Among the contents of Bloom's desk drawer described in Ithaca are three typewritten letters from Martha Clifford to Henry Flower, and, perhaps on a separate slip of paper, "the transliterated name and address of the addresser of the 3 letters in reversed alphabetic boustrophedonic punctated quadrilinear cryptogram (vowels suppressed)." The narrative reproduces Bloom's cryptic writing more or less (different editions vary slightly) as follows: "N. IGS. /WI. UU. OX/W. OKS. MH/Y. IM." Decoding the message is entertaining but not surprising, since the narrative has already disclosed its meaning: Martha's "name and address." And Bloom's effort to encode that information seems ludicrously unnecessary, given how difficult it would be to forget it—raising the question of whether he expects Molly to find the letters and wants her to track down the addresser's identity, or at least be teased by the possibility of doing so.

Martha's name and address have been "transliterated" in the common cryptographic sense that each letter has been replaced, consistently, with a different letter generated by a secret code. In this case, it is a very simple "reversed alphabetic" code created by laying out two alphabetic lines, forward and reversed, and mapping the true letters onto the ones beneath them, as so:

A B C  D  E  F  G H  I   J  K  L  M  N  O  P Q R S  T  U V W X Y Z
Z Y X  W V U  T  S  R Q P   O  N  M  L  K J  I  H G  F  E  D C B A

Bloom's code has its "vowels suppressed" in the sense that he writes down only the consonants, leaving the decoder to supply whatever vowels may fit, in the manner of ancient Hebrew writing. Gifford observes that eliminating the vowels "makes a simple code such as this harder to crack, since the patterned frequency of vowel recurrence makes the words relatively easy to spot." The text is "punctated" in the sense that, in each place where a vowel has been omitted, Bloom has inserted a period (full stop) followed by a space, as one would do in punctuating the end of a sentence. He has also marked the breaks between words by writing them on four separate lines, making the cryptogram "quadrilinear." In the text of the novel, the line breaks are indicated with forward slashes, just as is conventional when rendering verse lines in prose.

The slashes could conceivably be Bloom's, but that would weaken the sense of the final feature of the code: it is "boustrophedonic." In ancient Greek and some other ancient Mediterranean scripts, stone tablets were often covered with writing that moved from left to right across a line, then from right to left on the next line, and so on, in the manner of an ox plowing furrows in a field (bous = ox, strophe = turn, don = in the manner of). In these ancient inscriptions the appearance of individual letters was usually reversed in alternate lines, making mirror images of the usual characters. Bloom has not done that, but he apparently has written left to right, then right to left on the next line down, and so forth:

N. IGS.
WI. UU. OX
W. OKS. MH
Y. IM

Strangely, however, Bloom has followed the boustrophedonic principle only in lines 1-3, abandoning it in the fourth. His failure to turn back after the third line can be seen when one enacts the other rules of reversing the consonants and supplying the missing vowels, shown here in lower case:

MaRTHa   
DRoFFiLC       
DoLPHiNS       
BaRN

"Clifford" here follows the ox-plow principle, but "Barn" does not. Why? In December 1996 Richard Henninge proposed an answer on the j-joyce website: if the last line were written properly as "MIY," Henninge noted, it would decode to "NRB," which could be read as "Nora B.," i.e., Nora Barnacle. Putting the encoded text from the novel (Y. IM) together with the decrypted version of the boustrophedonic mirror image (NR. B), one gets "Why, I'm Nora Barnacle." This play on Joyce's part, if such it is, would support the view that Bloom's correspondence with Martha Clifford leads inevitably back to his marriage with Molly, no less than Joyce's adventure with Marthe Fleischmann left him committed to Nora. John Gordon affirms Henninge's supposition at the end of a 2002 JJQ article making essentially that point.

Instead of "reversed," which appears in the first edition and the Gabler edition, the Odyssey Press editions of the 1930s and the Penguin and Random House texts deriving from the 1960 Bodley Head edition all read "reserved," which, Gifford suggests, could mean "to keep from being known to others." But don't all cryptograms do that? The Gabler edition's change seems commendable, as it captures the sense of two alphabetic lines running in opposite directions. That edition also seems justified in amending "boustrophedontic," found in all earlier texts but nowhere else in the English language, to "boustrophedonic." None of these editions emends "punctated" to the usual "punctuated," as would seem desirable. (Punctation usually refers to patterns of small dots on the surfaces of leaves or human skin.)

Editions also disagree on the spaces between the coded consonants. The Gabler version corrects what appears to be a mistake in earlier texts by inserting a period and space between "WI" and "UU," where the "o" of Clifford belongs. But it does not supply one after "IGS," where the final "a" of Martha belongs. The final Odyssey Press edition of 1939 gets the "a" space right but misses the "o" one. No edition that I have looked at seems to convey Joyce's intention perfectly, so in the text displayed on this website I offer a unique version. I have not succumbed to the sheer madness of moving the periods in the second line to what would seem their proper places (WI .UU .OX), since no published text does that.

Read More

John Hunt 2021