R. Hayyim Soloveitchik’s Attitude Towards Girsaot

A Response to Dr. Marc Shapiro

Elisha Friedman

R. Hayyim Soloveitchik earned his towering stature in Jewish intellectual history for his revolutionary methodology of Talmudic analysis, but scholars have also noted another aspect of his insights, which is a disinterest, bordering on disregard, for textual veracity. In general, traditional Talmudists were not overly concerned with bibliographic issues and girsaot (textual variants), but even within this tradition R. Hayyim stands out as extreme. One of the most glaring, and by extension often quoted, examples of R. Hayyim using a faulty text is his analysis of the passage in Mishneh Torah, Hilkhot Nizkei Mamon 4:4.

In a recent post, Dr. Marc Shapiro responded to my objection to this example. Dr. Shapiro’s clarification was helpful in elaborating on the proof from Nizkei Mamon, as well as documenting a number of scholars who have used it. The basic outlook of this school is that R. Hayyim is entirely removed from girsah considerations. In this post I will articulate an alternative approach to this issue, arguing that R. Hayyim does take girsah factors into consideration. In section one, I sketch Shapiro’s perspective (in my understanding) and the proof from Nizkei Mamon. In section two, I present an alternative reading. In section three, I conclude with some broader reflections on R. Hayyim’s attitude towards girsaot.

1. R. Hayyim’s Aversion to Girsaot

Traditional Talmudists, with some notable exceptions, generally ignored issues of girsah, preferring to focus on the content and logical flow of classical texts. In many ways this is the defining hallmark of traditional Talmudic scholarship, as opposed to other methods of study which concentrate on textual issues. One of the notable exceptions to this overall trend was Vilna Gaon. As an example, the mishna towards the end of tractate Avot (5:20) reads,

יהודה בן תימא אומר, הוי עז כנמר, וקל כנשר, ורץ כצבי, וגבור כארי, לעשות רצון אביך שבשמים. הוא היה אומר, עז פנים לגיהנם, ובשת פנים לגן עדן. יהי רצון מלפניך יי אלהינו שתבנה עירך במהרה בימינו ותן חלקנו בתורתך.

Yehuda ben Tema said: Be strong as a leopard, light as an eagle, swift as a deer, and brave as a lion, to do the will of your Father in Heaven. He would say: the brazen-faced [is destined] to Gehinom; the shame-faced to Gan Eden. May it be Your will, O Lord our God, that Your city be rebuilt speedily in our days and set our portion in the study of Your Torah

After two statements reflecting on the proper role of aggression in religious life, the mishna suddenly includes an out-of-place prayer for the rebuilding of Jerusalem. It seems clear there is a textual problem with this mishna, and surveying how traditional commentators approached it is illustrative.

Vilna Gaon offered an obvious textual emendation – that this prayer was erroneously copied here, but properly belongs four mishnayot later at the end of the tractate.[1] Meiri offers another textual solution, that originally this mishna was the conclusion of the tractate, so it ended with a prayer, and then subsequently additional statements were added on, but the prayer remained where it had been originally.[2]

Yet the majority of commentators, including Rambam, Bartenura and Tiferet Yisrael, avoid a textual solution. Some suggest a homiletical reason for the insertion of this prayer, based on the notion that God blessed the Jewish people with a lack of aggression. R. Akiva Eger in his commentary cites an interpretation from Radach (R. Dovid Hacohen of Corfu), who explains that since this mishna discusses the proper religious uses of aggression, therefore the mishna prays for the messianic times when we will no longer require any aggression.

This example illustrates the general reluctance of traditional commentators, with some exceptions such as Vilna Gaon, to resolve questions through textual solutions. It also points towards a possible reason for this: avoiding textual solutions generates greater religious and homiletical creativity. Since, in this instance, commentators avoided the textual solution, they were enabled to formulate homiletical insights.

There is no question that R. Hayyim Soloveitchik reflects this traditional hesitancy regarding textual solutions. When R. Hayyim confronts a difficulty, he prefers a conceptual answer over a textual resolution, even when the latter seems evidently correct. As an example, in Hilkhot Hametz u’Matzah 1:3, Rambam ruled that one who did not burn his hametz before Passover does not receive the punishment of lashes because it is a passive violation not active, “for he did not perform a deed” (מפני שלא עשה מעשה). R. Hayyim asks that this statement contradicts the Talmudic rationale exempting lashes (Pesachim 95a), which is that this is a prohibition which is attached to a positive mitzvah to destroy the hametz (לאו הניתק לעשה).

This question is problematic, since there is a seemingly obvious textual answer, namely, that Rambam had a well-attested variant girsah of that Talmudic passage, which omits the entire לאו הניתק לעשה rationale. In fact, in the standard Vilna Shas, this line appears in parentheses to indicate the alternative girsah! Yet instead of offering this obvious girsah solution, R. Hayyim prefers to develop his conceptual insight, which illustrates his general aversion to girsah answers.[3]

When studying R. Hayyim’s insights one can intuit that his avoidance of girsaot goes further than other leading Talmudists, his relentless focus on conceptual analysis seemingly leaves no room for girsaot (or other) concerns. But it can be difficult to express this insight in more precise language. It is particularly challenging because we have no programmatic statement from R. Hayyim as to his outlook on textual issues, as such scholars must try to recreate his attitude from his hiddushim.[4]

There is a scholarly formulation of this which contends that R. Hayyim avoids girsaot to the extent that he is even willing to cite and analyze faulty texts which were retracted by their very own authors. The explanation for this is that when it comes to religious texts R. Hayyim maintains an extreme theory of textual autonomy, that religious texts transcend the author who wrote it. I will cite Shapiro’s formulation of this idea,

On occasion, [traditionalists] even ignored the explanations Maimonides himself later gave for his halakhot… A different justification, with a more modern appeal, is quoted in the name of R. Naphtali Zvi Judah Berlin: “All the works of our teachers, the rishonim, until the era of Pnei Yehoshua, were written with ruah ha-kodesh, and obviously Maimonides’ Code is in this category. What Maimonides wrote in a responsum is Maimonides’ explanation of what he wrote in his Code, but there is still room for other explanations, for there are 70 facets to the Torah.” In fact, even after Maimonides wrote to the sages of Lunel – who had asked about a difficulty in the Mishneh Torah (Nizkei Mamon 4:4) – informing them that their text of this halakhah was mistaken, we still find R. Hayyim Soloveitchik and others offering analyses of the mistaken text!
Marc Shapiro, Studies in Maimonides and His Interpreters, (University of Scranton, 2008), 54-5.

The primary evidence for this formulation derives from R. Hayyim’s comments in Hilkhot Nizkei Mamon, to which we proceed now. The printed editions of Mishneh Torah read,

When a person entrusts his animal to an unpaid watchman, a paid watchman, a renter or a borrower, these individuals assume the owner’s responsibilities. If the animal causes damages, the watchman is held liable. When does the above apply? When they did not guard the animal at all. If, however, they guarded the animal in an excellent manner, as they should, and it got loose and caused damage, the watchmen are not liable, and the owners are liable.

This was a problematic halakha because if the watchmen guarded it excellently no one should be liable. Rambam himself was asked this question as part of a series of questions posed to him on Mishneh Torah by the sages of Lunel. Rambam responded explicitly that this version was an error and instead it should read that the owners are liable in a case of inferior protection,

If, however, they guarded the animal in an excellent manner, as they should, and it got loose and caused damage, the watchmen are not liable. If the watchmen guarded the animal in an inferior manner, they are not liable if they are unpaid. Instead, the owners are liable, even if the animal kills a person. The watchman is liable if he is paid, a renter, or a borrower.

Rambam’s response was cited in Kesef Mishnah, which is printed on the side of the standard editions of Mishneh Torah, such that no reader could miss this retraction. Yet, R. Hayyim wrote two lengthy pieces on Nizkei Mamon, and in both he analyzes this passage using the erroneous text, without ever noting Rambam’s own emendation! This seems to be clear evidence that R. Hayyim disregards textual emendation, even when the text in question is clearly mistaken.

In his recent post Shapiro documents a scholarly consensus which has coalesced around this reading, including rabbis Eliyahu Soloveitchik, Yehoshua Inbal, Norman Solomon and most prominently, Baruch Shmuel Deutsch. This reading of R. Hayyim has widespread support because it is so striking that he is knowingly using a faulty text in his lengthy analysis of this topic. In fact, when I explained this essay in May 2023 as part of a series explaining Hiddushe Rabbenu Hayyim Halevi, I enthusiastically made the same point, influenced by the earlier scholarship, which you can hear at the beginning here.

But after processing the piece, I came to believe that a more careful reading of R. Hayyim shows that in fact he deftly avoids ascribing ideas to Rambam based on the faulty text, in recognition of Rambam’s own emendation. I then found that this view had already been exhaustively presented in 2007 by R. Avraham Lipschitz.[6] In the following section I will present Lipschitz’s approach, that R. Hayyim does not dismiss Rambam’s own emendation of his text, and more broadly that R. Hayyim is not as impervious to textual matters as the prevailing scholarship would indicate.

1. Re-Reading R. Hayyim

Twice in his comments, R. Hayyim refers to the erroneous version. First, in the opening paragraph in Nizkei Mamon 4:4, where R. Hayyim sets up the piece. In order to explain why R. Hayyim is not in fact basing his insights on the faulty text, let me sketch the broad purpose of this piece. The conceptual insight at the heart of this piece is the idea that when an animal damages something, while being watched by a watchman, the watchman’s financial liability is inherent to his responsibilities as a watchman, and not derivative from the owner’s liability.[7] In R. Hayyim’s framework, the paradigmatic case for this is when the damaging animal was a tam (new damager). When a tam damages, as opposed to a mu’ad (habitual damager), the owner does not pay from their own estate, only from the value of the animal itself. This creates a litmus test for R. Hayyim to explore the watchman’s liability, since there is no financial restitution outside the animal itself.

If we now work backwards to the opening paragraph, we can see that this insight is not dependent on the faulty girsah. In order for the piece to work R. Hayyim needs to establish two things: first, that this halakha is referring to a tam, as Maggid Mishneh in his commentary suggests, second, he needs to introduce the issue of a watchman’s liability, which Raavad in his glosses touches on. Now both Raavad and Maggid Mishneh had the faulty version of Mishneh Torah, so it seems reasonable that when presenting their approaches R. Hayyim sticks with the faulty version, which is a historically accurate reflection of these two commentators.

But as R. Hayyim progresses to his own analysis of this issue, it is no longer relevant which version of Mishneh Torah one uses. In both versions – the mistaken one and the corrected one – Rambam is discussing a tam, and rules that in some cases the owner is liable, while in others the watchman is. The juxtaposition of these two rulings sets up the premise of R. Hayyim’s piece, and it makes no difference whether owners are liable for excellent protection (erroneous version) or inferior protection (corrected version). Rambam’s girsah is incidental to R. Hayyim’s discussion, and as such this piece does not provide evidence for R. Hayyim’s attitude towards girsaot.

Let us move on to the second instance where R. Hayyim analyzes Rambam’s problematic text, in Nizkei Mamon 4:11. This is a much stronger proof than the first instance, since in this piece R. Hayyim cites the erroneous version to prove the point that Rambam sees the owner’s and watchman’s liability as mutually exclusive. R. Hayyim deduces this from the erroneous text, which states the watchman is exempt and the owner is liable, indicating that it is an either-or issue, i.e. only one party will be liable in any given case. He then proceeds to explain the faulty text further, and he does all this without acknowledging Rambam’s own retraction.

From this passage then we do seem to have proof for the extreme formulation of R. Hayyim’s approach to textual autonomy, and Hazon Ish in his glosses (as Shapiro notes in his post) certainly understands it this way. But I would note two counter points. First, Lipschitz counters that R. Hayyim is still not really relying on a faulty text, since his overall discussion is not dependent on this point. The conceptual idea in this piece is whether in a case where the first watchman gave the animal over to a second watchman, the first watchman is entirely removed from the situation, or even though he’s exempt he remains a relevant party. As part of this discussion, R. Hayyim is attempting to explain the related issue of the owner’s role in the situation, which brings him to the faulty text. But even without this digression, the overall piece can stand on its own. According to Lipschitz, this minimizes the implication of R. Hayyim using the faulty text.[8]

Second, I would add that if you follow the contours of R. Hayyim’s discussion, this step is deep into a discussion of Raavad’s glosses. One could argue that in this step he is explaining the debate from the viewpoint of Raavad, who never knew about Rambam’s retraction and had to try and make consistent sense of Rambam’s rulings. Although R. Hayyim does ascribe these ideas to Rambam, it is really Raavad’s strawman version of Rambam. Put differently, had the faulty text not been used by Raavad, R. Hayyim would not defend it.

Lipschitz’s approach (slightly) modifies the prevailing scholarly view that R. Hayyim is entirely impervious to textual emendations.[9] While it is clear that R. Hayyim is largely dismissive of textual issues, even he has his red lines, such as using a faulty text which the author himself corrected. Though, I will note, the broader difficulty of formulating precisely R. Hayyim’s attitude towards girsaot remains.

1. R. Hayyim’s Attitude Towards Texts

We have explored the debate between Shapiro and Lipschitz, over the extent of R. Hayyim’s disregard for girsaot, namely, whether he would build a hiddush on a retracted girsah. In this conclusion I reflect on the religious motivation for R. Hayyim’s cavalier attitude towards girsaot.

It is important to clarify that R. Hayyim’s opposition to emendations is not rooted in a reverence for the printed text, but rather in a reverence for texts which were used by Torah giants. In other words, if a mistaken girsah can be attributed to a Torah giant, R. Hayyim considers it valid, but if it has no traditional support amongst Torah sages, it is invalidated. For example, if someone were to discover an unknown girsah in the Cairo geniza, it would have little relevance for R. Hayyim. Likewise, a printed version, which was not endorsed by Torah giants, does not have standing in R. Hayyim’s outlook.

For this reason, R. Hayyim is willing to change the girsah at the end of his essay on Hilkhot Edut 21:9, because as he notes none of the commentators on Rambam are bothered by this question (אשר כן משמע מזה שלא העירו בזה המפרשים כלל), which seemingly indicates they had a different girsah. Since no leading Torah scholar can be said to endorse the printed version, R. Hayyim does not need to defend it.

But R. Hayyim is not entirely consistent in this approach, and in another parallel case he does not apply this logic. In Hilkhot Korban Pesah 6:2, Rambam rules that one who slaughters the Korban Pesah for the correct group, but intends to sprinkle the blood for the wrong group, the sacrifice is valid but this group does not fulfill their Pesah obligation. Not only does this ruling seem logically inconsistent – if it is a valid offering why does the group not fulfill their Pesah obligation? – it also seems to blatantly contradict the explicit Talmudic ruling that the group does fulfill their obligation. R. Hayyim’s opening paragraph suggests all sorts of girsah based solutions, except the most obvious one: that there is a misprint in Mishneh Torah.

Here too, the misprint theory could be supported by the fact that none of the classical commentators on Rambam address this question, indicating they had a corrected version. In fact, in a letter critiquing R. Hayyim’s approach, R. Chaim Mordechai Roller, cites this exact consideration as proof that it must be a misprint in Mishneh Torah, since none of the classical commentators raise the question.[10] R. Hayyim’s refusal to accept this girsah change seems to contradict his own logic in Hilkhot Edut.

Be that as it may, R. Hayyim’s attitude seems to be that we must defend any version of a text which was used by Torah giants. Presumably, and this is my own words, it is hard to accept that Torah giants use nonsensical texts, such that any problematic girsah they used must have a good explanation. There may be an additional element of Divine Providence, that we must assume God would not allow Torah giants to get tripped up by worthless texts, so there must be a profound meaning to explain it.

* * * * * *

Notes

[1] Vilna Gaon in his commentary to Avot (on mishna 20 and mishna 24):

טעות סופר כאן, וצריך לומר בסוף הפרק, כמו שכתבתי למטה

[2] This is one option Meiri proposes (in his commentary to mishna 20):

או אפשר שיהיה זה סיום המסכת אלא שנתוספו בה אח”כ דברים אחרים

See also R. Shmuel de Uceda’s related discussion in Midrash Shmuel (end of mishna 21 and beginning of mishna 22), and Tosafot Yom Tov (beginning of mishna 21), about the order of these ending mishnayot. Chanoch Albeck in his edition of the Mishna (Mossad Bialik, 1952) endorses Meiri’s suggestion.

[3] Elsewhere I have documented the nuances of R. Hayyim’s attitude towards girsaot in further detail, see כללי הגר”ח מבריסק בביאור הרמב”ם – ספריית אסיף, pp. 388-92.

[4] Avraham Lipschitz, “יחסם של אחרונים לתגליות נוסח”, סיני, גליון קמא, תשס”ח, עמ’ קג

suggests a possible formulation, pointing to one instance (which will be discussed below) where other traditional commentators explain Rambam’s faulty girsah, but at least pay homage to the girsah issue and note the corrected version, whereas R. Hayyim’s omission of any acknowledgement of the whole issue is more extreme:

לכאורה לפנינו מקרה ראשון של התעלמות גורפת מהמישור הפילולוגי ללא כל צורך בהצדקתה

[5] Marc Shapiro, Studies in Maimonides and His Interpreters, (University of Scranton, 2008), 54-5.

[6] Avraham Lipschitz,

“יחסם של אחרונים לתגליות נוסח”, סיני, גליון קמא, תשס”ח, עמ’ פא-קו

which takes this halakha in Mishneh Torah as a test case for traditionalist attitudes towards girsaot. Lipschitz’s son, R. Uri Lipschitz, is a Rosh Yeshiva at Yeshivat Siach Yitzchak, founded by Rav Shagar.

[7] See R. Baruch Ber Lebowitz’s formulation of this in Birkat Shmuel, Bava Kamma, #7 & 8

[8] Ibid, pp. 103-4.

[9] Lipschitz sums up his conclusion (p. 103):

קיצור הדברים, טענתי היא שאין לגר”ח התנגדות עקרונית להשתמש בנוסחים חדשים, וודאי לא כאלה המוזכרים בכסף משנה, אלא שהקו המנחה אותו הוא הרעיון המרכזי אותו הוא בא לחדש, או הקושי המרכזי בו הוא עוסק

[10] Teshuvot Be’er Chaim Mordechai (3:12).