Rhotics in Hebrew and Aramaic

[RCK]

I was wondering what you would think of the idea of Hebrew being a (quasi-)non-rhotic language in comparison to Aramaic, as the letter ר is dropped in the Hebrew forms of such Aramaic words as שרביט and כרסא to become the Hebrew words שבט and כסא. Does this make sense? Is there perhaps another explanation for the phenomenon of the disappearing ר? There's tons of words  with a disappearing ר like this including דרמשק/דמשק מרכלית/מכלית עריסה/עיסה and זוטא /זוטרא. There are probably more.

Shalom & Kol Tuv

Reuven Chaim Klein

Beitar Illit, Israel

Author of: God versus Gods & Lashon HaKodesh

ORCiD | LinkedIN | Google Scholar | Amazon

[Andrey Rozenberg]

I was also interested in this question some time ago. Cannot quickly find the original sources, but I believe it's the other way around: insertion of r in later stages of Aramaic as evidenced by the etymology of כורסא < Akkadian kussu and דרמשק ~ dmśk in Old Aramaic, see also similar strictly r-less forms in Arabic and Amarna and even Egyptian. The conditioning for this phenomenon is not clear though.

--
You received this message because you are subscribed to the Google Groups "Jewish Languages" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jewish-languag...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jewish-languages/CALpJrLYid9snAMa521wQRcyJng-S4D7XPGDnKTUDD7iYuA-v%3DQ%40mail.gmail.com.

[RCK]

Another scholar pointed out that we also find the opposite with the Hebrew אמר turning in to the Aramaic אימא יימא by dropping the ר. If you do end up finding the original sources your compiled on this, I'd love to hear about them. In cognates of these words in Proto-Semitic, would the r have already been there?

Shalom & Kol Tuv

Reuven Chaim Klein

Beitar Illit, Israel

Author of: God versus Gods & Lashon HaKodesh

ORCiD | LinkedIN | Google Scholar | Amazon

On Tue, Mar 21, 2023 at 8:32 PM Andrey Rozenberg <aleph...@gmail.com> wrote:

I was also interested in this question some time ago. Cannot quickly find the original sources, but I believe it's the other way around: insertion of r in later stages of Aramaic as evidenced by the etymology of כורסא < Akkadian kussu and דרמשק ~ dmśk in Old Aramaic, see also similar strictly r-less forms in Arabic and Amarna and even Egyptian. The conditioning for this phenomenon is not clear though.

On 21/03/2023 14:53, RCK wrote:

I was wondering what you would think of the idea of Hebrew being a (quasi-)non-rhotic language in comparison to Aramaic, as the letter ר is dropped in the Hebrew forms of such Aramaic words as שרביט and כרסא to become the Hebrew words שבט and כסא. Does this make sense? Is there perhaps another explanation for the phenomenon of the disappearing ר? There's tons of words  with a disappearing ר like this including דרמשק/דמשק מרכלית/מכלית עריסה/עיסה and זוטא /זוטרא. There are probably more.

Shalom & Kol Tuv

Reuven Chaim Klein

Beitar Illit, Israel

Author of: God versus Gods & Lashon HaKodesh

ORCiD | LinkedIN | Google Scholar | Amazon

--

[Andrey Rozenberg]

OK, so derivations of forms with reish from forms w/o it are provided in Jastrow's, Klein's and Sokolow's dictionaries - Jastrow and Klein are available from Sefaria.
E.g. Klein has an extensive elaboration on the question with examples here: https://www.sefaria.org.il/Klein_Dictionary%2C_%D7%A8?ven=Carta_Jerusalem%3B_1st_edition,_1987&lang=bi

What follows is the earliest treatment of the question that appears in Julius Fürst's Lehrgebäude der aramäischen Idiome... (1835), see https://archive.org/details/lehrgebaudederar00furs
As the dictionaries and other later sources, he treats it as a purely intra-Aramaic phenomenon of epenthesis but doesn't mention ד(ר)משק for some reason (I hope the text is readable, force LTR text direction if problems arise):

S. 60, Die epenthetisclien Liquiden:

Die übrigen Liquidae ל und ר, von welchen das ר am häufigsten ist z.B.
שַׁבִּיטָא = שַׁרְבִּיטָא (so viel als שֵׁבֶט Scepter);
כָּרְסֵא Thron, syr. ܟܽܘܪܣܝܳܐ für ࠊࠓࠎࠇ ;כֻּסֵּא, כּוּרְסֵא im Sam. = כַּסֵא ࠊࠎࠄ;
גַּמִּידָא = גַּרְמִידָא d. hebr. גׂמֶד;
סַרְבֵּל, כַּרְסֵם, קַרְדֵם für סַבֵּל, כַּסֵם, קַדֵם;
כּוּרְמִיזָא Faust (nicht Stab, Ezech. 21, 18.) von קְמַץ = כְּמַז zusammenralfen, ballen, (wie אֶגְרׂף von גָּרַף, für welches es steht), nicht zusammenhängend mit גׂמֶד.

Es werden durch dergleichen Epenthetica ganz natürlich, wenn man die eingeschobenen Buchstaben als Bildungsconsonanten betrachtet, viele Plurilittera gebildet, die eben nur insofern diesen Namen verdienen, als sie einer besondern Verbalform angehören. Das ל kommt als littera epenthetica seltner vor, erscheint aber doch nach dem Principe der Euphonie, nämlich nur die Verdoppelung einer Muta zu hindern, z. B. der wahrscheinlich aus Indien samt seinem Namen verbreitete Pfeffer: sanskr. pippali, heisst, um das p nicht zu verdoppeln, פִּלְפָּלִי pilpali, neben פִּרְפָּלִי griechisch πέπερι, woraus das Verbum פַּלְפֵּל mit Pfeffer bestreuen, würzen, streuen überhaupt, herstammt. *)

Auf diese Weise gebraucht man die liquidae als Einschiebungen zur Bildung der quadrilittera z. B.
כַּרְבֵּל binden, gürten = כַּבֵּל;
שַׁרְנֵק erwürgen = שַׁנֵּק so viel als חַנֵּק **);
קַרְסֵם abfressen = קַסֵם;
בַּלְסֵם würzen;
כַּרְסֵם Düfte verbreiten;
פַּלְדּס ausdehnen,
גַּנְדַּר wälzen,
aus גְּדַר ,(פְּטַשׁ) פְּדַס ,פְּסַם ,בְּסַם u. s. w. die gewöhnlich Paelformen sind. Oft werden dreiconsonantige Stämme durch solche Einschiebsel gebildet, die einen leichten flüssigen Consonanten deswegen einschieben, wie wir in der Folge bei der Stammbildung sehen werden.

The most convincing thing here is the existence of doublets and the fact that the inserted reish is in most cases associated with extension of otherwise triliteral roots.

[Andrey Rozenberg]

I must say some of the Hebrew forms you mention might require explanations separate from epenthesis in Aramaic:

* Hebrew זוטא is indeed mentioned as a shortened form of Aramaic זוטרא: https://www.sefaria.org.il/Klein_Dictionary%2C_%D7%96%D7%95%D6%BC%D7%98%D6%B8%D7%90.1?ven=Carta_Jerusalem%3B_1st_edition,_1987&lang=bi&with=all&lang2=en, this might be indeed an example of a reish drop in Hebrew, but I'd remain skeptical
* the relationship of the מַכֹּלֶת to מַרְכֹּלֶת is less than certain (see discussion in https://he.wiktionary.org/wiki/%D7%9E%D7%9B%D7%9C%D7%AA), Klein sees no connection between the two and derives one from אכ"ל and other one from from רכ"ל
* עיסה/עריסה - or more broadly, roots ער"ס/עס"ס - I'd very tentatively hypothesize that both are in fact extensions of an ancestral bi-consonantal root which is otherwise a very common phenomenon

[Andrey Rozenberg]

Sorry, I didn't realize at first that זוטא is well attested in Aramaic sources and the terminal aleph also betrays its Aramaic origin.
See Sokoloff, A dictionary of Jewish Babylonian Aramaic... p. 403 who also derives it from זוטר
This does not explain how the reish disappeared but at least it must have happened in Aramaic.

[RCK]

Very interesting! I asked this question to several experts who are not on this email list and I wanted to share some of their answers because each one adds another little point:

Dr. Geoffrey Kahn:

In some of the words you mention the /r/ seems to have developed secondarily by some kind of dissimilation of a geminate. This applies, for example, to כרסא and דרמשק. This dissimilation was somewhat more prevalent in Aramaic than in Hebrew. In some cases an /n/ appears, e.g. madda` > manda`. Both /r/ and /n/ are sonorant consonants.

Dr. Jared Greenblatt:

There are definitely some  cases of loss of /r/, although although "disappearing r" is sometimesa case of assimilation, and some r sounds in Aramaic are a result of dissimilation.  One case that springs to mind is Aramaix יימא, אימא etc for אמר ...
But I think to call it non-rhotic, even quasi, is to overstate it a bit.  You'd usually expect some regular predictable r loss for that....
Still, it's a nice point, especially if r was pronounced sometimes as a type of uvular approximant even in ancient times, in some contexts... Geoffrey Kahn's presents evidence of that from alKirkesani....I would think that approximants  would be less phonetically stable that a tapped r.....

Dr. Bill Schniedewind:

I would say that the issues with the resh are an inner Aramaic development, and not related to proto-West Semitic.  In other words, Hebrew without the resh in these forms is more original.  We have good early comparative linguistic forms in the Amarna letters as well as early northwest Semitic inscriptions. So, the resh didn’t disappear, it emerged. But it’s always interesting to explore different ideas. 
{RCK: Interesting. And what would you say would cause the REISH to randomly appear in certain words?} Dissimulation.  Aramaic has a number of other features like the prosthetic aleph, which are not derived from Proto-West Semitic. Hebrew is more original.  Also, Aramaic is not a Canaanite language, so it has linguistic developments distinct from the Canaanite branch (Hebrew, Phoenician, Moabite, etc).

Carlo Meloni (who wrote a dissertation rhotics in Hebrew):

Your question is intriguing, but I believe that reaching a definite conclusion is not possible due to the complexity of the situation when considering other Semitic languages. Firstly, it should be noted that the Hebrew word כסא contains a dagesh in the ס, which indicates its pronunciation as "kisse". This could be a sign of the assimilation of the rhotic to the preceding /s/. However, when examining cognates in other Semitic languages, we find that the Akkadian equivalent is "kussûm", while the Arabic equivalent is "kursiyy". It is known that the Arabic word is a loan word from Aramaic, and it is possible that the Hebrew word is also derived from Akkadian. Nevertheless, it remains uncertain whether the original word contained a rhotic that was later assimilated, or if it contained a geminated consonant that later dissimilated to /rs/. The latter process has been observed in a few cases in Aramaic, typically with /n/, which is also a liquid consonant. An example of this is "arnona", which is derived from Latin "annona".
With regards to the word שבט, its Akkadian equivalent is "šabbiṭum", its Geez equivalent is "səbṭ", and its Arabic equivalent is "sibṭ". Most of the cognates appear to lack the rhotic, and there is even a secondary form in Aramaic, שבטא.
Lastly, a similar process can be observed with the word דמשק. The word contains a dagesh in the מ, indicating its pronunciation as "dammeseq". In old Aramaic inscriptions, the word is written similarly as דמסק, but in later dialects, we find דרמשק. This suggests that some sort of dissimilation process may be occurring.

Shalom & Kol Tuv

Reuven Chaim Klein

Beitar Illit, Israel

Author of: God versus Gods & Lashon HaKodesh

ORCiD | LinkedIN | Google Scholar | Amazon