The word stupa has long been familiar to me; it’s (to quote the OED, entry updated June 2019) “A structure serving as a Buddhist monument, shrine, or mausoleum; esp. a domed or bell-shaped structure topped with a spire.” The name is “< Sanskrit stūpa, specific use of stūpa, stupa lock of hair, top of the head, further etymology uncertain.” But I didn’t know the equivalent Russian word субурган [suburgan], which I just had to look up; the Russian Wikipedia article has this useful paragraph on terminology:
The name “stupa” (Sanskrit for ‘top, peak’) is characteristic only of India and Nepal; in Sri Lanka the name dagoba is used, in Myanmar zedi and pato, in Thailand chedi and prann, in Laos that, in Tibet and Bhutan chorten, and in Mongolia suvarga; in Russia (Buryatia, Kalmykia, Tuva, Altai) it is suburgan, in China and Vietnam bao ta, pagoda, etc.
Название “ступа” (санскр. – “макушка”) характерно только для Индии и Непала; в Шри-Ланке применяется название дагоба, в Мьянме — зеди и пато, в Таиланде — чеди и пранн, в Лаосе — тхат, в Тибете и Бутане — чортен, в Монголии — суварга; в России (Бурятия, Калмыкия, Тыва, Алтай) — субурган, в Китае и Вьетнаме — бао та, пагода и т. п.
English Wikipedia adds “The Asian words for pagoda (tā in Chinese, t’ap in Korean, tháp in Vietnamese, tō in Japanese) are all thought to derive from the Pali word for stupa, thupa.” In the “Translations of Stupa” box in the right margin they have all the ones mentioned in the Russian article except for suburgan, which is obviously of the same origin as Mongolian suvarga, whatever that is. They also don’t explain the origins of chedi/zedi, prann, and chorten; for the last-named, M-W says “Tibetan mchod rten, literally, offering holder.” As always, your thoughts are welcome.
I know “prang” as a distinctive Cambodian-style type of spire… Wikipedia says
The term prang is a compound of the Sanskrit terms pra- (‘forward, in front’) and aṅga (limb of the body), with the contacting vowels united by sandhi.[1]
I also just read the very interesting if necessarily sometimes repetitive “Strange Tales of an Oriental Idol – An Anthology of Early European Portrayals of the Buddha” edited by Donald S. Lopez Jr., and it was fascinating to see the terms & facts get distorted as they filtered back to Europe. So much came via Sri Lanka that i think “dagoba” predominated in early accounts, not stupa.
One of the strangest threads of this anthology was that the Jesuits were extraordinarily interested in Buddha’s evil cousin Devadatta for generations – because of the *extremely* minor detail that he was said to have been crucified upside-down in one of the Buddhist Hells for his misdeeds.
For some reason I’m reminded of an Angry-Young-Man rock song from my teens in which the outro (starting around 2:50) has a repeated chant of “stupa stupa stupa.”* Or so I’ve always thought, although some online sources claim it’s “stupid stupid stupid.” https://www.youtube.com/watch?v=3KOLoWvHImw
*To be fair, a chant of “stupə stupə stupə,” which I guess could be spelled a variety of ways.
That’s an… odd… interpretation. The song is called “Stupefaction,” so one might think it’s just the first two syllables on repeat; the “stupid” version is supported by the line “I ask the neighbour why are you so stupid?” But “stupa” is right out.
Graham Parker and The Rumour were faves of mine back in the day; I worshiped Howlin’ Wind and Heat Treatment, and of course I bought Squeezing Out Sparks, but then I moved on and never wound up buying The Up Escalator, so I’m not familiar with that song. It seems to lack the utter ferocity and unstoppable rhythmic brio of his earlier work.
Well, how would you spell the first two syllables of “stupefaction” if the word is clipped back to just those two? It can’t be “stupe” because that leads the reader to assume a “silent e.” So it’s gotta be “stupa” or maybe “stupuh,” innit? Something homophonous with the AmEng pronunciation of “stupa” in the Buddhist-architecture sense even if semantically unconnected.
I lasted one album longer with Graham than you did, apparently. I saw him at Toad’s Place in fall ’83 when he was already two albums past The Up Escalator but didn’t feel inspired to go back and buy either of those.
Well, how would you spell the first two syllables of “stupefaction” if the word is clipped back to just those two?
Oh, OK, that’s fine then — I thought you meant he was singing the word “stupa” rather than using it as a representation of the pronunciation.
Well, tepe also (Turkic, etc.; absent from OED, and Wikipedia articles linked in LH’s post). I had it all through a book MS I edited recently. Ubiquitous in archaeology from the Middle East to the fringes of China, and roughly equivalent to similarly frequent tell (various spellings; Semitic, etc.). Possible connections with toupet, etc., etc.
But is a tepe a stupa? I thought it was just any artificial mound.
A tepe is not a stupa; and in its origins tepe is apparently just a hill, not an artificial one. The words are connected in etymology, which you included in your post. I thought from my mention of toupet it would be clear that I too was addressing etymologies.
Sorry, my brain is fuzzy today. I blame Covid Penumbra.
I don’t think I ever consciously realized (though of course in retrospect it is obvious) that the Buddhist stupa was unrelated to the normal Russian word ступа “mortar” (as in “and pestle”), these days probably best known as the vehicle that Baba Yaga flies in when she’s not riding a broom.
(I see that English Wiktionary just bundles both meanings into one entry, with an etymology for the “mortar” word that almost certainly does not apply to the “temple” word.)
As far as I can recall I’ve never seen субурган before. If I had I must have forgotten it.
EDIT: I wondered if there’s some s-mobile-esque etymology that could relate stupa to English top (Proto-Germanic *tuppaz, which apparently also meant “lock of hair”), but I don’t think the consonants allow it. Would have been so neat though…
JWB: Well, how would you spell the first two syllables of “stupefaction” if the word is clipped back to just those two? It can’t be “stupe” because that leads the reader to assume a “silent e.” So it’s gotta be “stupa” or maybe “stupuh,” innit? Something homophonous with the AmEng pronunciation of “stupa” in the Buddhist-architecture sense even if semantically unconnected.
Stupor (non-rhotically)?
January:
I wondered if there’s some s-mobile-esque etymology that could relate stupa to English top
Sure, all over the place. The OED etymology for toupet, to cite a favourite example, gives the lead:
“French toupet /tupɛ/ tuft of hair, especially over the forehead, derivative (in form diminutive) of Old French toup , top, tup , tuft of hair, foliage, etc.; < Low German topp- = Old High German zopf top, tuft, summit; compare Old Frisian top tuft, top, Old Norse toppr top, tuft, lock of hair: see top n.1 and adj."
But what an "s-mobile-esque etymology" might be, I cannot say.
Gauthiot offered an Iranian etymology (*spur-xān, ‘house of perfection’, transmitted to Mongol via Old Uyghur) for Mongol suburγan, as reported in Chavannes and Pelliot (1912) Un traité manichéen retrouvé en Chine, page 132 (=108), available here.
@Trond: I could be wrong about this on account of not being non-rhotic, but I think non-rhotic “stupor” would have a perceptibly longer vowel (and maybe a slightly different one?) in the second syllable than “stupə” does.
Gauthiot offered an Iranian etymology (*spur-xān, ‘house of perfection’, transmitted to Mongol via Old Uyghur) for Mongol suburγan
I like that!
But I didn’t know the equivalent Russian word субурган [suburgan], which I just had to look up
Are you reading Pelevin, by any chance?
LH readers who are curious about the etymology of Thai เจดีย์ chedi (from Pali cetiya, from Sanskrit caitya-) can easily trace it by going to the Wiktionary entry for the Thai word and clicking through the links in the etymologies.
some s-mobile-esque etymology that could relate stupa to English top
tepe seems to fit much better:
Turkish
Etymology
From Ottoman Turkish تپه (tepe, “crown of the head, apex, summit, mountain peak, hill”), from Proto-Turkic *tepü, *töpü (“hill, top; top of head”). Cognate with Old Turkic 𐱅𐰇𐰯𐰇 (töpü, “height”), Old Uyghur [script needed] (töpü, “top of head”), Karakhanid [script needed] (töpü, “summit of a mountain, hill, crown of a head”), Yakut төбө (töbö, “top, summit”).
Pronunciation
IPA(key): /teˈpɛ/
Hyphenation: te‧pe
Noun
tepe (definite accusative tepeyi, plural tepeler)
1. hill
2. top, peak
3. apex, vertex
4. crest, crown
https://en.wiktionary.org/wiki/tepe#Turkish
Your stupa needs a torana.
Now that the subject of South Asia has come up, I guess it’s the right place for:
Sensitive Reading
The Pleasures of South Asian Literature in Translation
https://www.luminosoa.org/site/books/e/10.1525/luminos.114/
Are you reading Pelevin, by any chance?
No (although I’m reading Furmanov’s Чапаев in preparation for reading Pelevin’s Чапаев и Пустота), I found that in «Страна происхождения» by Dmitry Bakin (a weird and wonderful writer):
@J.W. Brewer: For this non-rhotic speaker, it’s just a schwa in both, and as far as I know that’s true generally. I think it tends to break rhotic speakers’ minds in the way systemic accent differences often do*, but in non-rhotic accents there really is no difference between syllables that would be [ə] and [ər] in a rhotic accent.
*I still really have to work to remember that the majority of the world’s English speakers aren’t just faking the foot-strut split to sound posh.
My idiolect is rhotic, but I make a distinction between “stupor” in the everyday lay sense, where I have schwa in the second syllable, and “stupor” as a medical technical term, where I have a full vowel. I imagine some non-rhotics make a similar distinction (or have even generalised the latter at the expense of the normal lay pronunciation.)
Survives as Zopf “braid” (as opposed to “pigtail” for example, but extended to braided baked goods).
They do if we can let a PIE *-nó- drop from the heavens, and we need that anyway to account for the *-pp-.
I had to have it explained to me that in English non-rhotic accents* there’s really no difference between /ə/ and /(ə)r/, so that seven and Severn are exact homophones – the unstressed vowel can disappear completely in both words, too.
That’s not so in non-rhotic German, where [ɵ ~ ɘ ~ ə ~ ɛ] (in different accents) and [ɐ ~ a] are strictly kept apart, even in the weird accent that has shifted the former to [ɐ] and the latter all the way to [aː], and the former can disappear but the second never does.
* I don’t know if that’s true for the American ones, actually.
seven and Severn are exact homophones
I don’t think that’s right. The final syllable of “Severn” has the same vowel as the final syllable of “nocturne”, even for non-rhotic RP speakers, at least in slow speech.
Stupendous thread.
Seven and Severn would be homophones for me (and I did spend 5 years of schooling next to the Severn as a member of the Severn Hill boarding house at Shrewsbury School…
An unstressed unreduced vowel?
The person who told me is Justin B. Rye, who describes his near-RP accent at some length here. The Severn is not on the list of examples, but beta – beater, tuna – tuner and uvula – uvular are presented as pairs of homophones.
(There are snarky answers to the usual American shibboleth questions at the bottom. “What do you call it when rain falls while the sun is shining?” In short, “summer”.)
Harrumph. If furriners didn’t insist on pronouncing “beta” funny, they’d think it was homophonous with non-rhotic “baiter.”
DE, I’m curious — which full vowel do you have in (medical) “stupor”? Is it the same as in “Severn” and “nocturne”?
I have the same vowel distinction between senses of stupor as David Eddyshaw, although I don’t actually use the medical sense very often. There is a parallel difference between the generalized and more precise biological senses of the similar word torpor, although the adjective torpid is pronounced the same for either meaning.
https://en.wikipedia.org/wiki/Ice_Stupa
http://icestupa.org/
maybe this is a little offtopic but it may be seasonable…
@TR:
No, it’s different: /stjupɔɾ/ (or thereabouts.)
“Stew: pour.”
Both CEPD and LPD treat Severn and seven as homophones.
Och, well. I’m not actually a RP speaker. I just play one on TV.
Proto-Germanic *tuppaz
Hmm. Besides *tuppaz there is also a *þūbaz ‘mound’. Both have been proposed as sources of Finnic *tuppas ~ *tubas ‘mound, tuft (grass, hair, etc.)’ (semantics not distributed over the phonetic variants) but I wonder if maybe separating them is an unnecessary choice and *tuppaz is instead a back-loan from Finnic. Also PII *stuHpas would probably work just fine as a source of the Finnic variants, with *Hp substituted as either *pp or *p.
There is also Mari /tupka/ ‘tuft of hair’ and Komi /tup-jura/ ‘tufty’ (“tup-headed”); in olden times these were compared with e.g. Finnic *tukka ‘hair’ in a proposal for a PFU *-pk-, but more likely the Mari word is a suffixed reflex of similarly *tuppV- or *tupV-. The Komi word though would require either earlier *toppa, *tappV or late loan origin.
I don’t have a distinction for different senses of stupor (due to insufficient interaction with medical professionals?) Seven and Severn are homophones for me, even in careful speech, and I’ve lived close enough to the river most of my life for it to be a familiar word. I’ve known people from south-western England whose accents don’t have non-prevocalic rs but who I’m pretty sure would pronounce them differently – presumably their accents are rhotic at some level, but the rs don’t make it to the surface.
Is the schwaless medical version Stew-Pour or Stew-Purr? (I hope no one out there has a pour/purr merger …)
Finnic as a source for long consonants in Germanic should definitely be explored!
Hmm. Besides *tuppaz there is also a *þūbaz ‘mound’. Both have been proposed as sources of Finnic *tuppas ~ *tubas ‘mound, tuft (grass, hair, etc.)’
I immediately looked up the etymology of tuft; Wiktionary attributes it to *þūbaz via a Germanic loan in late Vulgar Latin, while Etymonline suggests either the Latin root given in Wiktionary (with no further etymology) or a later Germanic loan in Old French from a descendant of *tuppaz.
As I mentioned, the original question was whether the Germanic and Sanskrit roots (both of which apparently have no further known etymology) are connectable (are they? I don’t really understand enough about where the relevant consonants come from to tell one way or another), but the Finnic connection is also intriguing!
My Mongolian etymological dictionary gives:
Суварга < Uyghur supurɣan < Persian spurχān.
Анхандаа дээдэсийн шарил дээр босхоцон дурасхалын хөшөөг нэрлэж байгаад хожим давхарлаг маягаар өндөрлөн зассан шовтор оройтой дурасхалын хөшөөг нэрлэх болжээ. (Very roughly: Initially a memorial erected on the grave of ancestors, later a monument with a layered conical top.)
Cf. suburǧan ‘a tomb’ (Clauson 792), Manchu subarǧan ‘обелиск’.
Kha, Bur. suwarǧa, Ord. suwurǧa, Oir. suwvrǧan (The ‘v’ in the last word is supposed to be a superscript.)
Thanks!
The usual Japanese word for ‘stupa’ is 仏塔 buttō, ‘Buddha tower’.
The word 卒塔婆 sotōba or sotoba is from a Chinese transliteration of the Sanskrit word stūpa. It usually refers to (vaguely) pagoda-shaped wooden slats bearing lines from sutras, etc. that are placed behind the grave of the departed at the time of family grave visits and have as their purpose mourning the dead, praying for their souls, and praying for their Buddhahood.
See the Wikipedia article at 板塔婆 ita-tōba, which is an alternative name for them.
The word 塔 tō in Japanese is now used for any kind of tower-like structure. Tokyo Tower is, for instance, known as 東京塔 Tōkyō-tō. I’m wondering whether this expansion of usage is a result of the resemblance of 塔 tō to the English word ‘tower’. (‘The Watchtower’, which you may be familiar with from proselytising missionaries, is ものみの塔 monomi no tō.)
Incidentally, суварга in Mongolian is pronounced /sovraɣ/.
Bathrobe!
I would very much like to know the author and title of the dictionary from which you quote that etymology (Суварга < Uyghur supurɣan < Persian spurχān… Анхандаа дээдэсийн шарил дээр босхоцон дурасхалын хөшөөг…), for my own improvement. Could you tell us?
An Etymological Dictionary of Mongolian Language
by Tumurtogoo Domii
Mongolian Academy of Sciences Institute of Language and Literature
Ulaanbaatar 2018
ISBN 978-99978-892-9-4
МОНГОЛ ХЭЛНИЙ ҮГИЙН ГАРАЛЫН ТАЙЛБАР ТОЛЬ
Домийн Төмөртогоо
Монгол улс шинжлэх ухааны акадэми хэл зохиолын хүрээлэн
It was printed by Admon. I assume the publisher is the Mongolian Academy of Sciences Institute of Language and Literature.
The body of the dictionary goes from page 9 to page 377.
Tumurtogoo is a famous Mongolian linguist responsible for a huge output over the course of his life. He is still working at the Mongolian National University, although a stroke he had some years ago has apparently slowed him down.
You can see a copy here.
Thanks for all the bibliographic information, Bathrobe! I will try to have my library get this book.
I will also note here that Henning (1945) “Two Central Asian Words” Transactions of the Philological Society 44 ( https://doi.org/10.1111/j.1467-968X.1945.tb00214.x ) offered an alternative etymology. He associated Persian marzaġān, marġazan and Armenian գերեզման gerezman (a Iranian loanword in Armenian) and proposed an unattested Sogdian *zmrγʾn as the source of Old Uyghur suburğa:n “tomb” (supurğa:n?), subsequently transmitted to Mongolian:
Henning offers a very tentative suggestion that the sequence zmr preserves the original order, and can be associated to the zǝmar- “ground” in Avestan zǝmargūz, zamargūz “hidden underground, creeping away underground” (used of demons; root gaoz- “to hide”).
That sounds exceedingly speculative.
I wonder that Henning did not directly attribute the metatheses to taboo deformation, especially in the context of Zoroastrian disapproval of interment (an aversion he mentions in passing), as oppposed to exposure in a dakhma.
Definitely possible, and definitely so hard to test as to be close to useless. If we’re already speculating, how about a confusion of similar words instead of unlimited metathesis?
I’m glad you’ve weighed in on this. I’m no expert on either etymology or historical linguistics so I was happy to throw in what I had to hand. Good to see that it’s started a deeper discussion.
The connection with dakhma and Zoroastrianism is interesting. I assume there is no connection at all with Sinhala දාගැබ් dagab (dagoba).
For Japanese, English Wikipedia blithely gives 卒塔婆 as the translation of ‘stupa’ whereas it is actually derived from stūpa (via Chinese) but is now primarily used in a different sense. The dangers of relying on Wikipedia….
The connection with dakhma and Zoroastrianism is interesting. I assume there is no connection at all with Sinhala දාගැබ් dagab (dagoba).
Hobson-Jobson gives the etymology for dagoba in the entry for the word: from Pali dhātugabbha and Sanskrit dhātugarbhaḥ, literally ‘having an interior filled with relics; relic-receptacle’, a compound of dhātu- ‘constituent part, element, ashes of a cremated body, relics’ and garbha- ‘womb, interior, middle’ (also ‘inner chamber, temple sanctuary’).
As for the etymology of Avestan daxma- ‘tomb’ (Parthian dahmag ‘tomb’, Middle Persian daxmag ‘tomb, dakhma’, Persian دخمه dakhma ‘dakhma’), Karl Hoffmann proposed that Avestan daxma- was dissimilated from earlier *dafma-, from an Indo-Iranian *dhabhma-, from a virtual Indo-European *dhm̥bh-mo-, with a *dhm̥bh- also seen in Greek τάφος ‘tomb’ and Armenian damban, dambaran ‘tomb, grave’, perhaps with a further cognate in Old Prussian dambo, ‘ground’. Hoffmann’s proposal is here. (It’s a single-paragraph, one-page article.)
Wow, a Least Publishable Unit from 1965!
Is that impressively early or impressively late?
Impressively early.
…but of course, first, historical linguists have never felt “publish in a journal with an impact factor or perish” as hard as natural scientists; and, second, this paper may be too small to get into a prestigious journal, i.e. it may actually be smaller than a LPU.
second, this paper may be too small to get into a prestigious journal
That article was published in the Zeitschrift für Vergleichende Sprachforschung; that’s as prestigious as you could get in IE linguistics at that time.
No kidding. Getting published in Kuhns Zeitschrift seemed like the height of professional achievement when I was in grad school.
Also, Karl Hoffmann could probably have gotten published in whatever journal he wanted.
Ah. In my field the truly tiny papers (“I found an entirely unremarkable rib that probably belongs to a well-known species”) appear in that author’s in-house journal that he edits himself.
I don’t know how common this is for ZfvS specifically, but I’m familiar with one-page-or-less etymology articles especially from ca. 1850–1950. The Finnish linguistics mainstay Virittäjä or the Hungarian Magyar Nyelvőr have some of these pretty much every issue from the times. After that the easy pickings start running low enough that it mostly takes at least a page and a half to present a new finding or argue for a new alternative.
The theoretically least publishable unit in etymology is probably about one sentence long and, in olden times, when well-known data sources were allowed to be left uncited, could even have been formatted as just a title without any following paragraphs (“Swampy Foo abcde < Proto-Foo *awtʃəðə > Lumpy Foo uwxyz“). Though people who dig up things this simple are usually finding them from understudied languages and will likely present them instead in batches of anywhere between handful to a few hundreds… or in footnotes to a paper treating something else.
I don’t know how common this is for ZfvS specifically
Not uncommon, as I remember. The very first issue has at least two one-paragraph articles, on p. 288 and 512 (the latter, it’s true, is a long paragraph that continues onto the next page).
Note that the first article is Aufrecht on πέος, every budding Indo-Europeanist’s favorite s-stem.
will likely present them instead in batches of anywhere between handful to a few hundreds…
In the German tradition, that kind of article is called Miszellen, so in KZ or IF you’d find articles called, say, Iranische Miszellen, where each section would discuss separate etyma or groups of etyma.
Yeah, there’s a bunch of those in that first issue as well.
Reporting in that the idea this thread had given me is now out as an article (with also a due footnote nod to the other discussants here): “Uralic *tuppas — bridging Indic and Germanic“.
Excellent, thanks for letting us know! GT version of the abstract:
GT shits a few bricks but we get the general idea.
Very good!
A couple of speculations mentioned through the paper:
– Stupa may be borrowed from a substrate language.
– Stupa and Germanic *þuBaz ~ *þuBon could be related if we suppose s-mobile.
Uniting these, we might speculate that the word was borrowed from the (Globular Amphora?) substrate in Corded Ware. *tupa moved west and north and became Grimm-shifted, while stupa took the long route through Russia, was brought into the multi-ethnic system of the Southern Urals, entered Uralic, lost its s- and was brought back through Russia and into the Baltic Bronze Age, where it was geminated before eventually being picked up by Germanic.