He was a woman
He
was a woman
In Brief
In Lotus Eaters Bloom jumps from thinking that a
woman has recently played the part of Hamlet (fact), to
wondering whether the hero of Shakespeare's play might
secretly have been a woman (one literary critic's published
speculation), to supposing that this might explain Ophelia's
suicide (his own whimsical contribution). In Scylla and
Charybdis John Eglinton too links the recent performance
by a woman with the theory that "the prince was a woman." His
remarks are directed against overly ingenious theories in
general, Stephen's most certainly included, but it seems
possible that Joyce may have included the reference to Edward
Vining's theory here because it typifies 19th century
understandings of Hamlet that Stephen is strenuously opposing.
Read More
§ "Hamlet
she played last night. Male impersonator":
Bloom thinks this of Millicent
Bandmann-Palmer, who did perform in the play at the Gaiety Theatre on June
15. Gifford notes that a review of the show in the next day's
Freeman's Journal mentioned that she acted the
principal role and "to say the least sustained it creditably."
Strange as it may seem today (perhaps—it is becoming common
again in the 21st century), such cross-gender casting was a
frequent practice in 1904. Just as theater companies in
Shakespeare's time cast boys in women's parts because women
were viewed as inferior creatures and legally barred from
London's stages, late 18th and 19th century theater companies
often cast women in men's parts because beloved female actors
could find relatively few substantial female roles in
Shakespeare's plays. The conversion of Shakespeare's men into
"breeches parts" or "pants roles" (terms also commonly used in
opera) had become increasingly frequent by the end of the 19th
century.
In an article titled "Tragedy, Gender, Performance: Women as Tragic Heroes on the Nineteenth-Century Stage," Comparative Literature 30.2 (1996): 135-57, Anne Russell ably summarizes this history. Some of the performances she mentions were so influential that for decades afterward women continued to be cast in the same part: Sarah Siddons' Hamlet (1776-81), Ellen Tree's Romeo (1832), Patricia Horton's Fool in King Lear (1838). For theatrical producers of the time, the Danish prince seemed particularly suited to female interpretation, and "At least fifty English and American actresses played Hamlet in the latter part of the nineteenth century: Charlotte Cushman, Emma Walker, Fanny Wallack, Clara Fisher Maeder, Alice Mariott, Julia Seaman, Winetta Montague, and Millicent Bandmann-Palmer...who played Hamlet hundreds of times" (143). The tradition reached a kind of apex in Sarah Bernhardt's famous 1899 performances.
§ But "Perhaps
he was a woman": Bloom's focus shifts from seeing the
actress playing Hamlet as a male impersonator to wondering if
the character himself (or herself) somehow is (or was) a
woman. His source here is a 19th century American literary
critic named Edward Payson Vining whose thoughts about Hamlet
both reflected and influenced the female portrayals happening
on stage. In The Mystery of Hamlet; An Attempt to Solve an
Old Problem (1881), Vining argues that Queen Gertrude
gave birth to a girl. With war raging between Denmark and
Norway she dressed her child as a boy in hopes of ensuring her
offspring's succession to the throne. Vining devised this
backstory to explain Hamlet's feminine qualities: instead of
showing "the energy, the conscious strength, the readiness for
action that inhere in the perfect manly character" (46), the
prince spends his time deploring drunkenness, pondering
morality and religion, and emotionally leaping into graves.
Nor was his quasi-biographical way of reading a play all that unusual for his time. Literary critics in the second half of the 19th century often studied leading characters in Shakespeare's plays as if they might be real people, jumping off from mere dialogue to sketch detailed psychological portraits and inferring life histories that began well before the opening scene. (The most famous work of this kind, A. C. Bradley's Shakespearean Tragedy, was published in the year in which Ulysses is set, 1904.) Such an approach feels dated now, but it was all the rage in the Victorian and Edwardian era. It becomes truly untenable only if one begins to think of Shakespeare's characters as actual people. Bloom seems to flirt with this absurdity by thinking that Hamlet "was a woman," but he maintains comic detachment in the following sentence: "Why Ophelia committed suicide."
Eglinton has his own wry detachment: "The bard's fellowcountrymen," he supposes, "are rather tired perhaps of our brilliancies of theorising. I hear that an actress played Hamlet for the fourhundredandeighth time last night in Dublin. Vining held that the prince was a woman. Has no-one made him out to be an Irishman?" One Irishman whose brilliancies of theorizing he is aiming at is Stephen, who has been ransacking recent biographies of Shakespeare to construct a pseudo-critical account of the playwright's life just as fantastic as Vining's portrait of Hamlet. But seeing as how Eglinton, Lyster, Best, and Mulligan are always throwing out alternatives to Stephen's preferred authorities and explanations, Joyce may be evoking the Hamlet-was-a-woman theory here for a second reason. Vining's account of a character ill-suited to a masculine world of violent action sounds very much like the view of Hamlet advanced by Stephen's argumentative opponents: "The beautiful ineffectual dreamer who comes to grief against hard facts." Stephen's Hamlet is much more the murderous man of action.