Friday, December 31, 2010


The latest issue of Massorot (contents and abstracts in English, PDF) includes, among other cool stuff, a paper by Moshe Bar-Asher on Maghribi Judeo-Arabic šarḥ of Ecclesiastes. Šarḥ of Qohelet assumes a special place among all the translations of Tanakh, Talmud and targumim made by and for the Arabic-speaking Jews of North Africa. First, it is one of the five Megillot which, as Bar-Asher notes in the introduction to the paper, were traditionally not translated. Secondly and more importantly, even if exceptions were made for the other four since they were to be publicly read at various points throughout the year,

קהלת לא תורגמה לפי שלא נכללה מעולם בסדרי הקריאה של הציבור הרחב. למדנים שנדרשו לה יכלו להבין בעצמם פירושים שלה שנכתבו עברית, ולא נזקקו לשרח בערבית

Qohelet was not translated because it was never included in the readings for the general public. Scholars who studied it could understand its meaning even it if was written in Hebrew and they did not need a translation into Arabic.

Nevertheless, continues professor Bar-Asher, translations of Qohelet were done as scholarly exercises where teachers would assign to their students the task of translating the book into Arabic. If I'm not altogether mistaken, Bar-Asher lists the following such instances:
- in Marrakech at the beginning of the 20th century on at least two occasions;
- in Sefrou in 1920's or 30's;
- and from a village near Tefilalt some time between 1924 and 1926, created by Moshe Bar-Asher's late father and recorded by professor Bar-Asher in 1983-84, the subject of the paper.

I'm more than happy to add one more to the list, earlier than any of the above and not a manuscript, but rather a printed version dated 5657 / 1896-1897 from the famous press of Solomon Belforte & Co. in Livorno. This one is not a part of my humble collection of late Judeo-Arabic prints, but it can be found in the British Library under the shelfmark 1906.a.42. The booklet contains:
- Song of Songs with its targum and Judeo-Arabic šarḥ of the latter which is quite similar, but not entirely identical to the one in my collection (Livorno 5615 / 1854-55, henceforth ŠM1). The first part also contains commentary on the targum explaining "difficult words and concepts in the Targum" (המלות הקשות והענין שבתרגום);
- prayers for Pesakh (תפלת המחה של פסח);
- and finally, Qohelet with šarḥ.

The title page as well as the introduction on the following page identify the author as Chaim Ha-Kohen The Younger (הצעיר חיים הכהן) and indicate that the work was compiled in Tripoli, Libya (טראבלס המערב). Who the author actually was is still unclear to me. With the reference to Tripoli, the first name that springs to mind is that of Mordechai Ha-Kohen (1856-1929), the author of The Book of Mordechai, a study of the history and customs of the Jewish population of Libya. The other possible candidate is Joseph Chaim Ha-Kohen (1851-1921) who, so his Wiki page tells me, was born in Morocco, moved to Jerusalem at an early age, but then often returned to Maghrib. However, a quick search among the seforim would seem to indicate that either of the two used their full name and neither is identified with one of at least three seforim signed by הצעיר חיים הכהן.

The identity of the author remains a mystery for now, but one detail deserves mentioning: this is the first šarḥ from Libya I am aware of. Whether its language reflects the Libyan dialect still remains to be seen, but I offer here some preliminary remarks, old-school style, based on the first two chapters which you can find here.
(Abbreviations: QL - the British Library Livorno print; QM - Bar-Asher 2010)

- The non-assimilated definite article is transcribed without the aleph: לִכְּלָאמָאת "deeds (lit. words)" (1:8), לְכַּל "everything" (1:14). One notable exception is אַלְכַּל in 1:2.
- The assimilation of the definite article is indicated throughout, sometimes also by means of a dagesh (אַגָֹּאהַל in 2:15,16; אַצְּלָאם in 2:13; אַרִּיח in 1:6), but usually without it: אַסִמְס "the sun" (throughout); אַצְנָאיַע "works" (2:17).
- The loss of [h] found in many dialects of North Africa (and in Malta), which results in a number of hypercorrections: הָאנָא "I", הָאש "what" (throughout).

Side note: The coolest such hypercorrection resulting from the loss of [h] has got to be the date on ŠM1. As with many Jewish books (but also everyday items) all over the world, the year is given as passage from the Tanakh with some of the letters highlighted, either through size or in some other way. Add up their numerical values (hint: final forms count as regular ones) and you get the short form year (לפ''ק = לפרט קטן), i.e. without the thousands. This is what it looks like on the title page of ŠM1:

This time the letters to be counted are highlighted using a little crown of dots above them, so השירה נא ליד left to right ads up to 4+10+30+1+50+5+200+10+300+5 = 615, which is the year 5615, i.e. 1854-55 CE. Except if you were to look for these three words in any copy of Tanakh, you wouldn't find them. The actual verse is אָשִׁירָה נָּא לִידִידִי ("Let me sing for my beloved"), from Isaiah 5:1. The author, compiler or printer knew that where they came from, ה was not pronounced and thought - or had been taught - it applied to the text of Tanakh as well. A wonderful example, indeed, but can we rely on the date being correct?

- Assimilation and dissimilation of [s] [z] and [ʃ] [ʒ] (sifflant / chuintant alternation [1]), so typical for Maghribi Judeo-Arabic writing: אַסִמְס "the sun" throughout, לִיס "is not" (1:11 and beyond), but לִיש in the first 8 verses, צַזַר "trees" (2:5, see below), וְיִזְרֶק "and rises" (1:5, see modern Maghribi Arabic šṛəq).
- Much more detailed analysis will be required to fully understand the system (if any) behind the transcription of the vowels using niqqudot. It would appear, however, that both patah and schva stand for [ə] (כְּבַרְתְ וְזַדְתְ "grew and added" 1:16), while qamats stands for [a]. To underscore the point, qamats is usually followed by aleph: מָאשִׁי "goes" (1:6), וַלְנסָאן "human" (2:21). There are exceptions to this, such as צָלְטָאן "the king" (1:1, 2:12) or עָלְמָא "wisdom" (1:16-17). Whether there's a method to this and just what it means (as you will note, in both cases qamats follows an pharyngealized consonant) will remain to be seen.

- Typical Neo-Arabic tafḵīm (pharyngealization): צָלְטָאן "king" (1:1).
- A particularly neat example of both sifflant / chuintant transformation and AND tafḵīm can be found in 2:5:

צְנַאעְת לִי ג'נָאנָת וסְוָאנִי . וַגְרַסְתְ פִיהוֹם צַזַר גְ'מִיע תְמָאר

The highlighted word is a translation of hebrew עֵץ = "tree" and traces back to Arabic شجر. Except in Maghribi Judeo-Arabic, it underwent double chuintant > sifflant transformation to first סז'ר (thus for example in ŠM1 1:16 כַסַזְ'רָא) and then to סזר. Subsequently, the whole word was pharyngealized to something along the lines of ṣəẓəṛ. The translator, however, only had צ at their disposal to indicate pharyngealization, hence צַזַר.

- Much more literal than QM, QL often employs participles where the Hebrew original does, even if fluent native Arabic wouldn't. Thus for example in 1:5, Hebrew סֹובֵב סֹבֵב "turns and turns" is translated as דָאיֶיר דָאיֶיר, whereas QM has more idiomatic יצייר תצוויר, i.e. imperfect followed by the verbal noun, a construction which indicates intensity.
- QL shows preference for the preposition לְ to translate the Hebrew אֶל, whereas QM prefers אילא.
- Like QM, QL uses לִיס for both "is not" and the negative particle.

Word choice:
In most word choices, QL is much more conservative (i.e. less dialectal) than QM. Thus for example:
- QL adopts the Hebrew הֲבֵל (as הְבֵיל), whereas QM uses the Arabic חתוף (with some exceptions).
- QL uses אַלִנְסָאן for "human", while QM uses the typical Maghribi בנאדם.
- QL has יְרוּשָלָיִם for "Jerusalem", while QM uses מדינת אסלאם.
In one instance, it's QM that is extremely literal: in 1:6, QM uses the Hebrew דארום and צאפון for "south" and "north", respectively. Interestingly, QL has here קַבְלִי and בַחְרִי, both terms I would describe as very Egyptian.
- For the relative pronoun, QL uses אַלְדִי with some exceptions, like אֵלִּי in 2:9. QM, predictably, uses the typically Moroccan דדי.
- QL uses אַלְכַּל "everything" substantively, but - like QM - גְ'מִיע as a determiner.
- QM occasionally uses the Neo-Arabic רא for "to see" (e.g. 2:12) while QL sticks with the Classical נְצַֹר throughout.
- QL translates the Hebrew כִּי "because" as לָאייַן.

The rest is forthcoming in the form of a paper, hopefully soon. Ah well, at least I don't have to go far for my new year's resolutions. Boldog új évet, everybody!

[1] Sumikazu Yoda, "'Sifflant' and 'chuintant' in the Arabic dialect of the Jews of Gabes (south Tunisia)",
Zeitschrift für Arabische Linguistik 46 (2006): 7-25


John Cowan said...

Nice new look!

When I looked at those "crowns of dots", I immediately read them as /a/, and thought of the conventional vocalization of acronymic names like RaMBaM and MaHaRaL (rabbi and roboticist). But then I did a double-take, and thought, "What I am thinking? That's not the vowel sign for /a/ at all."

At which point I did a triple-take, and realized that I read them as /a/ because they are Tengwar vowel signs for /a/!

Mo said...

It's possible that the author chose to write the date with ה even though the verse uses א in order to reach the right number, as none of the other letters in the phrase (ידי) can be used to add up to 5...

bulbul said...


possible, but unlikely. That would be messing with the Holy Writ and that's not just done, at least not deliberately.

Unknown said...

asics,asics israel,asics shoes,asics running shoes,asics israel,asics gel,asics running,asics gel nimbus,asics gel kayano
ralph lauren uk
michael kors outlet
christian louboutin uk
nike air max 90
mlb jerseys
nfl jersey wholesale
swarovski outlet
michael kors handbags
rolex watches
nike free run
nba jerseys
nike trainers uk
ugg boots
ray ban sunglasses
rolex watches
mont blanc pens
nike uk store
coach outlet clearance
nike blazer pas cher
michael kors uk
mulberry handbags
vans outlet
ralph lauren outlet
nike store uk
cheap oakley sunglasses
canada goose jackets
nhl jerseys
louis vuitton pas cher
kobe shoes
coach outlet
nba jerseys
air max 90
nike foamposite
ralph lauren pas cher

Unknown said...

michael kors handbags clearance
green bay packers jerseys
hugo boss outlet
michael kors handbags cheap
tennessee titans jersey
ray ban sunglasses uk
pandora outlet
bengals jersey
tiffany and co
michael kors uk

chenmeinv0 said...

roshe run flyknit
cheap ugg boots
ugg australia
instyler max
adidas yeezy 350
ralph lauren
ralph lauren polo
adidas nmd runner
michael kors handbags
pittsburgh steelers jerseys

Unknown said...

true religion jeans
coach outlet
gucci outlet
yeezy boost 350
gucci handbags
parada handbags
michael kors canada
air jordan uk
marc jacobs handbags

theteamblogger said...

tutuapp apk
tutu app apk
tutu app download
tutuapp apk download

TutuApp gained its fame when it released the modded version of Pokemon Go ... They are available for free download.

yanmaneee said...

balenciaga sneakers
lebron shoes
kyrie irving shoes
kd 11 shoes
irving shoes
moncler outlet
hermes birkin
converse shoes

xiumeikahle said...

CASINO, LLC. VICIOS, LA | 718-369-5598
CASINO, LLC. VICIOS, LA | 718-369-5598. 청주 출장샵 Company 광양 출장마사지 Description. 경상북도 출장샵 Company Description. Company Description: Cancun, Louisiana, United States of America 서산 출장샵 is located in Vermilion, 안동 출장마사지

Anonymous said...

navigate to this website o3s33p5v26 louis vuitton fake replica bags joy g0h96n4n59 replica bags in gaffar market his response t6i79i6f71 7a replica bags philippines replica bags ru check here s7h47z8g26 replica bags india

Anonymous said...

o4p59h8m39 b6q46l3f86 m5b01a7w47 a2r86k6n32 y2f16b0e09 w8k89s7i11

Driveway Sealing Indianapolis said...

Very informative blog related to my business! Keep posting similar informative articles!!