … ‘I shall collect my funeral pile …'”—a glance at pages 640–641 in this edition will confirm that MWS wrote the sentence, and pages 769 and 816 will demonstrate the small alterations that PBS made to the original sentence when he fair copied it in May 1817. PBS was not the one who “wrote the dramatic denouement on the sea of ice,” and it was patently unfair to claim that “it is impossible to assign with absolute certainty either to Percy or to Mary Shelley complete creative responsibility for any part of the novel” (pages 156, 155). The words “absolute” and “certainty” and “complete” make such a statement difficult to refute, but this edition will supply enough evidence to indicate the advisory role that PBS had in the making of MWS's monster.

Huet had many important things to say, and it is unfortunate that she based much of her analysis on incorrect readings of the evidence. One good thing, however, that comes from her mistakes, is a reminder of just how important a manuscript is to literary analysis. This facsimile edition, of course, is both a substitute and not a substitute for the original: future scholarly work on Frankenstein might begin with these photofacsimiles, but some textual scholars may need to consult the disassembled Notebooks currently on deposit at the Bodleian Library. In some cases, however, this photofacsimile edition may be more useful to scholarship than the now disbound Notebooks, because it distills and arranges some otherwise difficult-to-interpret evidence. The mere type facsimiles may keep future scholars from misreading words or hands or from conjecturing in such a way as to misrepresent the composition of these Notebooks. The very complexity of the Notebooks, for example, has partly misled David Ketterer in his recent articles on the Frankenstein manuscripts. Ketterer has done good work in his “Readings” (1995), “Draft” (1996), and “Insert” (forthcoming, 1997)—see “Short Titles” for the full references—but his articles would have been more soundly based had they waited upon the publication of this edition. Not that my reading of the evidence is the only way, but this edition would have made it easier for him to consider, cite, and illustrate the evidence for his arguments—as it is, his use of 1818 Rieger as a guide makes some of his arguments difficult to follow. I have addressed some of his misreadings in the Short Titles, in the Introduction, and in the "Frankenstein Chronology;" but I acknowledge here that my work is the better because of the hours that Ketterer and I have debated the evidence. Such a scholarly debate, whether conducted in public or in private (in an Oxford pub or in telephone calls between Newark, Delaware, and Montreal), can improve scholarship, and it is my hope that the debate that this edition will occasion will increase our understanding of MWS's Frankenstein.

At press time in August 1996, I am benefiting by reading two forthcoming essays, both of which have been kindly provided me in page proofs: Ketterer's “Insert” and Johanna M. Smith's “Texts of Frankenstein.” In both cases I have reciprocated with suggested “alterations” to these proofs that reflect my working with The Frankenstein Notebooks for the past two years. For example, while it is true that sometimes only “tentative conclusions” can be drawn from the Abinger manuscripts, there is no reason to hypothesize any intervening text between what I have denominated the Draft and the Fair Copy. In fact, there is unequivocal evidence to prove that MWS transcribed directly from the Draft (which is in an “intermediate” state) to the Fair Copy (compare the texts as arranged in parallel columns in Appendix A; see the false starts in the Fair Copy that are apparently caused by the handwriting in the Draft; and, finally, recall that the margins of the Draft were used in the very process of the transcription to record and to calculate the pages in the Fair Copy). In the absence of any evidence to suggest a draft intervening between Draft and Fair Copy, there is all the more reason to conclude that PBS's transcriptions at the end of the Fair Copy were his alone. The rhetorical flourishes in these concluding Fair-Copy pages, where PBS embellished MWS's Draft, not only give evidence that these words were his but also suggest that MWS was not reading aloud from the Draft with PBS transcribing her words. In this regard, there is no evidence in any of these notebooks that either of the Shelleys wrote what the other dictated.

How and when the two Shelleys exchanged the Notebooks we will never know for certain, but there is ample pen and ink evidence to show that PBS read sections of the text at least twice—almost all of the Draft pages reveal PBS's hand, and many of the pages suggest that he brought different pens and inks to his editing. He most likely read each chapter (or groups thereof) as MWS wrote them, and then appears to have read all of Notebooks A and B at one final sitting, probably in April when MWS was correcting them. We know, e.g., that PBS read Draft: Vol. II, Ch. 10 at least twice: once (possibly in January 1817) he wrote in the margin to suggest that it should be “Victor's proposal” to journey to England; and then later (the postmark and other evidence proving that it was between 9 and 17 April) he read and corrected the pages that MWS rewrote.

There are times in the manuscript when you can actually “see” MWS and PBS at work on the Notebooks at the same time, possibly sitting side by side and using the same pen and ink to draft the novel and at the same time to enter corrections. We know from Sophia Stacey that in Florence PBS “at night has a little table with pen and ink, she [MWS] the same” (quoted in Angeli, page 97)—apparently, these were portable writing desks that allowed the two Shelleys to work near each other in their room at night. Evidence of that closeness is on folio 3 verso of Notebook A, the last of the six rough-draft insert pages that MWS used to introduce more information on Victor's scientific interests (see pages 34–35 in this edition). MWS wrote the page in very distinctive pale gray ink (quite different from the black ink she later used to add “the science of” in line 18). Because PBS seems to have been the one to have written “natural philosophy” in the same distinctive pale gray ink and because he certainly made emendations in the darker ink, there is a very considerable probability that the two Shelleys twice worked together on this page between 27 October and 4 November 1816 (see Frankenstein Chronology for these dates).

Much more research remains to be done on the collaboration between the two Shelleys, especially now that the pages revealing PBS's hand in Frankenstein are available in photofacsimile. We all owe a great debt to Bruce Barker-Benfield who established in December 1993 through his Quiring Charts that all of the now separated leaves from the Draft and Fair Copy were once part of bound notebooks, a circumstance that not only tells us a great deal about the composition of the novel but also more about the collaboration between the two Shelleys. As these notebooks were passed back and forth between MWS and PBS, so also were ideas and phrases that went into the making of Frankenstein. We know, for example, that MWS took wholesale a PBS Diary entry and incorporated it into her novel (see Frankenstein Chronology; for 23 June 1816), but James P. Carson has reminded us that collaboration goes both ways: PBS's significant addition about the “republican institutions of our country” (Victor here speaking about Switzerland—see pages 126–129 in this edition), which could not have been drafted until at least November 1816 (see Frankenstein Chronology;), seems to echo a passage that MWS apparently wrote in a letter dated 1 June 1816. The nature of the echo, however, is complicated by MWS having rewritten this now missing letter when she prepared her Geneva letters for publication in the Shelleys' collaborative History of a Six Weeks' Tour (published November 1817). Related to this conundrum about who echoes whom are PBS's and Godwin's later admissions in 1818 that MWS would be the perfect person to write about the “calumniated Republicans” in a book on the Commonwealth in British History, a subject in which PBS confessed that he was “little skilled” (see PBS Letters, II, 21 and n.).

It should be clear that any analysis of these complicated intertextualities demands a study of the facts of just who wrote what when, as well as an understanding that some of the facts are beyond discovery or resist reconstruction.⁴² Also required is an understanding of what Jack Stillinger has described as “Multiple Authorship and the Myth of Solitary Genius.” Moreover, a study of this particular kind of mutual influence demands facing the issue of what Carson (page 449) called multiple ventriloquisms (MWS writing as if she were Walton or Victor or the monster; and even PBS writing as if he were MWS writing as if she were Victor recording Elizabeth's letter). Even within the text of the novel (which so often reflects what happens outside the text), Victor actually corrected Walton's narratives about Victor and the monster and the De Laceys; consequently, we are reading Victor writing as if he were Walton writing about Victor. Victor himself, then, was a collaborator and an editor: he “corrected and augmented” Walton's text because he “would not that a mutilated one should go down to posterity”—an ideal that any writer or editor hopes to serve!⁴³

The first group of short titles below provide a convenient shorthand by which to distinguish various texts and editions of Frankenstein from each other, and all of the titles survey the scholarship that has contributed to this edition. For many other important books and articles on Mary Shelley, Frankenstein, and her other works, consult the bibliographies published separately (e.g., Lyles), annually (e.g., in the Keats-Shelley Journal), or electronically (e.g., the MLA International Bibliography on CD-ROM). For reviews of Frankenstein published between 1818 and 1832, see Frankenstein Chronology.