The extra vav that wasn't

Summary: As it appears in Rashi, Ibn Ezra, and Chizkuni. See also this earlier post, by the same title.

Post: There is a masoretically troubling word in parashat Terumah. In Shemot 25:22, we read:

22. I will arrange My meetings with you there, and I will speak with you from atop the ark cover from between the two cherubim that are upon the Ark of the Testimony, all that I will command you unto the children of Israel.

כב. וְנוֹעַדְתִּי לְךָ שָׁם וְדִבַּרְתִּי אִתְּךָ מֵעַל הַכַּפֹּרֶת מִבֵּין שְׁנֵי הַכְּרֻבִים אֲשֶׁר עַל אֲרֹן הָעֵדֻת אֵת כָּל אֲשֶׁר אֲצַוֶּה אוֹתְךָ אֶל בְּנֵי יִשְׂרָאֵל:

Rashi comments on this:

"and all that I command you unto the children of Israel" -- behold this vav is extra and extraneous, and there are many like it in Scriptures. And so shall you interpret: And that which I speak with you there, it is all that I command you to the children of Israel.		ואת כל אשר אצוה אותך אל בני ישראל: הרי וי"ו זו יתירה וטפלה, וכמוהו הרבה במקרא, וכה תפתר ואת אשר אדבר עמך שם, את כל אשר אצוה אותך אל בני ישראל הוא:

So too, Ibn Ezra notes the presence of this extra vav:
וטעם וי"ו ואת כל אשר אצוה. ככה הוא. 

Chizkuni also seems to have, and explain, this extra vav. While this printing omits the vav in the citation (three lines from the bottom):

This one does not:

"And all that I command" -- The explanation is: and I will speak with you from atop the kapores, and I will also speak to you there all that I command you.

There is also the Septuagint text which seems like it might be based a vav there:
21 καὶ γνωσθήσομαί σοι ἐκεῖθεν καὶ λαλήσω σοι ἄνωθεν τοῦ ἱλαστηρίου ἀνὰ μέσον τῶν δύο Χερουβὶμ τῶν ὄντων ἐπὶ τῆς κιβωτοῦ τοῦ μαρτυρίου καὶ κατὰ πάντα, ὅσα ἐὰν ἐντείλωμαί σοι πρὸς τοὺς υἱοὺς ᾿Ισραήλ.
22 And I will make myself known to thee from thence, and I will speak to thee above the propitiatory between the two cherubs, which are upon the ark of testimony, even in all things which I shall charge thee concerning the children of Israel. 

While I don't know of any masoretic note or codex which supports this ve'et, Vetus Testamentum lists a number of texts which have the vav in place:

Minchas Shai writes:

"et kol asher -- I am astonished at three gedolei haDor, who are great scholars, Ibn Ezra, Rashi and Chizkuni, who explain ve'et kol asher with a vav, as if such was written in their sefarim. And we have already investigated this matter, and we found it in the mesora rabbata that it places this pasuk among those three which have a command 'et kol asher'. And two of those are: et kol asher anochi metzaveh etchem of parshat Re'eh, and et kol asher atzavenu of Shoftim, and in those as well, it was written without a vav. And there is no room for argument. And so writes the Rav Eliyahu Mizrachi upon Rashi -- 'I have searched in many sifrei Torah and I have not found this."

But, while Mizrachi did not find it in any Sifrei Torah, we saw that the author of Vetus Testamentum did find a number of sefarim which had this reading. Also, Masoretic notes were composed on the basis of looking at texts, so this just demonstrates a masoretic tradition based on the same texts we have (and based on the Leningrad codex, etc.). Also, I would point out that this is an instance in which the principle of lectio difficilior  works against our Masoretic text. It is stranger, on the surface, to have the vav, yet it can work out grammatically, as Rashi and Chizkuni explain.

After writing this, I found Rabbi Dr. Shnayer Leiman's article on this: "Was Rashi's Torah Scroll Flawed?". I see that I hit on many of the same points, but he also delves deeper, and e.g. considers Chavel's suggestion that it is about אֶל בְּנֵי יִשְׂרָאֵל, the el is written without a vav, instead of ve'el. He also considers many variant manuscripts and printings that exist. His conclusion, though, is that Rashi's sefer Torah indeed had that extra vav.