Korach: One Donkey, or One Person's Donkey

Recently, I began penning a Jewish Link article on masechet Shevuot. In the end, I decided to suppress it, because it was a bit complex for my audience, so I went with a completely different article.

Here is the first portion of the article, which addressed a verse in the sidra of Balak, before it then transitioned to discuss the meaning of oaths in Shevuot 36.

My discussion continues after this lengthy quote.

Rabbi Meir's Existential Quantifier

As part of my Digital Humanities research, I’ve been exploring representing statements using first order logic (FOL), for use in reasoning-based Artificial Intelligence systems. Looking over masechet Shevuot, it struck me that many of the ambiguous utterances making up oaths, and especially their interpretations by Tannaim, Amoraim and Rishonim, could be well represented using first order logic.

Let’s take a Biblical example, לֹ֠א חֲמ֨וֹר אֶחָ֤ד מֵהֶם֙ נָשָׂ֔אתִי. (Bemidbar 16:15) So said Moshe to Hashem, aggrieved after being accused of self-aggrandizement by Datan and Aviram. Various interpretations / misinterpretations are possible. For instance, does the first אֶחָ֤ד modify מֵהֶם֙, so it’s the donkey of one-of-them? That’s Rashi, לֹא חֲמוֹרוֹ שֶׁל אֶחָד מֵהֶם נָטַלְתִּי. Or, does it modify חֲמ֨וֹר, so it’s a single-donkey of theirs? That’s Ramban, כִּי מֵעוֹלָם לֹא לָקַחְתִּי מֵהֶם אֲפִלּוּ חֲמוֹר אֶחָד. Given the parallel in Moshe’s continuation, וְלֹ֥א הֲרֵעֹ֖תִי אֶת־אַחַ֥ד מֵהֶֽם, “nor have I wronged one of them”, I’d side with Rashi. It could also e.g. mean “I didn’t take just one donkey, but took several donkeys”.

Moshe’s statement could be encoded in FOL as ¬∃x∃y (Person(x) ∧ Donkey(y) ∧ Owns(x,y) ∧ Took(I,y)) ∧ ¬∃z (Person(z) ∧ Wronged(I,z)). The ¬ symbol means “not”; the ∃ is the existential quantifier, meaning “there exists”; the ∧ symbol means “and”. Not shown are the ∨, meaning “or”, as well as ∀ which is the universal quantifier, means “for all”.

Thus, this FOL statement is: It is not the case that there exists x and y, such that x is a person, y is a donkey, x owns y, and I took that donkey y. Also, there does not exist z such that z is a person and I wronged z. Note that person x and person z do not have to be the same person.

OK, so let us examine Rashi, Rashbam, and Chizkuni. Together, of course, with the Eliyahu Munk translation, which needs some serious fixing.

The pasuk, in Bemidbar 16:5:

וַיִּ֤חַר לְמֹשֶׁה֙ מְאֹ֔ד וַיֹּ֙אמֶר֙ אֶל־יְהֹוָ֔ה אַל־תֵּ֖פֶן אֶל־מִנְחָתָ֑ם לֹ֠א חֲמ֨וֹר אֶחָ֤ד מֵהֶם֙ נָשָׂ֔אתִי וְלֹ֥א הֲרֵעֹ֖תִי אֶת־אַחַ֥ד מֵהֶֽם׃

Moses was much aggrieved and he said to יהוה, “Pay no regard to their oblation. I have not taken the ass of any one of them, nor have I wronged any one of them.”

Rashi explains:

לא חמור אחד מהם נשאתי. לֹא חֲמוֹרוֹ שֶׁל אֶחָד מֵהֶם נָטַלְתִּי. אֲפִלּוּ כְּשֶׁהָלַכְתִּי מִמִּדְיָן לְמִצְרַיִם וְהִרְכַּבְתִּי אֶת אִשְׁתִּי וְאֶת בָּנַי עַל הַחֲמוֹר, וְהָיָה לִי לִטֹּל אוֹתוֹ הַחֲמוֹר מִשֶּׁלָּהֶם, לֹא נָטַלְתִּי אֶלָּא מִשֶּׁלִּי; וְתַ"אֻ "שְׁחָרִית", לְשׁוֹן אֲרַמִּי כָּךְ נִקְרֵאת אַנְגַּרְיָא שֶׁל מֶלֶךְ "שַׁחֲוָר":

לא חמור אחד מהם נשאתי means, I have not taken the ass of any one of them; — even when I went from Midian to Egypt and placed my wife and my sons on the ass (Exodus 4:20), and I surely ought afterwards to have taken the price of that ass from their money, yet I took it only from my own (Midrash Tanchuma, Korach 7). The translation given by Onkelos of the word נשאתי is שחרית: in the Aramaic language a forced levy made by the king is so called, viz., שחרור (cf. Rashi on Bava Batra 47a. The translation of Onkelos therefore is: I did not press into my service the ass of any one of them).

Thus, with לֹא חֲמוֹרוֹ שֶׁל אֶחָד מֵהֶם נָטַלְתִּי, Rashi says it is the ass that belongs to one of them. אחד modifies the word מֵהֶם. As I noted above in that article excerpt, look to the end of the selfsame pasuk for such usage.

Rashbam invokes grammar to argue with his grandfather Rashi and writes:

לא חמור אחד מהם נשאתי - אפילו חמור אחד לא נשאתי מהם כשאר משתוררים על העם. ואם כן למה מפשיעים אותי בשררה?! ואם היה נקוד לא חמור אַחַד מהם (בפת"ח) פתרונו היה חמור של אחד מהם כמו: אַחַד העם, וכמו: את אַחַד מהם.

My own translation:

לֹ֠א חֲמ֨וֹר אֶחָ֤ד מֵהֶם֙ נָשָׂ֔אתִי — even a single donkey I have not appropriated from them like other rulers of the people. And if so, why do they accuse me of improperly exercising authority?! And if it [אֶחָ֤ד] were vocalized לא חמור אַחַד מהם, with a patach [אַחַד], then its explanation would be a donkey of single one of them, just like אַחַד העם [in Bereishit 26:10] or like את אַחַד מהם [at the end of this verse].

Note that Eliyahu Munk’s translation is a bit “off”. Besides not quoting the vert end of the verse, he also awkwardly translates the alternative meaning of לא חמור אַחַד מהם. Thus he writes:

לא חמור אחד מהם נשאתי, I have not even appropriated a single donkey of theirs (the entire people) as a form of taxation as do most others rulers from all of their subjects. Seeing that this is so, on what do they base their claim that I behave like a ruler, a despot? If the vowel pattern under the word אחד would not have been “segol” followed by “kametz,” but two successive vowels “patach,” the meaning would be: “I have not taken away a single one of their privately owned asses.” It would have been a construct mode as in Genesis 26,10 אחד העם, i.e. “one of the common people.”

No! I am not sure if this is just awkward translation or if he is missing the point. But he should not put the word “single” where he puts it, because then it still is modifying “asses”. Privately owned, sure, but that is where the word single belongs. As he writes it, you don’t get the sense that it modifies “them”. Instead, Munk seems overly focused on the fact that it is a construct form, thus “a single one of”. But he still seems to think that it is merely grammatical point, and still modifies “ass”.

Next, Chizkuni weighs in. Chizkuni is a very early supercommentary on Rashi, among other aspects of his commentary. He writes (but just look at the Hebrew):

The correct translation is:

חֲמ֨וֹר אֶחָ֤ד — according to the vocalization, it is in its expected import. However, according to Rashi’s commentary, it should have been vocalized like אַחַ֤ד הָעָם֙ [in Bereishit 26:10].

The פירוש כמשמעו is in disagreement with Rashi, and thus means “a single donkey”. If it were as Rashi meant, that it was a donkey of a single one of them, then it should be in the construct form, just like that earlier verse in Bereishit.

What is wrong with Eliyahu Munk’s translation? Here it is again. Besides expanding the actual words, he takes it to refer to the vocalization of the word chamor!

לא חמור אחד מהם נשאתי, “I have not appropriated a single donkey from anyone of them;” this remark by Moses is relevant to verse three, when he and Aaron had been accused of having elevated himself above the people. The vocalization of the word חמור corresponds to the plain meaning of the verse, whereas according to Rashi’s interpretation the word echad should have been achad, “one of,” as in Genesis 26,10.

Of course, Chizkuni was referring to the vocalization of the word echad.