In my article this past Shabbos (flipdocs, HTML, paywalled Substack) I champion the simple reading of the Mishnah, that Rabbi Eleazar maintains that the lishmah requirement — applies to the tofes, the standard part which I’d define as harei at muteret lechol adam — rather than the toref.
More after the image.
I develop it in elaborate fashion, but they are all in plain text, so a detailed summary doesn’t seem necessary.
In short:
Connect Rabbi Eleazar’s eidei mesirah karti and Rabbi Meir’s eidei chatima kartei to their positions of whether and what needs to be written lishmah. Rabbi Meir holds it words by handing over a shtar ra’aya. Rabbi Eleazar doesn’t require a shtar ra’aya, though its existence could be a benefit after the fact.
So, why should Rabbi Eleazar require lishmah on any aspect of the toref, which has the names? Just like the signature, this works to make a shtar ra’ayah. If so, the Mishnah reads fine to me, that vechatav lah works on the tofes.
Derashot requiring sefirat devarim yeteirim could either apply to the toref or the tofes.
Rabbi Yehuda, also in the Mishnah requiring lishmah alongside Rabbi Eleazar, requires וְדֵן דְּיֶהֱוֵי לִיכִי מִינַּאי סֵפֶר תֵּירוּכִין וְאִגֶּרֶת שִׁבּוּקִין וְגֵט פִּטּוּרִין לִמְהָךְ לְהִתְנְסָבָא לְכֹל גְּבַר דִּיתִצְבִּיִּין. This is also sefirat devarim yeteirim. This extra text is in the tofes.
The Vilna Gaon, as conceptualized by Rav Schachter, interprets the pasuk for Rabbi Yehuda as if there were a colon after sefer keritut. וְכָ֨תַב לָ֜הּ סֵ֤פֶר כְּרִיתֻת֙ : וְנָתַ֣ן בְּיָדָ֔הּ וְשִׁלְּחָ֖הּ מִבֵּיתֽוֹ וְיָצְאָ֖ה מִבֵּית֑וֹ וְהָלְכָ֖ה וְהָיְתָ֥ה לְאִישׁ־אַחֵֽר. So that is what he is writing. See the Vilna Gaon, as reported by his students, here.
I would personally move the colon a bit earlier, to right after vechatav lah.
I don’t know that lishmah is part of what Rabbi Eleazar said.
In two printings and Vatican 130, that word lishmah is missing. And when the Geonim cite it in the piskah, that last word lishmah is missing. (Though this might be because an excerpt must be short.) In the back of my mind, I have developing ideas of how this might be relevant, and how it applies better to tosef than toref. So this is the part he (or the scribe acting as his agent) must write.
It seems difficult at first glance to make this work with other sources in Tannaim and Amoraim. So, the preceding Mishnah about a get which happens to have her name and his name? Doesn’t that suggest toref? So I am not officially dealing with all of these, so work needs to be done. But I’d say that it is explaining how he came across such a get that wasn’t lishmah. The scribes were writing it for practice, using his and her name. But if we don’t ascribe this to a different Tanna than Rabbi Eleazar, then it is the tofes that requires lishmah and is absent on this get. But even without lishmah, the husband will be giving the get and it will function as a shtar ra’ayah. It is a way of exploring different ways that a typical get will be written, with different levels of lishmah, such as not intended for gittin at all, intended for a different wife, intended for one of two wives that he chooses. But that doesn’t mean that the names in the example, on the toref, are what require lishmah.