We were watching the great 1978 series Connections (available at the Internet Archive), and James Burke (the creator, writer, and presenter), discussing the history of timepieces, referred to “verge and foliot clocks.” I was, of course, intrigued, and as soon as the episode ended I investigated. Verge, which can mean any sort of rod, shaft, pole, or wand as well as “The spindle or arbor of the balance in the old vertical escapement,” is from Old French verge < Latin virga ‘rod, etc.’; foliot is more interesting, having the senses “? Foolish matter. Obsolete. rare,” “A kind of goblin. Obsolete. rare,” and “A type of clock escapement consisting of a bar with adjustable weights on the ends. (Disused.)” The etymology:
? < Old French foliot.
Notes
The Old French word is recorded only as meaning watch-spring; but according to Hatzfeld & Darmesteter it is derived from the verb folier to play the fool, to dance about, and so may have had other meanings related to this verb. Compare the surname Foliot, known from 12th cent. in English. How Burton obtained the word there is nothing to show; he evidently connects it with Italian folletto, = French (esprit) follet, hobgoblin, properly a diminutive of fol foolish. Can it be a misprint for follet?
As you can probably guess, that’s an ancient unrevised entry (from 1897). I have to say it’s a very attractive word, and I regret that it’s disused.
Treccani has for folietto
2. Nel linguaggio marin., ogni cavo sottile che assicura un oggetto mobile; in partic. quello che lega il remo allo scalmo.
To paraphrase :”in sailor’s slang, any flexible/pliable cable that holds in place a moving part, in particular the cable that fastens the oar to the oarlock.”
This is suggestive, but the Italian word does not apply to clockwork.
fastens the oar to the oarlock
If @Hat’s collecting obscure nautical terminology, you can’t go past thole pin(s), used to hold an oar to the gunwale, and as a fulcrum for rowing. With impeccable PIE etymology
Verge is much favoured in the more strait-laced of my dictionaries of West African languages for the sense (ahem) membrum virile. (The word, alas, does not seem to be reconstructable to proto-Oti-Volta. Nor its distaff counterpart.)
So oed removed the oed2 from the web:/
___
“strait-laced” – so verge is more straight-laced than membrum virile? In absence of stylistically neutral words for [need an emoji] I prefer dictionaries that use pictures….
drasvi said: “so verge is more straight-laced than membrum virile? In absence of stylistically neutral words for [need an emoji] I prefer dictionaries that use pictures….”
Good grief, of all things that should not be without an emoji of its own …
The vast majority of uses of the eggplant emoji are not referring to eggplants.
I saw “the yard” in a straight-laced 19th century dictionary. I had to look it up.
https://www.poeticous.com/robert-graves/ogres-and-pygmies
What is the difference (in this lexicographic context) between saying a word, or a particular sense thereof, is “Obsolete” and saying it is “Disused”?
Well, since 2019 there is a “drop of blood” period emoji, following a campain meant to defeat the stigma.
Usually when periods are mentioned in chats it’s women compainin at the pain (willingness to do so may depend on culture), so maybe something less physiological would have made more sense. Actually, associated taboos can make it less “private” and thus more appropriate topic for conversation.
WP: “In some societies it involves menstruation being perceived as unclean or embarrassing, inhibiting even the mention of menstruation whether in public (in the media and advertising) or in private (among friends, in the household, or with men).”
“Even the mention” is apparently a Western thing, and the word “even” is perhaps cpnfusing
thole pin(s), used to hold an oar to the gunwale, and as a fulcrum for rowing. With impeccable PIE etymology
Probably not. Thole ‘suffer, endure’ is certainly PIE, but thole as in thole pin is Common Germanic but thought to be unrelated.
“In some societies it involves menstruation being perceived as unclean or embarrassing, inhibiting even the mention of menstruation whether in public (in the media and advertising) or in private (among friends, in the household, or with men).”
“In Ohio, we say the mirror cracked from side to side.” —James Thurber
Thole as in pin, unrelated to the ‘suffer’ word sure:
Not appearing in English ’til mid-15c is no surprise: much nautical technology and terminology didn’t arrive until around then, via Dutch (cognate dol).
Well, the OED says “Ulterior etymology uncertain”; Etymonline agrees with Watkins; Wikt says < *tūl-, *twel- ‘bush; sphere’. Sounds like OED is right.
Danish has at tåle = ‘endure, tolerate’ and åretold = ‘oar fulcrum’. (Unetymological and silent d in the latter). Also told = ‘toll’.
any flexible/pliable cable that holds in place a moving part
I was thinking that perhaps the foliot was originally the ‘leaf’ (folium) of the arbor (cf. ‘the spindle or arbor of the balance in the old vertical escapement’ mentioned above). But then I found this enjoyable video here, in which the presenter discusses the verge and foliot after around the 2:40 mark. From his demonstration, especially after around 3:10, the OED’s description of the foliot as the watch-spring itself, does not seem exact. Rather, the verge and foliot mechanism moderates the rapidity with which thread unwinds from around the shaft (in this version) and in later versions, I suppose, the rate at which a coiled mainspring unwinds. The first cite for foliot in the TLFi is from Froissart Li Orloge amoureus (electronic version here), an interesting extended conceit apparently composed before 1373. It mentions the foliot several times, of which two here:
This evidently salient back-and-forth motion of the foliot in the mechanism would explain the derivation from Old and Middle French foloier ‘act like a fool, frolic, gambol’, Middle French folier.
(Short comment because I am on the road and I have that problem of kicked off the OED after a few minutes when I log in through a new location.)
“The first cite for foliot in the TLFi is from Froissart Li Orloge amoureus (electronic version here), an interesting extended conceit apparently composed before 1373. ”
Yes, cf. the earlies quotation in OED in this meaning (a link to OED2 entry* is in my comment above):
….compared to “a1250 Owl & Night. 866 Ne singe ih hom no foliot.” for the other meaning.
It’s unclear what they mean by “[Fr., 14th cent.: see etym. note.]” – is 14th century when the word first appears in French and is quotation from Froissart in 1899 the first appearance of the [foreign…] word in an English text?
Then we need TLFi and not OED, because we are still dealing with what essentially is a French word.
____
* OED2 used to be online in open access, just not searchable. It seems very recently they removed it.
OED2 used to be online in open access, just not searchable. It seems very recently they removed it.
I really don’t like the direction the OED is headed; it’s as if they’d been taken over by a hedge fund that didn’t give a damn about words or users but only cared about maximizing profits.
drasvi, the OED2 notation is clear and unambiguous if you know how to read it. Yes, “14th cent.” is when the word first appears in French (square brackets indicate information that is only etymological), and yes, 1899 is the earliest attestation they had in an English text (if they had anything earlier, they would have put it in).
OED2 pages are now behind the paywall, at a slightly different URL (“oed2” changed to “oedv2”). I didn’t realize the Internet Archive had been capturing them, thanks for pointing that out. I recall they used to appear occasionally in Google searches.
Most of the OED2 print volumes can also be checked out from the Internet Archive.
AntC: “Not appearing in English ’til mid-15c” — Etymonline specifically says it’s “from Old English” in the part you didn’t quote! Thole is found as a gloss for Latin scalmus in a couple of glossaries from the Old English period. The OED (1912) remarked that it had “a hiatus during nearly the whole Middle English period”, i.e. until c1440 (and that one is in the sense of a pin in a cart, not a boat). But it is very difficult to know whether there was a gap in the word’s usage in English, or just a gap in surviving attestations from the centuries when relatively little was written in English.
Please remember that Etymonline is, as usual, quoting very old scholarship and not the state of the art. Sources like account books and inventories were much less available to the OED1, compared to poetry and histories, and there has been a lot more work on them since. The Middle English Dictionary has since found three citations before 1440, the earliest in 1378-81 reported in a book on Middle English Sea Terms.
Even without the Old English attestations we would know that the word can’t be strictly a borrowing from Dutch, since in that case it would start with d-, not th-. However, the Middle English Dictionary points out “also cp., esp. for form dol-, MDu., MLG dolle”, i.e. there probably was some influence from those cognates, reflected in the spellings of some attestations, but it wasn’t enough to replace the original English th-.
In Eddison’s The Worm Ouroboros there’s a Red Foliot who rules over the Foliot Isles. I think he’s human rather than any kind of hobgoblin, just as the Witches, Demons and so on in the book are just names for human nations.
en branle
TIL about Middle French branle “to move back and forth”, underlying the name of branle the dance style – I’ve known of the latter but apparently never thought of looking up its etymology.
(A non-dance meaning of the same verb previously on LH; I realized I’ve read about that one after claiming the TIL, though admittedly I’ve seen it long before I learned about the dance style.)
I just discovered that alongside apheli[on/um] there exist a variant apoheli[on/um].
Is there a minimally respectable tradition of writing apo-h as apoh- and not as aph- (ἀφ-)?
I think that’s a barbarism. Post-Mycenaean Greek seems not to have allowed /h/ outside the beginnings of words at all.
Someone’s attempt to explain the difference in ru.WP resulted in total mess:
When I search for apohelium in google books, the first result is Svenskt och ryskt lexikon (på nådig befallning utarb. vin Stats-sekretariatet för storfurstendömet Finland) where “apohelium” is translated with Russian afelij, so I looked “aphelium” in Swedish wiktionary… And discovered a proposal that (1) the pronunciation with /f/ is dated (2) that it resulted from erroneous reading of ph as transcription of ϕ.
Again, a weird feeling: it is certainly assembled from Latin -p- and h-, so the transcribed ϕ exists only in Latin mind and the explanation is not fully wrong, but unless Greeks wrote something else than φ in new coinages with roots with h-, -ph- here is φ.
* Klein’s dictionary does not contain anything like this.
Also i guess I can write aphedron “latrine” as apohedron (dodecahedron, icosahedron, apohedron, polyhedron….), and the same with katahedra (and speaking of periods, ἄφεδρος “menses muliebres”, which I did not know).
The spelling apohelion previously on LH; at the time I was unaware of any actual instances, but they do apparently occur.
EDIT: see also previous LH discussion for aphedron.
DM:
Post-Mycenaean Greek seems not to have allowed /h/ outside the beginnings of words at all.
Hmm.
As for la verge and [se] branler, these often indeed mean in current French “membrum virile” and “masturbate [onself]” respectively. For the latter compare certain meanings of tripoter (“to fiddle with”, “to grope [someone]”) and se tripoter (“to masturbate onself”, “to grope [each other]”).
That’s more like “/rː/ gets devoiced automatically, like initial /r/-“.
Greek morpheme-initial /h/ seems to have been preserved in compounds, based on epigraphic and papyrus spellings that use Η/Ͱ/etc. and on Latin transcriptions (Euhemerus, synhistor). The only morpheme-internal /h/ I know of is in ταώς ‘peacock’ (a loanword of unclear origin), which according to Athenaeus was pronounced ταὧς in Attic.
(A previous version of this comment disappeared, and had some mistakes anyway. Please ignore the old version if it shows up in the moderation queue!)
John Cowan said: “Etymonline agrees with Watkins” (on PIE origin for thole, noun)
Rather, Etymonline *quotes* Watkins, but Douglas Harper doesn’t (and never pretended to) make his own judgement, so that doesn’t count as agreeing.
Watkins was following Pokorny in connecting thole to the widely accepted root *teuə- (in AHD’s notation, after loss of laryngeals), ‘to swell’, via “Probably suffixed zero-grade form *tu-l‑.” (Etymonline has e where there should be schwa in this and several other quotations from Watkins; old font problem, or what?)
But Kroonen does not follow this reasoning, if I understand him correctly; he does claim Greek and Lithuanian cognates, but does not take the reconstruction farther than “European”, i.e., no non-European cognates.
John said: “Wikt says < *tūl-, *twel- ‘bush; sphere’”
No source given, and the supposed PIE root is a red link (is there a way to tell whether it never existed, or was deleted?), so I think that one can be ignored. But meanwhile, somebody else in the entry for Proto-Germanic *þullaz tears down Kroonen’s and Pokorny’s extra-Germanic cognates as “prone to coincidence”! There’s nobody cross-checking these things.
In short, as you said, there’s still dispute and OED’s “uncertain” from 1912 is still true.
is there a way to tell whether it never existed, or was deleted?
Clicking on the redlink will start to create an article with that name. If one was previously deleted a message advises of this fact. For example, clicking googledork will warn you that it has been deleted no fewer than three times already.
@LH, sad to hear.
Of course IFF the money are invested in improvement of the dictionary (unlike journals which do not fund the research they publish and companies with strange names that are fighting with Asmahan’s videos on youtube) I can’t object.
@ktschwarz, thank you! I was not familiar witht heir notation and I’m surprised.
I’m used to the situation with words from Orinental languages, where the date of the first attestation in the Orient often says nothing about the word’s appearance in the Occident (and no one cares anyway).
1899 is quite late. Then in this meaning it is as English as amenokal (not to be confused with ամենակալ) is Irish. Also compare situations when their earliest quotation is not the earliest – I just did not know whether this means they were not able to find anything earlier or they may not mention the earliest use in a meaning.
@J1M I guess apohedron (tetrahedron, octahedron…) will be апоэдр in Russian.
(actually, -hedron in Russin becomes unstressed -əd(ə)r (or maybe a Russian would write -əd(ɨ)r) and it’s natural to treat it ass a suffix. tetrá-ədr, octá-ədr, pizdá-ədr…)
the supposed PIE root is a red link […], so I think that one can be ignored.
That’s an unwarranted inference. A red link merely means that no one has undertaken to write up the root in question, not necessarily that the etymology is disputed.
Oh, fascinating.
In German it’s uniformly -eder as well, but stressed, and the hiatus is kept: Tetraeder, Oktaeder, Dodekaeder, Ikosaeder…
John C: “A red link merely means that no one has undertaken to write up the root in question, not necessarily that the etymology is disputed.”
Really, there are legitimate PIE roots that at the end of 2023, still nobody has ever undertaken to write up? For example?
In this particular case, there are no references to *tūl-, *twel- ‘bush; sphere’ anywhere else on Wiktionary, or anywhere else that Google can find besides scrapes of this Wiktionary page. No reason to think this is more than a daydream of one self-described amateur. (Here’s the insertion.) By contrast, the comment at the Proto-Germanic page is at least engaging with the existing literature.
mollymooly, thanks!
Is there any *tewH-?
*tewh₂- is the Wiktionary entry for the verb ‘to swell’ mentioned above; it’s *teuh₂‑ in Watkins’s notation, or *teuhₐ in Mallory/Adams (hₐ is their notation for “indistinguishably h₂ or h₄”; they have an h₄, Watkins doesn’t).
I count 15 different routes in Watkins’s AHD appendix from English to forms of this root:
English < oldest documented ancestor < PIE form
thigh < Old English thēoh < *teuk‑
thousand < Old English thūsend < *tūs‑
thole < Old English thol(l) < *tu-l‑ “probably”
tylectomy < Greek tulos < *tu-l‑ “probably”
thumb < Old English thūma < *tūm‑
tumor < Latin tumēre < *tum-ē‑
tumulus < Latin tumulus < *tum-olo‑
tuber < Latin tūber < *tūbh‑
tyrosine < Greek tūros < *tū-ro‑
obturate < Latin obtūrāre < *tūros “possibly”
sorites < Greek sōros < *twō-ro‑
quark < Old Church Slavonic tvarogŭ < *twō-ro‑
soma < Greek sōma < *twō-mn̥
soteriology < Greek saos, sōs < *twə-wo‑
tomb < Greek tumbos < *tu-m-b(h)‑ or *tum‑ “perhaps”
But most of these have been challenged. For example, Wiktionary has a different origin for the Slavic source of quark, whereas Mallory/Adams sees that word and the Greek tūros ‘cheese’ as belonging to a “Slavic-Greek-Iranian isogloss” noun meaning ‘curdled milk’ and not related to the verb.
Of that list of 15, four have OED3 revisions so far, and two of them don’t consider that PIE root even worth mentioning:
obturate: the -tur- component is “of unknown origin”
quark: the Slavic word family is “of uncertain origin”
The other two only mention it as a possibility:
tomb: beyond Greek, “uncertain and disputed”; the ‘swell’ verb is one of three competing theories
tylo‑ comb. form.: the Greek is “perhaps < the same Indo-European base as thole n.1”
So *now* they’re willing to consider it for thole! Go figure.
I didn’t know Watkins went out so far on so many limbs. I had assumed the lists in the AHD appendix were mostly secure, I would’ve guessed 80%, but at least for this root it’s definitely less than that.
(I copy-pasted “Old Church Slavonic” above from AHD, but I know it’s not precisely correct, since German got Quark from West Slavic, not South Slavic. AHD probably meant that that’s the earliest evidence of this family of cognates.)
A good deal of the material from the AHD that ktschwarz quotes above dates from the first edition of the AHD (1969) and was never significantly revised in subsequent editions. It simply repeats what is in Pokorny’s Indogermanisches etymologisches Wörterbuch (1957). Compare the entry for the root tēu-, tǝu-, teu̯ǝ-, tu̯ō-, tū̆- in Pokorny here.
It looks like quark was added in the 3rd edition, but that Slavic etymology is in Pokorny (“Zugehörigkeit auch von abg. tvarogъ `lac coagulatum’ als einer Hochstufenform ist gut möglich”).
I didn’t know Watkins went out so far on so many limbs.
Oh, he definitely did. I love the appendix — how could you not? — but we at Yale, studying with Cowgill, were very aware of how cavalier the Harvard school was when it came to reconstructions.
Oh, nice, that looks like an exact cognate… if the y is long. Is it? The dictionaries over at Perseus don’t explicitly say it is…
if the y is long.
Wait a second – maybe I missed a step here, but how do we get from PIE /u:/ (/uH/) to English “o”? If the PIE was short /u/, there would be no problem.
Iваненко, languagehat: I knew it was based on Pokorny but I naively assumed that Watkins must have done some winnowing and vetting of Pokorny, if not by 1969, then at least by 2000 or 2011. Thanks for the reality check.
τύλος has an acute, so the υ is short.
By Dybo’s law: *-VHRV́- > Germanic and Italo-Celtic *-VRV-. (That’s assuming an adjective; *-ló- seems to have formed some sort of participle, and its stress would explain the zero-grade of the root. The Greek word is a noun, and nouns were often formed from adjectives by letting the stress default to initial.) From Proto-Germanic *u…a we get Proto-Northwest-Germanic *o…a.
On the quantity of the upsilon in Greek τύλος, Wiktionary’s entry for the alternate form τύλη says: “According to Liddell & Scott, the υ is short in earlier authors and long in later ones” (there’s a link there to the entry at Perseus). It also records two other proposed Indo-European cognate groups besides the Germanic group with thole, or “it may rather be of Pre-Greek origin”.
τύλος is one of the proposed cognates of thole given by Kroonen, the other is Lithuanian tulìs ‘nail on a car, plug, nail of an axle’. Kroonen’s Proto-Germanic for this has a short u, but he (unlike Watkins) does *not* go farther, he does not connect it with the PIE verb *teuh₂‑.
When Watkins/Pokorny connected this with the verb they must have thought the u was shortened for whatever reason, since Watkins writes *tu-l‑ with no macron, whereas he does have macrons in most of the other forms, such as the ancestor of thumb, the thick or swollen finger.
When Watkins/Pokorny connected this with the verb they must have thought the u was shortened for whatever reason
At least for Pokorny, no such assumption is necessary; his generation was still fine with an ablaut PIE /*u/ – /*u:/ (see the *tū̆- in the heading of the entry). Watkins is another matter.
@DM: Thanks, I keep forgetting about Dybo’s law, which won acceptance only after I learnt my sound laws. But even if the Greek “y” wasn’t long, the two words might still be related, just not through *tewH.
J.W. Brewer:
Well, that should be easy, it’s… checking the OED Terminology page… WTF, they define “obsolete” but not “disused”! And this isn’t leftover old terminology, they’re still using it in recently revised entries, e.g. acopic ‘That serves as a remedy for fatigue’, revised 2009, now labeled “rare. Now disused” (no label in the original 1884 entry).
Furthermore, you have pointed out a lacuna in the OED that neither Charlotte Brewer (Examining the OED) nor David-Antoine Williams (The Life of Words) has pointed out, which is *really* saying something. You should be admitted to the Charlotte-Brewer-level circle of close OED readers.
I would’ve guessed that “disused” meant no longer actively used by anyone, but still passively understood by some (as a disused factory is still standing and recognizable, though not functioning), and Google found some people who seem to feel similarly at alt.usage.english:
But is that what the OED means by the label? No way of knowing! I couldn’t find any support for that interpretation in any other dictionaries, either; the OED itself gives “obsolete” as part of its definition of “disused”. Does any other dictionary use “disused” as a label?
You can get an instant list of all senses labeled “disused” in the online OED: 1290 senses in 859 entries. Most of them seem to be scientific, technical, or medical, but some are slang: e.g., “R.A.F. slang. to play pussy: to fly under cover in order to avoid detection by another aircraft, etc. Now disused.” That’s from World War II, too recent for their “obsolete” cutoff, which was stated to be, I think, 1929 according to a blog post that I can’t find anymore, probably trashed in the July 2023 redesign *grrrrr*. So possibly “disused” means something like “alive too recently for the obsolete label, but we’re quite sure it’s extinct in the wild”? Stop making us guess! They’ve been making us guess for 140 years, but that’s no excuse!
Another possibility is something like “female hysteria”: I do use the term… when discussing misconceptions. ” Similarly if people are familiar with “to play pussy”, they may use it… in a novel.
There is a label for words that are still in use but only when describing the past, it’s historical. Annoyingly, the OED left that out of the terminology page as well — and they didn’t include this specific lexicographic sense in the entry for historical, either — but I know that’s what the label means because I’ve read about it, probably in the now-deleted blog post.
It is not the OED’s job to imagine how people could possibly use words, it is their job to describe how they documentably *do* use them. If they could find the aviation sense of “play pussy” in recent novels or history books about World War II, then it would be labeled historical, not disused. In fact they have over 200 words labeled both slang and historical, e.g. in Patrick O’Brian novels.
Yes, that’s what I call them.
(and yes, it is not their job but I’m not them, so I said “if”.
Having this said, I am not sure that “I could use it in a novel” should not affect dictionary definitions. Slightly different “we all understand it” clearly is not established based on analysis of a corpus.).
I no longer have access but IIRC “squarial”, which got into the Oxford Additions series just as it became technologically extinct, used to have a bespoke OED3 annotation like “a temporary coinage”.
https://en.wikipedia.org/wiki/Squarial
https://www.oed.com/dictionary/squarial_n
To me, disused seems to imply that a word is still part of someone’s vocabulary, but there is effectively no occasion to use it. I’m not sure whether it’s at all helpful, but I draw an analogy to what the Adept T’Lar said to Sarek in the last scene of Star Trek III, after Sarek had requested fal-tor-pan, the refusion, for Spock:
The Adept knows what the ritual entails, and she believes (correctly) that she knows how to do it. But it has simply never come up before.
Yup, “squarial” has been ridiculed here before, and yes, the label “temporary” is unique as best I can tell.
I had to dig through the Internet Archive to find the deleted blog post on date labeling, which said:
(That’s much, much later than most dictionaries put the boundary; Merriam-Webster and American Heritage put it at “not used since Johnson”, which was 1755.)
Anyway, they’ve never been rigorous or consistent with that boundary. There are lots of words with last uses before 1929 that are not marked obsolete (even after revision). There are also hundreds with much later uses that are, by mistake — one way this can happen is when obsolete words are re-invented and they add the new sense but forget to remove the “obsolete” label from the entry as a whole, e.g. baggie, formerly “Scottish diminutive of bag, i.e. stomach”, later “small sealable plastic bag”.
Then there are the authors who trawl the dictionary for obs. words to revive
As seen here a decade ago.
I previously noted that the OED originally marked ettin obsolete (and has never updated the entry), even though the word has actually been in continuous use since at least Old English—and it would be no surprise if it went continuously back to Proto-Germanic with little or no change in meaning.
I’m disappointed every time when I try to use ettin in the New York Times Spelling Bee.
I’m reasonably sure that “1929” is an error, and perhaps the reason the blog post was deleted. The “OED terminology” metapage says s.v. obsolete: “If an entry, meaning, or lemma is no longer in use in the English language, it may be considered obsolete. This usually means that no evidence for the term can be found in modern English.” Looking s.v. Modern English gives start dates of 1450 (including Early Modern English) and 1700 (excluding it). So I’m guessing that for “1929” read “1699”.
1929 is surprising, but don’t assume, check the entries. If you read the archived page and look up the examples it gives, you’ll find, for example:
raiker ‘a vagabond’, last use: 1866
sammyfoozle ‘to cheat’, last use: 1837
teeny ‘irritable, peevish’, last use: 1847
trade wind ‘A wind that blows steadily in the same direction for a long period’, last use: 1869 (Later, non-obsolete uses are specialized to ‘A wind blowing steadily towards the equator in tropical and subtropical regions’)
wasteheart ‘woe is me! (Yorkshire, Scottish)’, last use: a1912
And the page also says that “‘now rare’ is used if the word was quite common in the past but infrequent since 1930” — unlikely that there are two independent typos that agree like that. So look at some other entries. There are hundreds of entries labeled “Obsolete” with citations in the range 1925-1929, for example:
amiable, adj. 1. † Worthy of being loved, lovable; lovely. 1.b. Of a thing. Obsolete. c1384–1925
† Bonzoid, adj. Obsolete. Exhibiting characteristics of the comic strip character Bonzo… Apparently an isolated use. 1928.
locust tree, n. 6. † New Zealand. Any of several yellow-flowered leguminous trees of the genus Sophora … Obsolete. 1867–1927
† miff, n. Obsolete (colloquial and regional). A fit of pique, a huff … 1623–1926
† Monday pop, n. Obsolete. A concert of popular music held on a Monday. 1862–1927
You can also check the December 2021 update notes, which remark on an “obsolete” sense of the word dipstick for “a person, especially an excise officer, who checked the capacity of casks or barrels using a gauging rod” (yes! newly discovered cutthroat compound!), recorded 1804–1884.
So, yes, OED really does mean to put the boundary that late. Why 1929? Because the First Edition closed in 1928? Or that’s their idea of “beyond living memory”? Wouldn’t it be nice if they deigned to tell us mere customers things like that.
The terminology page is sloppy in using “modern English”, since that’s easily confused with “Modern English”, but if they had meant that, it would have been capitalized — I can’t believe they’d forget to capitalize it if they meant the historical period. They should have said “current English” or “contemporary English”.
Yeesh. Concedo. At this rate they should perhaps say that all entries are obsolete unless explicitly marked Current; it would save space.
Would be simpler not to mark anything obsolete; that way they wouldn’t have to change the label when they find words still alive somewhere they hadn’t looked before, like Jamaica. The count is at 170,780 meanings marked obsolete to 708,820 that aren’t, and I bet that gap is not going to get any smaller.