Vayishlach #4: How to address a business letter

Yaakov sends messengers, or angels, to his brother Esav, and commands them to deliver a message. As we read in Bereishit 32:4-5:

ד וַיִּשְׁלַח יַעֲקֹב מַלְאָכִים לְפָנָיו, אֶל-עֵשָׂו אָחִיו, אַרְצָה שֵׂעִיר, שְׂדֵה אֱדוֹם.	4 And Jacob sent messengers before him to Esau his brother unto the land of Seir, the field of Edom.
ה וַיְצַו אֹתָם, לֵאמֹר, כֹּה תֹאמְרוּן, לַאדֹנִי לְעֵשָׂו: כֹּה אָמַר, עַבְדְּךָ יַעֲקֹב, עִם-לָבָן גַּרְתִּי, וָאֵחַר עַד-עָתָּה.	5 And he commanded them, saying: 'Thus shall ye say unto my lord Esau: Thus saith thy servant Jacob: I have sojourned with Laban, and stayed until now.

The way the JPS translation parses pasuk 5 reflects the way it is most commonly taken. Namely, that he commands them, saying to them to tell his lord Esav something. The message then starts with "Thus saith thy servant Jacob." He then, in private conversation to his servants, is calling Esav his lord, which is strange.

Speiser (in Anchor Bible Genesis), following Ehrlich, writes:

Figure 1: Speiser, Anchor Bible Genesis, composite image

An epistolary formula is one "[o]f or associated with letters or the writing of letters." Thus, he notes that in Akkadian, there is a matching formula to that in the verse. (Well, not exactly. He puts the word speaks in parentheses, where in the pasuk we have the word speaks.) Plus there is the issue of why he would refer privately to Esav as his lord.

He claims that "So say to my lord Esav" is actually part of the message. Speiser's translation is not exactly literal, so it might be worthwhile to show how JPS and Speiser seem to take each part of the verse.

a) וַיְצַו אֹתָם, לֵאמֹר = 'And he commanded them, saying' = 'and gave them this message:'
b) כֹּה תֹאמְרוּן, לַאדֹנִי לְעֵשָׂו = 'Thus shall ye say unto my lord Esau:' = 'To my lord Esau say as follows'
c) כֹּה אָמַר, עַבְדְּךָ יַעֲקֹב = 'Thus saith thy servant Jacob' = 'Thus speaks your servant Jacob'

The difference is that for Speiser the message really starts at the beginning of (b) while for JPS the message starts at the beginning of (c).

Speiser also notes that this reading is against the trup. This is not necessarily the case, but I will delay that discussion to later.

I would note that, while not conclusive evidence, we need not turn to extra-Biblical evidence for this epistolary formula. Though this too is parsed differently in JPS, let us turn to Ezra 4:11-12:

יא דְּנָה, פַּרְשֶׁגֶן אִגַּרְתָּא, דִּי שְׁלַחוּ עֲלוֹהִי, עַל-אַרְתַּחְשַׁשְׂתְּא מַלְכָּא--עבדיך (עַבְדָךְ) אֱנָשׁ עֲבַר-נַהֲרָה, וּכְעֶנֶת. {פ}ו	11 this is the copy of the letter that they sent unto him, even unto Artaxerxes the king--thy servants the men beyond the River--and now {P}
יב יְדִיעַ, לֶהֱוֵא לְמַלְכָּא, דִּי יְהוּדָיֵא דִּי סְלִקוּ מִן-לְוָתָךְ, עֲלֶינָא אֲתוֹ לִירוּשְׁלֶם; קִרְיְתָא מָרָדְתָּא וּבִאישְׁתָּא, בָּנַיִן, ושורי אשכללו (וְשׁוּרַיָּא שַׁכְלִלוּ), וְאֻשַּׁיָּא יַחִיטוּ.	12 be it known unto the king, that the Jews that came up from thee are come to us unto Jerusalem; they are building the rebellious and the bad city, and have finished the walls, and are digging out the foundations.

The way this is parsed in JPS is that (in pasuk 11) 'they sent unto him, even unto Artaxerxes the king.' Then presumably, the letter begings after the --, with 'thy servants the men beyond the River.'

Here, though, we have a duplication. Why say 'they sent unto him' and then 'unto Artaxerxes the king.' The reason the word 'even' is inserted in the translation is to note this duplication. But in fact, if we turn to extrabiblical evidence, we see that this is how letters were addressed.

Thus, if this is so, it should really read: 'This is the copy of the letter that they sent to him: To Artaxerxes the king, the men of [the province] Across the River, and now.' In fact, this is how Jacob Myers translates the pasuk in the Anchor Bible Ezra-Nechemia, pg 31, without even noting in a note or comment that he is doing this.

I do not beleive that this was necessarily not known to Chazal, even though Speiser credits Ehrlich as his source for this explanation. In fact, we might surmise as much from reading midrash rabba on this week's parsha. Yehuda Nesia sent a letter to the Roman Emperor Antoninus, with whom he was friendly. (Now, elsewhere in midrash rabba for this sedra, there is an idea to use Yaakov's interactions with Esav as a template for how to interact with the Romans.)

מדרש רבה פרשה ע"ה

ה וַיְצַו אֹתָם, לֵאמֹר, כֹּה תֹאמְרוּן, לַאדֹנִי לְעֵשָׂו
רבינו אמר לרבי אפס
כתוב חד אגרא מן שמי למרן מלכא אנטונינוס
קם וכתב
מן יהודה נשיאה למרן מלכא אנטונינוס
נסבה וקרייה וקרעיה
אמר ליה כתוב
מן עבדך יהודה למרן מלכא אנטונינוס
אמר ליה רבי מפני מה אתה מבזה על כבודך?
אמר ליה מה אנא טב מן סבי?!
לא כך אמר
כֹּה אָמַר, עַבְדְּךָ יַעֲקֹב

'And he commanded them saying, so say to my lord Esav'
Rabbenu said to Rabbi Apas:
'Write a letter from me (lit. from my name) to my master the king Antoninus.'
He (R Apas) got up and wrote: From Yehuda Nesia (the Prince) to our master the king Antoninus.
He (Yehuda Nesia) got up and read it and tore it up.
He (Yehuda Nesia) said 'Write: From your servant Yehuda to our master the king Antoninus.'
He (R Apas) said, 'Rebbi, for what cause do you degrade your honor?'
He (Yehuda Nesia) said to him, 'What, am I better than my ancestor?! Does it not say: So says your servant Yaakov?'

Thus, Yehuda Nesia considers the message sent from Yaakov to Esav to be epistolic in nature, or at least a basis for an epistle. Note also that while the order of the sender and sendee is reversed, he still includes both components, and 'our master the king Antoninus' parallels 'my lord Esav.' The only thing at issue was how to refer to himself.

So, while not a clear-cut case, it appears that Yehuda Nesia may have parsed the pasuk the same way.

Now, on to the issue of the trup. Speiser noted that his reading was against the traditional accents. If so, I would guess this would be true for the pasuk in Ezra as well.

From Bereishit 32:5:
ה וַיְצַ֤ו אֹתָם֙ לֵאמֹ֔ר כֹּ֣ה תֹֽאמְר֔וּן לַֽאדֹנִ֖י לְעֵשָׂ֑ו כֹּ֤ה אָמַר֙ עַבְדְּךָ֣ יַֽעֲקֹ֔ב עִם־לָבָ֣ן גַּ֔רְתִּי וָֽאֵחַ֖ר עַד־עָֽתָּה׃

and from Ezra 4:11:
יא דְּנָה֙ פַּרְשֶׁ֣גֶן אִגַּרְתָּ֔א דִּ֚י שְׁלַ֣חוּ עֲל֔וֹהִי עַל־אַרְתַּחְשַׁ֖שְׂתְּא מַלְכָּ֑א עבדיך (עַבְדָ֛ךְ) אֱנָ֥שׁ עֲבַֽר־נַהֲרָ֖ה וּכְעֶֽנֶת׃

Note in both cases the etnachta (֑) is the accent that breaks the pasuk in twain. It appears on לְעֵשָׂ֑ו, to Esav, in Bereishit and on מַלְכָּ֑א , the king, in Ezra, thus appearing to break up the verse as in the JPS. Thus, in Bereishit, the "say to Esav" is divided from the rest of the message and the "so says your servant Yaakov," and in Ezra, the "unto the king Artaxerxes" is divided from the rest of the message and from "your servants on the other side of the river."

We would presumably like the division after the word לֵאמֹר in Bereishit, and after עֲלוֹהִי in Ezra.

However, the cantillation can be misleading in this regard, and you really need to know a lot about trup before you can make a statement that a reading is against the trup. In this case, it is advisable to see what one of te definitive books on trup has to contribute to the matter.

Let us turn to William Wickes' Two Treatises on the Accentuation of the Old Testament:In a chapter titled “The Dichotomy” of the second treatise, he writes:

(page 34)

III. α, It is on the same principle that the introductory part of the verse, although logically requiring the main accent (Athnach) after it, is constantly passed over, that this accent may be introduced where the weight of the meaning of the passage seems to lie. Observe the division in the following instances:

…

(page 35)

β. Particularly noteworthy is the way in which the words that introduce a speech – or anything similar, as a command, decree, oath, covenant, &c. – are treated. They constantly occupy a subordinate position, as far as the accents are concerned. The clause containing the speech itself, the command, &c, is counted the more important, and receives the main accentuation. In short, the division is made (as above) just as if the introductory words were absent, e.g.

‘And God said, Let there be a firmament in the midst of the waters, and let it divide the waters from the waters.' (Gen. i. 6).
'And the LORD said unto him: Therefore whosoever slayeth Cain, vengeance shall be taken on him sevenfold.' (Gen iv. 15). {J: The pasuk continues. Also, I took JPS's 'the LORD' in place of Wickes' writing of the name, and will do so in the other examples as well.}
'And Moses said unto the children of Israel, 'See, the LORD hath called by name Bezalel the son of Uri, the son of Hur, of the tribe of Judah. (Ex. xxxv. 30)
The Lord GOD hath sworn by His holiness, Lo, surely the days shall come upon you, that ye shall be taken away with hooks, and your residue with fish-hooks. (Amos iv. 2).
'...and commanded to destroy all the wise men of Babylon.' (Dan.. ii. 12).
' Thou, O king, hast made a decree, that every man that shall hear the sound of the cornet, ..... and all kinds of music, shall fall down and worship the golden image' (iii 10).

Such cases occur in every page.

Since these are words that introduce a speech, we would not expect to find the division on the word introducing that speech, but a bit later. Thus we should not expect it on לֵאמֹר but rather later in the actual message, and we should not expect it on עֲלוֹהִי but rather later, in the actual message. In fact that is where we find it, within the address: after 'So say to my lord Esav' and after 'to the king Artexerxes.'

Perhaps once we delay the etnachta from the word לֵאמֹר we would like to have it even later, so as not to divide the first part of the address from the second, but (one could say) this is already a syntactical and semantical division, since after all, even though it is part of the message, it is a command to relate the following message to the recipient.