Thursday, August 18, 2005

Carnival of the Etymologies

Usually, when I do this, I pluck words from headlines and trace them back through time to their reconstructed roots. That's a walk down the garden path of the scholar. This time, take one root and follow it up its branches into leafs and twigs of words in Modern English. That's the jungle of history.

I chose at random the productive Proto-Indo-European verb root *gno- "to know." This word enters Modern English by the three main paths words usually take to get here: Germanic, Latin, and Greek.

The *gno- seed scatters its blossoms through many non-English languages, too, of course. There are words in this family that never have been heard in an English sentence, such as Old Persian xšnasatiy "he shall know," Old Church Slavonic znati "to know," Russian znat "to know," Old Irish gnath "known," and Sanskrit jna- "know."

But in English, the most obvious native form of this root is know, recognizable in spelling and even pronunciation as the same word that scholars have reconstructed from 6,000 years ago, spoken by some horse-taming tribe near the shores of the Black Sea.

Know comes from Old English cnawan, which is the Anglo-Saxon form of Proto-Germanic *knoeanan, which in turn is the Germanic version of the prehistoric *gno- root.

*Knoeanan also had offspring in Old High German bi-chnaan "to know," Old Norse kenna "to know, make known," Old Frisian kanna "to recognize, admit," German kennen "to know," and Gothic kannjan "to make known."

It's only from the wider (non-Germanic) reconstruction that scholars determined the English kn- consonant cluster is close to the original sound, and the other modern Germanic languages have shifted away from it. English has shifted away from it, too, of course -- in pronunciation. But the reduction to "n-" in standard pronunciation came about after the spelling was set, and so the kn- remains on paper. The simple "n-" pronunciation was in use before 1750, and for about a century before that it had been pronounced "hn-," "dn-," or "tn-." It was fully voiced as "kn-" in Old English and Middle English.

The English word has a widespread application. We use know to mean both "to know as a fact" and "to be acquainted with (something or someone)." The Anglo-Saxons used two distinct words for this, witan and cnawan. The first survives now only in the phrase to wit, but it also is related to the noun wit meaning "mental capacity." The Proto-Indo-European root of this is *woida- the perfective form of the verb root *weid- "to see," and the sense is metaphoric; to "have seen," thus "to know."

Most Indo-European languages use their forms of these two words in the two senses of "to know" (e.g. German wissen/kennen; French connaître/savior; Latin novisse/cognoscire; Gothic witan/kunnan; Old Church Slavonic znaja/vemi). But in most of them the pure distinction of the words has blended, though English has gone further than most in shifting the workload to one verb.

What about knowledge? It turns up first in Middle English, cnawlece. The first element obviously is know, but the second element is obscure, and some linguists speculate that it may be identical with the -lock suffix of action or process that is found in wedlock.

Acknowledge is in this family, too, but it's an oddball. It's a 16th century blend of Middle English aknow (from Old English oncnawan "understand") and the Middle English verb knowlechen "admit." Somehow, in the merger, a parasitic -c- slipped in, so that, while the kn- became a simple "n" sound, the c stepped up to the plate to preserve, in this word, the ancient kn- sound.

Besides "to know as a fact" and "to be acquainted with (something or someone)," there's a third meaning in know, which is "to know how to do something." In English, this has adhered to the verb can.

Can represents the 1st and 3rd person singular of Old English cunnan "to know, have power to, to be able." This came up from the common Indo-European root via Proto-Germanic *kunnan "to be mentally able, to have learned."

The past tense of Old English cunnan was cuðe, with a "th" sound at the end. It forms the first element in the man's proper name Cuthbert, which literally means "famous-bright." But its ending changed in the 14th century to standard -d(e). An -l- was added in the 16th century on the model of would and should, where it is historic, and the word came out could.

The Old English form of the word survived, however, in uncouth (Anglo-Saxon uncuð) which originally meant "unknown, uncertain, unfamiliar." The meaning "strange, crude, clumsy" is first recorded in 1513. In the 1890s, couth was peeled off from this negative compound and re-born as a word of its own, with a new sense of "cultured, refined,"

The pure form of Old English cunnan has come down as the verb con meaning "to study." This is one of four cons in English, all unrelated. The one meaning "negation" (mainly in pro and con) is short for Latin contra "against;" the sense "swindle" is an American English shortening from confidence man, who was so called from the many scams in which the victim is induced to hand over money as a token of confidence. The meaning "to guide ships" is from French conduire, from Latin conducere.

The past participle of cunnan, meanwhile, has come down as cunning, which originally meant "learned." The sense of "skillfully deceitful" probably had emerged by the 14th century.

The ancient Greeks took a different path to form their word for "cunning." They took it from a word related to medos "counsel, plan, device, cunning," and median "to protect, rule over," and came up with medeia, which also was the name of the famous sorceress, daughter of the king of Colchis, in the mythologies.

Old English cuð "known" also yielded a noun, cyðð, which to the Anglo-Saxons meant "native country, home." It survives in the alliterative phrase kith and kin, attested from 1377, which originally meant "country and kinsmen."

[kin, which looks like it might be another relative, is from a different Proto-Indo-European root, *gen-, "to produce."]




In the Latin branch of the *gno- family, you meet gnoscere "to get to know, get acquainted (with)," from *gno-sko-, a suffixed form of the ancient root. This Latin verb dumped several dictionary pages worth of words into Modern English, some directly from Church Latin, some via French vernacular.

Among them is notice (from Latin notus "known," past participle of (g)noscere). Notice was a noun exclusively at first. The verb is attested from c.1450, and it originally meant "to notify," later "to point out" (1627). The main modern meaning "to take notice of" is attested from 1757, but it was long execrated in England as an Americanism (or occasionally as a Scottishism, the two crimes not being clearly distinguished).

Take Latin gnoscere and add the prefix com- meaning "together, with" and you get cognizance (from Latin cognoscere "to get to know, recognize") and connoisseur (from Old French conoisseor "a judge, one well-versed," from cognoscere).

Put the Latin prefix ad- "to" in front of that, and you get acquaint, from Latin accognitus, past participle of accognoscere "know well." Or put the Latin pre- "before" at the head of the word and you get precognition "foreknowledge," from Latin præcognoscere "to know before."

Put the negative prefix in- in front of it, however, and you get incognito, via Italian from Latin incognitus "unknown," from the past participle of cognoscere. Those scrupulous about Latin might want to retain the form incognita when the word is ued in reference to women.

Put the prefix re- "again" at the head of the word and you get recognize, from Latin recognoscere "acknowledge, recall to mind, know again, examine, certify."

A much less obvious relative here is quaint, which came into English about 1225 meaning "cunning, proud, ingenious," from Old French cointe "pretty, clever, knowing," from Latin cognitus. The sense of "old-fashioned but charming" is first attested 1795, and it could describe the word itself, which had become rare after c.1700, though it soon recovered popularity in this secondary sense. The word may have fallen from use because it suggested an obscene pun. Chaucer used quaint and queynte as spellings of cunt in "Canterbury Tales" (1386), and Andrew Marvell may be punning on it similarly in "To His Coy Mistress" (1650).

Another suffixed form of the ancient root was *gno-ro-. Stick a negative prefix at the front of that and you get, eventually, ignorant via Latin ignorantia, from the past participle of ignorare "not to know, disregard." Ignore comes from the same source.

Ignorance is widespread in the world, ancient and modern, so it's not surprising that similar compounds are widely attested. Old English uncuð is essentially a brother of ignorantia. So is Old Norse ukuðr, Gothic unkunþs, Sanskrit ajnatah, Armenian ancanaut', Greek agnotos, and Old Irish ingnad "unknown."

Ignoramus originally was an Anglo-French legal term, from Latin ignoramus "we do not know," the first person present indicative of ignorare. As a legal term, this was what a grand jury wrote on a bill when it considered the prosecution's evidence insufficient. The sense of "ignorant person" came from the title role of George Ruggle's 1615 play satirizing the ignorance of common lawyers.

From the ancient suffixed form *gno-dhli- comes noble. The family tree of this word goes through Old French to Latin nobilis "well-known, famous, renowned, of superior birth," earlier gnobilis, literally "knowable," from gnoscere. The prominent Roman families, which were "well known," provided most of the Republic's public officials.

In Middle English the word came to mean "distinguished by rank, title, or birth." The sense of "having lofty character, having high moral qualities" is recorded from 1601. But another quality of nobilities may be preserved in the scientific term noble gases (1902), said to have been so called for their "inactivity" or "interness."

Finally, the suffixed zero-grade form *gne-ro- comes up to English as narration, from Latin narrare "to tell, relate, recount, explain," literally "to make acquainted with," from gnarus "knowing." The verb narrate is first recorded in 1656, but it was stigmatized as Scottish and not in general use in English until the 19th century.




Latin generally dropped the gn- consonant cluster down to a simple "n-," but the Greeks, who had more supple tongues, tended to keep it.

Greek forms of the ancient root that have come into English include gnostic, a noun use of the adjective gnostikos "knowing, able to discern," from gignoskein "to learn, to come to know." This word was applied to various early Christian sects that claimed direct personal knowledge beyond the Gospel or the Church hierarchy.

The opposite of gnostic, of course, is agnostic, but this is a modern word, coined in 1870 by T.H. Huxley from Greek agnostos "unknown, unknowable," from a- "not" and gnostos. It has been ably defined as "one who professes that the existence of a First Cause and the essential nature of things are not and cannot be known." The word is sometimes said to be a reference to Paul's mention of the altar to "the Unknown God," but according to Huxley it was coined with reference to Gnosticism as an early Church movement.

"I ... invented what I conceived to be the appropriate title of 'agnostic,' ... antithetic to the 'Gnostic' of Church history who professed to know so much about the very things of which I was ignorant." [T.H. Huxley, "Science and Christian Tradition," 1889]

Also from Greek gignoskein, with prefixes attached, are diagnosis (literally "a discerning, a distinguishing") and prognosis "forecast of the probable course of a disease," from Greek progignoskein "come to know beforehand."

Greek gnomon "indicator," literally "one who discerns," is the root of English gnomon "pillar that tells time by the shadow it casts" and gnomic "full of instructive sayings."

Gnomic is not derived from gnome "dwarf-like earth-dwelling spirit," which is an obscure word out of Paracelsus, who may have abstracted it from Greek *genomos "earth-dweller."

Labels: