The Voice That Would Not Be Silenced: Kol Nidrei, Halacha, and Jewish Memory

Considering the matter of vows, the Ramban makes a point that is hard to ignore:

“For there is no need to teach all the Children of Israel that a father and husband can annul [vows]… Perhaps these laws should even be concealed from them, so that they will not treat vows lightly. Now, the release from vows is not stated explicitly in the Torah; rather, it is a halacha given to Moses at Sinai, which Scripture hangs by a hair… The reason is that He made the release from a vow or oath like the secrets of the Torah: they should be told only to those worthy of them, and written only by allusion.”[1]

The release from vows is like one of the great secrets of Torah. The Gemara itself makes precisely this point. When Rav Huna bar Hinena sought to teach publicly the law of one who preemptively nullifies his vows on Rosh Hashanah, Rava stopped him: “The Tanna deliberately concealed it, and you wish to expound it publicly?”[2] The Nimukei Yosef adds that even the Mishnah itself was deliberately composed in this manner, “for we are not guarantors for the ignorant, to give them advice as to what they should do.”[3]

And yet we stand on the eve of Yom Kippur, while it is still daylight, with a Torah scroll in the cantor’s arms, and proclaim that very thing three times, in a voice that grows louder each time. If this is indeed the same nullification of vows discussed in the sugya, then we are doing publicly what Rava prohibited even teaching publicly. And if it is not, then what exactly is it? Geonim and Rishonim have disputed that question for a thousand years.

Tosafot, on the page itself, ask precisely this question—and predicate it on exactly this identification:

“And if you ask: according to what we explained above, that Kol Nidrei constitutes the nullification of vows, how can we say it publicly in the synagogue before the entire congregation? Have we not said here that it is not taught publicly?”[4]

The Formula, and the First Opposition

The ancient version of Kol Nidrei was formulated in the past tense: “All vows that I have vowed and all prohibitions that I have undertaken… from the last Yom Kippur until this Yom Kippur.” So it appears in Seder Rav Amram, in Orchot Chayim, and in the manuscripts of Machzor Vitry. A person stands on the Day of Judgment and declares that the vows of the preceding year are retroactively void.

Against this came an objection of extraordinary severity. Seder Rav Amram itself, immediately after presenting the text of Kol Nidrei, records: “But they sent word from the holy academy that this is a foolish custom, and it is forbidden to do so.”[5] Rav Yehudai Gaon, as Rav Natronai transmits in his name, declares: “We do not study [the laws of] vows, and we do not know how to prohibit or permit in such matters—neither vow nor oath.” Rav Natronai concludes: “We have neither seen nor heard this from our teachers.”[6] Rav Hai Gaon writes more forcefully still: “Be stringent as we are, and do not depart from the practice of the yeshivot.”[7] And the author of Sefer Ha-Eshkol summarizes the matter in a single sentence: “All the heads of the yeshivot said that Kol Nidrei should not be recited, for it has no basis.”[8]

The Emendation, and the Second Opposition

In France, at the beginning of the sixth millennium, Rabbi Meir ben Samuel—the son-in-law of Rashi and father of Rashbam and Rabbenu Tam—reversed the formula. Not “from the previous Yom Kippur,” but “from this Yom Kippur”: no longer a release from the past, but a stipulation concerning the future. Rabbenu Tam reports the emendation in his father’s name and defends it:

“My father and teacher, of blessed memory, corrected [the text]… And one who says ‘from the previous Yom Kippur’ is simply mistaken, for a person cannot release himself, nor can this be done without prior regret, nor without a single expert or three ordinary judges; moreover, the halacha follows Rav Pappa, who is the later authority, and who states in HaShole’ach that the vow must be specified.”[12]

He lists four requirements here, all of them conditions of formal release from vows: someone other than the person who took the vow must release it; the release must be preceded by regret; it requires either a single expert or a court of three laymen; and the vow itself must be specified. In his view, the old text satisfies none of these conditions. The new text, by contrast, requires none of them: one who stipulates in advance that his future vows shall be void is not “releasing” anything and therefore requires neither a court, regret, nor specification of the vow. Accordingly, even the vocalization changes: all of the relevant verbs are pronounced with a kamatz, indicating the future tense.[13]

With a single emendation, then, all the old objections disappeared. But four new ones arose in their place. The Kol Bo, citing Rabbi Natan ben Rabbi Yehuda, lists them; the first is precisely what troubled Rava: “First, we ought not to publicize the matter.”[14] And if Kol Nidrei is merely a stipulation, what is the role of the regret expressed in the text? Why is it repeated three times? And what becomes of the words “and it shall be forgiven”?

Ran captures the point in a single sentence: “If not, why is he taking the vow now?”[15] He therefore rules: “It is improper to recite this, lest people treat vows lightly, for the Tanna deliberately concealed it.”

The greatest difficulty of all comes from the sugya itself. One who takes a vow and remembers at that moment that he previously stipulated that his vows should be void has thereby overridden his stipulation and upheld the vow. If he does not remember, the stipulation is effective and the vow is void. The entire force of the nullification therefore depends upon forgetfulness. This is precisely what makes Rava’s prohibition so severe: one who teaches the mechanism publicly does not merely accustom people to treating vows lightly; he destroys the mechanism itself, because one who knows of it will no longer forget it. Ran captures the point in a single sentence: “If not, why is he taking the vow now?”[15] He therefore rules: “It is improper to recite this, lest people treat vows lightly, for the Tanna deliberately concealed it.” Rabbenu Yeshaya, author of Tosafot Rid, says the same, as does the Nimukei Yosef: “The matter is difficult: how can this custom be practiced? It is as though one were teaching it publicly—or worse.”[16]

And yet. Rabbenu Tam himself, who defended the emendation with all his force, writes elsewhere about himself: “My heart did not permit me to change the established order of the community… Even with regard to the shofar blasts of Rosh Hashanah I did not wish to change the custom, although it appears to me that we are acting contrary to the law.”[17] His contemporary Rabbi Yosef Kara, a student of Rashi, openly admits that he is convinced by the emendation but does not practice it: “Because I feared changing the ancestral custom that has existed from time immemorial, I wrote ‘from the previous Yom Kippur’… Yet to my limited understanding, Rabbi Meir’s practice appears correct.”[18] And Rosh, who could have adopted the emendation and thereby freed himself from the entire set of objections, chooses the earlier custom: “Nevertheless, the practice of the ancients appears preferable.”[19]

The Release and the Stipulation

How, then, did it become permissible to proclaim publicly what Rava prohibited teaching publicly?

Tosafot answer briefly: “Because it is like the language of the Targum—not everyone understands it.” Rabbenu Tam, as cited by Or Zarua, elaborates: Rav Huna bar Hinena wished to teach the law “to speakers of the vernacular, in the vernacular, so that everyone would understand—and they would come to treat vows lightly. But now that we say it in the sacred language, not everyone understands.”[20] Ritva formulates the matter in terms taken directly from the sugya: the permissibility depends on the ignorant who “do not remember their Yom Kippur stipulation.” In other words, permission to proclaim it aloud depends upon precisely the same forgetfulness upon which the mechanism itself depends.[21]

The Levush, however, says precisely the opposite. Kol Nidrei was instituted in Aramaic “because it was instituted in Babylonia, where Aramaic was their principal language; therefore they instituted it in Aramaic so that everyone would understand it.”[22] This is not merely an alternative opinion. If everyone understands it, the very condition upon which the mechanism depends disappears. The author of Noheg KaTzon Yosef sets the two explanations against one another and states the contradiction explicitly: “This is contrary to the words of Maharam Yaffe in the Levush.”[23]

Ritva adds a third reversal. He accepts Rava’s concern but argues that reality has turned it on its head: “That was in their day, when vows were regarded seriously by the ignorant; but today, when vows are treated lightly, this is preferable for us.”[24] In generations when the public viewed vows with gravity, publicizing the release was dangerous; in generations when vows themselves had become trivialized, publicizing it was precisely the remedy. The author of Shibbolei HaLeket answers him in the same language from the opposite direction: “When you say this before ordinary and ignorant people, you turn their inadvertent transgression into deliberate sin.”[25]

In the midst of this argument, the Shitah Mekubetzet says something that turns the entire discussion around. A nullification of vows performed only in one’s heart has no legal force, because unspoken intentions do not count; only an explicit declaration constitutes a “clear and demonstrable intention.” Hence:

“Nevertheless, there is reason to require a public nullification, as indicated by the custom that we recite Kol Nidrei aloud on the eve of Yom Kippur, publicly; surely this custom did not arise for no reason.”[26]

Publicity is not a concession to longstanding custom. It is a condition of the act itself.

Yet two requirements now stand in direct opposition. The nullification must be proclaimed publicly; otherwise it is no more than an unspoken intention. Yet it must not be understood as a mechanism upon which one can rely, because if it is understood it will be remembered, and if remembered it is void. How can both requirements be satisfied at once? By means of a formula proclaimed aloud but not explained. Thus the Shitah Mekubetzet writes:

“The fact that it is recited aloud on the eve of Yom Kippur is not equivalent to teaching it publicly, since we do not tell them: ‘You are released.’”[27]

Rava did not prohibit proclamation. He prohibited exposition. Proclaiming an Aramaic formula that no one explains is not the same as teaching the public that it possesses a means of escaping its vows. Tosafot, too, conclude their answer with exactly this point: “Nor should one tell them: ‘You are permitted.’” But this answer, on its own terms, depends upon a condition: the formula must remain foreign and unfamiliar. Just as the mechanism itself depends upon forgetfulness, so too does the permission to proclaim it. Both rest upon the same fragile foundation, and both are destined to collapse together.

The same Shitah Mekubetzet also resolves the question of who constitutes the court: “If they gathered of their own accord, that is sufficient; and here the primary purpose of their gathering is certainly prayer, and having gathered for another purpose they may release one another.” The congregation that assembled for prayer rather than for annulment is fit to serve as a court for one another precisely because it came for something else. Rosh had already made the same point: “The entire congregation recites it individually and quietly; as for the cantor, the congregation releases him.”

Those Who Sought to Abolish It from Within

Rabbi Chaim ben Galipapa, rabbi of Pamplona, had already changed the text of Kol Nidrei in his community to the future tense and then wrote to Rivash. Rivash answered him at length, first rebuking him for another claim he had made—that in their own time there was no longer any reason to conceal release from vows from the ignorant, “for everyone is wise and discerning.” Rivash rejects this completely and summarizes in a single sentence:

“If some leniency appears correct to him as a matter of law because of his own reasoning, he should not act upon it, and from the masses he should conceal and hide that leniency.”[28]

Only afterward does he turn to Kol Nidrei. He congratulates Galipapa on the change he has already made and adds: since the people of your community have already accepted one correction from you, “you can now abolish it altogether… and if your strength proves sufficient to abolish it entirely, all the sages of the land will thank you.” In that same responsum he writes: “It is better not to say it at all, so that people will not treat vows lightly, for the Tanna deliberately concealed it… and throughout Catalonia it is not recited.”

Since the people of your community have already accepted one correction from you, “you can now abolish it altogether… and if your strength proves sufficient to abolish it entirely, all the sages of the land will thank you.”

Meiri, in Provence, had already done precisely that:

“After the great Geonim of the world themselves were in doubt concerning these matters, what are we, and what is our life, that we should enter into their dispute? We have nothing to do but make room for peace and flee from doubt. For this reason, our practice is not to recite the Kol Nidrei formula at all.”[29]

Radbaz testifies that the Musta‘arabi communities of Egypt likewise did not recite it.[30] And Ramban, in his own laws of vows, omitted the underlying law itself—not Kol Nidrei, but the rule of advance nullification found in the sugya. Ritva explains why: “All the early authorities established it as law, except our master of blessed memory, who refrained from teaching it in the academy for the reason we have stated.”[31]

We therefore find, across four lands and over three centuries, authorities of the highest stature who held that the practice ought to be abolished and said so explicitly.[32]

And yet, everywhere else, it continued to be recited.

At this point something must be stated clearly. In this entire debate, no one is answering an outside accuser. Rivash is responding to a rabbi; Meiri is discussing the doubts of the Geonim; Radbaz is answering a question concerning the wording. Outside accusers certainly existed, and had for centuries, as we shall shortly see. But they were not in the room, and they did not determine the outcome. This was an entirely legitimate internal dispute in which both sides possessed serious arguments, and it remained unresolved.

And Then Entered the One Who Had Not Been Asked

One point must be established before proceeding, without which this entire matter cannot be understood. The Mishnah in Chagigah lists the release from vows among the laws that “hover in the air and have nothing upon which to rely.” Rambam, who knew this very well, writes something striking in the Laws of Oaths:

“Although it is permissible to seek release from an oath, as we have explained, and there is no defect in doing so, one whose heart recoils from the matter possesses a trace of heresy. Nevertheless, it is proper to exercise caution in this matter.”[33]

One who recoils from the release of vows not because of a difficulty in the sugya but because his heart finds the very institution troubling—that recoil itself is the sign, a trace of heresy. For a halacha that lacks an explicit verse and rests solely upon received tradition is precisely where faith in the Oral Torah is tested. The author of Chavot Yair records the other side of the same coin, citing a tradition he had received: “One who refrains from releasing ordinary vows is regarded as deficient in faith in the tradition of our Sages, since the release from vows is not explicit in Scripture, as stated in Hagigah, where it is said to hover in the air.”[34]

The weakest point in the release from vows—its lack of an explicit Scriptural source—is therefore also the very point at which it becomes a test. Anyone who attacks it will attack precisely there.

Let us now return to those same centuries and see what was being said about Kol Nidrei from outside. Beginning in the ninth century, references to Kol Nidrei begin appearing from an entirely different quarter. When one reads them, it becomes clear that it was not the exact wording that troubled the critics.

The first were the Karaites, the sect that rejected the Oral Torah and whose arguments, as we have seen, had led the yeshivot to refrain from teaching the law. Daniel al-Qumisi denounced the release from vows as false swearing; after him came Salmon ben Jeroham and Sahl ben Mazliah, both of whom quote the formula in the past tense. Yehuda Hadassi, writing in 1148, describes what he himself witnessed:

“They gather in the synagogue on the eve of the fast of Yom Kippur… and the communal cantor stands ‘before the eyes of all the congregation,’ with the congregation standing beside him… and with a pleasant and loud voice he begins to chant in their ears, and they listen attentively… and they imagine that by this they are forgiven and atoned for.”[35]

The standing congregation, the voice, the melody, the experience—these are what anger him, and these are precisely the features that have no source in the sugya. He does not object to Rabbi Meir ben Samuel’s emendation, nor to the technical question of whether one “overrode his stipulation.” He objects to the voice.[36]

And yet the voice was not lowered.

In the year 5000 (1240), an apostate named Nicholas Donin stood in Paris and made the accusation in full: Jews do not honor their oaths, since every year they release themselves from their vows.[37] Rabbi Yehiel of Paris answered him, and his answer had already been written generations earlier:

“Look at the end: ‘And it shall be forgiven to the entire congregation of the Children of Israel, for all the people acted unwittingly.’ We release only inadvertent transgressions… But vows between one person and another cannot be annulled without the other party’s consent.”[38]

So too Tosafot and Rabbenu Tam had already stated explicitly that public vows and oaths imposed by a court “do not depend upon his intention alone, but upon the intention of Heaven, the court, and the government.”[39] And so too Rav Saadia Gaon, whom Ramban cites in Mishpat HaCherem: the release concerns only communal inadvertence and mistaken vows; “as for other vows, bans, and oaths, no one ever imagined such a thing.”[40]

By 1858, the Russian authorities had already ordered that prayer books print, at the beginning of Kol Nidrei, a Hebrew explanation stating that it had no bearing on oaths or obligations.

And yet nothing changed.

In 1700, Johann Eisenmenger, a Christian Hebraist, published his Entdecktes Judenthum—thousands of pages assembled from the Talmud and halachic literature for the purpose of attacking Judaism and Jewish tradition. He too reached Kol Nidrei, investigated the matter, and conceded that according to Jewish law Jews are not released from an oath sworn to a Christian or to the authorities, whether through Kol Nidrei or by any other means.[41] Yet despite that admission, by 1729, in the German town of Celle, Kol Nidrei ceased to be merely a matter of books. In the prescribed Jewish oath, the More Judaico,[42] a Jew was required to stand wrapped in his tallit, holding a Torah scroll, and declare that his oath could not be released by a Jewish court “nor annulled by the prayer of Kol Nidrei.”[43]

This is a moment worth dwelling upon. A Christian court was, in effect, attempting to decide a dispute among the Rishonim—and chose precisely the reading rejected by most of them: that Kol Nidrei constitutes a general release from oaths.

By 1858, the Russian authorities had already ordered that prayer books print, at the beginning of Kol Nidrei, a Hebrew explanation stating that it had no bearing on oaths or obligations.

Oldenburg, Yom Kippur 5600

On Yom Kippur of the year 5600 (1839), the state rabbi of the Duchy of Oldenburg instructed his community to omit Kol Nidrei from the service. He was thirty-one years old. His name was Rabbi Samson Raphael Hirsch.

The reasons he gave were “reasons of halacha,” and there is no reason to doubt them.[44] They apparently followed the same line we have already encountered in Ran, Rivash, and Meiri: concern that those who do not understand what is happening will come to treat vows lightly. In Horeb, Hirsch writes that an oath sworn in court cannot be released at all.[45]

A year later, on Yom Kippur 5601, he reversed himself. In a written declaration that has survived, he explained:

“Although last year Kol Nidrei was omitted for reasons grounded in halacha, I have nevertheless come to the conclusion that such a change—however well founded in law—is better not made by a single rabbi acting alone.”[46]

He instructed the cantor to resume reciting Kol Nidrei, once. That, incidentally, is Ritva’s own ruling.[47]

Years later, describing what the Reform movement had wrought, Hirsch wrote a sentence that does not concern Kol Nidrei at all—and yet explains it better than anything else:

“Only in this have you harmed us… You have made almost every remedy and every reform odious in the eyes of those who walk in innocence. Every innovation, even one founded in holiness, will now be rejected for fear that it springs from a polluted source resembling your counsel.”[48]

The touch of outsiders can make something untouchable even when their criticism had substance. The man who wrote those words had himself been burned by precisely this phenomenon years earlier, when rabbis older than he—chief among them his teacher Chacham Bernays—caused him to reverse his decision to abolish Kol Nidrei.[49]

Braunschweig, 25 Sivan 5604

Five years after Oldenburg, a conference of Reform rabbis convened in Braunschweig and sought to abolish Kol Nidrei altogether. The preserved discussion deserves attention, not because of the strength of their arguments, but because of what they inadvertently revealed.

First, they knew. The chairman opened:

“We recognize that the accusations leveled against us concerning Kol Nidrei are indeed slanderous, and nevertheless we shall not refrain from reform.”[50]

They said explicitly that the accusations against Kol Nidrei were false—and sought to abolish it anyway.

Second, they discovered that the rabbis were not with them. One participant lamented that he had recently encountered an opinion submitted by two rabbis to the authorities—one from the old school and one from the new—in which both declared Kol Nidrei to be “among the sacred elements of our prayers.”[51] In other words, when the government asked the question, even a rabbi of the new school stood up to defend it.

Third, they needed the Geonim. One speaker sought to prove that Kol Nidrei was of little importance by pointing out “that at the time of its introduction five Geonim opposed it.”[52] The internal debate we examined in the previous sections had become an instrument in the hands of those who had not been participants in it. That too is one way in which others attempt to decide for us.

Fourth, they knew it would not work. One of those advocating abolition conceded that “the melody does not outweigh the damage”[53]—meaning that even in his eyes the melody carried something that had to be weighed. At the end of the discussion, one rabbi stood up and said simply: “He fears that he will not succeed in implementing this in his community.”[54]

In the very room where the decision to abolish it was taken, someone already knew that the people would not permit it.

And so it proved.

And Yet It Has No Effect

At the end of this long road, the Aruch HaShulchan writes the sentence toward which everything has been leading. But the reason he gives is greater than the sentence itself:

“This is effective only when one stipulates concerning the future that every vow he may take shall be void, and then forgets this at the time of taking the vow. But now that it is a fixed formula, forgetfulness is no longer relevant, and it is merely a prayer like all other prayers.”[55]

He does not say that Kol Nidrei never constituted nullification. He says that it ceased to constitute nullification, and he explains exactly how: once it became a fixed formula, forgetfulness was no longer possible. And forgetfulness was the entire source of its force.

Nor was the Aruch HaShulchan the first to arrive there. Ritva had already written that the intention is merely “by way of supplication”; the Nimukei Yosef wrote that it is recited neither as release from past vows nor as a stipulation for future ones, but “as a prayer to the Holy One, blessed be He”; and Radbaz used precisely this point to explain why it may be said publicly: “They regard it as part of the order of prayer.”[56] Each, in his own way, removed Kol Nidrei from the original sugya. The text itself also continued to be edited: Maharam of Rothenburg directed that the words “as it is written in Your Torah” be omitted, and the formula “By authority of the heavenly court and by authority of the earthly court… we permit ourselves to pray with the transgressors” is attributed to him.[57] The Aruch HaShulchan merely stated, in the end, what had decided the matter.

What turned it into a fixed formula? One might say: the voice did.

The three repetitions, the melody, the rising volume—these turned the stipulation into something familiar and known. Once it was known, people remembered it; once they remembered it, they overrode it. The very publicity that the Shitah Mekubetzet had required as a condition for the nullification’s validity was what eventually destroyed the nullification.

The three repetitions, the melody, the rising volume—these turned the stipulation into something familiar and known. Once it was known, people remembered it; once they remembered it, they overrode it. The very publicity that the Shitah Mekubetzet had required as a condition for the nullification’s validity was what eventually destroyed the nullification.

There was therefore no solution to the contradiction we identified earlier, but rather its termination. As long as Kol Nidrei contained an operative nullification, the two demands stood opposed and were reconciled only with difficulty. Once the content ceased to function, both demands disappeared together: there was no longer a nullification requiring publicity, and no longer a teaching that was forbidden to publicize. The form alone remained. The halachic authorities did not abandon Kol Nidrei; they removed it from the sugya, leaving the voice without any independent halachic justification.

And this is the strange thing about the entire story. Over the course of a thousand years, halacha emptied Kol Nidrei of its operative content—and during those same centuries it became stronger rather than weaker. Why, then, did they not listen to Rivash?

It is the Rogatchover, in his notes to Zohar HaRakia, who answers with reference not to content but to form:

“Because the Boethusians do not acknowledge [this law]… therefore we perform it publicly, in order to publicize it… as with the harvesting of the Omer in Menahot 65a and the willow ceremony in Sukkah 43b. All matters that they do not acknowledge must be done publicly. This is why we perform the release of vows on Yom Kippur with such extraordinary publicity.”[58]

Something others deny, we perform aloud. Not in order to win an argument, but because the very denial transforms publicity into part of the act.

What is surprising is that this thread runs through the Rishonim even though none of them used the word “Boethusians.” Ask Rosh, Ritva, the author of Shibbolei HaLeket, the Nimukei Yosef, or the Levush why Kol Nidrei is said three times, and they all answer from the same source: because the enactments of the Sages are tripled—“This is the sickle, this is the sickle, this is the sickle”; “This is the basket, this is the basket, this is the basket”—in Menachot 65a.[59]

Many times I wished to correct it and taught the cantors the proper form, but when they came to pray they were unable to alter it because of the established melody that had become universally known.

The Levush adds the practical implication: “Therefore the practice is to raise the voice on the second recitation more than on the first, and on the third more than on the second, so that everyone will hear and make the stipulation for themselves.”[60] Mahzor Vitry preserves another explanation, more poetic:

“The first time he should recite it exceedingly softly, like a person astonished and afraid to enter the king’s palace and ask a gift from the king, fearing to approach him… On the third occasion he raises his voice more and more, like a person who has become accustomed to being among the king’s household and whose heart is at ease with him.”[61]

Anyone who wishes to know how powerful that melody had become should read the Levush himself. He believed that the text printed in the prayer books was fundamentally corrupt, attempted to correct it—and failed:

“Many times I wished to correct it and taught the cantors the proper form, but when they came to pray they were unable to alter it because of the established melody that had become universally known.”[62]

The great halachic authority, pen in hand, standing before a cantor who cannot change the tune.

The Decision

Rema, writing generations earlier, had already recorded three laws in a single breath in Orah Hayyim 619, and each is another element of the same decision:

“The custom is to say Kol Nidrei while it is still day, extending the melodies until night; it is recited three times, and each time the voice is raised more than before… And a person should not deviate from the custom of the community, even with regard to its melodies or the liturgical poems recited there.”[63]

While it is still day—because vows may not be released on Shabbat or Yom Tov, and the same applies to Yom Kippur.[64] With a rising voice—so that everyone will hear and stipulate for himself. And a person should not change the custom.

This is not a position born of thoughtlessness. It is stated after the Geonim prohibited the practice; after Ran ruled that it should not be recited; after Rivash advised abolishing it entirely; after Meiri wrote that this was the practice in Provence; and after Radbaz testified to the Egyptian custom. All of these remain standing; no one erases them. But a question that had opened in the study hall was closed in the synagogue—not because the congregation knew how to answer Rivash’s arguments, but because a point had been reached at which surrendering the practice no longer looked like reform. It looked like capitulation.

And that is precisely what happened. As long as the matter remained internal, it was an open dispute that could be decided in either direction. Once others laid their hands upon it—in Celle, in St. Petersburg, in Braunschweig—the question closed. Not because the abolitionist case had become weaker, but because the decision was no longer ours alone. At such a moment, one does not surrender that which others have come to determine in our place.

What We Did Give Up

But a principle stated in general terms must be tested at the point where reality puts it under pressure. If we have said that we do not relinquish what others seek to decide for us, we must ask whether we have in fact always acted that way.

We have not.

At the conclusion of every prayer service, the congregation recites Aleinu. In the prayer book before us appears a line that, for some four hundred years, did not appear there: “For they bow to vanity and emptiness, and pray to a god who cannot save.” Its language is entirely biblical, combining two verses from Isaiah, and it names no one. Yet beginning with the Rishonim it was interpreted through gematria: the author of Arugat HaBosem records a tradition that when reciting “vanity and emptiness,” one should have in mind the numerical value of “the Name”; for that same reason, the custom arose to spit at that point.[65]

From then on, the line was never left in peace. As far as is known, the earliest Christian text to quote it is Bernard Gui’s inquisitors’ manual of 1323–1324. He translates Aleinu into Latin, including the line in its proper place, and warns the reader that although Jews do not explicitly name Christians, their indirect formulations are aimed at them. Nor is Aleinu alone on his list: he includes alongside it the blessing against the sectarians and the blessing “who has not made me a gentile.” From then on, all three would stand beneath the same hand. Roughly three generations later, in 1399, an apostate named Pessah-Peter attacked Aleinu in Prague; Rabbi Yom Tov Lipmann Mühlhausen answered him at length in Sefer Nitzachon.[66]

The answer did not help. The line disappeared—first from manuscripts, then from printed editions. Some deleted it and left a blank space; some placed a small circle, an asterisk, or a comma in its place; most left nothing at all, so that the reader had no idea anything had ever been there. Thus it remained in Ashkenazic prayer books and mahzorim for centuries.[67]

Here there is no “and yet.”

Anyone wishing to see how far the process went should open Tur, Orach Chaim 133. In the earliest printings, the text reads: “One says Aleinu leshabe’ach, and when reaching ‘a god who cannot save,’ one should pause slightly before saying ‘but we bend the knee and bow,’ so that it should not appear that he is referring back to what preceded.” In nineteenth-century editions this changed. First it became “when reaching ‘the god of stars and constellations who cannot save’”—using the standard abbreviation found in censored Jewish books—and later “like all their multitude,” the reading found in the Vilna edition.[68]

These are not variant readings. The Tur instructs the reader to pause lest it appear that his bowing refers to the words immediately preceding, whereas “like all their multitude” is not derogatory at all and provides no reason for such a pause. The instruction remained in place after the phrase to which it referred had been erased. The censorship did not remain in the prayer book. It entered the Tur itself.

Now to the question. These are not two unrelated episodes. Antonius Margaritha, whose book was presented to the emperor at the Imperial Diet in 1530, attacks Kol Nidrei in that work—and at that very gathering Rabbi Joseph of Rosheim was required to defend Aleinu. One book, one author, two prayers. A generation later, an anonymous pamphlet listed the two side by side.[69]

Yet Aleinu yielded, and Kol Nidrei did not.

Why?

One answer lies in halacha itself and becomes apparent precisely from Bernard Gui’s list. All three prayers he mentioned were censored, but not in the same way. The blessing against the sectarians is a formal blessing instituted by the Sages and cannot simply be omitted. It was therefore not removed; instead, words were substituted where the censored object of the curse had previously stood.[70] Aleinu, by contrast, is fundamentally a liturgical poem whose authority derives from custom. There was no need to rewrite it at all: the line could simply be omitted, without impairing any formal obligation. A piyut could be sacrificed, and it was.[71]

Yet however correct this answer may be, it only intensifies the question regarding Kol Nidrei. For Kol Nidrei had even less standing. Aleinu is recited every day; it is cited in the Tur and the Shulchan Aruch; no authority ever described it as having no basis. Kol Nidrei, by contrast, was called a foolish custom by the Geonim; Rivash wrote that it would be better not to say it at all; and the Aruch HaShulchan concluded that it is merely a prayer like all other prayers.

By that measure, Kol Nidrei should have fallen first.

It alone did not fall.

In Aleinu, the voice became the instrument of the decree, and the text yielded. In Kol Nidrei, the voice was ours: the Shitah Mekubetzet required it, Rema commanded it, and the Levush taught cantors to raise it. All the authorities could do was add a prefatory explanation at the top of the printed page.

The second answer lies in the archive. In 1703 the King of Prussia issued an edict concerning Aleinu. It followed denunciations by two apostates, the second of whom explicitly identified the prayer. Despite a Christian professor from Göttingen defending the Jews, and despite the Jews swearing that all the charges were false, specific passages were prohibited “for all eternity,” and Jews were forbidden to teach them to their children on pain of expulsion. Nor was that all:

“From now on the prayer shall be recited, instead of as previously in a whisper, loudly and clearly by one member of the congregation, with all the others repeating after him.”[72]

That deserves to be read twice. The authorities did not ask that the prayer be silenced. They demanded that it be recited aloud—so that they could hear what had been removed.

Until that day, Aleinu had been said quietly; and quiet speech is precisely what allows one to say what cannot be printed.

Here the reversal is complete. In Aleinu, the voice became the instrument of the decree, and the text yielded. In Kol Nidrei, the voice was ours: the Shitah Mekubetzet required it, Rema commanded it, and the Levush taught cantors to raise it. All the authorities could do was add a prefatory explanation at the top of the printed page.

The page can be censored.

The voice cannot.

The proof comes from Prussia itself. Some seventy years after the edict, in Königsberg, the inspector complained that the congregation was mumbling and could not be heard. The cantor was interrogated and admitted: he recited Aleinu in full voice only on Yom Kippur; on all other days he used a moderate tone.[73] Even there, when the voice returned, it returned on Yom Kippur.

It follows that resistance is not merely a reflex. There is discrimination at work. In the very same generations it operated in both directions: Aleinu was surrendered; Kol Nidrei was not. What makes one thing a Kol Nidrei and another an Aleinu—that still requires study.

But one condition necessary for such discrimination emerges clearly from the comparison. Distinctions can be made only so long as the question being asked is: what is the value of this particular practice in itself? As long as that question is asked in the study hall, different practices can yield different answers. Here they yielded, there they stood firm—and often the very same sages made both decisions.

But once an outside hand touches everything at once, the question changes. We no longer ask what this practice is worth. We ask who gets to decide.

That question has only one answer, the same answer for every practice. Once the second question is asked, the first is no longer asked at all.

This, it seems to me, is the heavy price of indiscriminate external pressure. The problem is not simply that it produces rigidity—sometimes rigidity is entirely justified, and that is the Rogatchover’s whole point. The deeper problem is that it deprives the community under pressure of the ability to distinguish between what deserves to be defended and what does not.

Such distinctions take generations to build.[74]

❊

The congregation did one further thing, for which there is no trace in any sugya: it placed a Torah scroll in the cantor’s arms. The Kol Bo and Orchot Chaim give a halachic explanation: “They take out a Torah scroll in order to release vows in its presence.”[75] The scroll serves as an instrument of the court.

But perhaps another explanation may be suggested.

The Gemara in Chagigah asks about the Mishnah’s statement that the release from vows has nothing upon which to rely, and cites Tannaim who attempted to locate Scriptural support for it. One interpretation after another is rejected—and one survives. In the Yerushalmi it appears as follows:

“Rav Yehuda said in the name of Shmuel: It is written, ‘He shall not profane his word.’ He cannot render his own word profane; another may render his word profane. And who is that? A sage who releases him from his vow.”[76]

This is precisely the exposition of Ramban with which we began. He who wrote that the release from vows hangs by a hair immediately added: “Except that Scripture alluded to it in the words, ‘He shall not profane his word.’”

For a thousand years, halachic authorities searched for something upon which Kol Nidrei could rest, and many of them answered that it rested upon nothing.

The congregation did not surrender it.

And at the moment of its recitation, it raised aloft the very book from which the single surviving verse had been expounded—

to teach and make known that it does, after all, have something upon which to rely.

[1] Ramban, Numbers 30:2.

[2] Nedarim 23b.

[3] Nimukei Yosef, Nedarim 7a in the pagination of Rif.

[4] Tosafot ad loc., s.v. Tanna setumei. Their answer: “Here this applies specifically because of the ignorant who attend the public lecture… But Kol Nidrei, which is in the language of the Targum, is not understood by everyone; nor should one tell them, ‘You are permitted.’”

[5] Seder Rav Amram Gaon, ed. D. Goldschmidt, p. 163. The phrase “foolish custom” is cited by the Rishonim in the names of various Geonim: in Sefer HaManhig (p. 344), Rabbenu Yeruham in the name of Rav Natronai, and Hemda Genuza in the name of Rav Nahshon.

[6] Responsa of Rav Natronai Gaon, ed. Brody, p. 311. The words are those of Rav Yehudai, transmitted by Rav Natronai in his name. See also Orhot Hayyim, Laws of Yom Kippur.

[7] Sha‘arei Teshuva, no. 143. See also Kol Bo, no. 68, and Orhot Hayyim ad loc., where he is cited as “Rav Hai bar Rav Nahshon”; in Shibbolei HaLeket, no. 317, a “Seder Rav Hai Gaon” is cited without further identification, and scholars have disputed whom this refers to.

[8] Sefer Ha-Eshkol, Albeck ed., p. 97, Laws of Yom Kippur.

[9] Kol Bo, no. 68, in the name of the author of Sefer HaIttim. See likewise Orhot Hayyim ad loc. in the name of Rabbi Yehuda ben Barzilai.

[10] Lewin, introduction to Otzar HaGeonim, Nedarim, p. 9; cited also in Avoda Berura on Tractate Nedarim.

[11] Abudarham, Order of Yom Kippur.

[12] Sefer HaYashar of Rabbenu Tam, Novellae, no. 100, Schlesinger ed., p. 70. Rosh likewise lists these arguments (Nedarim 23b), except that instead of “a person cannot release himself” he asks: “The cantor who says Kol Nidrei—who released his vow?” Ramban likewise cites them in Mishpat HaHerem, writing that the release is ineffective “for the clear reasons explained by Rabbenu Tam”; another version reads “four reasons.”

[13] Ran, Nedarim 23b (“all with a qamatz”); Ritva, Nedarim 25b; Shitah Mekubetzet ad loc., which adds: “Those who pronounce them with a sheva… are mistaken.”

[14] Kol Bo, no. 68, in the name of Rabbi Natan ben Rabbi Yehuda. It adds: “It is true that Rabbenu Tam rejects some of them.”

[15] Ran, Nedarim 23b; likewise Nimukei Yosef ad loc.

[16] Tosafot Rid, cited in Shibbolei HaLeket, no. 317; Nimukei Yosef ad loc.

[17] Sefer HaYashar, Responsa section, Rosenthal ed., p. 99; cited by N. Wieder, “Past and Future in the Text of Kol Nidrei,” p. 190.

[18] Hamburg manuscript, cited by Wieder, ibid.

[19] Rosh, Nedarim 23b.

[20] Tosafot, Nedarim 23b; Or Zarua, vol. 2, no. 277.

[21] Ritva, Nedarim 25b.

[22] Levush HaHor, no. 619, p. 408.

[23] Noheg KaTzon Yosef, Laws of Yom Kippur, p. 384.

[24] Ritva, Nedarim 25b.

[25] Shibbolei HaLeket, no. 317.

[26] Shitah Mekubetzet, Nedarim 23b.

[27] Shitah Mekubetzet, Nedarim 23b. Tosafot conclude their answer similarly (ad loc., s.v. Tanna setumei): “Nor should one tell them: ‘You are permitted.’”

[28] Responsa of Rivash, no. 394.

[29] Meiri, Hibbur HaTeshuva (Meshiv Nefesh), p. 518.

[30] Responsa of Radbaz, vol. 4, no. 33. It should be noted that Radbaz himself did recite Kol Nidrei, using a formula combining past and future.

[31] Ritva on Nedarim, Mossad Harav Kook ed., p. 231.

[32] For independent testimony that Kol Nidrei was not accepted at all in the Catalonian and Algerian rites, see L. Zunz, Die Ritus des synagogalen Gottesdienstes, Berlin 1859, p. 106; and Responsa Beit Yehuda of Rabbi Yehuda Ayash, 107a, no. 7.

[33] Rambam, Laws of Oaths 12:12. See Mishnah Hagigah 1:8.

[34] Responsa Havvot Yair, no. 15. The statement appears incidentally in a responsum concerning another subject and is introduced with the words “I have received a tradition.”

[35] Hadassi, Eshkol HaKofer, p. 53. Before him, Daniel al-Qumisi, Salmon ben Jeroham, and Sahl ben Mazliah had already denounced the custom, all citing the past-tense formula. Some scholars have suggested that the public recitation developed in response to them; J. Mann (Texts and Studies, vol. II) rejects this and demonstrates that the custom predates them, noting that it continued “despite the protests of the Babylonian Geonim.”

[36] Scholars have already noted that Hadassi’s very complaint—his insistence that the formula is sung—reveals the power of the melody in preserving the practice. See I. Davidson, “Kol Nidre,” American Jewish Year Book 25 (1923/24), p. 191.

[37] The accusation predates Donin: it is already found in 1235 in the context of attacks on the Talmud. See H. Merhavia, The Church Confronting Talmudic and Midrashic Literature, p. 262. From then on the accusation recurs repeatedly: in Antonius Margaritha (1530), an anonymous pamphlet of 1560, Wagenseil (1681), and Eisenmenger. See Y. Deutsch, Leo Baeck Institute Year Book 55 (2010), p. 48.

[38] Disputation of Rabbenu Yehiel of Paris, Margaliot ed., p. 15.

[39] Tosafot, Nedarim 23b; Rosh ad loc.; and Rabbenu Tam, cited in Shibbolei HaLeket, no. 317.

[40] Ramban, Mishpat HaHerem.

[41] Cited in R. Tasch, Samson Raphael Hirsch, p. 138.

[42] The Jewish oath (More Judaico): a special form of oath imposed upon Jews in European courts on the assumption that an ordinary oath did not bind a Jew. The person swearing was required to stand wrapped in a tallit and holding a Torah scroll, and to accept curses upon himself should he lie. Beginning in the eighteenth century, such formulas began to include explicit reference to Kol Nidrei.

[43] Cited ibid., p. 137. See also V. Zimmermann, Die Entwicklung des Judeneids, p. 237.

[44] See M. Breuer, “On the Abolition of Kol Nidrei in the Holy Community of Oldenburg,” HaMa‘ayan, Tevet 5724. Some have connected the episode to the struggle against the Jewish oath; Heinrich Graetz wrote in his diary that the change had been made at his own initiative.

[45] Horeb, chapter 70, secs. 472–473.

[46] The declaration was printed in Allgemeine Zeitung des Judenthums, year 6 (1842), p. 170, in the issue of 8 Nisan 5602, and is cited in Breuer’s translation, ibid.

[47] Ritva, Nedarim 25b: “There are places where they say it only once, and this appears correct.” Likewise Or Zarua in the name of Rabbenu Tam: “The law is with them.”

[48] Cited in Y. Emanuel, ed., Rabbi Samson Raphael Hirsch—His Thought and Method, p. 335.

[49] M. Breuer, ibid.

[50] Protocolle der ersten Rabbiner-Versammlung, Braunschweig 1844, p. 35.

[51] Ibid., pp. 37–38.

[52] Ibid., p. 42.

[53] Ibid., p. 40.

[54] Ibid., p. 42.

[55] Aruch HaShulchan, Orah Hayyim 619:3.

[56] Ritva, cited in Responsa of Rivash no. 394; Nimukei Yosef ad loc.; Responsa of Radbaz, ibid.

[57] Abudarham, Order of Yom Kippur; Orhot Hayyim, Laws of Yom Kippur. See I. Davidson, “Kol Nidre,” American Jewish Year Book 25 (1923/24), pp. 189–190.

[58] Tzafnat Pa‘aneah, Notes to Zohar HaRakia, prohibition 71, p. 276. The phrase “to make a matter public” is itself language of the Jerusalem Talmud, Sukkah 4:4.

[59] Rosh, Nedarim 23b, and Rosh, Yoma ch. 8, no. 28; Ritva, Nedarim 25b; Shibbolei HaLeket, no. 317; Nimukei Yosef ad loc.; Levush, no. 619. An additional explanation is cited in the name of Raavan: perhaps one did not hear the first time and will hear on the second or third.

[60] Levush, no. 619, p. 410.

[61] Mahzor Vitry, A. Goldschmidt ed., pp. 774–775. The passage survives in a single manuscript recension.

[62] Levush, ibid.

[63] Rema, Orah Hayyim 619:1, in the name of Maharil.

[64] Mishnah Shabbat 24:5; Nedarim 77a. Rosh already gives this reason at the end of Yoma; see Mishnah Berurah ad loc., sec. 5.

[65] Arugat HaBosem of Rabbi Abraham ben Azriel, Urbach ed., vol. 3, pp. 468–469; and Sefer Maharil, Customs, Laws of Prayer. See N. Wieder, “On Anti-Christian and Anti-Muslim Gematria in the Aleinu Prayer,” Sinai 76 (5735), pp. 1–14.

[66] The Latin translation of Aleinu in the inquisitors’ manual of 1323–1324 (Practica inquisitionis) is discussed by Y. H. Yerushalmi, Harvard Theological Review 63 (1970), p. 359. On Pessah-Peter (1399) and the response of Maharil of Mühlhausen, see Sefer Nizzahon, no. 347.

[67] For a comprehensive mapping of the manuscripts and printed editions, including the symbols inserted where the line had been removed, see R. Langer, “The Censorship of Aleinu in Ashkenaz and Its Aftermath,” in The Experience of Jewish Liturgy, Leiden 2011, pp. 147–166.

[68] Tur, Orah Hayyim 133. The stages of the textual change are documented by Langer, ibid., p. 165: first in the Slavuta and Kopys editions early in the century, and later in the Zhitomir and Warsaw editions. From the Vilna printing onward this became the text found in most editions to the present. The Jerusalem Institute edition restored the earlier reading on the basis of manuscripts and early printings.

[69] A. Margaritha, Der gantz Judisch Glaub, Augsburg 1530. On his attack on Kol Nidrei, see Deutsch, ibid., p. 48; on the defense of Aleinu at the same gathering, see Langer, ibid., p. 163. The anonymous pamphlet of 1560 (Verzeychnuß und kurtzer Ausszug) lists the two side by side.

[70] Censors deleted the objects of the curse in the blessing against the sectarians, and substitute words were introduced so that the blessing could continue to be recited. Some have connected this process to the development of the wording ve-la-malshinim. See R. Langer, Cursing the Christians? A History of the Birkat HaMinim, Oxford 2011.

[71] Langer, ibid., p. 162.

[72] S. Stern, Der preussische Staat und die Juden, Tübingen 1962, vol. 1.1, pp. 116–117; official record in vol. 1.2, p. 240, no. 270. The edict is dated August 28, 1703. Christian inspectors paid by the state were subsequently appointed to attend services and ensure compliance.

[73] The episode occurred in 1777. Langer, ibid., p. 164, based on Stern, ibid., vol. 3.1, pp. 334–341.

[74] On another dimension of this dynamic, I wrote in “Breaking in a Year What Was Built in a Decade,” Tzarich Iyun, Elul 5786.

[75] Kol Bo, no. 68; Orhot Hayyim, Order of Yom Kippur, no. 29. See likewise Rosh, Yoma ch. 8, no. 28, and Tur, no. 619.

[76] Jerusalem Talmud, Hagigah 1:8. See also Babylonian Talmud, Hagigah 10a.

The Voice That Would Not Be Silenced: Kol Nidrei, Halacha, and Jewish Memory

The Formula, and the First Opposition

The Emendation, and the Second Opposition

The Release and the Stipulation

Those Who Sought to Abolish It from Within

And Then Entered the One Who Had Not Been Asked

Oldenburg, Yom Kippur 5600

Braunschweig, 25 Sivan 5604

And Yet It Has No Effect

The Decision

What We Did Give Up

Write a Comment