DEPARTMENTS |
MARCH 2006 – NO. 4
|
Lexicography
by Grant Barrett
Missing Austin Friars.
Last year the Austin Friars disappeared from my dictionary.
There were never more than a couple of terse lines — Austin Friars, plural n., another name for Augustinian Friars — but somewhere between the first and second editions of the New Oxford American Dictionary (NOAD), those words were cut by an editor's knife.
This is not remarkable. You wouldn't know it to read publishers' press releases and the belly bands heralding new words in new dictionary editions, but usually, when a specific number of new words are added, a similar number of old words have been cut.
A database comparison between the two NOAD editions shows that other terms besides Austin Friars have also disappeared. In roughly the same place where support system appeared, you will now find an entry for support group with the same definition. An editorial decision, no doubt: the first term was not quite so common, nor quite so right, as support group.
Dihydroepiandrosterone vanished: its column inches are now taken by DHEA, the much more common acronym for the same drug.
Agist was demoted from having its own headword — the bolded word at the beginning of an entry — to being a run-on under agism, where it is marked as a variant spelling of ageist. A run-on is a word or expression based on another word, common enough to warrant mention but not so common to warrant its own entry.
Other terms were cut for not being sufficiently American, words like oncost, a Briticism for an overhead expense. Oncost was a legacy word leftover from the New Oxford Dictionary of English, a comprehensive (but not exhaustive) dictionary sold domestically in the United Kingdom, which provided the core of the first NOAD edition in 2001.
In fact, few dictionaries are created from scratch. When publishers create new dictionary lines, existing works are usually bought or borrowed to provide the substantial nine-tenths or more of the lexicon that is shared by all of the world's Englishes. Sometimes out-of-copyright works are used as lexical seed corn — a favorite is the 1913 Webster's Revised Unabridged Dictionary, which is now widely available in digital form — and other times, newer works are licensed to avoid having to redefine the substantial technical, scientific, and slang words that have appeared in the ensuing 90-plus years.
The creation of new works from old always involves cutting material, sometimes called deaccessioning (a term borrowed from museum curation, where it means to remove a work from museum's collection and, usually, to sell it). For museums, it's a pronouncement: "We no longer desire this object."
For lexicographers, cutting entries is an act of desperation brought on by a goat rodeo of printer's signatures, page counts, and paper costs, of trim size and product lines, and by a second-guessing of previous editorial decisions. More entries mean more costs. To cut costs, you must control the number of entries. So editors in charge must ask themselves, Which of the children shall we apprentice to the knacker-man so that we can afford to support the rest of the family?
Changing editorial goals and judgments are a crucial part of determining what to cut. In 1965, Philip Gove, an editor for Merriam-Webster, wrote, "We have been asked to explain how the Third Edition can have 50,000 new words and yet have 100,000 words less than the Second Edition." By way of explanation, he listed 31 types of entries that were eliminated between the first printing of Webster's New International Dictionary, Second Edition, and Webster's Third New International Dictionary. He also included a sample list of more than 500 words from 50 pages in the letter D of the second edition that do not appear in the third edition.
Many cut entries were encyclopedic entries — those that describe a place or a person more than they provide a meaning — but a great number came from eliminated categories like "self-explanatory lists" and "the more recondite terms," which depended heavily upon editorial decision-making. The elimination of "nonstandard and substandard terms" was even dicier. Given that few lexicographers, even 40 years after the appearance of WNID3, can completely agree on the definitions of "slang" and "dialect," it's clear that a finished dictionary depends heavily upon the informed opinion and preferences of editors, which in turn, are a way that dictionaries acquire their distinctive characters. Despite the common lay use of "The Dictionary," — which Erin McKean, editor in chief of Oxford University Press's U.S. Dictionaries department, says makes it sound as if there is only one Dictionary of the Universe — the various dictionaries on the market are not interchangeable widgets. They are as different as one Song Dynasty vase from another.
It turns out that the second edition of the New Oxford American Dictionary underwent a substantial redesign, which permitted the inclusion of many more words in the same amount of space. Far fewer words were cut than is usually necessary between editions — but something as simple as a type-size change is not always an acceptable solution, because a good part of a potential readership may decide to buy a dictionary only if its failing eyes can easily make out the text. Anything extreme, like agate type on thin onion skin (which might allow too much of the obverse page to show through) is usually out of the question.
*
A dictionary user might guess that the fat volumes, even the ones marked "unabridged," are complete or at least close to it. No! More words, and meanings of existing words, never make it at all to the pages of a dictionary, any dictionary, than are ever cut. There is a greater bounty of language than the few hundred North American English-language lexicographers can make room for. Most dictionaries are synchronic, which means, roughly, that they are snapshots of language in time. They do not try to encompass all the words that ever were.
For example, there are millions of chemical words alone, like the above-mentioned dihydroepiandrosterone, that you usually will not find in general-use dictionaries. Add another affix and methylisothiazolinone becomes methylchloroisothiazolinone, but the question isn't whether these are real words — they are — but whether they merit space for entries. Most dictionary-makers settle for merely defining the common chemical prefixes, suffixes, and roots so the user who wants to puzzle out the contents of their shampoo bottle can do so.
Lexicographers don't simply let deaccessioned words fall behind the cabinets where mice build them into scrap nests. Lexicographers are hoarders. They have collections. They have stacks. They have editions and printings, and reprintings, and fascicles and supplements, and galleys and blues and lists — lots of lists — of words. Post-Its, index cards, reporter's notebooks, print-outs, napkins, Bazooka Joe wrappers. Some entries start as ballpoint ink on a forearm on a beery night. They store and keep language as insurance against future need.
If deaccessioning is a pitiable abandonment, the other side, accessioning from the great word hoards, is a rescue. And the discovery of durable neologisms — or even just previously unrecorded old words — is, ultimately, what makes it necessary to cut words.
Back to Top
AUTHOR BIO:
Grant Barrett is project editor of the Historical Dictionary of American Slang for Oxford University Press and a minor contributor to the New Oxford American Dictionary, second edition. He is also editor of the forthcoming Official Dictionary of Unofficial English (McGraw-Hill, May 2006), editor of the Oxford Dictionary of Political Slang, and editor of the Double-Tongued Word Wrester Dictionary website.