A Structural-Literary Analysis of the Piska “And It Happened at Midnight”

Piska 7, “And it happened at midnight,” is composed of three units: the opening unit contains four proems (sections 1–4, ed. Mandelbaum, pp. 122–25); the middle unit contains five interpretive derashot related mostly to the first verse of the Torah reading, Exodus 12:29 (sections 5–9, pp. 125–30); and the concluding unit contains three narrative derashot on the plagues and finishes with a concluding derashah (sections 10–12, pp. 130–34). For each of these units I will point out features that deviate from the usual features found in the Pesikta, distinguishing characteristics that differ from those noted by Goldberg. After noting these “deviations,” I will try to explain how they are a result of the editing process typical of this piska.

The Opening Section of the Piska

The opening unit of the piska contains four proems. The second and fourth proems contain features that deviate from the usual structure of proems in classical midrashic literature, including the Pesikta itself. The form of the second proem seems to be very close to the one that precedes it, and it also conceptually completes it. Thus the first two proems seem to be one literary unit. The first is a short proem ascribed to “R. Tanḥum of Jaffa in the name of R. Nunya of Caesarea.” The darshan explains David's words in Psalms 73:16, “So I applied myself to understand this, but it seemed a hopeless task” (ואחשבה לדעת זאת עמל היא בעיני) as referring to the gap between the divine ability to “determine the exact time of midnight” and the human inability to do the same. The proem ends with the typical formula: “And since no being except God can determine when exactly midnight is, that is why it says, ‘And it happened at midnight’” (ולפי שאין בריה יכולה לעמוד על חצי הלילה אלא הוא, לפי' אמ' ויהי בחצי הלילה). The second proem is structured similarly:

As in the first proem, the second proem also uses a verse (this time from Isaiah) to highlight God's ability to make distinctions that human beings cannot. However, the divine ability on which the midrash focuses is different—whereas in the first proem it was the ability to tell precise time, in the second it is the ability to distinguish between a child who is a firstborn and one who is not. Both proems relate to the mention of God in the verse, “And it happened at midnight, and the Lord struck down every firstborn.” The first proem connects the phrase “and the Lord” to the beginning of the verse, as if it implies that only God can determine the exact moment of midnight; whereas the second proem connects “and the Lord” with the words that follow—only God can distinguish between a firstborn and one who is not a firstborn.Footnote ¹⁴ However, the second proem does not conclude in the manner one would expect. Instead, it repeats the conclusion of the first proem, and even the words of the verse cited in the first proem: “But as for me this is too hard for me. And no other being can determine the exact moment of midnight except for Him, therefore Scripture says, And it came to pass [precisely] at midnight.” At this point we were expecting the conclusion to read something like the following: “And since no other being in the world is able to distinguish between seed of a firstborn and seed which is not a firstborn except for the Holy One, blessed be He, that is why Scripture says, And God struck every firstborn” (ולפי שאין בריה יכולה להבחין בין טיפה של בכור לשאינה של בכור אלא הב"ה לפי' וה' הכה כל בכור).Footnote ¹⁵

In my opinion this unusual phenomenon, the lack of accordance between the conclusion of the proem and its content, is a result of the process in which the proem was edited. The editor lifted the conclusion from the previous proem, but stopped in the middle of his creative work before he had finished polishing up the transferred material to accord with its new setting.Footnote ¹⁶ The editor of the Pesikta fashioned this proem out of raw material that we can reconstruct, patterning the conclusion after the structure of the opening proem, but he left the concluding line as it was, without emending it to match its new setting.Footnote ¹⁷ It seems that the editor took the body of the proem from a similar proem found in piska 21, “Arise, give light” (קומי אורי) (ed. Mandelbaum, p. 319):

The beginning of this proem is almost identical to the proem above, except that there the end of the verse, “And My glory I do not give to others” is connected more tightly to the verse, “And the glory of the Lord will shine on you” (וכבוד ה' עלייך זרח). The editor of our piska used the derashah on the verse, and wove into it the theme of God's ability to discern between firstborns and non-firstborns,Footnote ¹⁸ as if God claims that “My glory” is found in the ability to make a type of distinction that humans cannot make. The editor concluded the proem with a structure he borrowed from the previous proem without adapting it to its new context. Taken together, the first two proems are one literary unit that emphasizes the supernatural nature of the plague of the firstborn, both in terms of time (exactly at midnight) and in terms of deed (ability to know who is a firstborn and who is not).

The fourth proem is based on the verse, “At midnight [חצות לילה] I will arise to thank You for your righteous judgments” (Psalms 119:62), which exhibits clear linguistic affinity to the verse at the center of the entire piska, “At midnight” (בחצי הלילה). The midrashic idea is that David thanks God for all that happened to the forefathers at midnight. The midrash contains three alternative midrashim for the verse “for your righteous judgments.”Footnote ¹⁹

1. 1. For the judgments that God brought on evil Pharaoh, and the righteous acts He performed with Sarah, my ancestress.

2. 2. For the judgments He brought on the nations of the world, and the righteous acts He performed for my ancestor and ancestress (Boaz and Ruth).

3. 3. For the judgments he brought on the Egyptians in Egypt and the righteous acts He performed for our forefathers in Egypt.

The section concerned with Israel in Egypt is placed last—not according to its biblical order, which would have placed it second. This allows the darshan to conclude with the topic at hand, the plague of the firstborn. However, the derashah does not follow the typical pattern of a proem—concluding with the first verse of the public Torah reading. Rather it concludes:

As with the case above, Goldberg viewed this conclusion as evidence that the piska is not “original.” In my opinion, however, this is further evidence of an editorial process that was arrested at an incomplete stage.Footnote ²⁰ The editor used a derashah whose original form is found in midrash Ruth Rabbah (6:1). There, the sections are found in the correct and logical biblical order. The last section, which refers to Ruth, ends with the verse featured in the derashah, “And you put a blessing in his heart, as it says, ‘Blessed are you my daughter to the Lord’” (ונתת בלבו ברכה, שנא' ברוכה את בתי לי"י) (Ruth 3:10).Footnote ²¹ The editor of the Pesikta adjusted the order of the sections but did not adjust the ending of the derashah such that it would accord with the structure of a proem. It is possible that he preferred to retain the conclusion because it connected to the topic of Passover eve.Footnote ²²

The Middle Section of the Piska

As stated above, this section includes five interpretive derashot related mostly to the first verse of the Torah reading, “And it happened at midnight and the Lord struck every firstborn in the land of Egypt, from the firstborn of Pharaoh who sat on his throne to the firstborn of the captive who was in the dungeon, and all the firstborn of the cattle” (Exodus 12:29). All of these derashot exhibit close affinity to the Mekhilta.Footnote ²³

For example, the first derashah (section 5) opens with a direct quote from a tannaitic source very similar to the text found in Mekhilta de-Rabbi Ishmael: “R. Shimon b. Yoḥai taught: Moses knew neither its times. …”Footnote ²⁴ The second derashah (section 6) also concludes with a tannaitic source parallel to that found in the Mekhilta: “It was taught in the name of R. Natan: On the day that one of their first-born died. …”Footnote ²⁵

This derashah, which describes the impact the plague of the firstborns had on houses in which the firstborn had already died before the plague, is attached in the Mekhilta to v. 30, “For there was no house in which there was no dead” (כי־אין בית אשר אין־שם מת). It is noteworthy that in the Pesikta it directly continues the derashah on v. 29: “You should be perplexed—a house in which there is no firstborn to the woman or to the man, how can I fulfill [the verse] for there was no house in which there was no dead” (הגע עצמך בית שאין שם בכור לאיש ולא לאשה מה אני מקיים כי אין בית אשר אין שם מת). This shift is an excellent example of the overall aim of our piska's editor to focus on the first verse of the public Torah reading.

The following derashah (section 7) also opens with a teaching that has clear parallels in both Mekhiltot, “From here [we learn] that Pharaoh too was a firstborn”Footnote ²⁶ and continues with an Aramaic aggadah on the words, “He struck the Egyptian firstborns.”

The derashah in section 8 is of special interest:

This derashah explains why the firstborns of handmaidens died and were specified in the verse: The punishment is retribution for their wish to remain enslaved as long as Israel too would remain enslaved. While the derashah seems to relate to Exodus 12:29, the phrase, “the firstborn of the handmaiden,” is found only in a parallel verse that appeared earlier, in Moses's speech (Exodus 11:5): “And every firstborn in the land of Egypt will die, from the firstborn of Pharaoh until the firstborn of the handmaiden sitting behind the mill-stone, as well as all the firstborn of the cattle” (ומת כל־בכור בארץ מצרים מבכור פרעה היֹשב על־כסאו עד בכור השִפחה אשר אחר הרחיִם וכל בכור בהמה). The derashah itself is a tannaitic tradition ascribed to R. Elazar son of R. Yose Hagalili, and it is found in a similar form in Mekhilta de-Rabbi Shimon b. Yoḥai.Footnote ²⁷

There is no way to determine with certainty why the editor of the Pesikta placed this derashah here, out of its correct place. However, the realization that the editor had sources available to him that were mostly similar to those found in the two Mekhiltot can help us make an educated guess as to why he did so. While the derashah in Mekhilta de-Rabbi Shimon b. Yoḥai (pp. 28–29) is based on the words, “the firstborn of the captive” (בכור השבי), it already shifts the focus to the words “the firstborn of the handmaiden” (בכור השפחה) found in the earlier verse:

We can presume that the editor of the Pesikta wanted to include R. Eliezer's derashah on the verse at hand, “the firstborn of the captive,” but as occurred above, the process of adapting the derashah to its new location was not completed. Thus it looks as if the derashah is related to the words “the firstborn of the handmaiden.” Whatever the actual reason for why this derashah is not found in its proper context, it is clear that the derashot in the Pesikta exhibit a strong connection to those in the Mekhilta. This connection is also manifested in the next derashah (section 9), which also has a parallel in the Mekhilta, from which it too was likely drawn.Footnote ²⁸

The Concluding Section of the Piska

The third section, which concludes the piska, is composed of three summarizing narrative derashot on the ten plagues, followed by a concluding derashah (sections 10–12). The first derashah (section 10) emphasizes the idea that “the Master of Mercy does not first afflict human beings” (אין בעל הרחמים נוגע בנפשות תחילה). God is cautious not to hurt human beings until all other possibilities of affecting human behavior have first been exhausted. This principle is illustrated with four examples: Job, Mahlon and Kilyon (from Ruth), the scale diseases (נגעים) described in Leviticus 13–14, and the plagues in Egypt. A similar derashah is found in Vayikra Rabbah 17:4 (ed. Margulies, p. 381) but there the order is at first chronological: Job, Egypt, Ruth;Footnote ²⁹ it then concludes with the topic at hand in Leviticus, the scale diseases. Friedman (Ish-Shalom) seems correct in his proposal that the derashah's original setting was Vayikra Rabbah, where the references are in chronological order.Footnote ³⁰ The order in the Pesikta was created by moving the stanza concerning Egypt to the end, without also moving the scale diseases stanza before the stanza on Ruth. More significantly, the stanza dealing with scale diseases is interpreted based on verses taken from the portion of Leviticus on which the derashah is located, whereas the stanza referring to the plague of the firstborn is not interpreted at all in reference to verses from Exodus. Rather, the verses it employs are from Psalms. This is a strong indicator that the derashah was created in the context of Leviticus and only later brought to the Pesikta, where it was placed in the context of Exodus. Again, we can see that the editor used a derashah from an earlier source and adjusted it accordingly, such that it concluded with the stanza on Egypt. However, he did not fully adapt it to its new location by referring to verses in Exodus.

The next clause (section 11) includes two topical derashot that survey the ten plagues from an innovative interpretive perspective.Footnote ³¹ The first derashah counts the plagues as if they were the “the order of battle of kings” (טכסיס מלכים). The second derashah enumerates them with a look to the future, “the one who punished the former, He is the one who will punish the latter” (מי שפרע מן הראשונים הוא יפרע מן האחרונים). The derashah demonstrates that all of the plagues of Egypt will be meted out to Rome (אדום) in the future. Towards its conclusion the derashah reads:

Another derashah, which serves as a peroration (section 12), follows and continues this derashah:

This derashah connects directly with the verse cited in the previous derashah, Isaiah 34:11, concerning the plague of darkness. The motif of Rome being struck by dark chaos and emptiness is repeated. However, the connection between the peroration and the parashah is weak—it too deals with verses from Isaiah and to a certain extent with the plague of darkness. In reality, the source of the derashah is Vayikra Rabbah 6:6 (ed. Margulies, p. 146), where it serves as a conclusion to the section based on the verse, “If a person sins when he hears a public charge” (ונפש כי תחטא ושמעה קול אלה) (Leviticus 5:1). The conclusion is tightly connected there through the motif of “received the Torah” (קבלו את התורה) found in the words of the rabbis.Footnote ³² The same concluding derashah is also found in the Pesikta, in piska 21, “Arise, give light” (קומי אורי) where it is well connected to the piska's content. In the piska, “And it happened at midnight” the concluding derashah is not linked closely to the overall topic of the piska. Its main motif is “darkness,” which is not mentioned in the remainder of the piska. The weak connection that the peroration has in our piska, “And it happened at midnight,” with the larger parashah testifies again to the incomplete editorial process that the material in this piska underwent.Footnote ³³

Summary of the Literary Structure of the Piska

In sum, the structure of the piska shows that it has undergone a certain amount of literary editing. Its various sections highlight the plague of the firstborn, and the end of the piska provides a summary of all ten plagues, both topics appropriate to Passover. It is not completely disorganized, as Goldberg claimed. However, there are multiple instances in which the structure of the derashot deviates from the literary norms found elsewhere in the Pesikta. Instead of accounting for these differences by positing that these sections were imports into the Pesikta that occurred at a later stage in its transmission, I have explained these as instances in which the editorial process of adapting literary material to new settings did not come to its full fruition. The editor used available sources to fashion a new piska for a new public Torah reading that had not, as of yet, had a full piska or parashah created for it.Footnote ³⁴ The proof that these derashot were secondary literary creations by the editor is found in the traces of their original settings left in the derashot. The failure of the editor to fully integrate the material into its new setting is what allows us to trace their history and to reveal his editorial work.

The Liturgical Background of the Piska “And It Happened at Midnight”

One of the main proofs cited by scholars who held that this piska is not original to the Pesikta is the lack of documentation for such a festival Torah reading in early halakhic sources and piyyutim. This claim should be reexamined. According to M. Megillah 3:4, on Passover they read, “the section in Leviticus related to the appointed times” (בפרשת מועדות של תורת כהנים). The Mishnah does not distinguish between the first day of Passover and the other days. The Tosefta (Megillah 3:5), on the other hand, does offer such a distinction:

While the reading mentioned in the Tosefta for the first day of Passover is somewhat problematic, it seems to be consistent with that in the Mishnah.Footnote ³⁶ In any case, the second line of the Tosefta, “And on the other days of Passover they skip from one passage about Passover to another that is written in the Torah,” hints at a new custom, one that had not yet been fully determined, of reading portions from the book of Exodus on the other days of the festival, a custom that was later more firmly established in Babylonia and recorded in the Bavli.Footnote ³⁷

A similar shift in Torah readings from Leviticus/Deuteronomy to Exodus occurred with the reading for Shavuot. The Mishnah mentioned above relates that on Shavuot they would read the passage “Seven weeks” (שבעה שבועות) in Deuteronomy (16:9), whereas the Tosefta reports an alternative tradition, according to which they would read “On the third month” (בחדש השלישי), a passage from Exodus (19:1).Footnote ³⁸ Taken all together, it seems that the Tosefta demonstrates a process whereby readings from Leviticus and Deuteronomy, mostly halakhic passages, were replaced by more narrative passages from Exodus. These readings were perceived to serve as a better basis for derashot that would shape the character of the holiday and attract a greater audience.Footnote ³⁹

This shift developed and became entrenched over time, a phenomenon we can already detect within the Tosefta itself. For Shavuot, the new custom had already taken shape and therefore it is mentioned explicitly in the Tosefta. In contrast, the new custom of reading from Exodus on Passover seems to still be in formation, and thus it is cited in the Tosefta only for the remaining days of the festival, without any obligatory order as to which passage should be read on which day. Only at a later stage did the custom of reading from Exodus on the first day of Passover take root and replace the older custom. According to this reconstruction, the custom of reading “And it happened at midnight” on Passover is not a Babylonian custom, as Goldberg posited. It is indeed a Palestinian custom, although later than the earlier custom of reading from Leviticus. Ish-Shalom's assertion (adopted by Heinemann as well) that this is an “early Palestinian custom which the halakhah came to uproot” has no supporting evidence.Footnote ⁴⁰ Taking these two points into account, it seems likely that this development occurred in Palestine during the Byzantine period, probably sometime around the fifth century. This is also supported by the lists of liturgical poets (paytanim) from the end of the Byzantine period (Moses, Pinḥas) that take account of readings for the intermediate days of Passover, as cited by Heinemann in his article.Footnote ⁴¹ Goldberg rightfully rejected this as evidence for the antiquity of the custom. Nevertheless, in my estimation this finding does provide evidence for the fact that this was a new custom in Palestine itself.

This discussion concerning the custom of reading “And it happened at midnight” on the first day of Passover is also relevant for the custom of reading “And it happened when he sent out” (ויהי בשלח) on the seventh day of Passover. There is no halakhic or liturgical evidence that either of these customs existed in Palestine in antiquity (in the Mishnah and Tosefta).Footnote ⁴² In light of the argument above, we can assume that this custom also took shape during the Byzantine period in Palestine, and was alluded to in the Tosefta with the words, “the passages about Passover that are written in the Torah” (עניינות הפסח הכתובין בתורה).

I suggest that the editing of the Pesikta occurred during the same time period as the shift in Torah reading customs for Passover from Leviticus to Exodus was happening. This explains why the Pesikta contains a piska for “And it happened at midnight,” from Exodus, which accords with the new custom alluded to in the Tosefta, alongside a piska for “An ox or sheep,” from Leviticus, which accords with the Torah reading custom referred to in the Mishnah. The same is true for the piska, “And it happened when he sent out” for the seventh day of Passover, also alluded to in the Tosefta. In both cases, we can sense that the customs are in a state of transition, but that this change has not yet been completed. This state of transition is reflected in the halakhic texts by the differences between the Mishnah and the Tosefta, and in the midrashic texts, by the inclusion of piska'ot for both readings. In contrast, for Shavuot, the Pesikta includes a piska only for “On the third month,” which accords with the Torah reading custom mentioned explicitly in the Tosefta. There is no piska whatsoever for “Seven weeks,” the custom mentioned in the Mishnah. It seems that the new custom of reading from the book of Exodus on Shavuot had been sufficiently established to push aside the earlier custom entirely.

* * * * *

We can now evaluate the literary findings that emerge from the analysis of the piska “And it happened at midnight” in light of the theory proposed above concerning the editing of the Pesikta. The piska's editing is coherent; it is not “confused” as it was characterized by Goldberg. It includes three sections of derashot with a consistent theme: The plague of the firstborn is a summation of the entire process of the ten plagues. The composition of the piska and the relationship between the number of proems and the number of interpretive derashot on verses is not “unusual,” and does not serve to indicate any lack of “authenticity.”Footnote ⁴³ The unusual features that were uncovered in the piska can be understood as the result of an editing process that was not fully completed. This explains, for instance, the two proems whose conclusions do not match the content, as well as the concluding derashah that seems disconnected from its context. This editing process also explains the editor's massive dependence on the Mekhilta in the interpretive sections of the piska, and his adherence to the interpretation of the first verse of the parashah.Footnote ⁴⁴ Finally, the same phenomenon explains the secondary use that the editor made of the piska “Arise, give light” (קומי אורי) in the second proem and in the peroration.

All of these unusual features are indicators of purposeful and creative editing of material available to an editor whose aim was to fashion a piska for “And it happened at midnight,” in a midrashic format similar to Vayikra Rabbah.Footnote ⁴⁵ The editor of the Pesikta lived at a time when Torah reading customs were evolving, and he may have even wanted to support these changes. This historical background serves to explain why there are differences between the various piska'ot. In other words, the literary differences between the various piska'ot reflect the availability of materials to the editor, and this availability is largely a factor of the antiquity of the Torah reading custom at the heart of the individual piska. Piska'ot for Torah passages that had been part of the Torah reading for a long time, at least from the time of the Mishnah, such as “Shekalim,” “This month,” and others, were simpler for the editor to compose, because a significant amount of midrashic material suited to the common midrashic formula had already accrued on these passages.

In contrast, the composition of piska'ot for readings that had only just been instituted, such as “And it happened at midnight,” “And it happened when he sent out,” or “In the third month,” presented a greater challenge for the editor.Footnote ⁴⁶ Here he was forced to be much more creative and to use material that was not created specifically for holiday Torah readings. This explains the intensive use of the Mekhilta in the interpretive sections as well as the secondary use of the piska “Arise, give light” in the proem and conclusion.

The supposition raised by some scholars that the piska was copied, by the editor or by later copyists, from a lost interpretive midrashic composition on Exodus is implausible. First of all, there is no evidence that such midrashic collections existed, not even as fragments that were later embedded into Tanḥuma-Yelammedenu literature.Footnote ⁴⁷ But beyond this argument from silence, the analysis of the piska itself makes this supposition unlikely. If this hypothesis was correct, we would expect to find uniformity in the five piska'ot in the Pesikta that relate to readings from the book of Exodus. But in reality, the signs of incomplete editing found in “And it happened at midnight” are not present in other piska'ot based on readings from Exodus, “Shekalim” and “This month.”Footnote ⁴⁸

The literary differences between the various piska'ot are well explained by the theory proposed here. For Torah readings that had long been part of the reading cycle, there had already accrued ample midrashic material, whereas for readings that had only recently entered the cycle, the editor was forced to recycle midrashic material lifted from other settings: the Mekhilta, other classic midrashim, and other piska'ot in the Pesikta itself. Moreover, the connections that exist between this piska and the piska, “Arise, give light” negate the possibility that this piska was taken from an edited and fully formed “homiletic” midrash.Footnote ⁴⁹

Similarly, the suggestion that copyists took this piska from Pesikta Rabbati and imported it into the Pesikta is problematic. First of all, the piska is found in most Pesikta manuscripts (א1, א2, ק, כ)Footnote ⁵⁰ as well as a Genizah fragment (T-S 16.93).Footnote ⁵¹ More substantively, the piska is not structured in the typical style of Pesikta Rabbati, a composition that generally adheres to the Tanḥuma genre of midrashic literature.Footnote ⁵² Moreover, in the Parma manuscript of Pesikta Rabbati, which is the most complete manuscript, there is an additional “Tanḥumaic” piska that is parallel to “And it happened at midnight” (ed. Ish-Shalom, pp. 195–7). It is very unlikely that both piska'ot were originally part of the same composition.Footnote ⁵³ Goldberg is correct in his assertion that “a confusion of proems, topical derashot, a new proem and then a return to topical derashot following derashot based on verses” is typical of Tanḥuma literature, whose structures are looser than those found in classic midrashim.Footnote ⁵⁴ However, as demonstrated above, the piska is not entirely “confused.” It has an organized structure of three sections: a proem, derashot on the first verse, followed by a conclusion that summarizes the ten plagues. While this is not the typical structure found in the Pesikta, it is still not so “unusual” as to force the claim that it is a late “Tanḥumaic” creation.Footnote ⁵⁵

All in all, it seems that both the literary process—the creation of the Pesikta as described here—as well as the liturgical developments that occurred during the same period—the transition to holiday Torah readings from the book of Exodus—reflect a similar trend. Both of these phenomena are connected to a shift in the rabbinic world to appeal to a broader, less rabbinic audience by changing the festival Torah reading customs and the midrashic rhetoric that accompanied them.Footnote ⁵⁶ The Torah reading customs found in the Mishnah, those taken from the chapters in Leviticus and Deuteronomy concerning the festivals, are mostly halakhic in their content. Halakhic topics are more appropriate for a rabbinic audience steeped in the nuances and particulars of Halakhah. For the Torah reading for Shavuot, the Tosefta already documents a shift to the passage in Exodus concerning the revelation on Sinai. The next step in this process seems to have been a shift to reading the story of the Exodus in the book of Exodus itself for the first day of Passover, and the passage on the splitting of the sea for the seventh day of Passover. These readings are mostly narratives; they allow the darshan to relate to the history of the festival and the reason it is observed, and to shape rhetoric that would be more appealing and meaningful to less educated synagogue attendees, and not just rabbis well versed in law. In these historical and social circumstances there developed a need for a literary composition to aid the darshan in coping with the new customs.Footnote ⁵⁷

Analysis of the literary phenomena occurring in this piska reveals the rich, surprising, and dynamic world of cross influences between literary and liturgical processes that left a deep mark on the formation of one of the most important and puzzling midrashic compositions from the amoraic period, the Pesikta de-Rav Kahana.