Line-End Hyphenation as a Problem for Editors, with Case-Studies from De Quincey

Abstract

'I had an idiotic hatred of hyphens in those days', wrote Tennyson in a note to 'Oenone', explaining his youthful preference for such unhyphenated forms as 'glenriver' and 'tendriltwine'.[1] Idiotic or not, his anathema would probably be echoed by most textual editors. For such a small and inconspicuous character, the hyphen causes a disproportionate amount of trouble. The situation is not improved by the fact that, historically speaking, the problematic aspects of the hyphen appear to have gone unnoticed until quite recently. As a result there is no body of editorial tradition as to how the various hyphen-related problems should be dealt with. Nor is historical information readily available to clarify questions that arise. Most reference books, bibliographies, and databases contain, under the heading 'Hyphen', either nothing or items of interest mainly to linguisticians and software designers.[2] Editors, accordingly, tend to deal with hyphenation problems on an ad hoc basis, as if confronting a difficulty no-one had ever faced before. Constantly reinventing the wheel is a process likely to produce, on the whole, fairly poor wheels. One purpose of this essay is to assemble some information about the hyphen and its quirks, in the hope of providing a focus around which greater knowledge and further thought may coalesce. The hyphen -- not to be confused with the long dash which follows this phrase -- is a character resembling a very short typographical rule, used either to join the parts of a compound, or to indicate that a word has been broken at a line-ending in the interests of justification. In this ambiguity lies its problematic nature. As its appearance suggests, the word hyphen is originally Greek, [GREEK TEXT NOT REPRODUCIBLE IN ASCII], a contraction of [GREEK TEXT NOT REPRODUCIBLE IN ASCII] ('under one'), an indication that what the hyphen joins is to be treated as a single entity; to be read, that is, as a compound and not as two separate words. The problems raised by the hyphen in classical manuscripts are not part of my present concern, and indeed the real difficulties begin only with the arrival of print; for (as implied above) the ambiguity of the hyphen arises from its use in justification. Sometimes it joins the parts of a compound (of a 'hyphenated compound', as we often say): linking two words each of which can stand alone but which are felt to be closely linked, and which may function, when linked, as a different part of speech. Thus, above, I used the compound 'hyphen-related'. A case only slightly different, also used above, is 'no-one', where the hyphen is used not just to link the words into a compound (for in some contexts 'no one' would risk ambiguity), but also to keep them graphically apart (one might well feel uncertain about how to pronounce 'noone' aloud, and the other theoretical alternative, 'noone', would be even more worrying). So much for the hyphenated compounds, as I shall call them for clarity. But the hyphen is also used to indicate that a word has been broken at the line-end. In modern typographical practice such breaking is normally done between major elements of the word: certainly between syllables, but where convenient between semantic units too. And when a line-end hyphen is found between elements that look as if they may be either parts of a single unbroken compound word, or the elements of a hyphenated compound, the problems begin. For when the passage is retyped or reset, the chances are that most of the line-end hyphens will no longer be at the ends of lines. Should they disappear, or should they be retained? In the modern period, line-end hyphenation is mainly a problem of printed texts for the simple reason that most people when writing by hand tend to avoid breaking words, and even hyphenated compounds, across line-endings. The inherent flexibility of handwriting makes this avoidance fairly easy. Of course cases of ambiguous line-end hyphenation occur; but look at any handwritten page of your own notes, or at a hand-written letter (if you still receive such things) and you are unlikely to see many line-end hyphens. …