Oxford University Press's
Academic Insights for the Thinking World

Monthly etymology gleanings for February 2015

One month is unlike another. Sometimes I receive many letters and many comments; then lean months may follow. February produced a good harvest (“February fill the dyke,” as they used to say), and I can glean a bagful. Perhaps I should choose a special title for my gleanings: “I Am All Ears” or something like it. Unfortunately, every good title, witty pun, and memorable rhyme, including The Importance of Being Earnest and intellectual / hen pecked you all, has already been used by others. Perhaps my “ears” have also occurred to some gleaner more than a hundred years ago. I did not check.

House between Germanic and Slavic

How do we know that Slavic borrowed Germanic hus (long u, that is, a vowel of Modern Engl. boo, coo, woo), the older form of house, rather than the other way around? Could hus be a loanword from Slavic? The situation is apparently so obvious that no one found it necessary to discuss it, not even A. Stender-Petersen, the author of the old but still most useful book Slavisch-germanische Lehnwortkunde (1927). The same holds for the etymological dictionaries of Slavic and Germanic I have consulted. At best, they say: “The word was borrowed by Slavic.” This omission is irritating. I think the reason for the universally shared conclusion is of phonetic nature. Contrary to Slavic, all the Old Germanic languages have the same old form (hus). It is easy to show how, with regard to the vowel and the consonants, Polish, Russian, and other Slavic languages modified hus, but reconstructing the early Slavic protoform that was supposedly taken over by Germanic would pose a nearly insoluble problem. Also, the impression is that Germanic hus occurred earlier than its Slavic modifications.

Does French hameau have a suffix?

I am all ears.
I am all ears.

My statement that hameau does not have a suffix pertains to the modern form. Engl. hamlet has a pseudo-suffix: –let, as in ringlet or starlet, is a viable morpheme, but if we subtract the last three letters from hamlet, the stub will carry no meaning because hamlet is not a small ham. Nor would we have any luck with –et. But at least there is a feeling that hamlet contains two parts. The situation is familiar: some element (in the past, certainly a suffix) has attached itself to a sound complex that no longer has an independent existence. In Modern French hameau, there is not even anything to subtract. That is why I wrote that this word has no suffix. Its etymology is another matter, and I discussed it in my post.

Man, earth, world, and worm

Most etymologists recognize the kinship between Latin homo “man” and humus “earth.” But the path from Latin vir “man” (a cognate of Old Engl. wer “man,” which is also the first element of Old Engl. weorold “world”) to vermis “worm” (a cognate of worm) leads nowhere, because the root of the word for “worm” has been found in several verbs meaning “to turn” (one of them is Latin vertere, familiar to English speakers from revert, subvert, convert, verse, and others). Worms certainly wriggle (“turn”). (By the way, that is why the suggestion that squirm is a blend of some verb and worm looks plausible.)

Thresh versus thrash and threshold

It is tempting to refer the difference between the vowels in thrash and thresh to dialectal differences. But, as noted in the previous post, almost identical doublets existed in Old English, and this fact complicates the picture. Reference to dialects does not absolve us from going further and formulating certain phonetic regularities. It appears that no line separates English dialects with e preserved and e broadened before r. The case of marry, merry, and Mary in some varieties of American speech is irrelevant to thresh ~ thrash: there the three words have merged, while here the difference has been retained.

The importance of the threshold in various superstitions is a well-known and well-understood phenomenon, but the word seems to have been coined in a secular context. German Kobold, meaning “goblin” and sounding very much like goblin (a word cited by the same correspondent who offered his comments on threshold), does not seem to have anything to do with cave.

The initial group dw- in English

Initial dw indeed occurs in very few English words, and I mentioned this circumstance in my dictionary (the entry dwarf). Such gaps are always puzzling. Initial tw-, outside expressive or sound symbolic words like tweak, twitch, can be found almost exclusively in the words related to two (twin, twist, etc.) No English word begins with tl, though speakers have no trouble pronouncing little. In Russian, only two words begin with tl-. No one has been able to explain why in English kn– and gn– were simplified to n (knock, gnaw), while elsewhere in Germanic those groups do reasonably well. Phonotactics (the branch of phonetics dealing with the distribution of sounds) is full of such capricious rules.

Slavic mir

From an etymological point of view, Slavic mir “world” and mir “peace” are indeed the same word. Mir meant “community” (a sense still understood in Russian: compare, among others, phrases like vsem mirom “all together”). Living with those belonging to one’s own group promised peace. Later the word split into two. In Russian, for some time, they even had different shapes on paper: mir “world, cosmos” was at some time spelled with the roman letter (that is i) in the middle. A careless typesetter printed the second noun in the title of Tolstoy’s novel War and Peace with this i, a mistake that gave the publisher much trouble and resulted in a long, useless discussion about Tolstoy’s intentions. Did he mean War and the World? No, he did not. The best work on the etymology of Indo-European words for “peace” is still Karl Brugmann’s monograph.

Unrelated Words

Can host and guest go back to the root of Latin os “mouth”? From what we know about those words, no connection between host and os can be established. Similarly, os has nothing to do with the place name Hastings, which goes back to a tribal name. In host and Hastings, initial h is an integral part of their roots, while os never had the form hos.

Nor do the verb be, even though spelled bee in some early texts, and the noun bee go back to the same source. It is usually believed that the original meaning of be was “to grow” or “to become” and that of bee “trembling (insect),” from the movement of the creature’s wings. (Worms turn, bees quiver, and so it goes.)

I’ll now quote the beginning of a letter I received more than a year ago:

“I found it funny that the word for ‘food’ in Italian is like an exact description of where it comes from …ci in cibo [may] come from Greek gi for ‘earth’ and bo is a kind of being like a shortened bero for ‘borne’. So food being what grows from the earth, ‘earthborne’ [sic].”

The writer continues with a long list of similar hypotheses. This is the way some people tried to discover word origins in the Middle Ages. Since that time linguists have developed a more reliable method of treating their material. In etymology, ingenious speculation, unsupported by method, seldom results in viable conjectures. I can suggest to our correspondent that, whenever he feels an interest in such matters, he make use of some good dictionary. The answer given there may not be fully convincing, it may even say only “origin unknown,” but at least it won’t offer a fanciful guess.

Spelling bee, the bane of our being.
Spelling bee, the bane of our being.

A Note on Spelling and Spelling Reform

I agree that a good deal in the spelling of English makes sense from a historical point of view, but this is exactly the point. Like many people in my profession, I am conservative when it comes to usage, that is, I accept the fact that Old English became Middle English and study both periods with great pleasure, but I don’t want Modern English to become Postmodern and wince at he has showed (though American dictionaries say that the past participle showed is a legitimate competitor of shown) and keep saying sneaked for snuck. But in spelling loyalty to tradition is (even in my view) detrimental, the more so as tradition does not always deserve admiration. I see no sense in spell versus dispel, till versus until, full versus the suffix –ful, and even in all versus always and also.

Valerie Yule’s list of 38 very common words that cry out for change is excellent, but we need a more systematic approach to the reform. That is why I would support regularizing the use of double letters, abrogating x and q, as well as the digraph ph, and so on. (Apophthegm is not my favorite English word, though I have nothing against it in Greek.) However, my approach is opportunistic. I will accept any version of the reform the public will agree to instil(l). Certain things will be impossible to implement, for instance, “phonics” or the introduction of diacritics for sh, and ch, however reasonable their introduction may be. I see no point in fighting a battle we are sure to lose despite my acceptance of Cyrano’s aphorism (“aforism”): “One does not always fight to win.” As regards the causes of our erratic rules, see the comments by Masha Bell. Something could be added to her points, but for initial orientation her list will suffice.

Image credits: (1) A child listens discreetly as his mother and her friend discuss Christmas presents, 1913. Cartoon by Bob Satterfield. Public domain via Wikimedia Commons. (2) National Spelling Bee. Photo by erin m. CC BY-NC 2.0 via erin_m Flickr.

Recent Comments

  1. B Slade

    re: “No English word begins with tl, though speakers have no trouble pronouncing little.”

    In Jamaican patois these “tl” sequences are transformed to “kl”. See: “A Likkle Sound Change in Jamaican Creole English”: http://staefcraeft.blogspot.com/2009/10/likkle-sound-change-in-jamaican-creole.html

  2. John Cowan

    Indeed, sometimes meaningless suffixes arise because the word containing them is really a loanword. In the most famous case, cranberry < Low German kraanbere, the latter is transparently ‘crane+berry’. But when the word was imported into English, displacing the native marsh-whortleberry and fenberry, the second half of kraanbere was etymologically nativized but the first half was not: if it had been, we’d say *craneberry.

  3. EugeenLV

    Just wondering why strawberries are called strawberries and how the modern form has been coined (I hope this has not been covered earlier)

  4. Masha Bell

    Perhaps spelling reform could be started with a just a short list of words that could obviously do with improving, such as changing ‘you and I’ to ‘u and i’, dropping the useless –e from ‘have, give, live, are, gone’ and reducing the most misspelt heterographs ‘its/it’s’, ‘there/thare’ to just ‘its’ and ‘thair’? The still surviving British distinction of ‘a practice/to practice’ which UK teachers endlessly get wrong should surely be reduced to just ‘practice’ like in the US where this has caused no problems whatsoever?

    But with at least 4,219 quite common words containing some unpredictable letters, only a more systematic approach can make a substantial difference to literacy acquisition overall. Regularising consonant doubling in words of more than one syllable is one of the strongest candidates for it.
    With only 503 words having regular doubling (rabbit, reckon, offer …),
    567 without it (habit, record, profit….) and
    217 with redundant, phonically misleading doubling (abbreviate, account, offend)
    this aspect of English spelling is blatantly shambolic. It sabotages the already complex use of a, e, i, o and u as both long vowels (pop, pope, popper – proper).

    Surplus –e endings clearly do so too (inflate – delicate). As do irregularly spelt short vowels in nearly 200 very common words which children meet at the very start of learning to read and write (many, women, money – cf. penny, swimming, funny). A reform that combined regularising the spelling of short /a/, /e/,/ i/, /o/ and /u/,
    consonant doubling and cutting of surplus e’s would be easy to explain and understand and do enormous good, with little inconvenience to currently proficient readers and spellers.

    The spellings of /ee/, long /o/ and /oo/ could do with regularising too, if it should ever be decided to make English spelling even more learner-friendly and to bring it closer to the difficulty level posed by German or Dutch, but starting to make the use of a, e, i, o and u as long and short vowels more transparent would be a good beginning.

  5. John Cowan

    Eugeen: The word strawberry has no relatives and its meaning is unknown.

  6. Yves Rehbein

    Some words to compare to strawberry:

    Ger. Strauch (shrub), Ger. straucheln (strugle, stumble) Ger. Streif (stripe), but of course also:

    straw, cognate to stray and strewn, from PIE *strew (to spread, scatter), whence also Latin sterno (e.g. to spread, but also to besprinkle sth.), whence strata; Not to be confused with — or maybe that’s exactly what happened and gave the besprinkle sense to Lt. sterno — Proto Germanic *sterno (star, Ger. Stern), from PIE *h₂stḗr (star), from PIE h₂eHs- (to dry; to burn, to glow; hearth; ashes), whence e.g. arid (dry); Not to be confused with *h₁er- (earth), whence earth, Welsh erw (field; if not from *h₂erh₃-, to plough, whence Irish arbhar, corn, cereal) and Ger. Erde — compare Ger. Erdbeere (strawberry).

    Wiktionary links it and straw to sterh₃-, but explains it as “[…] the applicability to berries growing on a bush”.

    Now, I shouldn’t guess, but it’s no sperculation that they are growing close to the ground, glowing red, and are sprinkled with seeds, because I don’t even know since when it was known.

Comments are closed.