Land entitlement and confusing grammar
In the parasha read this past shabbat, Parashat Pinchas, I noted a sentence fragment that is grammatically troublesome (Num. 26:55):
אַךְ־בְּגוֹרָל יֵחָלֵק אֶת־הָאָרֶץ; לִשְׁמוֹת מַטּוֹת־אֲבֹתָם יִנְחָלוּ׃
I can only translate it approximately while retaining some of the problematic detail:
Moreover, by lot will be divided (ACC) the land; according to the names of the tribes of their fathers will they inherit.
The first stark problem here is that יֵחָלֵק (will be divided) is a passive (נפעל) form, and so should not take a direct object as is marked by the אֶת (ACC) particle. In the present case, the verb seems to take “the land” as exactly this direct object it shouldn’t have. Semantically, one should expect “the land” to be the subject of the passive verb, and not to be marked as an object.
Moreover, it cannot be the subject either, because “the land” is a feminine noun, and the verb requires a masculine subject. I should say that there are a few cases where אֶרֶץ clearly is masculine in the bible (Gen. 13:6; Isa. 9:18 for instance), but this is far from the general rule for what is a fairly common word.
If only the verb was תֵּחָלֵק (for feminine subject) and אֶת was dropped, the verse would read perfectly well. And this is what Onkelos gives in his translation:
בְרַם בְעַדבָא תִתפְלַג אַרְעָא לִשׁמָהָת שִׁבטֵי אֲהָבָתְהוֹן יַחסְנוּן׃
Well, that’s according to some manuscripts—those preferred by Bar Ilan University’s Mikraot Gedolot Haketer project, as well as Merkaz HaRav Kook’s Torat Hayim. Other editions, such as Artscroll’s popular humash, do include an object (accusative case) marker, and write “תִתפְלַג יָת אַרְעָא” and thus follow the grammatical anomaly found in the Hebrew Masoretic text. According to CAL, Targum Neofiti has the inclusion of יָת as a marginal variant, while Pseudo-Jonathan and Peshitta do not seem to be found mentioning it. So generally the Targumim fix the problem, which only further illustrates its presence!
Ibn Ezra also possibly takes a step to solving the issue, as one might expect him to. On “יחלק את הארץ” we find his words:
והטעם כי בגורל יחלק הארץ למטות.
Notably here he removes the object marker את that sees to not belong, but strangely he does not change the gender of the verb’s subject. The Torat Hayim suggests that this comment is really about the second part of the verse anyway, explaining that “according to the names of the tribes of their fathers” really means “to the tribes”. So it is possible, according to that note, that Ibn Ezra is not even attempting to correct the grammatical problems in our verse, although this would be surprising.
There is another alternative, in changing the vocalisation of the verb: reading יְחֵלֵּק instead of the Masoretic יֵחָלֵק changes passive “it/he will be divided” to “it/he will divide”. (This could also be a possible reading of Ibn Ezra.) In this way, the object marker is appropriate, and the verb need not agree with a feminine subject like “land”. Nonetheless, there is no appropriate masculine subject in the context that could divide the land. Moses, for instance, is spoken to in second-person (not third) in verse 54. The “lot” could be a potential subject, but then it should not have the preposition בְּ-, and I am not sure if such an abstract noun is allowed to be the agent of division semantically in Hebrew. So even this emendation to the Masoretic vocalisation does not solve teh issue.
So I would like any help I can get on what is going on here. Does the general rule that passives can’t take objects also fail to apply elsewhere in the bible? It should not be too hard to find out with a clever program, or better, an online parsing guiden and a program. Any scholarly or traditional references? Any suggestions otherwise? Help!
Waltke-O’Connor (§23.2.1b) note that ×�ת can also indicate the subject with a niph’al verb. The source they quote is Exodus 21:28, although Gesenius (§121a, b) notes other examples of this with the niph’al (such as Gen 4:18) and also with passive verbs in general (eg: Gen 27:42). A grammatical search on Accordance 6.9.1 yielded about 40 hits (for niph’al + object marker, within one to two words). Others include Gen 17:5, 17:11; Num 7:10; Jos 10:25.
Comment by Simon — 8 July, 2007 @ 7:16 pm
I thought that was the most likely solution, but still then we have ארץ being masculine, and so we have two rare features within the one verse. And to my great annoyance, Ibn Ezra is more than happy for it to be masculine, and doesn’t say a thing…
And I know that I never know what to call את but I was thinking in terms of passives generally which in syntactic terms block the
assignment of the accusative case.
I have to say I’ve never used Waltke-O’Connor… And that I don’t have a grammar to hand, and that Accordance sounds neat. I also know that Gesenius used to be online (I think I quoted it once in a comment on your Blogger), but that I’ve since lost it.
Having now looked at your examples, I now have more to say:
The Ex. 21:28 example is possibly the most similar semantically to what we have here. Still, there is a clear eater in that case, and no clear divider here.
In Gen. 4:18, 27:42, 17:5, each of these verbs have two thematic relations given, and so את is used to distinguish the patient from the agent, etc. In the first two examples, the agent is given with the preposition ל attached. The third example has no thematic actor/agent, but is a ditransitive verb, and so again את serves a role in identifying the syntactic subject of the passive verb. I would say in these cases that את cannot be considered the accusative case marker, but is marking a particular thematic relation as distinguised from another present. In our case, there is no other thematic role given: only הארץ.
Gen. 17:11 is an interesting case. At least at one point I had assumed that this is a variant root of the word, with root letters nun.mem.lamed. I remember looking up this root in a lexicon, such as BDB, and it suggesting so too. So at most this case is disputed.
Num. 7:10 is just strange, because “ביום המשח הוא” is also ungrammatical,
and somehow having a pronoun there (maybe along with the infinitive) seems to require the objective form of the pronoun. I don’t understand what to do here…
Jos. 10:25 is also to be distinguished because נלחם
is an active-style nif’al. It is also possible that את was
originally used with this verb to signify “with” and not the object
marker… but here where we have אותו it seems
more clearly to have been derived as the object (although if I recall,
there are other cases of אותו meaning ‘with’).
Thanks again: I learnt something, but I’m still not feeling completely
satisfied. Other cases more closely akin to this one would be more
appeasing. Are there other solutions?
Comment by Joel — 8 July, 2007 @ 7:23 pm
These passives with את seem to be a whole lot more frequent than I imagined…
It seems we have a similar example to the verse mentioned above in the coming parasha:
יֻתַּן אֶת-הָאָרֶץ הַזֹּאת לַעֲבָדֶיךָ (Num. 32:5)
Here Ibn Ezra lists verses with הארץ as masculine, including that from above, Num. 26:55.
I wonder whether anyone has studied the distribution of masculine reference to feminine nouns like this…
Comment by Joel — 9 July, 2007 @ 11:39 am
Having another thought. Maybe the insertion of את loses the requirement of gender and number agreement; the verb becomes subjectless grammatically, and the verb defaults to 3m sng. This would explain why we have the masculine passives in both these Numbers verses.
Also, the ostensive subject of וַיֻּגַּד in Gen. 27:42 is דִּבְרֵי עֵשָׂו, a plural, for which the verb is not properly inflected.
Or maybe this has nothing to do with having את and there is a general lack of gender agreement associated with passives? Numerous times in the bible we have מַצּוֹת יֵאָכֵל. Should we not expect מַצּוֹת תֵּאָכֵלְנָה or something?
Comment by Joel — 9 July, 2007 @ 11:57 am
Having spoken the other night to Prof. Chaim Cohen, he is under the opinion that:
1) When the verb precedes the subject it sometimes appears masculine singular irrespective of the number and gender of its subject. In Arabic and Syriac, this is the rule. In Hebrew it is not so common, but it can happen nonetheless and he prefers that to the suggestion that ארץ is here functioning as a masculine noun;
2) The את (DDOM) is here being utilised as an indicator of the casus pendens, that is the nominative absolute which is being highlighted morphologically as the primary element of the sentence (the subject). He argued this on the basis of parallels in Akkadian, although a quick look through both Gesenius and Waltke-O’Connor supports the idea of a casus pendens in Hebrew as well.
Comment by Simon Holloway — 11 July, 2007 @ 3:15 am
By the way, I’m having difficulty finding examples of 1) with Accordance. I can turn up thousands of instances where a masculine singular verb is followed by a feminine singular, feminine plural or masculine plural noun, but I have yet to work out how to sift out all the occurrences where that noun is the object. I’m working on it…
Comment by Simon Holloway — 11 July, 2007 @ 3:19 am
Okay, that was a bit fiddly but I think I got it. Here are the instances of a masculine singular verb preceding a feminine singular noun:
1 Sam 13:22 (נמצא חרב)
Isa 9:18 (נעתם ארץ)
Isa 16:10 (ונאסף שמחה)
Jer 8:16 (נשמע נחרת סוסיו)
Ezek 23:29 (ונגלה ערות זנוניך)
Ezek 28:15 (נמצא עולתה)
Masculine singular verb preceding a feminine plural noun:
Ezek 32:25 (נתן חתיתם)
Job 42:15 (נמצא נשים)
Masculine singular verb preceding a masculine plural noun:
2 Sam 21:10 (נתך מים)
Dan 1:15 (נראה מראיהם)
1 Chr 26:6 (נולד בנים)
Obviously this is not a conclusive list. I limited my searches to niph’als that appeared within one space from the verb that governed them. A search that takes other binyanim into account, and allows for greater difference between words, will yield too many results to conveniently transpose. Anyway, let me know what you think.
Comment by Simon Holloway — 11 July, 2007 @ 3:42 am
Wonderful! Thanks for the great research. I’m glad I noticed the difficulties, even if I had seen them before numerous times and not noticed them then. It is much more appealing to have it being a feature of the passive verb than a feature of a gender-changing noun.
Are you able to see if there are any examples wherein the verb follows the noun, for interest?
Comment by Joel — 11 July, 2007 @ 8:52 am
An example of which we see in מַצּוֹת יֵאָכֵל presumably?
Comment by Joel — 11 July, 2007 @ 8:55 am
Difficult… For some reason it is even harder to isolate examples where the noun precedes the verb. In the absence of a proper methodology for doing so, here are some examples where the verb is masculine singular and the noun is feminine singular:
Lev 19:20 (חפשה לא־נתן)
Num 11:12 (האדמה אשר נשבעת)
Num 14:16; Deut 6:10, 26:3 and 31:7 (אל־הארץ אשר נשבע)
Deut 1:8, 6:23 and 8:1; Jos 5:6 (את הארץ אשר נשבע)
Deut 7:8 (את השבעה אשר נשבע)
Deut 7:13, 11:9, 11:21, 28:11 and 30:20 (על האדמה אשר נשבע)
Jos 21:43 (את־כל־הארץ אשר נשבע)
Mal 2:6 (ועולה לא נמצא)
The above examples are all listed to niphal perfectives. Once I change that I can get (many!) more responses. Changing it for the example you gave me, I found a few responses where there was a feminine plural noun and a masculine singular niphal imperfect. Those are:
Ex 13:7, Num 28:17, Ezek 45:21 (מצות יאכל)
Prov 14:17 (מזמות ישנא)
It is also worth noting that in Lev 6:9 we find מצות תאכל.
I’d make a more comprehensive list of occurrences with different types of nouns and different types of verbs but there are some 47 to sift through for imperfect singular + feminine singular noun alone, and it would take too long to make my way through all combinations. It looks interesting! Let’s hope it proves of relevance to Nehemiah - then I might have an opportunity to go back and do it properly. Hope this helped.
Comment by Simon Holloway — 12 July, 2007 @ 2:27 am
An interesting collection of citations… but not all apply. מצות יאכל seems to be a strange case that we might need to work with separately as well. In Prov. 14:17, מזמות is actually adjoined to the previous word, and there is no gender-disagreement between איש מזמות and ישנא.
And as for the various נשבע cases, the preceding noun is not actually its subject. What this preceding noun is described as in syntactic terms, I am not sure. Nonetheless, the grammatical subjects in all these cases are the person who is making the oath. As should be obvious from the first example you give. So there need be no tense agreement with what is oathed. I don’t know whether we are to understand the verb as a reflexive.
חפשה לא־נתן and ועולה לא נמצא maybe say something. Need we have the restriction of noun before verb? I know there are too many cases to pull out from (or even each for in) accordance.
Comment by Joel — 14 July, 2007 @ 11:29 am