I find it hard to believe I’ve never posted about rebetika, since not only do I love the music (when I was in Athens I sought out a dusty record store where I could buy some LPs I then had to lug back to New York) but the word itself is very interesting. For one thing, there’s no unanimity on how to spell it; Wikipedia has it under Rebetiko (“plural rebetika […], occasionally transliterated as rembetiko or rebetico), while the OED (entry from 2002) has it s.v. rebetika (sadly, it’s not in M-W or AHD under any spelling). Here’s the OED definition, which is quite discursive:
A style of Greek popular song, characterized by lyrics depicting urban and underworld themes, a passionate vocal style, and an ensemble accompaniment played esp. on stringed instruments such as the violin, bouzouki, etc.; (with plural agreement) the songs themselves. Also (in form rebetiko): a song in this style. Frequently attributive.
First recorded commercially in Turkey before the First World War (1914–18), rebetika is assumed to have long existed (under various other generic names) as an oral tradition in Mediterranean seaports and prisons. Following the Greco–Turkish war of 1919–22, the genre became associated with the numerous Anatolian refugees settling in Athens. Extensively recorded and performed in the 1920s and 1930s, notably by immigrants from Asia Minor, Piraeus bouzouki players, and Greek Americans, rebetika also became known in English as ‘Greek Blues’ or ‘Piraeus Blues’.
But it’s the etymology that makes it a must-post, and happily Martin Schwartz has sent me a recent article of his on the subject. First I’ll provide the OED version:
< modern Greek ρεμπέτικα, plural of ρεμπέτικο eastern-style song of urban low life, use as noun of neuter singular of ρεμπέτικος of vagabonds or rebels, probably < ρεμπέτης rebetis n. + ‑ικος ‑ic suffix.
Notes
On the further etymology, compare note at rebetis n.
The forms with ‑mb‑ arise from the influence of an idiosyncratic transliteration of the modern Greek (in which the sequence ‑μπ‑ normally represents b), originally in G. Holst Road to Rembetika (1975).
(I think of it as rembetika because I was introduced to it by that Gail Holst book, which I recommend.) Now to Martin’s “A rebetic roundup: people, songs, words, and whatnot” (published as ch. 27 of The SOAS Rebetiko Reader); I’ll quote some bits and urge you to visit the link for more:
Today the adjective “rebetika”, as used by the majority of Greeks, refers to urban Greek music of the earlier half of the 20th century, and is associated with lyrics reflecting lower class culture – drugs, thugs, drink, pimps, prisons, poverty, illness, alienation and thwarted love – although the wide range of the genre makes it describable as an urban popular music, with a déclassé aspect. Indeed, its songs, which are for the most part based on several fixed dance rhythms, played an important role in the Greater Athenian recording and nightclub scene from shortly after the 1922 Asia Minor Catastrophe well into the 1950s and to some extent later. The term “rebetika” has, to shifting degrees, been applied to two successive but overlapping chronological varieties. The first, from ca. 1923 to 1937, is characterised by musical styles, instruments, and vocal techniques continuing, or much influenced by, those of the Greeks of Turkey, chiefly of Smyrna and Constantinople, and including material of Turkish origin. The second, from the early 1930s into the 1950s, while thematically and choreographically related to the first, featured the bouzouki, an earthier singing style, and an increasingly Greco-European profile. […]
Although I am marginally a “rebetologist”, my central discipline is as an etymologist, historical linguistics being my chief academic activity. It is from this perspective, with the aid of some “rebetological” data, that I shall address the history of the terms rebétis and rebétiko / rebétika.
A preliminary notice: I use the transcription rebétika as representing the pronunciation used by most Greeks, as against the often encountered “rembetika”; in Greek spelling, μπ (mp) is necessary to indicate the sound /b/, and in this instance the μ (m) is silent, but wrongly present as a frequent transcription into Latin letters.
After dismissing some other theories (deriving it from alleged Pre-Modern Turkish rebet asker, Greek rébelos ‘a rebel,’ and Arabic ribaṭ), he continues:
The most fruitful direction for our linguistic quest is to proceed from Ancient Greek PEMB- (rhemb-, Mod. Gr. remv-) ‘to wander’, which gives re(m)b- (with-μπ-) in various Late and Modern Greek verbs and nouns referring to loafing, laziness, relaxed enjoyment, etc.; see Gauntlett 1982: p. 90, fn. 51. With the base rebet- itself is the word rebéta found in several literary attestations from 1871 onward as an argot term in Smyrna and Constantinople for ‘a lower-class neighborhood populated by criminals’ (from ‘unruly place’, as still used in 1895 by N. Georgiadis for the festivals [pane(gh)iria] in Silivri). It is interesting that when in 1918 the Constantinopolitan N. Sofron, writing sketches of everyday life in his city, took as a nom de plume Rebétos derived from rebéta in its older usage, and not from rebétis, which shows that the latter form was not yet common. For rebétis, the first occurrence (date unclear to me) seems to be in Nikolaos G. Politis’ serial ethnographic volumes called Paradoseis, in which a character named Giannis the Rebétis figures, although nothing informative is said of him, and, as we shall see, rebétis is not found again until 1923. […]
There remains the question of the newly emerged earliest literary occurrence of “rebetiko” as connected with this designation on the record labels, and the relationship of rebétis to both, which gets us back to our linguistic inquiry. Vlisidis’ material indeed disproves the idea that the term “rebetiko” on record labels was (as proposed by Panos Savvopoulos) just an invention on the part of the recording companies. As Vlisidis indicates, the record labels from 1912-1913 bearing the characterisation “rebétiko” drew on a word which was current at the time. However, Vlisidis’ further proposal, that the literary material which calls itself rebétiko/a was reflected by these discs is problematic. The underclass nature of the diction, as well as the thematics of the four poems which are called “rebétiko / rebétika”, differ dramatically from what we find for the two 1912/13 light love songs called “rebétiko” on the record labels, and also from the many subsequent recordings bearing that epithet on the label. […]
We now have enough material to offer a solution to the problem of the term rebétiko. A linguistic approach would also involve distinguishing between and then reconciling the various usages of what are in fact complexly related words, rebéta, rebétiko, and rebétis. As a mannerism first used literarily in 1912, rebétiko would be an adjectival invention, ‘pertaining to the rebéta’, i.e. ‘that which belongs to the underclass realm’. From popular magazines of the period (cf. Vlisidis), it would have been noticed by Greeks involved in the recording industry, who however took it to be derived from the verb re(m)bo etc. referring to rambles, indolent or relaxed enjoyment, the word thereby providing for the categorisation of discs a trendy-sounding designation of miscellaneous light songs, such as we find in “Aponia” and “Tiki Tiki Tak”. Toward the mid-1920s, however, with the emergence of rebétis for a member of a lower-class subculture, music pertaining to the latter world began to enter the miscellaneous industrial category, explaining the diverse and contradictory range of recordings labeled “rebétiko”.
This now calls for an account of the origin of rebétis. Politis’ obscure attestation of rebétis may reflect a temporary neologism based on one hand on rebéta (cf. Georgiadis’ 1918 rebétos) and on the other hand constituting a regular derivation with -étis from the verb root ré(m)b-, see Gauntlett 1982, pp. 90-91 for parallels; note however that such a derivation is not “undermined” by nouns with -étis yielding adjectives with -etikós vs. the accentuation of rebétiko, which precedes, and is NOT derived from rebétis. For the formation of the more conclusive 1923 attestation of rebétis by “Smyrnios”, one has, alongside a deverbal explanation of rebétis, the possibility of a “back-formation” from rebétiko ‘pertaining to the underclass realm’. Given the 1923 attestation of rebétis and its continuation by Markos in his 1933 “O Harmanis” [The drug-deprived one], Pikros’ 1925 mention of rebéta as in effect the feminine equivalent of rebétis seems suspicious; one would rather expect rebétisa (cf. ghóis [Anc. Gr. góēs] ‘sorcerer’: ghóisa ‘sorceress’ continuing the ancient fem. suffix -issa), which is found canonically in our songs. Given that rebétis itself was still only marginally attested, perhaps Pikros had misunderstood a phrase with the probably already obsolescent rebéta ‘lower-class milieu’, taking the latter as its female personification, or, in a context referring to a group of people, he misinterpreted rebétes as a plural of rebéta rather than of rebétis.
There’s much more (e.g., “It is possible that the suffixation of rebétis was supported by a traditional underclass word of the same semantic field, serétis ‘tough guy’, of Turkish origin”), but I will reluctantly stop quoting and send you to the link. I just want to add something about the difficult issue of nasal + consonant clusters and how to transliterate them. Peter Mackridge, in his excellent 1987 The Modern Greek Language (Amazon, Internet Archive), writes:
To begin with the combinations of nasal + consonant that existed in traditional demotic, some dialects always pronounced the nasal fully, others always omitted it completely, while others displayed a certain variety. Grammarians, on the other hand, have taught that these combinations should be pronounced with or without the nasal according to whether the nasal was present in an earlier version of the word […]. With the rise of literacy, however, speakers have usually treated every instance […] alike, that is, either always with or always without the nasal, according to each speaker’s idiolect. Furthermore, it cannot be expected that speakers will know the etymological origin of all the words they use.
Most scholars now seem to have settled on nasalless versions, but I confess it makes me uneasy, since I always think of the Greek script with its nasals. I also have to point out that my two bilingual dictionaries, D. N. Stavropoulos’s Oxford English-Greek Learner’s Dictionary and J. T. Pring’s Oxford Dictionary of Modern Greek, handle these words very differently; the former has ρεμπέτης ‘outcast, scamp, rebetis’ and ρεμπέτικος ‘of/from a rebetis,’ while the latter has only ρεμπέτικο ‘sort of popular song in oriental style.’ And as I look at those entries, I note the following word in each: Stavropoulos has ρεμπούπλικα ‘trilby, homburg, felt hat,’ whereas Pring has ρε(μ)πούμπλικα ‘trilby or homburg hat’ (Wiktionary has it as ρεπούμπλικα). Truly Greek is a land of contrasts.
the sequence ‑μπ‑ normally represents –
– at the beginning of a word: “b”
– in the middle of a word: “b” or “mb” depending on your dialect (and possibly your idiolect? – not sure about that) I believe that just “b” is more a northern Greece thing, but happy to stand corrected by those more knowledgeable.
Hence the alternative English transliterations.
Yes, of course. The question is how widespread the “dry” versions (to use this guy’s terminology) are, and thus how much sense it makes to generalize them in transliteration.
I am amused and intrigued by the claim that the word initially became a standard style/genre descriptor in the music biz only because the relevant music biz suits misunderstood its meaning and etymology.
Odd use of diction in that article—not really fitting of either a linguist or a musicologist. Normally, when discussing singing, diction means “clarity of enunciation,” but that doesn’t seem to be what he meant by it.
So is rebéta Greek for jiānghú?
did the apparent lack of interest in looking at turkish (or romanes/rromani ćhib) parallels strike anyone else as a bit tendentious, when thinking about a genre centered in istanbul and smyrna (and sung to a significant degree by people whose home-community languages were not greek – judezmo/ladino, romanes/rromani ćhib, and i think maybe armenian)? i don’t know enough about subcultural/underworld languages in those places to really know whether to expect a turkish connection, but the little i know about lubunca makes it seem quite possible.
I still sometimes affectionately call the basketball powerhouse “Maccampi Tel-Aviv“, reminiscing on reading the scores in Greek broadcasts of away games in the ’90s.
When I see “rebetika”, I think of Misirlou, and Dick Dale, and surf music.
Greek-Nigerian μπάσκετ-μπολ (mbasketmbol) players include 4 brothers surnamed Antetokounmpo (< Adetokunbo).
1) Amalia Arvaniti & Brian D. Joseph. Variation in voiced stop prenasalization in Greek. Γλωσσολογία 11-12 (2000): 131-166.
2) For listening to Rembetiko on the Internet: https://rebetiko.sealabs.net/radioplayer.php
Not only μπάσκετ-μπολ, but also μπάντμιντον.
I had never heard of rebetiko until a few weeks ago when driving back from Italy I heard an hour long progam on Greek popular music on ORF, our excellent public radio station in Austria. There’s actually a link, if anyone is interested: https://oe1.orf.at/programm/20250823/804516/Diagonals-Feiner-Musiksalon-Sommer-Edition-Hellas
For ORF, read ÖRF? 😊 Unless you do not recognise the name change from Ostmark.
No, ORF is correct for the broadcaster and the TV channels even though the primary radio channel is called “Ö1”. Good question why the umlaut was dropped long before the internet might have made that a convenient idea. Short for Oesterreich ? Or maybe someone in 1968 just decided it made for a cleaner logo.
Even in Germany, in the 1960/70s, a lot of cheap typewriters didn’t have Umlauts (for example, the first typewriter I had as an adolescent). The typescripts of many of Paul Celan’s poems were written with such a typewriter.
rozele makes a perfectly fair point although of course these influences could have flowed in all directions. The variety of early-20th-centuryTurkish spoken by any Turcophone pimps and dope dealers and similar demimondaines could perfectly well have had a bunch of Greek loanwords absent from more elevated Ottoman registers.
Re “no unanimity on how to spell” you’d think there would be other loanwords taken into English from Modern Greek where there would be orthographic variation in English because of different transliteration approaches. It would depend on the word, of course, but beyond these nasal combinations there are a certain number of other contexts where “going with the traditional romanization of Ancient Greek” and “trying to capture the pronunciation of Modern Greek” will give you predictably different outputs. Does a beta yield a B or a V, does a gamma yield a G or a Y, etc etc. Although maybe no one is pretentious or classicizing enough to try “rhetsina” for retsina?
did the apparent lack of interest in looking at turkish (or romanes/rromani ćhib) parallels strike anyone else as a bit tendentious, when thinking about a genre centered in istanbul and smyrna (and sung to a significant degree by people whose home-community languages were not greek – judezmo/ladino, romanes/rromani ćhib, and i think maybe armenian)? i don’t know enough about subcultural/underworld languages in those places to really know whether to expect a turkish connection, but the little i know about lubunca makes it seem quite possible.
You know, you could actually click through and find out what he says. Here are a couple of paragraphs that might suggest he doesn’t exhibit a lack of interest in Turkish:
I would gently point out that he’s an etymologist, not some random Turkophobe; he just doesn’t happen to think this particular word is from Turkish.
Am I right that some Greeks would pronounce Y’s μπάντμιντον badminton and others badmidon? (And that, as per Julian’s post, none would say mbadminton?)
Right on both counts.
I was idly musing as to whether the spelling “baklava” might reflect word-internal inconsistency in romanization* but alas the corresponding Greek word is μπακλαβάς. (Plus it supposedly came into English most immediately from Turkish rather than Greek.)
In thinking about other Modern-Greek-origin loanwords in current AmEng it seemed they were heavily concentrated in food and beverage terms. Contrary to my speculation above about variable orthography, it strikes me that there are good functional reasons for proprietors of Greek restaurants in an Anglophone society to want their menus to use romanizations that are uniform from place to place and now I’m curious if there’s some historical process (formal or informal, centralized or decentralized) that has brought that about.
*Which would be amusing if nothing else. Maybe there’s another example out there? Hope springs eternal.
μπάντμιντον badminton and badmidon
What about bantminton?
apologies, @hat – reading through this morning, i see that he did just that!
ORF, OeNB, ÖNB – Oesterreichischer Rundfunk, Oesterreichische Nationalbank, Österreichische Nationalbibliothek.
ts is simply too blatantly postclassical for that.
Raidió Teilifís Éireann was established in 1960. Between about 1980 and 2005 its initials gradually changed from RTE to RTÉ
So many interesting comments, many of which I hope to be able to address by Saturday! I do discuss various other aspects of Turkish etymology in my article, which I hope those interested in the subject will read from the link kindly provided by Language Hat. The article contains much on non-linguistic matters: History, sociology, biography, poetics, many translations, melodic comparisons (Eastern Ashkenazic etc.), anecdotes of collecting old discs, spoofs, and cartoons.
Martin Schwartz
Electricité de France stubbornly remains EdF for typewriter reasons.
Gonna stick this here as the most convenient recent thread involving orthographic/transliterational variability. I was browsing through an update of perhaps dull/bureaucratic ecclesiastical news when I saw an item advising that the modest Eastern Orthodox congregation in Whitehorse, Yukon Territory is no longer named “St. Nikolai, Bishop of Zhitsa Mission” but has been officially renamed “St. Nikolai, Bishop of Zhicha Mission.” The “standard” romanization of the underlying Serbian toponym Жича per FYLOSC orthographic conventions is Žiča, but of course English is not FYLOSC* and one can well imagine that the supply of diacritical marks in the Yukon may be inadequate and/or of unpredictable reliability, thus motivating an alternative ASCII-friendly approach using digraphs. I couldn’t remember having previously seen the “Zhitsa” alternative, but google books does have an example showing it in use in Anglophone texts in the 1920’s.
*Indeed, per FYLOSC conventions the first name would be Nikolaj rather than Nikolai.
I definitely approve of that change — who could possibly know that “Zhitsa” was supposed to be pronounced Zhicha? (Hard enough for English-speakers to manage the “zh” in the first place!)
Yeah, I don’t know what motivated the “Zhitsa” variant, other than the fact that it was previously attested must mean something. I guess one possibility is variant dialect pronunciation within FYLOSC and someone working with a “non-standard” original. Another might be Ч being mistaken for Ц due to a typo in a Cyrillic original or sheer confusion. There are no doubt other possibilities, and once it’s floating around in some English texts it can be picked up by others who don’t know the original language or the Cyrillic spelling. The particular saint does have a certain following in North America that extends beyond the ethnically-Serbian communities so the Whitehorse congregation might well have been named without any FYLOSC speakers (even limited-proficiency “heritage” speakers) being involved.
Nikola rather, as in Tesla.
“Cakavian” is spoken on the Croatian coast somewhere, so that’s unlikely.
That’s likely; in some fonts (handwritten on icons in particular?) they can look pretty similar.
This particular saint was a monk-and-bishop so of course he was known by a name different from what his parents had given him at birth. But somewhat oddly he had originally been named Nikola/Никола but took Nikolaj/Николај as his monastic name. Which is a more subtle change than what is usually done. Sometimes monastic/religious subcultures use forms of names not otherwise particularly current in the relevant society – e.g. everyone knows that the Russian for “John” is Ivan, but there are Russian saints (plus I assume some monks and bishops not generally regarded as saints …) named Ioann/Иоа́нн instead.
Ah. Church Slavic (of Serbian, Russian etc. recension).
@languagehat and all: The OED definition in small print is problematic, because the urban songs on Greek discs recorded in Constantinople and Smyrna before WW I, while they contain MUSICAL genres found the Athenian rebetica, only very rarely and limitedly refer to underworld themes. To call these discs from Turkey rebetika is anachronistic.
@Brett: You’ll see that I use “diction” with regard to written prose and poetry, NOT singing. My usage accords with the s.v. online Oxford Language definition, 1) ‘Choice or use of words in speech or writing”.
@Jerry Friedman: I didn’t know jianghú, but from what I see, I’d say the social contexts are sufficiently different for comparison to have much value.
@rozele: I don’t know of Ladino, Romani, or Armenian words in rebetic songs or in Greek at large. The former languages were spoken in private; in public their speakers use Turkish
(or in Smyrna, probably Greek as well) Thanks for lubunca, which I didn’t know about. I doubt that any lubunca words entered lower class Greek argot. BUT Elias Petropoulos wrote a
fascinating (Greek) book, Kaliarda (Καλιαρντά), about homosexual Greek argot of a few decades back. I doubt that here, too, is there any lubunca, BUT the etymology of the word (contra Wiktionary’s and Petropoulos’ suggestions, resp. gaillard and Goliard), was shown by someone in either The Journal of Modern Greek Studies or Byzantine and Modern Greek Studies to go back to a Romani word for café habitués, where the “café/coffee’ part is a compound from Romani kali ‘black’, cf. Skt. kāli- ‘black’. Btw čhib is pleasantly from Skt. jihvā ‘tongue’.
More tomorrow.
We discussed Kaliarda back in 2018 (I linked to a great series of posts by Nick Nicholas of Ἡλληνιστεύκοντος).
I just want to thank Martin Schwartz for this article (and Hat for blogging about it). I’ve loved rebetika (probably most especially A Kostis) (& etymology) for a long time, & I now know so much more about it than I did before. I wish more articles rambled (rebo’d?) in the way this one does
Man, I posted the above before I finished the article. The record collecting details alone are… I dunno, I really liked reading them
@languagehat: Wow, 2018 Kaliarda ! I should probe NN’s posts on this further. I see he lists ∂ikélo “I see’; maybe < Romani dik- ‘to see’? Maybe I underestimated the role of Romani in Kaliarda; I should check when I get the chance.
I can’t resist mentioning my favorite Kaliarda kenning from Petropoulos: Zougklogougoulfú /zungloghughulfú/. The analysis is -ú Kal. abstract/institutional suffix; zungloghughulfís Kal. ‘lion’ < zungla Gr ‘jungle’ and Kal. ghúghulfos ‘dog’ (< Eng. wolf + Gr ghuf-ghuf, the sound of barking); thus the (commercial) company of the lion’ = Metro-Goldwyn-Mayer !
@rozele: 2 rebetic songs from the ’30s with Ladinoid phrases; “Xanthi Evreopoula” ‘Blonde Jewish Girl’ sung by Rita Abatzi, where there are Ladinoid and Araboid phrases; I don’t know whether te kyero mučo is just Span. te quiero mucho or a Ladino dialect with mučo vs. munčo. The word occurs among the Ladino phrases in Vidalis & Roza Eskenazy “Evreopoula”, but I can’t make out from the YouTube (I no longer have the 78 rpm disc) if Roza, who spoke Ladino, has the -n-. She ends with “vamos, vamos”, apparently having been convinced to run off with the drunken swain. I traced the melody in part to a Greek syrto dance riff, and in part to a klezmer instrumental circulating in Constantinople ca. 1911; long story.
@sh: The important Arvaniti-Joseph 2002 article was superseded by the same authors’ 2004 article “Early Modern Greek /b d g/: Evidence from Rembétika and Folk Songs“. All 14 tracks in Appendix I are from my archival reissue CD Greek-Oriental Rebetica: Songs and Dances in the Asia Minor Style. Brian Joseph told me that my CD inspired the article!
Recently I was working via Zoom with an excellent singer from Northern Greece, who has a talent for imitating foreign singing styles. Despite my coaching in her final recording she was not able to produce the n in “friend” (vel sim.) and could only produce “Fred”, as it were. An anecdote for the reverse effect: Long ago in Athens I made the acquaintance of a fellow who ran errands for antique stores and spoke English, after a fashion. I was taken by his story about a woman with a “bing” nose (imagined a Bing cherry which would light up and ring.) I roguishly gave him this English sentence to translate into Greek: “The man jumped from the 5th storey apartment and left his wife a widow.” He translated, with last part “k afise tin yineka (ghineka) tu hira” “and left his wife a window”. A satisfying defenestration report indeed. Equally when I had a Brazilian woman read aloud “The sight of the convicts in chains brought tears from the crowd”; she pleasingly had “cheers” for “tears”. But I digress (as usual).
@whoever (no time to check now) mentioned Radio Rebetiko Sealabs: I often listen to it for background music, but NB– the selections are not all rebetika; plenty of European-style songs and Greek rural-style songs. It’s like a commercial-free 24 hour (mainly) jukebox of Greek popular music from the 1930s and 1940s.
@J.W.Brewer: baklavá would be the Greek accusative of baklavás; note the Eng. lexicalization “mous(s)aka” from acc. of Gr. nom. /musakás/.
@all: rebetika would parallel words like adío ‘farewell’ where the voiced stop is not generally prenasalized. I do hear Greeks (for whom the lower-class musical culture is somewhat distant) say /rembetika/. Here the unavoidable written Greek letter m compounds the matter.
@languagehat and all: I just opened to the beginning of Petropoulos’ extensive Kaliarda glossary and I see avéle apokaté ‘come here’ and avéli go(n)doreliá ‘ wind is blowing’ (the second word a kenning ‘godfartery’). Clearly < Romani avel- ‘to come’. Kal. avélo is an important auxiliary verb with a great range of meanings, but ‘come’ is clearly the basis. I now suspect that the Romani component of Kal. is great.
Reading that comment made me hungry for the music, and I’m now listening to my copy of the CD Greek-Oriental Rebetica. Love that Rita Abatzi!
@Martin Schwartz: Thanks for the reply. I wasn’t entirely serious, more amused that two underworlds had come up here fairly close together, but I think you have a good point.
So /musáka/, the form I’m used to, has undergone the neoštokavian stress shift?
@Martin Schwartz: interesting point re English “moussaka” matching the accusative rather than nominative in Modern Greek, but if the references are right that English borrowed “baklava” straight from Turkish rather than via Greek the situations aren’t parallel since that’s apparently the nominative (or “absolute,” if you like) form in Turkish.
English gets moussaka not only from Greek but also from other Balkan languages, e.g. Serbian and Croatian musaka. AHD online made a last-minute update to the etymology in 2017:
“saqqā, to chill” doesn’t seem to fit here; I wonder if it was accidentally left over from their previous etymology:
which doesn’t seem to make sense since moussaka is baked, so it’s a good thing they updated it. (Though they didn’t get the update into the Semitic root appendix, perhaps because by 2017 AHD was on the verge of shutting down.)
AHD’s explanation about liquids being poured over the layer of meat is different from other sources; the OED (2003) takes it from the same Arabic word but finds it difficult to explain: “The Arabic etymon suggests a procedure of repeatedly adding liquid during cooking; however, this is not mentioned for moussaka in Turkish cookery books, old or new. It perhaps refers simply to the addition of the sauce.”
Davidson’s Oxford|Penguin Companion to Food (1999 and subsequent editions) has an entry on moussaka signed CP (Charles Perry), with a paragraph on the etymology:
Charles Perry would know if anyone does, but it raises the questions, how sure are we that the dish isn’t of Arab origin, and if it originated in Turkey why did they give it an Arabic name? And if it’s named because of tomato juices, it has to be post-Columbian — is that known for sure?
@rozele and all: Petropoulos, Kaliarda, gives such words as /lubunyá/, /lubína/, and /lúba/, all for effeminate gay guy; this word is clearly from Turkish slang lubunya in the same mg., whence Turkish lubunca, apparently < Turkish lubni ‘whore’. In the Petropoulos book I also saw /tsurnô/ ‘theft’, surely to Romani čor ‘thief’, and /dik/ (spelled NTIK) ‘look’, again < Romani, which Petropoulos takes as imperative of /∂ikélo/, but the latter belongs to a series of Kaliarda verbs with -élo based on avélo. The /∂-/ would be Hellenized from /d-/. Also latsós ‘nice/pretty/beautiful’ (Gr. /oréos) < Romani lačo ‘id. (not ‘safe’, as per the received transl. of Latcho Drom).
@J.W. Brewer and ktschwarz and all: The Wiki Moussaka is full of good stuff. The Turkish, as I think is now clear, would be from Ottoman Egyptian ‘chilled’, the Egyptian dish being “usually chilled for a day or so”. Btw for the etymologically different Arabic for ‘watered’, Perry transcribes musaqqâ, following the Arabist convention of noting -ā as -â when it is written with silent (etymological) -y. An interesting datum from the Wiki is that the familiar Greek moussaka was invented by Nikolaos Tselementes, who spent time as a chef in the US, in the 1920s; he innovated e.g. the béchamel sauce layer. Whatever the story of the 1862 (OED) muzakka (with grave accent on each vowel), one can hardly doubt that our English “moussaka” is from Greek, given the many Greek (incl. Cypriot) (vs. Turkish and non-Greek Balkan) restaurants in the U.S., Canada, the UK, and Australia. Re the history of our word, it should be noted that food names often give variant meanings. Cf. e.g. cutlet as defined by OED vs. Merriam-Webster (for my Bronx childhood the word referred to a lamb-chop), and then again Slavic kotleta (my Russophone wife served tasty chicken kotlety for dinner this evening).
To close with a rebetic note: Musical instrument names also, notoriously, yield different applications. The Turkish bağlama is usually a long lute with a resonant sound; the borrowed Greek term bag(h)lamás (with gamma for spirant g) is a bouzouki-like instrument with a high pitch, small enough to be concealed (by rebetes on their way to a hashish-den, cautious of police) is one’s coat sleeve.) This instrument was important for the Piraeus school of rebetika. There is a nice YouTube “Greek Professional Baglamas” posted by ArtPort May 8, 2020; the young man playing is wearing a cool T-shirt showing Markos Vamvakaris and below him (Giorgos) Batis, playing the baghlama(s); the melody is by Batis, recorded when he was a member of the “Famous Piraeus Quartette/Four”, whom the T-shirt photo originally depicted.
There is a nice YouTube “Greek Professional Baglamas” posted by ArtPort May 8, 2020
Is this it?
@languagehat: Yep. The original song can be heard at:
Giorgos Batis Fonogrfitzides YouTube.
The sepia-tone photo shows the whole Quartette. Batis is wearing a bow-tie.
With correct -a- restored, /fonoghrafidzí∂es/ = ‘phonograph vendors’, apparently itinerants
whom the song regards as very cool guys.
Martin Schwartz