an editing diary


A language is an overgrowth of words. A dictionary is a description of the trees. An etymological dictionary knows "the woods," the spaces between the trees. With hornet nests and old bottles and forgotten foundations of lost spring houses and rusting farm gear, and mud. The woods were my element as a boy. There are no grids. You remember that if you go between those two oaks and stay to the left of that tall stump you'll come to a sassafras tree with sweet leaf-sap.


The etymonline copy-edit/fact-recheck/text-rewrite wagon train has passed "pre-," and now stands a few steps into pri-, on the downhill slope of "P." That letter has been a slog. Started "P" in September, I think, but I took a couple months detour to rebuild the first half of "C" because I didn't like it. Now I'm starting a nine-days-straight shift at the desk job that pays most of the bills, so the journey through "P" will be on hold. "S" will be worse.


You can get a lot of dictionarying done with a week's vacation from the full-time job. Editing of "P" is finally finished — I hiked into the foothills in September and finally got down to the seaside docks the following spring — and I wanted to see if I could bull through Q in the one week. But Q is a dark hall of robed and silent inquisitors adept in the torture of Latin pronouns. Quondam, quorum, quantity, quality, quo, quam, quis — ablative or dative? relative, interrogative, or indefinite? Cui bono? (Why's it cui in that case, eh?) Quo vadis? Quatsit Tuia? I think I passed.

With a little time to spare, so I picked up a bunch of CL- I had lying around the house for a rainy day and burned through that. What a mess. Picking apart the threads of close (v.) and close (adj.) and ... this probably makes no sense to anyone; but it's not like there's a watering hole you can go to and drink with other dictionary fools and carp about the work.


Editing into R is refreshing. It's high ground, up-and-down but mostly down, rock-ribbed, arid desert country, vividly alive with rare, raspy, restless, rugged things, so wear rough clothes. The dog owns it. He trots in parallel with you, at a safe distance, out of rock-throwing range. Stops when you stop. Moves when you move. But now, at the end of RA-, the real desert opens out ahead. The sand-to-the-horizon desert where nothing is alive but the quicksand. RE-.

Any dictionary-writer would shudder at the mere mention of it. Crossing it risks sanity. The Romans concocted it, but English got addicted to it in the 19th century and affixed it to literally everything. So the list of re- words is more or less coterminous with the dictionary itself -- a dictionary within a dictionary. And of course it can be reduplicated (re-re-financed), so a third dimension opens up. Even the OED warns against this ground: "impossible to attempt a complete record of all the forms resulting from its use. The number of these is practically infinite ...." Dry bones and "turn back!" signs on weathered board.


"S" is where dictionary writers go mad; swamps of Germanic between long Latin deserts. You come up the road into SK-, and the place looks identical to that village just over the ridge, the one you just worked in, the one that begins with SH-. Except everyone in this place speaks with a Swedish accent and wears operatic viking hats.


Editing "S" began at the start of August, and the letter had 87 pages to itself, the most in the site. Now the editing nears the end of it, but "S" is going to be 108 pages long in a few days.

Paths out of that purgatory run through SU-. That place is all seedy villas, flat street after flat street, in some stale small city in Oklahoma or Argentina, late in the day. Each house a blank wall to the street, and you know the layout, because all are the same: trim little courtyard prefix (sub- or super- ); solid block of Latin root behind it. -due, -ject, -lime. What pass for the gentry homes here knowable only by a faded French sur- painted on the gate.

Hurry across the corner where svelte and Svengali and a few other malingerers loaf, and suddenly you're in SW- which is nothing like: it's a burg of Germanic tribes. Hotter, darker, and every word here has a bar-room brawl going on inside — Anglo-Saxons, Vikings, Dutch, Danes, Faeroese, who the hell knows, turning up the tables; kindred words mashed together by common use and stepped on and wrung out.

You stagger to the end of that block, that night, and dawn into the clear, still air of Athenaion — syn- after syn- after syn-. The weird, familiar touch of the tongue licked on ancient thought-prints.


"W" is the insane asylum of the dictionary. Weird, wary, worrisome. Sensible, sober Latin will have nothing to do with it. Down the halls stalk delusional French words that insist they're Englishmen because they wear a "W" on top of their heads. Echoes of chain-clatters sound down empty wings of lost sounds — wlanc, wlonc. By the time you sort out wrack, wreck, rack, wreak you're ready for a padded cell yourself.


Writing through the part of English that begins with EF- felt like crossing a briar patch. Mostly ex- words assimilated to -f-, but often with unpleasant knots in their past. On the continent this sometimes goes to -esf- and then they lop off the vowel (sforza = "effort") so it starts to look like the speech of Sylvester the cat from the Warner Bros. cartoons. But the real clutch-and-drag was the number of words that had valid senses in some branch of science or philosophy — or did 175 years ago — but no one outside that pigeonhole ever would need them.

So in the middle of "E" in Bierce's "Devil's Dictionary," this was good for a laugh, not just for the joke's sake, but because it came in exactly the right spot:

"Efferous, Effigiate, Efflagitate, Effodient, Effossion. See some other dictionary."


Editing the first half of letter "E" on a computer with a sticky "i" key. If things look unlikely, just keep adding "i's" till they make sense. And is there a more grammatical and coherent way to write the plural of the letter "i"? Also, the T-shirt of the year at the Jersey Shore for muscular dudes or busty females reads "suns [sic] out, guns out." Unless there's more than one sun in New Jersey (SOBs don't count) in which case ditch the "sic." At least it's not "gums out," which is sort of more apropos, at least in Ocean City. Mrs. Etymonline refuses to take a picture of me in one.


You there, em-, you're not fooling anyone. You're nothing more than ex- or en- before a "b" or "m" or "p." As far back as ancient Greece, apparently, people when they spoke the sequence inb turned the -n- to an -m-. It goes on right through Latin and its children. The -i- went to an -e- in French on the way to Hastings, but still you have to search pretty hard for an English word that starts enb-. But we don't have a problem with the words in unb- (unbalanced, unbeaten, unbecoming).


The "FL-" pages in an English dictionary start silly but turn spooky. Everything echoes in the train-car where doppelgangers meet. The dictionary at "FL-" holds words from far afield that have no mutual DNA, yet have come to mean much the same thing by way of the same sound — flood, fluid, flowing, fluvial. It holds words from the same tongue that seem to sound like what they mean — but mean very different things: dainty flutter or a sodden flop. Underfoot are horrid inbreds like fly and flee. I wanted to write about it here, but it's like writing about dreams.


Please pardon the digital silence. The etymonline elves are taking their annual week of vacation with family in the Carolinas to escape the ice and cold. Don't worry; the work continues. The ga- words are being cleaned and lubed, and the weed-whacker's started on ge-. I'm learning to use an i-pad thing which is like Shrek learning to do needlepoint and involved many breaches of the Second Commandment. Meanwhile I've seen Canopus for the first time and the cardinals are singing. And I'm pulling apart the history of the Latins for the seminar Gina Cooke and I are giving in Chicago in March. Fascinating stuff; I should post some previews when I get home.


Finished polishing the silverware in the GA- part of the alphabet today. There's always some trepidation when I approach one of the "most-quoted" entries (such as gay), because I'm going to break it down and check it and re-pack it, and if I find new or better information, or see something that is put together poorly, the entry might change a good deal. And that could pull rugs out from under other people who relied on the site for their information.

I'm pleased to say gay came through with only minor and unimportant alterations. But it's a timely reminder that etymonline necessarily trades authority for flexibility. It isn't ever carved in stone. It's always trying to improve and be better at what it is, but that means it changes, or can change.


Editing the early part of "G" is a pleasant jog through hedgerow country, long open stretches of level tillage and stone steeples — Anglo-Saxon, French-from-Germanic — then at intervals an enormous wall-high tangle of a verb — gear, get, give — that you have to sit down and spend a week picking apart, but it's pleasant work, really. Then, midway through ge- you suddenly sink into an algae bloom of "gen-" words, all of them ultimately from that fecund root in Indo-European for "give birth to." Now you're slogging a Carboniferous swamp, a vegetable riot in Latin and Greek. Oh, there are a couple jokers in the pack, a Greek word for "knee," etc. But till you get to geo- it's all birth and procreation for long stretches. At the end you climb out of it and sit down to scrape the muck, half afraid to find a butterfly, "very beautiful and very dead," under your boot.


Editing the im- section of etymonline is an appropriate Halloween activity; that stretch of the dictionary is a long graveyard of abortive words. Occasionally there's a tangle of something living (imperial/imperialism), or a sturdy tree (important), and even an exotic not from Latin (imp); but in general it's row upon row of lexical tombstones. Grass grows in the walk, and the wind across the shorn hills suspires a ghostly Latin.

Impartible in Shakespeare's day could mean "incapable of being divided" (im- + partable) or "capable of being imparted" (impart + -able). Today it means nothing and even the spell-checker knoweth it not.

To impassion was "to fill with passion," but impassible meant "emotionless."

Impassionate could be "free from passion" (in- "not" + passionate) or it could mean "strongly stirred by passion," from Italian impassionare "to fill with passion," which uses the other im-, the one that means "in, into." It, too has been left behind.

They reach up to you wordlessly from their crypts as you look into them, craving life, but you can't give it to them. The flaw in Latin is too much for our language to bear.


I'm trying to edit in- words, which are one long prairie of Latin, except the Regency beaux left their pants all over the place.

Actually the pantsing of in- begins a little earlier, and with the ladies, not the beaux. Around 1790, inexpressibles began to be used in polite conversation for trousers:

"I have retain'd the word BREECHES, as they are known by no other name amongst country folk.—The change from vulgarity to refinement, in cities and towns, has introduced other appellations; there they are generally called SMALL CLOTHES, but some ladies of high rank and extreme delicacy call them INEXPRESSIBLES." [footnote in "Poems Miscellaneous and Humorous," by Edward Nairne, Canterbury, 1791]

The silliness blossoms from there. By 1806, they are unmentionables, by 1819, indescribables. For the next decade the words pile on fast, some of them not even pretending to mean the same as inexpressibles: indispensibles (1820); ineffables (1823); unutterables (1826); innominables (1827); and inexplicables (1829).

One more example, from the days when women were not permitted in the gallery of the British House of Commons:

We cannot omit here to state, that, some years since, we recollect a rumour in the gallery [of the House of Commons], that Madame de Staël was sitting, en habit d'homme, in a surtout and military indescribables, listening to the debate, under the protection of Sir J. Macintosh. ["Privileges of Women," in "Retrospective Review," London, 1824]


They rode out each year after the ground froze, the Germans. They were hideous and stank of the butter they used to red their wild hair. Yet they had wonders in metal and could make harness from grass, lacking good, tough reindeer hides. They traded for the amber and the fish and the furs. We had a woman from their tribe to trade their words. Around the fires we told tales. Some of them had been to Rome. We told them that our grandmothers' grandmothers remembered when all was ice.

They told of the things they carried and of how they were well-ruled at home. At times our speaker could not find words, so she spoke theirs, until we understood.

Triangulation of linguistics, archaeology, anthropology say the Finns have lived a long time alongside the people who spoke what became German and Dutch and English. And some words in modern Finnish obviously are borrowed from Germanic languages.

Linguists have pieced together the language that those Germanic-speaking ancestors would have spoken. By reeling time backward over and over, word by word, they have tracked English and its sisters down into the trunk of Proto-Germanic, to the roots, reversing time till the sprout blooms backward into the earth.

But they can't know if they're right; their reconstructions always bear the albatross of asterisks.

Finnish is a non-Indo-European language, utterly unlike the Germanic or Slavic languages near it. It is like amber. It was isolated, then oppressed — who can say why, but it has conserved itself with little change for 2,000 years.

Ring is Modern English and German, and it has been ring since the Middle Ages or so. But the linguists trace English/German ring, and words like it in Danish, Dutch, etc., to an imaginary Proto-Germanic word *hringaz, which is their highly educated guess at how that word would have sounded when the speaker's breath still hung in the air a thousand years before anyone wrote down anything in Germanic.

But they can't go back in time, so they can't be sure.

Ring is one of the words Finnish borrowed from Germanic. But in Finnish it's not ring. It's rengas.

See it? It's a prehistoric Germanic word, not a modern one.

The most plausible explanation is, the word was borrowed before the first sound shift spread throughout Germanic and gave it a forward accent and left the endings to atrophy. Put that change at very roughly 500 B.C. It's a fly in amber, a mirror flash from down the well.

  • English king, German König from Proto-Germanic *kuningaz: Finnish has kuningas.
  • English soap, German Seife (probably the word they used for the gunk in their hair) from Proto-Germanic *saipon: Finnish has saippua.
  • English leek, German Lauch from Proto-Germanic *lauka-: Finnish has laukka.
  • English rope, German reif from Proto-Germanic *raipaz: Finnish raippa.

The Germanic roots are apparent in the Finnish words, in their full forms and with the vowels unchanged.

From an academic view, accidental apparent confirmation of the PIE educated guesswork. Perhaps there are other explanations. From a human view, perhaps the wind lay still long enough for once to let you eavesdrop Bronze Age conversation. Your hair should be on end; that was time-travel and the shiver you feel may be from that remembered ice.


The letter K feels Germanic. It clicks like hobnails on marble, it stands like a rune (no curves or upright right angles to cross the grain), we meet it in the purer air of German, in words where the bastard English has a sinuous C, in kalt for cold, Kreuz for cross, Klasse for class.

Yet K's as Semitic as any of them, from Phoenician kaph or something like it in scripts to keep accounts in a Levantine port where some Greek galley-master saw it and took it home and told the folks its name was kappa.

The use of K in German turns out to be, like most things about German, recent and deliberate; C was not fully regularized to K (or Z, depending on pronunciation) until the spelling reform of 1901.

Early Romans made little use of kappa and instead used gamma for both the "g" and the "k" sounds, the latter more frequently, so that they came to feel the "k" sound was the proper one for gamma.

The Romans got their alphabetic writing via an Etruscan script in which gamma was written as a crescent, thus the shape of modern C.

To restore a dedicated symbol for the "g" sound in Latin, a modified form of gamma was introduced as G. In classical Latin, C has only the value "k," and thus it passed to Celtic and, via Irish monks, to Anglo-Saxon, where K was known but little used.

In Late Latin, however, pronunciation of C began to shift (in the direction of "s"), and this accelerated after the break-up of Latin. In Old French, many "k" sounds drifted to "ts" and then "s," but they still were written with a C.

Greek names with a K that had been brought into Latin early also had been regularized with a C spelling. They then underwent the Late Latin sound-shift; hence the modern un-Greek English pronunciation of Cyrus, Circe, etc.

The many Greek words that arrived in Latin after the sound shift often were Church words (such as Kyrie). To keep their pronunciation clear, the Romans tended to write them with K.

K thus became a supplementary letter to C in Medieval Latin, used with Greek and foreign words. Most of the languages descended from Latin had little need of it, having evolved other solutions to the sound shifts.

After the Norman conquest, new scribal habits brought over from France restricted the use of the now-ambivalent C and expanded the use of hitherto-neglected K in English to clearly mark the "k" sound.

Since then, English K has been pressed into duty for words transliterated from Arabic, Hebrew, Turkish, Japanese, Hawaiian, and dozens more languages, where it represents several different sounds. In older borrowings these often followed traditional English spelling and were written with a C- (Corea, Caaba).

The life story of the letter leaves the K section of an English dictionary a lexical rag-and-bone shop.


[And now, almost seven years later, I'm at the tail end of P- again, where I was at the top of this]

 Letter P is a long slog to edit; it's right up there in the top three (with C and S) for popularity as a word-starter. It's taken me most of a year this time to traverse it.

But it's a pleasant range country to ride, or else I've grown used to the slog.  After you scramble down out of the mighty mountains of the pre- and pro- words, the land calms and greens and you leave the letter by a pleasant country path across the rest of PR-, then PU-, then a few PY-s.

With the border in sight, among the PULV- words, a plot of flowers caught my eye, and I came down to see. They were near a stream bank, out of the sun, amid moist grass. They might have been the flare of flowers Frost's mower left for his fellow to find. But no, those were wildflowers, and these had the feel of flowers meant to be looked at.

There was pulverescence "incipient powderiness; tendency to become powdery." It's attested by 1828 in floral descriptions, but it's better suited to poetry, or to fixing an angelic nimbus to the common dust-bunny.

There was pulveraceous "having a dusty or powdery surface." Which ought to be used thus: My apartment is in a perfectly pulveraceous condition.

There's pulverulence "dustiness; the state of being dusted over" (1727), which seems too good a word to languish in descriptions of chalks and insect scales.

The sound quality in pulv- makes a lovely word-opening, gentle but not insipid, smooth and lipped, clean and mellifluous. A gentle puff then an undulation. It is close to pulcher, the Latin for  "beautiful," but improves on it with -v- for -c- (or -ch-, thanks to Greek influence on Latin).

Then there's pulveratrix (1770). It means merely "a bird which cleanses itself by wallowing in dust," but such a rich, glittery, sensuous sound should enjoy a racier life than it can find in ornithology textbooks.

These words, in their little cluster, were achingly beautiful. I mean their sound, which is a word's being. I mean the linked sound qualities, the potent noise that mere typographical letters can only wish they were. That sound, any word-sound, has no meaning until given one. And the meaning it gets seldom quite fits it.

For some words the sound is the meaning, yes, but these pulveraceous blooms are not onomatopoeic. They have all got their roots in Latin. Hence their essential formality, their self-awareness. Not Frost's wildflower patch, but an eighteenth-century gentleman's strolling garden, in a generation that held botany a queen of science. Garden flowers quaintly gay, neatly drawn in verse by Collins, Cowper, or Gray.

But maybe Frost was on them after all. I enjoyed them (and lost all sense of editing anything). And perhaps other minds who write or read dictionaries for the love of words, since 1727 or so (Bailey's) had stepped out of their way just so, to breathe here and watch those anonymous flowers nod and think it as fine a sight as any in the book.



Latest stories