Catalectic
Catalectic
In Brief
Listening to the sounds his feet are making ("Crush, crack,
crick, crick"), the poet in Stephen
recalls a familiar fragment of verse and thinks, "Rhythm
begins, you see. I hear. A catalectic
tetrameter of iambs marching." Tetrameter is a metrical
verse line containing four feet such as iambs (dee DUM), and
catalectic, from a Greek word meaning "incomplete" or "left
off," refers to the omission of a syllable from the final
foot. But in the tetrameter line that Stephen recites (the
second one is trimeter), either the opening syllable
is missing or the meter is trochaic (DUM dee).
Additional confusion is introduced in the Gabler edition by
changing "A catalectic" to "Acatalectic," meaning "without
incompletion"––i.e., complete.
Read More
Scanned as iambic tetrameter, "Won't you come to
Sandymount" is missing a syllable from the
beginning––as if chopped down from "O, won't you come to
Sandymount." The Princeton Encyclopedia of Poetry and
Poetics observes that "The term 'initial truncation' is
used to describe the omission of the first syllable of a
(generally iambic) line. A line so truncated is also called a
'headless' (acephalous) line." Stephen appears to be thinking
of this kind of truncation, because when he continues to brood
on the fragment of verse he hews a syllable off the beginning
of the next line, changing "Madeline the mare"
to "deline the mare."
The result is two iambs, but with the missing syllable restored this line sounds trochaic. If that is the case, then what's missing is an unstressed syllable after "mare." The encyclopedia observes that "Truncation is frequent in trochaic verse, where the line of complete trochaic feet tends to create an effect of monotony." (The same is true of dactylic verse in English.) If both lines are catalectic, Stephen is wrong about "iambs marching." If the first line is iambic, he is wrong about "catalectic."
The two metrical rhythms flip over into one another easily,
so iambic and trochaic lines can readily be mixed, as in the
Keats poem shown here. Its first and last lines are catalectic
trochaic tetrameter, with the missing final syllable marked by
a caret. The second and fifth lines are regular, or
"acatalectic," iambic tetrameter. The third and fourth lines
are regular trochaic tetrameter. One could argue that the
second line gives evidence for scanning Stephen's first line
iambically: if initially truncated to read "Till I cry out,
'Hold, enough!'," it becomes rhythmically identical to "Won't
you come to Sandymount." But the same is true of the
trochaic line "Give me women, wine, and snuff." On balance, it
seems that Stephen should be thinking of a catalectic
tetrameter of trochees marching.
Whichever way the line is scanned, in no rational universe
could it ever be described as acatalectic––i.e., metrically
regular. The Gabler text of Ulysses does just that,
egregiously emending the two-word phrase to "Acatalectic."
All print editions from 1922 onward have "A catalectic,"
and the version that Joyce published in The Little Review
in 1918 reads "Catalectic tetrameter of iambs marching."
Gabler often seems to regard the Rosenbach manuscript as an
authoritative arbiter of disputes, and Joyce's handwriting
there does appears to read "Acatalectic." But plucking a word
that makes no sense from this document to replace a phrase
that does make sense, and that has been repeatedly scrutinized
and affirmed over the course of a long publishing history,
constitutes editorial malpractice.
In his revised collection of annotations, Slote offers a
lengthy explanation of Gabler's choice by observing that Joyce
himself wavered between the two variants. He wrote
"Acatalectic" in "several drafts" but changed it to "A
catalectic" on a galley proof, and then on the same proof page
went back to the one-word form. The two-word form appeared on
a later proof and thus entered the first edition, but the
confusion continued as Joyce compiled lists of errata. "Gabler
reverts to the one-word form," Slote observes, "because that
was the last version attested in Joyce's hand," but, "in an
unpublished letter to Harriet Shaw Weaver, from 3 November
1922, Joyce relented and endorsed the two-word form."
This interesting history provides context for Gabler's decision but does not excuse it: an author's errors cannot absolve the editor of his own sins. Joyce apparently made a mistake in moments of uncertainty and thought better of it when publication was imminent. (He knew no classical Greek and probably had nothing like The Princeton Encyclopedia on his shelves.) Replacing the more or less right word of other printed texts with a totally wrong one adds editorial caprice to authorial confusion.