Mongolian priests and bugs, with a note on the Japanese word for "bonze"
« previous post | next post »
An anonymous correspondent asked:
Are these actually related words, or just homonyms?
p. 127 of Jack Weatherford, Genghis Khan and the Quest for God: How the World's Greatest Conqueror Gave Us Religious Freedom:
Male shamans were treated with cautious respect, but they evoked suspicion and even disgust. As one saying put it, “the worst of men become shamans.” The word boo, Mongolian for “shaman,” is part of a cluster of words with loathsome connotations: foul, abominable, to vomit, to castrate, an opportunistic person without scruples; it is also the general term for lice, fleas, and bedbugs. 28
His footnote 28: бѳѳ (бѳѳδийн), to vomit (бѳѳлжих), to castrate (бѳѳрлѳх), an opportunistic person without scruples (бѳѳрѳний хн), and the basic term for lice, fleas, and bedbugs (бѳѳс). Хvлгийг муу жоро болох. A Modern Mongolian-English Dictionary, ed. Denis Sinor (Indiana University, Uralic and Altaic Series, vol. 150, 1997),
Someone else asked whether Japanese boosan / bōsan 坊さん ("monk") were somehow related.
I have an innate distrust of claims for sound symbolism. See:
"Phonosymbolism and Phonosemantics in Chinese" (1/13/12)
Facetiously, Jichang Lulu remarked:
The English word “shaman” is part of a cluster of words with loathsome connotations: shat, sham, shambles, shatter, shanty, shabby, shallow, shackles, shack, shaggy…
Other Mongolian words in бөө- böö- (traditional script böge- ᠪᠥᠭᠡ-): бөөм bööm ᠪᠥᠭᠡᠮ bögem (also böön) ‘bunch, cluster, ball lump’, with its many derivatives: бөөр böör ᠪᠥᠭᠡᠷᠡ böger_e ’kidney; testicle; affectionate word for a child’. The latter also has its own set of derivatives, including Weatherford's бөөрлөх böörlöx ᠪᠥᠭᠡᠷᠡᠯᠡᠬᠦ bögerelekü (also böör avax) ’castrate’, as well as бөөрөнхий böörönxii ᠪᠥᠭᠡᠷᠡᠨᠬᠡᠢ bögerenkei ‘round; vague, indecisive; opportunistic’, of which he conveniently picked that one sense.
‘Opportunistic’ and ‘castrate’ should obviously be removed from W.'s list. This also applies to his first and most crucial examples, ‘foul’ and ‘abominable’, for which he fails to provide any actual Mongolian words. Three genuine separate groups of words are left: бөө böö ᠪᠥᠭᠡ böge ‘shaman’ (and its derivatives); бөөлжих bööljix ᠪᠥᠭᠡᠯᠵᠢᠬᠦ bögeljikü ‘to vomit’; бөөс böös ᠪᠥᠭᠡᠰᠦ bögesü ‘louse etc.’. I don't know if they are related in any way. The ’vomit’ verb is derived; not sure if there's a connection to the many ‘round’ words, or to ‘shaman’. A Turkic comparandum meaning ‘magic’ has been proposed for Mong. böö ‘shaman’. There's also a suggestion that ‘louse’ could be related to a Turkic word. I can't give an opinion on those.
Lulu added:
The first paragraph was meant as sarcasm. Those words have all sorts of origins and little in common beyond the accident of having come to begin with sha-. Other sha- words aren't particularly loathsome: shampoo, shallop, shall, shank. Shallot is a lovely kind of Allium. ‘Loathsome’ is so vague a descriptor as to make it easy to apply to ‘clusters of words’ sharing a syllable in many languages.
I'm also generally sceptical of indiscriminate appeals to sound symbolism. From memory, a LL post on the topic discussed Korean japta 잡다 ‘grab’.
If I had to propose a cluster of Mongolian böö- (and also bö-) words, I would associate it with the meaning ‘round’.
While I find the böö- ‘loathsomeness’ argument weak, the saying Weatherford cites is genuine and his description of the attitude towards shamans shouldn't be dismissed. The böö- discussion doesn't seem to help his argument, but it does raise the question of whether bööljix ‘to vomit’ is somehow related to ‘shaman’.
Dotno Pount replied:
I’m pretty sure they’re just homonyms. Or not even. The root of “stranger” is böör, which means kidney, and the phrase comes from the idea that people outside one’s kin network do not share the same blood and bones, and therefore do not share internal organs like kidney and liver. And the phrase is incomplete as “böörnii hün.” There should be anothet word, like “hüren” brown, or “har’” foreign. Kidney is also the root of the word for castration here, but I’ve never heard anyone use this word for castration. However, I wouldn’t be surprised if this is a regional euphemism.
The root for flea is “böös”, not “böö,” I think. And it’s not in plural. Unless there’s some historical evidence that one led to the creation of the other….
Also, I don’t know whether the root of “to vomit” “böölj” is “böö” or not. The ending “lj” looks like a derivational suffix. However, there is certainly no connection in terms of meaning in my mind. I can see how a kid would try to use it as a pun.
From Christopher Atwood:
Although there's nothing immediately impossible about it (as there so often is with such popular etymologies) I think it's almost certainly a homonym. What there is is a root of Old Mongolian böge-n which means lump, ball, bunch or mass, and is certainly the root of böge-sün "lice" (semantically appropriate and sUn is a well-known suffix) and probably of böge-lji- "to vomit" (bring up masses of stuff?) and böge-re "kidney" and also "testicles" probably (semantically it works nicely, but -re is not a regularly attested derivational suffix). I would guess, however, that böge or böe "shaman" is probably a borrowing from some other, now extinct, language and only coincidentally looks like this family of genuine Mongolian words.
From Timothy May:
The actual modern word for shaman is böö, not boo, although I suspect umlauts were left off intentionally as I don’t see Weatherford using them anywhere in that book.
Böö is derived from the earlier böge (classical Mongolian) or bögechi.
See Lessing’s dictionary 3rd reprinting (1995), p. 123-124.
To vomit is bögelzi-
To castrate is bögerele-
Lice is bögesü
While all of these have “loathsome connotations”, we also have “bögem” which is “Lump, clot, cluster, ball, bunch; bögemne-to gather together. Neither of these are loathsome, unless one dies from a blood clot.
Bögenchüg is a ball of yarn and böger-e is a kidney or testicle.
I think it might be a stretch to say that böge—or böö is simply a root with “loathsome connotations”, but it is curious. While I hate to generalize, shamans historically have always been on the fringe of society as they do have powers beyond a normal person’s “ken”. Having respect for them is a healthy attitude as you might need them, and you certainly wouldn’t want to be on the bad side of a shaman.
From Bathrobe:
I'm pretty sure they aren't etymologically related and the random sample is suspicious.
* бѳѳлжих does mean vomit.
* бѳѳрлѳх is from the word бѳѳр meaning kidneys (primary meaning) or testicles. It thus means removal of the kidneys or testicles. Unless you regard the kidneys as disgusting I don't think this really supports his thesis.
* бөөрөнхий means 'round'. A secondary meaning is 'neutral, irresolute, vague, unprincipled'.
* бөөс means 'louse'
Other words using бөө include:
* бөөрөг a terrace running alongside a mountain
* бөөрөлжгөнө quince, stone bramble
* бөөм, бөөн heap, clump, crowd
* бөөдий unclean, foul, abominable
* бөөцгөр haggard, shrivelled, cadaverous
* бөөгнөрөх pile up, accumulate, collect
There are several unfavourable words here but also a few that are not particularly unfavourable.
I think there is no doubt that Mongolian does exhibit a certain amount of sound symbolism. For example, бондгор, бөндгөр, бүндгэр, бэндгэр, бундгар are all words meaning 'chubby' representing varying degrees of chubbiness. There is a whole series of words beginning in б that refers to sparrows, larks, and other small birds. But I don't think that this applies to бөө. Just to be sure, I checked with a Mongolian friend who told me that бөө just means 'shaman' and has neither positive nor negative connotations. Linking shaman with flea, vomit, or castrate is 'folk linguistics'. I don't think it holds any water.
For a conclusion, and to put things in context, this comes from Thomas Allsen:
Since I am not a Mongolian philologist I am not sure if this series of words are related or chance homophones. Charles Bawden published a dictionary some years back which, if I remember correctly pays greater attention to questions of etymology. As a matter of opinion, I think Weatherford has greatly confused modern ideas about religious liberty with the diffuse, mix and match religious traditions of North and East Asia.
Footnote on Japanese bōsan 坊さん ("monk")
From Linda Chance:
A short look at the online dictionaries suggests that it comes from 僧坊, quarters (lanes?) for monks (attested as early as 747). 坊主 is perhaps the full form, shortened for familiarity to 坊さん, which through sound change becomes ぼんさん in the Edo period. Or one might think of it as the usual reference to a person via their associated place, since there are no pronouns to perform that function.
From Nathan Hopson:
坊さん is informal for 坊主, bōzu, which gives us the English bonze.
The more formal term is 僧侶 (sōryo) or just 僧 (sō). Monks' quarters are 僧房 (sōbō), or 宿坊 (shukubō).
I had until now assumed that at one level at least 坊さん is then one who inhabits a 宿坊.
Digging a bit deeper, it looks like 房 and 坊 are basically synonymous in Japanese — or, as one dictionary (字通) puts it "同系の語とみてよい."
The explanation 字通 gives is:
方に方形・区画の意がある… 条里によって区画された一画を坊という。坊に坊門を設け、坊長をおいた。寺院の内部も坊に分かたれ、一坊の主を坊主といった。
IOW:
The 方 of 坊 contains the meaning of a precinct. Administrative plots were called 坊 (under the 条理 system), and each was gated and given a chief. Similarly, temple grounds/precincts were divided up and each given a head, called a 坊主.
I assume you get from lane/alley to precinct in a relatively straightforward way: the lanes/alleys are the demarcators, and the workshop, store (at least in non-contemporary Japanese), fields, temple, etc., is what is contained therein.
As for 坊 in Sinitic, we have:
Pronunciation 1: fāng
lane; alley
neighborhood; urban division; ward
store (on a street, in a market, or in a neighborhood)
paifang or other work of architectecture extolling the virtue of someone
a surname
Pronunciation 2: fáng, fāng
workshop, mill
Pronunciation 3: fáng
Alternative form of 防.
dam; embankment
to guard against
—
The definitions "priest's residence; (Buddhist) priest; boy" are considered to be Japanese.
"bonze" — American Heritage Dictionary of English, 5th ed.: French, from Portuguese bonzo, from Japanese bonsō : bon, ordinary (from Early Middle Chinese buam) + sō, monk (from Middle Chinese səng, from Sanskrit saṃghaḥ, sangha; see sangha).
The AH editors must be referring to bonsō / bonzō 凡僧 ("an ordinary, lowly monk"), which is at variance with the opinions of the Japanese specialists (and the dictionaries on which they are relying) cited above.
In Chinese, fán sēng 凡僧 ("an ordinary, lowly monk") may be traced back to Yán shì jiāxùn 顏氏家訓 (The Family Instructions of Master Yan) of 顏之推 (531-591)
martin schwartz said,
May 9, 2018 @ 3:36 pm
It may be relevant to the semantics of the post's mention of the "bug"
word to google:
bug-Wiktionary
but I seriously doubt etymological connection with the words in the post, as also applies to Eng. bug vis-à-vis Arabic baqq 'a bug'.
Martin Schwartz
John Rohsenow said,
May 9, 2018 @ 4:01 pm
"Böö is derived from the earlier böge (classical Mongolian) or bögechi."
==>>>> "bogey-man"? ;-)
Victor Mair said,
May 9, 2018 @ 4:10 pm
https://en.wikipedia.org/wiki/Bogeyman#Etymology
The word bogey is believed to be derived from the Middle English bogge / bugge ("hobgoblin") and is generally thought to be a cognate of the German bögge, böggel-mann (English "Bogeyman"). The word could also be linked to many similar words in other European languages: bogle (Scots), boeman (Dutch), Butzemann (German), busemann (Norwegian), bøhmand / bussemand (Danish), bòcan, púca, pooka or pookha (Irish), pwca, bwga or bwgan (Welsh), puki (Old Norse), pixie or piskie (Cornish), puck (English), mumus (Hungarian), bogu (Slavonic), buka (Russian, бука), bauk (Serbian), bubulis (Latvian), baubas (Lithuanian), bobo (Polish), bubák (Czech), bubák (Slovak), bebok (Silesian), papão (Portuguese), торбалан (Bulgarian), Μπαμπούλας (Greek), bua (Georgian, ბუა), babau (Italian), бабай (Ukrainian), baubau (Romanian), and papu (Catalan).
A related word, bugbear, from bug, meaning goblin or scarecrow, and bear, was imagined as a demon in the form of a bear that eats small children, and was also used to mean a general object of dread. The word bugaboo, with a similar pair of meanings, may have arisen as an alteration of bugbear.
~flow said,
May 9, 2018 @ 4:45 pm
FWIW there's also བོན, sometimes spelled བོཾན (I think), i.e. Bon or Bön; cf. https://en.wikipedia.org/wiki/Bon. Maybe also linked to Sanskrit?
Is there are chance to clear up whether somehow 凡 and/or 梵 got mixed up in Japanese with 房…坊? They all seem to share readings and all pop up around "Buddhism, monks, and their quarters".
David Marjanović said,
May 9, 2018 @ 6:43 pm
Perhaps he did that to avoid the misleading implication of ö: it isn't [œ], Mongolian does not sound like Turkish and never did. öö is [oː], oo is [ɔ̙ː] with retracted tongue root.
Victor Mair said,
May 9, 2018 @ 6:52 pm
From Seishi Karashima:
"bonze" < 坊主 (bauzu ばうず; ぼうず) < 僧坊主 (first appears in Kumārajīva’s translation) [pdf of article on this subject available upon request]
Linda Chance said,
May 9, 2018 @ 7:20 pm
The question put to me was not about the derivation of English bonze, although I too have heard that 坊主 is the source. As for 凡僧 ・ぼんそう, some reference works do seem to favor it as the source for bonze, through Portuguese. Clearing this up would require references I do not have at hand.
SO said,
May 9, 2018 @ 7:25 pm
The etymology for bonze given by the American Heritage Dictionary is wrong, whereas Prof. Karashima is right. It's based on 16th c. Middle Japanee boozu (< bauzu < baũzu) 坊主 which still had a prenasalized /z/ — which is where the n in bonze stems from.
Pickering said,
May 9, 2018 @ 8:45 pm
Off topic, but I first heard the word 'bonze' listening to that very western opera Madama Butterfly. https://www.youtube.com/watch?v=3OWWpWouLI4
In the opera he's a dick, but then so is Pinkerton.
AG said,
May 9, 2018 @ 8:48 pm
How does this all fit with akanbo, Japanese for baby, and boya, which means "boy"?
赤ん坊 = infant
坊や = boy
???
If the kanji mostly means "'quarters", why would the word for baby be "red quarters"?
Isn't it more likely that there was a Japanese word for "boy" that sounded like "bo" and got mixed up with a kanji character that sounded the same and also means "quarters"?
There are Japanese expressions where a person is referred to by an architectural term, like "mangaka" "漫画家" which literally means "manga house", but…?
AntC said,
May 10, 2018 @ 3:11 am
"'.. beware …,
If your Snark be a Boojum! "
~flow said,
May 10, 2018 @ 5:56 am
@AG not sure whether that -家 suffix to denote professionals originated in China or Japan (tho I'd guess China), but in Japanese at least it's not at all uncommon to conflate the place with the one who works there, e.g. 花屋さん (はなやさん) means first the owner of the flower shop—the florist—but then also the flower shop; cf. "the florist's".
Victor Mair said,
May 10, 2018 @ 6:07 am
From Juha Janhunen:
Well, Mongolian böö < *büxe 'shaman' is, of course, a borrowing from Pre-Proto-Turkic *bügö, from which Turkic bügü 'wise'. Considering the development of the vowels in the second syllable, this belongs to the same layer of borrowings as Mo altan = Turkic altun, both < Pre-Proto-Turkic *alton. In my opinion, these borrowings took place in the late 1st millenium in the context of the Xiongnu-Xianbei interaction.
Victor Mair said,
May 10, 2018 @ 6:09 am
From Marcel Erdal:
I think there was no Pre-Proto-Turkic *bügö, nor Turkic bügü 'wise'. There is pretty solid evidence for Turkic bögö 'sage', bökä 'hero' and bügü 'magic' (the activity), all apparently unrelated. Who took what from whom is a matter of conjecture.
Ian said,
May 10, 2018 @ 8:33 am
@David Marjanović
> Mongolian does not sound like Turkish and never did.
You're right that it currently doesn't, but there is much reason to believe that it did, around the time of Middle Mongol, or Proto-Mongolic. Some recent proposals suggest that Proto-Mongolic be reconstructed with tongue root harmony, like much of modern Mongolian, but Juha Janhunen and I are currently working on a paper arguing the contrary.
Indeed, for much of the current era it has been a given that Proto-Mongolic be reconstructed with palatal-velar harmony like Turkish. This is why the romanization of Mongolian -RTR vowels includes the umlaut, since it is believed that at one point they were true front vowels. Since then, however, a vowel rotation has lowered and pharyngealized the original back vowels and velarized the original front rounded vowels, resulting in the system we see today.
Chris Button said,
May 10, 2018 @ 8:51 am
Regarding 坊, the word "temple" has a similar etymology in English as a specifically separated off area.
Thanks for clarifying that. That's another great example of phonetic pre-nasalization being borrowed phonemically as a nasal (which is incidentally one of the reasons I remain somewhat skeptical of a nasal prefix in Old Chinese supposedly behind voiced obstruents rather than simple phonetic pre-nasalization being attested in loans).
Chris Button said,
May 10, 2018 @ 9:17 am
@ AG
Perhaps in a similar way to how "boy" originally referred to a male servant.
Alyssa said,
May 10, 2018 @ 12:36 pm
"The English word “shaman” is part of a cluster of words with loathsome connotations: shat, sham, shambles, shatter, shanty, shabby, shallow, shackles, shack, shaggy…"
This is fairly convincing, actually. For your average English speaker, a stereotypical shaman would probably look shaggy, shabby and in shambles, they'd live in a shack or shanty, and we'd probably consider them a sham. "Loathsome" is surely too broad to be meaningful, but I don't think it's crazy to suppose a cluster here with connotations like "ragged, primitive, poor"
David Marjanović said,
May 11, 2018 @ 8:50 am
That'll be interesting to read, I'm looking forward to it.
I think it'll be a tall order to argue that this unprecedented rotation is more likely to have happened than the reverse.
~flow said,
May 11, 2018 @ 8:52 am
@Chris Button "It's based on 16th c. Middle Japanese boozu (< bauzu < baũzu) 坊主 which still had a prenasalized /z/ […]. That's another great example of phonetic pre-nasalization being borrowed phonemically as a nasal"
Isn't it the other way round? 坊 does have a velar nasal coda in the originating language (Chinese), so the simplest assumption would be that nasality did occur phonemically in the borrowing language (Japanese) as well, from which it later waned.
Chris Button said,
May 11, 2018 @ 1:34 pm
@ ~flow
My cursory understanding is that the OJ phoneme in question at the time of the loan from Japanese that ultimately gave the nz cluster in "bonze" would have been/ⁿz/ (paralleling OJ pre-nasalized /ᵑɡ/, /ⁿd/, /ᵐb/). I'm not a specialist in OJ by any means, but connected to your point (although not relevant to the loan process above), is the suggestion that pre-nasalisation of medial voiced stops may be associated with a preceding nasalized syllable. However since voiceless obstruents apparently became voiced medially without pre-nasalization, it seems to me that the pre-nasalization is perhaps rather reflective of an articulatory mechanism to preserve voicing than the result of a nasal conditioning environment. Perhaps an OJ specialist can chime in here?
SO said,
May 12, 2018 @ 11:18 am
I should have taken this one step further earlier by writing "boozu (< bauzu < baũzu < baũ + s(y)u)" instead. That is, Middle Chinese -ng is reflected as the nasal vowel ũ at first, which later (approx. Kamakura times) merges with its oral counterpart u and still later (Muromachi) gives long oo in combination with the preceding a. Voiceless s- in the 2nd syllable of the word is turned into voiced/prenasalized (later: plain voiced) z- precisely due to the presence of a nasal (here a nasal V but a nasal C had the same effect) at the end of the 1st syllable. I.e. even the modern standard Japanese form boozu lacking any nasal element preserves a sound change induced by nasality that was originally there.
For a by and large parallel case see e.g. Spanish / Portuguese biombo 'folding screen' from Middle Japanese byoobu < byaubu < byaũbũ < byaũ + fũ 屏風.
However, not every voiced/prenasalized obstruent in earlier Japanese has a source involving a nasal of some kind. Plain voiced obstruents are usually likewise borrowed as voiced/prenas. obstruents. For instance, m in Korean tambae 'tobacco' preserves the prenasalization of b in Middle Japanese tabako, which wasn't there yet in its Portuguese (?) donor word.
<<< However since voiceless obstruents apparently became voiced medially without pre-nasalization, […]
This certainly does happen in some varieties in the NE and SW, but to the best of my knowledge there is no real, substantial evidence for the same phenomenon e.g. in Kyoto Jp. (The idea has been around since at least the 1950s however…)
Chris Button said,
May 12, 2018 @ 8:17 pm
@ SO
Thanks for the additional information.
If voiced obstruents emerged in OJ word medially and were characterized by pre-nasalisation, then would this not render inconsequential the role of any nasal component in a prior syllable? In short, how do we know that /bauⁿzu/ came from /baũzu/ with inherited nasality as opposed to from /bauzu/ (the original nasal component already lost) with the nasality of /ⁿz/ simply being a surface reflex of underlying /z/ in order to more easily maintain the voicing?
SO said,
May 12, 2018 @ 9:48 pm
主 in 坊主 should give s(y)u when borrowed from Middle Chinese into Jp., as the initial was voiceless in the donor language. Assuming a regular course of events, z in boozu can only be explained by a(n original) nasal final in the preceding syllable. The same is true for the second b in byoobu.
But of course the resulting form in Jp. as such doesn't necessarily give us much information in that regard. At least not its 16th c. or modern form. I would have to check this for 坊, but chances are that its Sino-Jp. reading is attested with explicitly marked ũ in the late Heian glossary / character dictionary Ruiju myōgishō 類聚名義抄 for instance.
(To avoid confusion: "your" /ⁿz/ and "my" /z/ refer to one and the same phoneme throughout. There is no evidence for a phonological contrast between plain voiced C and prenasalized voiced C at any textually attested stage of the language.)
Chris Button said,
May 13, 2018 @ 3:50 pm
@ SO
So, forgive my general ignorance here, if we take a word like Modern Japanese ha.da "skin" which presumably went back to something like Old Japanese pa.ⁿda, what is the evidence for an earlier nasal final in the first syllable? My assumption had been that the prenasalization in ⁿd was simply a surface articulatory phenomenon related to maintaining the voicing of the obstruent.
SO said,
May 13, 2018 @ 6:59 pm
In the case of boozu absence of a nasal in the first syllable should have resulted in *boosu instead. We can only tell because we know what the two components of the word are in Middle Chinese and Sino-Japanese. Otherwise there would be no way to tell why we get a z here.
For "native" Japanese words it's usually much more difficult to tell. In some cases …C[+v/n]… goes back to …NVC[-v/n]… in a transparent fashion (murazi < mura-nusi, kazasi < kami-sasi, or also Classical Japanese ikaga < ika=ni=ka etc), or also to …C[+v/n]VC[-v/n]… for that matter (kizi < kigisi, or also Classical Japanese -Ade < – Azu.te etc), but for many many other words we simply cannot tell.
I don't know of any good Japanese or Japonic etymology for OJ pada. Might be from *paNVta in analogy to cases such as those mentioned above but to the best of my knowledge there's no evidence for this. Might as well have been pada from the beginning. I guess we'll never really know.
Chris Button said,
May 13, 2018 @ 10:25 pm
Thanks SO – that makes sense. The idea that it might have been /pada/ [pa.ⁿda] from the beginning suggests to me that an earlier OJ voicing distinction was possibly lost unless the obstruent occurred in an environment conducive to voicing (e.g. intervocalically) in which case the voicing was retained through prenasalization.