On Being Pretty Ugly: A Nice But Quaint Oxymoron
By Anatoly Liberman
The etymology of the adjective pretty has been investigated reasonably well. Many questions still remain unanswered, but it is the development of the word’s senses rather than its origin that amazes students of language. The root of pretty, which must have sounded approximately like prat, meant “trick.” Judging by the cognates of pretty in Dutch, Low (Northern) German and Old Icelandic, the adjectives derived from this root first meant “sly, crafty, roguish, sportive.” Before us is evidently a slang word that has been current in Northwestern Europe since long ago, a circumstance that can perhaps account for some of the vagaries of its history. This adjective was well-known in Old English, but it suddenly disappeared from written sources and reemerged only in the 15th century with more or less predictable meanings. Perhaps it had descended into the underworld and resurfaced under the influence of its German cognates (thousands of people crossed the channel in both directions at that time) or returned to everyday use devoid of its too playful connotations (assuming that it ever had them), and did so without any push from outside. Conversely, for several hundred years, it may have been restricted to a few dialects and escaped the attention of authors by chance. This would be a common event in the life of words. As noted, Old Icelandic also had a related noun, and opinions are divided on whether it was native there or borrowed from English. Back in Middle English, the oldest recorded meaning developed from “crafty, wily, artful” to “clever, skillful” and “pleasing, fine, proper.” The original sense was forgotten.
In the history of the word pretty and its cognates, we notice numerous forms with r before rather than after the vowel. Such are, for instance, Dutch part, alongside pret, and Engl. dialectal purty. This type of variation is common and does not necessarily point in the direction of slang. (To cite a less known example: Engl. pang once had the doublet pronge, and the pr- variant may be the earliest one. Such doublets are not always amenable to the laws of historical phonetics.) None of the forms cited above has anything to do with Engl. pert or pertinent. Attempts to trace pretty to a Latin or Celtic etymon or to connect it with the English verb prate (which has a cognate in Dutch) have met with no or minimal approval.
The line between “proper; physically fit” and “good-looking” is easy to cross. Thus, handsome, so obviously derived form hand and some with the meaning “handy, easy to handle,” soon changed to “apt, happy; considerable” and “beautiful” (“handsome is as handsome does”). Likewise, clever seems to have been coined with the meaning “brisk, sprightly,” but in the 17th century it could also mean “handsome.” Somewhere in this loose conglomeration of senses, we usually find “considerable.” The most colorless of them all, it allows us to use the adverb pretty in phrases like pretty dark and pretty scary, in which the idea of prettiness is suppressed and pretty means “quite.” Only a deliberate joke, a combination like pretty ugly makes one aware of how incongruous such word groups are.
As long as we are dealing with the notion of cleverness, two more adjectives can be mentioned. One is nice, a favorite example in works on historical semantics. Latin nescius meant “ignorant” (sci-, as in science). Its Old French descendant nice “silly, simple” progressed so: “foolish” à “wanton” à “coy, shy; fastidious, dainty” à “difficult to manage; subtle; precise; minutely accurate” (from “foolish”!) à “agreeable, delightful.” If a nice pretty girl knew from what etymological depths she has risen, how from a wily, ignorant wench she became a pleasant and charming maiden! Luckily, speakers have no clue to the past of the language they speak. Another similar word is quaint. We realize with an effort that it is almost a homonym of acquaint. Its Latin etymon is cognitus “known.” In Middle English, it appeared (also from Old French) with the meanings “clever; skillfully made, elegant.” The final stage is familiar from the modern language: “uncommon but attractive.” “Uncommon” is important, for some early dictionaries define quaint as “unknown.” “Unknown,” and that about an adjective that came into being with the meaning “known”!
The spelling of pretty is also odd, and the pronunciation pritty has not been explained to everybody’s satisfaction. About half a millennium ago, if not earlier, all the long vowels of English began to change. Hence the difference between the value of Engl. a and e, for example, and that of Italian, French, or German a and e. Short vowels occasionally adjusted themselves to their long partners: long e (as in Italian or German) became what we now pronounce in be, he, we, and short e turned into short i. Sometimes we observe the adjustment happening in a well-defined context. Thus, the group -eng regularly yielded -ing. Suffice it to try nonsense words like reng –theng—seng to realize that none of them sounds native. The only two common words with eng are lengthen and strengthen. (Engine, often pronounced in dialects as ingine, may be added to them.) In all others, eng became ing. Wing, mingle, think, link, and so forth once had e, as do their cognates in German—for instance, denken and Gelenk. But the spelling did not always follow the new pronunciation, and we still spell English and England instead of Inglish and Ingland. Pretty has no n after e, but it probably shared the fate of wing and the rest. Pritty may be a dialectal pronunciation of pretty that has been accepted by the Standard. It is amazing how troublesome such a nice and simple and word may be.
Anatoly Liberman is the author of Word Origins…And How We Know Them. His column on word origins, The Oxford Etymologist, appears here each Wednesday. Send your etymology question to firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”