Post: In parshat Shlach, Bemidbar 14:
|24. But as for My servant Caleb, since he was possessed by another spirit, and he followed Me, I will bring him to the land to which he came, and his descendants will drive it[s inhabitants] out.||כד. וְעַבְדִּי כָלֵב עֵקֶב הָיְתָה רוּחַ אַחֶרֶת עִמּוֹ וַיְמַלֵּא אַחֲרָי וַהֲבִיאֹתִיו אֶל הָאָרֶץ אֲשֶׁר בָּא שָׁמָּה וְזַרְעוֹ יוֹרִשֶׁנָּה:|
The Judaica Press takes pains to translate in accordance with Rashi. Why 'drive out', which requires an insertion of an implicit '[its inhabitants]'? Because Rashi says:
|יורשנה: כתרגומו יתרכינה, יורישו את הענקים ואת העם אשר בה, ואין לתרגמו יירתינה, אלא במקום יירשנה:|
That is, Rashi makes a diyuk in Onkelos in that it states יתרכינה rather than יירתינה. And that would mean 'drive out' rather than 'inherit it'. (These would seem to be different roots, with תרכ defined by Jastrow [pg 1699] as 'to drive out, divorce, banish.) It seems quite possible that Rashi is aware of competing girsaot. Yet that this competing girsa shares so much in common with the Hebrew root does make one suspect that there is a taut sofer in play here.
The Teimanim indeed have a different girsa in Onkelos:
|יד,כד וְעַבְדִּי כָלֵב, עֵקֶב הָיְתָה רוּחַ אַחֶרֶת עִמּוֹ, וַיְמַלֵּא, אַחֲרָי--וַהֲבִיאֹתִיו, אֶל-הָאָרֶץ אֲשֶׁר-בָּא שָׁמָּה, וְזַרְעוֹ, יוֹרִשֶׁנָּה.||וְעַבְדִּי כָּלֵב, חֲלָף דַּהֲוָת רוּחַ אֻחְרִי עִמֵּיהּ, וְאַשְׁלֵים, בָּתַר דַּחְלְתִי--וְאַעֵילִנֵּיהּ, לְאַרְעָא דְּעָאל לְתַמָּן, וּבְנוֹהִי, יוֹרְתִנַּהּ.|
The word יוֹרְתִנַּהּ is not the same as Rashi's stated competing girsa of יירשנה, so perhaps his objection on grammatical grounds would fall away as well.
Here is what Shadal has to say, in Ohev Ger:
The majority of texts are against Rashi's girsa of Onkelos, and even if it is precise grammatically, Shadal does not believe that this reflects Onkelos' intent. For if so, Onkelos would have inserted a word or two to target the inhabitants. As it stands, it does not make sense to say that one will drive out the land!
I can add an interesting point to this. If we look at the oldest Rashi manuscript I have access to, from Munich, 1233, we find:
That is, in the main body of text, Rashi just says כתרגומו, but he does NOT specify just what this Targum of Onkelos IS. On the margin, if I can make it out well enough, someone puts in what the Targum of Onkelos should be to justify this statement of Rashi.
So, perhaps Rashi was not suggesting this girsa in Onkelos at all. Instead, he had in his Onkelos a word with the same root meaning inherit, but with a different mishkal, a different verb pattern. For example, as the Teimanim have it, יוֹרְתִנַּהּ. And he did not intend to select a different root, but rather was making a syntactic point. It was not יירתינה, which is (let us say) the kal, but rather something more forceful and causative. How about that they will dispossess the residents of the land. They will take possession, rather than simply inheriting it. I can imagine ways in which this works well even with Shadal's objections.
With all that said, see how the printer (divrei hamadfis) in HaMaamar handily responds to Shadal:
That is, there is a parallel in the Targum to sefer Yehoshua, in which one can apply that verb ותתרכון to a city, and not explicitly to the residents of the city; this, then, would be the natural pattern of this phrasing. I would note it would also work somewhat against my suggestion (based on the old manuscript of Rashi) that Rashi did not indeed have this nusach of Onkelos before him, for the nusach would not have the same difficulty. Still, I think it plausible that Rashi could have worked from a different girsa, in accordance with the majority, and that this was just the suggestion of some scribe in the time of the Rishonim.
Another, slightly tangential, point is that an earlier translation in this sidra by Onkelos gives us a third or fourth options. On pasuk 12, we had:
|יד,יב אַכֶּנּוּ בַדֶּבֶר, וְאוֹרִשֶׁנּוּ; וְאֶעֱשֶׂה, אֹתְךָ, לְגוֹי-גָּדוֹל וְעָצוּם, מִמֶּנּוּ.||אֶמְחֵינוּן בְּמוֹתָא, וַאֲשֵׁיצֵינוּן; וְאַעֲבֵיד, יָתָךְ, לְעַם רַב וְתַקִּיף, מִנְּהוֹן.|
It might be possible to select a destructive connotation, as above, and apply it here.