Difference makes for comparisons, comparisons give rise to uneasiness, uneasiness to wonderment, wonderment tends to admiration; and finally admiration turns to a yearning for mutual exchange and unity.
–Thomas Mann, Die Vertauschten Köpfe Footnote 1Introduction
Comparative studies of Indian and Jewish rituals played a distinct role in the development of the scholarly study of religion. A number of formative eighteenth- and nineteenth-century studies, from de la Créquinière to Hubert and Mauss, focused on Indian and Jewish rituals, exhibiting a special interest in ritual sacrifice.Footnote 2 Subsequently, it was primarily each on its own terms, rather than in juxtaposition with one another, that these traditions continued to inform the theoretical study of religion (as exemplified in the works of Staal, Heesterman, Douglas, and Halbertal);Footnote 3 but the last few decades have witnessed a renewed and growing interest in studying these sacrificial traditions jointly, as in the works of Holdrege, McClymond, and Stroumsa.Footnote 4
These joint studies of Vedic/Brahmanic and biblical/Jewish sacrifice have tended to focus on structures and meanings of ritual practices, using discourses on sacrificial ritual only inasmuch as they were instrumental in reconstructing those structures and meanings. Thus, despite their methodological sophistication and breadth of scholarship, and despite the fact that they refer to rabbinic and Brahmanical literature, these studies do not venture to ask explicitly whether and how a joint study of Indian and Jewish analytic and theoretical discourses about sacrifice might inform a comparative study of sacrifice and ritual more generally.
Nevertheless, scholars have noted that highly nuanced emic analytical frameworks, heuristically valuable for the investigation of the relation between sacrificial action and sacrificial text, were developed in Indian as well as Jewish literature already in antiquity. In India, during the few centuries before and after the turn of the era, a special field of knowledge developed that almost exclusively undertook this investigation, which later became famous under the name of Mīmāṃsā. In Jewish contexts, such analytical frameworks are found, admittedly in a less straightforward and less systematic form, within Talmudic literature broadly defined.Footnote 5
Thus, this study builds upon previous comparative studies dedicated to the Vedic/Brahmanical and biblical/Jewish sacrificial traditions. It differs from them in that it focuses neither upon the sacrificial rituals themselves (reconstructable from textual and archaeological sources) nor on the ritual prescriptive and descriptive texts that reflect a first order of conceptualization. Rather, this study focuses on a second order of conceptualization, namely, on the intellectual discourses on patterns, rules, principles, and problem-solving techniques developed for the study of those sacrificial texts that constitute the first order of conceptualization. These discourses, found in Mīmāṃsā and in Talmudic literature, developed rule-based hierarchies that govern the ritual complex as well as abstract categories such as primacy/secondariness and parity/disparity, and sets of epistemic means for deducing information from the earlier corpora. The scholastic texts on sacrifice are thus not a window through which one may desire to observe ritual practice but are themselves the object of comparative inquiry. Though they are at least twice-removed from the rituals themselves, focusing on the later scholastic texts offers a unique perspective inasmuch as these texts, in their attempt to understand the textual-ritual tradition of which they are part, are self-reflective in ways that the earlier texts are not.
Studying these scholarly traditions is therefore doubly informative for any science of ritual: in the ways they resemble any modern theory, and in the ways they consistently depart from the particular explanations and techniques favored by modern theoreticians. The former suggests the existence of a cross-cultural object of scientific study, while the latter may contain some insight into the logic of rituals that might not be apparent to modern theoreticians.
Moreover, if we desire to think not only about rituals but about how to study rituals, and how to analyze and conceptualize them, a comparison of these two scholarly traditions, which are not quite paralleled outside Indian and Jewish literatures, appears to be a sine qua non. Such a comparison offers a rare opportunity to characterize conceptual discourse about ritual, opening the door to a series of more general questions about ritual taxonomies; the relation between hermeneutics and ritual; the fundamentals of legal thought, inasmuch as it is indelibly linked to ritual thought in these traditions; and the contribution of ritual thought to the emergence of systematic scientific reasoning. This study is a modest first step in that direction.
I. Specific Goals of this Study
Scholars of Talmudic literature, while acknowledging the great service of comparative studies with Greco-Roman, Syriac, and Iranian (primarily Sassanian) materials, tend to characterize the Talmud as a work that is sui generis.Footnote 6 Similarly, in his overview of the scholastic tradition of Mīmāṃsā, James Benson reflects sentiments widespread among Indologists when he writes, “the subject [of Mīmāṃsā] has no familiar counterpart outside the Indian tradition. This is not to say that it is an intellectual tradition without parallels, but only that it is not clear, or at least not well-known, what those parallels might be.”Footnote 7
The first goal of this article is to offer a response to these claims by sketching out, in a preliminary fashion, some general convergences alongside major divergences between Mīmāṃsā and Talmudic literature on ritual sacrifice (section II). Thus, it offers both Mīmāṃsakas and Talmudists a rare opportunity to examine these idiosyncratic discourses from a comparative perspective.
The second goal is to demonstrate these convergences and divergences through the juxtaposition and subsequent “transposition” (a process described below) of two passages, one Talmudic, the other Mīmāṃsic, which share thematic, structural, and discursive properties (section III). Through the selected pericopes, we show that byproducts of the sacrificial process pose serious conceptual problems within both traditions—though not necessarily the same problems—and that from these conceptual challenges the two traditions derived strikingly similar dilemmas, debating whether a sacrificial ritual should be re-performed if byproducts are missing.
Finally, we point in the direction of future explorations of the intellectual space of Talmudo-Mīmāṃsā, asking how the convergences between the two systems might have arisen. We suggest, however tentatively, that both might contribute to an etic theory of ritual, setting into relief analytic concepts and operative categories implicit in both scholastic-ritual traditions, the full extent of which may be better appreciated by the joint analysis of Mīmāṃsic and Talmudic texts.
II. Talmudo-Mīmāṃsā
A. Introducing Mīmāṃsā and Talmud
The system of Mīmāṃsā originated in the second half of the first millennium BCE in South Asia.Footnote 8 It arose within the context of the Vedic textual and ritual tradition, reflected in an extensive corpus beginning with the Vedic hymns and a body of literature stemming from reflective engagement with sacrificial rituals. This body of literature extended from reflection upon the sacrificial rituals’ mythical and cosmological associations (as in the Brāhmaṇas) to systematic compilations of their structure and specific ways to execute them (as in the Śrauta-Sūtras).Footnote 9
Within this extensive intellectual enterprise, associated with and anchored in sacrificial rituals, a specialized “science,” Mīmāṃsā, developed. Its main focus was correlating and aligning ritual actions with the textual corpus that enjoins them, with the further claim that this “science” is knowledge of dharma, which is centered in Vedic injunction. Thus, this school of inquiry, encompassing both a “science of sacrifice” and a “science of hermeneutics,”Footnote 10 was grounded, on the one hand, in the vast preexisting ritual literature, and on the other hand in the contemporary performance of sacrificial rituals.
The earliest textual source of Mīmāṃsā is a collection of short statements (sūtras) attributed to Jaimini (Jaimini-Mīmāṃsā-Sūtras = JMS, ca. 3rd cent. BCE). The statements of JMS are terse and at times somewhat cryptic, as they do not mention the ritual and textual contexts in which they were made.Footnote 11 An extensive commentary (Bhāṣya) attributed to Śabara (ŚBh, ca. 3rd cent. CE) is the earliest extant comprehensive explanation of JMS, wherein Śabara provides the missing ritual and textual contexts, groups the short statements into thematic sections, and elaborates upon them following specific dialectical patterns.Footnote 12 ŚBh presents a well-knit fabric of complex argumentational threads, creating a debate interspersed with the voices of a moderator, who introduces the subject matter and formulates the doubt, followed usually by an opponent (Pūrvapakṣin), who advances the prima facie position, which is then countered by the proponent (Siddhāntin; other voices can be included as well). While the ensuing debate can run into several rounds, it usually culminates in a settled conclusion.Footnote 13
The sacrificial tradition of classical Jewish literature is based on the descriptive and prescriptive Israelite ritual texts in the Pentateuch, primarily in the priestly literature within the books of Exodus, Leviticus, and Numbers (compiled ca. mid-first millennium BCE), and on their further elaboration in postbiblical literature, specifically in the Mishnah and in the Sifra (ca. 3rd cent. CE).Footnote 14 Two of the Mishnah’s six orders, Qodashim (sancta) and Teharot (purities), as well as numerous other sections (such as tractate Yoma on the Day of Atonement and tractate Pesahim on the paschal ritual) are dedicated to the two interrelated systems of sacrifice and purity, which together form the most elaborate intellectual edifice in ancient Judaism. These mishnaic texts are a topically ordered collection of statements and disputes on ritual minutiae that reflect the beginnings of a systematization of the principles upon which these biblical/Jewish ritual systems are grounded.
The Babylonian Talmud (BT, compiled ca. 6th cent. CE, in Sassanian Mesopotamia), though not the earliest extant commentary on the Mishnah,Footnote 15 came to be the lens through which the Mishnah was read throughout the Middle Ages. The Babylonian Talmud follows the topical order of the Mishnah, presenting elaborate arguments and counterarguments regarding its particularities and generalities. Accompanied by a host of commentaries and sub-commentaries, BT became the quintessential text of traditional Jewish learning.Footnote 16
For the present study, it is helpful to consider the broad diachronic contours of the Mīmāṃsic and Talmudic scholastic traditions as follows: in both cases, an ancient (Vedic, Pentateuchal) corpus, which is primarily injunctive, underlies a ritual-legal compendium that is rather concise and topically ordered (JMS, ca. 3rd cent. BCE; Mishnah and Tosefta Order Qodashim, ca. 2nd–3rd cent. CE). This later legal-ritual compendium is anchored in and thoroughly permeated by the ancient injunctive corpus, but it often assumes it without citing it explicitly. Consequently, an extensive, discursive and dialectic composition is created as a commentary upon the legal-ritual compendium (ŚBh, ca. 3rd cent. CE, on JMS; BT, ca. 6th cent. CE, on the Mishnah), and among its many achievements and goals it identifies the scriptural grounding of many of that compendium’s basic legal-ritual principles. As a result of this process, parts of the concise compendium (JMS, Mishnah)Footnote 17 are often intelligible only in light of its commentary (ŚBh, BT), even while the commentary often acts as a lens through which fundamental aspects of the earlier compendium are distorted.Footnote 18 In later generations, throughout the Middle Ages and down to modernity, the composition of commentaries and subcommentaries has generated a vast ocean of scholastic literature dedicated to the elucidation of the ancient sacrificial-textual tradition.Footnote 19
In both cases, the scholastic traditions are not identical with the ritual traditions of practitioners. Mīmāṃsakas are, on the one hand, conversant with the ritual world, but they are, in the first place, theoreticians engaged in abstracting and developing theoretical categories and principles. Their expertise is thus distinct from that of the Yājñikas, experts in ritual practice.Footnote 20 In rabbinic literature the distinction between the two types of expertise is even starker. Since sacrifice outside the Jerusalem Temple is unacceptable in Jewish law, it essentially ceased with the destruction of the Jerusalem Temple in 70 CE. While some rabbinic traditions may trace back to a period from before this and therefore preserve insights from ritual practitioners, identifying them is difficult.Footnote 21 Rabbinic literature was compiled when sacrifice was no longer practiced, sometimes many centuries later.
B. Carving Out the Intellectual Space of Talmudo-Mīmāṃsā
Although the two textual traditions developed in isolation from one another in antiquity and evolved along separate trajectories in subsequent centuries, there are significant affinities between them in terms of contents and style. With a degree of simplification, both traditions can be characterized as containing dialectic, textually stratified theoretical discourses, which attempt to systematize a complex corpus, employing fixed and often nonintuitive hermeneutical tools for textual deduction and conventional scholastic catchphrases that are occasionally identical in their specific function.Footnote 22
Perhaps the most striking similarity between the two traditions is a thematic overlap between the types of problems they address. Similar nonobvious dilemmas are raised and nonintuitive hermeneutic techniques are employed in order to entertain readings that are improbable in terms of the plain sense of the earlier text, as we shall see in the example below.
Significant dissimilarities between the two scholastic traditions are no less noteworthy. Mīmāṃsā is dedicated almost exclusively to sacrifice,Footnote 23 whereas the Talmud and its hermeneutics cover a broad variety of domains, including matrimony, torts, and ritual purity, areas that in the Indian tradition would belong to the realm of Dharmaśāstra (which itself might be infused with Mīmāṃsā-like ways of thinking). Moreover, BT contains a significant amount of non-halakic materials as well (including narrative, ethical, and to a lesser extent mythological-cosmological), which lend themselves more easily to comparison with the Upanishads, the Brāhmaṇas, the Purāṇas, and Vedānta (Uttara-Mīmāṃsā) literature.Footnote 24
In terms of attribution vis-à-vis anonymity, ŚBh and BT (and to some extent, early Mīmāṃsā versus early rabbinic works in general) are nearly mirror images. With regard to authorship, ŚBh bears a much stronger stamp of a single author than the late antique rabbinic works, which are collective and anonymous in the sense that the individual positions and arguments are embedded within an anonymous, collective editorial layer.Footnote 25 Partially as a result of this situation, JMS and ŚBh reflect a degree of large-scale structural and organizational coherence not found in rabbinic works prior to the High Medieval period (e.g., Maimonides). As becomes a complex anthology, BT tractates often begin in medias res, are replete with digressions and duplicate pericopes, and the structure of each pericope (including the number of disputants participating) may vary greatly.Footnote 26
With regard to appeals to authority, Talmudic and Mīmāṃsic works also operate very differently: the statements and controversies compiled in early rabbinic works are often attributed to named authorities—Tannaim (2nd–3rd cent. CE) and Amoraim (3rd–5th cent. CE).Footnote 27 The controversies reflected in ŚBh, on the other hand, are almost exclusively between unnamed and perhaps ahistorical interlocutors—Pūrvapakṣins and their victorious opponents, Siddhāntins.Footnote 28
Finally, while both Mīmāṃsic and Talmudic discourses are characterized by a tension between local exegetical concerns and the formulation of underlying generalized principles, in antiquity Mīmāṃsā shows a greater degree of thorough systematization and explicit abstraction.Footnote 29 Mīmāṃsā’s explicit attention to abstract categories is also expressed in terms of its structural organization. Whereas mishnaic tractates tend to be organized around topical principles such as animal sacrifices, grain offerings, and the offering on the Day of Atonement (Tractates Zebahim, Menahot, and Yoma, respectively),Footnote 30 the organizing principles of JMS’s sections are often abstract categories, such as difference or non-difference between acts, the distinction between primary and subsidiary acts, or the order of their performance (books 2, 3, and 5, respectively).Footnote 31
Recognizing the vast cultural, religious, historical, geographical, and linguistic dissimilarities between the two corpora, this study does not aim to establish a shared historical origin, much less to hypothesize that one scholastic or ritual tradition influenced the formation and development of the other, directly or indirectly.Footnote 32 Neither do we suggest that the shared properties described here should be interpreted as evidence of underlying universals. In fact, Mīmāṃsic and Talmudic discourses on sacrifice appear to be exceedingly idiosyncratic when compared with the ritual-textual traditions from contiguous geographic areas from the Mediterranean basin to Iran,Footnote 33 including the copious Zoroastrian literature on ritual.Footnote 34
The convergences between these two distant scholastic traditions are best demonstrated by considering a control group consisting of works that are historically adjacent to rabbinic discourse on sacrifice: Philo (Alexandria, early 1st cent. CE), the New Testament (e.g., the Epistle to the Hebrews, second half of 1st cent. CE), Flavius Josephus (Jerusalem and Rome, late 1st cent. CE), Porphyry (Tyre, 3rd cent.), Iamblichus (Syria, 3rd–4th cent. CE), and, in a different way, Zoroastrian texts (such as the Vidēvdād, on ritual pollution, 6th cent. CE). Despite their robust and at times highly specialized discourses on ritual, these texts tend toward the symbolic (in the Jewish-Hellenistic texts), the theurgical (in Iamblichus), and the exegetical-commentarial (in the Zoroastrian texts); and they do not display an interest in the same kinds of hypothetical questions and syntactic hairsplitting exemplified below from the Mīmāṃsic and Talmudic texts, nor do they share similar conceptions of what types of arguments qualify as acceptable retorts.Footnote 35 These late antique texts belong to milieux that are, compared to JMS and ŚBh, much closer to those that produced classical rabbinic literature, due to their geographic overlap and due to the linguistic (Greek, Latin, Aramaic, Persian) points of contact between them. The affinity between rabbinic literature and Philo, Josephus, and Hebrews is particularly significant inasmuch as they all share a common biblical (including Levitical) textual heritage.Footnote 36 Yet, despite their geographical proximity and shared background, rabbinic discourse on sacrifice evidences a much greater affinity to Mīmāṃsā in terms of the thematic, discursive, and conceptual similarities discussed below than it does to the sections on sacrifice in these neighboring works.
The existence of such convergences despite the absence of any significant historical point of contact leads to the ineluctable question: if the similarities do not allow for cladistic, diffusionist or universalistic explanations, how are they to be accounted for? We will address this question in section IV, after first examining the texts themselves.
III. JMS 4.1.27 and BT 52b
A. Blood and Dung in JMS 4.1.27
JMS 4.1.27 is part of a series of pericopes (adhikaraṇas) beginning in 4.1.22, that contain test cases for determining what components within a sacrificial ritual are its “instigator” (prayojaka, prayoktṛ), i.e., its “prompter” or raison d’être,Footnote 37 which causes it to be carried out in the first place. The prima facie view, which serves as a baseline position for several debates in the subsequent sūtras, states that “when [several things are] brought about by the same [act, they] should all equally be regarded [as instigators of that act].”Footnote 38 As ŚBh expounds, the case alluded to in JMS 4.1.27Footnote 39 is part of the Agniṣṭoma, pertaining to the elaborate soma-sacrifice.Footnote 40 Within this ritual, portions of a sacrificial animal (paśu),Footnote 41 including cuts from the animal’s heart and tongue, are offered to the deities Agni and Soma (hence the animal is called agnī-ṣomīya).Footnote 42 The Vedic instructions pertaining to the agnī-ṣomīya, which include an injunction to remove its dung and blood, are the basis of the following discussion, which aims to determine what instigates the ritual—only the offering of the heart and tongue, or perhaps also the removal of the dung and blood.
For ease of reference, the various components of the text are reproduced here in different fonts,Footnote 43 and the Siddhāntin’s and Pūrvapakṣin’s voices are introduced by [S] and [P], respectively.Footnote 44
[Sūtra]: paśāv anālambhāl lohitaśakṛtor akarmatvam.
[Bhāṣya]: asti jyotiṣṭome paśur agnīṣomīyaḥ. tatra śrūyate: hṛdayasyāgre ‘vadyaty atha jihvāyā ity evam ādi .Footnote 45 tathā, lohitaṃ nirasyati, śakṛt saṃpravidhyati, sthavimato barhir aṃlktāFootnote 46 ‘pāsyatī ti. Footnote 47 tatra saṃdehaḥ, kiṃ hṛdayādibhir avadānair ijyā paśoḥ prayoktrī, uta śakṛtsaṃpravyādho lohitanirasanaṃ ca tad api prayojakam iti. kiṃ prāptam?
[P]: ekaniṣpatteḥ sarvaṃ samaṃ syād ubhayaṃ prayojakam iti.
[S]: evaṃ prāpte brūmaḥ, paśauFootnote 48 śakṛllohitayor aprayojakatvam, na hi tadarthaḥ paśor ālambhaḥ, śakṛt saṃpravidhyati lohitam apāsyatīty ucyate, na paśor anyasya veti, paśur agnīṣomīyo vākyena, yo dīkṣito yad agnīṣomīyaṃ paśum ālabhata iti. śakṛllohite paśoḥ prakaraṇena bhavetām, prakaraṇaṃ ca vākyena bādhyate.
[P]: nanv ete śakṛllohite pratipādyete, tena yāgārthasya paśor nānyasyeti niścayaḥ.
[S]: evaṃ cet, aprayojake śakṛllohita iti.
kiṃ bhavati prayojanam? sāmye sati śakṛllohitābhāve ‘nyaḥ paśur ālambhanīyaḥ, śakṛllohitayor aprayojakatve lopaḥ.
[Sūtra]: In (the case of) animal (sacrifice)—(the treatment of) blood and dung is not (considered ritual) action—since the sacrificial killing Footnote 49 is not (for that purpose).
[Bhāṣya]: In the Jyotiṣṭoma ritual there is the agnīṣomīya animal. It is said there he takes from the heart first, then from the tongueFootnote 50 etc. and then he pours out the blood, removes the dung, wipes it with the broad side of the barhi-grass and disposes of it.Footnote 51 Regarding this there is a doubt: is the raison d’être of the animal sacrifice the offering of the heart, etc., or is the removal of the dung and the pouring-out of the blood also a raison d’être? What is the conclusion?
[P]: (Since) in case of simultaneous coming-into-being, all are to be considered equally (JMS 4.1.22), both are raisons d’être.
[S]: This having been claimed, we argue: in (the case of) an animal, (there is) non-instigation of dung and blood, for it is not the case that the animal’s killing is for that purpose! (Proof:) it is said he removes the dung, he pours away the blood, not ‘of (this) animal’ or ‘of a different one’. That the animal (sacrificed) is the agnīṣomīya offering (is deduced) by means of syntactical connection (vākyena)—“… the one who is initiated; in that he sacrifices the agnīṣomīya (animal) …” (TS 6.1.11.6).Footnote 52 That the dung and the blood are to be (this) animal’s (is deduced) by means of context (prakaraṇena); context is overpowered by syntactical connection.
[P]: But now, it is undeniable (iti niścaya) that these blood and dung are disposed of, and hence (that they are) of the animal to be sacrificed, not of a different one.
[S]: In that case, the dung and the blood are (certainly) not what instigates the offering.
What is the ramification of this distinction? If they are on a par (= equally instigators), then if dung and blood are lacking, Footnote 53 a different animal is to be killed; but if blood and dung are not instigators, it is skipped.
The sūtra, which assumes outright that an animal is not killed for the sake of disposing its blood and dung, is fleshed out by Śabara as a response, as it were, to a theoretical dilemma: is the instigator of the sacrifice only the offering of the animal’s heart and tongue, or is it perhaps the case that the removal of the animal’s blood and dung, too, are its instigators?Footnote 54
At first, the imaginary interlocutors engage in a hermeneutic debate. The Pūrvapakṣin, adhering to the general rule voiced in 4.1.22, argues that since the injunction to offer the heart and tongue and the injunction to remove the blood and dung both appear in the Vedic text as necessitated by one and the same sacrificial rite, and since there are no clear grounds for drawing a distinction between these two pairs of acts (in terms of their status as instigators),Footnote 55 there ought to be a parity among them (sarvaṃ samaṃ syāt, as in 4.1.22): both the heart-and-tongue apportionment and the blood-and-dung removal are raisons d’être for killing the animal.
However, the Siddhāntin responds, rather commonsensically, that the animal is not killed for the sake of treating its blood and its dung. In fact, he suggests that the Vedic phrasing does supply grounds for drawing a distinction between the two pairs of acts: the fact that the animal is killed to serve as an agnīṣomīya offering is derived from the very syntax of the Vedic statement “… the one who is initiated; in that he sacrifices the agnīṣomīya (animal)….”Footnote 56 But the Vedic injunction (“he removes the dung, he pours away the blood”) does not explicitly state whether it is this particular animal’s, or some other animal’s (na paśor anyasya vā)—which information can only be derived from considering the broader context of the entire passage.Footnote 57 Since syntactic connection (vākya) is a more immediate—and hence epistemologically a more powerful—form of textual deduction than is context (prakaraṇa),Footnote 58 he claims that there are grounds—syntactic grounds—for drawing a distinction between the two pairs of acts. The former are instigators, the latter are not.Footnote 59
A quibble (in shorthand) follows. If Ganganatha Jha’s sensitive reading is to be accepted, then the undertones of the ensuing back-and-forth are as follows. The Pūrvapakṣin quite reasonably insists (nanv ete … pratipādyete) that the blood and dung are undeniably the agnīṣomīya animal’s and not another animal’s, since the Vedic text itself designates these two as disposables (as the verbs nirasyati and sampravidhyati suggest)—implying that they are residual byproducts from the same animal that is under discussion. However, by insisting on this, the Pūrvapakṣin appears to have fallen into a trap set by the Siddhāntin, who now retorts that if so (evaṃ cet)—i.e., if the blood and dung are merely disposable byproducts—then it is all the more evident that they cannot be what prompted the sacrifice in the first place.Footnote 60
The commentary on this sūtra culminates in a summary that draws the practical ramifications from the opponents’ debate.Footnote 61 If, as the prima facie view would have it, the disposal of the blood and dung is no less a raison d’être of the sacrificial ritual than the offering of the heart and tongue, then absent blood and dung a whole new animal must be brought, and the entire ritual must begin anew only to ensure that the requisite blood and dung removal is carried out properly. But if the blood and dung are not instigators and merely byproducts generated in the process of obtaining the materia sacra (as the ultimately victorious interlocutor opines), then, in the event they are lacking, the ritually inconsequential act of removing them is simply skipped (lopaḥ).
B. Blood Applications in BT Zeb. 52b
In the fifth chapter of BT Zebahim (“Animal Sacrifices”), a pericope (b. Zeb. 52b infra) discusses the blood-applications on the Day of Atonement. The pericope harks back to the mishnaic expositions in Tractates Yoma and Zebahim, and ultimately back to the biblical injunctions in Lev 16. On this day, the high priest is to sprinkle parts of the sanctuary with the blood of a bull and of a goat offered as ḥatta’t (“purification-” or “sin-”) offerings, and then decant the leftovers—a detail not explicit in Lev 16 but perhaps inferable from Lev 4:7, 18, etc.—at the base of the bronze altar in the sanctuary’s courtyard.Footnote 62 For ease of reference, italics, bold, and underlining are used to distinguish between various strataFootnote 63 in the text and in the translation.Footnote 64
תנו רבנן: “וכלה מכפר את הקדש ואת אהל מועד ואת המזבח—אם כיפר [כ]לה ואם לא כיפר לא כילה.” דברי ר' עקיבה.
אמ' לו ר' יהודה: “מפני מה לא נאמ' 'אם כילה כיפר—שאם חיסר אחת מן המתנות לא עשה כלום'?”
מאי ביניהו? איתמר: ר' יוחנן ור' יהושע בן לוי. חד אמ' שיירין מעכבין איכא ביניהו. וחד' אמ' משמעות דורשין איכא ביניהו. תסתיים דר' יהושע בן לוי דאמ' שיירין מעכבין, דאמ' ר' יהושע בן לוי: לדברי האומ' שיירין מעכבין מביא פר אחד Footnote 65 ומתחיל כתחלה בפנים.Footnote 66
Our sages taught: “(Scripture reads:) and having finished purging the adytum and the tent of meeting and the altar” (Lev 16:20)—if he has purged, then he has completed, and if he has not purged, he has not completed.” This is the opinion of Rabbi Akiba.
Rabbi Judah said to him: “why shall we not rather say,Footnote 67 ‘if he has completed, then he has purged, Footnote 68 such that if he omitted one of the (blood-)applications, he has accomplished nothing’?”
What is the [practical] distinction between them? It is said: R. Johanan and R. Joshua ben Levi (disagree). One said they (= R. Akiba and R. Judah) differ as to whether the (pouring out of the) residue is indispensable. And the other said they differ (merely) on the mode of interpretation (but not regarding the legal ramifications). Footnote 69 Let it be proven that it was R. Joshua ben Levi who said (they differed as to whether the pouring out of) the residue is indispensable. For R. Joshua ben Levi said: According to the view that the leftovers are indispensable, he (the priest) must bring another bull and begin anew inside (the sanctuary).Footnote 70
The pericope opens with the exposition of a hermeneutic dilemma. Leviticus 16:20 states that “having finished purging (wǝkillâ mikkappēr) the Holy of Holies, the Tent of Meeting, and the altar” with the blood of a bull and of a he-goat, the high priest should proceed to perform the “scapegoat” ritual (vv. 21–22). Grammatically, the words “and having finished purging” constitute a verb and its complement in Biblical Hebrew. But through a syntactic twist, when read within the rabbinic hermeneutic framework, they are creatively understood as if they comprised a protasis and an apodosis: according to R. Akiba, “if he has purged, then he has completed,” or conversely, “if he has completed, then he has purged” according to R. Judah.Footnote 71
The implications of this grammatical dispute between the two second-century Tannaim is then debated by two third-century Amoraim, R. Yohanan and R. Joshua ben Levi. One holds that R. Akiba and R. Judah were merely following two different exegetical paths toward the same conclusion.Footnote 72 The other holds that their debate has concrete legal implications: R. Akiba, who takes the phrase “and having finished purging” to indicate that “if he has purged, then he has completed,” maintains that once the high priest has performed “purging” (a process that does not include decanting the leftovers upon the base of the altar), the ritual can be considered complete. Conversely, Rabbi Judah, who takes the same phrase to imply that “if he has completed, then he has purged,” opines that it is not until the leftover blood is decanted upon the base of the altar that purgation can be considered effective.
The section translated here concludes with an attempt to identify which of the two Amoraim held that R. Akiba and R. Judah disagree on legal matters, namely, on whether or not failure to decant leftover blood renders the ritual invalid. R. Joshua ben Levi is recorded to have stated: “according to the view that the leftovers are indispensable, he (the priest) must bring a new bull and begin anew inside (the sanctuary).” Since he said “according to the view,” this reveals his opinion that there was a Tannaitic controversy on this matter. Therefore, it must have been R. Joshua ben Levi who opined that the controversy between R. Akiba and R. Judah was about whether the decanting of leftover blood is a sine qua non.Footnote 73
While the Tannaitic debate is articulated in exegetical terms, and while exegetical forces may have actually been at play here,Footnote 74 the sages appear to be split on a fundamental question about the “grammar” of ritual, the ramifications of which are spelled out by some of the Amoraim. If decanting the leftover blood is perceived as mere waste-disposal—the elimination of ritually insignificant ritual byproducts—then in the event that, for whatever reason, there happens to be no residue to decant at the foot of the altar,Footnote 75 this “non-cultic” act is simply skipped.Footnote 76 If, however, the decanting of leftover blood is perceived as an integral part of the ritual of purgation for carrying out the sacrificial rite, then absent residual blood, a whole new animal must be brought, and the entire ritual must begin anew, only to ensure that this time around the requisite blood-disposal is carried out properly.Footnote 77
IV. Discussion
A. An Experiment in Transposition
Focusing on the criteria enumerated in section 2, it appears that both pericopes exemplify theoretical discourse on ritual sacrifice that is characterized by (synchronic and diachronic) multivocality, with the various strata showing through the lines of the text, as reproduced here (in different fonts). Both reflect discursive and dialectic systematizations of a complex textual tradition, and even share similar conventional shorthand scholastic catchphrases.Footnote 78 Both also employ nonintuitive hermeneutic moves, relying on morphological and syntactic sensitivity (more than on symbolism, folk etymology, and other techniques found elsewhere in both Indian and Jewish traditions).Footnote 79 Moreover, they display an interest in the same kinds of hypothetical questions and shared conceptions of what types of arguments qualify as acceptable retorts.
In addition to these general convergences, JMS 4.1.27 and the pericope in BT 52b share a common nonintuitive hypothesis that serves as a driving force for both pericopes, namely, that the absence of the byproducts of a ritual process might conceivably invalidate the entire process and require its re-performance. Though this may not be a commonsense hypothesis, ruling out the nonintuitive offhand runs the risk of reducing ritual to the capacities of the reasoning mind. To sense the degree to which this shared hypothesis is nonintuitive, consider the following (admittedly imperfect) analogy to a modern birthday party ritual. If one imagines a birthday party that leaves no byproducts such as wrapping paper and leftover cake, a dilemma would arise. According to one view, the ritual removal of byproducts can be skipped, but according to another view, such an omission would invalidate the entire party. The guests would need to be re-invited, gifts presented and unwrapped, and a new cake baked, sliced, and served, only to ensure that the second time around some byproducts remain so that they can be disposed of properly.
Couching the convergence in philological rather than logical terms, one may resort to an experiment in “transposition,” in which we excise a section of each text and append it to the other. In this case, the culmination of each of the two excerpts can be removed and attached to the end of the other text. Only a bare minimum of editorial intervention is required: since BT speaks of a bull (not a goat), and since JMS mentions also dung (not just blood), the word par (bull) needs to be emended to śā‘îr (he-goat) and śakṛllohita- (dung and blood) to lohita- (blood).
As a result of this transposition, each of the two texts would remain perfectly intelligible, logically intact, and fundamentally unchanged. To the question “kiṃ bhavati prayojanam?” (“what is the practical ramification?” [= nafka minah]), Śabara’s moderator would respond with the statement, לדברי האומר שיירין מעכבין, יביא שעיר אחר ויתחיל כתחלה בפנים (“according to the view that considers the disposal of byproducts indispensable”—which is the Pūrvapakṣin’s view—then, if there is no blood to be disposed of, “he brings a different goat and begins all over again”). Conversely, the Talmudic pericope would prove that it was R. Joshua who held that there was a debate about whether or not the treatment of leftover blood is a sine qua non, since he would have been accorded with saying, “sāmye sati lohitābhāve ‘nyaḥ paśur ālambhanīyaḥ, lohite ‘prayojakatve lopaḥ” (“If they are on a par [= equally instigators], then if dung and blood are lacking, a different animal is to be killed, but if blood and dung are not instigators, it is skipped”).
Note that in this process, the last nine words in ŚBh on JMS 4.1.27 are transposed with the last ten of the Talmudic pericope. As a result of this Kopfvertauschung, each of the two “heads” suits its new “body” so seamlessly that, had it not been for the intermingling of Sanskrit and Aramaic/Hebrew, the products of this transposition might not have been recognized by readers from each tradition as hybrids. Such a correspondence calls for an explanation.
B. Caveats and Potentials
The “transposition” carried out here is admittedly an unsolicited modern intrusion into the two ancient scholastic traditions, which themselves intruded on the older ritual texts with inconvenient questions. Thus, in demonstrating that the two pericopes remain intact after the experiment, we do not wish to imply that a conversation between members of the two traditions would have been very likely in antiquity, nor that the two pericopes are fundamentally identical.Footnote 80
In many ways, the experiment in transposition highlights the distance between the two traditions, underlining some of the fundamental dissimilarities enumerated above (section 2), such as the anonymity versus historicity of the interlocutors, and the stamp of authorship. Moreover, the fact that both pericopes happen to deal with the disposal of a common substance—blood—should not obscure the fact that there is a major dissimilarity between their attitudes toward blood. In the ancient Israelite sacrificial system, blood is arguably the most sacred of all materia sacra, and often a major reason for killing an animal in the first place.Footnote 81 By contrast, blood (lohita) very rarely figures in the Vedic sacrificial tradition as oblatory material (havis).Footnote 82
Most importantly, the emic models underlying the surface similarities between the two pericopes differ in important ways. While both pericopes point to an identical dilemma—should the ritual be performed anew just because there happens to be no blood to be disposed of—the dilemma arises from different concerns. In JMS 4.1.27, it has to do with the status of blood qua potential member of an abstract category of instigators, while in BT 52b, it has to do with the status of leftover blood qua residual materia sacra.
This divergence reveals a double problem with byproducts of the ritual procedure, which are not mutually exclusive but are analytically distinct. In JMS 4.1.27, the main problem with byproducts is that they tend to occupy an ambiguous place within the fabric of ritual. They are inevitably present within the ritual arena and yet ostensibly useless. As such, their treatment participates in Mīmāṃsā’s systematic analysis of which ritual elements are primary and which nonprimary. In BT 52b, the problem motivating the discussion of the byproducts is their sacrality: materia sacra is perceived as charged with a ritual potential that must be exhausted in the ritual process. Therefore, byproducts threaten to become potent and even dangerous residues, which must be offered or eliminated.
Nevertheless, it is precisely the analytical distinction between the two problems potentially posed by ritual byproducts, a distinction that comes to light more clearly through the juxtaposition of the two pericopes than through the analysis of each on its own terms, that points to the potential explanatory power of the experiment in transposition for the study of these two ritual-textual traditions, and perhaps other traditions as well. With this distinction in mind, it will be possible to return in the future to each of the textual traditions individually and examine how the dilemma, in both of its forms, might inform each.Footnote 83
C. Accounting for the Data
The primary thrust of this study has been to take a first step toward establishing common grounds for discussion between two distinct emic “sciences of sacrifice.” Nevertheless, since each of the scholastic traditions discussed here emerged and evolved within specific historical and social contexts, some account must be given of the various factors that may have contributed to the convergence exemplified in this study. It appears that several distinct factors may account for the similarities we have discussed thus far.
First, both the ancient Vedic and the ancient Israelite traditions, in conversation with which Mīmāṃsā and the Talmud emerged, were characterized by an intense engagement in sacrificial ritual, in the sense that sacrifice played an important role both in action and in thought. Second, both Jewish and Indian intellectual elites in the first few centuries BCE and CE, who inherited the ancient sacrificial ritual texts from the Vedic and biblical traditions, were engaged in a form of scholasticism that was preoccupied with the development of hermeneutic techniques extending well beyond simple lexical equivalences, including counterintuitive textual maneuvers. Third, faced with a significant gap between an authoritative ritual text and ritual practice that did not align with that text, both systems found it necessary to bridge the gap. In the Talmudic context, this situation was determined to a great extent by the fact that sacrifices were no longer permissible after the destruction of the Jerusalem Temple in 70 CE, though gaps between ritual text and ritual practice existed also before the destruction. In the context of Mīmāṃsā, where there was no comparable abrupt cessation of sacrificial praxis, the gap is primarily between Vedic/Śruti texts and their alignment with ritual practice. Nevertheless, since the Mīmāṃsakas, the ritual theoreticians, are in many cases distinct from the yājñikas, the experts in ritual practice,Footnote 84 it appears that here, too, systematization and theorization of sacrificial ritual gained impetus at a time when the discussion of sacrifices was to some extent theoretical.
Finally, a fourth factor, partially intertwined with the first three, is worthy of consideration. Both traditions reflect a lack of appeal to divine intentions as a determining factor in problem-solving: such desires or intentions are left out of the interpretive frame, which must be pressed hard to yield the truth within the text (and thus, within the practice). Thus, according to the theology that is operative in both traditions, sacrifice is not understood and valorized directly in terms of communication with the divine, but rather seems internalized: the text has an infinite surplus of meanings that god-ward language does not anticipate.Footnote 85
Conclusions
This article reveals an unexpected affinity between two vastly different corpora, deriving from the Talmudic and the Mīmāṃsic traditions, in terms of dialectic discourse, terminology, ritual concepts, thought structures, hermeneutic assumptions, and a tendency toward formalization and abstract categorization—in a way that is, as far we know, unique in the ancient world. This affinity allows Talmudists and Mīmāṃsakas, for the first time, to study the discourse about sacrifice from a comparative perspective.Footnote 86
One of the many horizons that such a comparative project opens—and the one pursued in this preliminary study—pertains to the development of a nascent “fundamental science” of ritual. A “fundamental science” of ritual must address questions such as: Faced with a continuum of ritual activity, how can we know where one act ends and the next begins? What determines whether one ritual is “like” another (within a single cultural context, or even cross-culturally)? What is the hierarchic relation between action and material: is materia sacra merely a means for carrying out ritual action; or is the sacrificial process merely a means for harnessing the ritual potential of the material, which must be exhausted “to the dregs”? What does it mean for a ritual to fail or to succeed? And finally—as in the example treated here—where does one draw the line between ritual and nonritual material and action?
Significant sections of Mīmāṃsic and Talmudic literature are dedicated to precisely these questions, both explicitly and implicitly. Thus, scholarly discussion on these fundamental aspects of ritual can be nuanced by incorporating the insights of these emic perspectives, stripped, through the comparative lens, of incidental considerations that derive from local exegetical concerns rather than from an underlying ritual “grammar.” The abstract principles and categories developed and refined through a joint study of these two emic “sciences” will assist in developing and testing operative categories for a more general study of ritual, allowing one to analyze rituals from additional cultural domains, including many that lack (second- or even first-order conceptualization) reflective texts on ritual.
This study addressed one fundamental question in such a “science of ritual,” namely, the status of ritual byproducts. The treatment of byproducts, leftovers, and residues is, ironically, a cardinal aspect of sacrificial discourse. This theme, which was already the focus of attention in many Vedic and biblical (first-order) ritual texts,Footnote 87 became the locus of intense sophisticated analysis in (second-order) Mīmāṃsic and rabbinic texts.Footnote 88 Perhaps this is so because leftovers, byproducts, residues, and other nonprimary components reveal an instability inherent in ritual sacrifice: offering materia sacra inevitably generates byproducts, which occupy an ambiguous place within the ritual domain—they are not offerable and yet they are sacred and cannot be disposed of in any which way. Their treatment thus serves as a litmus test for distinguishing between degrees of ritualization; in fact, it is a way of asking a most basic question: What is ritual and what is not?Footnote 89
Additionally, examining this aspect of discourse on ritual points to a potential ramification of Talmudo-Mīmāṃsā. Both traditions are involved in correlating rituals and texts in ways that lead them to highlight elements that lie in the shadows, as it were— byproducts of the ritual process whose ritual potential is questionable, and textual residues such as superfluous morphemes and apparent pleonasms whose hermeneutic potential must be exhausted. Certain scholarly attitudes toward ostensibly superfluous elements within sacred texts are thus, in a sense, a mirror image of ritual attitudes toward materia sacra derived from sacrificial victims.
Above (“Discussion,” part A), we likened the textual experiment carried out in this study to a Kopfvertauschung, a “transposition of heads,” whereby switching the final arguments of two pericopes—one Mīmāṃsic and the other Talmudic—left the line of reasoning fundamentally intact in both. We have now also considered some of the factors that may have contributed to the similarity between the two distant intellectual traditions—rendering such an unlikely experiment feasible—and the horizons that such an endeavor may open up. Head-replacement can be a violent and risky business, even when rooted in a commendable fascination with differences and an admirable “yearning for mutual exchange.”Footnote 90 If performed in haste, the results can be unpredictable, judging from precedents ranging from the formation of the elephant-headed Gaṇeśa in Purāṇic loreFootnote 91 to R. Judah the Hindu’s testimony on the sea monster redivivus in a Talmudic tall tale.Footnote 92 Nevertheless, a tradition extending back to the Ṛgveda suggests that, if performed with due caution, as in the case of the Aśvins, Dadhyañc, and the horse’s head, it can lead to the obtainment of precious knowledge about the secrets of sacrifice.Footnote 93
Open access
