Hostname: page-component-68c7f8b79f-r8tb2 Total loading time: 0 Render date: 2025-12-26T22:25:38.642Z Has data issue: false hasContentIssue false

Does an Arrow a Day Keep Satan Away? Late Antique Magical Subtexts in Babylonian Rabbinic Narratives

Published online by Cambridge University Press:  22 October 2025

Simcha Gross*
Affiliation:
University of Pennsylvania; simgross@upenn.edu
Rights & Permissions [Opens in a new window]

Abstract

A remarkable tale in the Babylonian Talmud (b. Qidd. 81a–b) recounts how Pelimo, who regularly exclaimed “an arrow in Satan’s eye!,” ironically attracts Satan’s personal attention. Disguised as a pauper, Satan wreaks havoc until he ultimately offers an alternative apotropaic formula—a biblical verse (Zech 3:2)—for Pelimo to use against him. While often read as a moral allegory, this article argues that the story is rooted in late antique Babylonian notions about demons and apotropaic practices, as evidenced in incantation bowls, amulets, and related objects. The narrative not only presumes this cultural context but actively engages it—contrasting two protective formulae and casting Satan as an advocate for one. In doing so, it reveals how rabbis participated in broader debates over the legitimacy and efficacy of different magical practices, using narrative as a tool of persuasion. Comparison with other Talmudic passages further highlights evolving rabbinic views on ritual power and the shifting role of biblical verses in Jewish magic.

Information

Type
Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of the President and Fellows of Harvard College

Introduction

According to a remarkable story in the Babylonian Talmud (b. Qidd. 81a–b), a figure named Pelimo regularly recited the formula “An arrow in Satan’s eye!”Footnote 1 This caught the attention of none other than Satan himself, who, in response, disguises himself as a pauper and takes advantage of Pelimo’s generosity to enter his house. After wreaking havoc, Satan reveals himself and admonishes Pelimo about his daily recitation. Pelimo asks Satan how else to repel him, and Satan advises Pelimo to recite a version of Zech 3:2, “The Merciful rebuke Satan!”

The sensational nature of the story—a direct confrontation between a man and Satan—has inspired subsequent adaptations and retellings, including in early twentieth-century Hebrew literature and modern American cinema.Footnote 2 It has likewise elicited many scholarly interpretations. Typically, these approach the story as a foundational moral tale, regarding the figures and their dialogue as fundamentally “metaphorical” and intended to convey a broader didactic message about human behavior.Footnote 3 Satan is “symbolic” of what a culture deems its opposite.Footnote 4 Different explanations have been offered for the story’s didactic message, ranging from a critique of Pelimo’s personal hubris or his lackluster hospitality. To others, the moral of the story is that “Satan” —figurative of one’s internal desires —is simply unconquerable without God’s help.Footnote 5 In short, “Satan” in the story is either a divine figure exacting punishment for misdeeds, or a personification of Pelimo’s “inner demons.”Footnote 6 The formula recited by Pelimo and the alternative offered by Satan are not in themselves meaningful, but somehow emblematic of the larger underlying values contested in the story.Footnote 7 Pelimo’s daily recitation and Satan’s alternative of Zech 3:2 represent the character development that Pelimo supposedly undergoes from the opening to the conclusion of the story.

According to these interpretations, the story contemplates the virtues and characteristics of humankind; it is therefore not, we are told, “enslaved” to its “historical or sociological context.”Footnote 8 The assumed self-contained and transcendent quality of this story is a lingering product of an approach deriving from New Criticism according to which each rabbinic story “supplies all the information necessary for its interpretation.”Footnote 9 These stories, instead, were intended to provide lasting instruction.

An emphasis on the enduring didactic quality of rabbinic stories forecloses possible historicist and contextual interpretations. Yet over the past few decades, an ever-growing body of scholarly literature has demonstrated the advantages of examining Babylonian rabbinic stories against the backdrop of their broader social and cultural milieux, in which they are deeply embedded.Footnote 10 Attention to the historical context makes legible not only crucial details, motifs, and realia mentioned in and presupposed by rabbinic stories, but also situates their instructive purpose, agenda, and tendenz in salient concerns of the time. These works recast the rabbis not as expositors of timeless values, but as subjects informed by and intervening in their own particular historical moment.

A more recent subset of research has demonstrated that Babylonian rabbinic texts share a common late antique Babylonian cosmology replete with a complex and highly delineated angelology and demonology controllable through various apotropaic and prophylactic practices.Footnote 11 That is, Babylonian rabbinic texts are conversant with the contemporary and geographically proximate beliefs and practices conventionally grouped under the problematic modern category of “magic.”Footnote 12 These studies therefore trouble older binaries that distinguished between the “popular masses,” who were supposedly mired in superstitions of the unlearned, and the scholastic elite represented by the rabbis, who rejected them.Footnote 13 Instead, these studies show that the rabbis were active participants in a larger world of discourses about controlling their enchanted cosmos, sublimating and “rabbinizing” earlier practices and even preexisting textual units to better accord with rabbinic ideas and ideals.Footnote 14 Key to these conceptual developments have been the ongoing publication and study of the Aramaic Incantation Bowls, ceramic bowls adorned with varying incantations in three dialects of Eastern Aramaic, but mainly in Jewish Babylonian Aramaic.Footnote 15 The bowls were discovered in the very regions in which the Babylonian rabbis dwelled, and divulge various direct relationships with both rabbinic literature and individual rabbis.Footnote 16

Through a study of the story of Pelimo, this article explores how rabbinic stories and discussions not only assumed knowledge of common apotropaic practices, but also served as vehicles through which rabbis adjudicated between these practices and determined which were licit and illicit, effective and ineffective. Rather than attesting a situation in which rabbinic piety involved the rejection or renunciation of popular practices, the story exemplifies how the rabbis at times selected between already widely available options, using narrative as a tool of persuasion. In turn, a comparison of the story of Pelimo with other late Babylonian rabbinic texts suggests a broader chronological development within Babylonian rabbinic thought about apotropaic practices, and demonstrates the benefits of attending to the composition history of the Talmud in the study of so-called magical and medical practices.

The Story of Pelimo

The story of Pelimo appears within a collection of interrelated stories.Footnote 17 It is preceded by two others in which Satan disguises himself as a woman to tempt and expose the hypocrisy of rabbis who scoffed at sinners, and appears just before a story about a rabbi who would regularly perform proskynesis and recite, “The Merciful One, save me from the evil inclination,” but is then tempted by his wife after she disguises herself.Footnote 18 The various stories in the collection have different associative links, including supernatural beings, disguise, and daily recitations. Our focus, however, will be on the story of Pelimo itself, which reads as follows:Footnote 19

פלימו הוה רגיל לומר גירא בעינא דשטנא

יומא חד מעלי יומא דכיפורי הוה

אידמי ל’ כעניא אתא קרא אבבא

אפיקו ל’ ריפתא

אמ’ יומא כי האידנא כולי עלמא אגואי ואנא אבראי

עיילו קריבו ריפתא קמיה

אמ’ יומא כי האידנא כולי עלמ’ אתכא ואנא לחודאי

אתיוה אותבוה אתכא

הוה מלי נפשיה שחינא וכיבי והוה קא עביד מילי דמאיסותא

אמ’ ל’ תיב שפיר

הבו ליה כסא אכמר שדא ביה

נחרו שקא ומית

שמע דהוו קאמרי פלימו קטל גברא

ערק טשא בבית הכסא אזל נפל קמיה

חזייה דהוה מצטער גלי ליה נפשיה א”ל ומאי טע’ אמרת הכי

ואלא היכי כו נימא

א”ל אימ’ רחמ’ יגער יי בך השטן

Pelimo would regularly say, “An arrow in Satan’s eye!”

One day, it was the eve of the Day of Atonement.

He [Satan] disguised himself as a poor man, came, and cried out at [Pelimo’s] door.

They brought out bread to him [Satan].

[Satan] said: “On a day like this, when everyone is within, and I am outside!?”

They brought himFootnote 20 in and served him bread.

[Satan] said to him: “On a day like this, when everyone sits at the table, and I am sitting alone!?”

They led him and sat him at the table.

He covered himself with boils and sores, and he was behaving repulsively.

[Pelimo] said to [Satan]: “Sit properly!”

They gave him a cup. He turned and spat into it.

They scolded [him], he [Satan] sunk and died.

He [Pelimo] heard that people were saying, “Pelimo killed a man!”

[Pelimo] ran and hid himself in a privy. [Satan followed him and]Footnote 21 he [Pelimo] fell before him.

When [Satan] saw that [Pelimo] was suffering, he revealed himself and said to him: “Why did you say this [‘An arrow in Satan’s eye’]?”

[Pelimo replied:] “And what then should we say [to keep you away]?

He [Satan] said to him: “Say ‘the Merciful One, may the Lord rebuke you Satan’ (version of Zech 3:2).”

The story raises several interpretive questions. Why does Pelimo regularly recite “An arrow in Satan’s eye”? Indeed, Pelimo’s recitation structures the entire narrative, as it initiates the story and its replacement by a different formula concludes it. Why does Satan take umbrage at this invocation and react the way he does? Why does Satan instead recommend the recitation of Zech 3:2, and what makes it preferable to Pelimo’s recitation?

“An Arrow in Satan’s Eye”

Pelimo’s invocation is found in two other pericopae in the Babylonian Talmud. One appears in the context of several anecdotes about rabbis who extol early marriage as a means of achieving purity of thought and deed. There, Rav ḥisda (c. early fourth century) declares:Footnote 22

[ב]הא עדיפנא מחבראי דנסיבני בשיתסר,[ואי]נסיבנא בארבסר, הוה אמינא ל’ לסטן גיר’ בעיניה

In this I am superior to my colleagues that I married at sixteen, and had I married at fourteen, I would have said to Satan, “An arrow in his eye!”

Rav ḥisda argues that his early marriage set him apart from his colleagues; however, had he married even earlier, he claims, he would have been able to declare to Satan, “An arrow in his eye!” This statement is often understood euphemistically, suggesting that his conduct would have been so pure that it would have been tantamount to fending off Satan with an arrow. However, considering the Pelimo story, it is clear that “An arrow in Satan’s eye” is not an expression but an invocation that was recited by individuals. Rav ḥisda is therefore saying one of two things. He may be saying that he would only have been qualified to make such a bold invocation had he married earlier and thereby been pure of thought and deed. The implication would be that the invocation is accompanied by some risk and should only be deployed by the especially righteous. Alternatively, Rav ḥisda is saying that marriage at fourteen is as effective a means of repelling Satan as reciting “An arrow in Satan’s eye.” Through this comparison, Rav ḥisda praises both the ability of early marriage to remove temptation and the invocation as a particularly potent means of repulsing Satan.

The formula appears again in a discussion concerning the two commandments that include a requirement to wave a ritual item on a festival: the offering of loaves of bread and lamb on Shavuot, and the four species on Sukkot.Footnote 23 A Mishnah in MenaḤot (5:6) details how the bread and lamb offerings brought on Shavuot are to be waved: back and forth and up and down. This is based on the seemingly redundant verbs of Exod 29:27 concerning the offering, which is both “waved” and “heaved up.” In the Talmudic pericope, Rabbi YoḤanan explains that this act of waving in four directions is intended to recognize that God is the ruler over heaven, earth, and the four winds, i.e., above, below, and all around.

The Babylonian Talmud continues with the comment that “in the West,” they teach that the reason the priest waves this offering in two sets of directions is to restrain wicked spirits (רוחות) and wicked shadows (טללים), common terms for malevolent forces that harm humans. Another Palestinian rabbi derives from this that even non-essential parts of commandments, such as waving, prevent calamities (מעכבין את הפורענות). The ritual act therefore wards off malevolent forces which reside in all directions.

At this point, the Babylonian rabbi Rava declares: “So too with regards to lulav,” meaning that like the sacrificial offerings, lulav, too, is a ritual act involving waiving that effectively repels malevolent forces.Footnote 24 Rava’s determination is followed by a brief anecdote about another Babylonian rabbi who used the lulav to repulse not just any demons, but Satan:Footnote 25

רב אחא בר יעקב ממטי ליה ומייתי ליה ואמ’ האי גירא בעינא דשטנא.

R. AḤa b. Jacob waved it [the lulav] to and fro, saying, “This is an arrowFootnote 26 in Satan’s eye.”

Even as Rav AḤa’s practice is clearly relevant to the larger discussion about the ability of ritual acts to ward off malevolent forces, his practice departs from it in two significant ways: he does not wave the lulav in four directions, but apparently only to and fro, and he accompanies this act with the utterance, “This is an arrow in Satan’s eye.” These two departures are interrelated; it appears that Rav AḤa is waving the lulav forward and backwards to mime the act of piercing Satan’s eye, meaning that the lulav is here functioning as the arrow attacking Satan’s eye, as the demonstrative pronoun in his invocation makes clear.Footnote 27

Intriguingly, this brief anecdote is followed by a short anonymous editorial interpolation that censures Rav AḤa’s practice:

.ולאו מילתא היא משום דאתי לאיגרויי ביה

This is not appropriate, because it will incite him [Satan] against him.

The anonymous comment condemns Rav AḤa’s action because it is likely to incite Satan against the performer, the opposite of the outcome that Rav AḤa intends.Footnote 28 The brief gloss shows a disagreement among the rabbis; Rav AḤa believed in the utility of the utterance, whereas the later gloss considers it dangerous. This same apprehension might also be found in Rav ḥisda’s statement that if he were married at fourteen, he would be qualified to say “An arrow in Satan’s eye”; only those entirely free from temptation could recite the formula without fear of inciting Satan against them. Rav ḥisda recognizes the apparent efficacy of this utterance, but also that only some are qualified to recite it.

For its part, the story of Pelimo narrativizes the tension between Rav AḤa’s practice and the critical anonymous gloss. Pelimo, like Rav AḤa and Rav ḥisda, believes in the efficacy of this utterance, yet by reciting it, Pelimo incites Satan against him, just as the anonymous gloss warned in the case of Rav AḤa. Yet whereas in the case of Rav AḤa, the anonymous gloss represents a secondary editorial interpolation, a later condemnation of an act Rav AḤa deemed appropriate, the story of Pelimo appears crafted from the outset to question the wisdom of reciting “An arrow in Satan’s eye.” This suggests a chronological development, between earlier Amoraim, like Rav AḤa and Rav ḥisda, who employed some version of this formula, and the anonymous gloss and the story of Pelimo, which were composed to combat it.

“An Arrow in Satan’s Eye” and the Much Suffering Eye

The fact that the same formula appears in three different Talmudic pericopes suggests that it enjoyed some degree of popularity. But the basis for this formula remains unclear, as does the reason it came to be rejected by the composer(s) of the story of Pelimo and the editorial gloss of the story of Rav AḤa. As a few scholars have noted, this utterance is an apotropaic formula of some sort, similar in nature to other non-scriptural apotropaic formulae found within rabbinic literature and in Christian and Muslim sources in the first millennium.Footnote 29

Yet the particular expression of a sharp object piercing a demonic eye is more directly reminiscent of an extremely popular and pervasive apotropaic image found across the Mediterranean in late antiquity: the so-called much suffering eye. The image—found on mosaics, marble plaques, and pendant amulets—depicts an eye attacked by various forms of sharp objects and aggressive animals, the former of which includes tridents, swords, spears, and daggers or arrows, and the latter includes scorpions, lions, snakes, and various birds and horned creatures. Spatially, the sharp inanimate objects typically strike from above, whereas the animals mainly strike from below. Like other late Hellenistic and Roman amuletic and apotropaic images, the depicted action serves by means of analogy to threaten the eye and thereby rebuff it.Footnote 30

Mosaic in the House of the Evil Eye, Antioch; Second Century (Wiki Commons)

Pendant Amulet; Walters Art Museum 54.2653 (Wiki Commons)

Significantly, the image of the much suffering eye appears on a ceiling tile in the synagogue in Dura Europos in upper Mesopotamia, the only ancient synagogue discovered proximate to the regions in which Babylonian rabbis lived.Footnote 31 There, the eye in the image is attacked by a beetle from below, by two snakes on the side, and by three identical objects from above, each labeled with a letter from the Greek name of Yahweh, Iaô.

The much suffering eye served to repulse a range of different malevolent forces, whose identities regularly bled together, from particular demons, to the evil eye, to the demonic more generally.Footnote 32 The amalgamation of these possibilities is seen from the Testament of Solomon (18:39), which describes the apotropaic use of the image.Footnote 33 There, a demon named Rhyx Phtheneoth declares: “I cast the evil eye on every man. But the much suffering eye (ὃ πολυπαϑὴς ὀφϑαλμὸς), when inscribed, thwarts me.” To this text, the much suffering eye is a recognized amuletic motif which thwarts a particular demon, but that demon is responsible for deploying the evil eye against humanity. The evil eye may similarly have been conceived as simply one minion or manifestation of Satan himself.Footnote 34 Indeed, some late antique and later incantations depict the evil eye as an independent demon, an anthropomorphized figure with an eye for a body that is stabbed by the holy rider.Footnote 35

The formula “An arrow in Satan’s eye” is therefore likely an oral iteration of a common apotropaic image intended to symbolically attack the eye of the demonic force in order to deter it.Footnote 36 Indeed, this background explains why the formula in the Talmud targets the single eye of Satan rather than both of his eyes. The existence of oral iterations of an apotropaic image is not unique to this circumstance; we find it, for instance, in the case of Greek incantations.Footnote 37 However, whereas in those cases the oral iteration likely precedes the image, here it seems likelier that the oral iteration is an adaptation of the image.Footnote 38

The Babylonian Talmud may not be the only late antique Babylonian text to reflect a textual or oral iteration of the much suffering eye. The Shafta ḏ-pishra ḏ-ainia or the Scroll of the Exorcism of the Eyes, a Mandaic collection of incantations against the evil eye(s), includes a striking, even more comprehensive, verbal description of the much suffering eye.Footnote 39 Though the text only survives in modern copies, the work contains several near identical parallels with the Aramaic Incantation Bowls, suggesting that its recipes, or some versions thereof, largely date to late antiquity.Footnote 40 One formula reads:Footnote 41 “Tremble and be scared off, Evil Eye and Dimmed, from the body of N. son of N.! May the snake bite you and the scorpion sting you and the centipede sting you and the gander peck you and the sword cut you. . . .” This incantation almost perfectly reproduces the image of the much suffering eye. Indeed, every single attacking object described here appears on the much suffering eye mosaic in Antioch, and the individual elements are found in other iterations of the image. Another recipe in this Mandaic work similarly threatens the evil eye with various harms, including a snake, scorpion, crab, falcon, crane/goose, and kite.Footnote 42 These Mandaic incantations are further proof that the much suffering eye was rendered into words by others in late antique Babylonia.Footnote 43

However, as opposed to the two Mandaic incantations, the version used by the rabbis selects only one sharp object to use against the eye: the arrow. The significance of this may lie in the fact that in many manifestations of the much suffering eye, the sharp objects attacking the eye from above are wielded by a deity. In particular, in several iterations of the motif, a trident is rammed into the eye from above by Poseidon himself, as depicted in the marble plaque from the collection of Woburn Abbey. The trident also appears without Poseidon, such as in the Mosaic in the House of the Evil Eye in Antioch, but one might suggest that Poseidon’s presence is here implied. Over the course of late antiquity, the trident, and in time, three distinct smaller sharp objects which derive from the three spikes of the trident, continue to appear above the eye in pendant amulets, without Poseidon grasping it. However, these are now accompanied by formulae naming the deity, such as Ἰἁω Σαβαὠθ Μιχαἠλ βοἡθι (Iao Sabaoth, Michael, help), or more basically, Eἷς Θεός (the “One God”).Footnote 44 Fascinatingly, the ceiling tile in Dura Europos seems to have a simplified version of this motif.Footnote 45 Instead of a trident from above, it too has three spikes, and each is accompanied by one letter from God’s name in Greek, Iaô. These amulets and images therefore make explicit that it is a divine actor threatening the eye with violence.Footnote 46

The development of the motif may explain the tension in the Talmud between the use of the formula by earlier Amoraim and its rejection by later Talmudic editors and composers. I suggest that originally the oral utterance of “An arrow in Satan’s eye” was formed not simply by picking a single weapon at random from the motif of the much suffering eye, but by selecting the sharp object thrust by a deity. The formula, in its origins, may have intended that it was God wielding the arrow to attack Satan, and the implied divine authority behind the invocation was clear to the rabbis who recited it.Footnote 47 In time, however, the implied presence of the deity, and the association with the image of the much suffering eye, may have been lost.Footnote 48 As a stand-alone utterance, it troubled later editors and composers, to whom it seemed audaciously to pit the reciter as the wielder of the arrow against Satan himself.

Satan’s Resistance and Misbehavior

As a story about the (in)effectiveness of a particular apotropaic utterance to repulse Satan, Satan’s behavior throughout his encounter with Pelimo is consistent with what we would expect based on the prevalent demonology in late antique Babylonia.

First and foremost, in seeking entry into Pelimo’s house, Satan attempts to accomplish precisely what incantations sought to prevent: the entrance of malevolent forces into the home of the client.Footnote 49 Indeed, for this reason, incantation bowls and other apotropaic incantations and objects were most commonly placed at the threshold of a house or individual rooms therein.Footnote 50 A regular class of demons listed in the bowls among those seeking entry is the “satans” (סטני).Footnote 51

The means by which Satan in the story enters the home takes advantage of real social practices evidenced by several Babylonian rabbinic sources and incantation bowls.Footnote 52 Satan approaches Pelimo’s home on the eve of the Day of Atonement disguised as a pauper and asks for sustenance. Charity in the form of food, liquid, and other needs regularly occurred at the entrance or doorway of homes in Babylonia, as depicted in rabbinic stories of highly charitable individuals.Footnote 53 Festivals in particular were occasions for charity of this kind, and it was especially meritorious to eat before the Day of Atonement and provide for others to do the same.Footnote 54 It is likely not by chance that Satan arrives on the eve of the Day of Atonement, as elsewhere in the Babylonian Talmud it is identified as the one day a year when Satan does not have permission to “accuse.”Footnote 55 Perhaps the setting is meant to heighten the sense of the ineffectiveness of the invocation “An arrow in Satan’s eye,” which incites Satan even on the one day a year when he was supposedly powerless.

That Satan is asking for something as basic as food and liquid is consistent with the idea found in both the Babylonian Talmud and the incantation bowls that demonic forces had corporeal qualities, among them the consumption of food.Footnote 56 Indeed, several bowl incantations feature a particular formula that, paradoxically, fends off malevolent forces from one’s home precisely by inviting them to enter and consume food and liquid, but solely on the condition of peaceful entry, requiring them otherwise to leave altogether.Footnote 57 This background underscores the extent to which Pelimo’s original utterance failed: Satan accepts Pelimo’s charity but uses it to infiltrate his home. Predictably, once in the house, Satan wreaks havoc, by flouting social conventions and eventually jeopardizing the head of the household, Pelimo.Footnote 58 Meaningfully, the language used to describe how Satan’s body “was covered with boils and sores” has a verbatim parallel to another story in the Talmud, which also appears in a context related to “magic.”Footnote 59 There, a rabbi utters some form of incantation to generate “boils and sores” on his body in order to repel a woman seeking to seduce him, whereas here it is Satan who generates the boils and sores to repel the members of Pelimo’s household.

Upon being accused of killing the pauper, Pelimo flees to a privy where he is directly confronted by Satan. The privy was a common site of malevolent forces in antiquity.Footnote 60 Both Jewish and non-Jewish sources thematize the dangers represented by the concentration of malevolent forces in bathrooms. Within rabbinic literature, the concern for the demonic in bathrooms is particularly pronounced in Babylonian sources.Footnote 61 The Babylonian Talmud offers multiple blessings for entering and exiting the toilet, some of which parallel those found on Aramaic incantation bowls.Footnote 62 Pelimo therefore encounters Satan in two places where one would expect to find him based on the demonology prevalent at the time: the home and the privy.

Indeed, the demonological context of the story can be seen by examining a clear rabbinic literary parallel, both in terms of content and language.Footnote 63 In another Babylonian rabbinic story, the Angel of Death is prevented from ending Rabbi Ḥiyya’s life because he never ceases studying Torah. To distract him, the Angel of Death appears as a pauper (אידמי ליה כעניא) and knocks on his door to ask for bread (אתא טריף אבבא, אמר ליה: אפיק לי ריפתא). When bread is brought to him (אפיקו ליה), the Angel of Death asks why R. Ḥiyya is not honoring him, presumably by personally greeting him. When R. Ḥiyya opens the door the Angel of Death reveals himself (גלי ליה) to him and immediately takes his life by means of his instrument, the fiery rod.

Despite their obvious similarities, the stories depart precisely in terms of the character traits and objectives associated with the Angel of Death and Satan respectively. The Angel of Death is not provoked in any way, he is simply doing his job and seeking to take R. Ḥiyya’s life at the appropriate time. By contrast, Satan is provoked by Pelimo’s utterance, and responds accordingly. The Angel of Death seeks to distract R. Ḥiyya from his learning in order to end his life, and so begs for him to greet him at the entrance. Satan wishes to enter the house itself, where he wreaks havoc. The Angel of Death’s encounter with R. Ḥiyya ends immediately upon his death, whereas Satan does not kill or even physically harm Pelimo as much as harass him, even relocating to the privy, a site of demonic habitation. Both stories end with the “reveal” of the supernatural figure behind the disguise.

The composer of the story of Pelimo therefore employed a shared literary template of an encounter between a human and supernatural being, but in so doing highlighted crucial differences between them. A story about the successful repulsing of one supernatural being (the Angel of Death) by means of a speech act (Torah study) is transformed into a story about inciting another supernatural being (Satan) through provocative speech (“An arrow in Satan’s eye”), and ends precisely by instructing Pelimo—and the story’s audience—about how to successfully thwart Satan in the future.

Zechariah 3:2

At the story’s conclusion, Satan recommends that Pelimo recite a version of Zech 3:2, which in one set of manuscripts is rendered entirely in an Aramaic paraphrase—“May the Merciful One rebuke Satan”—while in the other is a hybrid of Aramaic (“O Merciful One,” רחמנא) and the Hebrew of the verse.Footnote 64 Why does Satan recommend this formula in particular?

Zechariah 3:2 appears in the context of Zechariah’s vision of Joshua the High Priest, a crucial if elliptical figure in the return of Jews from the Babylonian exile. In the vision, Joshua stands before the “angel of the Lord,” while Satan stands to Joshua’s right and accuses him.Footnote 65 At this point God intervenes:

ויאמר יהוה אל השטן יגער יהוה בך השטן ויגער יהוה בך הבחר בירושלם הלוא זה אוד מצל.מאש

The Lord said to Satan, “The Lord rebuke you, Satan! The Lord, who has chosen Jerusalem, rebuke you! Is this one not a burning stick snatched from the fire?”

In the verse, God himself “rebukes” Satan.Footnote 66 The verb for “rebuke” (גער) appears elsewhere in the Bible in similar interpersonal contexts: Jacob rebukes Joseph for his dream (Gen 37:10), a letter rebukes Jeremiah (Jer 29:27), God rebukes the sea, restoring order to chaos (2 Sam 22:16; Nah 1:4; Ps 18:16; Job 26:11–13), and God’s rebuke is likened to fire (Isa 66:15). While the verbal root is conventionally translated as rebuke, as I have rendered it throughout, the semantic range can be more accurately described as denoting an effective speech act, and in the case of God’s usage, as “the deity’s effective power against his enemies via the use of speech.”Footnote 67

This expanded semantic range explains a second related sense of the same root in Hebrew, Aramaic, and Ugaritic apotropaic texts: to exorcise or combat malevolent forces.Footnote 68 According to the most recent edition, this usage is already found in the Ketef Hinnom Amulets, Hebrew incantations dating from the seventh to sixth centuries BCE.Footnote 69 The reconstructed text of Amulet B now includes the line “May h[e]/sh[e] be blessed by Yahweh, the warrior [or: helper] and the rebuker (הגער) of [E]vil.”Footnote 70 This usage is found as well in several of the Dead Sea Scrolls.Footnote 71 The root continues to be highly productive in late antique Jewish amulets in both Palestine and Babylonia.Footnote 72 The verb is similarly used in the New Testament, where ἐπιτιμάν describes Jesus’ exorcistic activities, the same root used to translate גער in the LXX.Footnote 73

Given the semantic range of this root, Zech 3:2 came to be understood not only as God “rebuking” Satan similar to Jacob rebuking Joseph, but as God exorcising or dismissing Satan himself. Invoking the verse came to reenact God’s rebuke of Satan.Footnote 74 The verse therefore appears on a couple of late antique Jewish amulets.Footnote 75 Strikingly, in a passage in Jude 9, in what seems to be an allusion to a now lost Second Temple text describing an exchange between the angel Michael and Satan, Michael chooses not to “accuse Satan of blasphemy” (κρίσιν ἐπενεγκεῖν βλασφημίας), but instead recites Zech 3:2 to rebuke Satan.Footnote 76 As in the story of Pelimo, here Michael prefers Zech 3:2 over a different utterance against Satan. The appeal of this verse as an utterance in amulets and in Jude 9 is clear: like the analogical operation behind the much suffering eye, here God’s own action in the past is invoked to apply in the present.Footnote 77

Though this verse appears on only a few western late antique Jewish amulets, it is one of the most cited verses in the incantation bowls, featured on roughly twenty-five of those published to date.Footnote 78 The verse’s appearance on the bowls is itself telling. In most cases, it appears at the very end of the incantation, often following a natural concluding point, such as the affirmation, “Amen Amen Selah.”Footnote 79 This suggests that scribes saw it as a potent additional invocation, or even as a means of applying God’s rebuke of Satan to the entirety of the preceding incantation.Footnote 80 Indeed, a set of parallel incantations provide a window into this scribal operation, as one scribe saw fit to add the verse to the end.Footnote 81 To be sure, verses are often cited at the end of bowl incantations, but the frequency of Zech 3:2 in this position is noteworthy.Footnote 82

While Zech 3:2 often appears alone at the end of incantations, it is also integrated alongside other scriptural citations in this position.Footnote 83 When these passages are compared, one can surmise that what made them particularly attractive as apotropaic formulae was that all were likely understood to reflect God’s sovereignty and power over other forces in the cosmos. In two bowls, Zech 3:2 appears alongside Deut 6:4 and Ps 91:1. The former is of course the verse from the Shema which declares that God is “one” or “alone,” implying his supremacy over all other divine forces, while Ps 91:1 is the opening of a Psalm in which the protagonist is under God’s shadow, and God serves as his “refuge and fortress” against various forces, including pestilence, terror of the night, arrows, plague, and more. Similarly, in two cases Zech 3:2 appears alongside Num 9:23 (in one case also with Deut 6:4), another popular verse in the bowls. Though in its original context this verse is about the Israelites’ journeying and camping at God’s command, in the context of the incantation bowls the verse appears to imply that all forces—including the angelic and demonic— are ultimately commanded by God Himself.Footnote 84 Other verses—Num 12:13, Ps 121:7, and Is 51:14— refer to God’s ability to heal, to protect, and to loosen those bent down by “the oppressor” (המציק), or in Ps 55:9, man fleeing from “the stormy wind,” all easily applicable to the context of repelling demons.Footnote 85

The shared practice behind the use of Zech 3:2 in the bowls and in the story of Pelimo is further on display in the fact that, in the story, Satan does not recommend the recitation of Zech 3:2 in Hebrew, but in an (at least partially) Aramaic paraphrase.Footnote 86 The use of an Aramaic paraphrase is itself reflective of the practice in some bowl incantations to cite Aramaic translations of biblical verses, sometimes without any clear relationship to the surviving formal literary Targumim.Footnote 87

The story of Pelimo is in fact not the only Babylonian rabbinic text to embrace the recitation of Zech 3:2. In another passage (b. Ber. 51a), Rabbi Joshua b. Levi lists three pieces of advice given to him by the Angel of Death regarding situations to avoid, in order to escape inadvertently bringing harm upon oneself. The Angel of Death’s final piece of advice is to avoid standing before women when they return from visiting the dead. The Angel of Death explains that in this scenario, “I go leaping before them with my sword in hand, and I have permission to harm.” This Hebrew saying is followed by an anonymous Aramaic gloss which undoubtedly postdates the original statement attributed to Joshua b. Levi. It asks, “And if one happens upon [these women], what is the remedy?” That is, the gloss seeks a solution if the initial attempt to avoid complications was unsuccessful. It offers a series of possibilities, which include turning aside four cubits, crossing the river, taking another path, and hiding behind a wall. The gloss continues, “And if he cannot do any of these things, let him turn his face away and say, “The Lord said to Satan, ‘The Lord rebuke you, Satan,’ until they have passed.” Here a late Babylonian rabbinic gloss again presents Zech 3:2 as an effective apotropaic remedy even after other solutions are unavailable.Footnote 88 This source shares a thematic relationship with the story of Pelimo, in that both the Angel of Death and Satan offer advice on how to successfully keep them away and avoid harm at their hands.Footnote 89

The story of Pelimo therefore rejects one apotropaic formula popular in late antique Babylonia in favor of another.Footnote 90 Far from departing from or critiquing popular practices, here the rabbis select between them, using Satan himself as their mouthpiece. Indeed, the story offers a rationale for the popularity of Zech 3:2 in the incantation bowls and apparently in the eyes of the later Talmudic editors responsible for b. Ber. 51a. On a basic level, the story reflects a preference for scriptural citations over other invocations with no basis in scripture. The story however likely goes even further, suggesting that Zech 3:2 corrects for the problems with “an arrow in Satan’s eye”; where the latter pits the reciter directly against Satan, thereby endangering himself, the former invokes God as the prototypical rebuker of Satan. The story therefore expresses the importance of the authority behind invocations, a concern we find in other rabbinic texts and a familiar trajectory in the development of ancient incantations more generally.Footnote 91 Moreover, judging by the story, scriptural texts were not selected willy-nilly, as a general act of “piety,” but rather for their applicability to the circumstances.Footnote 92 This is borne out from an examination of the scriptural citations used in the incantation bowls; though bowl scribes display some flexibility in terms of the scriptural passages they invoked, the most commonly cited texts were those in which God’s activity could be applied to the context of the incantations themselves. Not only do the rabbis not stand apart from common practice, but they offer a rationale for it, through the particular medium of the rabbinic story.

Apotropaic Use of Scripture and the Later Babylonian Rabbis

As we have seen, the story of Pelimo offers a similar critique of “an arrow in Satan’s eye” as that of the anonymous gloss to the story of Rav AḤa, and it recommends the recitation of Zech 3:2 as does the anonymous gloss in b. Ber. 51a. Coupled with its use of a literary template and features found in other stories in the Talmud (e.g., b. Moʿed Qaṭ. 28a), this suggests that the story originated in the period of the Talmud’s redaction, generally dated to the sixth century, precisely the period of the incantation bowls’ production.Footnote 93 Together, the correspondences between the story and these various editorial additions in the Talmud indicate a growing interest among the Talmud’s later editors in delimiting the use of scripture in prophylactic contexts.

This pattern is visible throughout the Babylonian Talmud. For instance, in one place, the anonymous editors distinguish between licit and illicit uses of biblical verses.Footnote 94 They reconcile an apparent contradiction between an anecdote about a rabbi who, on the one hand, recited verses before bed and, on the other, stated that one may not recite verses to heal oneself, by distinguishing between reciting verses to heal, which is proscribed, and reciting verses for protection, which is permissible.Footnote 95 In this same pericope, the anonymous editors further qualify the famous early Tannaitic statement by Rabbi Akiva that “one who whispers over a wound” (הלוחש על המכה) the verse Exod 15:26 (“Every illness that I placed upon Egypt I will not place upon you, for I am the Lord, your Healer”) has no share in the world to come.Footnote 96 The anonymous editors cite the opinion of a R. YoḤanan b. Beroka that Rabbi Akiva’s opinion only applies if one spits on the wound while reciting the passage, because one may not recite the name of God over spit. While R. YoḤanan’s cryptic limiting statement is cited elsewhere in the Talmud without elaboration (b. Sanh. 101a), seemingly only commenting on the permissibility of the recitation of Exod 15:26, the anonymous editors craftily deploy his opinion here to undermine a major Tannaitic precedent and dismiss any possible objection to the use of scripture for prophylactic purposes more generally.Footnote 97

In another case, in the so-called “Talmudic Dreambook,” the anonymous editors again update an earlier discussion to include the prophylactic recitation of verses. In the pericope, a chain of rabbis state that if one has an ominous dream, they should “go and have it interpreted before three [people].”Footnote 98 The anonymous editors interject that this opinion contradicts that of Rav ḥisda who said that an uninterpreted dream was like an unread letter, either meaning that it had no force until interpreted or that its meaning could only be known upon interpretation.Footnote 99 The anonymous editors, again switching to Aramaic instead of the Hebrew of the original dictum, clarify that one should “spin [the dream] positively before three.”Footnote 100 They continue to detail the procedure:

He should go and sit before three [some mss: who love him] and say to them “I had a good dream.”

And they should respond to him seven times “it was good, and may it be good, and may the good Merciful One make it good, and from heaven may it be decreed upon himFootnote 101 that it be good, and may it be good, and good may it be.”

And they should say to him three [verses with the word] “overturn” (הפך), and three with “redeem” (פדה) and three with “peace” (שלום) concerning saving, and three concerning peace: [continues to cite the following verses for overturn: Ps 30:12, Jer 31:12, Deut 23:6; for redeem: Ps 55:19, Isa 35:10, I Sam. 14:45; for peace: Isa. 57:19, I Chr. 12:19, I Sam. 25:6].Footnote 102

The anonymous editors introduce an entirely new ritual to counteract a bad dream, one which includes recharacterizing the nature of the dream, repetitive formulae, and the use of biblical verses with words that directly address the concern posed by a bad dream.

Like the prescription to recite Zech 3:2 in b. Ber. 51a, the anonymous editors elsewhere also prescribe verses in specific contexts, in each case using the same introductory formula: “what is its remedy” (מאי תקנתיה). In b. PesaḤ. 111a, if a man encounters a woman returning from purifying herself after her menstrual period, whichever of them has intercourse first is taken by the “spirit of harlotry.” The anonymous editors ask “what is its remedy” (מאי תקנתיה), and prescribe the recitation of Ps 107:40, according to some manuscripts, or the very similar Job 12:21, according to others. The passage includes other anonymous glosses that ask “what is its remedy” (מאי תקנתיה) before offering solutions.Footnote 103 A few lines earlier, it says that men should avoid having a menstruating woman walk between them. The editors ask “what is its remedy” (מאי תקנתיה), and answer that they should recite a verse that begins and ends with God/אל, which commentators suggest refers to Num 23:22–23. Curiously, just before this, the Talmud presents a Tannaitic statement about never passing between or being passed between by a dog, palm tree, or woman, and according to some, also a pig and snake. The Talmud once again asks if one does indeed pass them, “what is its remedy” (מאי תקנתיה). Yet here, the remedy is attributed to Rav Papa, who says that one should recite a verse that begins and ends with God (אל), just as the ensuing anonymous comment suggests. An alternative view attributed to “others” however apparently suggests that one should instead say a verse that begins with no (לא) and ends with “to him” (לו), which likely refers to Num 23:19. It is possible that the opinion attributed to Rav Papa was in fact stated by the anonymous editors and only received an attribution at a later date, but even if the attribution to Rav Papa is original, it is noteworthy that the proposed remedy of the recitation of particular verses are mainly made by the anonymous editors.

In each of these cases, the anonymous editors express an interest in prescribing the use of scripture for protection and stipulating its proper use in incantatory contexts. To be sure, there are several medico-magical statements in the Babylonian Talmud attributed to earlier sages that prescribe the recitation of scripture, and in other cases the anonymous editors recommend invocations other than scripture.Footnote 104 But the preponderance of apotropaic uses of scripture is in later anonymous editorial comments. The chronological stratification suggests some change or at least intensification of rabbinic attitudes and ideas about incantations and the place of scripture within them.Footnote 105 Indeed, rather than take for granted that scripture would always be welcome in apotropaic contexts, the greater prevalence of apotropaic uses of scripture in the later layers of the Talmud contrasts markedly with the near absence of any practical recipes with scripture in some early medieval handbooks like Sefer HaRazim and Ḥarba DeMoshe.Footnote 106

This chronological shift in rabbinic notions of incantations allows us to recognize historical developments even within the rabbinic corpus. Further work must similarly move not only beyond older monolithic characterizations of a single “rabbinic” attitude towards “magic,” but also beyond flattened portraits that collapse all of rabbinic literature into a messy stew of diverse and conflicting opinions.Footnote 107 While acknowledging rabbinic diversity, methods like literary source criticism of the Babylonian Talmud enable scholars to identify developments and changing trends among the rabbis that map onto place and time.Footnote 108

Conclusion

The story of Pelimo exemplifies the benefits of historicist approaches that situate rabbinic stories in their ambient social and cultural milieu. The story is predicated on a rich set of common notions about the behavior of demons and the use of particular apotropaic formulae current in late antique Babylonia. It not only assumes knowledge of this context, but directly participates and intervenes in it by contrasting two different apotropaic formulae, using Satan as a mouthpiece in favor of one over the other. The rabbinic story thereby creatively articulates the rationale behind why at least some Jews, as evidenced especially by the incantation bowls, and by several other late Babylonian rabbinic passages, preferred to invoke scripture. Rather than stand apart from prevalent assumptions and practices among Babylonian Jews, the rabbis rejected some common practices while endorsing and amplifying others. In turn, the story of Pelimo is part of a nexus of late Babylonian rabbinic texts that express a growing interest in the use of scripture in apotropaic contexts. This development among the Talmud’s editors calls on scholars to not only problematize older binaries between rabbis and non-rabbis, elite and non-elite, but to attend to developments within rabbinic literature and among the rabbis in their conceptions of magic and medicine.

Footnotes

*

I am grateful to the journal editors, the reviewers, Rivka Elitzur-Leiman, and Yakir Paz for their insightful comments and constructive feedback on earlier versions of the article.

References

1 Pelimo appears several times in rabbinic literature, mainly in the Babylonian Talmud (e.g., b. MenaḤ. 37a), but his identity is unclear; one suggestion is that it is a shortened form of “so-and-so”; see Reuven Margoliot, LeḤeqer Shemot VeKinuyim beTalmud (Jerusalem: Mosad HaRav Kook, 1959) 28 (Hebrew). It more likely derives from a Greek name like Φιλήμων.

2 For the former, see Shmuel Yosef Halevi Czaczkes (later Shai Agnon), “The Little Hero,” HaMiṣpeh 1.4 (May 6, 1904) 6. For the latter, see A Serious Man, directed by Joel Coen and Ethan Coen (Focus Features and Working Title Films, 2009).

3 Jeffrey L. Rubenstein, The Land of Truth: Talmud Tales, Timeless Teachings (Philadelphia: JPS, 2018) 116–40; Ido Hevroni, “ ‘An Arrow in Satan’s Eye’: Contexts and Meaning in a Talmudic Polemic Story,” Jerusalem Studies in Hebrew Literature 23 (2009) 15–51 (Hebrew), esp. 18. See also Admiel Kosman, “Pelimo and Satan: A Divine Lesson in the Public Latrine,” CCAR Journal 57 (2010) 3–13; idem, Men’s World: Reading Masculinity in Jewish Stories in a Spiritual Context (Würzburg: Ergon Verlag, 2009) 113–21.

4 Hevroni, “ ‘An Arrow in Satan’s Eye,’ ” 2.

5 Shulamit Valler, Nashim beḤevra HaYehudit beTequfat HaMishnah veHaTalmud (Tel Aviv: Ḳibutz ha-meʾuḤad, 2000) 46–51; Ishay Rosen-Zvi, “The Evil Inclination, Sexuality, and the YiḤud Prohibitions: A Chapter in Talmudic Anthropology,” Te’oriah u-viqoret 14 (1999) 73–74 (Hebrew).

6 Rubenstein, The Land of Truth, 119.

7 Ibid., 130; Hevroni, “ ‘An Arrow in Satan’s Eye,’ ” 28.

8 Hevroni, “ ‘An Arrow in Satan’s Eye,’ ” 5.

9 For an overview and critical assessment, see Jeffrey Rubenstein, Talmudic Stories: Narrative Art, Composition, and Culture (Baltimore, MD: Johns Hopkins University Press, 1999) 8–10, and Hillel Newman, “Closing the Circle: Yonah Fraenkel, the Talmudic Story, and Rabbinic History,” in How Should Rabbinic Literature be Read in the Modern World (ed. Matthew Kraus; Piscataway, NJ: Gorgias Press, 2006) 105–35.

10 See Geoffrey Herman and Jeffrey Rubenstein, “Introduction,” in Aggadah of the Bavli and its Cultural World (ed. Geoffrey Herman and Jeffrey Rubenstein; Providence, RI: Brown Judaic Studies, 2018) xi–xxxv, and Matthew Goldstone, “The Babylonian Talmud in its Cultural Context,” Religion Compass (May, 2019) 1–11.

11 Gideon Bohak, Ancient Jewish Magic: A History (Cambridge: Cambridge University Press, 2008) 351–425; idem, “I Beg You: Appeals to Angels in the Talmudic and Gaonic Periods,” Tarbiz 89 (2022) 39–72 (Hebrew); Yuval Harari, “The Sages and the Occult,” in The Literature of the Sages, vol. 3b, Midrash and Targum (ed. Shmuel Safrai et al.; Philadelphia: Fortress Press, 2006) 521–64; Avigail Manekin-Bamberger, “A Jewish Magical Handbook in the Babylonian Talmud,” Jewish Studies Quarterly 29 (2022) 242–60; eadem, “Healing and Protection in Rabbinic Liturgy and in the Jewish Aramaic Incantation Bowls,” Jerusalem Studies in Jewish Thought 27 (2022) 95–104. Contrast with earlier approaches described in Natalie Polzer, “The Bible in the Aramaic Magic Bowls” (M.A. Thesis, McGill University, 1986) 59.

12 Yuval Harari, Jewish Magic before the Rise of Kabbalah (Raphael Patai Series in Jewish Folklore and Anthropology; Detroit: Wayne State University Press, 2017) 15–67; David Frankfurter, “Ancient Magic in a New Key: Refining an Exotic Discipline in the History of Religions,” in Guide to the Study of Ancient Magic (ed. David Frankfurter; Leiden: Brill, 2019) 3–20; Bernd-Christian Otto, Magie: Rezeptions- und diskursgeschichtliche Analysen von der Antike bis zur Neuzeit (Religionsgeschichtliche Versuche und Vorarbeiten; Berlin: DeGruyter: 2011).

13 For a review and critique of this binary, see Simcha Gross and Avigail Manekin-Bamberger, “Babylonian Jewish Society: The Evidence of the Incantation Bowls,” Jewish Quarterly Review 112 (2022) 1–30.

14 On rabbinization, see Raʿanan Boustan, “Rabbinization and the Persistence of Diversity in Jewish Culture in Late Antiquity,” in Diversity and Rabbinization: Jewish Texts and Societies between 400 and 1000 C.E. (ed. Gavin McDowell, Ron Naiweld, and Daniel Stökl Ben Ezra; Cambridge: OpenEdition Books, 2021) 427–49.

15 For helpful overviews, see Dan Levene, Curse or Blessing, What’s in the Magical Bowl? (Parkes Institute: University of Southampton, 2002); Gideon Bohak, “Babylonian Incantation Bowls: Past, Present, and Future,” Pe‘amim 105–6 (2005–2006) 253–65 (Hebrew).

16 Gross and Manekin-Bamberger, “Babylonian Jewish Society.”

17 For a study of the cultural poetics of the collection, including the surrounding legal materials, see Gail Labovitz, “Of Proper and Unrestrained Men: Reading Law, Narrative, and Desire in the Babylonian Talmud,” Hebrew Union College Annual 79 (2008) 43–68. For Babylonian rabbinic story collections more generally, see Eli Yassif, “Story-Cycles in Rabbinic Aggadah,” Jerusalem Studies in Hebrew Literature 12 (1990) 103–45 (Hebrew); Shamma Friedman, “Regarding Historical Aggadah in the Babylonian Talmud,” in Saul Lieberman Memorial Volume (ed. Shamma Friedman; Jerusalem and New York: Jewish Theological Seminary of America, 1993) 119–64 (Hebrew); Eliezer Diamond, “Wrestling the Angel of Death: Form and Meaning in Rabbinic Tales of Death and Dying,” Journal for the Study of Judaism 26 (1995) 76–92.

18 On the latter story, see Shlomo Naeh, “Freedom and Celibacy: A Talmudic Variation on Tales of Temptations and Fall in Genesis and its Syrian Background,” in The Book of Genesis in Jewish and Oriental Christian Interpretation (ed. J. Frishman and L. Van Rompay; Leuven: Peeters, 1997) 73–89, and Michal Bar-Asher Siegal, “Syriac Monastic Motifs in the Babylonian Talmud: The Heruta Story Reconsidered (b. Qiddushin 81b),” in Jews and Syriac Christians: Intersections Across the First Millennium (ed. Aaron Michael Butts and Simcha Gross; Tübingen: Mohr Siebeck, 2020) 27–46.

19 b. Qidd. 81a–b according to MS Vatican 111. The manuscript variants are minor and will be referred to when relevant. All translations of rabbinic literature are my own.

20 The pronoun is explicit in the other witnesses.

21 According to the other manuscript witnesses.

22 b. Qidd. 29b–30a according to MS Vatican 111.

23 b. Sukkah 37b–38a = b. MenaḤ. 62a.

24 On the prophylactic power of the lulav in ancient Jewish visual culture, see Naama Vilozny, Lilith’s Hair and Ashmedai’s Horns: Figure and Image in Magic and Popular Art; Between Babylonia and Palestine in Late Antiquity (Jerusalem: Yad Ben Zvi, 2017) 255–56 (Hebrew). Other ritual items, like the shofar, are similarly said to serve as a vehicle by which “to confuse Satan.” See Pseudo-Jonathan to Num 10:10, b. Sukkah 54a, 55a; b. Roš Haš. 16a–b, 29a, 32a; b. ʿArak. 3b.

25 According to MS British Library 400.

26 The manuscript witnesses are divided as to whether Rav AḤa explicitly called the lulav an arrow or simply declared “this in Satan’s eye,” though the latter is still undoubtedly an allusion to the “arrow in Satan’s eye” formula.

27 The Babylonian Talmud includes other remedies against the evil eye, e.g., in b. Ber. 55b, which also includes both a physical action and utterance. Incidentally, in another story (b. B. Bat. 16a), Rav AḤa b. Yaaqov interprets Satan’s actions in the book of Job favorably and in response “Satan came and kissed his feet.” In another story (b. Qidd. 29b), Rav AḤa conquers a demon who resides in the study house, on which see Sara Ronis, “A Seven-Headed Demon in the House of Study: Understanding a Rabbinic Demon in Light of Zoroastrian, Christian, and Babylonian Textual Traditions,” AJS Review 43 (2019) 125–42. For an attempt to connect these various stories into a biographical portrait of Rav AḤa, see Hevroni, “An Arrow in Satan’s Eye,” 31–33, which confuses demonology with moral theology and metaphor.

28 The fear of inciting Satan by challenging him is distinct from the idea of “opening one’s mouth to Satan” (e.g., b. Ber. 19a; b. Ber. 60a; b. Ketub. 8a) by claiming to be just and free of sin. For danger associated with following improper procedure related to amulets, see b. Šabb. 66b.

29 Sophie Lunn-Rockliffe, “Demons between the Desert Fathers and the Rabbis,” Jewish Studies Quarterly 25 (2018) 285, and Adam Silverstein, “On the Original Meaning of the Qurʾanic Term al-shayṭān al-rajīm,” Journal of the American Oriental Society 133 (2013) 21–33.

30 See esp. Christopher Faraone, The Transformation of Greek Amulets in Roman Imperial Times (Philadelphia: University of Pennsylvania Press, 2018) 105–8; Campbell Bonner, Studies in Magical Amulets, Chiefly Graeco-Egyptian (Ann Arbor: University of Michigan Press, 1950) 97–100 (and the additional amulets on the The Campbell Bonner Magical Gems Database [http://cbd.mfab.hu/]). See also J. Robert C. Cousland, “The Much Suffering Eye in Antioch’s House of the Evil Eye: Is it Mithraic?” Religious Studies and Theology 24 (2005) 61–74. On analogy in incantatory formulae, see Stanley J. Tambiah, “Form and Meaning of Magical Acts: A Point of View,” in Modes of Thought (ed. Robin Horton and Ruth Finnegan; London: Faber and Faber, 1973) 199–229.

31 Bohak, Ancient Jewish Magic, 322; Carl H. Kraeling, The Excavations at Dura Europos Conducted by Yale University and the French Academy of Inscriptions and Letters: Final Report VIII, Part I: The Synagogue (New Haven, Yale University Press 1956) 48–49; Karen Stern, “Mapping Devotion in Roman Dura Europos: A Reconsideration of the Synagogue Ceiling,” American Journal of Archaeology 114 (2010) 473–504. For the apotropaic eye elsewhere in Dura, see Lucinda Dirven, “Religious Competition and the Decoration of Sanctuaries: The Case of Dura-Europos,” Eastern Christian Art 1 (2004) 1–19, esp. 8.

32 On the evil eye in Rabbinic literature, see Richard Kalmin, “The Evil Eye in Rabbinic Literature of Late Antiquity,” in Judaea-Palaestina, Babylon and Rome: Jews in Antiquity (ed. Benjamin Isaac and Yuval Shahar; Tübingen: Mohr Siebeck, 2012) 111–38; Rivka Ulmer, The Evil Eye in the Bible and in Rabbinic Literature (Hoboken, NJ: Ktav, 1994); in the ancient Near East, see James Nathan Ford, “ ‘Ninety-Nine by the Evil Eye and One from Natural Causes’: KTU2 1.96 in its Near Eastern Context,” Ugarit-Forschungen 30 (1998) 201–78; in Aramaic incantations, see Dan Levene, “Jewish Liturgy and Magic Bowls,” in Studies in Jewish Prayer (ed. Robert Hayward and Brad Embry; Oxford: Oxford University Press, 2005) 175–80.

33 Greek text in C. C. McCown, The Testament of Solomon (Leipzig: J.C. Hinrichs, 1922) 58*, translation in D. Duling, “The Testament of Solomon,” in The Old Testament Pseudepigrapha (ed. James H. Charlesworth; vol. 1; Garden City, NY: Doubleday, 1983) 981.

34 According to b. ʿAbod. Zar. 20b and Mesekhet Kallah 3, the Angel of Death is “filled with eyes.” The demon קטב מרירי is similarly said, in rabbinic literature, to be covered with eyes. See Lam. Rab. 1 (Midrasch Echa Rabbati [ed. Salomon Buber; Vilna: Romm, 1899] 63).

35 Ford, “ ‘Ninety-Nine by the Evil Eye,’ ” 211–15.

36 Rivka Elitzur-Leiman, “Jewish Metal Amulets from Late Antiquity” (Ph.D. diss., Tel Aviv University, 2021) 111–14, suggests that a Jewish Palestinian Aramaic amulet incantation may also refer to the much suffering eye.

37 Faraone, Transformation of Greek Amulets, 210–11.

38 The rabbinic formula and the image resonate in a more general sense with Aramaic incantations that threaten demons with physical harm, for instance, one which threatens that malevolent forces will be shot with an arrow (גירא) in the heart (MS 1928/21:9–10; I thank James Nathan Ford for sharing this text with me). For Mandaic parallels, see James Nathan Ford, “Notes on the Mandaic Incantation Bowls in the British Museum,” JSAI 26 (2002) 258–59. A possibly similar formula may be recommended by an anonymous gloss in b. PesaḤ. 111a, but the meaning is uncertain. In b. Sanh. 95a and 107a, Satan turns into a deer and bird respectively, and David tries shooting him with arrows. On these texts, see Rebecca Lesses, “Image and Word: Performative Ritual and Material Culture in the Aramaic Incantation Bowls,” in Practicing Gnosis: Ritual, Magic, Theurgy, and Liturgy in Nag Hammadi, Manichaean and Other Ancient Literature (ed. April D. DeConick, Gregory Shaw, and John D. Turner; Leiden: Brill, 2013) 394. On demonic shapeshifting in the incantation bowls, see Marco Moriggi, “Devilish Apparitions in Mesopotamian Incantation Bowls: Preliminary Remarks about Demons in the Guise of Animals,” in Animals, Gods and Men from East to West: Papers on Archaeology and History in Honour of Roberta Venco Ricciardi (ed. Alessandra Peruzzetto, Francesca Dorna Metzger, and Lucinda Dirven; Oxford: British Archaeological Reports, 2013) 119–22. Demonic manifestations are counteracted by damaging their eyes in b. Šabb. 156b and b. Yoma 69b.

39 Published by Ethel Drower, “Shafta d-Pishra d-Ainia: Exorcism of the Evil and Diseased Eyes,” Journal of the Royal Asiatic Society 4 (1937) 591–611; eadem, “Shafta d-Pishra d-Ainia: Exorcism of the Evil and Diseased Eye (Conclusion),” Journal of the Royal Asiatic Society 5 (1938) 1–20.

40 See e.g., Christa Müller-Kessler and Theodore Kwasman, “A Unique Talmudic Aramaic Incantation Bowl,” Journal of American Oriental Society 120 (2000) 159–65, and Matthew Morgenstern, “A Mandaic Lamella for the Protection of a Pregnant Woman: MS 2087/9,” Aula Orientalis 33 (2015) 217–86.

41 Drower, “Shafta d-Pishra d-Ainia” (1938), 2 (Mandaic) and 9 (English).

42 Ibid., 3, 11, revised translation in Matthew Morgenstern, “A New Mandaic Dictionary: Challenges, Accomplishments, and Prospects,” in From Ancient Manuscripts to Modern Dictionaries: Select Studies in Aramaic, Hebrew, and Greek (ed. Tarsee Li and Keith Dyer; Piscataway, NJ: Gorgias Press, 2017) 160. The word for crane can also mean goose (כורכיא), another bird present on the mosaic in the House of the Evil Eye. Likewise, a few lines later the eye is again threatened with crabs and centipedes.

43 The much suffering eye may explain the appearance on two incantation bowls of Prov 30:17, published by Shaul Shaked, “Dramatis Personae in the Jewish Magic Texts: Some Differences between the Incantation Bowls and Geniza Magic,” Jewish Studies Quarterly 13 (2006) 373–76 (M4), and Ali Faraj, “An Incantation Bowl of Biblical Verses and a Syriac Incantation Bowl for the Protection of a House,” in Proceedings of the 13th Italian Meeting of Afro-Asiatic Linguistics, Held in Udine, May 21st24th, 2007 (ed. Frederick Mario Fales and Giulia Francesca Grassi; Padua: Sargon Editrice e Libreria, 2010) 205–12, where it appears alongside Zech 3:2 and Exod 15:14–17.

44 For the former, see Faraone, Transformation of Greek Amulets, 375 n. 76, and for the latter, see the image of Walters Art Museum 54.2653 above.

45 Robert du Mesnil du Boisson, Les peintures de la synagogue de Doura-Europos, 245256 après J.-C. (Roma: Pontificio istituto biblico, 1939), fig. 96.1.

46 Not coincidentally, the reverse of the many pendant amulets with the motif of the much suffering eye is the motif of “the holy rider,” a figure on horseback stabbing a demon with a spear, typically accompanied with an inscription identifying the rider’s identity either as “One God who conquers evil,” or particular figures like Solomon, St. Sissinios, and rarely, St. George. On the holy rider, see Faraone, Transformation of Greek Amulets, 113–14.

47 In b. Taʿan. 25a, according to some manuscripts, God says to a rabbi “I cast my arrows at you.”

48 For the possibility that later editors misunderstood earlier medical recipes, see Markham Geller, “An Akkadian Vademecum in the Babylonian Talmud,” in From Athens to Jerusalem: Medicine in Hellenized Jewish Lore and in Early Christian Literature (ed. Samuel Kottek and Manfred Horstmanshoff; Rotterdam: Erasmus Publishing, 2000) 18 n. 26 and passim.

49 On demons demanding entry, with negative consequences, see e.g., Joseph Naveh and Shaul Shaked, Amulets and Magic Bowls: Aramaic Incantations of Late Antiquity (Jerusalem: Magnes Press, 1985) 188–93.

50 James Montgomery, Aramaic Incantation Texts from Nippur (Philadelphia: University of Pennsylvania Museum, 1913) 40–45.

51 See generally Michael Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmudic and Geonic Periods (2nd ed.; Ramat Gan, Israel: 2020) 765 (s.v. סטנא).

52 See Simcha Gross, “ ‘Whoever is Hungry, Come and Eat’: On the Origins and Later Development of a Puzzling Passover Passage,” Aramaic Studies 18 (2020) 171–97.

53 b. Ketub. 67b for the Day of Atonement. In general, see Gross, “ ‘Whoever is Hungry, Come and Eat.’ ” Other Babylonian rabbinic passages include b. Šabb. 151b concerning the imperative to give bread to those who come to one’s door, and b. Šabb. 156b concerning a poor man who requests food on the day of a wedding feast. See also b. Moʿed Qaṭ. 28a, discussed below.

54 E.g., b. PesaḤ. 68b and b. Ber. 8a–b. A story set on the eve of Yom Kippur, where the townsfolk display poor hospitality, is found in b. Ḥullin 110a–b, discussed by Aaron Amit, “The ‘Halakhic Kernel’ as a Criterion for Dating Babylonian Aggadah: Bavli Ḥullin 110a–b and Parallel,” AJS Review 36 (2012) 187–205.

55 b. Yoma 20a.

56 b. Ḥag. 16a, and discussion in Gideon Bohak, “Conceptualizing Demons in Late Antique Judaism,” in Demons and Illness from Antiquity to the Early-Modern Period (ed. Siam Bhayro and Catherine Rider; Leiden: Brill, 2017) 111–33.

57 See Gross, “ ‘Whoever is Hungry, Come and Eat.’ ”

58 Interestingly, the verb used to describe Satan “turning” and spitting appears in various incantation bowls to mean to “turn back” the malevolent forces. See Sokoloff, A Dictionary of Jewish Babylonian Aramaic, 543.

59 b. Qidd. 39b–40a, with Bohak, Ancient Jewish Magic, 398.

60 Ilaria Briata, “Demons and Scatology: Cursed Toilets and Haunted Baths in Late Antique Judaism,” Demons in Early Judaism and Christianity: Characters and Characteristics (ed. Hector Patmore and Josef Lössl; Leiden: Brill, 2022) 256–72. In general, see the early study of Campbell Bonner, “Demons of the Bath,” in Studies Presented to F. Ll. Griffith (ed. Stephen R. K. Glanville; London: Oxford University Press, 1932) 203–8, and more recently Gemma Jansen, “Cultural Attitudes: Interpreting Images and Epigraphic Testimony,” in Roman Toilets: Their Archaeology and Cultural History (ed. Gemma Jansen, Ann Olga Koloski-Ostrow, and Eric Moormann; Leuven: Peeters, 2011) 165–76.

61 Rafe Neis, “ ‘Their Backs toward the Temple, and Their Faces toward the East’: The Temple and Toilet Practices in Rabbinic Palestine and Babylonia,” Journal for the Study of Judaism 43 (2012) 328–68, and Avigail Manekin-Bamberger, “An Akkadian Demon in the Talmud: Between Šulak and Bar-Širiqa,” Journal for the Study of Judaism 44 (2013) 282–87.

62 E.g., the parallel formulae in y. Ber. 9:4; 14b and b. Ber. 60b, with the alternative version offered by the Babylonian sage Abaye in the latter. See discussions in Avigail Manekin-Bamberger, “Healing and Protection in Rabbinic Liturgy,” 95–104; Briata, “Demons and Scatology.”

63 b. Moʿed Qaṭ. 28a. On Babylonian rabbinic stories concerning the Angel of Death, see Diamond, “Wrestling the Angel of Death.” Intriguing parallels exist as well between this rabbinic motif and its appearance in the Testament of Job, where Satan disguises himself as a pauper and approaches Job’s home, on which see Hevroni, “An Arrow in Satan’s Eye,” 23–25.

64 The former is in MS Munich 95 and the printed editions; the latter in MS Vatican 110–11.

65 On the identity of “Satan” in this passage, see Ryan E. Stokes, “Satan, YHWH’s Executioner,” Journal of Biblical Literature 133 (2014) 251–70. This scene receives a conventional midrashic interpretation in b. Sanh. 93a.

66 Interestingly, in the Peshitta, it is “the angel of the Lord” (ܡܠܐܟܗ ܕܡܪܝܐ), mentioned in the previous verse and throughout the chapter, who rebukes Satan in Zech 3:2. See The Syriac Peshiṭta Bible with English Translation: The Twelve Prophets (ed. Donald Walter et al.; Piscataway, NJ: Gorgias Press, 2012) 226–27.

67 Theodore J. Lewis, “ʿAthtartu’s Incantations and the Use of Divine Names as Weapons,” Journal of Near Eastern Studies 71 (2011) 207–27, esp. 212; Andrew A. MacIntosh, “A Consideration of Hebrew gʿr,” Vetus Testamentum 19 (1969) 471–79.

68 Lewis, “ ‘Athtartu’s Incantations.”

69 Gabriel Barkay, Andrew G. Vaughn, Marilyn J. Lundberg, and Bruce Zuckerman, “The Amulets from Ketef Hinnom: A New Edition and Evaluation,” Bulletin of the American Schools of Oriental Research 334 (2004) 41–71.

70 Barkay et. al., “The Amulets from Ketef Hinnom,” 68, with 65–66. See also Theodore J. Lewis, “Job 19 in the Light of the Ketef Hinnom Inscriptions and Amulets,” in Puzzling Out the Past: Studies in the Northwest Semitic Languages and Literatures in Honor of Bruce Zuckerman (ed. Steven Fine et al.; Leiden: Brill, 2012) 99–113, 319–20.

71 E.g., 1Q20:28–29. For text, see Daniel Machiela, The Dead Sea Genesis Apocryphon: A New Text and Translation with Introduction and Special Treatment of Columns 13–17 (Leiden: Brill, 2009) 76, and 1QHa Frg. 4, discussed in Esther Eshel, “Apotropaic Prayers in the Second Temple,” in Liturgical Perspectives: Prayer and Poetry in Light of the Dead Sea Scrolls (ed. Esther Chazon et al.; Leiden: Brill, 2003) 82–83.

72 Jonas Greenfield, “The Genesis Apocryphon: Observations on Some Words and Phrases,” in Studies in Hebrew and Semitic Languages Dedicated to the Memory of Prof. Eduard Yechezkel Kutscher (ed. Gad B. Sarfatti et al.; Ramat-Gan: Bar-Ilan University Press, 1980) xxxii–xxxix; Naveh and Shaked, Amulets and Magic Bowls, 268 (s.v. gʿr); Joseph Naveh and Shaul Shaked, Magic Spells and Formulae: Aramaic Incantations of Late Antiquity (Jerusalem: Magnes Press, 1993) 266 (s.v. gʿr); Gordon Hamilton, “A New Hebrew-Aramaic Incantation Text from Galilee: ‘Rebuking the Sea,’ ” Journal of Semitic Studies 41 (1996) 215–49, citing Nahum 1:4. The verb is used in the passive, e.g., James A. Montgomery, “Some Early Amulets from Palestine,” JAOS 31 (1911) 272–81, and compare with Naveh and Shaked, Amulets and Magic Bowls, 82–84 (Amulet 9), and in the active and infinitive, e.g., Naveh and Shaked, Amulets and Magic Bowls, 44–49, updated in Joseph Naveh, “A Recently Discovered Palestinian Jewish Aramaic Amulet,” in Studies in West-Semitic Epigraphy: Selected Papers (Jerusalem: Magnes, 2009) 159–66. For another active use, see Ohad Abudraham, “ ‘Joshua Son of Nun and the Seven Angels’: a Hebrew Lamella from the Wolfe Collection,” Journal of Jewish Studies 122 (2021) 49 (lines 3–5) and 53. For incantation bowls, see e.g., M117 in Dan Levene, A Corpus of Magic Bowls: Incantation Texts in Jewish Aramaic from Late Antiquity (London: Kegan Paul, 2003) 77–80, and for a Babylonian magical handbook, see Gershom Scholem, “Havdala De-Rabbi ʿAqiva – A Source for the Tradition of Jewish Magic During the Geonic Period,” Tarbiz 50 (1980) 253–68.

73 See e.g., Matt 17:18 (compare with Mark 9:25; Luke 9:42); Mark 1:25–26 (Luke 4:35), with Howard Clark Kee, “The Terminology of Mark’s Exorcism Stories,” New Testament Studies 14 (1968) 232–46.

74 Polzer, “The Bible in the Aramaic Magic Bowls,” 158–59; Daniel Waller, The Bible in the Bowls: A Catalogue of Biblical Quotations in Published Jewish Babylonian Aramaic Magic Bowls (Cambridge: Open Book Publishers, 2022) 22 n. 29. In general, passages about God’s power and activities were regularly applied to his control of the demonic and unseen, e.g., Num 6:4 in Targum Pseudo-Jonathan (Clarke, Targum Pseudo-Jonathan: Numbers, 205), and Sifre Num. 40 (Menahem Kahana, Sifre Numbers: An Annotated Edition [Jerusalem: Magness Press, 2011] 1.1, 108, and 2.2, 316–17), where “to preserve you” includes from demons.

75 Naveh and Shaked, Amulets and Magic Bowls, 40–44 (Amulet 1); Roy Kotansky, Joseph Naveh, and Shaul Shaked, “A Greek-Aramaic Silver Amulet from Egypt in the Ashmolean Museum,” Le Muséon 105 (1992) 5–25.

76 On this and other likely Second Temple Jewish traditions in Jude, see Ryan E. Stokes, “Not over Moses’ Dead Body: Jude 9, 22–24 and the Assumption of Moses in their Early Jewish Context,” Journal for the Study of the New Testament 40 (2017) 192–213.

77 “Typologically,” so Polzer, “The Bible in the Aramaic Incantation Bowls,” 104–5.

78 On the verse’s prominence, see Naveh and Shaked, Magic Spells and Formulae, 25; Charles Isbell, Corpus of Aramaic Incantation Bowls (Missoula, MO: Scholars Press, 1975) 195; Joseph Angel, “The Use of the Hebrew Bible in Early Jewish Magic,” Religion Compass 3 (2009) 789–90. For the list of bowls, see Waller, The Bible in the Bowls, 153–62, and Christa Müller-Kessler, “The Use of Biblical Quotations in Jewish Aramaic Incantation Bowls,” in Studies in Magic and Divination in the Biblical World (ed. Helen R. Jacobus et al.; Piscataway, NJ: Gorgias Press, 2013) 239–41.

79 Judah B. Segal, Catalogue of the Aramaic and Mandaic Incantation Bowls in the British Museum (London: British Museum Press, 2000) 63–65 (023A); Montgomery, Aramaic Incantation Texts,127–32 (#3), 138–40 (#5); 188–89 (#16; compare with Montgomery #12 for the same client, which ends instead with Ps 121:7, another popular verse); Levene, A Corpus of Magic Bowls, 12, 34–37 (M59); Shaul Shaked, James N. Ford, and Siam Bhayro, Aramaic Bowl Spells: Jewish Babylonian Aramaic Bowls (2 vols.; Leiden: Brill, 2022) 2:169–76 (JBA 104; MS 2053/238); Haggai Misgav, “Jewish-Aramaic Incantation Bowls,” in Finds Gone Astray: ADCA Confiscated Items (ed. Dalit Regev and Hananya Hizmi; Jerusalem: The Antiquities Department of the Civil Administration, 2018), #1, #2, #6; Siam Bhayro, “An Aramaic Magic Bowl for Fertility and Success in Childbirth, Museu da Farmácia (Lisbon, Pharmacy Museum), Inv. No. 10895,” Aramaic Studies 15 (2017) 106–11; Cyrus Gordon, “Two Aramaic Incantations,” in Biblical and Near Eastern Studies: Essays in Honor of William Sanford LaSor (ed. Gary A. Tuttle; Grand Rapids: Eerdmans, 1978) 233–36 (ZRL 48); Gideon Bohak, “From Qumran to Cairo: The Lives and Times of a Jewish Exorcistic Formula (with an Appendix by Shaul Shaked),” in Ritual Healing: Magic, Ritual and Medical Therapy from Antiquity until the Early Modern Period (ed. Ildikó Csepregi and Charles Burnett; Florence: SISMEL, 2012) 47–48 (MS 2053/7); Tatyana Fain, James Nathan Ford, and Alexey Lyavdansky, “Aramaic Incantation Bowls at the State Hermitage Museum, St. Petersburg,” Babel und Bibel 9 (2016) 300–302 (S-448); Gaby Abousamra, “Semomit in a New Incantation Bowl,” in Between the Worlds: Magic, Miracles, and Mysticism (ed. M. Maeva et al.; 2 vols.; Sofia: EFSEM, 2020) 2:455–464; Isbell, Corpus of Aramaic Incantation Bowls, 100–101 (#42).

80 So Polzer, “The Bible in the Aramaic Magic Bowls,” 107; Siam Bhayro, “The Use of Quotations from the Psalms in the Aramaic Magic Bowls,” in You Who Live in the Shelter of the Most High (Ps. 91:1): The Use of Psalms in Jewish and Christian Traditions (ed. Ida Fröhlich et al.; Göttingen: Vandenhoeck & Ruprecht, 2021) 76; Waller, The Bible in the Bowls, 16. Misgav, “Jewish-Aramaic Incantation Bowls,” 55. C.f. the two attested appearances of this verse in Jewish lamellae (fn. 75 above), where it appears in the midst of the incantations.

81 M59, with Levene, A Corpus of Magic Bowls, 12, 34–37, where it is absent from the parallel incantations listed by Levene, including B16019 (Montgomery, Aramaic Incantation Texts #33; Marco Moriggi, A Corpus of Syriac Incantation Bowls: Syriac Magical Texts from Late-Antique Mesopotamia [Leiden: Brill, 2014] 36–42 [#4]).

82 On the distribution of verses in bowl incantation, see Polzer, “The Bible in the Aramaic Incantation Bowls,” 87–88; Müller-Kessler, “The Use of Biblical Quotations,” 227–28.

83 Naveh and Shaked, Amulets and Magic Bowls, 184–87 (Bowl 11, with Deut 6:4; Ps 91:1); Markham Geller, “Eight Incantation Bowls,” Orientalia Lovaniensia Periodica 17 (1986) 108–9 (Aaron Bowl B, with Deut 6:4 and Ps 91:1); Shaked et al., Aramaic Bowl Spells, 2, 25–27 (JBA 67; MS 2053/56, with Ps 121:7–8), 163–67 (JBA 103; MS 2053/196, with Num 9:23); 190–92 (JBA 108; MS 2053/218, with Num 12:13), 194–97 (JBA 109; MS 2053/230, where Zech 3:2 is followed by “Amen Amen Selah,” then a citation from Ps 55:9, also again by “Amen Amen Selah”); Gaby Abousamra, “An Incantation Bowl from the Matenadaran,” Semitica 61 (2019) 139–46 (Matenadaran MS 132, appears with Is 51:14, followed by a coda recapitulating the name of the clients and purpose of the bowl). In Montgomery, Aramaic Incantation Texts, 209–11 (#26), three verses [Deut 6:4, Num 9:23; Zech 3:2] all precede the incantation.

84 Zech 3:2 appears in Levene, A Corpus of Magic Bowls, 71–74 (M108), following the client’s name and a brief description of the malevolent forces around him. The bowl otherwise largely consists of verses that correspond with the evening prayer as laid out in Amram Gaon’s liturgical handbook (Daniel Goldschmidt, Seder Rav ‘Amram Ga’on, Edited on the Basis of Manuscripts and Printed Editions, with Supplements, Variant Readings, and an Introduction [Jerusalem: Mosad HaRav Kook, 2004] 52–53). The verse also appears in a bowl comprised entirely of only two additional verses: Prov 30:17 and Exod 15:14–17; see Faraj, “An Incantation Bowl of Biblical Verses.” The bowl is divided into four segments, which Faraj arranges in an unusual way; following the natural flow of the verses, especially Exod 15:14–17, then Zech 3:2 again appears at the end of the incantation.

85 For wind as demons, see Sokoloff, A Dictionary of Jewish Babylonian Aramaic, 1040–41 (s.v. רוחא).

86 Admittedly there are few manuscripts attesting this passage, but one (MS Vatican 110–111) includes a citation of the Hebrew of Zech 3:2 introduced with Aramaic, suggesting that the use of Hebrew was a secondary scribal change.

87 E.g., Levene, A Corpus of Magic Bowls, 11, 79; Shaul Shaked, “Rabbis in Incantation Bowls,” in The Archeology and Material Culture of the Babylonian Talmud (ed. Markham J. Geller; Leiden: Brill, 2015) 109–10; Waller, The Bible in the Bowls, 143.

88 A marginal gloss in MS Florence 7 prescribes the recitation of Zech 3:2 not once but seven times (שבעה זמני).

89 Similarly, in b. PesaḤ. 110a, the “chief sorceress” provides Amemar with an incantation to use against sorceresses, on which see Simcha Fishbane, “ ‘Most Women Engage in Sorcery’: An Analysis of Sorceresses in the Babylonian Talmud,” Jewish History 7 (1993) 33–34, as does “Joseph the Demon.”

90 Zech 3:2 continued to serve as an effective apotropaic formula in early medieval Geonic materials. See Goldschmidt, Seder Rav ‘Amram Ga’on, 55, with Levene, A Corpus of Magic Bowls, 72–74, and Dan Levene, Dalia Marx, and Siam Bhayro, “ ‘Gabriel is on their Right:’ Angelic Protection in Jewish Magic and Babylonian Lore,” Studia Mesopotamica 1 (2014) 185–98.

91 For rabbinic texts, see Avigail Manekin-Bamberger, “Taming Nature and Gaining Authority—Rabbinic Decrees Reconsidered,” Aramaic Studies 21 (2023) 1–21. For ancient incantations, see Faraone, Transformation of Greek Amulets, 234–35 and passim.

92 For “piety,” see Bhayro, “The Use of Quotations from the Psalms,” 71–72.

93 On scholarly attempts to date the formation and redaction of the Talmud, see Simcha Gross, “Editorial Material in the Babylonian Talmud and its Sasanian Context,” Association of Jewish Studies Review 47 (2023) 51–76. For the date of the bowls, see Gross and Manekin-Bamberger, “Babylonian Jewish Society,” 5–6 n.18.

94 b. Šebu. 15b, with Gideon Bohak, Ancient Jewish Magic, 381.

95 A similar distinction is attributed to Rabbi Yudan in y. Šabb. 6:2; 8b = y. ʿErub. 10:12; 26c, though he may refer less to the use of any biblical citations in incantations than the recitation of a particular set of Psalms “for the afflicted” in the Temple. Other early rabbis excluded practices for healing from the proscribed category of “the ways of the Amorites,” on which see Bohak, Ancient Jewish Magic, 364–65.

96 m. Sanh. 10:1.

97 The same view about spitting while reciting the verse is attributed to Rav in y. Sanh. 10:1; 28b. Rabbi Joshua b. Levi there says that the same punishment applies to one who spits while reciting Lev 13:9. Noteworthily, Exod 15:26 does indeed appear on late antique Jewish amulets. See Bohak, Ancient Jewish Magic, 378–80.

98 b. Ber. 55b, acc. to MS Oxford 366. The manuscripts fluctuate quite extensively, a common feature of editorial additions to the Babylonian Talmud. For a discussion of the larger pericope, see Philip Alexander, “Bavli Berakhot 55a–57b: The Talmudic Dreambook in Context,” Journal of Jewish Studies 46 (1995) 230–48, and Haim Weiss, “All Dreams Follow the Mouth”: A Reading in the Talmudic Dreams Tractate (Or Yehudah: Devir, 2011) 82–89 (Hebrew). On ancient Jewish dream interpretation, see Alessia Bellusci, “Jewish Oneiric Divination: From Daniel’s Prayer to Genizah Seilat Halom,” in Unveiling the Hidden—Anticipating the Future: Divinatory Practices among Jews between Qumran and the Modern Period (ed. Josefina Rodríguez-Arribas and Dorian Gieseler Greenbaum; Leiden: Brill, 2021) 101–39.

99 On this dictum set within the larger context of ancient Jewish oneiric exegesis, see Maren Niehoff, “A Dream which is not Interpreted is like a Letter which is not Read,” Journal of Jewish Studies 43 (1992) 58–84.

100 On the alternation from a brief Hebrew dictum to a lengthy Aramaic statement as an indication of redaction, see e.g., Richard Kalmin, “The Formation and Character of the Babylonian Talmud,” in Cambridge History of Judaism IV: The Late-Roman Period (ed. Stephen T. Katz; Cambridge: Cambridge University Press, 2006) 840–41.

101 Some manuscripts read “and we decree,” on which see Manekin-Bamberger, “Healing and Protection in Rabbinic Liturgy,” 104–6.

102 The order of the citations is slightly different in MS Strasbourg 4028/50b.

103 One is very similar in form to b. Ber. 51a.

104 E.g., b. Ber. 5a where Rav Isaac says that the evening Shema keeps demons at bay, and b. Šabb. 67a, where a cure for fever is attributed to Rabbi YoḤanan, according to which one applies a knife to a thorn hedge on subsequent days and recites successive passages from the narrative of the burning bush (Exod 3). This unit is in Aramaic, and is slightly modified by the later fifth-century Babylonian rabbi Rav AḤa son of Rava, who similarly modifies a recipe by the mother of the fourth-century Babylonian rabbi Abaye a few lines earlier (b. Šabb. 66b), in structurally similar ways (in the former “O thorn bush, O thorn bush” [הסנה הסנה], in the latter “O river, O river” [נהרא נהרא]). This is the only recipe in this lengthy medical unit to prescribe the recitation of verses, on which see Daniel Picus, “Trimming the Text: Reading Ritual and Narrative Healing in the Babylonian Talmud,” Archiv für Religionsgeschichte 26 (2024) 315–38. For the ensuing spells in the unit, see Manekin-Bamberger, “A Jewish Magical Handbook.” b. Yoma 83b–84a includes an elliptical story that may portray a rabbi uttering a verse prophylactically. A rabbi in the “Talmudic dreambook” recommends that if one experienced a bad dream they should stand before a congregation mentioning several biblical precedents and then recite the priestly blessing (Num 6:24–26) along with the priests (b. Ber. 55b). For Palestinian rabbinic discussions of the apotropaic use of verses, see Bohak, Ancient Jewish Magic, 378–81.

105 The anonymous editors similarly qualify a series of views expressed by Reish Laqish pertaining to dangers of health, the demonic, and sorcery (b. PesaḤ. 111a), and the view of Rabbi Shimon bar YoḤai regarding dangers associated with disposed nail clippings (b. Nid. 17, with Yaakov Elman, “The World of the ‘Sabboraim’: Cultural Aspects of Post-Redactional Additions to the Bavli,” in Creation and Composition: The Contribution of the Bavli Redactors (Stammaim) to the Aggadah [ed. Jeffrey Rubenstein; Tübingen: Mohr-Siebeck, 2005] 412–15). In another discussion (b. Giṭ. 67b) concerning the Mishnah’s tangential mention of the illness Kordiakos, the anonymous editors explain that the Mishnah’s inclusion of the illness’ name rather than its causes provides the knowledge necessary to compose effective amulets against it. On this text, see Bohak, Ancient Jewish Magic, 374–75. For the anonymous editors on the relative sanctity of amulets, see b. Šabb. 61b–62a, and b. Šabb. 115b, where a citation of a Tannaitic tradition (t. Šabb. 13:4) about the writing of “blessings” is altered to read “blessings and incantations” (Saul Lieberman, Tosefta Kifshuta [12 vols.; New York: Jewish Publication Society, 2002] 3:205). On the editors’ creation of extended discussions concerning demons and remedies, see b. PesaḤ. 109b–112a, with Sara Ronis, Demons in the Details: Demonic Discourse and Rabbinic Culture in Late Antique Babylonia (Oakland, CA: University of California Press, 2022), and on astrology, see Jeffrey Rubenstein, “Talmudic Astrology: Bavli Shabbat 156a–b,” Hebrew Union College Annual 78 (2007) 109–48. Elman, “World of the ‘Sabboraim,’ ” mistakes statements about the character of demons by the later editors for a kind of rationalizing discourse that questions the very prevalence and power of demons altogether.

106 See Polzer, “The Bible in the Aramaic Incantation Bowls,” 190–204. The same is true for the handbook published since Polzer’s study: Gideon Bohak and Matthew Morgenstern, “A Babylonian Jewish Aramaic Magical Booklet from the Damascus Genizah,” Ginzei Qedem 10 (2014) 9–44. By contrast, Havdala de-Rabbi Akiva, whose formulae are closely related to the incantation bowls, regularly deploys biblical texts.

107 Bohak, Ancient Jewish Magic, 351–425.

108 On source criticism of the Babylonian Talmud in pericopes pertaining to “magic,” see Shamma Friedman, “Now You See it, Now You Don’t: Can Source-Criticism Perform Magic on Talmudic Passages about Sorcery?” in Rabbinic Traditions between Palestine and Babylonia (ed. Ronit Nikolsky and Tal Ilan; Leiden: Brill, 2014) 32–83.