publisher colophon

The publisher asks the reader's indulgence for typographical errors unavoidable in the exceptional circumstances.

-slip pasted into the first edition of Ulysses (1922)

There is no consensus about the extent of the "typographical errors" for which Sylvia Beach felt compelled to apologize in 1922. Jeri Johnson lists 293 errata supplied by Joyce,1 Jack Dalton estimated in 1972 that Ulysses contained "over 2,000 corruptions," and in 1984, Hans Walter Gabler claimed that his synoptic edition reported "well over 5,000 departures from the author's own text as established from the documents of composition."2 The considerable size of Gabler's estimate, as Jerome McGann has observed, reflects the particular "genetic" approach he adopted in relation to the text of Ulysses.3 Genetic criticism focuses upon the genesis of texts, attending to "the documents of composition"-drafts and manuscripts relating to the printed work, usually known as the "avant-texte."4 Gabler's edition does not describe only transient printer's errors; it traces the text of Joyce's work back through his earliest extant drafts and notes for each chapter, recording inconsistencies and "departures" along the way. But Ulysses is full of mistakes and not all of them are undesirable. It is rife with deliberate inaccuracies, gaffes, and misquotations that are frequently associated with Leopold Bloom. Vicki Mahaffey proposes a seeming oxymoron-" intentional error"-to describe this aspect of the novel. She argues that Gabler's editorial approach is peculiarly congenial to Joyce's own emphasis upon the contingency of texts and the frailties of authorial intentions.5 This article readdresses these issues, employing some of the methods of genetic criticism to examine selected examples of misquotation in Ulysses. While reaffirming Mahaffey's basic conclusions, I shall [End Page 44] question the tendency to use such material to discredit the notion of authorial intentions altogether. Instead, I will argue that the mixture of theoretical and practical textual considerations associated with genetic criticism provides a rich and complex picture of authorial intentions.

As Mahaffey points out, Joyce's creative interest in the potential for error sometimes overlapped with the fortunes and misfortunes of the text of Ulysses. Leopold Bloom is, for example, "nettled not a little" to note that his surname has been misspelled as "Boom" on the list of mourners at Paddy Dignam's funeral in the evening edition of the Dublin Evening Telegraph. Bloom consigns this error to "the usual crop of nonsensical howlers of misprints" found in newspapers, but it also dramatizes Joyce's sensitivity to the contingency and fallibility of print (Johnson, 602).6 Unfortunately, this passage fell victim to the same pitfalls of the printing process it depicts, since Bloom's surname was spelled correctly in the first edition. Similarly, the erroneous "world" in Martha Clifford's letter to Henry Flower ("I called you naughty boy because I do not like that other world" [5.244-45]) appeared as "because I do not like that other word" when it was first printed as part of a serial installment in the Little Review,7 and Stephen's telegram reading "Nother dying come home father" (3.199) was printed as "Mother dying come home father" in all versions of Ulysses until Gabler's edition. These examples confirm Fritz Senn's observation that "things have a way of going wrong. This is shown by Joyce and it happens to him."8

For Mahaffey, this is not happenstance. Joyce, she claims, uses error to allude reflexively to the vulnerability of authorial intentions:

If we look to Joyce's texts for evidence of his intentions, we discover him minimizing the importance of authorial intentions by stressing the ways in which they are modified and reframed by the variable processes of writing, transmission, and reception. Joyce, then, uses his authority to recontextualise that authority against the broader backgrounds of history and production, insisting upon the irreducible oscillation between intention and circumstance.


When the text of Ulysses suffers from misprints, then, it would seem to confirm the point Joyce wished to make. For Colleen Lamos, the coincidence between Joyce's subject matter and his experience with printing [End Page 45] Ulysses constitutes a "discursive errancy" that calls into question the stability of all linguistic representation. Lamos pushes Mahaffey's description of the "irreducible oscillation between intention and circumstance" further: "Cut loose from the narrative and linguistic anchors that make possible the distinction between truth and error, or between meaningful, willed purposes and mere mistakes, random accidents and stochastic arrangements, Ulysses has come to be seen by many as a wandering text ungoverned by authorial orderings." Presumably this group includes Lamos herself, who uses the notion of "textual errancy" as the starting point for a fascinating reading of "omissions, displacements and disavowals" in Ulysses as the anxious vessels for same-sex desire.9

Yet, as Mahaffey acknowledges, there is a paradox here. "Minimizing the importance of authorial intention" is itself an intentional act. It is an irony that Joyce's depictions of error should themselves be plagued by error, but it also reflects the high standard of textual accuracy required by the decision to depict such errors. Far from being a "stochastic" arrangement "ungoverned" by orderings, Ulysses pointedly establishes connections between corrupted texts and fallible, human acts. Bloom is "tickled to death" (16.1263) by the inclusion of C. P. M'Coy and Stephen Dedalus on the list of mourners, since he knows that they were absent. In "Lotus Eaters," M'Coy asks Bloom to "put down my name at the funeral" (5.169) and Bloom fulfills this undertaking in "Hades" at the same time as Hynes the journalist makes a point of checking his "christian name" (6.881-83). The same passage in "Hades" also accounts for the erroneous inclusion of "M'Intosh" in the Evening Telegraph, as Hynes misunderstands Bloom's reference to the "Macintosh" worn by the unknown mourner (6.891-98).

Such interconnections implicitly compare the Evening Telegraph's report with the preceding narrative description of the funeral in "Hades." The resulting contrast might imply a wide-ranging skepticism about the relation between language, representation, and experience, confirming Bloom's pun on the paper's title as "tell a graphic lie" (16.1232). This phrase, from the narration of "Eumaeus," is often treated as emblematic of Joyce's general attitude toward language, truth, and error (Mahaffey, 183). But inferences about the unreliability of the newspaper report depend upon crediting the (fictional) truth and linguistic accuracy of the narrative provided in "Hades." They rely upon the premises they seek to undermine.

It is true that this passage from "Eumaeus" and its links with other parts of the novel convey the kind of acute sensitivity to the medium of [End Page 46] print and the ways texts are mediated that Mahaffey describes. But it is also true that the structure of the book acts to emphasize the role of particular individuals and subjective factors in the process of textual corruption. Personal favors and petty misunderstandings play significant roles here. Even apparent mechanical failures are attributed to human causes. Bloom, for example, notices a line of "bitched type" in the funeral report (".)eatondph 1/8 ador dorador douradora" [16.1262, 16.1257-58]) and recalls his visit to the print shop in "Aeolus" where he bumped into Hynes with a copy of this report and then watched a compositor begin to set it up. As he reads the lists of mourners, Bloom infers that the typographical error is explained by the distraction of answering a telephone call: "must be where he called Monks the dayfather about Keyes's ad" (16.1258-59). Behind the impersonal face of the printed word, Ulysses insinuates the play of multiple, personal contingencies.

This interest in tracing lines of transmission and the source of textual interference is something Ulysses shares with genetic criticism. The manuscripts of the avant-texte serve a function similar to the allusive interconnections within Ulysses in relation to unwanted typographical errors. For example, Bloom's surname is correctly misspelled on the list of mourners in all of the extant proofs of Ulysses. However, the final proof of "Eumaeus" at the Harry Ransom Center in Austin contains written instructions by Joyce requesting a change in the typeface. The whole passage had been set in ordinary type, but he wanted parts of it to be reset in order to represent a "graphic," visual distinction between Bloom's thoughts and the Telegraph obituary. Joyce underlined the phrases he wanted to stay in ordinary print with a green crayon and requested that the rest of the paragraph be set in italics.10 This necessitated breaking up the type and resetting the passage. It seems likely that in doing this the compositor mistakenly included Bloom's surname as it is spelled elsewhere in the novel. "Eumaeus" was one of the last chapters of Ulysses to be set up in print. Date stamps and signatures on the first page of this proof indicate that it was sent to Joyce from the printers on January 18, 1922, and that Joyce had returned it with instructions to print it with corrections by January 20 (JJA 27:113). As it was the final page proof and the publication deadline of February 2, 1922, was not far off, it seems that Joyce did not see the passage again, which is why the error slipped into the published text.

The proofs of Ulysses provide one indication why many genetic critics share the suspicions of Lamos and Mahaffey regarding authorial intention. They illustrate what Jerome McGann calls the "collaborative status" [End Page 47] of printing and how prone the process is to error.11 This is also the subject of the passage from "Eumaeus." Yet in addition to revealing deviations from his intentions, Joyce's extensive corrections to printer's errors also confirm his basic desire to achieve an accurate text of Ulysses. He suffered persistent "irritation" with the misprints and errors ("I am extremely irritated by all those printer's errors").12 As Roy Gottfried points out: "Joyce would want his text free of unintentional errors so that he could have it filled with intentional ones."13 The coincidence of Joyce's subject matter with his experience of printing Ulysses may be ironic, but it is not always felicitous. The proofs of "Eumaeus" give a good idea of the degree to which Joyce's intentions were compromised, but they do not eradicate or even obviate the notion of authorial intentions.

Working within conventional textual criticism, McGann cites postpublication emendations and errors of transmission to call into question authorial integrity. Similar effects can be observed when genetic criticism explores prepublication documents.14 Gabler observes: "No creation of the human mind springs to instant life and perfection without revision. Whether preserved or not, there must always have been discrete textual states, in temporal succession, of a literary composition. Thus the work may be said to comprise all its authorial textual states."15 This final inductive leap reflects the way that genetic criticism frequently co-opts Roland Barthes' theoretical definition of the "Work" as individual instances of a continuously unfurling "Text."16 Such "discrete textual states" may contradict or contrast with each other, indicating the occasions upon which an author changes his or her mind during composition. From this perspective, the author's intentions can seem febrile, vacillating, and capricious. Jean Bellemin-Noël explicitly introduced the term "avant-texte" as an alternative to brouillon ("rough draft"), to avoid the notion that "authors have a presentiment of a perfect state that they are reaching for." "In reality," he argues, "their words, at first, are potentially acceptable formulations. Only afterwards do authors discover that they are dissatisfied with their words; only then do they return to work on them."17 The notion of authorial intention, he argues, is something falsely imposed in retrospect.

There is, however, no consensus or great consistency among genetic critics upon this matter. In contrast with Bellemin-Noël, Gabler urges that "we never come closer to an author's willed structuring of design and meaning than through his conscious choices of language, expression and style. Where revisional variants manifest themselves, they make evident [End Page 48] crisis points of articulation through which the work passed in writing" (Synchrony, 309).

If "willed structuring of design and meaning" glosses the idea of authorial intention, then Gabler here describes the common feeling that handling drafts and autograph manuscripts is a form of intimate contact with an author and the writing process. Gabler's fondness for "revisional variants" strongly informs his desire to record Joyce's work on Ulysses in a synoptic form that simultaneously presents as many of the varying states of the novel's composition as possible. He hopes, in this way, that the reader will be able to construe a "text in progress."18 For Gabler, as for Bellemin-Noël, "revisional" activity is characterized by flux. The act of writing has a "process-character" that renders "authorial intention . . . statically conceived" unsuitable for "editorial performance": "Instead . . . authorial intention, as the dynamic mover of textual processes, requires to be editorially set forth for critical analysis. So viewed, authorial intention is not a metaphysical notion to be fulfilled, but a textual force to be studied."19 Rather than dispensing with authorial intentions, as Bellemin-Noël pleads, Gabler suggests that genetic criticism may offer alternatives to received wisdom about the relationship between authors, intentions, and works of literature.

Gabler indicates that editors should worry less about what Joyce meant in order to track down the sequence of what he actually wrote. Such critical impartiality is a laudable aim in many ways, but it is problematic when it comes to Ulysses. For it is a significant and defining feature of Joyce's writing that his creative use of error makes it extremely difficult to separate text and intentions. A deliberate error associated with a narrator, character, or narrative event shares the same formal characteristics as an inadvertent slip on the part of Joyce or some other form of typographical error. The reader must always guess as to what was intended. This is why the proofs of "Eumaeus" are so helpful. They verify that the correct spelling of Bloom's surname in the list of mourners is a misprint, providing evidence from outside the published text of Ulysses that confirms Joyce's intentions.

This crux is central to Joyce's repeated use of misquotation in Ulysses. Many of these occur in passing and without elaboration, as, for example, when Bloom mangles an old Latin tag: "Daren't joke about the dead for two years at least. De mortuis nil nisi prius" (6.793-94). In contrast with his reaction to the misprints in the Evening Telegraph, nothing in the narrative spells out that this phrase should be de mortuis nihil nisi bonum. [End Page 49] Some misquotations betray Bloom's ignorance, but others seem like deliberate jokes. He recalls, for example, an erroneous version of the chorus from act 2 of The Pirates of Penzance: "After their feed with a good load of fat soup under their belts. Policeman's lot is oft a happy one" (8.408-9). By switching "oft" for "not," he probably indicts the complacency of Dublin's police. There is, however, no further comment to indicate that Bloom (or Joyce) has not misremembered W. S. Gilbert's libretto. Misquotations can be measured empirically against the source text of allusion, but ambiguities of intention are still relevant.

Even when a passage is misquoted more than once, it may not be transparent that the error was intended to be part of the dramatic fabric of the novel. In "Lestrygonians," for example, Bloom misquotes a scene from Hamlet that is central to the treatment of fathers and the idea of paternity in Ulysses. He recalls the words of the ghost of Hamlet's father ("I am thy father's spirit, / Doom'd for a certain term to walk the night" [1.5.10-11]) as "Hamlet, I am thy father's spirit / Doomed for a certain time to walk the earth" (8.67-68). Bloom trivializes the original, substituting "earth" for "night" and "time" for "term." He also adds an address from the Ghost to his son. Stephen Dedalus repeats this part of the misquotation when he quotes the same lines:

He speaks the words to Burbage, the young player who stands before him beyond the rack of cerecloth, calling him by a name:

Hamlet, I am thy father's spirit,

bidding him list.


Stephen's theory of Hamlet's psychodynamics depends on this misquotation, because it explicitly relies upon "calling" Hamlet "by a name" at this point in the play. It provides a template for allusion to this crucial scene in Ulysses and dictates the form of Paddy Dignam's appearance in "Circe" ("Bloom, I am Paddy Dignam's spirit. List, list, O list!" [15.1219]) and Zoë's parody of Stephen's abstruse language ("Hamlet, I am thy father's gimlet!" [15.3655]). While quarto and folio versions of the play disagree about how often and in what form the ghost addresses Hamlet, no published version of the text known to Joyce contains this address.

Such repetition does not, however, confirm whether or not Joyce realized that the phrase in this form was a misquotation. The scene is crucia [End Page 50] to the novel, so it would seem criminally irresponsible to allow such an error to predominate inadvertently, but it is not outside the bounds of possibility. Fortunately, a manuscript draft of "Scylla and Charybdis" at the National Library of Ireland provides some clarification. MS II.ii.2a is one of three notebooks containing an early draft of this episode. Following the loss of a similar manuscript in transit between Paris and Buffalo, New York, in 1950, this is the earliest known draft of "Scylla." As Joyce was drafting the passage that includes Stephen's misquotation, one page of the manuscript records a correctly quoted version of the lines from Hamlet that Bloom misquotes ("I am thy father's spirit / Doom'd for a certain term to walk the night"), and the next page contains an abbreviated version of these lines as they are misquoted by Stephen ("Hamlet, I am thy father's spirit") in the final text.20 Thus, the "Scylla" draft seems to provide strong evidence, if it was really needed, that Joyce knew the correct form of the quotation.

Since he crossed out the correct quotation, this manuscript may even show Joyce in the process of choosing to misquote these lines in the manner that then becomes characteristic of the novel's engagement with Hamlet. This represents one of the "revisional variants" that Gabler cherishes. Localized upon a single draft, such a textual alteration is certainly congenial to genetic criticism's understanding of writing as process, subject to change. The development of a passage may, however, take place over several drafts and even within the proofs of a printed text. My next example of misquotation illustrates this complicated kind of textual evolution and explores the way that the circumstances of composition can shed light on Joyce's intentions.

Watching the printer set up the Evening Telegraph in "Aeolus" reminds Leopold Bloom of his father's religious observances: "Poor papa with his hagadah book, reading backwards with his finger to me. Pessach. Next year in Jerusalem. Dear, O dear! All that long business about that brought us out of the land of Egypt and into the house of bondage alleluia. Shema Israel Adonai Elohenu. No, that's the other" (7.206-10). The Book of Exodus records that Moses asked the Israelites to "Remember this day, in which ye came out from Egypt, out of the house of bondage" (13:3).21 Bloom's switch of prepositions ("into" for "out of ") is redolent with irony. Daniel Fogel points out that this quotation is found three times in the Haggadah Bloom mentions. It is part of the rituals of the Seder, the celebratory meal held during Passover ("Pessach") to celebrate the Jewish exodus from Egypt.22 The Seder enacts Moses's demand to "remember," [End Page 51] but Bloom has forgotten the words of the Old Testament. Although this may seem ephemeral, whatever lies behind it is firmly rooted, for Bloom makes the same mistake later in the day while thinking about various rituals associated with departure: "And the tephilim no what's this they call it poor papa's father had on his door to touch. That brought us out of the land of Egypt and into the house of bondage" (13.1157-59). On this occasion, the error in quoting is compounded by a vocabulary error. As Marilyn Reizbaum points out, Bloom has mixed up "mezuzah . . . a ritual, talisman-like object that Jews are meant to attach to their doorposts" with "tephilim," phylacteries associated with prayer.23

This repetition makes it unlikely that Joyce was mistaken, but not impossible. Usefully, the second instance of this misquotation can be found in an autograph draft of "Nausicaa," the second of two notebooks at Cornell University.24 The physical layout of this manuscript is very revealing about Joyce's methods of composition and the kinds of association that influenced his writing. As he began inscribing the notebook, Joyce worked on the right-hand side of the right-hand page (the recto) of the open book, producing a largely continuous column of text on this area of the manuscript. At this early stage, he probably left the page facing (the verso of the previous recto) blank. Figure 1 shows a transcription from the top portion of the fourth recto (4r.) of the second notebook at Cornell (item 56), a page that Joyce numbered "32".

The column of text on the right-hand side represents this initial level of inscription. Various embedded changes indicate that he was copying from a previous draft and either correcting his own transcription or making running changes to the text. For example, a sentence reading "Then next morning you have a beautiful calm without a cloud on the beautiful smooth sea" originally referred to a "beautiful sunrise" instead of a "beautiful calm." Other discontinuous pieces of text elsewhere on the page suggest that once Joyce had finished copying out the manuscript he went back over the draft, adding new material. He probably started by using the large margin he had left for himself on the left-hand side of the open recto, although he may have made some marginal additions as he was drafting. It is likely that the segment of text on the left of figure 1, beginning "lifebelt round him," belongs to this second stage of work on the draft, since it is cued by a mark (transcribed as the letter F) for insertion into the column of text on the right.

The scattered location of further segments of text on this manuscript indicates that Joyce returned repeatedly to the draft to make changes. [End Page 52]

Figure 1. From Cornell Item 56 fol. 4r (JJA 13:237).
Click for larger view
View full resolution
Figure 1.

From Cornell Item 56 fol. 4r (JJA 13:237).

Although he deleted or revised some wording, it looks as though most of Joyce's work on the manuscript after its initial drafting involved adding material. This manuscript confirms that the general tendency of Joyce's compositional habits was accretive (the published version of this passage is nearly three times longer than the text in the first column on the right). As he ran out of space for additional material in the margin of the page he was working on, Joyce carried on writing further text on the blank page opposite. Figure 2 transcribes the top portion of the page (3v) facing 4r:

The letter M at the head of the column of text on the right indicates that this segment should be inserted at the point marked by another M in the column of text on the right side of 4r. This fragment elaborates the [End Page 53]

Figure 2. Cornell Item 56 fol. 3v (JJA 13:236).
Click for larger view
View full resolution
Figure 2.

Cornell Item 56 fol. 3v (JJA 13:236).

imagined fortunes of the sailor "Johnny." Notably, the phrase "hanging on to a plank" was deleted on 4r and can be found at the bottom of column M on 3v. By the time he completed the draft, the sense of this passage ran back and forth across the surface of the manuscript. Different densities of ink and the occasional entry in pencil (transcribed here in boldface) suggest that this process was not simultaneous. It seems that Joyce sustained the creative process by returning to the draft and adding to it on different occasions. Robert Scholes dates this manuscript from the autumn of 1919, but Michael Groden indicates Joyce may have been working on it up to January 1920 (JJA 13:xii).25 There is, though, little way of telling how long he spent making additions or with what frequency.

The misquotation from Exodus is located in the column of text on the left of 3v. marked for insertion into column M. This might suggest the misquotation was provisional, a relatively late thought that only found its [End Page 54] place on a second or third return to the draft after the initial inscription. It does not exist in any of the notes that I have seen for Ulysses, so this entry on the manuscript may represent the moment of its composition. The text in column M also includes the first extant formulation of Bloom's inaccurate reference to the "tephilim." If, as seems likely, the chronological sequence of Joyce's work on this manuscript runs from right to left across the open pages of his notebook, then this vocabulary error preceded the misquotation. This indicates that Joyce conceived of the misquotation as a further detail to reinforce Bloom's shaky grasp on Jewish language and lore, as witnessed by the mix-up of "tephilim" and "mezuzah." The doubling of this misquotation in the published text of Ulysses can make these errors of vocabulary seem secondary. However, this manuscript records the first composition of the Exodus misquotation, so it reveals that, initially, the reverse was true.

By confirming the strong link between Bloom's inability to recall the Old Testament correctly and his faulty knowledge of Hebrew, this evidence about the genesis of his apparently trivial misquotation supports what the novel reveals elsewhere about the vexed question of his Jewish background. For example, Bloom makes contrary statements about this matter. In dispute with the Citizen, he asserts "Christ was a jew like me" (12.1808-9), whereas in conversation with Stephen Dedalus he qualifies this position: "So I without deviating from plain facts in the least told him his God, I mean Christ, was a jew too and all his family like me though in reality I'm not" (16.1083-85). This denial may reflect the Judaic laws of matrilineal descent. Bloom's father converted to Protestantism in order to marry Bloom's mother, Ellen Higgins, who as the daughter of Julius Higgins (né Karoly) may or may not have been Jewish.26 Bloom himself converted from Protestantism to Catholicism and has since lapsed. His remarks to Stephen indicate that Bloom may not be Jewish genetically, legally, or culturally: there is no indication as to whether the memory of his father reading from the Haggadah was an odd, singular occasion or a regular occurrence. In these circumstances, a degree of uncertainty about the wording of the Haggadah might be understandable.

Nevertheless, Bloom seems to identify with his Jewish background. In a running change to the text of the "Nausicaa" manuscript, Joyce confirms this by altering the beginning of his first formulation, "that brought thee out of the land of Egypt" (which is closer to the original text) to read "that brought us out of the land of Egypt [my emphasis]." Less devastating to the sense, the introduction of this first-person plural epitomizes [End Page 55] Bloom's sense of connection to the fate of his Jewish ancestors. This acquires major significance within Ulysses. It is one source of a larger pattern of reference to Jewish matters throughout the novel and it informs the way that other characters in the novel perceive him, from confused ignorance ("Is he a jew or a gentile or a holy Roman or a swaddler or what the hell is he?") to anti-Semitism ("He's a perverted jew" 12.1635).27 His ambiguous national and racial identity even prompts Andrew Gibson to describe Bloom as "Joyce's paradigmatic modern Irishman."28

This interpretation of Joyce's intentions depends on the assumption that the sequence of his work on the manuscript corresponds to my account of its physical layout. It is possible, for example, that he composed segments of the text around the misquotation from Exodus instead of adding it to already written text. There is evidence that this is how he worked on other occasions. The working notes for "Circe" include an abbreviated (but largely accurate) version of the policeman's chorus from act 2 of The Pirates of Penzance: "policeman's lot is not a happy one," suggesting that Joyce decided to use the phrase before he knew where it would fit.29 The misquotation from Exodus may have been transferred to the text of "Nausicaa" from a similar set of notes that has been lost.

These examples reveal why Hans Walter Gabler is wrong to think that recording manuscript evidence accurately bypasses specific questions of authorial intention in favor of "textual force[s]." Transcribing and interpreting the avant-texte cannot be separated from implicit judgments about how an author worked and what he or she meant. This is particularly the case with the synoptic edition of Ulysses, which utilizes an extremely complex form of diacritical annotation in order to preserve editorial judgments about the differing stages and hierarchy of Joyce's work. As Peter Shillingsburg remarks, "editors are critics, too; an edition reflects the editor's critical biases."30 Accepting this does not compromise the achievement of Gabler and his team. Rather, it credits their critical acumen and acknowledges the responsibility of editors toward their authors.

Embracing Shillingsburg's claim may also facilitate the shift between the realms of editorial work, textual study, and criticism. Genetic criticism of Joyce's work has tended to dwell on his methods of composition and their relation to his depictions of the creative process.31 However, the manuscripts of "Nausicaa" also provide substance for other critical perspectives. They may confirm Marilyn Reizbaum's view that Bloom's misquotation from Exodus inadvertently alludes to painful feelings associated with neglect of his Jewish roots. It bespeaks the "sentiment of remorse" [End Page 56] he feels at treating "certain beliefs and practices" with "disrespect" (17.1893-94). Some of these feelings of guilt about his Jewish origins are inherited from his father. Bloom recalls him quoting from Augustin Daly's play, Leah the Forsaken (1862): "I hear the voice of Nathan who left his father to die of grief and misery in my arms, who left the house of his father and left the God of his father" (5.203-5). Significantly, Daly's play turns upon family ties, fathers and sons, Judaism, and the tensions between religious background and cultural inheritance. For Bloom's father it painfully evokes his own secession from Judaism ("Every word is so deep, Leopold" [5.206]). Note too the echo of the phrase from Exodus that Bloom misquotes ("the house of "). This textual connection indicates that the misquotation may even convey feelings of oedipal guilt about his father's suicide.

A reading of the misquotation in these terms evokes Freud's theory of parapraxis. In The Psychopathology of Everyday Life, he argues that slips of the tongue, errors, and misquotations express unconscious desires and anxieties. Joyce owned a copy of this work and, according to Richard Ellmann, discussed such Freudian slips with various language students before World War I.32 The manuscripts might provide evidence for the claim that Joyce put the psychological theory to creative use within Ulysses by using an allusive technique built upon misquotation. This introduces a new model of intention, however, in which the most significant mental processes occur within the Unconscious. Precisely this sense of the displacement of intentional action from conscious activity to some other region motivates some critics to reject authorial intention. Jean Bellemin-Noël, for example, proposes that the avant-texte stands in the same relation to the published version as Freudian free associations stand in relation to the Unconscious. His avant-texte is "other," simultaneously intimate and alien.33

Of course, the book of Exodus also has a resonance within Ulysses beyond the details of Bloom's personal life. Professor MacHugh, for example, cites the misquoted phrase accurately when quoting John F. Taylor's speech on the Irish Revival: "But, ladies and gentlemen, had the youthful Moses listened to and accepted that view of life, had he bowed his head and bowed his will and bowed his spirit before that arrogant admonition he would never have brought the chosen people out of their house of bondage, nor followed the pillar of cloud by day" (7.862-67). Joyce heard Taylor deliver this speech in person in October 1901 and may have owned a copy of the pamphlet The Language of the Outlaw (1904), which summarizes [End Page 57] the address.34 However, in the version of this pamphlet reproduced by Richard Ellmann, there is no allusion to the book of Exodus. This seems to have been Joyce's addition to his own fictionalized version of the speech. Robert Spoo argues that this is, nevertheless, historically true to the rhetoric of Irish nationalism, in which comparison between Ireland's struggle for independence from English rule and the bondage of the Israelites was a common trope.35 Joyce invokes this comparison in his own critical writings in comparing Parnell to Moses36 and in the "Cyclops" episode of Ulysses, where the Citizen describes the perspective of Irish emigrants fleeing the hardships of the Great Famine: "Twenty thousand of them died in the coffinships. But those that came to the land of the free remember the land of bondage" (12.1372-73).

Comparisons between the Irish and the Israelites are also important to the symbolic parallels that underlie the book's whole structure. Joyce described Ulysses to Carlo Linati as "the epic of two races (Israelite-Irish)" (Letters I 46) and seems to have taken seriously the Semitic prototypes for Homer's Odysseus proposed by Victor Bérard in Les Phéniciens et l'Odyssée (1902-3). Allusions to the book of Exodus-both quotations and misquotations-play a diverse role in the novel: they belong to its mimetic engagement with history and ideology and they participate in a fantastic play of symbolism and parallelism.

In terms of Joyce's intentions, the presence of accurate and inaccurate forms of reference to this text in the same episode of Ulysses might render the manuscript evidence under consideration unnecessary. It further decreases the likelihood that Joyce's own pen slipped. However, the draft of "Nausicaa" is still suggestive in its relation to his other work on Ulysses. Although the misquotation in "Aeolus" is the first allusion to the passage from Exodus in the published text of Ulysses, this is not true of Joyce's work on the novel. Genetic study of its composition reveals that the two misquotations were written in the reverse order. The general pattern of Joyce's working methods has been amply detailed by Michael Groden in Ulysses in Progress.37 Joyce first worked toward publishing each episode of the book in serial installments as he wrote them. These appeared in The Little Review between March 1918 and December 1920. Working from notes, he drafted each episode by hand several times, producing manuscripts such as the "Nausicaa" draft, and eventually passing on a fair copy to a typist. Often, although not always, this ended up in the complete fair copy of Ulysses he gave to John Quinn (now known as the Rosenbach Manuscript). Joyce then corrected the typescript, making some additions, [End Page 58] before sending it off for publication. When serial publication was forced to stop by a prosecution for obscenity brought by the New York Society for the Prevention of Vice, the pattern of Joyce's work altered. With Sylvia Beach's help, he arranged for Ulysses to be published in book form, using Maurice Darantiere's printing firm in Dijon. Between April 1921 and January 1922, Joyce wrote the final chapters of Ulysses and heavily revised those chapters that had already been published.

This return to already written material accounts for discrepancies between the sequence of the published narrative and the sequence of Joyce's work on the book. In late August 1921, working on a galley proof printing (or placard) of "Aeolus" (now in the Houghton Library at Harvard), Joyce altered a sentence that read "All that business about brought us out of Egypt alleluia" to read "All that long business about brought us out of the land of Egypt and into the house of bondage alleluia" (JJA 18:17).38 Prior to this, both allusions to the book of Exodus in "Aeolus" had been accurate. These dated back to the earliest extant draft of this episode-the fair copy collected in the Rosenbach manuscript-and can be found in the version published in The Little Review in October 1918.39

As well as confirming the degree to which Bloom's misquotation is an intentional feature of the book, this "revisional variant" is revealing in other ways. Having thought of it to iterate the shakiness of Bloom's knowledge of Jewish customs in the "Nausicaa" episode, Joyce doubled the misquotation by including it in "Aeolus." The superseded accurate allusion to Exodus in "Aeolus" compassed many of the associations that prevail in the subsequent version: it is linked to Bloom's family history, it echoes the rhetoric of Irish nationalism, and it invokes symbolic connections between Israel and Ireland. Doubling the misquotation confirms Bloom's particular intellectual weakness and deepens its association with his family background. But this process of iteration occurs the other way round in the novel's published text. Introducing the reader to this passage from Exodus via misquotation means that its resonance is already qualified by irony before Professor MacHugh starts quoting Taylor. The misquotation belongs to a characteristic vein of humor that seeks to temper political statement or grandiose mythos by evoking the seemingly trivial foibles of Leopold Bloom. The avant-texte indicates that this is a deliberate feature of Ulysses.

Some question may remain as to how much this was premeditated. The alteration to "Aeolus" belongs to a late stage of Joyce's work on Ulysses. Although he had originally anticipated that he would need to see three [End Page 59] proofs of the novel, Joyce saw some parts of it nine times in different stages of proof-printing (LIII 30, 31). In contrast with Darantiere's original estimate of 27,876 FF, Ulysses cost 42,492.55 FF to print, and Laura Barnes records that nearly half the excess costs were absorbed by man-hours spent making authorial corrections after typesetting.40 Joyce's imagination seems to have been stimulated by working on the proofs in the accretive manner that he worked on earlier drafts, but it is not clear how much of this was foreseen. Michael Groden argues that this late creative surge and discrepancies between the narrative sequence of Ulysses and the sequence of its composition point to major changes in Joyce's artistic intentions (64-112). For example, Joyce only added the distinctive newspaper headlines that punctuate "Aeolus" when he was revising proofs of the novel. The episode appeared in The Little Review without them. For Groden, this epitomizes an artistic shift during the writing of Ulysses in which Joyce adjusted his initial commitment to realism and heightened the novel's play with the representational possibilities of language. It became, in Groden's view, a novel about language and "different methods of narration" (155).

This would seem to substantiate Bellemin-Noël's claim that writers lack full presentiments about their own artistic ends. It explains why many genetic critics are uncomfortable with readings, such as my own, which use the avant-texte to make critical judgments about the published text. Pierre-Marc de Biasi describes this as "teleological" (where the published text of a work functions as "telos") and claims that "in many cases this . . . will falsify the perspective."41 There are three reasons for this: first, the claim, already discussed, that it implies an impossible foreknowledge of the final text. Genetic study reveals both how authors change their minds and the degree to which other agents, such as printers and publishers, interfere with the transmission of texts. But Ulysses presents mixed evidence here. Consider Joyce's decision to revise Bloom's allusion to Exodus into a misquotation during August 1921. I have argued that this represents a significant adjustment to the resonance of this phrase. It occurred at the same time as he was introducing the newspaper headlines and may be concomitant with the change of artistic direction that Groden describes. Yet it also marks a point of textual and allusive connection with his previous work on "Nausicaa" from the end of 1919. In turn, this first manuscript composition of the misquotation reaches back to his original work on "Aeolus," where he first thought of alluding to Exodus. These [End Page 60] unfolding verbal continuities point to a realization of literary intentions that was purposeful, even if not foreknowing.

The second grounds for rejecting "teleological" readings relates to the radical autonomy that some genetic critics claim for individual drafts of a work. Laurent Jenny asserts that "to present [an avant-texte] for reading is obviously to inaugurate it as a text" with literary and critical merits on its own rights.42 But there are also those who admit that some kind of teleology is inevitable within genetic criticism: documenting and classifying manuscripts places them on a continuum with some end in view, and most critics are drawn to the avant-texte through a first acquaintance with the published work. Daniel Ferrer proposes that genetic criticism both traces the text's development forward through different stages of drafting and confers retrospective meaning upon drafts via the published work.43

Most manuscripts are not autonomous: they are written with the hope of getting somewhere, even if the author cannot know fully where that will be. There is always something provisional about a document like the "Nausicaa" manuscript. Even on the hypothesis that when Joyce began writing he hoped that it would be the fair copy for publication, it seems that as he added more material to it the draft became intermediate, requiring further work to produce a clean copy. The wide margins he left during the initial inscription indicate that Joyce probably knew he would add material and that what he was writing was at a draft stage. Then again, he left similar margins on the fair copies in the Rosenbach manuscript, so this is not conclusive. This does not mean that he knew what "Nausicaa" would look like when published, nor does it mean that he wrote without a goal or without some sense of the relation between what he was doing, what he had written, and what he planned to write. That is what it is like to have intentions. A singular conception that posits intending as a kind of fiat is a false model. This is the "statically conceived" notion of authorial intention that Gabler rejects. However, dismissing it does not constitute a convincing dismissal of authorial intentions.44 Instead, genetic criticism points us toward the "oscillation between intention and circumstance" described by Mahaffey (182). Rather than discounting authorial intentions, this view permits a more realistic, qualified picture of authorial intentions at work, one that can legitimately involve an author discovering his or her meaning in the course of writing.

A final ground for discomfort with teleological readings of the avant-texte is the desire to keep texts "open" to interpretation. De Biasi distances [End Page 61] himself from "finalist presuppositions" in favor of "continuous unfurling," just as Gabler's synoptic edition of Ulysses aims at reproducing a "continuous manuscript text."45 This may be compared to Michel Foucault's observation that

The author . . . is a certain functional principle by which, in our culture, one limits, excludes, and chooses; in short, by which one impedes the free circulation, the free manipulation, the free composition, decomposition, and recomposition of fiction. . . . The author is therefore the ideological figure by which one marks the manner in which we fear the proliferation of meaning.46

Foucault rejects authorial intention by questioning the political implications. But the "proliferation" of conflicting critical accounts of Joyce's work is enough to prove that his worry about the repressive effects of citing authorial intentions is groundless. Nothing guarantees consensus about those intentions. Nor does thinking of manuscripts in relation to the published text curtail the richness of their variant readings or stifle the intriguing alternatives they suggest.

Tracing Joyce's deliberate, determinate verbal changes leads us to some of the deliberate indeterminacies in his writing. I hope my reading of the avant-texte of Ulysses is faithful to Bloom's situation and the complex vision of Irish national identity that Joyce created by choosing him as the hero of his novel. Nationalists such as the Citizen are prepared to invoke the book of Exodus in order to depict the plight of the Irish, but their sympathies do not extend to those, like Bloom, whom they perceive to be Jewish. Bloom's predicament is reflected by one final allusion to the phrase from Exodus in "Ithaca": "In what order of precedence, with what attendant ceremony was the exodus from the house of bondage to the wilderness of inhabitation effected?" (17.1021-22). Both of Joyce's male protagonists leave the house at this point in the narrative, so the catechism form leaves it unclear whether this "exodus" is that of Stephen Dedalus or Bloom. This narrative indeterminacy allows the phrase "the house of bondage" to become resonant for both of them. Bloom's house may be a "house of bondage" for Stephen because he is bored and possibly scared by his host and because he fears he may have embarrassed himself by singing an anti-Semitic song. He may also facetiously recall Professor MacHugh quoting this phrase when reciting Taylor's speech earlier in the day. For Bloom, 7 Eccles street is a "house of bondage" because of his [End Page 62] wife's affair. The earlier misquotation "into the house of bondage" may also reflect the stalemate between Molly and Bloom. In context with Stephen's recent rude behavior, the phrase may speak more generally to Bloom's experience of racial prejudice. It is significant that Bloom most strongly asserts his Jewish identity in the face of hostile openly anti Semitic pressure from the Citizen. Since his family made their own exodus from Hungary to Ireland, Bloom's misquotation and the resonance of this phrase may indicate that they discovered a further "house of bondage" where they are trapped by the stereotypes and prejudices of others.

This sequence in "Ithaca" was a late addition to the novel. Joyce added it to a placard of the episode at the beginning of January, along with reference to the "intonation secreto" of a "commemorative psalm"-"The 57th, modus perigrinus : In exitu Israel de Egypto : domus David de populo barbaro" (JJA 22:68). On the next and final page proof of this passage, he corrected the printer's slip of "perigins" for "perigrinus" and two of his own mistakes. The psalm in question is the "113th" in the Vulgate and it refers to the house of "Jacob" rather than "David" (JJA 27:172). This further allusion to the exodus of the Jews reveals that Ulysses was not "ungoverned by authorial orderings" as Lamos intimates: Joyce was still reinforcing the novel's patterning in its final stages. His corrections indicate that he was not lax about the distinction between "meaningful, willed purposes and mere mistakes" either (Lamos, 118). At the same time, Ulysses does not confirm whether this psalm is sung by both men under their breath. The same ambiguities of attribution and motive relating to the "house of bondage" are relevant. Determining that such features of the text, including Bloom's misquotations, are intended does not close its meaning. I have assumed that Bloom's error is unconscious and that the irony of his misquotation is that he is persecuted for a heritage that is slipping away from the grasp of his memory. The interpretive crux in "Ithaca" regarding the attribution of these allusions, may, however, cause the reader to wonder whether Bloom is not capable of the same quotational wit as Stephen Dedalus. The pathos of his misquotation may, after all, reside in Bloom's awareness of this painful personal situation.

Matthew Creasy
University of Glasgow
Matthew Creasy

Matthew Creasy is a Lecturer in the Department of English at the University of Glasgow. He is a graduate of Oxford University and a Kennedy Scholar. He is currently completing a monograph on Joyce, misquotation, and theories of intertextuality.


1. James Joyce, Ulysses, ed. Jeri Johnson (Oxford: Oxford World Classics, 1993), 746-61. Further references to this work will be cited parenthetically in the text. [End Page 63]

2. Jack Dalton, "The Text of Ulysses," in New Light on Joyce from the Dublin Symposium, ed. Fritz Senn (Bloomington: Indiana University Press, 1972), 107; James Joyce, Ulysses: A Critical and Synoptic Edition, ed. Hans Walter Gabler, Wolfhard Steppe, and Claus Melchior (New York and London: Garland, 1984), 1:vii.

3. Jerome McGann, "Ulysses as a Postmodern Text: The Gabler Edition," Criticism 27 (1985): 283-305. Further references will be cited parenthetically in the text.

4. See Jean Bellemin-Noöl, Le Texte et l'Avant-texte: les brouillons d'un poème de Milosz (Paris: Librairie Larousse, 1972). Some critics translate this term as "pre-text."

5. Vicki Mahaffey, "Intentional Error: The Paradox of Editing Joyce's Ulysses," in Representing Modernist Texts: Editing as Interpretation, ed. George Bornstein (Ann Arbor: University of Michigan Press, 1991), 171-91. Further references will be cited parenthetically in the text. Regarding the sheer number and wealth of misquotations in Ulysses, see Matthew Creasy, "James Joyce, Ulysses, and Misquotation" (D.Phil. diss., Oxford University, 2003).

6. Unless otherwise indicated, quotations from Ulysses are taken from the 1993 edition published by Bodley Head, based upon the working text of Gabler's synoptic edition. Further quotations will be cited by episode and line number in the text. James Joyce, Ulysses (Oxford: Bodley Head, 1993), 16.1260, 16.1267.

7. The Little Review 5, no. 3 (1918): 42.

8. Fritz Senn, " 'All the earears and erroriboose': Joyce's Misconducting Universe," in International Perspectives on James Joyce, ed. Gottlieb Gaiser (New York: Whitson, 1986), 196.

9. Colleen Lamos, Deviant Modernism: Sexual and Textual Errancy in T. S. Eliot, James Joyce, and Marcel Proust (Cambridge: Cambridge University Press, 1998), 118-19.

10. The James Joyce Archive, ed. Michael Groden et al. (New York: Garland, 1977-79), 27:118/119. The abbreviation JJA will be used hereafter, and further references cited in the text.

11. Jerome McGann, A Critique of Modern Textual Criticism (Chicago: University of Chicago Press, 1983), 68.

12. To Harriet Shaw Weaver, November 6, 1921, in James Joyce, The Letters of James Joyce, ed, Stuart Gilbert and Richard Ellmann (London: Faber & Faber, 1957-1966), 1:176. Further references will be cited parenthetically in the text.

13. Roy Gottfried, Joyce's Iritis and the Irritated Text: The Dis-lexic "Ulysses" (Gainesville, University of Florida Press, 1995), 9; also see 154-59.

14. Genetic critics do not exclude postpublication material from consideration, but they usually concentrate upon prepublication documents. See Mahaffey, 175-77, for an account of the criticism this drew upon Gabler's edition of Ulysses.

15. Hans Walter Gabler, "The Synchrony and Diachrony of Texts: Practice and Theory of the Critical Edition of James Joyce's Ulysses," Text 1 (1981 [pub. 1984]): 309.

16. Roland Barthes, "From Work to Text," in Image-Music-Text, trans. Stephen Heath (London: Fontana, 1977), 155-64. [End Page 64]

17. Jean Bellemin-Noël, "Psychoanalytic Reading and the Avant-texte," in Genetic Criticism, 30. Also see Le Texte et l'Avant-textes, 6.

18. Hans Walter Gabler, "Joyce's Text in Progress," in The Cambridge Companion to James Joyce, ed. Derek Attridge (Cambridge: Cambridge University Press, 1990), 213-36.

19. Hans Walter Gabler, "The Text as Process and the Problem of Intentionality," Text 3 (1987): 11-12.

20. National Library of Ireland MS II.ii.2a ff. 5r and 6r.

21. King James Version of the Bible. This phrase also occurs twice more in Exodus, at 13:13-14.

22. Daniel Fogel, "Symbol and Context in Ulysses: Joyce's 'Bowl of Bitter Waters' and Passover," ELH 46, no. 4 (1979): 712-14.

23. Marilyn Reizbaum, James Joyce's Judaic Other (Stanford, Calif.: Stanford University Press, 1999), 116.

24. Part of this draft of "Nausicaa" is also contained in MS V.A.10 at SUNY Buffalo, New York.

25. Robert Scholes, The Cornell Joyce Collection: A Catalogue (Ithaca, N.Y.: Cornell University Press, 1961), 25.

26. See Morton P. Levitt, "The Family of Bloom," in New Light on Joyce, 141-43.

27. In addition to Reizbaum's work, see Neil R. Davison, James Joyce, Ulysses, and the Construction of Jewish Identity: Culture, Biography, and "The Jew" in Modernist Europe (Cambridge: Cambridge University Press, 1996).

28. Andrew Gibson, James Joyce (London: Reaktion, 2006), 125.

29. "Circe" 19:52 in Joyce's "Ulysses" Notesheets in the British Museum, ed. Phillip Herring (Charlottesville: University of Virginia Press, 1972), 357. This note is crossed out in red pencil.

30. Peter Shillingsburg, Scholarly Editing and the Computer Age (1986), quoted in Michael Groden, "Contemporary Textual and Literary Theory," in Bornstein, 259-86.

31. See Daniel Ferrer and Jean-Michael Rabaté, "Paragraphs in Expansion (James Joyce)," in Genetic Criticism, 132-51; and Sam Slote, "Epiphanic 'Proteus,' " Genetic Joyce Studies 5 (2005), available online at GJS/articles.htm (last accessed April 28, 2007).

32. Richard Ellmann, James Joyce, 2nd ed. (Oxford: Oxford Lives, 1982), 340.

33. See "Psycho-analytic Reading and the Avant-texte," 28-35.

34. Richard Ellmann, The Consciousness of Joyce (London: Faber and Faber, 1972), 33-39.

35. Robert Spoo, James Joyce and the Language of History (Oxford: Oxford University Press, 1994), 123-25.

36. James Joyce, Occasional, Critical, and Political Writings, ed. Kevin Barry (Oxford: Oxford World Classics, 2000), 193.

37. Michael Groden, Ulysses in Progress (Princeton: Princeton University Press, 1977). Further references to this work will be cited parenthetically in the text. [End Page 65]

38. An unrevised duplicate copy of this placard at Buffalo was dated August 23, 1921, by Sylvia Beach (JJA 18:12).

39. See pages "6" and "26" of "Aeolus" in Ulysses: A Facsimile of the Manuscript,ed. Clive Driver (London: Faber and Faber, 1975).

40. Laura Barnes, "Darantiere, A Printer's Archive" (paper delivered at the Seventeenth Annual International James Joyce Symposium at Goldsmiths College, University of London, June 28, 2000). These figures were verified via private correspondence.

41. Pierre-Marc de Biasi, "Towards a Science of Literature: Manuscript Analysis and the Genesis of the Work," in Genetic Criticism, 49.

42. Laurent Jenny, "Genetic Criticism and its Myths," Yale French Studies 89 (1996): 15.

43. See Daniel Ferrer, "Clementis's Cap: Retroaction and Persistence in the Genetic Process," Yale French Studies 89 (1996): 223-36.

44. Sean Burke argues that Roland Barthes and Michel Foucault attack authorial intentions by setting up a similar, oversimplified target supposedly derived from theology, in The Death and Return of the Author: Criticism and Subjectivity in Barthes, Foucault, and Derrida, 2nd ed. (Edinburgh: Edinburgh University Press, 1998), 20-103, esp. 23-25.

45. Pierre-Marc de Biasi, "What is a Literary Draft? Towards a Functional Typology of Genetic Documentation," Yale French Studies 89 (1996): 33; Synoptic III, 1895.

46. Michel Foucault, "What Is an Author?" in Textual Strategies: Perspectives in Post-Structuralist Criticism, trans. Josué Harari (Ithaca, N.Y.: Cornell University Press, 1979), 159. [End Page 66]

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.