University of Virginia Library

I

One of the most common situations involving external allusion occurs when a writer quotes from an earlier piece of writing. Insofar as the emphasis of the reference is on a verifiable independent source, the quotation should be exact. But insofar as the writer's intention is to adapt the quotation, it becomes a created element in the new work and cannot then be deemed incorrect merely because it fails to correspond with an external source. In many cases the motivation is mixed: the writer wishes to call on the authority of a previous author (expecting readers to recognize the author or the work cited) but at the same time wishes to alter the quotation to serve a particular purpose in the new context. Of course, a writer sometimes simply misquotes without intending to, and if no consequences follow from the misquotation, it is merely an error and

nothing more; but if the misquotation becomes the basis for discussion or implication, then it has become an integral part of the new work, whether the misquotation was consciously intended or not.[4]

Many of the possible editorial problems involving quotations can be illustrated by a single famous instance, the section of "Extracts" prefixed to Moby-Dick. In this section Melville draws together eighty quotations, ranging from the Bible to mid-nineteenth-century fiction, constituting a massive epigraph to the book. One might at first feel that epigraphs are not part of the text they introduce and that there would thus be no legitimate reason for their not being accurate; but a moment's reflection reminds one that an author selects an epigraph in order to set up a relationship between its implications and those of the text to follow and that one should not be surprised, therefore, if the epigraph were intentionally slanted to make the relationship clearer. Epigraphs are as much a part of a text as the quotations embedded in it. In the case of Melville's "Extracts," the creative nature of epigraphs is evident: the sweep of the assembled material is intended to suggest the greatness and universality of the subject of whales and whaling.[5] Melville furthermore places his quotations in a dramatic framework: they are said to be "Supplied by a Sub-Sub-Librarian," who has "gone through the long Vaticans and streetstalls of the earth" in search of them. The fact that supposedly they have been prepared by a created character does not, of course, mean that any errors in them must necessarily be accepted as contributing to the characterization, but it does strengthen the point that misquotations may at times be functional, and intentionally so. Whether misquotations are in fact intended as part of a characterization can only be determined by the context, and in this instance there is nothing to indicate that Melville wished the reader to regard any errors as lapses on the Sub-Sub-Librarian's part; on the other hand, he may well have wished to alter certain quotations to make them more appropriate as epigraphs to the work that follows, and misquotations in the "Extracts" must be judged critically with this possibility in mind.[6] A survey of some of the editorial

questions raised by the "Extracts" can provide a convenient introduction to the issues involved in dealing with external references in general.

Perhaps the most straightforward situations are those in which misquotations result from obviously intended alterations by Melville. In the second extract, for example, from Job 41:32, the wording of the printed text[7] (the manuscript does not survive) exactly matches that of the King James Bible except that "Leviathan" is substituted for "He" in "Leviathan maketh a path to shine after him." Clearly no one, in the process of transmission from manuscript to print, could have misread Melville's "He" as "Leviathan"; furthermore, the indefinite reference of "He" calls for some explanation when the passage is quoted out of context. It seems certain that Melville wrote "Leviathan," intentionally altering the wording of his source. Similarly, the extract from Montaigne contains the clause "the sea-gudgeon retires into it in great security," whereas the passage in Hazlitt's Montaigne (Melville's source) reads "this little fish" instead of "the sea-gudgeon." Again, the change cannot have resulted from a misreading of handwriting. Although it is perhaps conceivable that Melville wrote "sea-gudgeon" as a result of losing his place momentarily—since "sea-gudgeon" occurs in an earlier (unquoted) part of Montaigne's sentence—it is much more likely that he wished not to lose this term and substituted it in what is otherwise essentially an accurate quotation. Even undistinctive words can sometimes be recognized as Melville's alterations: the extract from Waller consists of two couplets, separated by a row of asterisks indicating ellipsis; in the second couplet "his" (twice) and "he" appear, rather than Waller's "her" and "she"— substitutions obviously made so that the gender of the pronouns would match that in the first couplet, now that the two couplets are juxtaposed (in the original, forty lines separate them).[8] In instances of this kind the

editor of a critical edition will retain what are in fact misquotations, recognizing that the aim is to reproduce Melville's intended form of the quotations. (The accompanying apparatus should of course inform the reader in each case of the relation between the source passage and the passage of text being edited, for the retention, as well as the alteration, of an "error" in a quotation constitutes an editorial decision that must be put on record.)

Thinking about these examples leads one to see some of the conditions under which emendations in the "Extracts" would have to be made. Unintended slips—authorial, scribal, compositorial—can be present in the text of the "Extracts" just as in the body of the book, and a critical approach to the text demands that all "misquotations" be evaluated and not automatically accepted as intended alterations. When, in the quotation from Blackstone, "caught near the coast" appears instead of "caught near the coasts," and, in the extract from Frederick Debell Bennett, "these weapons" replaces "those weapons"—or when the passage from Uno von Troil contains "lime-stone" instead of "brim-stone"—the substituted word in each case could easily have resulted from a simple transmissional error (such as a memorial lapse or a misreading of handwriting), and in none of these cases does there seem to be any reason for an authorial change. A number of such examples occur in poetic quotations from prominent sources: in the second extract from Paradise Lost, Leviathan is said to be stretched like a promontory "in," rather than "on," the deep and to spout out a sea "at his breath," rather than "at his trunk"; and in the extract from Cowper we read that "rockets blew [rather than 'flew'] self driven, / To hang their momentary fire [not 'fires'] / Around [not 'Amid'] the vault of heaven." All these misquotations are conceivable misreadings of handwriting or slips in copying, and it is difficult to see why Melville (or anyone else) would wish to make them intentionally ("fire" for "fires" is a clear instance of error, because the word is supposed to rhyme with "spires" two lines earlier). Slips of this kind, which probably occurred in the process of transmission from authorial manuscript to printed book, call for emendation by the critical editor.

Of course, some of these erroneous readings may have been present in Melville's manuscript, but as long as they can be argued to be unintentional slips the case for emendation is not altered.[9] When the printed

text reads "Hosmannus" at a point where the name in Browne's Pseudodoxia Epidemica is "Hofmannus," the error may well have been Melville's own: since he was probably using his copy of the 1686 edition and was accustomed to transcribing long s as "s," he may have mistaken the "f" in this proper name for a long s. Some instances are less clear-cut but still make the same point. A line from Elizabeth Oakes Smith is printed in the "Extracts" as "A mariner sat in the shrouds one night," although the original reads "on the shrouds." The reading "in" could simply be a scribal or compositorial misreading of Melville's "on"; but it is also possible that Melville wrote "in," not because he wished to alter the wording but because he was not copying carefully, the "in" perhaps coming naturally to him as the more idiomatic wording. Unless there is reason to believe that Melville intended to revise the quotation—which seems unlikely here—the possible presence of the "in" in his manuscript should not deter the editor from emending to "on."[10] In other words, whenever the possibility of a misreading of handwriting or of an authorial slip outweighs the possibility that the misquotation is an intended one, the editor seeking to establish what the author wished will emend to correct the quotation. Whatever interest there may be in Melville's writing "Hosmannus" and "in the shrouds"—if indeed he did so— belongs to a different level of concern; such evidence will be preserved in the notes but does not belong in a text aimed to satisfy another concern.

An additional example or two may serve further to clarify the role of arguments based on possible slips or misreadings of handwriting. In the Smith quotation, four lines after "in the shrouds," there is the clause "it floundered in the sea," where "it" refers to "whale"; in the original the subject is "he," not "it," but here the conservative editor is likely to feel that the possibility of an intentional change is enough stronger to warrant

the retention of the "it." This misquotation could of course have been a mere slip, but at least a misreading of handwriting does not seem to be involved in this case, and that in itself lends some weight to the argument against emendation—though it cannot be the decisive factor. The opening line of the extract from Waller reads "Like Spenser's Talus with his modern flail," but the adjective in Waller is actually "iron," not "modern." There would seem little possibility that "modern" could result from a misreading of handwriting (or from a slip of the pen, for that matter); but "modern" makes no sense, and the editor may well take the position that the word cannot have been intended (even if its presence as an error cannot be explained) and that an emendation is in order. The argument is somewhat strengthened by the fact that Melville a few years later, in "The Bell-Tower," referred to Talus as an "iron slave." But this information is fortuitous: the point is that the editor's critical judgment carries more weight than inconclusive speculation about the transmissional process. That "modern" cannot be explained as arising from a particular kind of error of transmission does not mean that it must therefore be retained by a conservative editor, if that editor considers the change unlikely to have been intended by the author. The critical editing of a text must extend to the quotations that are a part of the text. Because quotations have external sources, the editor has access to one more stage of antecedent document at these points than elsewhere in the text and thus is in a more informed position for detecting erroneous readings.

The process of locating those external sources, however, raises some important questions of editorial procedure. First is the problem of deciding what particular edition of a source text is the proper one to use for comparison. If the text of a quotation in the "Extracts" matches the text of the corresponding passage in the first edition of the work quoted from, the problem does not exist, for it does not matter whether Melville used the first edition or some other edition, as long as the resulting quotation is accurate.[11] But when the text in the "Extracts" does not correspond with that of the first edition, one cannot assume that the difference necessarily results from a transmissional error in the process of writing and printing Moby-Dick or from a deliberate change on Melville's part; it may be that Melville copied accurately, but from a different edition. If

so, the editor's thinking about the passage will be affected, and it is therefore important to know, if possible, the immediate source of each quotation. Yet in the case of many classics that have gone through numerous editions and have been excerpted and quoted even more often the editor cannot be expected to have searched through all possible sources. It is conceivable, for example, that Melville happened to take the Waller passage from a secondary source that misquoted "iron" as "modern"; but an editor cannot begin to find all the places where Waller's lines may have been quoted and has no choice but to proceed on the basis of the available knowledge. Frequently, however, an editor will know some of the favorite sources that an author is likely to use and may even have some information about the particular copies read. That Melville's ninth extract—identified only as "Other or Octher's verbal narrative taken down from his mouth by King Alfred. A.D. 890"—comes from Robert Henry's The History of Great Britain (1771) is not difficult to learn when one knows that J. Ross Browne's Etchings from a Whaling Cruise (1846), one of Melville's principal sources, also quotes this passage; furthermore, it is clear that Melville took the passage directly from Browne, and did not go back to Henry, because his extract agrees with Browne in reading "this country" at a point where the first edition of 1771 reads "these parts." A similar instance is the extract from Bunyan, which does not reproduce the relevant passage from the 1682 edition of The Holy War but instead follows the wording (except for a sixteen-word ellipsis) of a paraphrase of this passage in Henry T. Cheever's The Whale and His Captors (1849).[12]

The question that all this leads to is how the editor should handle errors or alterations that were already present in the immediate source of the quotations. If Melville quotes a corrupt text under the impression that he is providing the reader with another author's words, is it part of an editor's duty to replace that corrupt text with an accurate text? Answering this question goes to the heart of the concept of scholarly critical editing. Expecting editors to make such "corrections" of quotations is in effect asking them to establish the text of each quoted passage so as to fulfill its author's intentions. Such a procedure would mean treating each quotation as if it were an individual item in an anthology, not a part of a context created by another writer. The reader comes to Melville's "Extracts" not to seek established texts of Waller and Bunyan but

to see what Melville does with those authors; the editor's job is to establish Melville's versions of Waller and Bunyan, which may turn out to be considerably different from the texts that would appear in scholarly editions of them. If a corrupt text of a quotation is the one that Melville knew, responded to, and wished to set before his readers, then that text is the intended one under these circumstances. Emendation may be necessary if it is clear that the copy-text version of a quotation contains readings unintended by Melville, but the test for emendation is not whether the readings were unintended by the original writer. In order to be in a position to make intelligent decisions on this question, an editor is obviously required to perform some textual research among editions of the quoted work: if the copy-text version of a quotation does not match the text of the first edition of the quoted work, the editor must attempt to locate another edition that does match, for otherwise there is no basis for judging whether the variants were present in Melville's immediate source or originated at a later point (either in his own copying—intentionally or inadvertently—or in the succeeding steps of transmission). The scholarly editor must be able to draw the line between restoring an author's intended wording of a quotation and collaborating with the author by pushing the process of "correcting" the quotation to a point never contemplated by the author. To claim that Melville intended to quote accurately is not to the purpose: aside from the historical question of the degree of accuracy implied at a given time in the past by the act of "quoting" (discussed below), this claim mixes up different kinds of intention. That Melville may have "intended" in advance to quote accurately the sentiments of various writers does not alter the fact that the scholarly editor's concern is with Melville's active intention as he wrote, reflected in the quotations themselves.[13] The immediate sources he accepted and used become the authoritative sources for the quotations in this context.

Thus the extract from Thomas Fuller's The Holy State, and the Profane State reads "mighty whales which swim" at a point where the first edition of 1642 and the "second edition enlarged" of 1648 read "mighty whales who swim"; but because "which" is the reading of the London 1841 edition—the edition borrowed by Melville, according to Merton M. Sealts's Melville's Reading (1966)—there would seem to be no reason

to emend. The editor is not called upon to investigate whether any authority could attach to "which" as a reading in Fuller's text, for Melville evidently copied accurately the wording of the passage that struck him in the 1841 edition; whether or not that wording corresponds exactly to Fuller's intention is irrelevant, because it is the wording that Melville encountered and used. In the case of the Waller extract, if an edition or secondary source available to Melville could indeed be found containing the phrase "modern flail," no emendation would be required, however peculiar the reading seems, because the passage with that reading in it would be the one that Melville reacted to and found appropriate for inclusion in the "Extracts." Until such a source is found, however, the inherent unlikelihood of the reading will weigh more heavily with an editor than the theoretical possibility that Melville came upon the reading somewhere; the editor is acting responsibly if, after a reasonable search in Waller editions and books known to have been used by Melville, the reading is regarded as an error to be emended. Some textual research is nevertheless clearly necessary. An editor who looked only at the original 1645 edition of Waller's Poems would find that the line reads "Like fairy Talus with his iron flail" and might conclude that "Spenser's Talus," as well as "modern flail," is an erroneous reading (though possibly one intended by Melville to identify the allusion). But a little further research would reveal that "fairy" was changed to "Spenser's" in the 1664 edition and would thus place "Spenser's" in a different class of readings from "modern." Without such textual investigation, editors are not in a position to make informed judgments; but pursuing that research by no means implies that they are shifting their focus from the intentions of the quoter to those of the quoted.[14]

A related element in considering a writer's intentions in making a quotation is an understanding of the contemporary conventions of quoting. Generally before the twentieth century (and in some cases even into the century) quotations were not thought of as "inaccurate" or "incorrect" if they occasionally departed from the wording—to say nothing of the punctuation and spelling—of the source, as long as they did not distort the gist of its meaning. It was not considered wrong, even in expository writing (that is to say, writing not usually classed as "imaginative" or belletristic), to place between quotation marks what we would now think of as a paraphrase or an adaptation. For an editor to make such "quotations" conform to modern standards of accuracy, therefore, would be to modernize (that is, to employ a modern approach—for the corrected quotation would often be less "modern" in form); and the scholarly editor will not wish to engage in modernizing here any more than with the punctuation and spelling of the rest of the text. In checking Melville's extracts against their sources, then, an editor need not be concerned with spelling, punctuation, capitalization, or other formal matters except to the extent that discrepancies markedly affect meaning (or obviously result from slips or nonauthorial styling) or that agreements point to Melville's immediate sources. It clearly never occurred to Melville to be troubled about taking a twenty-word middle section out of a long sentence of Davenant's and beginning it with a capital letter; or juxtaposing, without ellipsis marks (and actually in reverse order), two sentences from Bacon's History Naturall and Experimentall of Life and Death (1638) that are in fact separated by six of Bacon's "Items"; or running together two lines of verse without indicating the line break, as in the extracts from Bacon's version of Psalm 104 and from I Henry IV. When Melville inserts "Fife" in parentheses after "this coast" in his quotation from Robert Sibbald and "whales" in parentheses after "these monsters" in his extract from Darwin, he is using parentheses to mark explanatory insertions in the way that we would now use square brackets.[15] An editor who injects ellipsis dots, virgules, and brackets into these quotations is modernizing, by requiring Melville's quotations—and each of the extracts is in fact printed in quotation marks—to conform to present-day standards. The place for showing these relationships between the

quotations and their sources is the textual apparatus or other editorial end-matter; in the text itself the scholarly editor will wish to respect nineteenth-century customs in the use of quotation marks (and what they imply about the enclosed material) just as much as nineteenth-century practices in placing apostrophes, commas, and other punctuation.

This custom of allusive quotation is represented among Melville's extracts by a wide diversity of situations, which thus help further to define the nature of the accuracy that is attempted. In addition to substitutions, which often could be the result of a slip of the pen or a misreading of handwriting, there are instances of insertion, omission, and paraphrase that cannot reasonably be considered inadvertent. For example, the extract from the account of Schouten's sixth circumnavigation in John Harris's Navigantium atque Itinerantium Bibliotheca (1705) begins, "Here they saw such huge shoals of whales," whereas the passage in Harris reads "saw an incredible number of Penguins, and such huge shoals of whales." Obviously Melville wished to omit the six words after "saw" as irrelevant to his purpose; deleting the reference to penguins focuses more attention on the whales, but Melville saw no reason to note his ellipsis.[16] Similarly, in the quotation from Jefferson there is an unmarked omission of fifty-one words between the subject and the verb; the sentence from Daniel Tyerman and George Bennet's Journal (1831) silently omits eleven words; and five are left out of the sentence from James Colnett's Voyage (1798).[17] Sometimes omissions and substitutions occur together, as when four words are omitted and four other alterations are made in the sentence from Richard Stafford, causing it to refer only to one man and one whale instead of to a group of each. The motivation for some of these changes is not always as obvious as in the omission of the reference to penguins or the insertion of "Whale-" in "The Whale-ship Globe" (the extract from William Lay and Cyrus Hussey), but there can be no doubt that such alterations are intentional and that they did not, in Melville's view, prevent the results from being regarded as "extracts" from the works named. Indeed, passages placed in quotation marks could depart even further from the originals and consist entirely of paraphrase: the sentences from Stowe, Boswell, and James Cook are far enough from the original wording that they have to be considered paraphrases made by Melville (unless he was following secondary sources

that have not yet been located). And the first sentence of the extract from Uno von Troil was apparently constructed by Melville's rearranging parts of the original sentence. It is impossible to analyze precisely the various reasons underlying these changes; but it is clear that the desire to alter passages so as to emphasize their connection with whaling is not the sole explanation. The pattern of the extracts as a whole shows that the concept of what constitutes "quotation" here is a much looser one than present-day writers are accustomed to. As in any other piece of writing, intention is ultimately defined by the work itself, and the extracts, as a group, establish their own standards. When Melville's departures from his sources result simply from his practice of approximate quotation, they cannot be thought of as "unintended." Errors, to be emended editorially, can certainly be located in the extracts, but the process of identifying them must be founded on an understanding of the level of accuracy attempted in the first place. (And in a scholarly edition the information used by the editor for determining this level will be available to the reader in the notes that record or explain the differences between the extracts and their sources.)

Melville's twisting of quotations for his own purposes—beyond any customary casualness in quoting—does, however, play a significant role in producing the wording found in the "Extracts." When Melville paraphrases a passage, places the result in quotation marks, and labels the source, he is engaging in allusive quotation but is approaching the border line—even by nineteenth-century standards—between quotation and fresh composition. He apparently crosses that line in the passage that purports to be from Antonio de Ulloa, describing the breath of the whale "attended with such an insupportable smell, as to bring on a disorder of the brain"; these words seem in fact to be Melville's own elaboration of the three-word phrase "an insupportable smell," which refers in Ulloa to a fish called "cope." The next step is to create an entirely new passage and provide it with a fictitious source: the extract following the one from John Ramsay McCulloch is labeled "From 'Something' unpublished" and is presumably Melville's own extension of a point raised by the McCulloch quotation, for it is clearly designed to follow McCulloch's statement but does not occur there in the original. Melville does not engage in this practice often, but the presence of one or two examples further strengthens the view of the "Extracts" section as a creative work and not a mere anthology.

Suggesting that something created on the spot has an independent existence outside the work tends to break down any rigid boundary between what is external and what is internal, and references to "real" sources can sometimes partake more of the internal world of the work

than of external reality. One extract, which describes some white crew members returning from the pursuit of a whale to find "their ship in bloody possession of the savages enrolled among the crew," is credited to a "Newspaper Account of the Taking and Retaking of the Whale-ship Hobomack." In fact, however, no such mutiny took place on board the Falmouth ship Hobomok. Melville's recollection of stories he must have heard probably resulted in the mixing together of details of two different events: the 1835 fight between some Namorik Islanders and the crew of the Falmouth ship Awashonks, and the 1842 mutiny by some Kings-mill Islanders on board the Fairhaven ship Sharon. The details of the Sharon mutiny fit more closely with those described in the extract; but one of the officers of the Awashonks in 1835 was captain of the Hobomok in 1841, when Melville's ship encountered it, and Melville may therefore have been thinking partly of his account. After investigating this tangle, Wilson Heflin decided that this extract must be "a piece of Melville's invention."[18] If so, no substantive emendation would be appropriate. To replace "Hobomack" with "Awashonks" or "Sharon" (which one?) would probably not restore what Melville intended to write; and, while either one would fit the facts somewhat better, there is no reason why Melville should be required to follow facts here. Whether or not the spelling of the ship should be corrected to "Hobomok" is a separate question. Because Melville did know of the actual Hobomok and because "o" and "a" are sometimes difficult to distinguish in his handwriting, it may be that he wrote "Hobomock" rather than "Hobomack," and one could defend an emendation to "Hobomock" (as a permissible variant of the correct spelling) or possibly to "Hobomok." Recognizing that what the extract describes never took place aboard the real Hobomok does not prevent one from correcting the spelling of the ship's name on the assumption that the actual ship Hobomok is being referred to in the citation—for the likelihood is that Melville was thinking of the real ship but confusing what happened on it.[19] The supposed quotation is thus inspired by real events and refers to a real ship; but the "Newspaper Account," and the Hobomok mutiny it reports, exist only within Melville's "Extracts."

Discussion of "Hobomack," which occurs in a citation rather than an extract, calls attention to the fact that problems of external reference are just as likely to occur in the citations. Some of the questions they raise are the same as those connected with quotations in general. Thus the citations of Darwin's "Voyage of a Naturalist" and Lay and Hussey's "Narrative of the Globe Mutiny" should not be considered errors simply because these are not the actual titles of the two books; the works alluded to are easily identifiable from such references, which are examples of the widespread nineteenth-century custom of allusive citation.[20] And when Robert P. Gillies's Tales of a Voyager to the Arctic Ocean (1826) is reported as "Tales of a Whale Voyager to the Arctic Ocean," one knows that the inaccurate citation, with "Whale" inserted, is intended by Melville. Or when "Most Extraordinary and Distressing" is omitted and "Spermaceti-Whale" becomes "Sperm Whale" in the long title of Owen Chase's Narrative (1821), one can allow the altered wording to stand on the grounds that it seems more likely to have resulted from intentional alteration than inadvertent slip. But another long title, for Henry T. Cheever's The Whale and His Captors (1849), is transcribed so precisely as to suggest that exact quotation is intended, and the one slight omission —an "as" introducing the last phrase—should therefore probably be rectified.

Citation of an altogether wrong title raises a more interesting issue. The extract from James Montgomery is credited to "World before the Flood" but actually comes from his "The Pelican Island"; the error is one that Melville takes from a secondary source, because Cheever's book quotes the same lines from Montgomery and provides the same citation. The question, raised earlier, whether an editor is called upon to correct the errors of a secondary source, requires further thought in a case of this kind. Misquotations in the text derived from a secondary source—and there are two in the Montgomery passage deriving from Cheever—generally do not require emendation because they constitute part of the passages as the quoter knew them.[21] But allowing an erroneous citation of this sort to stand is a different matter. It is true that Melville was equally

trusting of Cheever here and accepted the title as "World before the Flood"; but surely his intention in writing it down, judging from his practice in the "Extracts" as a whole, was simply to provide a factual reference. Of course, one can also argue, as with the misquotations in the text, that he may have responded to the wording of the title he found in his source and that the title should similarly not be corrected. Another extract (although the situation is not quite parallel) may provide some relevant evidence: "Pilgrim's Progress" is corrected in the first English edition to "Holy War," a correction that was evidently among those made by Melville on the proofs sent abroad and one that reflects a concern for correct citations. In any case, the decision on the Montgomery citation is a difficult one. Editors could argue either way; but there would seem to be enough difference in function and effect between a citation of source and a quotation to justify differing treatments, and a case can be made for correcting the Montgomery reference. (Even if no emendation is made, a note should of course call attention to the correct title and explain Melville's source of the incorrect one.) Other corrections of factual errors in citations are less debatable: the man who wrote on the Bermudas in the Philosophical Transactions of the Royal Society in 1668 was named Stafford, not "Strafford," and the date of Jefferson's "Whale Memorial to the French minister" was 1788, not "1778."[22] There is no pattern in the "Extracts" suggesting the deliberate alteration of facts of this kind: although the "Extracts" section can be called an imaginative work, it maintains a firm link with external reality.

Another, much shorter, preliminary section precedes the "Extracts" at the front of Moby-Dick, and it raises similar problems because it, too, is made up of material having an existence outside the work and is assigned a fictional compiler, a "Late Consumptive Usher to a Grammar School." Called "Etymology," this section consists of three quotations, followed by a list of the words for "whale" in thirteen languages. Such a list would appear to be purely a factual matter, but the critical editor, interested in Melville's intention, will find that it raises some intricate questions. One of them can serve as a kind of conspectus of the considerations involved in dealing with external fact in a literary work. Just before the English word "WHALE" in the list appears the entry for the Icelandic, and the word given in the original edition is "WHALE," identical with the English. Because this is not the Icelandic for "whale"

and because it seems unlikely that Melville would have wished to have two identical words in his short list (one of the purposes of which appears to be to display a variety of words)[23] it would at first seem reasonable to regard the Icelandic "WHALE" as a scribal or compositorial error (influenced by the word in the next line), or as an authorial slip. But the situation is not that simple: Melville himself may very well have intended to write the word "whale," because in Uno von Troil's Letters on Iceland (1780), in the paragraph just preceding the one from which Melville took one of his extracts, there occurs the expression "illwhale (bad whales)," in which the first word is offered as the Icelandic and the parenthesis as the English translation. From this Melville may have concluded that if he dropped the "ill" from "illwhale" he would be left with the Icelandic for "whale." If so, he was doubly wrong: in the first place, "illwhale" is not an Icelandic word, and the second edition of Troil (1780) corrects it to "Illhwele"; in the second place, removing the "Ill," even from this corrected form, does not produce the word for "whale." What is the editor to do? If Melville is misled by an error in a source and bases a discussion on the error, nothing can be done; but here there is no discussion, only a simple listing in which Melville apparently intended to give the correct word. But if one applies to Troil's corrected text the operation Melville seemingly performed on the first text, one still has an incorrect word; the editor would be in the position of making an emendation no more correct than the original reading. If one decides to correct the text, then, one must bypass Melville's presumed source entirely and insert the modern Icelandic word "hvalur."[24] One could reasonably defend this action by arguing—as with certain facts in the citations of sources for the extracts—that Melville's intention, evident in the text, to provide correct facts justifies the editor's going beyond Melville's knowledge to make the correction, so long as the error does not achieve a possible significance of its own within the text. On the other hand, one could argue against the emendation by saying that, if Melville did indeed, on the basis of consulting an outside source, regard "whale" as the correct Icelandic word, "whale" is thus his intended form and should remain. But Melville's use of Troil's erroneous "illwhale," while highly likely, is after all conjectural and should not be elevated to the status of fact. One cannot dismiss entirely the possibility that first suggested itself: "whale" as a scribal or compositorial error for the correct word (presumably "hvalur") in Melville's manuscript, or even Melville's own lapse (with "hvalur" as what he meant to write). Even if one finally emends on this basis, the speculation about Melville's possible use of Troil is not wasted effort, for in suggesting one explanation for the appearance of "WHALE" in the text it focuses attention on a crucial issue: the degree to which Melville wished to respect external fact in this instance. Besides, the critical editor cannot be in a position to make informed judgments at such points without investigating all available leads to external sources. This illustration draws together a remarkable number of basic editorial questions and shows how references to external fact can provide peculiarly effective test cases for revealing how thoroughly an editorial approach has been thought through.