THIS SITE IS NO LONGER SUPPORTED            בית מדרש הוירטואלי עבר דירה
PLEASE FIND US AT OUR NEW TORAT HAR ETZION WEBSITE                                  
     English shiurim @          לשיעורים בעברית @

Daf 19 - Ketav al Gabei Ketav

  • Rav Moshe Taragin


1) Gittin (19a) mishna... Rav Cahana;
(20a) Amar Rav Chisda (first one)... lo;
Shabbat (104b) mishna, gemara ketav... muvchar;
Tosafot Gittin (19a) s.v. deyo;
Rashba Gittin (20a) s.v. Rav Chisda

2) Yerushalmi Gittin (2;3) Taman taninan... kotev;
Ramban (19a) s.v. deyo

3) Gittin (9b) Eidim... lahem;
Rashi s.v. mekar'in;
Tosafot s.v. mekar'in


1) How do Tosafot and the Rashba differ in their resolution to the contradiction between (19a) and (20a)?

2) How do the Bavli and the Yerushalmi differ?  How do they understand the problem of ketav al gabei ketav?

3) What aspect does the Ramban introduce into our sugya?

4) Explain Rashi (9b) in light of our sugya on (19a).


While the gemara continues to outline the manner by which a 'get' is composed, it introduces a concept known as 'ketav al gabei ketav'- literally, writing on top of already extant text. The gemara studies the status of ketav al gabei ketav regarding the prohibition of writing on shabbat with an eye toward its viability in the composition of 'get.'


            From the gemara's discussion on (19a) it appears that despite some debate regarding the status of the bottom layer of text, both Reish Lakish and Rebbi Yochanan agree that the upper layer is not considered writing.  Consequently, a person does not violate Shabbat with such writing. This seems to contradict the gemara (20a) which discusses the predicament of someone who forgot to compose a 'get' lishma or similarly forgot to write the name of Hashem (within a sefer Torah) lishma. Rebbi Yehuda allows him to rewrite the 'get' with the proper kavanot. If indeed the second act is not considered writing, it should not be employed to validate a text written without proper kavanot. The Rishonim differ in their treatment of this contradiction. The Rashba claims that the two gemarot differ and that indeed Rebbi Yehuda DOES consider ketav al gabei ketav as halakhic ketav. The Rashba supports his claim by citing the gemara in Shabbat (104b), where the mishna adopts a similarly dismissive view of ketav al gabei ketav as ketav.  The ensuing gemara claims that the mishna's perspective contradicts that of Rebbi Yehuda who validates ketav al gabei ketav. Just as Shabbat 104b rejects Rebbi Yehuda, similarly, Gittin 19a opposes him.


            Tosafot disagree with this view primarily because the gemara in Gittin does not register this rejection of Rebbi Yehuda. Tosafot claim that Rebbi Yehuda validates ketav al gabei ketav only when the upper layer adds or in some way improves that bottom layer. As the upper layer in Gittin (20a) conveyed the status of lishma, it is considered a 'tikkun' (improvement) of the bottom layer and is valid ketav al gabei ketav. Simple redundant upper layers are meaningless – as described in Gittin (19a).


            Tosafot's explanation of Rebbi Yehuda suggests a certain understanding of the deficiency of ketav al gabei ketav. Evidently, we do not consider it meaningful composition since a person merely retraces his steps without adding anything new.  However, when the top layer renders something new and creative, ketav al gabei ketav is considered a viable act of writing.


            A slightly different understanding of the problem of ketav al gabei ketav emerges from the Yerushalmi. The Bavli validates one and only one form of ketav al gabei ketav. If the bottom layer consists of sikra (an inferior form of ink) and the upper layer uses superior ink, then ketav al gabei ketav is valid (and a Shabbat violation has occurred). Since the top layer improved the bottom layer, its composition is meaningful (much in the same manner that a halakhic improvement carried by the top layer renders ketav al gabei ketav valid according to Tosafot).  The Yerushalmi, however, claims that ketav al gabei ketav is INVALID ONLY if the exact same material were used for each layer. If the bottom layer of sikra is covered with a second coat of deyo, OR if the bottom layer of deyo is covered with an upper coat of sikra, it is considered halakhic ketiva. What compels the Yerushalmi to validate even this form of ketav al gabei ketav?


            In truth, this debate surrounds a fundamental issue. If the upper material is different from the lower layer, even the Bavli admits that an act of erasure has occurred. In fact, the Bavli claims that sikra al gabei deyo - though not considered writing and not in violation of the act of ketiva - is still considered mochek (erasing) and entails a violation of Shabbat. Anytime a text is made illegible – even with a different substance – an act of erasure has occurred. The Bavli merely suggested that in a case of sikra al gabei deyo, even though the bottom layer was erased, since no creative act of writing occurred (since the bottom layer was superior to the ultimate replacement top layer) we cannot consider this writing. The flaw of ketav al gabei ketav was the absence of a creative act of writing, and this absence is not remedied by the erasure of the bottom layer.


Quite possibly, the Yerushalmi argues that – in principle - the concern with ketav al gabei ketav is the absence of text. We are not interested in whether a creative composition process occurred, but rather whether any NEW textual entity is rendered by the second act. In general, if the steps are merely retraced then no new ITEM has been imposed on the paper. The ultimate disqualification of ketav al gabei ketav is the absence of new text. If, however, the two layers come from different materials and we consider the bottom layer as being erased by the top, we cannot help but validate this form of ketav al gabei ketav. Once the bottom layer is erased, inevitably new text appears. Even though a creative ACT of writing did not occur, NEW TEXT was provided. The Bavli and Yerushalmi might have been debating the basis of the ketav al gabei ketav problem. Is the absence of a writing process a concern or merely the absence of new text? Can we accept a state of erasure and still disqualify ketav al gabei ketav? How do we assess sikra al gabei deyo? Tosafot might have sided with the Bavli that the problem with ketav al gabei ketav lies in the absence of a creative composition. According to Rebbi Yehuda, if the upper writing adds a new legal element then it can be considered writing.


A similar conclusion might be drawn from the Ramban's resolution to the contradiction. He suggests that Rebbi Yehuda validated ketav al gabei ketav in Gittin (20a) because the bottom layer was still wet when the top layer was applied. The gemara in Gittin (19a) refers to a situation in which the bottom layer had already dried. Why should the state of the bottom layer during the second act of writing influence the halakha? Perhaps the Ramban also viewed the problem of ketav al gabei ketav as surrounding the absence of a creative act of writing. If, however, the bottom layer has not dried and its ultimate shape is still undecided, the top layer provides a crucial service in determining its final image. This type of rewriting can easily be considered a creative act of writing.


            This question could potentially influence a very important machloket between Rashi and Tosafot in Gittin (9b). The gemara (parallel to our sugya) discusses possible solutions for illiterate eidim who cannot sign their own names. The gemara suggests 'mekaryin lahem neyar' to guide them in signing their names. Though this solution is intended to assist their signing, it is not altogether clear what the gemara means. Tosafot claim that we cut a stencil of the eidim's names and they merely fill the ink within the prearranged cut-out shapes corresponding to the letters of their names. Rashi, however, presents a more interesting solution: carve out their names in the actual document and let them trace these indentations with a pen. Tosafot is bothered by this suggestion. Wouldn't this situation be considered ketav al gabei ketav, and hence invalid for signatures!!


            Tosafot themselves suggest two answers for Rashi. The first answer dismisses the bottom layer from being significant ketav. After all, it is merely indentations carved out on the parchment rather than actual writing. As the bottom layer is not considered writing, the top layer cannot be defined as ketav al gabei ketav. This solution is somewhat problematic since the gemara itself (19a) appears to validate carved indentations as halakhic text for the composition of a 'get.' If carved indentations are considered independent text, they should form a legitimate bottom layer and disqualify subsequent top layers as ketav al gabei ketav.


            Tosafot suggest a second answer. Since the top layer converts indentations into actual ink, it entails an improvement which is an exception to the ketav al gabei ketav disqualification. Just as the gemara (19a) validated ketav al gabei ketav in the case of deyo al gabei sikra, so can we accept ketav al gabei ketav if ink is applied to indentations. This answer, though more plausible, still assumes a similarity between sikra as a bottom layer and indentations. This analogy can certainly be debated.


            We can suggest a different approach to justify Rashi's position.  Perhaps, Rashi viewed the entire concern of ketav al gabei ketav as inapplicable to the case of eidim signing a document. If he held like the Yerushalmi, that ketav al gabei ketav is not even considered new text, then indeed it would be legally problematic for eidim to append their names through ketav al gabei ketav. If, however, Rashi (as Tosafot 19a) claimed that ketav al gabei ketav entailed an absence of an act of writing, he might claim that this problem is irrelevant to eidim. A person writing a 'get' or a sefer Torah or someone involved in desecrating Shabbat must indeed execute a halakhically recognized act of writing. Ketav al gabei ketav, as it is redundant, is not considered ketiva and hence does not qualify for composition of a 'get' or sefer Torah, and similarly does not entail a Shabbat violation. By contrast, eidim do not necessarily have to perform an act of writing. Supposedly, it would suffice if they merely affix their signatures to the document even if they have failed to execute a creative act of writing. Rashi might have claimed that as ketav al gabei ketav entails an absence of writing, it is a problem which has no relevance to signing a document.


By claiming against Rashi, that ketav al gabei ketav does indeed apply to the signing of a document, Tosafot in Gittin (9b) might be lodging one of two claims:


1.  Ketav al gabei ketav does not provide new text (as we developed earlier in the Yerushalmi) and could subvert that signature process. Though the eidim do not have to perform a creative act of writing, they must provide new text – their signatures. If this text has already been supplied by the previous layer and they don't provide any text, they cannot be considered signatories.


2.  For some reason, eidim must perform an act of writing and not merely append their names to a document. Even if ketav al gabei ketav suggests an absence of 'ketiva,' it would potentially threaten the witnesses. This question will be further amplified in later shiurim which will iy"H discuss the relationship between the signing and the writing of the 'get.'


This explanation – that Rashi denies the applicability of the ketav al gabei ketav disqualification to the signing of a document- must be held up to the gemara in Gittin (19a). Gittin (9b) suggested the 'mekaryin' remedy without considering the ketav al gabei ketav ramifications. Tosafot and Rashi were left to decide the identity of mekaryin as well as the impact of this sugya upon ketav al gabei ketav.  Gittin (19a), however, directly addressed the applicability of ketav al gabei ketav to the signing of a document. It first established the inadmissibility of ketav al gabei ketav for Shabbat violation and wanted to extend this theory to the signing of a document. Doesn't the gemara itself assume that the rule IS relevant to the signing?


In truth, we must read the gemara's conclusion carefully. Reish Lakish suggested to Rav Yochanan that Shabbat and signing a 'get' should be equivalent (and in either case ketav al gabei ketav should be invalid). Rav Yochanan responded, "Because of our logical equation, we should actually rule this way practically?" This equivocal response invites many explanations. Rashi himself claims that Rav Yochanan was expressing uncertainty regarding the status of ketav al gabei ketav EVEN in the case of Shabbat. This would then qualify the gemara in Shabbat (104b), which seemed to outright and absolutely invalidate ketav al gabei ketav for Shabbat. Another reading would suggest that while the ruling in the case of Shabbat remains, the applicability to the signing of a 'get' can easily be questioned – similar to the manner we suggested in Rashi Gittin (9b). Clearly, if our logic is correct, Rashi himself would have to differentiate between Gittin (9b) - which was not worried about ketav al gabei ketav for signatures - and Gittin (19a), which engaged in a debate about both its general status and its particular relevance to signing a document.  



Sources for the next shiur:
1. Reading the get
A. Tosafot 19b s.v. Tzerichi
Mordechai, Paragraph 343
Avnei Milu'im 31:4
2. Delivering a get that appears to contain no text
A. Tosafot 19b s.v. Ta'ama
Rabbeinu Crescas s.v. Amar Shmuel
B. Rashi 19b s.v. Hashta
Rabbeinu Crescas s.v. Ve-parkinehu amar Shmuel...
Part One:
1. Why must witnesses to the delivery of the get read the document?
2. Why should they read it before the issue?
3. How does the Avnei Milu'im alter our understanding of the eidei kiyum?
Part Two:
1. How does the Tosefta contradict Shmuel's opinion?
2. Ultimately what is the basis of Shmuel's "safek"?