Gen 2:4b - 4:26 is J

The assignment is made chiefly on six grounds:

1. Contradictions with Gen 1 already discussed in a previous post.
2. The duplication of the lineage of Kain, already discussed in a previous post.
3. The use of the divine name of J (2:4, 5, 6, 7, 8, 9, 15, 16, 18, 19, 21, 22,
etc.)
4. The use of JE's nomenclature including
a. אֲדָמָה
2:5, 2:6, 2:7, 2:9, 2:19, 3:17, 3:19, 3:23, 4:2, 4:3, 4:10, 4:11, 4:12, 4:14
b. עֶצֶב
3:16, (twice)
c. טֶרֶם
Gen 2:5 (twice),
d. יָלַד
(4:18 x 3)
e. Anoki (3:10, 4:9)
5. The complete lack of P's words, in contrast to P texts that are saturated with these words.
6. The depiction of God in Anthropomorphic terms which is a J phenomenon.

אֲדָמָה

The word אֲדָמָה is concentrated in JE far more than in P. It appears approximately 5 times in P, approximately 53 times in JE, and approximately 37 times in D and twice in other sources.

The 5 in P are

Gen 1:25 6:20 9:2
Lev 20:24, 20:25

The 53 times in JE are

Gen 2:5, 2:6, 2:7, 2:9, 2:19, 3:17, 3:19, 3:23, 4:2, 4:3, 4:10, 4:11, 4:12, 4:14, 5:29, 6:1, 6:7, 7:4, 7:8, 7:23, 8:8, 8:13, 8:21, 9:20, 12:3, 19:25, 28:14, 28:15, 47:18, 47:19 (4 times), 47:20, 47:22 (twice), 47:23 (twice), 47:26 (twice)

Ex 3:5, 8:17, 10:6, 20:20, 23:19, 32:12, 33:16, 34:26,

Nu. 11:12, 12:3, 16:30, 18:15, 32:11,

Is there a traditional response to this? One possible response would be that the word אֲדָמָה is more common in narrative and JE is weighted towards narrative. There are several problems with this point of view. Most notably, the synonym for אֲדָמָה, Eretz, is not weighted towards JE as intensely is אֲדָמָה is. Eretz comes up approximately 842 in Chumash, with approximately 306 times in P, 185 times in D, 349 times in JE and 2 times in other sources. If the word “land” was indeed expected to be more weighted towards JE, we would have expected this to be for both אֲדָמָה and Eretz.

Another possible answer is to distinguish between the meaning of Eretz and אֲדָמָה. What comes to mind first is that אֲדָמָה means soil and Eretz means land, as in country. However, when I do a quick perusal through the p’sukim, I am not sure that that distinction holds. I’ll have to think about it more I guess.

עֶצֶב

The word עֶצֶב is limited to JE. It appears only 7 times though, so that can be explained by coincidence. The list is as follows

Gen. 3:16, (twice) 5:29, 6:6, 34:7 45:5, and the 7th escapes me at the moment.

Parsha Insights Miketz 2006

Genesis 42:

כו וַיִּשְׂאוּ אֶת-שִׁבְרָם, עַל-חֲמֹרֵיהֶם; וַיֵּלְכוּ, מִשָּׁם. כז וַיִּפְתַּח הָאֶחָד אֶת-שַׂקּוֹ, לָתֵת מִסְפּוֹא לַחֲמֹרוֹ
בַּמָּלוֹן; וַיַּרְא, אֶת-כַּסְפּוֹ, וְהִנֵּה-הוּא, בְּפִי אַמְתַּחְתּוֹ. כח וַיֹּאמֶר אֶל-אֶחָיו הוּשַׁב כַּסְפִּי, וְגַם הִנֵּה בְאַמְתַּחְתִּי; וַיֵּצֵא לִבָּם, וַיֶּחֶרְדוּ אִישׁ אֶל-אָחִיו לֵאמֹר, מַה-זֹּאת עָשָׂה אֱלֹהִים, לָנוּ. כט וַיָּבֹאוּ אֶל-יַעֲקֹב אֲבִיהֶם, אַרְצָה כְּנָעַן; וַיַּגִּידוּ לוֹ, אֵת כָּל-הַקֹּרֹת אֹתָם לֵאמֹר.
...........
לה וַיְהִי, הֵם מְרִיקִים שַׂקֵּיהֶם, וְהִנֵּה-אִישׁ צְרוֹר-כַּסְפּוֹ, בְּשַׂקּוֹ; וַיִּרְאוּ אֶת-צְרֹרוֹת כַּסְפֵּיהֶם, הֵמָּה וַאֲבִיהֶם--וַיִּירָאוּ

Since Yosef’s brothers had already discovered at the lodge that their money had been returned, why were they so surprised to see it at their father’s house? One answer is that at the lodge only one brother opened his sack. While that fits into the text, the story still reads quite strangely. A better answer is that these verses are from different hands. Perhaps though, as support to the first approach we can note the emphasis on the singular in the lodge, וַיִּפְתַּח הָאֶחָד אֶת-שַׂקּוֹ vs the emphasis in Yaakov’s house on the fact that each had the problem, אִישׁ צְרוֹר-כַּסְפּוֹ, בְּשַׂקּו. Still, it really does not read well. Suggestions appreciated.

A guten shabbos and Channuka to all.

טֶרֶם

Scholars say that the word טֶרֶם does not appear in P. They are, however, incorrect, as it appears 17 times in the Pentateuch, once in P, 15 times in JE, and once that can be viewed as D or as coming from another source. The 15 occurrences in JE are
Gen 2:5 (twice), 19:4 24:15, 24:45, 27:4, 27:33, 37:18, 41:50, 45:28
Ex. 1:19, 9:30, 10:7, 12:34
Nu. 11:33

The once occurrence in P is Lev 14:36, and the final time, D or other, is Deut 31:21.

Since the word is not common, this whole phenomenon may be chalked up to coincidence. I hope to, in the near future, present my analysis from a statistical standpoint of the coincidental nature of certain words according to the DH. For now, I’ll just note that the odds of flipping a coin 16 times and getting the same result 15 or more times of the 16 times is approximately equal to 1 in 2,000. The relevance of that point will be considered in future posts.

But, coincidence is not needed here. One may make a case that the word טֶרֶם is more frequent in narrative than in law and JE is weighted heavily towards the narrative section of the Bible. That weighting is not complete, as P does have narrative sections, but more on that again in future posts. Moreover, it’s not entirely clear to me that טֶרֶם should be limited to narrative, but I can not offer any hard evidence on that question.

Parsha Insights Vayeishev 2006

I will have limited internet access this week, so I’ve decided to put up this week’s parsha insights up early.

וַיֵּלֶךְ יוֹסֵף אַחַר אֶחָיו, וַיִּמְצָאֵם בְּדֹתָן. יח וַיִּרְאוּ אֹתוֹ, מֵרָחֹק; וּבְטֶרֶם יִקְרַב אֲלֵיהֶם, וַיִּתְנַכְּלוּ אֹתוֹ לַהֲמִיתוֹ. יט וַיֹּאמְרוּ, אִישׁ אֶל-אָחִיו: הִנֵּה, בַּעַל הַחֲלֹמוֹת הַלָּזֶה--בָּא. כ וְעַתָּה לְכוּ וְנַהַרְגֵהוּ, וְנַשְׁלִכֵהוּ בְּאַחַד הַבֹּרוֹת, וְאָמַרְנוּ, חַיָּה רָעָה אֲכָלָתְהוּ; וְנִרְאֶה, מַה-יִּהְיוּ חֲלֹמֹתָיו. כא וַיִּשְׁמַע רְאוּבֵן, וַיַּצִּלֵהוּ מִיָּדָם; וַיֹּאמֶר, לֹא נַכֶּנּוּ נָפֶשׁ. כב וַיֹּאמֶר אֲלֵהֶם רְאוּבֵן, אַל-תִּשְׁפְּכוּ-דָם--הַשְׁלִיכוּ אֹתוֹ אֶל-הַבּוֹר הַזֶּה אֲשֶׁר בַּמִּדְבָּר, וְיָד אַל-תִּשְׁלְחוּ-בוֹ: לְמַעַן, הַצִּיל אֹתוֹ מִיָּדָם, לַהֲשִׁיבוֹ, אֶל-אָבִיו. כג וַיְהִי, כַּאֲשֶׁר-בָּא יוֹסֵף אֶל-אֶחָיו; וַיַּפְשִׁיטוּ אֶת-יוֹסֵף אֶת-כֻּתָּנְתּוֹ, אֶת-כְּתֹנֶת הַפַּסִּים אֲשֶׁר עָלָיו. כד וַיִּקָּחֻהוּ--וַיַּשְׁלִכוּ אֹתוֹ, הַבֹּרָה; וְהַבּוֹר רֵק, אֵין בּוֹ מָיִם. כה וַיֵּשְׁבוּ, לֶאֱכָל-לֶחֶם, וַיִּשְׂאוּ עֵינֵיהֶם וַיִּרְאוּ, וְהִנֵּה אֹרְחַת יִשְׁמְעֵאלִים בָּאָה מִגִּלְעָד; וּגְמַלֵּיהֶם נֹשְׂאִים, נְכֹאת וּצְרִי וָלֹט--הוֹלְכִים, לְהוֹרִיד מִצְרָיְמָה. כו וַיֹּאמֶר יְהוּדָה, אֶל-אֶחָיו: מַה-בֶּצַע, כִּי נַהֲרֹג אֶת-אָחִינוּ, וְכִסִּינוּ, אֶת-דָּמוֹ. כז לְכוּ וְנִמְכְּרֶנּוּ לַיִּשְׁמְעֵאלִים, וְיָדֵנוּ אַל-תְּהִי-בוֹ, כִּי-אָחִינוּ בְשָׂרֵנוּ, הוּא; וַיִּשְׁמְעוּ, אֶחָיו. כח וַיַּעַבְרוּ אֲנָשִׁים מִדְיָנִים סֹחֲרִים, וַיִּמְשְׁכוּ וַיַּעֲלוּ אֶת-יוֹסֵף מִן-הַבּוֹר, וַיִּמְכְּרוּ אֶת-יוֹסֵף לַיִּשְׁמְעֵאלִים, בְּעֶשְׂרִים כָּסֶף; וַיָּבִיאוּ אֶת-יוֹסֵף, מִצְרָיְמָה. כט וַיָּשָׁב רְאוּבֵן אֶל-הַבּוֹר, וְהִנֵּה אֵין-יוֹסֵף בַּבּוֹר; וַיִּקְרַע, אֶת-בְּגָדָיו. ל וַיָּשָׁב אֶל-אֶחָיו, וַיֹּאמַר: הַיֶּלֶד אֵינֶנּוּ, וַאֲנִי אָנָה אֲנִי-בָא.
…………………..
לו וְהַמְּדָנִים--מָכְרוּ אֹתוֹ, אֶל-מִצְרָיִם: לְפוֹטִיפַר סְרִיס פַּרְעֹה, שַׂר הַטַּבָּחִים. {פ}


Almost all who read the above story note that it actually makes no sense at all. There are a number of problems with it, ranging from the slightly problematic to the completely unanswerable. Here’s a list of the questions:
1. Pasuk 28, “וַיַּעַבְרוּ אֲנָשִׁים מִדְיָנִים סֹחֲרִים. This seems like a total non sequitur. What does it have to do with the story of the brothers selling Yosef to the Yishmaelim?
2. Pasuk 28. “וַיִּמְשְׁכוּ וַיַּעֲלוּ אֶת-יוֹסֵף מִן-הַבּוֹר, וַיִּמְכְּרוּ אֶת-יוֹסֵף לַיִּשְׁמְעֵאלִים, בְּעֶשְׂרִים כָּסֶף" Who, exactly, is doing this? There are two possibilities, neither of which makes any sense.
a. it’s the brothers, but in that case the sentence is grammatically flawed, if it starts talking about the midyanim, then all of a sudden shifts to the brothers. And, on top of that, question (1) would then be strengthened. What, then, do the Midyanim have to do with anything and why are they in the pasuk?
b. The second possibility is that it is the midyanim, but, in that case the question is, where did they brothers go? Weren’t they going to sell yosef to the Yishmaelim? Also, if it was the Midyanim, then we are contradicting Gen (45:4) and similar p’sukim, which say it was the brothers.
3. Pasuk 29. Why is Reuven surprised that Yosef is not in the pit? He just sold him to the Yishmaelim!! Rashi says Reuven had left, but that’s an odd reading because it seems like this is one story. There’s also no textual basis for it.
4. Why did the brothers not answer Reuven that they had sold Yosef to the Yishmaelim?
5. In pasuk 36, how did the Midyanim bring Yosef to Mitzrayim, didn’t the Yishmaelim have him at this point of the story?
6. In 40:15, Yosef says “כִּי-גֻנֹּב גֻּנַּבְתִּי, מֵאֶרֶץ הָעִבְרִים; וְגַם-פֹּה לֹא-עָשִׂיתִי מְאוּמָה, כִּי-שָׂמוּ אֹתִי בַּבּוֹר,” but is it not true that he was sold, not stolen?
7. In 42:13, the brothers say, וַיֹּאמְרוּ, שְׁנֵים עָשָׂר עֲבָדֶיךָ אַחִים אֲנַחְנוּ בְּנֵי אִישׁ-אֶחָד--בְּאֶרֶץ כְּנָעַן; וְהִנֵּה הַקָּטֹן אֶת-אָבִינוּ הַיּוֹם, וְהָאֶחָד אֵינֶנּוּ. In other words, Yosef is missing. Why is Yosef missing? They know what happened to him. He was sold as a slave.


These questions are at varying level of difficulty. Some have balabatesh answers, but I’ve never really heard a solution that satisfactory deals with all of the problems well. In one fell swoop, the DH answers all these questions quite beautifully. According to DH, there are, in reality two stories here combined together. The J one reads as follows:

וַיֵּלֶךְ יוֹסֵף אַחַר אֶחָיו, וַיִּמְצָאֵם בְּדֹתָן. יח וַיִּרְאוּ אֹתוֹ, מֵרָחֹק; וּבְטֶרֶם יִקְרַב אֲלֵיהֶם, וַיִּתְנַכְּלוּ אֹתוֹ לַהֲמִיתוֹ.
כא וַיִּשְׁמַע רְאוּבֵן, וַיַּצִּלֵהוּ מִיָּדָם; וַיֹּאמֶר, לֹא נַכֶּנּוּ נָפֶשׁ
כה וַיֵּשְׁבוּ, לֶאֱכָל-לֶחֶם, וַיִּשְׂאוּ עֵינֵיהֶם וַיִּרְאוּ, וְהִנֵּה אֹרְחַת יִשְׁמְעֵאלִים בָּאָה מִגִּלְעָד; וּגְמַלֵּיהֶם נֹשְׂאִים, נְכֹאת וּצְרִי וָלֹט--הוֹלְכִים, לְהוֹרִיד מִצְרָיְמָה. כו וַיֹּאמֶר יְהוּדָה, אֶל-אֶחָיו: מַה-בֶּצַע, כִּי נַהֲרֹג אֶת-אָחִינוּ, וְכִסִּינוּ, אֶת-דָּמוֹ. כז לְכוּ וְנִמְכְּרֶנּוּ לַיִּשְׁמְעֵאלִים, וְיָדֵנוּ אַל-תְּהִי-בוֹ, כִּי-אָחִינוּ בְשָׂרֵנוּ, הוּא; וַיִּשְׁמְעוּ, אֶחָיו
כח וַיִּמְכְּרוּ אֶת-יוֹסֵף לַיִּשְׁמְעֵאלִים, בְּעֶשְׂרִים כָּסֶף;


The E version reads as follows:
כב וַיֹּאמֶר אֲלֵהֶם רְאוּבֵן, אַל-תִּשְׁפְּכוּ-דָם--הַשְׁלִיכוּ אֹתוֹ אֶל-הַבּוֹר הַזֶּה אֲשֶׁר בַּמִּדְבָּר, וְיָד אַל-תִּשְׁלְחוּ-בוֹ: לְמַעַן, הַצִּיל אֹתוֹ מִיָּדָם, לַהֲשִׁיבוֹ, אֶל-אָבִיו. כג וַיְהִי, כַּאֲשֶׁר-בָּא יוֹסֵף אֶל-אֶחָיו; וַיַּפְשִׁיטוּ אֶת-יוֹסֵף אֶת-כֻּתָּנְתּוֹ, אֶת-כְּתֹנֶת הַפַּסִּים אֲשֶׁר עָלָיו. כד וַיִּקָּחֻהוּ--וַיַּשְׁלִכוּ אֹתוֹ, הַבֹּרָה; וְהַבּוֹר רֵק, אֵין בּוֹ מָיִם
כח וַיַּעַבְרוּ אֲנָשִׁים מִדְיָנִים סֹחֲרִים, וַיִּמְשְׁכוּ וַיַּעֲלוּ אֶת-יוֹסֵף מִן-הַבּוֹר,; וַיָּבִיאוּ אֶת-יוֹסֵף, מִצְרָיְמָה.
כט וַיָּשָׁב רְאוּבֵן אֶל-הַבּוֹר, וְהִנֵּה אֵין-יוֹסֵף בַּבּוֹר; וַיִּקְרַע, אֶת-בְּגָדָיו. ל וַיָּשָׁב אֶל-אֶחָיו, וַיֹּאמַר: הַיֶּלֶד
אֵינֶנּוּ, וַאֲנִי אָנָה אֲנִי-בָא.
...........................
לו וְהַמְּדָנִים--מָכְרוּ אֹתוֹ, אֶל-מִצְרָיִם: לְפוֹטִיפַר סְרִיס פַּרְעֹה, שַׂר הַטַּבָּחִים. {פ}



Thus, in the J version, the brothers intended to kill Yosef. Yehudah convinces them to sell Yosef to the Yishmaelim. To J belongs also 45:4 and the similar p’sukim.

In the E version, the brothers throw Yosef in a pit and leave. Reuven secretly intends to rescue Yosef from the pit, but before he gets to do that, some midyanim take Yosef out of the pit and bring Yosef to Mitzrayim. Reuven is dumbfounded at why Yosef is not there and the brothers can not explain it to him. To E belongs also 40:15 and 42:13. Yosef was stolen and the brothers don’t know about it.

Different scholars have different approaches here in terms of the details, but the crux of the issue is Gen 37:28, which when you look at it, really sounds like there are two stories going on at once.

A good shabbos to all

יָלַד

The Hebrew word for a female giving birth is יָלְדָה. But, what about for a male giving birth? Sometimes, the word used is יָלַד meaning to have a kid. Sometimes though, the word הוֹלִיד is used meaning to cause to give birth in that the male causes the female to have the child. All in all, the form יָלַד appears 11 times, all of which are in JE. The form הוֹלִיד appears 62 times, 61 times in P and once in D. I am excluding verses that are of neither form, such as those in the nifal form. The complete list of the form יָלַד in JE is as follows

Gen (4:18 x 3) (10:8) (10:13) (10:15) (10:24 x 2) (10:26) (22:23) (25:3)

The complete list of the הוֹלִיד form in P is as follows:

Gen 5 (27 times) Gen (6:10) (11:10-27 27 times) (17:20) (25:19) (48:6)
Lev. (25:45)
Nu. (26:29) (26:58)

The הוֹלִיד in D is in Deut 4:25.

How can this phenomenon be explained. One approach would be to argue for a difference between the forms to explain the variant usages. As with the case of Ani/Anoki form last week, such a distinction eludes me. As far as I can tell, the forms are used interchangeably. That being the case, the fact that the form used always aligns with the author according to the DH is striking. There are 2 outs here:
1. Cassuto argues that the DH assigns the author based on the form of the verb הוֹלִיד, hence the alignment of the two is not surprising. Cassuto’s premise is that the DH has no other basis other than the form of the word הוֹלִיד. Sometimes Cassuto might be right, but I think he’s wrong in most of the cases. I can’t consider every single case in one post, but as I go through the Pentateuch, I’ll be looking at this question. Another answer is coincidence. A coincidence happening 73 times seems highly unlikely, but if we posit that the choice of verb form is not made by each and every pasuk but by each and every passage, we only have 14 passages so coincidence is a possibility. Of course, the chance of getting 14 heads in a row is less than 1 in 10,000, but that is a bad analogy for reasons I hope to explore in future posts.

Parsha Insights - Vayishlach 2006

Gen 26:
לד וַיְהִי עֵשָׂו, בֶּן-אַרְבָּעִים שָׁנָה, וַיִּקַּח אִשָּׁה אֶת-יְהוּדִית, בַּת-בְּאֵרִי הַחִתִּי--וְאֶת-בָּשְׂמַת, בַּת-אֵילֹן הַחִתִּי. לה וַתִּהְיֶיןָ, מֹרַת רוּחַ, לְיִצְחָק, וּלְרִבְקָה. {ס}

Gen 28:
ט וַיֵּלֶךְ עֵשָׂו, אֶל-יִשְׁמָעֵאל; וַיִּקַּח אֶת-מָחֲלַת בַּת-יִשְׁמָעֵאל בֶּן-אַבְרָהָם אֲחוֹת נְבָיוֹת, עַל-נָשָׁיו--לוֹ לְאִשָּׁה. {ס}

Gen 36:
ב עֵשָׂו לָקַח אֶת-נָשָׁיו, מִבְּנוֹת כְּנָעַן: אֶת-עָדָה, בַּת-אֵילוֹן הַחִתִּי, וְאֶת-אָהֳלִיבָמָה בַּת-עֲנָה, בַּת-צִבְעוֹן הַחִוִּי. ג וְאֶת-בָּשְׂמַת בַּת-יִשְׁמָעֵאל, אֲחוֹת נְבָיוֹת. ד וַתֵּלֶד עָדָה לְעֵשָׂו, אֶת-אֱלִיפָז; וּבָשְׂמַת, יָלְדָה אֶת-רְעוּאֵל.

These p’sukim obviously don’t agree with each other, in really any way at all. Traditional DH though is no help since all 3 are P . Hence, we’ve got three choices
1. Posit some extension to DH such as different strands of P to answer the questions. I, personally, am not a fan of this approach, because I think there is scant evidence for it.
2. Argue against DH that in reality Biblical texts were taken from many sources, not the 4-5 traditional ones. There is a weakness here as well because it seems clear, at least within each document according to DH, that these must be from one author due to tremendous amount of unity in the text.
3. Argue that all the above verses came from one hand and that one of the apologetic answers is in fact correct. This is weak as well because apologetics are always week.

So, I’ve got a good question and 3 bad answers. If we pick answer 3 though, DH has a huge problem because this would establish that even it must recognize the validity of apologetics which would rob of it of much of its evidence. Hmm..... Don't know what to conclude here. I originally had written this post as a scathing critique of DH but then blogger crashed on me and I lost it all and had to rewrite. I think God was trying to tell me something.

A gut shabbos to all

Ani/Anoki

I've considered this topic in the past, but I wanted to revisit it in order to facilitate my current discussion of Genesis. In P, Ani is used 123 times, but Anoki only once. In D, the score is 53 anoki's to 2 ani's. JE uses both, using Ani 51 times and Anoki 85 times. The list of Anoki's in JE is as follows:

Gen 3:10, 4:9, 7:4, 15:1, 15:2, 15:4, 16:14, 16:8, 18:27, 19:19, 20:6 (twice), 21:24, 21:26, 24:3, 24:13, 24:24, 24:27, 24:31, 24:34, 24:37, 24:42, 24:43, 25:22, 25:30, 25:32, 26:24 (twice), 27:11, 27:19, 28:15, 28:16, 28:20, 29:33, 30:1, 30:2, 30:3, 30:30, 31:5, 31:13, 31:38, 31:39, 32:12, 37:16, 38:17, 38:25, 43:9, 46:3, 46:4 (twice), 47:30, 48:21, 50:5, 50:21, 50:24

Ex 3:6, 3:11, 3:12, 3:13, 4:10 (twice), 4:11, 4:12, 4:15, 4:23, 7:17, 7:27, 8:24, 8:25, 17:9, 19:9, 20:2, 20:4, 23:20, 32:18, 34:10 34:11

Nu. 11:12 (twice), 11:14, 11:21, 22:30, 22:32, 23:15

Deut 31:23

The complete list of Anoki's in P is Gen 23:4.

It was requested than in considering various words, I begin to also consider the meaning of the word and why it would be limited to a specific source. In this case though, as far as i can tell, there is none. Other than saying that two authors used different words due to a different style, there is really no way at all to explain this phenomenon. This is because the words Ani/Anoki have the exact same meaning, as far as I can tell. I've only ever heard of two possible suggestions to explain the difference in the words. Rabbi Shalom Carmy of YU once told me thought Anoki was more pompous. An interesting theory, but I don't see any such pattern actually being born out in the usage of the words. Even if you did see such a pattern, the fact remains that that criteria is highly subjective. Cassuto tries to argue that there is a grammatical difference in the words. In order to save myself some typing, I'll just copy and paste the rules according to Cassuto from Dovid Gottlieb's website:


1. If subject of verbal-clause, `anokhi is used (examples G 16:5, 30:3). Single exception (G14:23) is sentence of unusual rhythm
2. If pronoun is in compound subject, following the verb, always `ani (G37:10)
3. If pronoun in nominativus pendens, and subject of rest of sentence is that pronoun, `ani is used (G17:4); if subject of rest of sentence is someone else, `anokhi is used (G24:27)
4. If pronoun emphasises pronominal suffix in verb then `ani used (G27:34, 38)
5. In noun-clauses, if wants to emphasise subject, `anokhi (G15:1); if does not want to emphasise subject, `ani (G41:9)

In my opinion, Cassuto's grammatical principle is a huge load of complete garbage. Since my response to his grammatical argument is quite long, verbose, and complex, I will leave it to a comment.