Text: Burton R. Pollin, “Digraphs and Diereses,” The Collected Writings of Edgar Allan Poe — Vol. II: The Brevities (1985), pp. xxxvi-xxxx (This material is protected by copyright)


[page xxxvi:]


In attempting to transmit a text as faithful as possible to the original, as prepared and finally printed by the author, one finds several odd discrepancies from the norms in The Brevities which demand explanation and resolution. These concern Poe’s handling of the digraph (æ and œ) and of the dieresis indicating the sounding of two successive vowels. To my knowledge his regular flouting of the conventions governing the printing of these two forms has been largely ignored or avoided, even in editions which claim to be presenting an accurate and edited text. The resolution of editors has usually been in terms of correcting “errors.” Yet the regularity of these occurrences throughout the corpus of Poe’s writings suggests a rationale for them. Let us consider each form separately in terms of The Brevities and the practice of his day, starting with the digraph.


There is no need to detail the history of the form “x” and “ce” which is usually derived from the Latin, which has substituted “e” for the original Greek “i“. We find it frequently as the ending of the feminine plural for Latin words of the first declension which have been adopted in English. Sometimes Greek roots, used as prefixes (e.g., “aero-”) will produce the form. The editors of the University of Virginia Press, in the Preface to their edition of Marginalia, acknowledge the problem, uniquely as far as I know, with this statement: “One small group of [page xxxvii:] words . . . containing the digraphs æ and œ requires different treatment since the words contain puzzling errors that may not be compositorial [see Illustration, facing page]; these are corrected in text . . .” (p. xiv [but we find “Thomœ” on p. 202 for M 255, left uncorrected]). The strange thing is that the illustration does explain the “puzzle,” for it shows us that Poe’s fine penmanship, worthy of the term calligraphy, often leads to a printer’s error. In these combined letters Poe invariably wrote as a single loop, cursive “a” so well shaped that the conjoined “e” concealing the start at the top right left it looking like an “o,” at least to some typesetters. I became aware of this feature of his writing in editing “Hans Pfaall” in the Imaginary Voyages, for the manuscript (p. 10) in the Morgan Library showed an apparent “aeronautics” (q.v. in the facsimile reprint following p. 385 of the text [Boston, 1981]). This dated from 1835, but all the manuscripts of Poe show this handwriting trait which apparently misled printers of his “tales and sketches” and the nonfiction. Sometimes the editors or Poe corrected the digraph errors, but often they were allowed to stand in the text. (Also, in italics the two digraphs much resemble each other.) Mabbott corrected them with a note about the “error” in his edition, but not consistently: “anemonæ” in the 11/38 “Blackwood’s Article” (TOM 345), “animalculæ” in the 6/41 “Island of the Fay” (600), “Printer” in the 5/39 “Devil in the Belfry” (364); but in the “Gold-Bug” of 1843 the “bug” is left as “scarabœus” throughout (pp. 808-10, 814-15, 817, 822, 829, 831) whereas in the 4/45 “Mummy” for the first of several occurrences TOM footnotes his editing thus: [save for the first instance] “all have been corrected, as have the plural forms” (1187- 88). It is time for us to recognize that Poe was not so poor a student of Latin (and, in a sense, Greek) as to be unaware of the difference between the two digraphs; every lad learning his declensions would know this. How we can explain his indifference to correcting them in his proof is another matter.

We must face the problem for many words in The Brevities. For example Tennyson’s poem cannot retain the erroneous “Ænone” in M 44 (as it does in the University of Virginia edition). This, being a proper noun and a title, I list it in my “Typos” specially and correct it in the text. In general, the many confused digraphs in the text are silently corrected editorially. In the printing of the digraph there are a few inconsistencies. The link between the two letters is sometimes missing in the text, especially for my commentary notes, although an effort was made to reproduce it faithfully in Poe’s text; thus, Æsop is listed in the Index, but in my notes, at least one of which is indexed, he becomes “Aesop”; likewise, the two letters as in “Æschylus” may sometimes be reduced to the second (“e” alone) for some commonly used words. The latter should be consulted in the Index under the digraph form.

Obviously, a full treatment of this subject is needed, examining [page xxxviii:] Poe’s usage and taking into account his chirography and also the usage of the period.


Poe was very individualistic in many of his theories of art, society, and other matters, including punctuation. He was aware of its importance in influencing, even at times determining meaning, as in M 197 where he pleads for a reasoned treatise on the subject. I know of no study of Poe’s habits of punctuation; yet his use or treatment of the dieresis (or diaeresis) requires a short study prior to presenting a text according to Poe’s intentions and preferences. The standard texts on printers’ rules, habits and demands in turning manuscripts into published works seem to indicate that the rules for the dieresis were not hard and fast but that custom almost dictated one course — that when the dieresis was used (instead of a hyphen between two sounded vowels), it was to be placed over the second vowel. The practice of printers in printed works uniformly establishes this as an invariable procedure. A basic work, Typographia, by J. Johnson (London, 1824), 2.38-39, avers: “The rules for placing the diaeresis being . . . unsettled, we . . . recommend it to authors to mark them in their copy, according to their own . . . fancy. . . .” But later, he “advises authors to leave the pointing entirely to the printers [who] have acquired a uniform mode of punctuation (2.56). John Wilson, in his authoritative, oft reprinted Treatise on English Punctuation (1826; Boston, 1856; 6th ed.), advises: “Place a diaeresis over the latter [of two vowels] instead of a hyphen between them” (p. 219). Concerning Poe’s texts, as a whole, there seems to be a slight bit of contention between compositors who prefer the hyphen and Poe himself, who prefers the dieresis. What seems to be unique in general practice is his placing it over the first vowel, and this becomes a mark of the writer, even a means of identifying his authorship for texts, and his supervisory hand for posthumous works (see Supplementary Marginalia, Introduction). The sixteenth century Spanish writer G√≥ngora, mentioned in SP 45, shared Poe’s passion for the “first position” dieresis (q.v. in the Aguilar ed. of Obras Completas note on p. 1158 for “Soneto” number 357 on p. 516 [this information through courtesy of Alice M. Pollin]).

A survey of his practice outside The Brevities’ entries is in order to demonstrate his custom, taken here as the basis for the texts that we have and the instances requiring resolution. Alphabetically let us run through a score of examples, covering all periods of his career, numerous texts, and different types of material. Dates, loci, and brief notes for some will be given.

äerienne: thus pointed in the 1845 BJ for “The Spectacles” but changed to aérienne by Moldenhauer in SAR 1977, p. 221, presenting the manuscript, on the basis of French lexicons and other editors regarding the “ambiance” of the tale. [page xxxix:]

Antinöus: “Never Bet the Devil,” late insertion by Poe, not given by TOM but by H 4.311 (i. e. for the posthumous ed.)

cöexist: found in the 1848 Eureka, para. 20, H.194 — carefully proofread by Poe

fäery: in BGM rev. of 12/39 (5.332), not reprinted as such by Harrison (for instances in Brevities see below)

homöomeria: in the 4/35 “Lionizing” (TOM 181/3)

Isitsöornot: a humorous and repeated use (seven times) in the 2/ 45 “Scheherazade” (TOM 1151-69)

Maelström: a unique instance in his revised tales called “Phantasy-Pieces” in the Table of Contents (q.v. in Quinn 337-38). The second dieresis over “strom” is preserved by Poe, probably under the impression that this gave it a Norwegian or “Nordic” provenance, although the place name is Dutch in origin and has no dieresis.

Michäel: in a letter of 8/18/44 (Letters 261), as part of the name of Michelangelo in a peculiar spelling probably deemed by Poe to be Italianate

mösaiques: in the 4/35 “Lionizing” and the 11/38 “Blackwood’s Article” (TOM 175, 345, but Introduced into the 1842 and 1845 texts for the latter)

näive: in “Hans Pfaall,” the 1839 and subsequent versions (Imaginary Voyages 449), see also below

näivete: in the 1845 “Spectacles” in Tom 883 (changed to “naivete” in both TOM and SAR 1977)

prëeminent in 9/6/45 rev. in BJ (1.205); the same in the derived SM 7, q.v. below (see also “prëeminence” below)

pröem: in the 12/27/45 BJ rev (2.388). See below for M 140. reaction: in the 1848 Eureka (H 16.232, 234)

Stäel (for Mme. de Stäel): in “Lionizing” (version of 1845; TOM 180/30); also in “Landscape Garden” of 10/42 (TOM 712) and “Domain of Arnheim” (TOM 712)

Zäire: in the 1831 “Siope” (TOM 195) and “Blackwood’s Article,” 1840 version (H 2.381)

zöology: editorial filler in the 11/1/45 BJ (2.264)

The following instances of this idiosyncratic use of the dieresis in The Brevities covering the earliest to the latest sections (1844 through 1850, prepared in 1849) help to prove the persistence of Poe’s view of the proper use of this mark of sounding: cacöethes (M 25); Camöens (M 76, Poe’s footnote); fäery (M 44 [see note f] and M 202); prëeminence (SM 7); prëeminent (SM 7); pröem (M 134 and M 140); Raffäellian (FS 50); rëaction (M 165, para. 2). The two instances in the Supplementary Marginalia are especially significant because they show Poe’s solution of the “two vowel” problem in the source-text of the 9/6/45 BJ article (2.136); Originally Poe had “printed” the words as “pre-eminence” and “preeminent” and in the posthumous volume both assumed the dieresis. [page xxxx:]

The data above help to resolve editorial problems, such as obvious omissions or errors, which occur in Pinakidia 107, Marginalia 291, and Literary Small Talk 7 (q.v. in the section of the Introduction on Typographical Errors). They have a greater importance: to signalize Poe’s independence in matters of pointing and his keen desire to underscore the correct sounding of the syllables of the words being read. His ingenuity or oddity may be inappropriate, but his intentions are commendable and his finely tuned ear, as in his essays on prosody and his poems, is in charge.






[S:0 - BRP2B, 1985] - Edgar Allan Poe Society of Baltimore - Editions - The Collected Writings of Edgar Allan Poe (B. R. Pollin) (Digraphs and Diereses)